NASASpaceFlight.com Forum

SpaceX Vehicles and Missions => SpaceX Starship Program => Topic started by: Nevyn72 on 02/15/2020 10:18 pm

Title: Starship On-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/15/2020 10:18 pm
Rather than clutter up the Off Earth modification and Assembly thread it's probably time that the In-orbit refueling concept gets it's own thread.

My own thoughts are that some concepts are getting a bit complicated and convoluted for their own good, especially at such an early stage in the process.

The first stage is and should be SS to SS refueling. This has the advantages of;
- simplest hardware
- you are only manufacturing one type of orbital hardware
- greatest flexibility (of mission profile)
- proof of concept with the most capable vehicle (ie. a fully maneuverable space ship)

In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

You could have a small nose section containing header tanks for the RCS system, batteries, avionics and even a deployable solar array. The existing SH design (with a SS thrust structure) is already devoid unnecessary external fittings found on the SS such as landing legs, flaps and thermal tiles.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Coastal Ron on 02/15/2020 10:33 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I like the simplicity of this idea, and as you pointed out, it doesn't require any new hardware design.

Initially you might have to use a Super Heavy that is fairly new, but over time you could use older Super Heavies that need to be retired, and retiring them in space to use as a tanker might be an option.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/15/2020 10:40 pm
the problem of too little thrust at lift off is not due to the fact that there is too little fuel, but to the physical limitations of the materials of which the engine is built, so there is a limit of maximum thrust per unit of surface area of the rocket base, which implies the maximum height of the rocket and the starship is already close to this limit, so adding more tanks on top makes t/w ratio worse not better
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/15/2020 10:46 pm
the problem of too little thrust at lift off is not due to the fact that there is too little fuel, but to the physical limitations of the materials of which the engine is built, so there is a limit of maximum thrust per unit of surface area of the rocket base, which implies the maximum height of the rocket and the starship is already close to this limit, so adding more tanks on top makes t/w ratio worse not better
My thought was to use a partly filled SH (tank) section, thus reducing total takeoff weight ratio to an amount similar to a SS+SH combo.

Using the connection between the SH components you could feed some additional fuel to the SH (launch) section to allow it to burn for a longer duration thus imparting a greater Delta V to the SH (tank) section.

I'm not a rocket engineer so this might just be a pointless thought bubble.....

EDIT: Clarity & grammar
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: fael097 on 02/15/2020 10:50 pm
You think the tanker variant will have just two huge fuel/ox tanks (plus header tanks) or separate regular tanks and storage tanks?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/15/2020 10:54 pm
if super heavy had more dv and used it, it would cause problems with entering the atmosphere and landing
I'm not a rocket engineer either, I'm just an engineer  ;)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/15/2020 10:55 pm
You think the tanker variant will have just two huge fuel/ox tanks (plus header tanks) or separate regular tanks and storage tanks?

My concept is literally a standard SH with SS thrust structure and a small nosecone for support equipment.....
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/15/2020 10:56 pm
if super heavy had more dv and used it, it would cause problems with entering the atmosphere and landing
I'm not a rocket engineer either, I'm just an engineer  ;)

Would extra fuel for extra boost back help?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: fael097 on 02/15/2020 10:59 pm
You think the tanker variant will have just two huge fuel/ox tanks (plus header tanks) or separate regular tanks and storage tanks?

My concept is literally a standard SH with SS thrust structure and a small nosecone for support equipment.....

I guess I mean the SS tanker variant.
On that note, I don't think a SH could be a tanker, SH wouldn't survive orbital reentry, would it?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/15/2020 11:00 pm
if super heavy had more dv and used it, it would cause problems with entering the atmosphere and landing
I'm not a rocket engineer either, I'm just an engineer  ;)

Would extra fuel for extra boost back help?

Yes, but you couldn't take it
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/15/2020 11:08 pm
Current design of starship barely works on paper, there really is no room for modifications, and I don't see any need for the tanker to be bigger than a starship. Refueling anyway will probably be done with one tanker parked in orbit which will be refueled several times from Earth and then it will refuel the starship parked in orbit (thanks to this the starship will be fully refueled with one refueling which will increase safety)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BrianPeterson on 02/15/2020 11:44 pm
Current design of starship barely works on paper, there really is no room for modifications, and I don't see any need for the tanker to be bigger than a starship. Refueling anyway will probably be done with one tanker parked in orbit which will be refueled several times from Earth and then it will refuel the starship parked in orbit (thanks to this the starship will be fully refueled with one refueling which will increase safety)

Musk hates the idea of an orbiting tanker. What has been stated is one Starship fueled up by an additional 3 to 4 other Starship launches.

Don't have any idea what lOX or Liquid methane weight, but payload bay converted to a tank is probably how this will work. .
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 02/16/2020 12:01 am
From what we've seen of the starship thrust structure, there is a single pipe from the CH4 header tank, to the engines, and down to  the skirt where it will presumably form the refueling system.

A single pipe means the connector will need to be androgonus.

I'm less familiar with airticht androgonus designs- any suggestions for the pipe cap?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: fael097 on 02/16/2020 12:19 am
How much propellant mass a tanker Starship could haul into LEO? Regular SS carries 1200 t of prop and 100 t of payload, so tanker SS would only carry 1300 t of prop? I mean, since it can carry more prop it can burn more prop, and it has internal space for carrying 2100 t of propellant without compromising fairing space for header tanks, batteries, actuators, etc. but could a booster even lift that up? and in the end, would a 2100 t prop tanker have more fuel left when it gets to LEO than say a 1300 t prop tanker?

This is a tanker with capacity for 2100 t:
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: joek on 02/16/2020 12:23 am
...
Musk hates the idea of an orbiting tanker. What has been stated is one Starship fueled up by an additional 3 to 4 other Starship launches.
...

"Hate" seems a bit strong; do you have a cite?  Pragmatic is likely a more appropriate description.  Not to mention that Musk appears to be trying to stay out of the "no depots" debate.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/16/2020 12:30 am
You think the tanker variant will have just two huge fuel/ox tanks (plus header tanks) or separate regular tanks and storage tanks?

My concept is literally a standard SH with SS thrust structure and a small nosecone for support equipment.....

I guess I mean the SS tanker variant.
On that note, I don't think a SH could be a tanker, SH wouldn't survive orbital reentry, would it?

It wouldn't need to survive re-entry, it stays up there.  ;)

When it finishes the current mission (could be multiple refueling events for a particular synod) it can reposition to new orbit as required/desired using the onboard SS thrust structure and surplus onboard fuel.

Note: this is all based on the concept that you don't a crewed SS sitting around waiting for multiple refueling events. You want a tanker up there waiting for it so refueling and departure can occur as quickly as possible. Having a SH based unit means you have surplus fuel available to top up the departing SS as much as possible.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/16/2020 12:32 am
if super heavy had more dv and used it, it would cause problems with entering the atmosphere and landing
I'm not a rocket engineer either, I'm just an engineer  ;)

Would extra fuel for extra boost back help?

Yes, but you couldn't take it

Well you could, the extra fuel for a longer boost back is part of the transfer from the SH (tank) being launched.

The only variable is how much fuel is left in the SH (tank) when it reaches the desired orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 02/16/2020 12:38 am
if super heavy had more dv and used it, it would cause problems with entering the atmosphere and landing
I'm not a rocket engineer either, I'm just an engineer  ;)

Would extra fuel for extra boost back help?

Yes, but you couldn't take it

Well you could, the extra fuel for a longer boost back is part of the transfer from the SH (tank) being launched.

The only variable is how much fuel is left in the SH (tank) when it reaches the desired orbit.
are you trying to SSTO? Because that never works out as well as you think it does.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BrianPeterson on 02/16/2020 12:39 am
"Hate" seems a bit strong; do you have a cite?  Pragmatic is likely a more appropriate description.  Not to mention that Musk appears to be trying to stay out of the "no depots" debate.

It was a video interview, where you could say he scoffed at the idea. Derided the idea. He doesn't like the idea much. I've watched so many interviews with him, finding the exact one is a needle in a haystack. 

I on the other hand think an orbital platform could be the best idea. What happens if one of the four launches required to refuel fails? Or weather suddenly makes it impossible to launch. You could have to scrub the mission or have a starship and crew hang out for days/weeks/months waiting for fueling. Having a platform that is constantly stocked up means you can launch knowing the required fuel is there and get the mission under way.

Musk has other ideas: this was easy to find.
<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Orbital refilling is vital to humanity’s future in space. More likely spacecraft to spacecraft (as aircraft do aerial refueling), than a dedicated depot, at least at first.</p>&mdash; Elon Musk (@elonmusk) August 1, 2019 (https://twitter.com/elonmusk/status/1156970909258829824?ref_src=twsrc%5Etfw) <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/16/2020 12:43 am
...
Musk hates the idea of an orbiting tanker. What has been stated is one Starship fueled up by an additional 3 to 4 other Starship launches.
...

"Hate" seems a bit strong; do you have a cite?  Pragmatic is likely a more appropriate description.  Not to mention that Musk appears to be trying to stay out of the "no depots" debate.

I believe (my opinion, no citation) that the issue Musk has with an orbiting tanker is that it's located for where you need it initially but the desired location changes with each mission so quickly becomes useless.

My suggestion means that when needed you can reposition the unit to the next ideal staging orbit.

You will be leaving a SS in orbit in this role anyway but with the SS(tanker)-SS you are launching a whole lot of additional weight to allow the SS(tanker) to return to Earth, just so you can then re-launch the whole lot again to a different orbit for the next mission.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/16/2020 12:45 am
are you trying to SSTO? Because that never works out as well as you think it does.

Not at all, the SH (tank) is a second stage using a conventional SS thrust structure (you could even delete the Sea level Raptors).

You use a conventional SH for the first stage.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 02/16/2020 12:56 am
...
Musk hates the idea of an orbiting tanker. What has been stated is one Starship fueled up by an additional 3 to 4 other Starship launches.
...

"Hate" seems a bit strong; do you have a cite?  Pragmatic is likely a more appropriate description.  Not to mention that Musk appears to be trying to stay out of the "no depots" debate.

I believe (my opinion, no citation) that the issue Musk has with an orbiting tanker is that it's located for where you need it initially but the desired location changes with each mission so quickly becomes useless.

My suggestion means that when needed you can reposition the unit to the next ideal staging orbit.

You will be leaving a SS in orbit in this role anyway but with the SS(tanker)-SS you are launching a whole lot of additional weight to allow the SS(tanker) to return to Earth, just so you can then re-launch the whole lot again to a different orbit for the next mission.
the "extra weight" of a tanker let's you use Earth's atmosphere to adjust inclination and Earth's resources to reset the rocket equation for each mission.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: aero on 02/16/2020 01:01 am
I know that there is a very large, if not an infinite number of possible orbits for the tankers, but practically speaking, how many tankers would be needed to cover 90% of the orbits that would be used? For example, would the LEO parking orbit for all Mars-bound Starships be the same or does the parking orbit depend on the Earth/Mars locations so strongly that the Mars Transfer Orbit can't be efficiently adjusted during or after TMI? By efficiently adjusted, I mean "adjusted without serious impact to the payload or fuel reserves."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 02/16/2020 01:05 am
I know that there is a very large, if not an infinite number of possible orbits for the tankers, but practically speaking, how many tankers would be needed to cover 90% of the orbits that would be used? For example, would the LEO parking orbit for all Mars-bound Starships be the same or does the parking orbit depend on the Earth/Mars locations so strongly that the Mars Transfer Orbit can't be efficiently adjusted after launch? By efficiently adjusted, I mean "adjusted without serious impact to the payload or fuel reserves."
I recall reading that Earth Mars cyclers could reasonably adjust their orbits to match the double flyby every 6 synods. That would imply six reasonable ejection angles for a Mars transfer.
But what about the moon? What about outer solar system destinations? What about different launch inclinations to Mars for that matter.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/16/2020 01:10 am
How much propellant mass a tanker Starship could haul into LEO? Regular SS carries 1200 t of prop and 100 t of payload, so tanker SS would only carry 1300 t of prop? I mean, since it can carry more prop it can burn more prop, and it has internal space for carrying 2100 t of propellant without compromising fairing space for header tanks, batteries, actuators, etc. but could a booster even lift that up? and in the end, would a 2100 t prop tanker have more fuel left when it gets to LEO than say a 1300 t prop tanker?

This is a tanker with capacity for 2100 t:

A simple example with a tanker that takes 200t of fuel instead of 100t:

SH 280t dry, 3300t propellant
SS 220t dry (with payload), 1200t propellant
stack weight 5000t
final mass for SH 1700t

SH with SS assuming isp 350 (simplification, isp increases during flight when the ambient pressure drops) has dV 3,7km/s
SS assuming isp 380 has dV 6,95km/s

now with a tanker that has 200t of fuel as a load so it weighs 1520t
stack weight 5100t, final mass for SH 1800t

SH dV 3,57km/s (130m/s less and slightly more gravity drag)
tanker dV  must be 7,08km/s to compensate and you have to burn 1293t of fuel to do that, so you're left with 107t of fuel instead of 100t in orbit

and the more fuel you take, the more gravity drag (which is not taken into account in this calculation) eats

Orbital mechanics is brutal

Only real way to increase tanker capacity is to shave dry weight, or you need bigger booster (which MUST be larger diameter)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: 50_Caliber on 02/16/2020 02:00 am
A depot at EML-1 makes sense for Lunar missions, if they build a reusable Lunar lander with similar mass ratio to SS, then they can refuel at EML-1, go to the moon, land and come back ready to refuel for next mission. The SS has a 6km+ delta V, so a lander with similar delta V could be refueled with perhaps 3-4 tanker flights to EML-1 station to refuel the lander. It can make the trip to the moon, land and come back to EML-1 for around 5km/s delta V. If it uses methane\lox , then it's a pretty straightforward refueling exercise that meshes perfectly with the SS architecture. We bypass any need for ISRU mining on the moon and it fast-tracks lunar development.

This also gives us an excuse to make a legit 2001 space station at that point.  8)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/16/2020 04:48 am
Orbital mechanics is brutal

Only real way to increase tanker capacity is to shave dry weight, or you need bigger booster (which MUST be larger diameter)

Interesting, so you take the first phase of In-orbit refueling to be SS to SS with a goal of having a delivered weight of around 100t of fuel. We already believe this will be the case so can use it as a starting point.

Now in my speculated second phase where you have a dedicated, and movable, In-orbit refueling tanker/depot you now have a starting target weight.

The question then becomes how much would my postulated SH (tank) weigh?
Would it be more or less than a fully functional SS with 100t of fuel?

Alternatively, if you stripped the aero-surfaces, landing structures, thermal tiles and sea-level Raptors off a SS how much less would it weigh and hence how much extra fuel could it carry to orbit?
(This option wouldn't give you the additional capacity of the SH (tank) but would function in the same way)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: zodiacchris on 02/16/2020 05:20 am
Not quite sure where the SH as tanker second stage would be required. Wouldn’t it make more sense to simply build the tanker SS with extended tanks by fitting a few extra hoops in the tanks and deleting the payload bay? With the exception of flight relevant systems no fit out of the nose, no decks, air locks, life support, windows etc. That would be the most KISS solution and not change SS aerodynamics or mould line.

The weight savings of the reduced equipment plus the normal SS payload capacity would give you around 150 -200 tonnes of refuelling capacity. Send Tanker 1 up, refuel that from Tankers 2 -6, then launch the Mars SS last to take on the fuel from Tanker 1.

That avoids loiter time for the manned Mars SS in orbit and possible hang ups with bad weather.

Of course Elon did say a while ago that the tanker version would look weird, so yeah, what do we know? ???
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: libra on 02/16/2020 06:30 am
How much propellant mass a tanker Starship could haul into LEO? Regular SS carries 1200 t of prop and 100 t of payload, so tanker SS would only carry 1300 t of prop? I mean, since it can carry more prop it can burn more prop, and it has internal space for carrying 2100 t of propellant without compromising fairing space for header tanks, batteries, actuators, etc. but could a booster even lift that up? and in the end, would a 2100 t prop tanker have more fuel left when it gets to LEO than say a 1300 t prop tanker?

This is a tanker with capacity for 2100 t:

A simple example with a tanker that takes 200t of fuel instead of 100t:

SH 280t dry, 3300t propellant
SS 220t dry (with payload), 1200t propellant
stack weight 5000t
final mass for SH 1700t

SH with SS assuming isp 350 (simplification, isp increases during flight when the ambient pressure drops) has dV 3,7km/s
SS assuming isp 380 has dV 6,95km/s

now with a tanker that has 200t of fuel as a load so it weighs 1520t
stack weight 5100t, final mass for SH 1800t

SH dV 3,57km/s (130m/s less and slightly more gravity drag)
tanker dV  must be 7,08km/s to compensate and you have to burn 1293t of fuel to do that, so you're left with 107t of fuel instead of 100t in orbit

and the more fuel you take, the more gravity drag (which is not taken into account in this calculation) eats

Orbital mechanics is brutal

Only real way to increase tanker capacity is to shave dry weight, or you need bigger booster (which MUST be larger diameter)

Excellent post, really. Another way of putting it is as follow.

What is the tanker ?

Basically, it should be the usual Starship, its propellant mass, and its payload.

Then in the case of a tanker, the payload, obviously, should be the maximum amount of propellant you can haul into orbit by sacrificing everything but the kitchen sink, to rise payload.

That is, the cargo area becomes another fuel tank.

Then there is the problem of mass: what part of the "standard Starship"are you willing to sacrifice and turn into propellant mass, that is, useful tanker payload ?

Do you want a dumb prop tank inside the cargo area ? or do you start modifying or removing cargo area structure ?

There it depends of two things a) commonality with a "standard Starship" and b) reusability or not.

The more you sacrify commonality with the standard Starship, the more propellant raw mass you get into orbit, but of course tanker development costs rise...

An interesting historical exemple are the Boeing 707 innumerable variants, tankers included. While the KC-135 was a separate development of 707 from day one (different fuselage), there were also civilian 707s turned into makeshift, cheaper tankers - also far less efficient.
Well then the kerosene plumbery, the distribution of kerosene between the aircraft own tanks and the tanks dedicated to refueling, where vastly different.
Musk will face the same issues with tanker Starship.

Quote
Orbital mechanics is brutal

That's an understatement. The basic Starship has 1200 mt of propellant in the tanks when separating from SH, yet most of this vanishes during ascent. If it carries zero payload instead of 150 mt, then somewhat automatically, 150 mt of props remain in the tanks. That's a standard Starship. Then by scrounging whatever can be scrounged of structure and weight, maybe you can rise this to 200 mt. Maybe. Since Starship needs to be fully-fueled for Mars departure, then you need at least 6 and probably 7 tanker flights to fill that thing. Brutal numbers, sure.

How about starting from this number - 7 tankers - and start stockpiling 10 - 12 of them in orbit, docked to a dumb truss ? Would not be that expensive and would give margin.

But that's for 1 flight; Imagine for 1000 or more, the sheer number of tankers that will be needed...

As for different orbits with different inclinations - no panic. There is a simple formula to calculate plane changes delta-v
https://en.wikipedia.org/wiki/Orbital_inclination_change

Musk advantage here is the high specific impulse of his methalox engines. I can see "Starship depots" making reasonable plane changes by firing their engines.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ultrafamicom on 02/16/2020 06:52 am
How much propellant mass a tanker Starship could haul into LEO? Regular SS carries 1200 t of prop and 100 t of payload, so tanker SS would only carry 1300 t of prop? I mean, since it can carry more prop it can burn more prop, and it has internal space for carrying 2100 t of propellant without compromising fairing space for header tanks, batteries, actuators, etc. but could a booster even lift that up? and in the end, would a 2100 t prop tanker have more fuel left when it gets to LEO than say a 1300 t prop tanker?

This is a tanker with capacity for 2100 t:

A simple example with a tanker that takes 200t of fuel instead of 100t:

SH 280t dry, 3300t propellant
SS 220t dry (with payload), 1200t propellant
stack weight 5000t
final mass for SH 1700t

SH with SS assuming isp 350 (simplification, isp increases during flight when the ambient pressure drops) has dV 3,7km/s
SS assuming isp 380 has dV 6,95km/s

now with a tanker that has 200t of fuel as a load so it weighs 1520t
stack weight 5100t, final mass for SH 1800t

SH dV 3,57km/s (130m/s less and slightly more gravity drag)
tanker dV  must be 7,08km/s to compensate and you have to burn 1293t of fuel to do that, so you're left with 107t of fuel instead of 100t in orbit

and the more fuel you take, the more gravity drag (which is not taken into account in this calculation) eats

Orbital mechanics is brutal

Only real way to increase tanker capacity is to shave dry weight, or you need bigger booster (which MUST be larger diameter)

On the other hand SH now needs 130m/s less dv for boostback and stages closer to launch pad, so more propellant can be used to boost SS, which gives a further increase in payloads.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 02/16/2020 10:14 am
I think there needs to be a distinction made between the tanker that remains in orbit and stockpiles fuel for the mission SS and the tanker that is used to transport fuel to the orbital stockpile.

The one that remains in orbit just needs to be delivered there (whether it is designed to return to Earth or not) and the amount of fuel it retains on arrival in orbit is largely academic. Design for the function becomes the priority.

The one(s) that deliver the fuel to orbit need to be optimised to maximise the fuel delivered and expedite launch cycle frequency, that becomes the design priority for these units.

The differing requirements mean that each unit are not necessarily identical in design.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Eka on 02/16/2020 11:46 am
I've never been able to pencil a scenario where an orbital fuel depot is better than just sending up the SS tankers. It all comes down to flexibility, the cost of putting that infrastructure into space, and how refueling is done. That last one, how refueling is done, is the killer. Once part or all of the propellants are available from outside Earth's gravity well, then the math may change, but not until then, and not always.

On how in orbit refueling is done. With the current Starship design, it MUST be done under acceleration. You must take that into account for any system you develop. Will that inflatable tank be able to handle the repeated accelerations and decelerations needed? Think of how much more mass you are accelerating and decelerating when you have a propellant load in that depot greater than than one SS can receive.
An fuel depot who can be one tanker or an dedicated tanker with an sort of sun shade, think an air mattress on an beam makes sense in that you can top it up during slow days and the Mars or Moon mission only need to refuel once.

Acceleration is needed to settle the fuel at the bottom of the tank, this does not to be an high acceleration, The dedicated fuel depot tanker will have pumps doing most of the work the trust is just to have the liquids settle at the bottom.
With a pump, you need enough acceleration and propellant head or tank pressure to keep the pump from cavitating.* Might as well just use a bit of acceleration, and tank pressure alone. Keep it as simple as possible. One thing to remember is both the source tank, and the receiving tank need the acceleration. The propellant must also be settled in the receiving tank to vent the gasses being displaced by the fuel. Want to transfer propellant without acceleration, one still needs the pressure, but another form of containment is needed. Fuel bladders in both source and destination tanks would work, but they add weight and can fail. Yes, many gotchas for fuel transfer in orbit. Gravity makes a lot of things easier here on earth.

* This is why 6 bar is the design goal for SS tanks. They need that pressure to keep the Raptor turbopumps from cavitating as they suck in huge amounts of fuel.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 02/16/2020 02:14 pm
To synthesize from previous posts, I agree that the tanker version of SS has the  function of delivering maximum feasible propellant to LEO. We should minimize structural mass and eliminate any feature that does not contribute to this function. It could be perceived in stages:

First, consider the standard SS with extra propellant tanks in the cargo area. We note that cargo SS features a large payload volume to accommodate a variety of payloads, such as equipment that occupies only part of the volume. In this case the mass carried in the cargo bay leaves some empty space in the cargo bay.  Liquid propellant, however, could occupy every cc of available cargo volume, provided propellant density allows propellant mass to be less than cargo mass capacity of 1100 tonnes. (I am no longer adept to calculate the excess volume.)

Second iteration, shorten tanker SS cargo bay. Determine the exact length required to carry the mass of propellant that is (edit:) to be transferred to another SS. The resulting mass savings allows propellant load to increase by an equal amount.

Third iteration, eliminate the cargo bay. Increase the length of main propellant tanks to accommodate both the fuel required to reach LEO and the fuel to be transferred.

Fourth iteration, increase diameter of fairing. Note that the most efficient shape to carry liquid is a sphere, which has less surface area than a cylinder for equal volume. In this step we would widen the nose section of tanker SS to be more like a sphere. It would be similar to the wide fairings used for some missions of Falcon 9, only more so. Tanker SS would assume the appearance of a Lollypop, which may be why Musk has said tanker SS would look kinda weird.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 02/16/2020 02:30 pm
To synthesize from previous posts, I agree that the tanker version of SS has the  function of delivering maximum feasible propellant to LEO. We should minimize structural mass and eliminate any feature that does not contribute to this function. It could be perceived in stages:

First, consider the standard SS with extra propellant tanks in the cargo area. We note that cargo SS features a large payload volume to accommodate a variety of payloads, such as equipment that occupies only part of the volume. In this case the mass carried in the cargo bay leaves some empty space in the cargo bay.  Liquid propellant, however, could occupy every cc of available cargo volume, provided propellant density allows propellant mass to be less than cargo mass capacity of 1100 tonnes. (I am no longer adept to calculate the excess volume.)

Second iteration, shorten tanker SS cargo bay. Determine the exact length required to carry the mass of propellant that is feasible to launch to LEO. The resulting mass savings allows propellant load to increase by an equal amount.

Third iteration, eliminate the cargo bay. Increase the length of main propellant tanks to accommodate both the fuel required to reach LEO and the fuel to be transferred to another SS.

Fourth iteration, increase diameter of fairing. Note that the most efficient shape to carry liquid is a sphere, which has less surface area than a cylinder for equal volume. In this step we would widen the nose section of tanker SS to be more like a sphere. It would be similar to the wide fairings used for some missions of Falcon 9, only more so. Tanker SS would assume the appearance of a Lollypop, which may be why Musk has said tanker SS would look kinda weird.
This all begs the question, why?

A tanker exists to refuel Starship. Therefore the most fuel it needs at any one time is enough to fully fuel one starship. Is there an existing ship with tanks that size, one you're already mass producing? Well, yes... Starship itself.

Any permanant depot must consider the trade between reentry mass, which allows the rocket to come home and resupply, and plane chance fuel, which must be lifted to orbit and docked to the existing depot in order to move the depot and any residual fuel to the new orbit.

If a plane change requires more than a mass ratio of 1- that is, if the mass of fuel needed for a useful "next plane" change of orbit is more than the (already reduced, by the removal of reentry mass) dry mass of the tanker, without fuel, it's easier to launch an entirely new tanker in the new orbit than move the old one.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 02/16/2020 03:07 pm
This all begs the question, why?
...
...
The reason that first comes to mind is an insurance policy. If there are unexpected fuel leaks during launch to orbit or minor spills during fuel transfers, we would still be able to deliver enough propellant over the planned number of launches to complete the mission.

Nevertheless, your point is well taken and SS tanker version is low priority.

Added: We may be addressing two different types of “tanker.” I am addressing the version of SS used to deliver propellant to orbit where it transfers fuel to another SS in orbit or to a “tanker-in-orbit,” which is a substitute for a propellant depot. I would call the tanker-in-orbit an SS Depot or some different name to avoid confusion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ZChris13 on 02/16/2020 03:19 pm
...
Musk hates the idea of an orbiting tanker. What has been stated is one Starship fueled up by an additional 3 to 4 other Starship launches.
...
"Hate" seems a bit strong; do you have a cite?  Pragmatic is likely a more appropriate description.  Not to mention that Musk appears to be trying to stay out of the "no depots" debate.
I believe (my opinion, no citation) that the issue Musk has with an orbiting tanker is that it's located for where you need it initially but the desired location changes with each mission so quickly becomes useless.
My suggestion means that when needed you can reposition the unit to the next ideal staging orbit.
You will be leaving a SS in orbit in this role anyway but with the SS(tanker)-SS you are launching a whole lot of additional weight to allow the SS(tanker) to return to Earth, just so you can then re-launch the whole lot again to a different orbit for the next mission.
Is that an issue?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: dnavas on 02/16/2020 03:52 pm
If a plane change requires more than a mass ratio of 1- that is, if the mass of fuel needed for a useful "next plane" change of orbit is more than the (already reduced, by the removal of reentry mass) dry mass of the tanker, without fuel, it's easier to launch an entirely new tanker in the new orbit than move the old one.

Just so.  In LEO there are a large number of possible waypoints, each of them mutually "inaccessible."  The question is whether there are a small enough "significant" set that the idea is tenable.  Because it's likewise a small iteration from fuel depot to electric LEO/GEO tug dock, but that iteration only makes sense if there are reasonable targets for LEO.  If only we had SST-lagrange point, but we don't.

Is there some way to determine, given a single SS tanker load of fuel, what LEO coverage exists for about 100mt of fuel?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wannamoonbase on 02/16/2020 04:11 pm
Only real way to increase tanker capacity is to shave dry weight, or you need bigger booster (which MUST be larger diameter)

I would not be surprised to see SH have a larger diameter than SS. 

The engines under the 9 meter SH have been crammed in and little or no room for growth.

The number of refueling flights seem like they could be reduced with a larger booster or more staging (3rd stage or FH style.)

If these two vehicles are going to be so cheap to build then why not have a different diameter for each? 

Edit: If the vehicles can eventually be as cheap to build and easy to operate as projected then the cost of fuel for each launch drives cost then at some point there are more efficient ways to get 100 tons of fuel on orbit. 

I suppose there is a long way to go and there is plenty of time to optimize.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TrevorMonty on 02/16/2020 04:14 pm
If SS needs refuelling it is going to GEO, moon or mars. Which should reduce number of possible tanker/depot orbits.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: aero on 02/16/2020 04:40 pm
If SS needs refuelling it is going to GEO, moon or mars. Which should reduce number of possible tanker/depot orbits.

Remember, as pointed out above, plane change delta-V is linear with velocity. So if it is going beyond LEO then the plane change maneuver becomes cheaper the lower the velocity of the Starship. My orbital mechanics is beyond rusty but I think I recall that doing a plane change maneuver during the Mars transfer is the most economical. That means all this discussion of doing LEO maneuvers to calculate plane change delta-V is a sub-optimal topic.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: hkultala on 02/16/2020 05:30 pm
Only real way to increase tanker capacity is to shave dry weight, or you need bigger booster ....

yes.

Quote
(which MUST be larger diameter)

No, it does not have to, unless going to MUCH bigger.

The current specifications in SpaceX web site says "72 MN" for Superheavy thrust and 4600 tonnes total propellant + payload.

Assuming the spaceship + superheavy together have empty mass of  less than 400 tonnes, this means maximum of 5000 tonnes liftoff weight.

This is T/W ratio of 1.44. WHich is a lot

They can increase the first stage fuel tank size to something like 4000 tonnes, increasing the empty weight by only ~30-40 tonnes.

This would mean T/W ratio of 1.25, which is still enough. There would be more gravity losses, but payload would be greater.

There has also been talk about 250-tonne-thrust version of Raptor. The optimal nozzle size for that would be slightly bigger than the nozzle size of current raptor, but they can make it have the same nozzle and lose ~1% of isp, and fit the same number of those engines under the craft. And then they can increase the thrust from 7.2 MN to about 8.8 MN.

That would allow making the tanks much longer, adding considerable amount of propellant, while still keeping the T/W ratio very good.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/16/2020 06:01 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Eer on 02/16/2020 06:36 pm
I look forward to the day when fuel is sourced from space itself - comets, asteroids, whatever. At that point, raising fuel from earth surface becomes unnecessary for most uses.

To begin with, though, “just in time” fuel supply can still benefit from “cross dock” delivery by filling an SS depot version (maybe never intended to land) so cruise SS with passengers can load and go just makes sense for many reasons (not least reducing suspense on part of passengers and crew whether all the tankers will arrive on schedule). That sort of risk reduction makes sense to me.

A small squadron of such depot SS versions can be deployed and filled in any number of orbits and gateway destinations, just like filling stations and recharging stations serve traffic where traffic regularly need services.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/16/2020 06:43 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.

Unnecessary complexity, such manoeuvres can be done with a simple starship tanker, fael097 calculated that the bulkhead shift alone will give a capacity of 2100t, so if it sends such a tanker with 1300t of fuel on the LEO will arrive with 100t, then it will be 20x refuelled to full. consumption of 700t gives dV 1.4km/s and in the tanker is 1200t to refuel the starship and 200t to return to the LEO (there would still be 100t left). when the starship enters the LEO it will be once refueled with another tanker which will give it dV 1.4km/s then both can meet in a higher orbit and the starship can be refueled to the full.

Instead of refuelling and docking, you have two refuellings and don't need a special tanker that can dock with its nose  ::)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/16/2020 06:57 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.

Unnecessary complexity, such manoeuvres can be done with a simple starship tanker, fael097 calculated that the bulkhead shift alone will give a capacity of 2100t, so if it sends such a tanker with 1300t of fuel on the LEO will arrive with 100t, then it will be 20x refuelled to full. consumption of 700t gives dV 1.4km/s and in the tanker is 1200t to refuel the starship and 200t to return to the LEO (there would still be 100t left). when the starship enters the LEO it will be once refueled with another tanker which will give it dV 1.4km/s then both can meet in a higher orbit and the starship can be refueled to the full?

Instead of refuelling and docking, you have two refuellings and don't need a special tanker that can dock with its nose  ::)

This is all discussed in the thread, but briefly: the problem is that refueling in a higher orbit means multiple trips through the (edge of the) Van Allen belt. It also requires a greater number of rendezvous/docking events (ie more risk).

I'm also drawn to the Pusher Tanker architecture for the mission possibilities of an ultra-high delta-v "Starkicker." Starkicker only gets one chance for its escape burn, and Oberth says you want your engine firing to be as short a duration as possible. So there's no time to use re-fueling during the delta-v slam, and another approach (Pusher Tanker, or simply StarPusher (ie Pusher Tanker but with no propellant transfer to the departure stage)) is needed if you need/want more delta-v than Starkicker alone can deliver.

https://twitter.com/elonmusk/status/1111760133132947458
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/16/2020 07:05 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.

Unnecessary complexity, such manoeuvres can be done with a simple starship tanker, fael097 calculated that the bulkhead shift alone will give a capacity of 2100t, so if it sends such a tanker with 1300t of fuel on the LEO will arrive with 100t, then it will be 20x refuelled to full. consumption of 700t gives dV 1.4km/s and in the tanker is 1200t to refuel the starship and 200t to return to the LEO (there would still be 100t left). when the starship enters the LEO it will be once refueled with another tanker which will give it dV 1.4km/s then both can meet in a higher orbit and the starship can be refueled to the full?

Instead of refuelling and docking, you have two refuellings and don't need a special tanker that can dock with its nose  ::)

Problem is (and this is all discussed in the thread), refueling in a higher orbit means multiple trips through the Van Allen belt. It also requires a greater number of rendezvous/docking events (ie more risk).

I'm also drawn to Pusher Tanker for the mission possibilities of an ultra-high delta-v "Starkicker." Starkicker only gets one chance for its escape burn, and Oberth says you want your engine firing to be as short duration as possible. So there's no time to use re-fueling during the delta-v slam, and another approach is needed if you need/want more delta-v than Starkicker alone can deliver.

https://twitter.com/elonmusk/status/1111760133132947458

We don't know how long it will take to refuel, and since both the starship and the tanker can enter a higher orbit at the same time, Van Allen belts may not be such a problem if refueling will be fast.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/16/2020 07:17 pm
... don't need a special tanker that can dock with its nose  ::)

And as to the eye-rolling about docking with the nose, in this thread we seem to be all treating rockets like LEGO (lengthening, stacking two SH together, swapping SH interstage for a nose cone), so surely swapping SS's nose for SH's interstage adapter is fair game. 8)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/16/2020 07:22 pm
[snip]

We don't know how long it will take to refuel, and since both the starship and the tanker can enter a higher orbit at the same time, Van Allen belts may not be such a problem if refueling will be fast.

Both the apogee-raise burn to HEEO and the departure burn are optimally performed at perigee, because Oberth. There are large performance losses (ie you can't reach the Moon) if you do it any other way.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/16/2020 07:45 pm
[snip]

We don't know how long it will take to refuel, and since both the starship and the tanker can enter a higher orbit at the same time, Van Allen belts may not be such a problem if refueling will be fast.

Both the apogee-raise burn to HEEO and the departure burn are optimally performed at perigee, because Oberth. There are large performance losses (ie you can't reach the Moon) if you do it any other way.

Yes, probably impossible to do faster than one full orbit
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/16/2020 08:33 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.

Unnecessary complexity, such manoeuvres can be done with a simple starship tanker, fael097 calculated that the bulkhead shift alone will give a capacity of 2100t, so if it sends such a tanker with 1300t of fuel on the LEO will arrive with 100t, then it will be 20x refuelled to full. consumption of 700t gives dV 1.4km/s and in the tanker is 1200t to refuel the starship and 200t to return to the LEO (there would still be 100t left). when the starship enters the LEO it will be once refueled with another tanker which will give it dV 1.4km/s then both can meet in a higher orbit and the starship can be refueled to the full.

Instead of refuelling and docking, you have two refuellings and don't need a special tanker that can dock with its nose  ::)

Twark covered a lot of this, but:

1) Refueling in HEEO incurs an extra two passes through the VA belts, which is a big chunk of total radiation exposure, given average solar activity.  That won't quite double the total exposure for a lunar missions, but it'll come pretty close.

2) Whether this turns out to be "unnecessary complexity" depends in large part how much risk you incur with a docking/refueling operation.  P/T minimizes that.

An option to P/T that uses ordinary tankers (i.e., non-payload-carrying Starships) is to do the refueling rendezvous post-TLI.  That way, you don't incur the extra pair of VA belt transits.  The tanker then does a free return and reenters at TEI speeds.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/16/2020 08:54 pm
...

There has also been talk about 250-tonne-thrust version of Raptor. The optimal nozzle size for that would be slightly bigger than the nozzle size of current raptor, but they can make it have the same nozzle and lose ~1% of isp, and fit the same number of those engines under the craft. And then they can increase the thrust from 7.2 MN to about 8.8 MN.

I believe the 250 tonne version is simply a change of injectors. No throttling, but higher thrust.

In another thread I proposed a slightly more extreme version of this: cram the Raptor engines together in a honeycomb pattern as tight as the combustion chamber / plumbing will allow, and simply truncate the nozzle to a hexagonal profile (imagine taking a hexagon cookie cutter to the existing nozzle, without otherwise changing its shape). Bonus: the cut off nozzles will have a wicked "sawtooth" pattern when viewed from the side. :)





this means

Darn, they got to him...  :D
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: mikegi on 02/16/2020 09:03 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.
I'm definitely no rocket scientist (understatement of the year!) but ... why not make the Pusher more of a large, reusable service module (LSM) for LEO->Moon->LEO? You attach a simple capsule and lunar lander to the front. The whole thing enters LLO, the lander does its thing, then the LSM returns them to Earth. The capsule detaches and reenters as the LSM passes by Earth. With no people onboard, the LSM can take as long as necessary to get back to a stable LEO (e.g. maybe multiple short passes through the atmosphere, make burns when optimal, etc.).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 02/16/2020 09:24 pm
I on the other hand think an orbital platform could be the best idea. What happens if one of the four launches required to refuel fails? Or weather suddenly makes it impossible to launch. You could have to scrub the mission or have a starship and crew hang out for days/weeks/months waiting for fueling. Having a platform that is constantly stocked up means you can launch knowing the required fuel is there and get the mission under way.

You use one of the tankers as a pseudo-depot, you launch the mission ship itself last. If weather delays a fuel launch (or it fails), your crew are still on the ground. You only launch once the tanker-cum-depot has enough fuel for the mission. Then the tanker-depot lands and goes back into ordinary service.

That latter (ordinary service) is also why (IMO) the tankers wouldn't be modified from the standard cargo vehicles for a long time, even if in practice it meant launching 5 cargo-ships-as-fuel-tankers instead of 4 mass-optimised tankers. The additional cost of that one extra launch is not worth building a whole extra type of Starship that has such limited use.

[Aside: The same reasoning says that the first "passenger" Starships would just be cargo ships with a passenger pods. Different pods for different missions. Lets you do Shuttle-like missions with mixed cargo and personnel. Such as satellite recovery missions, service missions, mixed science missions, etc. And the latter lets researchers use Starships as single-purpose specialised space-stations, by swapping out mission pods.]

And the ease of using a cargo-ship-cum-tanker as a single-mission orbital depot is why you don't need an actual permanent depot for a long time.



Of course, I intend to contradict myself in my very next post:
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 02/16/2020 09:25 pm
If a plane change requires more than a mass ratio of 1- that is, if the mass of fuel needed for a useful "next plane" change of orbit is more than the (already reduced, by the removal of reentry mass) dry mass of the tanker, without fuel, it's easier to launch an entirely new tanker in the new orbit than move the old one.

If anyone's curious, assuming 380s Isp and 350nm LEO, that's a roughly 20 degrees plane change. (Maths below. You might be able to improve that with a Low/High/Low plane change, but I haven't worked that out.)

However, for the Moon or Mars, there's virtually no inclination requirement (Figures below). Hell, you can boost from polar orbit if you want, with barely a hit. So any LEO depot works for BEO missions. LEO-to-GEO is more restrictive, but low-inclination-LEO to equatorial-GEO hides most sins. So a general LEO depot for equatorial boost is likely to see a reasonable amount of use, IMO.

If there's enough traffic beyond GEO, a large general-purpose depot definitely make sense. And if there's enough traffic to a specific inclination of GEO (such as equatorial), then a depot for that makes sense.



Rocket equation: delta_V = Isp * 9.8 * ln(wet_mass/dry_mass)
Target is fuel_mass = dry_mass, so 2n/n, so 2.
dV = 380* 9.8 * ln(2) = ~2.5km/s.

Maximum plane change we can get from a 2.5km/s burn in a 7.5km/s orbit:
delta_V = 2*sin(phi/2)*Vc
:: sin(phi/2) = 2500 / (2*7500)
:: phi = ~19 degrees



For Mars/Moon/Eq_GEO, the effective plane-change is reduced by the distance of the target. Starting at, say 27 degree LEO, Equatorial GEO is around a 4.5 degree effective plane change. The Moon is a sub-half degree plane change. And Mars is so close to zero as to not be worth mentioning. (Mars' solar inclination relative to Earth's is vastly greater, and nothing done in Earth orbit changes that.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TrevorMonty on 02/16/2020 09:28 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.

Unnecessary complexity, such manoeuvres can be done with a simple starship tanker, fael097 calculated that the bulkhead shift alone will give a capacity of 2100t, so if it sends such a tanker with 1300t of fuel on the LEO will arrive with 100t, then it will be 20x refuelled to full. consumption of 700t gives dV 1.4km/s and in the tanker is 1200t to refuel the starship and 200t to return to the LEO (there would still be 100t left). when the starship enters the LEO it will be once refueled with another tanker which will give it dV 1.4km/s then both can meet in a higher orbit and the starship can be refueled to the full.

Instead of refuelling and docking, you have two refuellings and don't need a special tanker that can dock with its nose  ::)

Twark covered a lot of this, but:

1) Refueling in HEEO incurs an extra two passes through the VA belts, which is a big chunk of total radiation exposure, given average solar activity.  That won't quite double the total exposure for a lunar missions, but it'll come pretty close.

2) Whether this turns out to be "unnecessary complexity" depends in large part how much risk you incur with a docking/refueling operation.  P/T minimizes that.

An option to P/T that uses ordinary tankers (i.e., non-payload-carrying Starships) is to do the refueling rendezvous post-TLI.  That way, you don't incur the extra pair of VA belt transits.  The tanker then does a free return and reenters at TEI speeds.
Does the SS need to pass through VA belt twice?. If it is refuelled close earth before orbit hits belt, it can then do its earth departure burn, only passing through belt once.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 02/16/2020 09:43 pm
We shouldn't be too hysterical about the Van Allen belts. Apollo 14 (which came through near the thickest parts of both VA belts, both out and back) gave the crew a total dose of about 10 milligreys. (A CT scan is about 8mgr.)

The particles also aren't overly penetrating, compared to cosmic radiation, and secondary (shotgun) radiation from the skin isn't an issue. So combining the skin of Starship with a radiation shelter made from your water storage should protect the passengers for a good month of exposure, if they wanted to.

The VA belts are extremely dangerous for an unprotected human. But an "unprotected human" in space died before the airlock finished depressurising.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/16/2020 11:50 pm
In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.

Unnecessary complexity, such manoeuvres can be done with a simple starship tanker, fael097 calculated that the bulkhead shift alone will give a capacity of 2100t, so if it sends such a tanker with 1300t of fuel on the LEO will arrive with 100t, then it will be 20x refuelled to full. consumption of 700t gives dV 1.4km/s and in the tanker is 1200t to refuel the starship and 200t to return to the LEO (there would still be 100t left). when the starship enters the LEO it will be once refueled with another tanker which will give it dV 1.4km/s then both can meet in a higher orbit and the starship can be refueled to the full.

Instead of refuelling and docking, you have two refuellings and don't need a special tanker that can dock with its nose  ::)

Twark covered a lot of this, but:

1) Refueling in HEEO incurs an extra two passes through the VA belts, which is a big chunk of total radiation exposure, given average solar activity.  That won't quite double the total exposure for a lunar missions, but it'll come pretty close.

2) Whether this turns out to be "unnecessary complexity" depends in large part how much risk you incur with a docking/refueling operation.  P/T minimizes that.

An option to P/T that uses ordinary tankers (i.e., non-payload-carrying Starships) is to do the refueling rendezvous post-TLI.  That way, you don't incur the extra pair of VA belt transits.  The tanker then does a free return and reenters at TEI speeds.
Does the SS need to pass through VA belt twice?. If it is refuelled close earth before orbit hits belt, it can then do its earth departure burn, only passing through belt once.

Sure, but that wouldn't be "refueling in HEEO," it would be refueling in LEO. And without HEEO refueling, Starship doesn't have the performance go to the Moon and back.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: joek on 02/17/2020 12:24 am
...
Musk has other ideas (https://arstechnica.com/science/2019/08/rocket-scientist-says-that-boeing-squelched-work-on-propellant-depots/)
...

Right.  That article (https://arstechnica.com/science/2019/08/rocket-scientist-says-that-boeing-squelched-work-on-propellant-depots/) is what I was referring to--and as I noted in a previous post in another thread at that time)--Musk wants to stay out of that fray.

Clearly Musk sees on-orbit refueling as a requirement (obvious from Starship conops).  However, Musk likely wants to stay out of the kerfuffle which would potentially put him at odds with certain major NASA programs-stakeholders (as articulated in Eric Berger's article):
Quote
But we were banned from even saying the 'd' word out loud.
...
Sowers' suggestion that "depots" should not be uttered in public is consistent with observations at the time that a US Senator from Alabama, Richard Shelby, had told NASA to stop talking about propellant depots. The NASA spaceflight center that manages the SLS rocket's development, Marshall Space Flight Center, is based in Alabama.

The operative question is: How to best achieve on-orbit refueling?  Musk's opinion is clearly "tankers"--at least in the near-term and for the missions SpaceX is considering.  Not to mention avoids the 'd' word.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: MATTBLAK on 02/17/2020 12:43 am
We shouldn't be too hysterical about the Van Allen belts. Apollo 14 (which came through near the thickest parts of both VA belts, both out and back) gave the crew a total dose of about 10 milligreys. (A CT scan is about 8mgr.)

The particles also aren't overly penetrating, compared to cosmic radiation, and secondary (shotgun) radiation from the skin isn't an issue. So combining the skin of Starship with a radiation shelter made from your water storage should protect the passengers for a good month of exposure, if they wanted to.

The VA belts are extremely dangerous for an unprotected human. But an "unprotected human" in space died before the airlock finished depressurising.
Didn't the Apollo 14 crew receive the highest overall dose of radiation of any Apollo crew? I'm sure I read that somewhere. EDIT: found table of dosage (see attachment) from bottom of this page - https://history.nasa.gov/SP-368/s2ch3.htm
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 02/17/2020 03:55 am
We shouldn't be too hysterical about the Van Allen belts. Apollo 14 (which came through near the thickest parts of both VA belts, both out and back) gave the crew a total dose of about 10 milligreys. (A CT scan is about 8mgr.)
Didn't the Apollo 14 crew receive the highest overall dose of radiation of any Apollo crew?

That's why I picked them as my example. (1rad is 10milligrays. Symptoms start in the low single digit grays, or low hundreds of rads.) They received the highest dose because of their trajectory, passing through the middle of the VA belts. The other missions were able to avoid most of the inner belt.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/17/2020 09:43 am
We shouldn't be too hysterical about the Van Allen belts. Apollo 14 (which came through near the thickest parts of both VA belts, both out and back) gave the crew a total dose of about 10 milligreys. (A CT scan is about 8mgr.)

The particles also aren't overly penetrating, compared to cosmic radiation, and secondary (shotgun) radiation from the skin isn't an issue. So combining the skin of Starship with a radiation shelter made from your water storage should protect the passengers for a good month of exposure, if they wanted to.

The VA belts are extremely dangerous for an unprotected human. But an "unprotected human" in space died before the airlock finished depressurising.

There's a lot of difference between a TLI trajectory, which transits both the inner and outer belts at something fairly close to maximum velocity, and, say, an LEO+2500 HEEO, which has a period of 11.5 hours and would likely spend a third of that deep in the outer belt.  I haven't done the time-of-flight computations, but my guess is that we're talking about several times the length of exposure that you get on a straight TLI burn.

Beyond that, what are the contingencies for a difficult rendezvous and docking?  You have close to 10 hours to achieve the R&D and transfer prop.  That should be plenty of time, but what if something goes wrong?  Do you abort the mission if you can't do a one-pass refueling?  Or do you expose the crew to still a third pair of passes through the VA belts?

Using either a TLI-based R&D for refueling or the pusher/tanker eliminates all of that risk, and incurs very little additional risk of a failed refueling, because both the tanker and the payload SS can always do a free-return around the Moon if they can't hook up.  Beyond that, the P/T guarantees a single R&D operation in LEO, where it's easiest to come fetch the crew if something goes wrong.

For cargo missions, none of this likely matters.  For crewed missions, it's a different story.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/17/2020 09:47 am
However, for the Moon or Mars, there's virtually no inclination requirement (Figures below). Hell, you can boost from polar orbit if you want, with barely a hit. So any LEO depot works for BEO missions. LEO-to-GEO is more restrictive, but low-inclination-LEO to equatorial-GEO hides most sins. So a general LEO depot for equatorial boost is likely to see a reasonable amount of use, IMO.

Two reasons you might want a high-inclination departure for the Moon:

1) It can minimize LLO insertion delta-v for lunar polar missions, which a likely to dominate the manifest.

2) High-inclination insertions avoid the thickest part of the VA belts, reducing radiation exposure.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/17/2020 04:00 pm
If 250t thrust raptor will exists, SH will be able to raise the tanker with 2100t of fuel, which should provide 160t of fuel on LEO instead of 100t, from my calculations even bigger tankers works very well and are superior to multiple launches of smaller tankers. If SH has 250t engines, delta V is smaller and final angle is more vertical which makes landing easier, "only" problem is that this big tanker due to its weight and more vertical start has bigger gravitational losses so it should probably have more engines

I assume a dry weight of 120t, I don't think the tanker will be lighter, and I'm even afraid it may be heavier because it requires more 4mm rings.

(hkultala was right, you don't need a bigger booster if you equip SH with 250t engines)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/17/2020 04:49 pm
You use one of the tankers as a pseudo-depot, you launch the mission ship itself last. If weather delays a fuel launch (or it fails), your crew are still on the ground. You only launch once the tanker-cum-depot has enough fuel for the mission. Then the tanker-depot lands and goes back into ordinary service.

I've been calling this the "aggregation tanker".  It's no different from the vanilla-flavored tankers (unless you drink the Pusher/Tanker Kool-Aid); it's just the one that receives the prop from the other tankers to minimize the number of rendezvous/dock/refuel operations that have to be performed against the live payload.  You potentially need two aggregation tankers, one for LEO top-off and one for either HEEO or post-TLI top-off.  (Of course, if you use P/T, it is the agg tanker, and you only need one, and one rendezvous/dock/refuel.)

Quote
[Aside: The same reasoning says that the first "passenger" Starships would just be cargo ships with a passenger pods. Different pods for different missions. Lets you do Shuttle-like missions with mixed cargo and personnel. Such as satellite recovery missions, service missions, mixed science missions, etc. And the latter lets researchers use Starships as single-purpose specialised space-stations, by swapping out mission pods.]

And this because of its likely shape, needs to be called "StarKist".

I think this is no-brainer absolutely the way to engineer a Starship-based crew system, but it's surprisingly controversial.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: hkultala on 02/19/2020 03:55 am
[snip]

We don't know how long it will take to refuel, and since both the starship and the tanker can enter a higher orbit at the same time, Van Allen belts may not be such a problem if refueling will be fast.

Both the apogee-raise burn to HEEO and the departure burn are optimally performed at perigee, because Oberth. There are large performance losses (ie you can't reach the Moon) if you do it any other way.

There are much larger performance losses if you don't refuel at highly elliptic trajectory.

Argument that something should not be used unless it's perfect (when it's still much better than the alternatives) is stupid.

And there is no problem of raising the apogee at pegiree, then having the whole ~20-hour orbit time to transfer the fuel, then burning for the moon on the next pegiree.

Needs to go through the Van Allen belts three times, not a big deal.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Keldor on 02/19/2020 04:23 am
[snip]

We don't know how long it will take to refuel, and since both the starship and the tanker can enter a higher orbit at the same time, Van Allen belts may not be such a problem if refueling will be fast.

Both the apogee-raise burn to HEEO and the departure burn are optimally performed at perigee, because Oberth. There are large performance losses (ie you can't reach the Moon) if you do it any other way.

There are much larger performance losses if you don't refuel at highly elliptic trajectory.

Argument that something should not be used unless it's perfect (when it's still much better than the alternatives) is stupid.

And there is no problem of raising the apogee at pegiree, then having the whole ~20-hour orbit time to transfer the fuel, then burning for the moon on the next pegiree.

Needs to go through the Van Allen belts three times, not a big deal.

You could potentially use a high inclination orbit and avoid the Van Allen belts almost completely.  This should only cost DV during the initial launch to orbit, before refueling, since once you're in orbit, the rotation of the Earth doesn't effect you.  You'd have a slight loss of efficiency landing on the Moon, but it's rotation speed is so slow that it won't make much difference.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 02/19/2020 05:17 am
Or just use several micro-sats, with charged tethers, to drain the belts, and stop having to worry about it...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/19/2020 08:20 am
[snip]

We don't know how long it will take to refuel, and since both the starship and the tanker can enter a higher orbit at the same time, Van Allen belts may not be such a problem if refueling will be fast.

Both the apogee-raise burn to HEEO and the departure burn are optimally performed at perigee, because Oberth. There are large performance losses (ie you can't reach the Moon) if you do it any other way.

There are much larger performance losses if you don't refuel at highly elliptic trajectory.

Argument that something should not be used unless it's perfect (when it's still much better than the alternatives) is stupid.

And there is no problem of raising the apogee at pegiree, then having the whole ~20-hour orbit time to transfer the fuel, then burning for the moon on the next pegiree.

Needs to go through the Van Allen belts three times, not a big deal.

I'm too lazy to do the math here, so correct me if I'm wrong:  You do better refueling at the highest energy possible, as long as you use the energy optimally.  For any EO, that means burning at the next perigee.  But TLI is the highest energy you'll need to get to the cis-lunar orbits, and it doesn't even need to do a second burn to complete the maneuver.  So refueling after TLI should be somewhat more efficient than refueling at HEEO.  (Technically, TLI is an HEEO...)

20 hours is likely enough time to do the rendezvous, dock, prop transfer, and undock.  But any glitch in the system leaves the crew in a dicey situation.  For example, if you had some problem that prevented you from doing the perigee burn with the main engines, you probably need to abort by chopping the perigee enough to reenter, using thrusters (it's not much delta-v).  Doing an indefinite number of passes through VAB would be bad.

You're in no worse shape refueling in TLI, and arguably in better shape because a free return will bring you back to reentry.

Of course, none of this is an issue with the pusher/tanker, since the P/T does one rendezvous in LEO and stages off, leaving the SS with enough prop to do the rest of TLI and everything needed to land and get home.  It's absolutely not as efficient as a tanker rendezvous in TLI, but it's fewer docking operations.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/19/2020 10:31 am
[snip]

We don't know how long it will take to refuel, and since both the starship and the tanker can enter a higher orbit at the same time, Van Allen belts may not be such a problem if refueling will be fast.

Both the apogee-raise burn to HEEO and the departure burn are optimally performed at perigee, because Oberth. There are large performance losses (ie you can't reach the Moon) if you do it any other way.

There are much larger performance losses if you don't refuel at highly elliptic trajectory.

So? No one is arguing for that.

Argument that something should not be used unless it's perfect (when it's still much better than the alternatives) is stupid.

No one is arguing for that either. Also, rude.

And there is no problem of raising the apogee at pegiree, then having the whole ~20-hour orbit time to transfer the fuel, then burning for the moon on the next pegiree.

Needs to go through the Van Allen belts three times...

That is exactly the problem.

...not a big deal.

That's a value judgement. TheRadicalModerate and I are just pointing out the trade-off.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/19/2020 10:38 am
(Technically, TLI is an HEEO...)

Technically it's hyperbolic, not elliptical. ;) But yes, your point about refilling post-TMI is a good one.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 02/19/2020 11:37 am
(Technically, TLI is an HEEO...)

Technically it's hyperbolic, not elliptical. ;) But yes, your point about refilling post-TMI is a good one.
it's only hyperbolic if the moon cuts off the orbit.

I've played enough kerbal to know you can have an orbit that wont inject to lunar SoI until the next time it goes around.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/19/2020 03:30 pm
(Technically, TLI is an HEEO...)

Technically it's hyperbolic, not elliptical. ;) But yes, your point about refilling post-TMI is a good one.
it's only hyperbolic if the moon cuts off the orbit.

I've played enough kerbal to know you can have an orbit that wont inject to lunar SoI until the next time it goes around.

Right, it's hyperbolic if it's moving at greater than Earth escape velocity (eg Apollo). An elliptical trajectory increases the trip time, but it doesn't really save that much propellant.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 02/19/2020 03:56 pm
Or just use several micro-sats, with charged tethers, to drain the belts, and stop having to worry about it...

"Don't you know that the Van Allen belts protect the Earth from solar storms? You corporate-lackey so-called 'scientists' only know how to destroy things." (Actual comments I've seen dominating the responses to articles about the VA belts on pop.sci sites.)

Isn't there also a way to drain at least the inner belt by injecting ELF from polar ground stations? (I might be misremembering, but I recall being amused by ELFs at the North Pole.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/19/2020 05:55 pm
(Technically, TLI is an HEEO...)

Technically it's hyperbolic, not elliptical. ;) But yes, your point about refilling post-TMI is a good one.

TLI C3 is -2 km²/s², more or less.  Model it as about 200 x 380,000 km and you'll be pretty close.

For TMI, things are a lot easier, because there's no hope of taking enough prop with you to return (even though the hope of doing Sabatier ISRU with some martian water and CO2 is much higher).  You can put an SS into a shorter-than-Hohmann TMI, direct from LEO, no problem.  All this HEEO stuff is a lunar idiosyncrasy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wannamoonbase on 02/19/2020 06:06 pm
….all this HEEO stuff is a lunar idiosyncrasy.

The HEEO and Lunar Gateway non-sense seems like NASA hedging itself.  That if lunar surface funds never arrived they would have a destination to use SLS and Orion.

If you're actually going to the Lunar surface then go to the surface.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/19/2020 08:16 pm
….all this HEEO stuff is a lunar idiosyncrasy.

The HEEO and Lunar Gateway non-sense seems like NASA hedging itself.  That if lunar surface funds never arrived they would have a destination to use SLS and Orion.

If you're actually going to the Lunar surface then go to the surface.

If nobody's ever going to the lunar surface, then this really doesn't matter.  Personally, I'm guessing that 90% of the revenue that SpaceX will earn from BEO missions in this decade will come from lunar surface and cis-lunar missions.

Note that HEEO or TLI-based refueling is a solution to a Starship-specific problem, and has nothing to do with the Gateway.  The Gateway, on the other hand, is a sorta-kinda solution to a set of SLS/Orion/HLS sorta-kinda problems, and has nothing to do with HEEO.  (Gateway is basically a risk reduction measure, to ease assembly of the HLS components.  If Boeing has its way with the "whole lander/ascender on a separate Block 1B" proposal, the Gateway will go away.)

FWIW (and O/T), I'm a big proponent of replacing the SLS/Orion method of getting to the Gateway and back with using Starship as an LEO-to-Gateway-back-to-LEO shuttle.  This:

1) Doesn't require launching a crew with SH/SS.  They can board from a separately-launched (and crew-qualified) F9/D2.

2) Doesn't require landing SS on the lunar surface.  They can use the HLS components to do that.

3) Doesn't require reentering and landing a crew on Earth with SS.  SS can propulsively return to LEO, then re-board the D2 for reentry.

4) Requires only refueling in LEO, and requires zero refuelings with the crew on board.  If you've got Isp=375s, then you can get to NRHO and back to (propulsive) LEO with 1190 t of prop.

This gets rid of the expensive SLS and Orion components ASAP, because crew-qualifying a Starship for only in-space use is a lot easier than crew-qualifying for launch, reentry, or landing on the Moon.  So you start with this, using the Gateway and HLS to land on the Moon, then qualify the lunar landing segment, then the earth reentry, and finally the launch.

Of course, the real problem is that getting rid of the expensive SLS and Orion is exactly what Congress wants to avoid...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TrevorMonty on 02/19/2020 11:37 pm
….all this HEEO stuff is a lunar idiosyncrasy.

The HEEO and Lunar Gateway non-sense seems like NASA hedging itself.  That if lunar surface funds never arrived they would have a destination to use SLS and Orion.

If you're actually going to the Lunar surface then go to the surface.

If nobody's ever going to the lunar surface, then this really doesn't matter.  Personally, I'm guessing that 90% of the revenue that SpaceX will earn from BEO missions in this decade will come from lunar surface and cis-lunar missions.

Note that HEEO or TLI-based refueling is a solution to a Starship-specific problem, and has nothing to do with the Gateway.  The Gateway, on the other hand, is a sorta-kinda solution to a set of SLS/Orion/HLS sorta-kinda problems, and has nothing to do with HEEO.  (Gateway is basically a risk reduction measure, to ease assembly of the HLS components.  If Boeing has its way with the "whole lander/ascender on a separate Block 1B" proposal, the Gateway will go away.)

FWIW (and O/T), I'm a big proponent of replacing the SLS/Orion method of getting to the Gateway and back with using Starship as an LEO-to-Gateway-back-to-LEO shuttle.  This:

1) Doesn't require launching a crew with SH/SS.  They can board from a separately-launched (and crew-qualified) F9/D2.

2) Doesn't require landing SS on the lunar surface.  They can use the HLS components to do that.

3) Doesn't require reentering and landing a crew on Earth with SS.  SS can propulsively return to LEO, then re-board the D2 for reentry.

4) Requires only refueling in LEO, and requires zero refuelings with the crew on board.  If you've got Isp=375s, then you can get to NRHO and back to (propulsive) LEO with 1190 t of prop.

This gets rid of the expensive SLS and Orion components ASAP, because crew-qualifying a Starship for only in-space use is a lot easier than crew-qualifying for launch, reentry, or landing on the Moon.  So you start with this, using the Gateway and HLS to land on the Moon, then qualify the lunar landing segment, then the earth reentry, and finally the launch.

Of course, the real problem is that getting rid of the expensive SLS and Orion is exactly what Congress wants to avoid...
While I like idea of LEO-Gateway-LEO shuttle, SS is not best vehicle for this when only transferring few tons. Most of tanker missions fuel is used to move SS dry mass back and fore. A small purpose built shuttle could do round trip using 150 fuel, lot less if hydrolox.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 02/20/2020 02:43 am
Or just use several micro-sats, with charged tethers, to drain the belts, and stop having to worry about it...

There is the classic Tethers Unlimited paper showing they can drain the belts with 10 small sats with charged tethers in a few weeks or so. But, who pays for it, and why? Because everyone benefits from it whether they paid in or not, especially GEO birds that slowly spiral orbit up to GEO with electric thrusters (since they avoid roasting their solar panels). One of those tragedy of the commons moments, best fixed by a multinational Space Guard service charged with keeping the spacelanes clear (space junk included). Though I could easily see SpaceX doing just because it was incidental to their ops.

https://web.archive.org/web/20190213154045/http://www.tethers.com/papers/ES_Remediation_IEEE_Paper.pdf (https://web.archive.org/web/20190213154045/http://www.tethers.com/papers/ES_Remediation_IEEE_Paper.pdf)

For a fuel depot service though, maybe cheat with equatorial LEO as the baseline as that's the lowest exposure? Though that doesn't help for other orbits...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/20/2020 05:40 am
….all this HEEO stuff is a lunar idiosyncrasy.

The HEEO and Lunar Gateway non-sense seems like NASA hedging itself.  That if lunar surface funds never arrived they would have a destination to use SLS and Orion.

If you're actually going to the Lunar surface then go to the surface.

If nobody's ever going to the lunar surface, then this really doesn't matter.  Personally, I'm guessing that 90% of the revenue that SpaceX will earn from BEO missions in this decade will come from lunar surface and cis-lunar missions.

Note that HEEO or TLI-based refueling is a solution to a Starship-specific problem, and has nothing to do with the Gateway.  The Gateway, on the other hand, is a sorta-kinda solution to a set of SLS/Orion/HLS sorta-kinda problems, and has nothing to do with HEEO.  (Gateway is basically a risk reduction measure, to ease assembly of the HLS components.  If Boeing has its way with the "whole lander/ascender on a separate Block 1B" proposal, the Gateway will go away.)

FWIW (and O/T), I'm a big proponent of replacing the SLS/Orion method of getting to the Gateway and back with using Starship as an LEO-to-Gateway-back-to-LEO shuttle.  This:

1) Doesn't require launching a crew with SH/SS.  They can board from a separately-launched (and crew-qualified) F9/D2.

2) Doesn't require landing SS on the lunar surface.  They can use the HLS components to do that.

3) Doesn't require reentering and landing a crew on Earth with SS.  SS can propulsively return to LEO, then re-board the D2 for reentry.

4) Requires only refueling in LEO, and requires zero refuelings with the crew on board.  If you've got Isp=375s, then you can get to NRHO and back to (propulsive) LEO with 1190 t of prop.

This gets rid of the expensive SLS and Orion components ASAP, because crew-qualifying a Starship for only in-space use is a lot easier than crew-qualifying for launch, reentry, or landing on the Moon.  So you start with this, using the Gateway and HLS to land on the Moon, then qualify the lunar landing segment, then the earth reentry, and finally the launch.

Of course, the real problem is that getting rid of the expensive SLS and Orion is exactly what Congress wants to avoid...
While I like idea of LEO-Gateway-LEO shuttle, SS is not best vehicle for this when only transferring few tons. Most of tanker missions fuel is used to move SS dry mass back and fore. A small purpose built shuttle could do round trip using 150 fuel, lot less if hydrolox.

If all Starship was ever going to do was an LEO-GW-LEO shuttle, I'd agree with you.  But the shuttle is merely the first capability, and the easiest one to get certified.  After that, SpaceX can bite off Earth EDL, followed by lunar landing/ascent, and finally launch from Earth with crew.  The order of these is up for debate, but the shuttle to and from NRHO is by far the lowest-hanging fruit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 02/20/2020 09:39 am
While I like idea of LEO-Gateway-LEO shuttle, SS is not best vehicle for this when only transferring few tons. Most of tanker missions fuel is used to move SS dry mass back and fore. A small purpose built shuttle could do round trip using 150 fuel, lot less if hydrolox.

Question is then, which such purpose-built shuttle?  When will that be available?  And will it be significantly cheaper than a twice-refuelled Starship (even if I don't believe the $2M/launch figure Elon envisions will be reached within that timeframe)?

My understanding is that ACES is not currently being developed.  Will Blue Moon or some auxillary part of New Glenn be able to shuttle back and forth between LEO and lunar/cis-lunar?  Any other crafts in development?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TrevorMonty on 02/20/2020 12:24 pm
While I like idea of LEO-Gateway-LEO shuttle, SS is not best vehicle for this when only transferring few tons. Most of tanker missions fuel is used to move SS dry mass back and fore. A small purpose built shuttle could do round trip using 150 fuel, lot less if hydrolox.

Question is then, which such purpose-built shuttle?  When will that be available?  And will it be significantly cheaper than a twice-refuelled Starship (even if I don't believe the $2M/launch figure Elon envisions will be reached within that timeframe)?

My understanding is that ACES is not currently being developed.  Will Blue Moon or some auxillary part of New Glenn be able to shuttle back and forth between LEO and lunar/cis-lunar?  Any other crafts in development?
Blue are working OTV which is what BE7 is for besides lander. My guess is lander and OTV will be similar.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 02/20/2020 12:48 pm
Or just use several micro-sats, with charged tethers, to drain the belts, and stop having to worry about it...

There is the classic Tethers Unlimited paper showing they can drain the belts with 10 small sats with charged tethers in a few weeks or so. But, who pays for it, and why? Because everyone benefits from it whether they paid in or not, especially GEO birds that slowly spiral orbit up to GEO with electric thrusters (since they avoid roasting their solar panels). One of those tragedy of the commons moments, best fixed by a multinational Space Guard service charged with keeping the spacelanes clear (space junk included). Though I could easily see SpaceX doing just because it was incidental to their ops.

https://web.archive.org/web/20190213154045/http://www.tethers.com/papers/ES_Remediation_IEEE_Paper.pdf (https://web.archive.org/web/20190213154045/http://www.tethers.com/papers/ES_Remediation_IEEE_Paper.pdf)

For a fuel depot service though, maybe cheat with equatorial LEO as the baseline as that's the lowest exposure? Though that doesn't help for other orbits...
This particular TotC has a Gordian knot solution- One person does it for awhile, everyone gets used to benifiting, then the person announces they will stop doing it unless they receive donations to continue.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: dcengineering on 02/21/2020 07:56 pm

If all Starship was ever going to do was an LEO-GW-LEO shuttle, I'd agree with you.  But the shuttle is merely the first capability, and the easiest one to get certified.  After that, SpaceX can bite off Earth EDL, followed by lunar landing/ascent, and finally launch from Earth with crew.  The order of these is up for debate, but the shuttle to and from NRHO is by far the lowest-hanging fruit.

Might be a bit awkward when the "shuttle" has more living space than the Gateway does  ;D

I think you hit the nail on the head with your previous post. After reading through the whole thread, it seems a lot of the issues stem from the complexities and risk associated with these maneuvers while a crew is on board, when the likelihood of NASA crew certifying Starship anytime soon is almost zero. D2 has been in development for 6 years now and is still not certified even though Falcon 9 is a relatively simple rocket if you ignore the whole landing part (which is immaterial to its crew rating anyways).

What sort of insane timeline are we even looking at to crew certify Starship? While I am somewhat confident SpaceX can beat NASA's 2024 Moon landing goal, it would be little more than a capabilities demonstration when crew aren't allowed on board during this timeframe anyways.

In the short to medium term, we should probably assume that manned operations involving SS is going to require F9/D2 to get astronauts back and forth from orbit. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 02/21/2020 08:08 pm
NASA can only certify for their missions, besides there is no such requirement as NASA certification for human flights
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 02/22/2020 01:35 am
NASA can only certify for their missions, besides there is no such requirement as NASA certification for human flights

FAA does.

Currently they are allowing suborbital (and presumably orbital) human rocket flight to classify vehicles as "experimental", passengers as "test crew", and exempt the operators from strict requirements of passenger aircraft. However, that's essentially a courtesy. It would take only a slightly more hostile Administration or a single anti-SpaceX senior senator to make the FAA adopt stricter airline-like rules.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: joek on 02/22/2020 01:58 am
NASA can only certify for their missions, besides there is no such requirement as NASA certification for human flights

Of course there is.  That is what CCP "certification" is all about (of which there are reams of documents).  In any case, this has nothing to do with this thread; take it elsewhere.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: joek on 02/22/2020 02:04 am
Currently they are allowing suborbital (and presumably orbital) human rocket flight to classify vehicles as "experimental", passengers as "test crew", and exempt the operators from strict requirements of passenger aircraft. However, that's essentially a courtesy. It would take only a slightly more hostile Administration or a single anti-SpaceX senior senator to make the FAA adopt stricter airline-like rules.

They are referred to as "participants" or "non-crew".  And yes, there have been pushes to institute FAA certification rules, which have been a work-in-progress for years, and resisted by the industry as a whole; SpaceX is not the only one with a stake in the game (and arguably less than some others).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/22/2020 05:05 am
NASA can only certify for their missions, besides there is no such requirement as NASA certification for human flights

I'm much more interested in how Starship fits into--and eventually supplants--the Artemis architecture.  For that to happen, there is absolutely a requirement that NASA crew-rate Starship.

Could SpaceX do an all-private lunar surface mission?  I suppose so.  But it's a massive risk for very little reward.  There's not enough money to be made doing it, and an accident is a huge setback.  I'm betting that they'll be very, very conservative with lunar surface missions, and do them only under the aegis of Artemis.

They can likely get wilder and crazier with CLPS, and I'd expect CLPS missions to prove out a lot of the tech that will be used for crewed landings later on.  But that just gives something to feed to NASA to get the crew certification ball rolling for landings.  It's a lot easier to get the LEO-NRHO-LEO stuff up and running sooner.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: QuantumG on 02/22/2020 05:52 am
Could SpaceX do an all-private lunar surface mission?  I suppose so.  But it's a massive risk for very little reward.  There's not enough money to be made doing it, and an accident is a huge setback.  I'm betting that they'll be very, very conservative with lunar surface missions, and do them only under the aegis of Artemis.

I believe they're very interested in Artemis and I'd L O V E to see what they've submitted already.

I expect they'll do an automated lunar landing on their own dime.

Then I think they'll do a flight around the Moon with private astronauts (with an eccentric Japanese flight assignment officer!)

Around then NASA will look pretty silly if they don't put government astronauts on the next flight. NASA doesn't like to be put in these situations, so they've probably already secured a seat on the free-flyer.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ThomasGadd on 02/22/2020 04:25 pm

While SpaceX is developing crewed Starship for NASA flights they can use Dragon as a transfer vehicle for non-NASA flights people could choose...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/23/2020 07:44 pm

While SpaceX is developing crewed Starship for NASA flights they can use Dragon as a transfer vehicle for non-NASA flights people could choose...

Are you suggesting doing LEO to NRHO in a D2?  Not gonna happen; there's not enough delta-v and D2 only has consumables for 7 days.  But using an F9/D2 to get to LEO so that a crew can transfer to/from a Starship, which takes it to NRHO, is pretty easy, and the crew system for Starship isn't very challenging.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: GregTheGrumpy on 02/23/2020 08:44 pm

While SpaceX is developing crewed Starship for NASA flights they can use Dragon as a transfer vehicle for non-NASA flights people could choose...

NASA choosing to go this route **when** SpaceX has already sent and returned non-NASA individuals will be seen as churlish as the Russians seemed with DM-1 visited the ISS.  I just don't see that happening.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: docmordrid on 02/24/2020 12:14 pm

While SpaceX is developing crewed Starship for NASA flights they can use Dragon as a transfer vehicle for non-NASA flights people could choose...

Are you suggesting doing LEO to NRHO in a D2?  Not gonna happen; there's not enough delta-v and D2 only has consumables for 7 days.  But using an F9/D2 to get to LEO so that a crew can transfer to/from a Starship, which takes it to NRHO, is pretty easy, and the crew system for Starship isn't very challenging.

And a lot cheaper than doing it with Starliner or SLS/Orion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacexfanatic on 04/15/2020 01:39 pm
Hi everyone, I did love the idea of Starship refueling another Starship but I was thinking about a simpler way to get extra fuel by mean of attaching external  tanks (cheap, disposable)  to  SS that will rendezvous somewhere ? think it simpler than later concept that still  has to be proven.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: SkyRate on 04/15/2020 02:26 pm
Hi everyone, I did love the idea of Starship refueling another Starship but I was thinking about a simpler way to get extra fuel by mean of attaching external  tanks (cheap, disposable)  to  SS that will rendezvous somewhere ? think it simpler than later concept that still  has to be proven.
Sure, the aft cargo pods could be made into tanks without hurting the ascent aerodynamics, but:

1. It's not that unproven. Progress tanker/cargo craft have auto-docked with Salyut-6, MIR and ISS 100+ times. Not to mention all the air-to-air refueling going on every day.

2. It's not that much simpler. Those tanks would each have to have connectors, valves, plumbing and sensors. The main tanks already have them (for being loaded via the SH). And if you transfer the tanks using a robotic arm, you have to add the weight and complexity of the arm. If not, it's a big, dangerous, time-consuming EVA job. "Best part is no part."

3. It's heavier. This is an upper stage. Every kg of structure is a kg less to orbit. Strap-on boosters are acceptable because they are dropped so early during the launch.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacexfanatic on 04/16/2020 10:01 am
Hi everyone, I did love the idea of Starship refueling another Starship but I was thinking about a simpler way to get extra fuel by mean of attaching external  tanks (cheap, disposable)  to  SS that will rendezvous somewhere ? think it simpler than later concept that still  has to be proven.

Sure, the aft cargo pods could be made into tanks without hurting the ascent aerodynamics, but:

1. It's not that unproven. Progress tanker/cargo craft have auto-docked with Salyut-6, MIR and ISS 100+ times. Not to mention all the air-to-air refueling going on every day.

2. It's not that much simpler. Those tanks would each have to have connectors, valves, plumbing and sensors. The main tanks already have them (for being loaded via the SH). And if you transfer the tanks using a robotic arm, you have to add the weight and complexity of the arm. If not, it's a big, dangerous, time-consuming EVA job. "Best part is no part."

3. It's heavier. This is an upper stage. Every kg of structure is a kg less to orbit. Strap-on boosters are acceptable because they are dropped so early during the launch.

My point is that it's more cost evective to only send the necessery to space which in our case is the fuel, sending the ship and get it back to earth is unessery if you could only send the fuel (and of course lightweight and cheap package) .

For the issue of rendezvous and connecting, I suggest the use of a flexible pipe like those used for air refueling which will use small thruster to move, once attached to the tank they will start suck the fuel or they will retract to connect the tank directly to straship. no plumbing needed just an orifice in the tank.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: RonM on 04/16/2020 12:49 pm
Hi everyone, I did love the idea of Starship refueling another Starship but I was thinking about a simpler way to get extra fuel by mean of attaching external  tanks (cheap, disposable)  to  SS that will rendezvous somewhere ? think it simpler than later concept that still  has to be proven.

Sure, the aft cargo pods could be made into tanks without hurting the ascent aerodynamics, but:

1. It's not that unproven. Progress tanker/cargo craft have auto-docked with Salyut-6, MIR and ISS 100+ times. Not to mention all the air-to-air refueling going on every day.

2. It's not that much simpler. Those tanks would each have to have connectors, valves, plumbing and sensors. The main tanks already have them (for being loaded via the SH). And if you transfer the tanks using a robotic arm, you have to add the weight and complexity of the arm. If not, it's a big, dangerous, time-consuming EVA job. "Best part is no part."

3. It's heavier. This is an upper stage. Every kg of structure is a kg less to orbit. Strap-on boosters are acceptable because they are dropped so early during the launch.

My point is that it's more cost evective to only send the necessery to space which in our case is the fuel, sending the ship and get it back to earth is unessery if you could only send the fuel (and of course lightweight and cheap package) .

For the issue of rendezvous and connecting, I suggest the use of a flexible pipe like those used for air refueling which will use small thruster to move, once attached to the tank they will start suck the fuel or they will retract to connect the tank directly to straship. no plumbing needed just an orifice in the tank.

Lightweight cheap packaging that can survive launch isn't possible for cryogenic propellants. This "package" will be the same construction as the SS propellant tanks, so why not skip the "middleman" and use SS tanks?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacexfanatic on 04/16/2020 01:33 pm
Hi everyone, I did love the idea of Starship refueling another Starship but I was thinking about a simpler way to get extra fuel by mean of attaching external  tanks (cheap, disposable)  to  SS that will rendezvous somewhere ? think it simpler than later concept that still  has to be proven.

Sure, the aft cargo pods could be made into tanks without hurting the ascent aerodynamics, but:

1. It's not that unproven. Progress tanker/cargo craft have auto-docked with Salyut-6, MIR and ISS 100+ times. Not to mention all the air-to-air refueling going on every day.

2. It's not that much simpler. Those tanks would each have to have connectors, valves, plumbing and sensors. The main tanks already have them (for being loaded via the SH). And if you transfer the tanks using a robotic arm, you have to add the weight and complexity of the arm. If not, it's a big, dangerous, time-consuming EVA job. "Best part is no part."

3. It's heavier. This is an upper stage. Every kg of structure is a kg less to orbit. Strap-on boosters are acceptable because they are dropped so early during the launch.

My point is that it's more cost evective to only send the necessery to space which in our case is the fuel, sending the ship and get it back to earth is unessery if you could only send the fuel (and of course lightweight and cheap package) .

For the issue of rendezvous and connecting, I suggest the use of a flexible pipe like those used for air refueling which will use small thruster to move, once attached to the tank they will start suck the fuel or they will retract to connect the tank directly to straship. no plumbing needed just an orifice in the tank.

Lightweight cheap packaging that can survive launch isn't possible for cryogenic propellants. This "package" will be the same construction as the SS propellant tanks, so why not skip the "middleman" and use SS tanks?

If Spacex envision to make reusable spacecrafts with 4mm thin stainless steel, it won't be that hard to imagine a fuel tank that goes to space with the same mensurations and with the same materials.skiping   the "middleman" may save weight and money.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Redclaws on 04/16/2020 01:39 pm
And now you’re discarding that item.  It has engines and avionics on it, doesn’t it?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacexfanatic on 04/16/2020 01:42 pm
And now you’re discarding that item.  It has engines and avionics on it, doesn’t it?

Engines and avionics like those in satellites.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DistantTemple on 04/16/2020 05:27 pm
Two other points. Both questionable.
Those tanks you postulate have to then reenter and either burn up or splash down, as it would not be appropriate to leave them in orbit. So assuming SX does get sixSS's off to Mars, soon (2024 or 20260) that would be 5 x 6 = 30 reentry and spashdown (as stainless will not burn up, and CF is too expensive) events. And after a few more years, 100's of dropped tanks raining down, with practical, safety, environmental, and public comment concerns.

And two they all cost to make, in $, manpower, resources, and factory space.

The refuelling system is envisaged as one suitable for hundreds and then thousands of launches. It is supposed to moving towards airline-like operations, with only fuel as expendable. And driving the cost <$10M etc.... Mass manufacturing tanks, with navigation and propulsion doesn't marry easily with that. (I guess mass of tank+fuel/oxidiser ~150t )

And a second stage is still needed, which needs to be 100% reusable, else the tanks would become a second stage themselves!!!!!

ISTM EM called this right.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 04/17/2020 12:26 am
And now you’re discarding that item.  It has engines and avionics on it, doesn’t it?
Engines and avionics like those in satellites.

I can't picture what you are describing. How do these tanks, loaded with propellant, get into orbit?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 04/17/2020 09:03 am
And now you’re discarding that item.  It has engines and avionics on it, doesn’t it?
Engines and avionics like those in satellites.

I can't picture what you are describing. How do these tanks, loaded with propellant, get into orbit?

Yeah. Super Heavy by itself has about zero payload to LEO, expendable.  So you need a Starship second stage to get to orbit with payload. 

Sure, you could just have the propellant for transfer sitting in a disposable tank attached to a satellite bus, that gets dropped off in orbit.

But that costs a lot more and reduces payload per launch, compared to having it sit in Starship's own tanks, and let SS do all the maneuvering. 

Also, you can't just have an orifice in the tank and poke a hose in. The two vehicles need to be firmly connected, have thrust applied to settle the fluids, and have two pipes for each propellant. One for fluid and one for gas, so pressure differential will move the fluid one way and the gas the other.

Maybe you could do all that with a flexible 4-pipe and tether connector, but no need for thrusters. There are technologies for extending and guiding snake-like connectors.  (See Tesla auto-charger). Besides which, SS needs fueling pipes on the tail anyway, so it can be fueled through SH on the pad.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 11:02 am
Hi,

Just wanted to start a thread discussing orbital refueling technology proposed by SX, this is (for me) a very critical technology for SS concept success in the future, very little is known about SX approah to the matter, so any insight on technologies existing or in R&D stages is welcome here to improve our understanding.

Quote
One of SpaceX's principal engineers behind the Starship project, Paul Wooster, has identified orbital refueling as one of most difficult technology challenges the company will have to overcome in order to realize its Mars ambitions.


Thanks.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 11:47 am
The main issues to deal with are:

1-How to prevent cryogen from boiling-off (keep the fuel cool) for long duration.

2-How to transfert the cryogen from one container to the other.

NASA with Robotic Refueling Mission 3 (RRM3) has been trying to developpe the technology needed with relative success, but the scale needed for SX plans will require a complete review of the process for scaling up which may lead to using completly different technology.

The technology is at it early stages of maturation which means that time is still needed and this means that the technology (orbital refueling) most likely (IMHO) wont be ready for SS developement timeframe.

(https://www.nextbigfuture.com/wp-content/uploads/2017/11/a75bd7c23f0f273c52da444f936bedf3-730x430.jpg)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 07/27/2020 12:04 pm
From the info extracted from Elon tweets and plugging holes with informed guesses we have rough general picture of how it would be done:

- Dock the vehicles bottom-to-bottom and obtain seal of the piping
- Vent receiving end to low internal tank pressure (probably well below atmospheric[*])
- Use thrusters (probably receiving end, see below why[*][**]) to produce ullage thrust in the same it's done before re-firing of any stage in orbit
- Open valves and let the transfer begin
- Pressure difference pushes the liquids while ullage thrust (in the order of milli-gee) keeps them settled
- Keep venting the receiving end. There's one possible difficulty here, see below [***]
- If the sending end is close full (for example transfer from accumulation tanker to receiving ship) you need also to keep it pressurized at the required level.
- Once liquid fuel is done pressure would equalize quickly. Possibly bubble detection would shut the valves before this happens as this would be a bit violent. Or maybe the system is made robust enough for it not to be a problem.
- Terminate the ullage thrust
- Purge & vent the piping between main valves on both vehicles.
- Undock

Notes:
*] Venting below atmospheric ensures the remaining fuel would be at a super-cooled temperature as intended.
**] Receiving end needs to be kept at low pressure so it has obvious source of gas for ullage thrusters
***] Receiving end may see liquid foaming and condensation foring in the ullage space. Venting liquid would be wasteful and could be a source of problems as liquid venting to vacuum will partly evaporate and partly freeze and frozen liquid may plug holes. So venting piping could use "cyclone" separator and possibly some mild heating.

From the above one could notice that we need some gas generating subsystem which could pressurize the vehicle without engines firing. So some burner, pump for it and heat exchanger to vaporize the liquid.


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 12:32 pm
From the info extracted from Elon tweets and plugging holes with informed guesses we have rough general picture of how it would be done:

- Dock the vehicles bottom-to-bottom and obtain seal of the piping
- Vent receiving end to low internal tank pressure (probably well below atmospheric[*])
- Use thrusters (probably receiving end, see below why[*][**]) to produce ullage thrust in the same it's done before re-firing of any stage in orbit
- Open valves and let the transfer begin
- Pressure difference pushes the liquids while ullage thrust (in the order of milli-gee) keeps them settled
- Keep venting the receiving end. There's one possible difficulty here, see below [***]
- If the sending end is close full (for example transfer from accumulation tanker to receiving ship) you need also to keep it pressurized at the required level.
- Once liquid fuel is done pressure would equalize quickly. Possibly bubble detection would shut the valves before this happens as this would be a bit violent. Or maybe the system is made robust enough for it not to be a problem.
- Terminate the ullage thrust
- Purge & vent the piping between main valves on both vehicles.
- Undock

Notes:
*] Venting below atmospheric ensures the remaining fuel would be at a super-cooled temperature as intended.
**] Receiving end needs to be kept at low pressure so it has obvious source of gas for ullage thrusters
***] Receiving end may see liquid foaming and condensation foring in the ullage space. Venting liquid would be wasteful and could be a source of problems as liquid venting to vacuum will partly evaporate and partly freeze and frozen liquid may plug holes. So venting piping could use "cyclone" separator and possibly some mild heating.

From the above one could notice that we need some gas generating subsystem which could pressurize the vehicle without engines firing. So some burner, pump for it and heat exchanger to vaporize the liquid.

Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 07/27/2020 12:52 pm
What's the benefit?  The fuel will still need to transfer out of the docked external tank into the engines, the system they envision has them doing that into existing tankage that's already present and used to get it into earth orbit (when it operates as the second stage) and the rest of the challenges aren't materially different in terms of timing for transferring fuel in during boost versus during the RCS-ullage+pressurization scenario another user shared above.

When adding new elements to a problem to solve it, it's useful to know why the added complexity is required and what unique problem it's solving, otherwise perhaps there's a better way.

In this instance, it's not clear what shipping up external tanks that will need to be plumbed solves that doing the ship-ship transfer doesn't.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 01:05 pm
What's the benefit?  The fuel will still need to transfer out of the docked external tank into the engines, the system they envision has them doing that into existing tankage that's already present and used to get it into earth orbit (when it operates as the second stage) and the rest of the challenges aren't materially different in terms of timing for transferring fuel in during boost versus during the RCS-ullage+pressurization scenario another user shared above.

When adding new elements to a problem to solve it, it's useful to know why the added complexity is required and what unique problem it's solving, otherwise perhaps there's a better way.

In this instance, it's not clear what shipping up external tanks that will need to be plumbed solves that doing the ship-ship transfer doesn't.

May be because it's simpler to do so, just imagine space shuttle docking to her external tank and going all the way to Mars, no air drag issue, the tank is light weight, no time and energy waisted in fuel transfert, the external tank could also be imagined as a payload of a starship fairing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 07/27/2020 01:12 pm
May be because it's simpler to do so, just imagine space shuttle docking to her external tank and going all the way to Mars, no air drag issue, the tank is light weight, no time and energy waisted in fuel transfert, the external tank could also be imagined as a payload of a starship fairing.
I think I might not have written my question clearly, please allow me to try again: What makes this simpler?  You're adding parts, you still have docking operations, you still have to transfer fuel into the vehicle (the only thing this changes is when it happens; the fuel needs to come in during the burn as opposed to during the fueling operation)....

What part of this is simpler?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Alvian@IDN on 07/27/2020 01:16 pm
We ofc want a thought experiment....

But at the end of the day it's up to SpaceX who actually does it. Remember that they have a very smart engineers, who knows everything they're doing = considerations of choosing this refueling method

They obviously already has NASA on supports, just to keep in mind
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 01:23 pm
May be because it's simpler to do so, just imagine space shuttle docking to her external tank and going all the way to Mars, no air drag issue, the tank is light weight, no time and energy waisted in fuel transfert, the external tank could also be imagined as a payload of a starship fairing.
I think I might not have written my question clearly, please allow me to try again: What makes this simpler?  You're adding parts, you still have docking operations, you still have to transfer fuel into the vehicle (the only thing this changes is when it happens; the fuel needs to come in during the burn as opposed to during the fueling operation)....

What part of this is simpler?

I didn't add parts I just proposed a rethinking of the design, I also didn't suggest a fuel transfert but a direct fuel burn from external tank in SS engines, I think the proposed idea of mating two ends of SS is also hard to imagine, because it's not possible to get the fluid flow in a reversed direction without affecting all parts like pumps and turbines and you can't get extra plumbing in a place where it's already  congested (with engines).

We ofc want a thought experiment....

But at the end of the day it's up to SpaceX who actually does it. Remember that they have a very smart engineers, who knows everything they're doing = considerations of choosing this refueling method

They obviously already has NASA on supports, just to keep in mind

This a passive way to see things.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nomadd on 07/27/2020 01:26 pm
What's the benefit?  The fuel will still need to transfer out of the docked external tank into the engines, the system they envision has them doing that into existing tankage that's already present and used to get it into earth orbit (when it operates as the second stage) and the rest of the challenges aren't materially different in terms of timing for transferring fuel in during boost versus during the RCS-ullage+pressurization scenario another user shared above.

When adding new elements to a problem to solve it, it's useful to know why the added complexity is required and what unique problem it's solving, otherwise perhaps there's a better way.

In this instance, it's not clear what shipping up external tanks that will need to be plumbed solves that doing the ship-ship transfer doesn't.

May be because it's simpler to do so, just imagine space shuttle docking to her external tank and going all the way to Mars, no air drag issue, the tank is light weight, no time and energy waisted in fuel transfert, the external tank could also be imagined as a payload of a starship fairing.
You'd still have to transfer the fuel to the engines, which wouldn't be any simpler than transferring it to the SS tanks, and you'd be dragging an extra 30 or 40 tons of weight around.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Alvian@IDN on 07/27/2020 01:28 pm
because it's not possible to get the fluid flow in a reversed direction without affecting all parts like pumps and turbines...
What's make it not possible? It's not like they flow the propellant directly into the engine anyways


...and you can't get extra plumbing in a place where it's already  congested (with engines).
It still had quite enough space

With extra tanks, you will throw the full reusability right off the window, about as far as a billion light years away
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 07/27/2020 01:30 pm
May be because it's simpler to do so, just imagine space shuttle docking to her external tank and going all the way to Mars, no air drag issue, the tank is light weight, no time and energy waisted in fuel transfert, the external tank could also be imagined as a payload of a starship fairing.
I think I might not have written my question clearly, please allow me to try again: What makes this simpler?  You're adding parts, you still have docking operations, you still have to transfer fuel into the vehicle (the only thing this changes is when it happens; the fuel needs to come in during the burn as opposed to during the fueling operation)....

What part of this is simpler?

I didn't add parts I just proposed a rethinking of the design,
A rethinking that adds external tanks to the tanks used by starship to reach orbit in the first place.
Quote
I also didn't suggest a fuel transfert but a direct fuel burn from external tank in SS engines,
"Direct fuel burn" IS a fuel transfer.

Quote
I think the proposed idea of mating two ends of SS is also hard to imagine, because it's not possible to get the fluid flow in a reversed direction without affecting all parts like pumps and turbines and you can't get extra plumbing in a place where it's already  congested (with engines).
We're not talking the height-restricted SLS engine bay here- there's plenty of room for plumbing if it's needed to get the job done. The only question is how much mass it takes- and it is very easy to be less mass than a few external tanks.

Edit: You arnt thinking the fuel will be transferred from engine bell to engine bell, are you? because there are easier ways.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 07/27/2020 01:35 pm
Indeed, I think spitballing ideas is fun, for this one it just seems like it adds complexity without reducing it elsewhere.  More parts, different loading, etc yet the core problem of transferring the fuel in from another tank still exists, but now it exists under heavy thrust instead of being performed during an ullage-settled transfer between two Starships.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: guckyfan on 07/27/2020 01:35 pm
Starship gets fueled on the pad through internal connections from Superheavy, the first stage. Which means the connecting plumbing is already there.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 01:37 pm
What's the benefit?  The fuel will still need to transfer out of the docked external tank into the engines, the system they envision has them doing that into existing tankage that's already present and used to get it into earth orbit (when it operates as the second stage) and the rest of the challenges aren't materially different in terms of timing for transferring fuel in during boost versus during the RCS-ullage+pressurization scenario another user shared above.

When adding new elements to a problem to solve it, it's useful to know why the added complexity is required and what unique problem it's solving, otherwise perhaps there's a better way.

In this instance, it's not clear what shipping up external tanks that will need to be plumbed solves that doing the ship-ship transfer doesn't.

May be because it's simpler to do so, just imagine space shuttle docking to her external tank and going all the way to Mars, no air drag issue, the tank is light weight, no time and energy waisted in fuel transfert, the external tank could also be imagined as a payload of a starship fairing.
You'd still have to transfer the fuel to the engines, which wouldn't be any simpler than transferring it to the SS tanks, and you'd be dragging an extra 30 or 40 tons of weight around.

Fuel transfert to engines is easy because turbopumps and vacuuem will take care of it, and for the weight you did mention space shuttle external tank weight which is true but the proposed concept may propose multiple external tanks jettisoned when empty.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Alvian@IDN on 07/27/2020 01:42 pm
What's the benefit?  The fuel will still need to transfer out of the docked external tank into the engines, the system they envision has them doing that into existing tankage that's already present and used to get it into earth orbit (when it operates as the second stage) and the rest of the challenges aren't materially different in terms of timing for transferring fuel in during boost versus during the RCS-ullage+pressurization scenario another user shared above.

When adding new elements to a problem to solve it, it's useful to know why the added complexity is required and what unique problem it's solving, otherwise perhaps there's a better way.

In this instance, it's not clear what shipping up external tanks that will need to be plumbed solves that doing the ship-ship transfer doesn't.

May be because it's simpler to do so, just imagine space shuttle docking to her external tank and going all the way to Mars, no air drag issue, the tank is light weight, no time and energy waisted in fuel transfert, the external tank could also be imagined as a payload of a starship fairing.
You'd still have to transfer the fuel to the engines, which wouldn't be any simpler than transferring it to the SS tanks, and you'd be dragging an extra 30 or 40 tons of weight around.

Fuel transfert to engines is easy because turbopumps and vacuuem will take care of it, and for the weight you did mention space shuttle external tank weight which is true but the proposed concept may propose multiple external tanks jettisoned when empty.
Hello, full & rapid reusability..
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 07/27/2020 01:48 pm
This is something that needs to be solved, but I consider it at least an order of magnitude easier than, say, reusable orbital reentry of the upper stage. You don’t need massive pumps or whatever. Just some light ullage thrust and a pressure differential.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: meekGee on 07/27/2020 02:15 pm




Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Eer on 07/27/2020 02:17 pm
For me (IANARS or E) is the active cooling (and it's energy requirements) for sub-cooling the O2 and CH4.  Yes, the issue exists at the ISRU production for return, but it also seems likely to be required for the propellent accumulation before TMI.

Transfer is part of the issue, though the pressure gradient discussion in OP is enlightening.  But accomplishing and maintaining the subcooling densification will require something more than just venting gas, I would think.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 07/27/2020 02:27 pm
My idea is not to have a crew in an outbound for Mars Starship waiting and having to have 6 tankers dock and transfer fuel.

My idea is to put tankers in orbit with refrigeration equipment and some solar panels for power.  Then fill the tankers with other tankers.  Then the outbound crewed ship can dock only once to a full tanker and then go. 

So essentially orbital tankers will become fuel depots for only one stop and fill to go per Starship.  Orbital tankers can be filled at reasonable pace during the off synods.  Then you have several full orbital tankers that can be used for any outbound Starships to the moon or Mars.  SpaceX could also let others license their technology to fill tankers with their rockets for a mission they want to do in deep space. 

For massive flotillas of Starships bound for Mars colony during the 6 month synod when Mars is closest to earth, fuel depots of some sort is the only way to save time.  Using Starship tankers would be easy without building anything new.   Starship will be a do-all spacecraft.  Tanker, fuel depot, moon lander, Mars lander, satellite orbiter, cargo carrier.  Modified for whatever is needed. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 07/27/2020 02:37 pm
For me (IANARS or E) is the active cooling (and it's energy requirements) for sub-cooling the O2 and CH4.  Yes, the issue exists at the ISRU production for return, but it also seems likely to be required for the propellent accumulation before TMI.

Transfer is part of the issue, though the pressure gradient discussion in OP is enlightening.  But accomplishing and maintaining the subcooling densification will require something more than just venting gas, I would think.
As I understand it (if I understand it ...), the purpose of prop densification is to get more prop in the tanks to help with the very difficult task of getting out of the lower bits of Earth's relatively deep gravity well.  Do we really think they're going to be using densified props for TMI, or Mars ascent?  I had always assumed sub-cooling would be a getting-off-of-Earth thing, and otherwise props would be run of the mill (i.e., at the tank pressure determined  boiling point) cryogens.  Is Raptor so finely tuned that it requires sub-cooled prop, or is sub-cooling just used for a needed performance boost?  I had always assumed the latter.  My own (totally uninformed) speculation is that sub-cooling is a total non-issue for on-orbit prop transfer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 02:51 pm




Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: meekGee on 07/27/2020 02:58 pm




Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: whitelancer64 on 07/27/2020 03:06 pm
My idea is not to have a crew in an outbound for Mars Starship waiting and having to have 6 tankers dock and transfer fuel.

My idea is to put tankers in orbit with refrigeration equipment and some solar panels for power.  Then fill the tankers with other tankers.  Then the outbound crewed ship can dock only once to a full tanker and then go. 

So essentially orbital tankers will become fuel depots for only one stop and fill to go per Starship.  Orbital tankers can be filled at reasonable pace during the off synods.  Then you have several full orbital tankers that can be used for any outbound Starships to the moon or Mars.  SpaceX could also let others license their technology to fill tankers with their rockets for a mission they want to do in deep space. 

For massive flotillas of Starships bound for Mars colony during the 6 month synod when Mars is closest to earth, fuel depots of some sort is the only way to save time.  Using Starship tankers would be easy without building anything new.   Starship will be a do-all spacecraft.  Tanker, fuel depot, moon lander, Mars lander, satellite orbiter, cargo carrier.  Modified for whatever is needed.

Was there someone suggesting they would launch crew first? That's entirely backwards.

I am pretty sure that using tankers as temporary fuel depots is already SpaceX's plan.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 03:09 pm




Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 07/27/2020 03:10 pm
Just here to point out that the Shuttle External Tank cost $75 million, for a launch vehicle whose total payload is about 20-27 tonnes to LEO. And doesn’t last long in LEO due in part to popcorning of the insulation. Drop tanks are much too expensive except for maybe specialized missions. It’s not much different from just being totally expendable...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: whitelancer64 on 07/27/2020 03:13 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..

Why does there have to be an interim solution?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 03:16 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..

Why does there have to be an interim solution?

Because time is ticking and SX plans are too ambitiuos (IMHO) for the timeframe, and it's like a black hole money siphoning endeavor.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: meekGee on 07/27/2020 03:18 pm




Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..
Why do you think that after figuring how to build a reusable heavy launcher, orbital fueling will be any sort of issue, or an issue big enough to warrant something as problematic for reuse as an external tank?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Alvian@IDN on 07/27/2020 03:24 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..

Why does there have to be an interim solution?

Because time is ticking and SX plans are too ambitiuos (IMHO) for the timeframe, and it's like a black hole money siphoning endeavor.
Source? Or just for throwing shade at SpaceX instead of refueling technique discussion?

You are at it again, dedicated expensive lander, and now this?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/27/2020 03:27 pm




Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..
Why do you think that after figuring how to build a reusable heavy launcher, orbital fueling will be any sort of issue, or an issue big enough to warrant something as problematic for reuse as an external tank?

You certainly mean theorically figuring how, but in practice and in a financially viable manner I dont think so.

Alvian@IDN
Quote
Source? Or just for throwing shade at SpaceX instead of refueling technique discussion?

You are at it again

I made my point of view considering NASA R&D progress on the matter (orbital refuelling).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: whitelancer64 on 07/27/2020 03:27 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..

Why does there have to be an interim solution?

Because time is ticking and SX plans are too ambitiuos (IMHO) for the timeframe, and it's like a black hole money siphoning endeavor.

If "time is ticking" and money is short, then the last thing they'd want to be doing is working on a more complicated solution to the problem that also requires more money for development, hardware fabrication, tooling, etc.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 07/27/2020 03:54 pm
Starship already needs to be refueled on Mars for return.

If you want to get started and send equipment or even people to Mars one-way and you don’t care about the cost to do it, Red Dragon on falcon Heavy would work. They’d have to pay for finishing the powered landing capability development for Dragon plus probably a Dragon XL module for habitable space on the way there, but it’d work.

But it’s orders of magnitude more expensive per kg to Mars. It’s wasting precious capital (including human capital, which may be rarer than financial capital) and time on a dead end that nonetheless is risky.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 07/27/2020 03:58 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..

Why does there have to be an interim solution?

Because time is ticking and SX plans are too ambitiuos (IMHO) for the timeframe, and it's like a black hole money siphoning endeavor.
I can't tell whether this is a troll, some strange sort of irony, or just a really uninformed but serious suggestion.

The idea is to launch a tank (pretty much a Starship with most bits thrown away somehow) and then dock with that, connect plumbing, and keep it attached for propellant transfer during the trans-Mars burn, yes?

And the justification is that it is simpler than launching another Starship, docking, connecting plumbing, transferring the propellant and then undocking?

Am I missing something here?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsnellenberger on 07/27/2020 06:09 pm
Refueling Starship on orbit requires an automated/robotic umbilical connection mechanism. It can be more robustly built and complex than a “standard” rocket umbilical, since it’s going into a large, reusable vehicle. It also probably wants to be androgynous, since Starships will be refueling back-to-back.

My thought is - once they’ve developed that androgynous, automated refueling mechanism, why wouldn’t SpaceX use that same mechanism everywhere it applies?  Pad umbilicals & SH tanking umbilicals, SH-Starship tanking umbilicals, Starship-Starship refueling umbilicals, MarsPort pad umbilicals - the works. One added benefit would be that they’d be testing the bejeezus out of the mechanism every time they tanked or flew one of the vehicles.

Another benefit would be that they could move away from having time-critical T-0 umbilicals. Once the vehicle is in the final count and the tanks are pressurizing for flight (T-30-ish), disconnect and retract the umbilicals into a protective housing. If the launch is aborted, the umbilicals just hook themselves back up. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: philw1776 on 07/27/2020 07:05 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?

At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...

You're right to point out to the vocabulary it might be deceiving, for me "Just" and "simply" means proven technology have been in use for a while.
"Complexe" means that it's at the early stages of concept has to be put in practice in reliable manner.
So..  it's proven to be prohibitively expensive?
 
Starship is anything but a Shuttle repeat.  It's the anti-shuttle.  Consider each element in isolation, fine, but certainly don't give it points just because it was used in shuttle..

Remember Starship has to land on Mars and take off again, then re-enter Earth.  Are you "just" going to keep docking with new tanks all the time and throwing them away afterwards?

Refueling is fundamental to any mode of transportation, and compared to other challenges like, say, launch and EDL, it's relatively easy.

I'm from those who consider Starship as a mean to get to Mars and beyond, the primary goal is to get to Mars, getting it cheap getting it reausable getting it through re-fueling or external tanks, this is for me a secondary issue that could be worked out, it must be some interim solution to get to Mars, but to get it in the 2030's IMHO we must get some compromises..

Why does there have to be an interim solution?

Because time is ticking and SX plans are too ambitiuos (IMHO) for the timeframe, and it's like a black hole money siphoning endeavor.

Yet another thread disguised as SpaceX plans for X that is in reality "My alternative architecture because SpaceX is on a track to fail"
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Vanspace on 07/27/2020 08:33 pm
I have been wondering about the capabilities of the anti-boiloff systems. For Mars ISRU the last step is to condense the gases to liquid.

Can the anti-boiloff system on the now empty landed Starships be fed gaseous propellants and condense them down?

If it can, that saves a bunch of payload. If it can't is there a simple pre-stage like cooling or compressing that will help?

If the androgynous GSE adapter also has the vent lines can those be used to feed gaseous propellants to the anti-boiloff system?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 07/27/2020 10:22 pm
Can the anti-boiloff system on the now empty landed Starships be fed gaseous propellants and condense them down?

I can't recall hearing anything about cryocoolers being standard equipment on Starships, even on the Mars-bound Starships.  As far as I have heard, the only systems for preventing boiloff during the Earth-Mars and Mars-Earth transits, will be insulation (header tanks inside the main tanks; the empy main tanks then act as a giant thermos bottle), and minimize sunshine hitting the tanks by pointing them away from the Sun.  This will lower the amount boiloff enough that there will be propellant left when they arrive, but it will not entirely prevent boiloff.  And it is definitely not enough to chill down the propellant.

To actually chill down oxygen and methane to condense into liquids, you either need active cooling, i.e. cryocoolers, or get much further out in our solar system.

So they will need to bring cryocoolers to Mars as normal cargo.

(It would make sense to have cryocoolers on accumulation tankers in LEO, though.  It is much more difficult to keep the propellant cool in LEO, as A) you will fill the main tanks, so there is no thermos bottle protecting the propellant, and B) there is no direction to point the ship to minimize incoming heat radiation, as the Earth takes up almost half the surroundings, reflecting sunlight onto you from almost everywhere.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Vanspace on 07/27/2020 10:30 pm
Can the anti-boiloff system on the now empty landed Starships be fed gaseous propellants and condense them down?

I can't recall hearing anything about cryocoolers being standard equipment on Starships, even on the Mars-bound Starships.  As far as I have heard, the only systems for preventing boiloff during the Earth-Mars and Mars-Earth transits, will be insulation (header tanks inside the main tanks; the empy main tanks then act as a giant thermos bottle), and minimize sunshine hitting the tanks by pointing them away from the Sun.  This will lower the amount boiloff enough that there will be propellant left when they arrive, but it will not entirely prevent boiloff.  And it is definitely not enough to chill down the propellant.

To actually chill down oxygen and methane to condense into liquids, you either need active cooling, i.e. cryocoolers, or get much further out in our solar system.

So they will need to bring cryocoolers to Mars as normal cargo.

(It would make sense to have cryocoolers on accumulation tankers in LEO, though.  It is much more difficult to keep the propellant cool in LEO, as A) you will fill the main tanks, so there is no thermos bottle protecting the propellant, and B) there is no direction to point the ship to minimize incoming heat radiation, as the Earth takes up almost half the surroundings, reflecting sunlight onto you from almost everywhere.)

Thank you!

If there needs to be cryocoolers on accumulation tankers, could they be added as part of the spec for mars bound ships? If you have to send it anyway, preassembled and already functioning might be better than as cargo.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 07/27/2020 10:54 pm
From the info extracted from Elon tweets and plugging holes with informed guesses we have rough general picture of how it would be done:

- Dock the vehicles bottom-to-bottom and obtain seal of the piping
- Vent receiving end to low internal tank pressure (probably well below atmospheric[*])
- Use thrusters (probably receiving end, see below why[*][**]) to produce ullage thrust in the same it's done before re-firing of any stage in orbit
- Open valves and let the transfer begin
- Pressure difference pushes the liquids while ullage thrust (in the order of milli-gee) keeps them settled
- Keep venting the receiving end. There's one possible difficulty here, see below [***]
- If the sending end is close full (for example transfer from accumulation tanker to receiving ship) you need also to keep it pressurized at the required level.
- Once liquid fuel is done pressure would equalize quickly. Possibly bubble detection would shut the valves before this happens as this would be a bit violent. Or maybe the system is made robust enough for it not to be a problem.
- Terminate the ullage thrust
- Purge & vent the piping between main valves on both vehicles.
- Undock

Notes:
*] Venting below atmospheric ensures the remaining fuel would be at a super-cooled temperature as intended.
**] Receiving end needs to be kept at low pressure so it has obvious source of gas for ullage thrusters
***] Receiving end may see liquid foaming and condensation foring in the ullage space. Venting liquid would be wasteful and could be a source of problems as liquid venting to vacuum will partly evaporate and partly freeze and frozen liquid may plug holes. So venting piping could use "cyclone" separator and possibly some mild heating.

From the above one could notice that we need some gas generating subsystem which could pressurize the vehicle without engines firing. So some burner, pump for it and heat exchanger to vaporize the liquid.

Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)
Sadly there is no way around the rocket equation. Adding a huge and heavy external tank will just slow it down (not to mention the design complexities). In fact more engines would be needed just to get it off the ground.

What SpaceX are doing with Starship in a crude sense is enabling a very fat rocket that would pose very difficult engineering problems to be launched in multiple thinner segments (if that makes sense).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: envy887 on 07/28/2020 12:42 am
Can the anti-boiloff system on the now empty landed Starships be fed gaseous propellants and condense them down?

I can't recall hearing anything about cryocoolers being standard equipment on Starships, even on the Mars-bound Starships.  As far as I have heard, the only systems for preventing boiloff during the Earth-Mars and Mars-Earth transits, will be insulation (header tanks inside the main tanks; the empy main tanks then act as a giant thermos bottle), and minimize sunshine hitting the tanks by pointing them away from the Sun.  This will lower the amount boiloff enough that there will be propellant left when they arrive, but it will not entirely prevent boiloff.  And it is definitely not enough to chill down the propellant.

To actually chill down oxygen and methane to condense into liquids, you either need active cooling, i.e. cryocoolers, or get much further out in our solar system.

There's anther option. With a good sunshield you can freeze methane or oxygen solid, even at 1 AU.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 07/28/2020 03:26 am
Starship as said, will point engines toward sun traveling to and from Mars.  Then they will have fold out solar panels for power to and from Mars.  These will have a tendency to have some shade also since they will hing near the bottom and turn 90 degrees to the rocket toward the sun.  So the engines, which can take heat and the solar panels will shield the fuel and the crew from the sun.  They will have equipment on board to keep the fuel and lox liquid.  Also, lox and methane are only about 20 degrees different in liquid form.  They both would have far less boil-off than liquid hydrogen which is over twice as cold.  Space is already very cold in the shade.  Also, like someone up said, in LEO, you have more boil-off problems because of heat reflected off the earth. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thrustpuzzle on 07/28/2020 04:20 am
Starship gets fueled on the pad through internal connections from Superheavy, the first stage. Which means the connecting plumbing is already there.
Starship's fueling has always been planned to be from the bottom, but is it for certain that SS's fuel on the Earth launch pad is delivered though Superheavy? The other alternative is a Starship specific GSE connection on the side of Superheavy's interstage, which has the small amount of plumbing to redirect the fuel and oxidizer to Starship's bottom fittings.

The SH interstage GSE connection plan has several tradeoffs, but the interstage method gives isolation between SH and SS plumbing systems and less weight and complexity (a lot of the plumbing and pumping is moved to GSE).  Downside is trickier GSE hardware design and operation, which may interfere with the fast (1-hour!) turnarounds Elon hopes for.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 07/28/2020 09:50 am
If there needs to be cryocoolers on accumulation tankers, could they be added as part of the spec for mars bound ships? If you have to send it anyway, preassembled and already functioning might be better than as cargo.

I don't think there are as much commonality as one might think between the two use cases.

The cryocoolers on an accumulation tanker (if they will have such coolers) will be designed to keep the already cold propellant cold, presumably thus handling liquids, and have radiators sized and designed for operation in vacuum, with significant influx of heat radiation from the Sun and Earth.  Cryocoolers for Mars ISRU will need to actually chill the propellants from maybe -60°C (if they are pre-cooled to ambient Mars temperatures) and in gas form, down to more than -180°C and liquid, and their radiators will want to be optimized for atmospheric use (thin though it may be).  I'm not sure how much difference this makes to their designs, though.

The normal tanking connections on Starship will be designed for liquids, and are basically straight pipes from the bottom of the ship up to the bottom of their respective tanks.  Cryocoolers on-board the ship will need to have a separate connector for receiving gas, and then either have their own pipes into the tanks for dumping their liquids into them, or have hoses that are connected to the normal tanking connectors.  (Having the cryocoolers sit inbetween the normal tanking connectors and the tanks, seems even more difficult: those connectors would then need to be able to handle both gas and liquid; there is limited space under the skirt; and the plumbing gets much more complicated.)

Later on (i.e, not the first few missions), SpaceX wants to be able to return ships within a couple of weeks from arriving at Mars, in order for them to be back on Earth to be reused every synod instead of every other synod.  Then you need to have cryocoolers on the ground, to cool the propellant as it is produced, as there won't be enough cryocooling capacity on the few ships that are landed.  Designing on-board cryocoolers to be used only for the first couple of missions would be a detour.


Those are the negatives.  On the other hand, for the first several missions, the ships themselves will need to act as tanks for the liquid propellants as it is produced, as there won't be enough room to bring separate tanks.  And you will need to keep that propellant cold for a long time, while the ships are surrounded by relatively warm atmosphere.  It seems unlikely that can be done efficiently without active cooling.  (Unless you build the tanks as large dewar flasks, but that will increase the dry mass of the ships significantly.)

But I think it still makes more sense to have the cryocoolers separate.  Then you can leave them on Mars, and have more space for other stuff on the next mission.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 07/28/2020 10:11 am
To actually chill down oxygen and methane to condense into liquids, you either need active cooling, i.e. cryocoolers, or get much further out in our solar system.

There's anther option. With a good sunshield you can freeze methane or oxygen solid, even at 1 AU.

Fairly difficult in LEO, though.  Half your viewfield has a warm planet which needs to be blocked, and the Sun shines in from some other direction.

And if the James Webb Space Telescope is anything to go by, a good sunshield is not cheap.  But possibly, if you stay away from LEO and LLO, it doesn't need to be quite as good and expensive. 8)

But yes, good point in general.  If one can be made simple and cheap, a sunshield could possibly be something for the HLS Starship waiting for the crew to arrive in NRHO at the Lunar Gateway.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/28/2020 10:45 am
From the info extracted from Elon tweets and plugging holes with informed guesses we have rough general picture of how it would be done:

- Dock the vehicles bottom-to-bottom and obtain seal of the piping
- Vent receiving end to low internal tank pressure (probably well below atmospheric[*])
- Use thrusters (probably receiving end, see below why[*][**]) to produce ullage thrust in the same it's done before re-firing of any stage in orbit
- Open valves and let the transfer begin
- Pressure difference pushes the liquids while ullage thrust (in the order of milli-gee) keeps them settled
- Keep venting the receiving end. There's one possible difficulty here, see below [***]
- If the sending end is close full (for example transfer from accumulation tanker to receiving ship) you need also to keep it pressurized at the required level.
- Once liquid fuel is done pressure would equalize quickly. Possibly bubble detection would shut the valves before this happens as this would be a bit violent. Or maybe the system is made robust enough for it not to be a problem.
- Terminate the ullage thrust
- Purge & vent the piping between main valves on both vehicles.
- Undock

Notes:
*] Venting below atmospheric ensures the remaining fuel would be at a super-cooled temperature as intended.
**] Receiving end needs to be kept at low pressure so it has obvious source of gas for ullage thrusters
***] Receiving end may see liquid foaming and condensation foring in the ullage space. Venting liquid would be wasteful and could be a source of problems as liquid venting to vacuum will partly evaporate and partly freeze and frozen liquid may plug holes. So venting piping could use "cyclone" separator and possibly some mild heating.

From the above one could notice that we need some gas generating subsystem which could pressurize the vehicle without engines firing. So some burner, pump for it and heat exchanger to vaporize the liquid.

Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)
Sadly there is no way around the rocket equation. Adding a huge and heavy external tank will just slow it down (not to mention the design complexities). In fact more engines would be needed just to get it off the ground.

What SpaceX are doing with Starship in a crude sense is enabling a very fat rocket that would pose very difficult engineering problems to be launched in multiple thinner segments (if that makes sense).

Orbital refueling/ISRU are technologies in the early stages of developpement this will require more time to be ready for SX planed timeframe and regarding the scale of requirement which is almost industriel to add to the diffucilty.

IMHO the external tank option seems to me the interim solution, I was also imagining putting those tanks in different positions like LEO, HEO, HTO, MO.............like gas stations. Those tanks may get some small thrusters, they dont have to travel with high velocity as is required for human travel.

Can someone give some data about fuel requirement for 100 tons dry mass SS round trip to Mars?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tyrred on 07/28/2020 10:59 am
From the info extracted from Elon tweets and plugging holes with informed guesses we have rough general picture of how it would be done:

- Dock the vehicles bottom-to-bottom and obtain seal of the piping
- Vent receiving end to low internal tank pressure (probably well below atmospheric[*])
- Use thrusters (probably receiving end, see below why[*][**]) to produce ullage thrust in the same it's done before re-firing of any stage in orbit
- Open valves and let the transfer begin
- Pressure difference pushes the liquids while ullage thrust (in the order of milli-gee) keeps them settled
- Keep venting the receiving end. There's one possible difficulty here, see below [***]
- If the sending end is close full (for example transfer from accumulation tanker to receiving ship) you need also to keep it pressurized at the required level.
- Once liquid fuel is done pressure would equalize quickly. Possibly bubble detection would shut the valves before this happens as this would be a bit violent. Or maybe the system is made robust enough for it not to be a problem.
- Terminate the ullage thrust
- Purge & vent the piping between main valves on both vehicles.
- Undock

Notes:
*] Venting below atmospheric ensures the remaining fuel would be at a super-cooled temperature as intended.
**] Receiving end needs to be kept at low pressure so it has obvious source of gas for ullage thrusters
***] Receiving end may see liquid foaming and condensation foring in the ullage space. Venting liquid would be wasteful and could be a source of problems as liquid venting to vacuum will partly evaporate and partly freeze and frozen liquid may plug holes. So venting piping could use "cyclone" separator and possibly some mild heating.

From the above one could notice that we need some gas generating subsystem which could pressurize the vehicle without engines firing. So some burner, pump for it and heat exchanger to vaporize the liquid.

Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)
Sadly there is no way around the rocket equation. Adding a huge and heavy external tank will just slow it down (not to mention the design complexities). In fact more engines would be needed just to get it off the ground.

What SpaceX are doing with Starship in a crude sense is enabling a very fat rocket that would pose very difficult engineering problems to be launched in multiple thinner segments (if that makes sense).

Orbital refueling/ISRU are technologies in the early stages of developpement this will require more time to be ready for SX planed timeframe and regarding the scale of requirement which is almost industriel.

IMHO the external tank option seems to me the interim solution, I was also imagining putting those tanks in different positions like LEO, HEO, HTO, MO.............like gas stations. Those tanks may get some small thrusters, they dont have to travel with high velocity as is required for human travel.

Can someone give some data about fuel requirement for 100 tons dry mass SS round trip to Mars?

Orbital refueling is *not in early stages of development.

Orbital refueling has been standard operating  practice with the ISS for many years. It just doesn't *look like what orbital refueling for Starship will look like.

That is a completely separate topic from ISRU.

External tanks for Starship, as has been pointed out, are a pipe dream. The plan is for orbital refueling with tanker Starships. Stick to the plan, it's what the experts are doing.

Refueling airplanes in mid-air is nearly a century old. Rendezvous and docking in space is more than a half-century old. These are solved problems. Implementation is a different story.

Methinks you're making a mountain out of a molehill.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/28/2020 12:18 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)

I think you are missing two points that many others are trying to make, in different ways:

Point 1:

How do these tanks get into orbit?

These fully fuelled tanks, holding 5 launches worth of fuel.

Without refuelling.

Point 2:

What makes an external tank superior to the internal tanks already carried within each Starship?

An external tank that has to dock and be reliably attached and plumbed into the Mars-going Starship.

Versus the internal tanks of the same capacity already built into the Starship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/28/2020 03:34 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)

I think you are missing two points that many others are trying to make, in different ways:

Point 1:

How do these tanks get into orbit?

These fully fuelled tanks, holding 5 launches worth of fuel.

Without refuelling.

Point 2:

What makes an external tank superior to the internal tanks already carried within each Starship?

An external tank that has to dock and be reliably attached and plumbed into the Mars-going Starship.

Versus the internal tanks of the same capacity already built into the Starship.

Just imagine the space shuttle without orbiter (only small engines on the external tank) how much fuel would be saved on the external tank?

-An external has not to do EDL or ascent again he could be jettisoned when empty.

-Imagine the fuel saved on getting rid of the extra weight of the tankers to get back to earth.

-Time saving on refueling.
-Saving on ullage fuel.
-Technologie available.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wes_wilson on 07/28/2020 03:52 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)

I think you are missing two points that many others are trying to make, in different ways:

Point 1:

How do these tanks get into orbit?

These fully fuelled tanks, holding 5 launches worth of fuel.

Without refuelling.

Point 2:

What makes an external tank superior to the internal tanks already carried within each Starship?

An external tank that has to dock and be reliably attached and plumbed into the Mars-going Starship.

Versus the internal tanks of the same capacity already built into the Starship.

Just imagine the space shuttle without orbiter (only small engines on the external tank) how much fuel would be saved on the external tank?

-An external has not to do EDL or ascent again he could be jettisoned when empty.

-Imagine the fuel saved on getting rid of the extra weight of the tankers to get back to earth.

-Time saving on refueling.
-Saving on ullage fuel.
-Technologie available.

I get it, I've had ideas I was attached too that the group had to dissuade me from.  However, your idea is missing one single and totally not-negotiable item.  The system has to be 100% re-usable because that's what the company owner (Elon Musk) has set as a requirement.

So modify your concept of using the external fuel tank from a shuttle to include re-use.  Make whatever changes are needed for it to re-enter, be recovered, be refurbished, and then used a 2nd time and see where that leads you.  Eventually, it leads to something like Starship.

Any concept that isn't 100% re-usable isn't worth discussing because it doesn't meet the criteria required by the owner/operator of the project.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 07/28/2020 04:12 pm
[snip]

Just imagine the space shuttle without orbiter (only small engines on the external tank) how much fuel would be saved on the external tank?

Such a vehicle wouldn't be able to reach orbit. The Shuttle could only reach orbit because of the SSMEs, which you have replaced with "small engines."

-An external has not to do EDL or ascent again he could be jettisoned when empty.

-Imagine the fuel saved on getting rid of the extra weight of the tankers to get back to earth.

See wes_wilson's comments on reuse.

-Time saving on refueling.

-Saving on ullage fuel.

Extra structural mass because you're strapping a tank to the side which has to withstand the high-thrust TMI burn.

-Technologie available.

See SLS for an example of how that logic worked out.

And no, the technology isn't available. The Shuttle ET was never autonomously/telerobotically attached. It was never attached in orbit. Those are very different problems from attaching the ET on the ground in a purpose-built processing facility.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: SkyRate on 07/28/2020 04:13 pm
To actually chill down oxygen and methane to condense into liquids, you either need active cooling, i.e. cryocoolers, or get much further out in our solar system.
There's anther option. With a good sunshield you can freeze methane or oxygen solid, even at 1 AU.
Fairly difficult in LEO, though.  Half your viewfield has a warm planet which needs to be blocked, and the Sun shines in from some other direction.
You could borrow a trick from the lunar polar craters, and make a cylindrical sunshade with the open ends pointing out of the ecliptic.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AC in NC on 07/28/2020 04:36 pm
I'd like to gently suggest that Star-Dust's solution-in-search-of-a-problem is so ill-formed as a concept that it deserves to be ignored and permitted a face-saving death on the vine.

These are a hodge-podge of internally inconsistent notions and it's unproductive to engage then with rational responses.

-  just imagine space shuttle docking to her external tank and going all the way to Mars

-  the external tank could also be imagined as a payload of a starship fairing.

-  I didn't add parts I just proposed a rethinking of the design, I also didn't suggest a fuel transfert but a direct fuel burn from external tank in SS engines

-  because it's not possible to get the fluid flow in a reversed direction without affecting all parts like pumps and turbines and you can't get extra plumbing in a place where it's already  congested (with engines).

-  and for the weight you did mention space shuttle external tank weight which is true but the proposed concept may propose multiple external tanks jettisoned when empty

-  Because time is ticking and SX plans are too ambitiuos (IMHO) for the timeframe, and it's like a black hole money siphoning endeavor.

-  I was also imagining putting those tanks in different positions like LEO, HEO, HTO, MO.............like gas stations.

-  Just imagine the space shuttle without orbiter (only small engines on the external tank) how much fuel would be saved on the external tank?

-  An external has not to do EDL or ascent again he could be jettisoned when empty.

-  Imagine the fuel saved on getting rid of the extra weight of the tankers to get back to earth.


Here we've got:

1.  Docking and fuel transfer from an external tank (Shuttle-style) solving (on an interim basis) the problem of docking and fuel transfer from an external tank (SS)
2.  Direct fuel burn from an externally mounted Shuttle-style tank not "adding parts" (which is a metaphor for resdesigning and adding complexity)
3.  An external tank as a payload inside a SS.
4.  The contention that it's not possible to reverse flow props though that's the central design principles of the aft GSE connects.
5.  Solving the time/energy/cost/complexity of multiple SS retankings with the obvious time/energy/cost/complexity savings of multiple external tanks jettisoned.
6.  The baseless black-hole siphoning nonsense.
7.  Boosting multiple expendable tanks as gas stations rather than direct burn tankage.
8.  Saving the expense of the the fuel to recover SS's at the cost of multiple expendable tanks.

Please.  Just let this die.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 07/28/2020 04:44 pm
To actually chill down oxygen and methane to condense into liquids, you either need active cooling, i.e. cryocoolers, or get much further out in our solar system.
There's anther option. With a good sunshield you can freeze methane or oxygen solid, even at 1 AU.
Fairly difficult in LEO, though.  Half your viewfield has a warm planet which needs to be blocked, and the Sun shines in from some other direction.

You could borrow a trick from the lunar polar craters, and make a cylindrical sunshade with the open ends pointing out of the ecliptic.
You'd need the sunshade to extend well past the ends of the tank to shade out heat from the Earth, and at that point your view solid angle is very small, so you wouldn't get much natural cooling.

Two cones that meet at the vehicle's "waist" would work better, since you'd get more solid angle for cooling. You'd want both cones to be angled so they point away from Earth, blocking out both the unidirectional Sun and the Earth hemisphere.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: matthewkantar on 07/28/2020 05:15 pm
It is so simple:

-Tanker to orbit.
-tanker to orbit, fill tanker in orbit,RTLS.
-repeat until prop requirement is met.
-Starship crew and gear to orbit.
-rendezvous, prop transfer, burn for Mars.

Any parts or steps added to this are dumb.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 07/28/2020 05:48 pm
It is so simple:

-Tanker to orbit.
-tanker to orbit, fill tanker in orbit,RTLS.
-repeat until prop requirement is met.
-Starship crew and gear to orbit.
-rendezvous, prop transfer, burn for Mars.

Any parts or steps added to this are dumb.

Adding drop tanks? Yeah, I agree that's dumb.

Adding sunshades and/or cryocoolers? I invite you to do the heat transfer calculation through a 4 mm stainless steel wall before calling it dumb. Even if we assume the heatshield is perfectly insulative, you're still getting a lot of Earthshine on the day side of your orbit.

Personally I think cryocoolers are most likely, considering the desire for full reuse and the complexity of "un-deploying" a large sunshade.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Lemurion on 07/28/2020 07:28 pm
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)

I think you are missing two points that many others are trying to make, in different ways:

Point 1:

How do these tanks get into orbit?

These fully fuelled tanks, holding 5 launches worth of fuel.

Without refuelling.

Point 2:

What makes an external tank superior to the internal tanks already carried within each Starship?

An external tank that has to dock and be reliably attached and plumbed into the Mars-going Starship.

Versus the internal tanks of the same capacity already built into the Starship.

Just imagine the space shuttle without orbiter (only small engines on the external tank) how much fuel would be saved on the external tank?

-An external has not to do EDL or ascent again he could be jettisoned when empty.

-Imagine the fuel saved on getting rid of the extra weight of the tankers to get back to earth.

-Time saving on refueling.
-Saving on ullage fuel.
-Technologie available.

If you could actually get this tank to orbit with a full load of propellant, you'd also be able to get a fully fueled Starship to orbit so you wouldn't need the tank in the first place.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: aero on 07/28/2020 09:04 pm
It seems that using the heavy as a tanker doesn't work very well. I've calculated (using the rocket equation) that the heavy without a Starship could "probably" make it to orbit, there doesn't seem to be much leeway for residual prop upon reaching orbit. I do think that using more accurate data and math, some residual prop to orbit could be achieved but requiring an expendable heavy really precludes the approach.

The key factor in my calculation is that the heavy leaves the launch pad with a 2 to 1 thrust to weight ratio. High acceleration limits gravity losses tremendously. I calculate, using constant ISP = 340 seconds, that the heavy can achieve 8.5 km/sec delta-V. It should do better because ISP increases as the rocket leaves the sensible atmosphere and at ISP = 360 seconds, the heavy could achieve delta-V = 9.1 km/sec. A better estimate is somewhere between the two extremes.

The above still fails utterly as a tanker though, because the heavy reaches orbit with no residual prop and is incapable of returning to Earth.

It would work as a fuel depot if there were reasons to do so.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: CJ on 07/28/2020 09:23 pm
My working assumption is that the most efficient way to refuel a Starship in a high-energy orbit would be to refuel another Starship in LEO with several prop launches, and then have it transfer the fuel load to the actual mission Starship. (assuming the mission starship can't do the mission from LEO)

The reason I think this is that attempting to send tanker launches to a higher orbit means pushing a lot of mass though a lot of delta-v; the entire dry mass plus landing prop on each prop launch, pushed through the delta/v between LEO and the target orbit. 

Do I have this right?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Stan-1967 on 07/28/2020 09:34 pm
Do I have this right?

I would add that there should be more daily opportunities to launch propellant to an intermediate SS in LEO vs. the mission SS in HEO.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 07/28/2020 09:46 pm
My working assumption is that the most efficient way to refuel a Starship in a high-energy orbit would be to refuel another Starship in LEO with several prop launches, and then have it transfer the fuel load to the actual mission Starship. (assuming the mission starship can't do the mission from LEO)

The reason I think this is that attempting to send tanker launches to a higher orbit means pushing a lot of mass though a lot of delta-v; the entire dry mass plus landing prop on each prop launch, pushed through the delta/v between LEO and the target orbit.

Do I have this right?

Yes, I think that's correct.

Extrapolating a bit from this same logic, you should make sure that 1) you transfer all the fuel to the mission Starship (leaving just enough fuel left for the tanker to return and land) and 2) that transfer fully tops off the mission Starship's tanks.

If you violate condition 1 and the tanker has extra fuel left over, you could have used fewer tankers to fill it in LEO, or (if it's a fractional tanker load) you could have staged in a higher orbit and put more delta-v into the mission tanker.

If you violate condition 2 and the mission Starship tanks aren't completely filled, you could have staged in a lower orbit and wasted less propellant dragging the tanker to a higher orbit than necessary.

Incidentally, these two conditions (which can be expressed mathematically) give you enough "knowns" that you're able to calculate exactly the best elliptical orbit to use for a given mission. Of course you still need propellant margins so you won't hit it exactly, but at least we can figure out where the bullseye is. :)


One last thing: what if you want more delta-v than you can get from two fully fueled vehicles in LEO? The fuel optimal strategy is to start with three fully fueled vehicles, raise orbit until one tanker can use all its fuel to completely fill the other two, then proceed as before. And you still have enough information to calculate the (two) intermediate orbits.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Norm38 on 07/28/2020 10:48 pm
There's a cool 2001ish architecture that comes out of that.  Surface Tankers want to stop in LEO and drop their cargo.  Orbital Tankers want to move out of LEO full  (and establish a supply line and the tankers can use SEP orbit raising and lowering to get back to LEO)

So you get a LEO fuel depot.  That has regular service. Then you need to launch Starships, and launch cargo.  Maybe crew launches on Dragon for a while?

So Starships could launch to LEO, maybe dock to take on crew and cargo, fuel to reach the tanker orbit, then refuel and go.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 07/28/2020 11:19 pm
There's a cool 2001ish architecture that comes out of that.  Surface Tankers want to stop in LEO and drop their cargo.  Orbital Tankers want to move out of LEO full  (and establish a supply line and the tankers can use SEP orbit raising and lowering to get back to LEO)

So you get a LEO fuel depot.  That has regular service. Then you need to launch Starships, and launch cargo.  Maybe crew launches on Dragon for a while?

So Starships could launch to LEO, maybe dock to take on crew and cargo, fuel to reach the tanker orbit, then refuel and go.
The problem with that is that there is no singular LEO. A LEO with a given inclination can only be reached  twice a day from any given launch site, assuming no launch direction restrictions and the ability to turn the pad around for both launches.And an equitorial LEO is unreachable by ANY launch site without a course correction burn. (A polar orbit can be reached by any site twice a day, but loses all benifit from launching  with the rotation of the earth) It wouldnt be one fuel depot/station in orbit, it would need to be dozens, which pretty much eliminates the benifits from a unified depot/station.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: aero on 07/28/2020 11:38 pm
So does that mean they need 12 LEO depots to launch every hour? By how much time could the Depots be separated such that the needed course correction would be acceptable? 6 seconds? 6 minutes? 1 hour? 3 hours? 6 hours? I know that they use instantaneous launch windows quit a bit. Is that absolutely required for tanker launches?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 07/28/2020 11:52 pm
So does that mean they need 12 LEO depots to launch every hour? By how much time could the Depots be separated such that the needed course correction would be acceptable? 6 seconds? 6 minutes? 1 hour? 3 hours? 6 hours? I know that they use instantaneous launch windows quit a bit. Is that absolutely required for tanker launches?

I believe the ISS rendezvous window was only 4 minutes, but the recycle time was greater so they just launched instantaneously at the best point in the window.

The problem is the RAAN. If your launch site isn't under the orbital plane, you either need to burn a lot of fuel, or wait a long time while your orbit precesses around. Both are really painful if your goal is efficient tanker utilization.

TheRadicalModerate worked out that a circular 263 km orbit would give you one launch opportunity per day without the need for orbital phasing (this shortens rendezvous time substantially), so that would be a good candidate orbit for an accumulation tanker. A 555 km circular orbit gives two opportunities per day, but the debris risk is a lot higher, you're dragging the tanker dry mass higher so you get less propellant delivered per launch, and you lose some Oberth effect on your burn.

https://forum.nasaspaceflight.com/index.php?topic=50851.msg2080345#msg2080345
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 07/29/2020 03:19 am
A side-ish note on launch opportunities, there are potentially a lot more than one or two launch opportunities for a given depot orbit in LEO if there are more than one launchpads.  I know there's a lot (read: loooooooot) of skepticism about the E2E plans for Starship, but....  that _would_ mean there's a lot of potential places that can be chucking fueling tankers up to meet at depots. 

Even if E2E specifically doesn't happen, an active orbital site at Boca Chica, KSC, and then however many offshore launch pads they may or may not be working on right now might all contribute themselves to multiple launch opportunities each day.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/29/2020 08:25 am
Interesting but seems very very complexe,to add to the difficulty this will have to be done 5 times to get to Mars....................can't we just attach an external tank like for space shuttle?
(https://www.fourmilab.ch/autofile/e5/figures/external_tank.jpg)

I think you are missing two points that many others are trying to make, in different ways:

Point 1:

How do these tanks get into orbit?

These fully fuelled tanks, holding 5 launches worth of fuel.

Without refuelling.

Point 2:

What makes an external tank superior to the internal tanks already carried within each Starship?

An external tank that has to dock and be reliably attached and plumbed into the Mars-going Starship.

Versus the internal tanks of the same capacity already built into the Starship.

Just imagine the space shuttle without orbiter (only small engines on the external tank) how much fuel would be saved on the external tank?

-An external has not to do EDL or ascent again he could be jettisoned when empty.

-Imagine the fuel saved on getting rid of the extra weight of the tankers to get back to earth.

-Time saving on refueling.
-Saving on ullage fuel.
-Technologie available.

I get it, I've had ideas I was attached too that the group had to dissuade me from.  However, your idea is missing one single and totally not-negotiable item.  The system has to be 100% re-usable because that's what the company owner (Elon Musk) has set as a requirement.

So modify your concept of using the external fuel tank from a shuttle to include re-use.  Make whatever changes are needed for it to re-enter, be recovered, be refurbished, and then used a 2nd time and see where that leads you.  Eventually, it leads to something like Starship.

Any concept that isn't 100% re-usable isn't worth discussing because it doesn't meet the criteria required by the owner/operator of the project.

I was thinking that the primary goal for EM was to reach Mars in the 2024? so an interim solution is needed to keep the goal of 2024 credible.

The aim of reusability is to keep space flight cheap, I dont think aircrafts fuel drop tanks are that expensive, it must be some point where the cost of fuel to get 1kg of hardware to space will be equal (or more) to the cost of one kg of hardware, reaching that point is for me the objectif to keep costs relatively low, in the end a tank is only some sort of package.


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 07/29/2020 10:15 am
It seems that using the heavy as a tanker doesn't work very well. I've calculated (using the rocket equation) that the heavy without a Starship could "probably" make it to orbit, there doesn't seem to be much leeway for residual prop upon reaching orbit. I do think that using more accurate data and math, some residual prop to orbit could be achieved but requiring an expendable heavy really precludes the approach.

The key factor in my calculation is that the heavy leaves the launch pad with a 2 to 1 thrust to weight ratio. High acceleration limits gravity losses tremendously. I calculate, using constant ISP = 340 seconds, that the heavy can achieve 8.5 km/sec delta-V. It should do better because ISP increases as the rocket leaves the sensible atmosphere and at ISP = 360 seconds, the heavy could achieve delta-V = 9.1 km/sec. A better estimate is somewhere between the two extremes.

The above still fails utterly as a tanker though, because the heavy reaches orbit with no residual prop and is incapable of returning to Earth.

It would work as a fuel depot if there were reasons to do so.

What dry mass did you assume? Because my calc shows about 9.5km/s dV which should be plenty to reach orbit even witch much more moderate TWR.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 07/29/2020 11:15 am
It seems that using the heavy as a tanker doesn't work very well. I've calculated (using the rocket equation) that the heavy without a Starship could "probably" make it to orbit, there doesn't seem to be much leeway for residual prop upon reaching orbit. I do think that using more accurate data and math, some residual prop to orbit could be achieved but requiring an expendable heavy really precludes the approach.

The key factor in my calculation is that the heavy leaves the launch pad with a 2 to 1 thrust to weight ratio. High acceleration limits gravity losses tremendously. I calculate, using constant ISP = 340 seconds, that the heavy can achieve 8.5 km/sec delta-V. It should do better because ISP increases as the rocket leaves the sensible atmosphere and at ISP = 360 seconds, the heavy could achieve delta-V = 9.1 km/sec. A better estimate is somewhere between the two extremes.

The above still fails utterly as a tanker though, because the heavy reaches orbit with no residual prop and is incapable of returning to Earth.

It would work as a fuel depot if there were reasons to do so.

What dry mass did you assume? Because my calc shows about 9.5km/s dV which should be plenty to reach orbit even witch much more moderate TWR.
It can reach orbit as a massive depot, but it cant SSTO significant fuel mass the way Stardust wants.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 07/29/2020 11:52 am
I was thinking that the primary goal for EM was to reach Mars in the 2024? so an interim solution is needed to keep the goal of 2024 credible.
Do you have any evidence at all for this statement (an interim solution being needed)? Because if you don't, people will simply ignore it.

Quote
The aim of reusability is to keep space flight cheap, I dont think aircrafts fuel drop tanks are that expensive, it must be some point where the cost of fuel to get 1kg of hardware to space will be equal (or more) to the cost of one kg of hardware, reaching that point is for me the objectif to keep costs relatively low, in the end a tank is only some sort of package.
What on earth makes you think that the primary cost in your idea is the material cost of the tank?
Please, please re-read all the problems with your idea that people have pointed out above.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 07/29/2020 11:54 am
It seems that using the heavy as a tanker doesn't work very well. I've calculated (using the rocket equation) that the heavy without a Starship could "probably" make it to orbit, there doesn't seem to be much leeway for residual prop upon reaching orbit. I do think that using more accurate data and math, some residual prop to orbit could be achieved but requiring an expendable heavy really precludes the approach.

The key factor in my calculation is that the heavy leaves the launch pad with a 2 to 1 thrust to weight ratio. High acceleration limits gravity losses tremendously. I calculate, using constant ISP = 340 seconds, that the heavy can achieve 8.5 km/sec delta-V. It should do better because ISP increases as the rocket leaves the sensible atmosphere and at ISP = 360 seconds, the heavy could achieve delta-V = 9.1 km/sec. A better estimate is somewhere between the two extremes.

The above still fails utterly as a tanker though, because the heavy reaches orbit with no residual prop and is incapable of returning to Earth.

It would work as a fuel depot if there were reasons to do so.

What dry mass did you assume? Because my calc shows about 9.5km/s dV which should be plenty to reach orbit even witch much more moderate TWR.
It can reach orbit as a massive depot, but it cant SSTO significant fuel mass the way Stardust wants.

That's for sure, no SSTO is good for bringing up significant mass in one go.

Stardust's proposal makes no sense whatsoever and we'd better let it die, but I don't think this subthread is related. I think it's rather about accumulation tanker or Starpusher stuff.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 07/29/2020 01:10 pm
So does that mean they need 12 LEO depots to launch every hour? By how much time could the Depots be separated such that the needed course correction would be acceptable? 6 seconds? 6 minutes? 1 hour? 3 hours? 6 hours? I know that they use instantaneous launch windows quit a bit. Is that absolutely required for tanker launches?

I believe the ISS rendezvous window was only 4 minutes, but the recycle time was greater so they just launched instantaneously at the best point in the window.

The problem is the RAAN. If your launch site isn't under the orbital plane, you either need to burn a lot of fuel, or wait a long time while your orbit precesses around. Both are really painful if your goal is efficient tanker utilization.

TheRadicalModerate worked out that a circular 263 km orbit would give you one launch opportunity per day without the need for orbital phasing (this shortens rendezvous time substantially), so that would be a good candidate orbit for an accumulation tanker. A 555 km circular orbit gives two opportunities per day, but the debris risk is a lot higher, you're dragging the tanker dry mass higher so you get less propellant delivered per launch, and you lose some Oberth effect on your burn.

https://forum.nasaspaceflight.com/index.php?topic=50851.msg2080345#msg2080345

Phasing requires VERY little fuel.
Phasing would tie up an orbiter for a longer time.
So either you have enough orbiters and don't worry about them being in orbit longer or you launch once per day.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 07/29/2020 01:30 pm
I was thinking that the primary goal for EM was to reach Mars in the 2024?
Nope - you're thinking here is just wrong.  EMs overarching goal is to safeguard the long-term future of human civilization.  A subordinate objective of this is to make human civilization multi-planetary.  Humans on Mars in 2024?  That would be great, but if its only flags-and-footprints (which is the very best you'll ever get with your idea), then Musk isn't interested.  Anything less than full colonization would be considered a failure.  Musk is in a hurry, but I think he'd much rather do it right than do it now.


so an interim solution is needed to keep the goal of 2024 credible.
Nope - see above.

The aim of reusability is to keep space flight cheap, I dont think aircrafts fuel drop tanks are that expensive, it must be some point where the cost of fuel to get 1kg of hardware to space will be equal (or more) to the cost of one kg of hardware, reaching that point is for me the objectif to keep costs relatively low, in the end a tank is only some sort of package.
As others have pointed out, your idea has been well and truly shredded here.  Please just let it go.  Your enthusiasm is commendable, but just as important as coming up with new ideas is knowing when to move on from an unworkable concept. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 07/29/2020 01:51 pm
I was thinking that the primary goal for EM was to reach Mars in the 2024? so an interim solution is needed to keep the goal of 2024 credible.
Do you have any evidence at all for this statement (an interim solution being needed)? Because if you don't, people will simply ignore it.

Quote
The aim of reusability is to keep space flight cheap, I dont think aircrafts fuel drop tanks are that expensive, it must be some point where the cost of fuel to get 1kg of hardware to space will be equal (or more) to the cost of one kg of hardware, reaching that point is for me the objectif to keep costs relatively low, in the end a tank is only some sort of package.
What on earth makes you think that the primary cost in your idea is the material cost of the tank?
Please, please re-read all the problems with your idea that people have pointed out above.

I have got one solid piece of evidence; refueling plans are gonna take a lot of time to be validated if ever financially viable option, regarding the cost of tankers fleet O&M costs, the engeering technical complexity.

A tank is by definition a large receptacle or storage chamber, it's all about materials (insulation, residtance, weight,........), I know there is some difficulties regarding boil-off, high pressure, rendevous, engines/thrusters but still there is no high value parts in it like engines/avionics/habitat.

When you travel light you will consume less fuel this is the idea, interim solution of course until we get advanced technologies that will allow ISRU, until we get  costs reduction per flights, which is still not in the foreseeable future.

I'm pretty sure if this option has been proposed by Elon a lot of educated experts here would have considered a genius idea and absolute truth but my name is not EM. ;D
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: t3kboi on 07/29/2020 02:36 pm
I can't do the math myself.  (Not lazy, just literally cannot do the math...)


Assumptions -
A Starship tanker has exactly the same configuration as SN5/6/8 we see being built now.
The Payload section has the same configuration as well.

There are two tanker types -
  Tanker 1 - delivery tanker.
   The Payload - additional cylindrical or spherical tankage up to the limits of either volume or mass.

  Tanker 2 - linger-on-orbit tanker (accepts deliveries, and performs final transfer to prime mission Starship)
   Tanker 1 Payload + cryo-coolers plumbed to both the Payload and main tankage.

*caveat - production Starships may ALL have the cryo-coolers, since that is part of operational necessity.  They may be removed from the delivery tankers to swap cryo-cooler mass for payload fuel mass.

Questions -

Given the known main tank volume and projected tankage volume available to the nose - how much can a tanker deliver to orbit? 

Total = (Main tank residuals - landing reserve) + Payload tank volume

How many refueling trips does it take to completely fill (assumed completely empty) prime mission Starship?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 07/29/2020 04:57 pm
I was thinking that the primary goal for EM was to reach Mars in the 2024? so an interim solution is needed to keep the goal of 2024 credible.
Do you have any evidence at all for this statement (an interim solution being needed)? Because if you don't, people will simply ignore it.

Quote
The aim of reusability is to keep space flight cheap, I dont think aircrafts fuel drop tanks are that expensive, it must be some point where the cost of fuel to get 1kg of hardware to space will be equal (or more) to the cost of one kg of hardware, reaching that point is for me the objectif to keep costs relatively low, in the end a tank is only some sort of package.
What on earth makes you think that the primary cost in your idea is the material cost of the tank?
Please, please re-read all the problems with your idea that people have pointed out above.

I have got one solid piece of evidence; refueling plans are gonna take a lot of time to be validated if ever financially viable option, regarding the cost of tankers fleet O&M costs, the engeering technical complexity.

A tank is by definition a large receptacle or storage chamber, it's all about materials (insulation, residtance, weight,........), I know there is some difficulties regarding boil-off, high pressure, rendevous, engines/thrusters but still there is no high value parts in it like engines/avionics/habitat.

When you travel light you will consume less fuel this is the idea, interim solution of course until we get advanced technologies that will allow ISRU, until we get  costs reduction per flights, which is still not in the foreseeable future.

I'm pretty sure if this option has been proposed by Elon a lot of educated experts here would have considered a genius idea and absolute truth but my name is not EM. ;D
All you've done there is demonstrated that you don't understand what "evidence" means. Have a look at some other ideas being discussed on these forums and see how other people provide calculations or references to support their assertions, when they want them to be considered anything other than an opinion.  Suggesting that the objections are about you personally really isn't going to help your case, either.

Really, though, it's time to drop this now.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 07/29/2020 08:30 pm
So does that mean they need 12 LEO depots to launch every hour? By how much time could the Depots be separated such that the needed course correction would be acceptable? 6 seconds? 6 minutes? 1 hour? 3 hours? 6 hours? I know that they use instantaneous launch windows quit a bit. Is that absolutely required for tanker launches?

I believe the ISS rendezvous window was only 4 minutes, but the recycle time was greater so they just launched instantaneously at the best point in the window.

The problem is the RAAN. If your launch site isn't under the orbital plane, you either need to burn a lot of fuel, or wait a long time while your orbit precesses around. Both are really painful if your goal is efficient tanker utilization.

TheRadicalModerate worked out that a circular 263 km orbit would give you one launch opportunity per day without the need for orbital phasing (this shortens rendezvous time substantially), so that would be a good candidate orbit for an accumulation tanker. A 555 km circular orbit gives two opportunities per day, but the debris risk is a lot higher, you're dragging the tanker dry mass higher so you get less propellant delivered per launch, and you lose some Oberth effect on your burn.

https://forum.nasaspaceflight.com/index.php?topic=50851.msg2080345#msg2080345

Phasing requires VERY little fuel.
Phasing would tie up an orbiter for a longer time.

Indeed. "This shortens rendezvous time substantially." I conspicuously said nothing about fuel savings. ;)

So either you have enough orbiters and don't worry about them being in orbit longer or you launch once per day.

The implication, in other words, is that you can have multiple launches per day from the same launch pad rendezvousing with the same tanker, if you use orbital phasing.

This tripped my up at first too. However, it's not correct. :(

Again, the problem is RAAN. If your tankers aren't launching simultaneously, they'll be out-of-plane with the tanker. The only way it would work is if you launch those multiple tankers simultaneously, which would require multiple pads. So you still can't exceed 1 launch (or 2, if you use the problematic higher orbit) per pad per day to the same tanker.

You can certainly have multiple pads on the same launch site, of course. I expect SpaceX will do this.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 07/29/2020 08:38 pm
So does that mean they need 12 LEO depots to launch every hour? By how much time could the Depots be separated such that the needed course correction would be acceptable? 6 seconds? 6 minutes? 1 hour? 3 hours? 6 hours? I know that they use instantaneous launch windows quit a bit. Is that absolutely required for tanker launches?

I believe the ISS rendezvous window was only 4 minutes, but the recycle time was greater so they just launched instantaneously at the best point in the window.

The problem is the RAAN. If your launch site isn't under the orbital plane, you either need to burn a lot of fuel, or wait a long time while your orbit precesses around. Both are really painful if your goal is efficient tanker utilization.

TheRadicalModerate worked out that a circular 263 km orbit would give you one launch opportunity per day without the need for orbital phasing (this shortens rendezvous time substantially), so that would be a good candidate orbit for an accumulation tanker. A 555 km circular orbit gives two opportunities per day, but the debris risk is a lot higher, you're dragging the tanker dry mass higher so you get less propellant delivered per launch, and you lose some Oberth effect on your burn.

https://forum.nasaspaceflight.com/index.php?topic=50851.msg2080345#msg2080345

Phasing requires VERY little fuel.
Phasing would tie up an orbiter for a longer time.

Indeed. "This shortens rendezvous time substantially." I conspicuously said nothing about fuel savings. ;)

So either you have enough orbiters and don't worry about them being in orbit longer or you launch once per day.

The implication, in other words, is that you can have multiple launches per day from the same launch pad rendezvousing with the same tanker, if you use orbital phasing.

This tripped my up at first too. However, it's not correct. :(

Again, the problem is RAAN. If your tankers aren't launching simultaneously, they'll be out-of-plane with the tanker. The only way it would work is if you launch those multiple tankers simultaneously, which would require multiple pads. So you still can't exceed 1 launch per pad per day to the same tanker.

You can certainly have multiple pads on the same launch site, of course. I expect SpaceX will do this.

If the inclination of the orbit is greater than your latitude you can get 2 opportunities. One on the ascending node and one on the descending node. Not that I think it is worth it though. Better to keep inclination the same as latitude and keep the earth rotation benefit. You could always put 2 pads at one launch facility. Imagine dual SH/SS's launching at the same time from Boca Chica. :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 07/29/2020 08:48 pm
[snip]

Phasing requires VERY little fuel.
Phasing would tie up an orbiter for a longer time.

Indeed. "This shortens rendezvous time substantially." I conspicuously said nothing about fuel savings. ;)

So either you have enough orbiters and don't worry about them being in orbit longer or you launch once per day.

The implication, in other words, is that you can have multiple launches per day from the same launch pad rendezvousing with the same tanker, if you use orbital phasing.

This tripped my up at first too. However, it's not correct. :(

Again, the problem is RAAN. If your tankers aren't launching simultaneously, they'll be out-of-plane with the tanker. The only way it would work is if you launch those multiple tankers simultaneously, which would require multiple pads. So you still can't exceed 1 launch per pad per day to the same tanker.

You can certainly have multiple pads on the same launch site, of course. I expect SpaceX will do this.

If the inclination of the orbit is greater than your latitude you can get 2 opportunities. One on the ascending node and one on the descending node. Not that I think it is worth it though. Better to keep inclination the same as latitude and keep the earth rotation benefit.

Yes, that's where you'd use the higher (555 km) orbit. But it has the aforementioned problems.

You could always put 2 pads at one launch facility. Imagine dual SH/SS's launching at the same time from Boca Chica. :)

Indeed. Didn't I say that?  8)  But since most of the expenses are per-pad instead of per-launch, there's not a lot of savings to be had. Non-zero, but not a lot.

Multiple pads on the same site might share payload processing facilities, but each pad needs its own foundation, flame trench, deluge system, propellant tank farm (or at least one farm with twice the capacity, "six of one half-dozen of the other"), GSE, etc.

Again I expect SpaceX will do it, but unfortunately it's not a buy-one-get-one-free situation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 07/29/2020 11:11 pm
[snip]

Phasing requires VERY little fuel.
Phasing would tie up an orbiter for a longer time.

Indeed. "This shortens rendezvous time substantially." I conspicuously said nothing about fuel savings. ;)

So either you have enough orbiters and don't worry about them being in orbit longer or you launch once per day.

The implication, in other words, is that you can have multiple launches per day from the same launch pad rendezvousing with the same tanker, if you use orbital phasing.

This tripped my up at first too. However, it's not correct. :(

Again, the problem is RAAN. If your tankers aren't launching simultaneously, they'll be out-of-plane with the tanker. The only way it would work is if you launch those multiple tankers simultaneously, which would require multiple pads. So you still can't exceed 1 launch per pad per day to the same tanker.

You can certainly have multiple pads on the same launch site, of course. I expect SpaceX will do this.

If the inclination of the orbit is greater than your latitude you can get 2 opportunities. One on the ascending node and one on the descending node. Not that I think it is worth it though. Better to keep inclination the same as latitude and keep the earth rotation benefit.

Yes, that's where you'd use the higher (555 km) orbit. But it has the aforementioned problems.
Technically you can use both oppotunities even at the lower orbit. But only one opportunity would benefit from the perfect rendezvous- the other would require a predictable level of phasing to reach the accumulation tanker.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: john smith 19 on 08/02/2020 03:05 pm

The cryocoolers on an accumulation tanker (if they will have such coolers) will be designed to keep the already cold propellant cold, presumably thus handling liquids, and have radiators sized and designed for operation in vacuum, with significant influx of heat radiation from the Sun and Earth.  Cryocoolers for Mars ISRU will need to actually chill the propellants from maybe -60°C (if they are pre-cooled to ambient Mars temperatures) and in gas form, down to more than -180°C and liquid, and their radiators will want to be optimized for atmospheric use (thin though it may be).  I'm not sure how much difference this makes to their designs, though.
Quite a lot. What people normally call "radiators" are really "convectors." Gases are usually pretty poor conductors of heat and don't have much heat capacity but they can move quickly, extracting substantial heat from a surface. Vacuum OTOH leaves you solely with radiation and the stefan-boltzman law on black bodies. doubling the temperature increase radiator output by 16, but the size of the constants is so tiny the power per unit area is tiny. (IIRC the ISS radiators radiate 47w/Sq m. Improving radiator efficiency in vacuum could give serious mass savings) So a "radiator" on earth, or even on mars, can be quite small relative to one that has to operate in vacuum solely by actual radiation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: john smith 19 on 08/02/2020 03:18 pm
A tank is by definition a large receptacle or storage chamber, it's all about materials (insulation, residtance, weight,........), I know there is some difficulties regarding boil-off, high pressure, rendevous, engines/thrusters but still there is no high value parts in it like engines/avionics/habitat.
Except it's not just a tank, is it?

It's a whole different mission architecture.

You're right that Musk would like to have something land on mars by 2024. However he does not expect it to be carrying crew and it doesn't have to be the full 150 tonne payload. 
Quote from: Star-Dust
I'm pretty sure if this option has been proposed by Elon a lot of educated experts here would have considered a genius idea and absolute truth but my name is not EM. ;D
I'm pretty sure people would be asking themselves what the hell happened to "Rapid and fully  reusable launch systems" and the goal of sustainable settlement.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: john smith 19 on 08/02/2020 03:25 pm
At work I outlawed the use of the words "just" and "simply" when proposing alternative ideas.

You're not allowed to grade your own work...
Excellent point.

I'd call it "Originator bias."

At best you could say it's probably simple to do if  you'd had a fair bit of experience doing something like it already. If they don't have that experience then it's just their (biased) opinion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DusanC on 08/03/2020 07:03 am

The cryocoolers on an accumulation tanker (if they will have such coolers) will be designed to keep the already cold propellant cold, presumably thus handling liquids, and have radiators sized and designed for operation in vacuum, with significant influx of heat radiation from the Sun and Earth.  Cryocoolers for Mars ISRU will need to actually chill the propellants from maybe -60°C (if they are pre-cooled to ambient Mars temperatures) and in gas form, down to more than -180°C and liquid, and their radiators will want to be optimized for atmospheric use (thin though it may be).  I'm not sure how much difference this makes to their designs, though.
Quite a lot. What people normally call "radiators" are really "convectors." Gases are usually pretty poor conductors of heat and don't have much heat capacity but they can move quickly, extracting substantial heat from a surface. Vacuum OTOH leaves you solely with radiation and the stefan-boltzman law on black bodies. doubling the temperature increase radiator output by 16, but the size of the constants is so tiny the power per unit area is tiny. (IIRC the ISS radiators radiate 47w/Sq m. Improving radiator efficiency in vacuum could give serious mass savings) So a "radiator" on earth, or even on mars, can be quite small relative to one that has to operate in vacuum solely by actual radiation.

Utilising Carnot cycle you could have much higher temperature at the radiator.

Selecting the working fluid would be a nice problem but IMHO if it could work in vacuum only employing radiation then on Mars with convection it's efficiency would just go up.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/09/2020 03:52 pm
Can the anti-boiloff system on the now empty landed Starships be fed gaseous propellants and condense them down?

I can't recall hearing anything about cryocoolers being standard equipment on Starships, even on the Mars-bound Starships.  As far as I have heard, the only systems for preventing boiloff during the Earth-Mars and Mars-Earth transits, will be insulation (header tanks inside the main tanks; the empy main tanks then act as a giant thermos bottle), and minimize sunshine hitting the tanks by pointing them away from the Sun.  This will lower the amount boiloff enough that there will be propellant left when they arrive, but it will not entirely prevent boiloff.  And it is definitely not enough to chill down the propellant.

To actually chill down oxygen and methane to condense into liquids, you either need active cooling, i.e. cryocoolers, or get much further out in our solar system.

So they will need to bring cryocoolers to Mars as normal cargo.

(It would make sense to have cryocoolers on accumulation tankers in LEO, though.  It is much more difficult to keep the propellant cool in LEO, as A) you will fill the main tanks, so there is no thermos bottle protecting the propellant, and B) there is no direction to point the ship to minimize incoming heat radiation, as the Earth takes up almost half the surroundings, reflecting sunlight onto you from almost everywhere.)

Thank you!

If there needs to be cryocoolers on accumulation tankers, could they be added as part of the spec for mars bound ships? If you have to send it anyway, preassembled and already functioning might be better than as cargo.
Poking my nose into a thread I haven't finished reading yet.


This is a good reason for dedicated accumulator/depot ships. Launch the cryocooler once and be done with it. Not the highest priority, but at some point the benefits outweigh the specialized build. A depot without boiloff = operational flexibility.


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/09/2020 08:04 pm
There's a cool 2001ish architecture that comes out of that.  Surface Tankers want to stop in LEO and drop their cargo.  Orbital Tankers want to move out of LEO full  (and establish a supply line and the tankers can use SEP orbit raising and lowering to get back to LEO)

So you get a LEO fuel depot.  That has regular service. Then you need to launch Starships, and launch cargo.  Maybe crew launches on Dragon for a while?

So Starships could launch to LEO, maybe dock to take on crew and cargo, fuel to reach the tanker orbit, then refuel and go.
The problem with that is that there is no singular LEO. A LEO with a given inclination can only be reached  twice a day from any given launch site, assuming no launch direction restrictions and the ability to turn the pad around for both launches.And an equitorial LEO is unreachable by ANY launch site without a course correction burn. (A polar orbit can be reached by any site twice a day, but loses all benifit from launching  with the rotation of the earth) It wouldnt be one fuel depot/station in orbit, it would need to be dozens, which pretty much eliminates the benifits from a unified depot/station.
True, there is no One LEO but is we're talking refueling we're talking about a payload destined for something other than LEO. I'm not a real orbital mechanic but from what I've worked out using vectors (ya gotta use what ya got) and from discussion on NSF, the higher you go the less energy needed for an orbital plane change.


If the payload ship is going to GEO, TLI or TMI launch to the orbital plane suggested by the launch site, which would most likely be the same plane as the depot. After tanking, boogie on in the same plane and change planes where it's cheap.


From discussion it sounds like a launch from BC needs a dog leg to avoid danger to population centers in case of abort so it sounds like there is a budget for maneuvering.


BC is 25.99 N, Kennedy is 28.39 N, a difference of 2.40 degrees. BC is further south and has a tad more energy from earth rotation plus it needs to dogleg anyway, so let it head out at 25.99 after the plane change.  AIUI from doing vectors, plane changes are cheap at small angles and increasingly expensive up to 90 or more deg. I could probably work up dV for any instantaneous plane change using vectors if I weren't so lazy.


Four fuel launches to LEO per day theoretically possible if I didn't trip over my pencil.


Phil


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/09/2020 11:14 pm
There's a cool 2001ish architecture that comes out of that.  Surface Tankers want to stop in LEO and drop their cargo.  Orbital Tankers want to move out of LEO full  (and establish a supply line and the tankers can use SEP orbit raising and lowering to get back to LEO)

So you get a LEO fuel depot.  That has regular service. Then you need to launch Starships, and launch cargo.  Maybe crew launches on Dragon for a while?

So Starships could launch to LEO, maybe dock to take on crew and cargo, fuel to reach the tanker orbit, then refuel and go.
The problem with that is that there is no singular LEO. A LEO with a given inclination can only be reached  twice a day from any given launch site, assuming no launch direction restrictions and the ability to turn the pad around for both launches.And an equitorial LEO is unreachable by ANY launch site without a course correction burn. (A polar orbit can be reached by any site twice a day, but loses all benifit from launching  with the rotation of the earth) It wouldnt be one fuel depot/station in orbit, it would need to be dozens, which pretty much eliminates the benifits from a unified depot/station.
True, there is no One LEO but is we're talking refueling we're talking about a payload destined for something other than LEO. I'm not a real orbital mechanic but from what I've worked out using vectors (ya gotta use what ya got) and from discussion on NSF, the higher you go the less energy needed for an orbital plane change.


If the payload ship is going to GEO, TLI or TMI launch to the orbital plane suggested by the launch site, which would most likely be the same plane as the depot. After tanking, boogie on in the same plane and change planes where it's cheap.


From discussion it sounds like a launch from BC needs a dog leg to avoid danger to population centers in case of abort so it sounds like there is a budget for maneuvering.


BC is 25.99 N, Kennedy is 28.39 N, a difference of 2.40 degrees. BC is further south and has a tad more energy from earth rotation plus it needs to dogleg anyway, so let it head out at 25.99 after the plane change.  AIUI from doing vectors, plane changes are cheap at small angles and increasingly expensive up to 90 or more deg. I could probably work up dV for any instantaneous plane change using vectors if I weren't so lazy.


Four fuel launches to LEO per day theoretically possible if I didn't trip over my pencil.


Phil

This is a great point.

If you look carefully into the trajectories F9 has historically used to reach GTO, you'll see that after launch they're actually at a slightly lower inclination than the launch site latitude (<1 degree), implying a small amount of dogleg maneuvering. During the apogee raise burn there's also a small inclination correction, with the remainder being done by the upper stage during the GEO insertion burn. This may seem counterintuitive at first, but if you do the math you'll find it actually achieves slightly higher efficiency than doing the entire plane change during the GEO insertion burn.

So with Starship you could easily rendezvous first with a tanker or depot staged at inclination ~= launch site latitude, then perform the apogee raise and optional GEO insertion in a nearly identical manner to a standard F9 trajectory.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/09/2020 11:46 pm
For launches to the Moon and Mars, wouldn't you always want to use the ecliptic plane? That is, wouldn't you want orbits inclined 23.4 degrees to the equator and in the plane of the ecliptic?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/10/2020 12:09 am
For launches to the Moon and Mars, wouldn't you always want to use the ecliptic plane? That is, wouldn't you want orbits inclined 23.4 degrees to the equator and in the plane of the ecliptic?

You generally want to launch from a higher inclination to avoid passing through the worst of the van Allen belt post-TMI.

There's no need for your parking orbit to be right on the ecliptic from an orbital mechanics perspective. It doesn't matter if you're a little above or below the ecliptic plane with respect to Earth's center of mass, you just account for that in your trajectory calculations.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/10/2020 05:03 pm
For launches to the Moon and Mars, wouldn't you always want to use the ecliptic plane? That is, wouldn't you want orbits inclined 23.4 degrees to the equator and in the plane of the ecliptic?
I've a hunch Twark shot over your head so I'll throw in some detail. I'll go crawl under a rock if you got it.


For a moon shot the intent might be a polar orbit (example only) so coming in along the ecliptic would be of no help. If you want a lunar orbit over the ecliptic, where you do the maneuver putting yourself in that plane has different propellant costs.


The minimum propellant costs is where the ship is moving slowest. Using a launch profile that illustrates the point, picture the ship in circular LEO at an inclination of 26 deg. Where this orbit crossed the ecliptic is where the correction burn has to be.


Doing the plane change at LEO, ship has to kill all velocity taking it out of the ecliptic and add a bit along the ecliptic to stay at orbital velocity. Then it has to orbit until its roughly opposite the moon and do a burn to, in effect, raise its perigee up to L1 plus a little bit more so it coasts through and on to the moon.


Or, it can stay at 26 deg and do the burn that will pass it through L1. When it gets there it will be passing through the ecliptic and moving slow. Compared to the plane change in the first example it will just be a couple of puffs.


There are some refinement possible but I think these examples illustrate the core differences.


Mars transit can play a similar game but without that handy L1, I'm not sure exactly where the most fuel efficient plane change is. Maybe Mars/Sun L1?


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 08/10/2020 05:25 pm
ELEO Starship

A LEO with a given inclination can only be reached  twice a day from any given launch site...

At Guiana Space Center you could launch efficiently to equatorial LEO ~ 15 times a day, if necessary.

Assuming SpaceX sublease, you could load 150 t cargo at GSC, or even hop that cargo from Boca Chica and then SH stack at GSC.  ELEO tankers would cycle from GSC.

Conceivably GSC and Alcântara Launch Center (http://www.parabolicarc.com/tag/alcantara-launch-center/) might compete for such a SpaceX sublease.

There are other benefits to ELEO, notably the potential for airliner-class radiation environment (https://forum.nasaspaceflight.com/index.php?topic=34036.msg1836275#msg1836275).  In ELEO, crews can undertake multiple long-duration missions without incurring career exposure penalty (https://www.nasa.gov/sites/default/files/atoms/files/space_radiation_ebook.pdf).

Image:  South Atlantic Anomaly.  ROSAT.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 08/11/2020 01:01 am
ELEO Starship

A LEO with a given inclination can only be reached  twice a day from any given launch site...

At Guiana Space Center you could launch efficiently to equatorial LEO ~ 15 times a day, if necessary.

Assuming SpaceX sublease, you could load 150 t cargo at GSC, or even hop that cargo from Boca Chica and then SH stack at GSC.  ELEO tankers would cycle from GSC.

Conceivably GSC and Alcântara Launch Center (http://www.parabolicarc.com/tag/alcantara-launch-center/) might compete for such a SpaceX sublease.

There are other benefits to ELEO, notably the potential for airliner-class radiation environment (https://forum.nasaspaceflight.com/index.php?topic=34036.msg1836275#msg1836275).  In ELEO, crews can undertake multiple long-duration missions without incurring career exposure penalty (https://www.nasa.gov/sites/default/files/atoms/files/space_radiation_ebook.pdf).

Image:  South Atlantic Anomaly.  ROSAT.


Alcantara might be more attractive for a US based company due to the technology protection agreement that was signed specifically for that spaceport.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/11/2020 01:04 am
Some thoughts on commercial refueling.


The SX refueling plan is innovative but idiosyncratic. I also think it can be made to work quite well. I also doubt any other launch provider could easily physically and operationally interface. They would have to latch on across 9m and have comparable fluid interconnects across what, eight meters? Then in the SX system the guest ship would have to provide ullage thrust. This is asking too much.


Commercial service would require an agnostic physical interface and maneuvering would most reasonably (opinion) be done by the depot.


New related thought: O2 is usually the more massive part of the propellant and it is common in all non hypergolics no matter what fuel they use. Tanking some O2 by itself would increase capabilities of other rockets.


Needing only one fluid connection, some creative engineering might find a way to dock asymmetrically without some gonzo structure being in the way of everything else. Obviously, I've got no idea how to do this.


Phil

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AntiSanity on 08/11/2020 02:30 am
One thing that popped into mind is that most vehicles design their fuel and oxidizer tank ratios for their specific fuel mixture ratios. Vehicles would either have tanks with ratios that aren't optimized for non-orbit refueling or, more likely, they launch with less than the maximum possible oxidizer load and fill up on oxidizer in orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/11/2020 03:01 am
One thing that popped into mind is that most vehicles design their fuel and oxidizer tank ratios for their specific fuel mixture ratios. Vehicles would either have tanks with ratios that aren't optimized for non-orbit refueling or, more likely, they launch with less than the maximum possible oxidizer load and fill up on oxidizer in orbit.
That's the idea. Second stage makes orbit with a lot more fuel than can be justified but then they top off the O2 and it all makes sense. Not as elegant and advantageous as complete refueling but better than none.


I doubt SX would ever stock anything but CH4 and O2 and there aren't that many rockets that use methane but that may change over the next decade.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TomH on 08/11/2020 03:47 am
IDK if this would simplify things or just make them more complex, but another possibility is to have two tanker variants: an O2 tanker and a CH4 tanker. Prop transfer on orbit might be simpler, but operational logistics on the ground would be more complicated.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: corpsband on 08/11/2020 12:04 pm
Seems to me the simplest solution is make a tanker/super heavy with enough payload capacity to fully refuel a single starship.  The eliminates all the complexity of multiple tanker launches, does not require the construction of depot, and fits the "less is more" principal.   Not sure exactly how large a stack would be necessary but surely it's less than 18m.  Once SS/SH is flying,  there is very little engineering required to upsize.  Most of the work now is (a) learning to manufacture a stainless steel ship as efficiently as possible (b) fine tuning the raptor design. 

A larger diameter tanker/SH neatly solves all the issues with refueling.  1 SS launch / 1 tanker launch / 1 rendezvous and you're on your way to wherever. 

Four or five launches and fuel transfers is fine to bootstrap the process but it is needlessly complex. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/11/2020 12:34 pm
Seems to me the simplest solution is make a tanker/super heavy with enough payload capacity to fully refuel a single starship.  The eliminates all the complexity of multiple tanker launches, does not require the construction of depot, and fits the "less is more" principal.   Not sure exactly how large a stack would be necessary but surely it's less than 18m.  Once SS/SH is flying,  there is very little engineering required to upsize.  Most of the work now is (a) learning to manufacture a stainless steel ship as efficiently as possible (b) fine tuning the raptor design. 

A larger diameter tanker/SH neatly solves all the issues with refueling.  1 SS launch / 1 tanker launch / 1 rendezvous and you're on your way to wherever. 

Four or five launches and fuel transfers is fine to bootstrap the process but it is needlessly complex.
No building an entirely new launch vehicle is needlessly complex. There would be a lot of issues at 18m diameter. Not necessary show stoppers but not insignificant. Whilst it might be inconvenient to have to make 4 or 5 launches, its a lot less inconvenient than designing a new rocket. I may be wrong - we shall see what SpaceX decides to do, they are the experts here.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Alvian@IDN on 08/11/2020 12:43 pm
Seems to me the simplest solution is make a tanker/super heavy with enough payload capacity to fully refuel a single starship.  The eliminates all the complexity of multiple tanker launches, does not require the construction of depot, and fits the "less is more" principal.   Not sure exactly how large a stack would be necessary but surely it's less than 18m.  Once SS/SH is flying,  there is very little engineering required to upsize.  Most of the work now is (a) learning to manufacture a stainless steel ship as efficiently as possible (b) fine tuning the raptor design. 

A larger diameter tanker/SH neatly solves all the issues with refueling.  1 SS launch / 1 tanker launch / 1 rendezvous and you're on your way to wherever. 

Four or five launches and fuel transfers is fine to bootstrap the process but it is needlessly complex.
Don't forget that the fact that they're many launches for refueling is ACTUALLY a good thing! Why you might be asking? Well because it can proves reliability & bringing cost down per launch of the system

I think there's no substantial complexity of reviving a rapid launch cadences & very fast rendezvous like in the past on Gemini 11 & its Agena target vehicle. With that fast pace they can do all of the 6-8 refuelings in a matter of a single week
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/11/2020 02:48 pm
Seems to me the simplest solution is make a tanker/super heavy with enough payload capacity to fully refuel a single starship.

Why wouldn't you then use that superheavy architecture as the mission ship as well?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/11/2020 03:05 pm
I could probably work up dV for any instantaneous plane change using vectors if I weren't so lazy.

Already done:

Change in velocity required for a plane change of angle phi in a circular orbit:
delta_V = 2 * sin(phi/2) * sqrt(GM/r)
Vc = sqrt(GM/r)
Hence: delta_V = 2 * sin(phi/2) * Vc

So a 5 degree plane change requires nearly 9% of orbital velocity. (Using an arbitrary 7km/s, that's 610m/s.) A 10 degree plane change requires 17% (1.2km/s.) A 45 degree plane change requires over 75% (5.4km/s.) 60 degree plane change requires 100%. At some point, you might as well land and re-launch.

Speaking of which:

This is a good reason for dedicated accumulator/depot ships. Launch the cryocooler once and be done with it. Not the highest priority, but at some point the benefits outweigh the specialized build. A depot without boiloff = operational flexibility.

Since one ship, fully fuelled on orbit, holds enough propellant to fully fuel one ship, then you are going to have one accumulator for each mission ship. ISTM that the architecture for accumulator/depot ships would be to pre-launch them into the preferred orbital plane of the target mission for each ship going to that target.

After refuelling its mission-ship, then the accumulator returns to Earth to reset for its next missions.

This month (well, last month), it's an accumulator for Mars missions. Next month, after a quick refurbish, it goes back up as an accumulator for a future Jupiter/Europa mission, or subs out another accumulator in the Moon fleet that's due for refurb...

IMO, you don't need or want a permanent depot, just cheap reusable tanker-variants with some extra mods. Hence you aren't concerned about orbital planes not lining up with orbital transfer windows (the usual objection to depot architecture), because you launched the accumulator(s) into the window you wanted for each specific mission. And since you have one accumulator for each mission-ship (including cargo ships), these things are going to be as common as dirt.

This also fits in with SpaceX's iterative development cycle. Early accumulators can be made from superseded last-gen Starships while the depot technology is being proven. Doesn't matter if you lose a few in early experiments, since they were going to be scrapped anyway.

[I always find it amusing that SpaceX's take on reusability leaves them with a bunch of low cost expendable hardware available.]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 08/11/2020 03:46 pm
Seems to me the simplest solution is make a tanker/super heavy with enough payload capacity to fully refuel a single starship.  The eliminates all the complexity of multiple tanker launches, does not require the construction of depot, and fits the "less is more" principal.   Not sure exactly how large a stack would be necessary but surely it's less than 18m.  Once SS/SH is flying,  there is very little engineering required to upsize.  Most of the work now is (a) learning to manufacture a stainless steel ship as efficiently as possible (b) fine tuning the raptor design. 

A larger diameter tanker/SH neatly solves all the issues with refueling.  1 SS launch / 1 tanker launch / 1 rendezvous and you're on your way to wherever. 

Four or five launches and fuel transfers is fine to bootstrap the process but it is needlessly complex.

If a normal 9m Starship can bring up 150 tonnes of propellant to orbit, then a special ship that can bring up 1200 tonnes would need to be 9m×sqrt(1200t/150t) ≈ 25.5 meter in diameter.  If you can increase thrust-per-engine-bell-area, you can make the rocket taller, and don't need to increase the diameter quite as much.

It might be possible to increase the amount of propellant to orbit by stretch the tanks into the nosecone, which should increase the tankage of the 9m Starship from 1200 tonnes to almost 2000 tonnes.  However, most of that 800 tonne increase would be consumed before reaching orbit.  Staging would need to happen some 20% earlier, and the second stage would need to make up the difference in Δv.  Not trivial to calculate how much more propellant you could bring to orbit, since that depends on gravity losses, which will increase when you increase the mass.  If you could get 300 tonnes of propellant to orbit with a tank-stretched 9m Starship, then a similar 18m ship could get 1200 tonnes to orbit.  But that to reach that, you likely need higher thrust density than the current SH/SS or Raptor (31 engines at 2MN each on SuperHeavy gives a thrust-to-weight ratio of 1.1 if you fill the Starship with 2000 tonnes of propellant, which is pretty bad).

"Surely less than 18m", no.  18 meter diameter is probably the minimum you can get away with.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 08/11/2020 04:01 pm
I may be stupid here.

Any orbit inclination in LEO should be able to inject into any orbit inclination on the moon, on mars, on etc.
What am I missing?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 08/11/2020 04:27 pm
I may be stupid here.

Any orbit inclination in LEO should be able to inject into any orbit inclination on the moon, on mars, on etc.
What am I missing?

Nothing.

To elaborate, there'd by very very slight dV penalty from equatorial LEO to high inclination lunar orbit, but it's so slight it vanishes in necessary performance margins.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 08/11/2020 04:49 pm
I may be stupid here.

Any orbit inclination in LEO should be able to inject into any orbit inclination on the moon, on mars, on etc.
What am I missing?

Nothing.

To elaborate, there'd by very very slight dV penalty from equatorial LEO to high inclination lunar orbit, but it's so slight it vanishes in necessary performance margins.

So just use one LEO orbit and avoid any plane changes for interplanetary missions. Interplanetary includes the moon.

And while we are at it. A Mars mission is possible from LEO with a full tank?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/11/2020 05:34 pm
Any orbit inclination in LEO should be able to inject into any orbit inclination on the moon, on mars, on etc.
[yes]
So just use one LEO orbit and avoid any plane changes for interplanetary missions. Interplanetary includes the moon.

Yes. The inclination is close to irrelevant. Transfer from a polar Earth orbit to lunar equatorial orbit if you want (not that there's a reason to do that.) Or from ISS orbit. Or from equatorial orbit. Escape trajectories are very forgiving. So just pick the inclination that works best for the launch site, or has some other logistical advantage (like using ISS for check-out, fit-out, crew xfer, etc.)

With one small quibble:
The orbital plane is relatively fixed as Earth moves around the sun, hence it will move in and out of sync with the ecliptic. So the orbital plane of a a particular LEO orbit will have two opportunities per year to launch into an interplanetary trajectory without a massive plane-change manoeuvre. If the target requires a transfer window outside of those two opportunities, you can't use that plane. You therefore need to pick a plane such that it will line up for the intended mission window.

It's not a huge burden, the mission window will last days (or rather, dozens of windows per day, for several days). But it means you aren't going to have a single LEO orbit that can be used as a staging ground for many different mission types and targets. Even for the same target, a plane that works for one window (eg, 2020 Mars window) won't work for the next (eg, 2022.) Each mission will need its own orbit (barring happy coincidences.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/11/2020 05:48 pm
Seems to me the simplest solution is make a tanker/super heavy with enough payload capacity to fully refuel a single starship.  The eliminates all the complexity of multiple tanker launches, does not require the construction of depot, and fits the "less is more" principal.   Not sure exactly how large a stack would be necessary but surely it's less than 18m.  Once SS/SH is flying,  there is very little engineering required to upsize.  Most of the work now is (a) learning to manufacture a stainless steel ship as efficiently as possible (b) fine tuning the raptor design. 

A larger diameter tanker/SH neatly solves all the issues with refueling.  1 SS launch / 1 tanker launch / 1 rendezvous and you're on your way to wherever. 

Four or five launches and fuel transfers is fine to bootstrap the process but it is needlessly complex.
Well, if this is based on a larger diameter SH it's not going to happen any time in the near future. Making a rocket longer is relatively easy. It's more of what you already have. Making it fatter is way more difficult. It's closer to a new rocket than a redesign.


If it's based on a longer SS, it's unclear that it could make orbit with any appreciable propellant payload. IIRC, somebody that knows how ran a calculation on this. It would also be a throw away. It'll never make it back.


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TomH on 08/11/2020 10:23 pm
Making a rocket longer is relatively easy. It's more of what you already have. Making it fatter is way more difficult. It's closer to a new rocket than a redesign.

If you make a rocket longer while maintaining the same diameter you are increasing mass while maintaining the same planar area for engine thrust. You either have to increase engine thrust or take a hit to your T/W (thrust to weight ratio). Any T/W has to be > 1 to attain liftoff. The lower the T/W, the greater your gravity losses. If you can cram in more engines, that's fine. If not, you have to increase prop flow rate and combustion chamber pressure. Sometimes that is possible, sometimes not. And when it is possible, it is not necessarily easy. Instead of a new rocket you may wind up with a new engine.

Having substantial T/W is critical on a S1. It is far less important on upper stages where iSP is a more significant factor.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 08/11/2020 10:53 pm
Haven't they already stretched the Starship and Superheavy and packed as many engines in the Superheavy as possible?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wila on 08/11/2020 11:57 pm
To anyone/everyone talking about doing one shot refueling with 18m+ diameter tanker to refuel a SS, why even make a tanker to carry up 1200+ tonnes of propellant when you could just make a slightly larger vehicle that could orbit the full 1500 tonnes of a fully fueled starship [1200t propellant, 150t dry, 150t payload] ???  - therefore not even bothering at all these "risky" docking and propellant transfer operations you are afraid of?

I thought the whole purpose of in space refueling is to not need to build a something that is over 75% of the way to just doing it directly.

Please explain your logic to me, because as I see it,as an individual vehicle it would then be an even simpler design without all the rendezvous/docking/propellant hardware.

[EDIT taker to tanker :)]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Redclaws on 08/12/2020 01:00 am
Because refueling is not primarily about LEO payload.  It’s about getting a rocket full of fuel in orbit, to carry that payload to high energy orbits.  If you build an extremely large rocket as your solution, you lose out hugely on mass fraction - it’s mostly empty, so there’s a huge amount of wasted tank and engine (etc) mass.

Build moderate sized rockets and refuel and you can both avoid a larger rocket - which is hard for a lot of reasons - and you get far, far better results to high energy trajectories.  (eg, fast Mars transfers)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: corpsband on 08/12/2020 03:42 am
Interesting to read the replies.  They fall into a couple of categories:

1) Multi-flight refueling is best. 

This ignores all the points I made about a specialized large tanker is more efficient.


2) Increasing the diameter mean starting from scratch on a new rocket design


Perhaps so but the number of tankers needed would be *far* smaller than the number of starships.   I'm also a bit skeptical that this is simply hand-waving.  SpaceX is learning to build in stainless but that knowledge doesn't evaporate when scaling up.  It may be a new design but the construction practices will already be in place.  A lot of the iteration in Boca Chica right now presumably centers on lowering manufacturing cost and maximizing manufacturing efficiency.  For a large fuel tanker neither of those need apply. 



3) If you have a system with that much payload,  why not just scrap the 9m design and use that for everything?

Because Starship is being optimized for MASS PRODUCTION.  In Elon's vision, SS's will be everywhere:  flying P2P , flying to the Moon, launching other payloads,  making trips to Mars.

The tankers could be a low volume, relatively high cost design very different from SS.  But the very fact that the design in NOT versatile (like SS) means simplifying a lot of things.  You basically get a second stage which is all tanks.  No provisions for adaptability.  No ECLSS.  No cargo doors.

Getting fuel to orbit efficiently is going to be a reality for a long time -- realistically until some sort of fusion engine is flown or perhaps space elevators become a reality.  Getting things out of Earth's gravity well will always be expensive in the foreseeable future. 

Oil supertankers are excessively large.  Yet we keep building them larger because they're efficient even if they are expensive to design and build.  The same applies to transporting fuel to LEO. Bigger is better.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/12/2020 03:47 am
Because refueling is not primarily about LEO payload.  It’s about getting a rocket full of fuel in orbit, to carry that payload to high energy orbits.  If you build an extremely large rocket as your solution, you lose out hugely on mass fraction - it’s mostly empty, so there’s a huge amount of wasted tank and engine (etc) mass.

Build moderate sized rockets and refuel and you can both avoid a larger rocket - which is hard for a lot of reasons - and you get far, far better results to high energy trajectories.  (eg, fast Mars transfers)
The "why dont you just" is about making 9m starship a 3rd stage, lifted to LEO fully fueled by a hypothetical 20m super-starship, with no orbital rendevus, docking or refueling needed.

If you're building a new rocket anyway, what does "launch 9m starship + launch 18m full fuel load" have over "Launch 9m starship+ Launch 9m fuel+ launch 9m fuel+ launch 9m fuel..." that isnt done even better by "Launch 20m starship and full fuel load for 9m 3rd stage in one launch."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: alang on 08/12/2020 04:13 am
If it's a given that cryogenic fuels are more difficult to manage due to 'boil off' which will be ameliorated during the cruise to Mars by having a 'tank within a tank', then would there be any advantage in having some large sun/earth shades parked in a few earth orbits so that initial earth orbit refuelling can be slower?
This is based on the assumption that such a solution would be relatively 'dumb' from a technical point of view compared to having fuel depots or dedicated tankers.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/12/2020 04:25 am
Well, maybe we need the 36m. Maximo Supremo booster lofting the 18m Semi Maximo second stage with the 12m Maximo Simpatico tanker. Launch the whole sucker To Mars on a rail gun and do away with ISRU. With some small operational modifications it could stop off in LEO and refuel SS.  :o
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TomH on 08/12/2020 10:12 pm
Interesting to read the replies.  They fall into a couple of categories:

1) Multi-flight refueling is best. 

This ignores all the points I made about a specialized large tanker is more efficient.

Sorry, but you are the one ignoring all the evidence that you are incorrect. With all factors considered, it just is not.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: corpsband on 08/12/2020 11:34 pm
Interesting to read the replies.  They fall into a couple of categories:

1) Multi-flight refueling is best. 

This ignores all the points I made about a specialized large tanker is more efficient.

Sorry, but you are the one ignoring all the evidence that you are incorrect. With all factors considered, it just is not.
Re-read the thread and that "evidence" is missing.  It amounts to "this is the ship we have" so it's the best we can do.   If you read my original post, this is a given at the start of real
 SS missions and for some period of time thereafter.  But it's not evidence that fueling up by 1/5th of a tank at a time is superior to fueling by "fill-er-up please" .   In fact I fully expect the multi-flight refueling to target a place-holder tanker (with regen equipment on board)  which will then fill the "mission" SS all at once.   So full-tank fueling will still be used. 

In any case,  I humbly request that you re-post or quote this additional evidence.  Saying "no it's not" is just a rebuttal without supporting arguments.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: r8ix on 08/13/2020 01:18 am
Interesting to read the replies.  They fall into a couple of categories:

1) Multi-flight refueling is best. 

This ignores all the points I made about a specialized large tanker is more efficient.

Sorry, but you are the one ignoring all the evidence that you are incorrect. With all factors considered, it just is not.
Re-read the thread and that "evidence" is missing.  It amounts to "this is the ship we have" so it's the best we can do.   If you read my original post, this is a given at the start of real
 SS missions and for some period of time thereafter.  But it's not evidence that fueling up by 1/5th of a tank at a time is superior to fueling by "fill-er-up please" .   In fact I fully expect the multi-flight refueling to target a place-holder tanker (with regen equipment on board)  which will then fill the "mission" SS all at once.   So full-tank fueling will still be used. 

In any case,  I humbly request that you re-post or quote this additional evidence.  Saying "no it's not" is just a rebuttal without supporting arguments.

Well, they’ve already mentioned depots, so yes, full tank refueling would seem to be on the agenda.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/13/2020 01:35 am
Interesting to read the replies.  They fall into a couple of categories:

1) Multi-flight refueling is best. 

This ignores all the points I made about a specialized large tanker is more efficient.

Sorry, but you are the one ignoring all the evidence that you are incorrect. With all factors considered, it just is not.
Re-read the thread and that "evidence" is missing.  It amounts to "this is the ship we have" so it's the best we can do.   If you read my original post, this is a given at the start of real
 SS missions and for some period of time thereafter.  But it's not evidence that fueling up by 1/5th of a tank at a time is superior to fueling by "fill-er-up please" .   In fact I fully expect the multi-flight refueling to target a place-holder tanker (with regen equipment on board)  which will then fill the "mission" SS all at once.   So full-tank fueling will still be used. 

In any case,  I humbly request that you re-post or quote this additional evidence.  Saying "no it's not" is just a rebuttal without supporting arguments.
How is a 18m single launch refueling of a 9m starship better than a 20m 3 stage rocket that puts a fully fueled 9m starship into orbit without any refueling at all?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Lars-J on 08/13/2020 01:44 am
Interesting to read the replies.  They fall into a couple of categories:

1) Multi-flight refueling is best. 

This ignores all the points I made about a specialized large tanker is more efficient.

Sorry, but you are the one ignoring all the evidence that you are incorrect. With all factors considered, it just is not.
Re-read the thread and that "evidence" is missing.  It amounts to "this is the ship we have" so it's the best we can do.   If you read my original post, this is a given at the start of real
 SS missions and for some period of time thereafter.  But it's not evidence that fueling up by 1/5th of a tank at a time is superior to fueling by "fill-er-up please" .   In fact I fully expect the multi-flight refueling to target a place-holder tanker (with regen equipment on board)  which will then fill the "mission" SS all at once.   So full-tank fueling will still be used. 

In any case,  I humbly request that you re-post or quote this additional evidence.  Saying "no it's not" is just a rebuttal without supporting arguments.
How is a 18m single launch refueling of a 9m starship better than a 20m 3 stage rocket that puts a fully fueled 9m starship into orbit without any refueling at all?

A smaller rocket is cheaper. Fewer stages are cheaper. Flying more often makes each flight cheaper. It seems rather obvious, but perhaps you are judging "better" by another metric than cost? What metric do you use for "better"?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/13/2020 01:50 am
Interesting to read the replies.  They fall into a couple of categories:

1) Multi-flight refueling is best. 

This ignores all the points I made about a specialized large tanker is more efficient.

Sorry, but you are the one ignoring all the evidence that you are incorrect. With all factors considered, it just is not.
Re-read the thread and that "evidence" is missing.  It amounts to "this is the ship we have" so it's the best we can do.   If you read my original post, this is a given at the start of real
 SS missions and for some period of time thereafter.  But it's not evidence that fueling up by 1/5th of a tank at a time is superior to fueling by "fill-er-up please" .   In fact I fully expect the multi-flight refueling to target a place-holder tanker (with regen equipment on board)  which will then fill the "mission" SS all at once.   So full-tank fueling will still be used. 

In any case,  I humbly request that you re-post or quote this additional evidence.  Saying "no it's not" is just a rebuttal without supporting arguments.
How is a 18m single launch refueling of a 9m starship better than a 20m 3 stage rocket that puts a fully fueled 9m starship into orbit without any refueling at all?

A smaller rocket is cheaper. Fewer stages are cheaper. Flying more often makes each flight cheaper. It seems rather obvious, but perhaps you are judging "better" by another metric than cost? What metric do you use for "better"?
Are those the only reasons? Because the program of record (9m tanker spam) has the 18m supertanker handilly beat, by that metric.

And the 3 stage design has the supertanker beat in all the ways the supertanker is better than the program or record, for only small increases in problems the supertanker already needs to solve.

So what exactly is the supertanker design best at, compared to both designs at once? (9m tanker spam, and 3 stage tankerless)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Lars-J on 08/13/2020 02:02 am
How is a 18m single launch refueling of a 9m starship better than a 20m 3 stage rocket that puts a fully fueled 9m starship into orbit without any refueling at all?

A smaller rocket is cheaper. Fewer stages are cheaper. Flying more often makes each flight cheaper. It seems rather obvious, but perhaps you are judging "better" by another metric than cost? What metric do you use for "better"?
Are those the only reasons? Because the program of record (9m tanker spam) has the 18m supertanker handilly beat, by that metric.

Indeed it does. Which is why don't expect to see an 18m supertanker for a veeeery long time. I'm not advocating for a 18m+ design, on the contrary. But perhaps you are confusing me for someone else.

And the 3 stage design has the supertanker beat in all the ways the supertanker is better than the program or record, for only small increases in problems the supertanker already needs to solve.

None of them are better than the program of record. And you just asserted that your "3 stage design" is better without anything to back it up with. So I ask again - By what metric? How is it better? Other than being like Apollo...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: corpsband on 08/13/2020 02:15 am
How is a 18m single launch refueling of a 9m starship better than a 20m 3 stage rocket that puts a fully fueled 9m starship into orbit without any refueling at all?
The supertanker is being proposed as the next logical step in increasing the efficiency of the 1000's of SS already on the drawing board.  SS is being designed to be mass-produced.  Supertanker is a higher cost,  low volume vehicle.   

If you want to go hypothetical,  SS is the Jeep.  It's tough, adaptable, and very useful.  It's also available in large numbers.   Tanker semi's are more efficient at carrying useful loads of fuel but they are specialized vehicles.  You don't go the grocery store, do some off-roading at the beach, and take the kids to school in the tanker truck.  Specific vehicles for specific use cases.

Bringing fuel to orbit is expensive and SSTO will never (at least with current fuels) get around it.  So long as you have lots of jeeps sitting around,  it makes sense being able to efficiently fuel them up.  Does that logical chain lead to fuel depots?  Perhaps it does.  Or maybe  a load and go tanker is more efficient than an on-orbit depot. 

But even if they aren't, fuel depots want tanker trucks to deliver fuel,  not jeeps. 

In the case of your super super heavy,  why not use that first stage to get a much larger vehicle to LEO and then refuel it with multiple flights.   The circle goes round and round until you literally can't build any larger.

To me at least it seems very rational to build fuel delivery vehicles which are much larger than the rest of your fleet.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/13/2020 02:41 am
I've been noodling SS based depots and was wondering if anybody has WAGS on some weights. This might fit better in the engineering thread but it's directly related to refueling so I ask some indulgence.


Here's numbers I've been working with but don't have that tight a feel for some of them. Ideas most welcome.
-Final leg design: 12t (we haven't seen them yet so this number is crude)
-Heatshield: 10t (I've seen this number within the last few months but new detail is emerging)
-Fins, mechanism, hull reinforcement: 18 t (a total WAG and probably too low)
-Cargo bay stringers: .75t/ring. (A WAG)
-Landing propellant: ??  (I have no idea)
-Top and bottom dome volumes: 1 ring equivalent (A stab in the dark)


Here's a couple of hard ones. A preliminary guesstimate allows a tank stretch of two rings. The top three rings could go away and save 7 tons which isn't worth trying to convert into another rings worth of tankage.


What sort of sun shade and deployment mechanism can be done for seven tons?
How much cryo cooler, PV, batteries and radiators, can 7t buy? What would average power consumption be?
Is there some combination of cooler and shade that would do a better job than either individually within a 7t budget?


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: r8ix on 08/13/2020 02:57 am
Not now. Eventually, perhaps; but as you noted, building the supertanker implies a heavier lift capacity, which means even cheaper $/kg to orbit, making larger vessels useful which would require multiple refueling trips…

For now, we will have 4-5 variations on a single basic design. Later, we may add an 18m version, with 3-4 variations. As a space ecosystem develops, and as those initial efforts reduce the cost of gettin to LEO (and higher) there will be greater opportunity and need for specialization. If Musk is right about StarShip launch costs, and if Supertanker development cost $1 billion, you’d have to eliminate 400 launches just to pay for the development costs. Full fuel-up in orbit under the current scenario would cost $15-20 million. If each supertanker launch got that down to $10 million, you’re looking at in the neighborhood of 1000 tanker launches (to be replaced by 125 supertanker launches) to break even, assuming no time cost of money, and no time loss to develop the supertanker…

So, like I said, maybe someday. But it doesn’t make sense now.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/13/2020 03:10 am
How is a 18m single launch refueling of a 9m starship better than a 20m 3 stage rocket that puts a fully fueled 9m starship into orbit without any refueling at all?
The supertanker is being proposed as the next logical step in increasing the efficiency of the 1000's of SS already on the drawing board.  SS is being designed to be mass-produced.  Supertanker is a higher cost,  low volume vehicle.   

If you want to go hypothetical,  SS is the Jeep.  It's tough, adaptable, and very useful.  It's also available in large numbers.   Tanker semi's are more efficient at carrying useful loads of fuel but they are specialized vehicles.  You don't go the grocery store, do some off-roading at the beach, and take the kids to school in the tanker truck.  Specific vehicles for specific use cases.

Bringing fuel to orbit is expensive and SSTO will never (at least with current fuels) get around it.  So long as you have lots of jeeps sitting around,  it makes sense being able to efficiently fuel them up.  Does that logical chain lead to fuel depots?  Perhaps it does.  Or maybe  a load and go tanker is more efficient than an on-orbit depot. 

But even if they aren't, fuel depots want tanker trucks to deliver fuel,  not jeeps. 

In the case of your super super heavy,  why not use that first stage to get a much larger vehicle to LEO and then refuel it with multiple flights.   The circle goes round and round until you literally can't build any larger.


To me at least it seems very rational to build fuel delivery vehicles which are much larger than the rest of your fleet.
But it ISNT significantly bigger than your 18m supertanker. it's literally 1200 tons payload vs 1470 tons payload, a difference of only 22%. This is in comparison with your existing jump of 800% to go from 150 tons payload to 1200 tons of payload.

A rocket that's only 22% larger removes all orbital docking and refueling events, compared to reducing the rendevus and refueling events to 1.

This is where comparisons to jeeps and such break down. When literally 90% of the vehical is propellant, bringing the extra 10% of non-propellant is nothing in comparison.

The advantage of the program of record is that everything is common- that tankers and cargo flights and crew flights can all use the same pads, the same boosters. That you maximize your benifits from mass production.

But if you are making the jump to a larger size, you might as well bring the whole thing, instead of just the fuel.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: corpsband on 08/13/2020 03:23 am
Not now. Eventually, perhaps; but as you noted, building the supertanker implies a heavier lift capacity, which means even cheaper $/kg to orbit, making larger vessels useful which would require multiple refueling trips…
...
So, like I said, maybe someday. But it doesn’t make sense now.
I agree and never suggested that a supertanker program be undertaken until after the 9m SS ecosystem is fully operational. 

Getting fuel to orbit is always going to be a major bottleneck until some significant breakthrough in propulsion technology occurs.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TomH on 08/13/2020 03:49 am
I humbly request that you re-post or quote this additional evidence.  Saying "no it's not" is just a rebuttal without supporting arguments.

I don't need to. The last few pages are full of evidence that rebut your ultra-tanker notion. You are the one ignoring evidence and reality.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/13/2020 04:13 am
How is a 18m single launch refueling of a 9m starship better than a 20m 3 stage rocket that puts a fully fueled 9m starship into orbit without any refueling at all?
The supertanker is being proposed as the next logical step in increasing the efficiency of the 1000's of SS already on the drawing board.  SS is being designed to be mass-produced.  Supertanker is a higher cost,  low volume vehicle.   

If you want to go hypothetical,  SS is the Jeep.  It's tough, adaptable, and very useful.  It's also available in large numbers.   Tanker semi's are more efficient at carrying useful loads of fuel but they are specialized vehicles.  You don't go the grocery store, do some off-roading at the beach, and take the kids to school in the tanker truck.  Specific vehicles for specific use cases.

Bringing fuel to orbit is expensive and SSTO will never (at least with current fuels) get around it.  So long as you have lots of jeeps sitting around,  it makes sense being able to efficiently fuel them up.  Does that logical chain lead to fuel depots?  Perhaps it does.  Or maybe  a load and go tanker is more efficient than an on-orbit depot. 

But even if they aren't, fuel depots want tanker trucks to deliver fuel,  not jeeps. 

In the case of your super super heavy,  why not use that first stage to get a much larger vehicle to LEO and then refuel it with multiple flights.   The circle goes round and round until you literally can't build any larger.

To me at least it seems very rational to build fuel delivery vehicles which are much larger than the rest of your fleet.
I see the logic in your position and will offer a practical reason it will not happen real soon.


SX has been working 11 months to get a nine meter rocket to hop. They need to develop a heat shield, refueling, a booster... I think you underestimate the difficulty of building a bigger rocket. Despite the fact it would probably be the same material and share construction techniques, it is very difficult. Big things do not scale linearly. Every aspect needs new calculations and trade off analysis. It's about as straight forward as figuring out SS.


Nothing now at BC is scaled to it, from the tents to the pads, the to the road between the build and launch site. There is barely room for doing the SS at BC. It would have to be another complete facility somewhere else. Started from scratch. Money is not infinite. Indeed what has already been done has been done on a shoestring by aerospace standards. Elon has to watch the purse very carefully.


The last announced intention was to build 100 rockets a year for 10 years. After two or three years of refueling experience Elon might say "Ok, enough of this crap. Let's build a big tanker." If this happens it will be because the pressure of immediate needs is gone and cash flow (from Starlink?) is improved. But he is a business man and he makes decisions based on cost effectiveness.


The super tanked is to be built in limited numbers. An equivalent of BC has to be built but bigger, and the R&D has to be done. Divide this by the number to be built (10?) and you have a base unit cost before any metal is bent. There will be little savings  through volume purchasing except for stainless and welding wire. Where the SS might cost 20 million each (including all R&D and capital outlay amortized over the fleet) the super tanker might be as high as three or four hundred million each. The big nut at the front end is spread over fewer ships.


Then the next question is, how much will it save? Even a launch facility substantially larger than any in existence, has to be built. What will it cost per launch? What is it's lifetime?


Only if it can do substantially more than a break even would it be considered.


Nobody here on NSF, including you and those who disagree, have these numbers. Elon probably has enough in his plate to keep him from putting much energy into the question but if he does decide to look at it, it is these numbers that will influence his decision.


For now Elon has to get SS running, launch some starlinks, go to mars and maybe go to the moon.


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: r8ix on 08/13/2020 05:08 am
Another point for consideration (given the ship analogy above), is that container ships, bulk cargo ships, and cruise ships are not significantly smaller than oil tankers. But they don’t have the “used fuel is 80-90% of total mass” issue.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: corpsband on 08/13/2020 12:25 pm

I see the logic in your position and will offer a practical reason it will not happen real soon.
...
Nobody here on NSF, including you and those who disagree, have these numbers. Elon probably has enough in his plate to keep him from putting much energy into the question but if he does decide to look at it, it is these numbers that will influence his decision.


For now Elon has to get SS running, launch some starlinks, go to mars and maybe go to the moon.


Phil
Well the very title of the thread invites speculation :)   We only recently have grain silo's hopping and we're discussing on-orbit refueling options!   In any case thanks for at least seeing the logic of my argument.   Getting fuel out of the gravity well is a huge barrier to putting a lot of mass on Mars (or the Moon). And it doesn't really take very many Mars-bound Starships to logistically overwhelm the current plan. 

If we're really going to have a self-supporting colony on Mars, a handful of SS's isn't going to cut it.  You need to ship a LOT of infrastructure-building cargo before that first 100 settlers arrive. 

Otherwise you're not doing much besides planting a flag. 

Personally I want to see the settlement not 10 guys in a tent camping on Mars. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: corpsband on 08/13/2020 12:30 pm
Another point for consideration (given the ship analogy above), is that container ships, bulk cargo ships, and cruise ships are not significantly smaller than oil tankers. But they don’t have the “used fuel is 80-90% of total mass” issue.
Indeed (although cruise ships are not really transportation per se -- they're floating resorts). Huge cargo ships will probably have an analog,  but a little further down the road.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 08/13/2020 01:09 pm
Sea Dragon was to be about 70'-75' in diameter or about 23m in diameter.  It was to be built at a shipyard that builds submarines. 

Anything over about 12m would require an entirely new infrastructure and launched at sea and land at sea.  Shear size and sonic boom wouldn't allow anything much bigger than 12m to be launched from any existing facilities.  Pads 39A and B were built to launch a Nova size (12m) rocket, nothing larger. 

I don't forsee an 18m ship for a long time.  Maybe a 12m.  12m is almost double the volume of a 9m Starship.  So almost double the payload.  An 18m Starship would give 4 times the capacity of a 9m starship. 

Reusing existing Starships (9m version) would allow for more launches for tanking a fuel depot (modified Starship tanker) to fill during the 18 month off synod between Starship flotillas going to Mars during the 6 month synod.  That way Starships going to Mars would only have to make one stop and fill up before going to Mars or the moon. 

Starship/Superheavy will be like Falcon 9 once fully operational.  Lots of launches and reuse.  Then SpaceX could work on a larger Starship/Superheavy, maybe a 12m-14m diameter.  18m seems a bit of a stretch. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/13/2020 02:17 pm
Interesting to read the replies.  They fall into a couple of categories:
1) Multi-flight refueling is best. 
This ignores all the points I made about a specialized large tanker is more efficient.

Based on emphasising the wrong kind of "efficiency".

3) If you have a system with that much payload,  why not just scrap the 9m design and use that for everything?
Because Starship is being optimized for MASS PRODUCTION.  In Elon's vision, SS's will be everywhere: [...]
The tankers could be a low volume, relatively high cost design very different from SS.  But the very fact that the design in NOT versatile (like SS) means simplifying a lot of things.  You basically get a second stage which is all tanks.  No provisions for adaptability.  No ECLSS.  No cargo doors.

Those things you list for the tanker are all bad things. It's more expensive to make (both per-unit and per-program), it launches less often (making it also much more expensive to operate), and it can't be used for anything else by design.

The whole business case for low cost Earth launch is based on flexible, rapidly launched, reusable ships that you use for everything. There might be a reason to replace Starship with a larger version, 12m, 16m, 18m, whatever SpaceX decides makes sense, but there's no good reason to have a single-purpose vehicle that is much more expensive than their work-horse.

"Expensive and rarely used" is the mindset that SpaceX was founded to get away from.

That said: There might be brief periods where SpaceX is building a larger Starship, where the basic launcher is flying, but not yet proven, where it is used as a fuel tanker and treated as essentially disposable, while SpaceX is iterating its design. But even then, a big tanker fuelling a previous generation Starship won't be the point of the design, and it won't be optimised for it. It's just a handy test mission. The moment the larger hardware is proven, it replaces that previous generation entirely as the general purpose workhorse launcher. They are not "jeeps", they are mass produced, standardised commercial trucks.

(Semi-trailer tankers are similarly based on standard general purpose semis. Designed to use standard infrastructure (roads). Even the trailer chassis are mostly standard, only the tanks themselves (and fuel handling equipment) are specialised. They are more like Starship tankers, a variant on the standard workhorse cargo vehicle.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: r8ix on 08/13/2020 02:27 pm
Another point for consideration (given the ship analogy above), is that container ships, bulk cargo ships, and cruise ships are not significantly smaller than oil tankers. But they don’t have the “used fuel is 80-90% of total mass” issue.
Indeed (although cruise ships are not really transportation per se -- they're floating resorts). Huge cargo ships will probably have an analog,  but a little further down the road.
Because we don't use ships for "transportation" any more. But the Queen Mary 2 was about the same size (345m), and the Harmony of the Seas is 362m. Long-term crewed vessels will have to be a little more "resort-like" than the ISS, and once we move past leading-edge exploration, resort vessels are not out of the question. Avenue 5, anyone?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wila on 08/16/2020 05:58 am
Firstly, just to clarify, I have never "advocated" a 3 stage Star Ship, I personally think that is a expensive Bad Idea.
But...
I do think it makes more sense than making a dedicated 20-22m diameter 1200t capacity SuperTanker AND the UltraSuperHeavy booster needed to loft it AND the launch site/GSE for all of it (which would likely have to be far out at sea [is it still GSE when at sea, not on the ground?]). And all that for the sole purpose of reducing what, by then, should be standard, well practiced and reliable rendezvous/docking/fuel transfer events from 5 or 6 down to 1, about an 80% risk reduction to this one part of the overall mission. If instead, by making this system 23m diameter and treat a fully fueled StarShip as cargo you eliminate the need for them all-together, giving a 100% risk reduction? (Okay, maybe 99.9% due to added stage seperation event, but that has got to be a lot less risky than even one rendezvous/docking/fuel transfer/undocking event.)

As I said though, I do not view it as a good idea, just better than a dedicated single purpose SuperTanker/UltraSuperHeavy setup.

Warning and disclaimer - all numbers used in the rest of this post are derived by very detailed arm waving, and then simplified to nicely rounded numbers - this is fully justifiable as even the great Elon himself knows not yet what these the numbers may come to be ;)

I do not believe an 18m diameter tanker could get 1200t propellant to orbit. As is, Raptor has an almost unbeatable thrust to nozzle area ratio.  As has be discussed to death several times  across multiple NSF forums, doubling the diameter only increases the area four time, so with the same thrust per area, you are only getting 4 times the thrust and therefor can only lift 4 times as much meaning your payload only increases by ~4 times, all else being equal (ISP, full-burnout mass ratio, etc.).   StarShip as described needs ~1200t propellant to fully refuel when empty.  The cargo version is to have ~150t cargo capacity.  The Tanker version has to be able to carry over 200t propellant to fully refuel a StarShip in 6 tanker runs, and over 240t to do it in 5 runs - the over bit being whatever amount is needed to compensate for boil-off. At the best value of 240t, 4 times that is only 960t, 240t short of a full load, so you'll need a second tanker run and associated rendezvous/docking/transfer/undocking event to fully fuel, and unless you can use the excess ~720t left in that second tanker to dock with and top off two or three more Starships it is a waste and you are still doing multiple rendezvous/docking/fuel transfer/undocking events.  So you need a SuperTanker larger than 18m.  Over 20m diameter if the ratios of a 240+t 5 run tanker are scaled, and almost 22m diameter if the 200+t 6 run tanker ratios are scaled, in order for the SuperTanker to be able to refuel a 9 m StarShip in just one rendezvous/docking/fuel transfer/undocking event.  And unless you've developed a new 5-6 times bigger engine, you are talking 155-186 or more current Raptors on that booster alone  :P .

Musk has said that 9m StarShip is what he considers to be the minimum needed to start setting up and populating a sizable (10,000 person plus) Mars BASE, but to get to the 1 million person colony level WILL need a bigger vehicle.  Mars trips will be the vast majority of flights that will need fully refueled StarShips in LEO.  The realistic need for fully refueled starships for GTO/DirectGEO/Lunar/DeepSpace  trips is only going to be in the few tens per year at very best for the next decade plus.  So these possibly 22m SuperTankers and associated facilities are going to be super busy 3 or 4 out of every 26 month Mars synodoic period, then just sit around almost unused till the next synod?  This is going to jack the cost of StarShip flights thru the roof in order to cover the costs of developing and maintaining such massive and overall seldom used system (shades of SLS).

By the time 9m SS and SH can't keep up with trips to mars, new bigger StarShip and associated booster will be needed anyway, and it will be sized to its needs, whether 12, 18 or 20+m diameter.

And if this specifically designed just to fill a 9m StarShip in one go SuperTanker/UltraHeavyBooster isn't usefull to it, said SuperTanker/UltraHeavy/GSE/Pad will just be a very expensive one trick pony.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/16/2020 05:06 pm
Firstly, just to clarify, I have never "advocated" a 3 stage Star Ship, I personally think that is a expensive Bad Idea.
But...
I do think it makes more sense than making a dedicated 20-22m diameter 1200t capacity SuperTanker AND the UltraSuperHeavy booster needed to loft it AND the launch site/GSE for all of it (which would likely have to be far out at sea [is it still GSE when at sea, not on the ground?]). And all that for the sole purpose of reducing what, by then, should be standard, well practiced and reliable rendezvous/docking/fuel transfer events from 5 or 6 down to 1, about an 80% risk reduction to this one part of the overall mission. If instead, by making this system 23m diameter and treat a fully fueled StarShip as cargo you eliminate the need for them all-together, giving a 100% risk reduction? (Okay, maybe 99.9% due to added stage seperation event, but that has got to be a lot less risky than even one rendezvous/docking/fuel transfer/undocking event.)

As I said though, I do not view it as a good idea, just better than a dedicated single purpose SuperTanker/UltraSuperHeavy setup.

Warning and disclaimer - all numbers used in the rest of this post are derived by very detailed arm waving, and then simplified to nicely rounded numbers - this is fully justifiable as even the great Elon himself knows not yet what these the numbers may come to be ;)

I do not believe an 18m diameter tanker could get 1200t propellant to orbit. As is, Raptor has an almost unbeatable thrust to nozzle area ratio.  As has be discussed to death several times  across multiple NSF forums, doubling the diameter only increases the area four time, so with the same thrust per area, you are only getting 4 times the thrust and therefor can only lift 4 times as much meaning your payload only increases by ~4 times, all else being equal (ISP, full-burnout mass ratio, etc.).   StarShip as described needs ~1200t propellant to fully refuel when empty.  The cargo version is to have ~150t cargo capacity.  The Tanker version has to be able to carry over 200t propellant to fully refuel a StarShip in 6 tanker runs, and over 240t to do it in 5 runs - the over bit being whatever amount is needed to compensate for boil-off. At the best value of 240t, 4 times that is only 960t, 240t short of a full load, so you'll need a second tanker run and associated rendezvous/docking/transfer/undocking event to fully fuel, and unless you can use the excess ~720t left in that second tanker to dock with and top off two or three more Starships it is a waste and you are still doing multiple rendezvous/docking/fuel transfer/undocking events.  So you need a SuperTanker larger than 18m.  Over 20m diameter if the ratios of a 240+t 5 run tanker are scaled, and almost 22m diameter if the 200+t 6 run tanker ratios are scaled, in order for the SuperTanker to be able to refuel a 9 m StarShip in just one rendezvous/docking/fuel transfer/undocking event.  And unless you've developed a new 5-6 times bigger engine, you are talking 155-186 or more current Raptors on that booster alone  :P .

Musk has said that 9m StarShip is what he considers to be the minimum needed to start setting up and populating a sizable (10,000 person plus) Mars BASE, but to get to the 1 million person colony level WILL need a bigger vehicle.  Mars trips will be the vast majority of flights that will need fully refueled StarShips in LEO.  The realistic need for fully refueled starships for GTO/DirectGEO/Lunar/DeepSpace  trips is only going to be in the few tens per year at very best for the next decade plus.  So these possibly 22m SuperTankers and associated facilities are going to be super busy 3 or 4 out of every 26 month Mars synodoic period, then just sit around almost unused till the next synod?  This is going to jack the cost of StarShip flights thru the roof in order to cover the costs of developing and maintaining such massive and overall seldom used system (shades of SLS).

By the time 9m SS and SH can't keep up with trips to mars, new bigger StarShip and associated booster will be needed anyway, and it will be sized to its needs, whether 12, 18 or 20+m diameter.

And if this specifically designed just to fill a 9m StarShip in one go SuperTanker/UltraHeavyBooster isn't usefull to it, said SuperTanker/UltraHeavy/GSE/Pad will just be a very expensive one trick pony.
Wow! Arm waving for sure, but half the trick to arm waving is finding the right numbers to wave over. I think ya dun good.


IMO, once refueling is proven and reliable, and the need is established, a custom built depot makes sense. It allows a steady tempo of launches over time instead of a flurry all at once. Between a sun shade and PV powered cryo cooling losses should be acceptable. A steady tempo with dedicated SS based tankers takes a lot of stress off launch and supply operations.


Keeping a Depot orbit at the inclination of BC or Kennedy (only ~2deg different) gives an opportunity to refuel missions to any location beyond LEO and with the correct mission profile, at minimum dV for a plane change. All without a massive one off SuperDuper booster.


Once there is a depot established for general, rather than mission specific use, the
small size might become an issue. Figure on a mars mission with two crewed and 10 cargo ships and a lot of refueling has to happen. Throw in a few GEO and lunar launches and it's a problem.


Down the road (2-3 mars missions?) there might be reason for a custom 12m hammerhead tank perched atop an SH with only enough propellant to make LEO. Total mass the same as SS but with no cargo, heatshield, fins or legs. Figure a nominal 100 tons dry mass for a standard SS and a guesstimate of 190 tons savings that can become dry mass and that tank can hold quite a bit of propellant.


I have no pencil or paper handy so my arm waving does not cover structural feasibility through launch. Probably not very good. Still, it can shrink back to 9m, include a cryo cooler, PV and a sunshade and some starter propellant and still be quite large. And a lot less hassle than mega launchers. A total WAG, 2-3 SS refueling.


Phil


Edit: I was editing for clarity and sanity when by battery quit, so here goes.
- the 190t savings came from legs, heatshield and fins: 40t estimate. No cargo =150t. These numbers are provisional.
- following that number in the main body where I spoke of the depot holding quite a bit of propellant I meant once on orbit. It would launch dry except for propellant to reach LEO.
-Converting the 190t savings into 1.6t rings allows 118 additional 9m rings with no launch mass change. Give it 33% penalty for additional structure and it's 79 additional rings. Make the depot 12m and it gets 52 rings. No way this is going to happen.
- The current SS HAS 9+ propellant rings. For every nine+ additional rings an SS can be topped off. What might be the longest it can be  safely stretched? Even with reinforcing and maybe greater ullage pressure. Would 12m make it easier? I doubt the engineering effort is worth it.
- Even with any reasonable stretch, more than one would be needed to refuel a modest Mars fleet. Does a long term depots impact on launch cadence justify its existence?




Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/16/2020 10:58 pm
Firstly, just to clarify, I have never "advocated" a 3 stage Star Ship, I personally think that is a expensive Bad Idea.
But...
I do think it makes more sense than making a dedicated 20-22m diameter 1200t capacity SuperTanker AND the UltraSuperHeavy booster needed to loft it AND the launch site/GSE for all of it (which would likely have to be far out at sea [is it still GSE when at sea, not on the ground?]). And all that for the sole purpose of reducing what, by then, should be standard, well practiced and reliable rendezvous/docking/fuel transfer events from 5 or 6 down to 1, about an 80% risk reduction to this one part of the overall mission. If instead, by making this system 23m diameter and treat a fully fueled StarShip as cargo you eliminate the need for them all-together, giving a 100% risk reduction? (Okay, maybe 99.9% due to added stage seperation event, but that has got to be a lot less risky than even one rendezvous/docking/fuel transfer/undocking event.)

As I said though, I do not view it as a good idea, just better than a dedicated single purpose SuperTanker/UltraSuperHeavy setup.

Warning and disclaimer - all numbers used in the rest of this post are derived by very detailed arm waving, and then simplified to nicely rounded numbers - this is fully justifiable as even the great Elon himself knows not yet what these the numbers may come to be ;)

I do not believe an 18m diameter tanker could get 1200t propellant to orbit. As is, Raptor has an almost unbeatable thrust to nozzle area ratio.  As has be discussed to death several times  across multiple NSF forums, doubling the diameter only increases the area four time, so with the same thrust per area, you are only getting 4 times the thrust and therefor can only lift 4 times as much meaning your payload only increases by ~4 times, all else being equal (ISP, full-burnout mass ratio, etc.).   StarShip as described needs ~1200t propellant to fully refuel when empty.  The cargo version is to have ~150t cargo capacity.  The Tanker version has to be able to carry over 200t propellant to fully refuel a StarShip in 6 tanker runs, and over 240t to do it in 5 runs - the over bit being whatever amount is needed to compensate for boil-off. At the best value of 240t, 4 times that is only 960t, 240t short of a full load, so you'll need a second tanker run and associated rendezvous/docking/transfer/undocking event to fully fuel, and unless you can use the excess ~720t left in that second tanker to dock with and top off two or three more Starships it is a waste and you are still doing multiple rendezvous/docking/fuel transfer/undocking events.  So you need a SuperTanker larger than 18m.  Over 20m diameter if the ratios of a 240+t 5 run tanker are scaled, and almost 22m diameter if the 200+t 6 run tanker ratios are scaled, in order for the SuperTanker to be able to refuel a 9 m StarShip in just one rendezvous/docking/fuel transfer/undocking event.  And unless you've developed a new 5-6 times bigger engine, you are talking 155-186 or more current Raptors on that booster alone  :P .

Musk has said that 9m StarShip is what he considers to be the minimum needed to start setting up and populating a sizable (10,000 person plus) Mars BASE, but to get to the 1 million person colony level WILL need a bigger vehicle.  Mars trips will be the vast majority of flights that will need fully refueled StarShips in LEO.  The realistic need for fully refueled starships for GTO/DirectGEO/Lunar/DeepSpace  trips is only going to be in the few tens per year at very best for the next decade plus.  So these possibly 22m SuperTankers and associated facilities are going to be super busy 3 or 4 out of every 26 month Mars synodoic period, then just sit around almost unused till the next synod?  This is going to jack the cost of StarShip flights thru the roof in order to cover the costs of developing and maintaining such massive and overall seldom used system (shades of SLS).

By the time 9m SS and SH can't keep up with trips to mars, new bigger StarShip and associated booster will be needed anyway, and it will be sized to its needs, whether 12, 18 or 20+m diameter.

And if this specifically designed just to fill a 9m StarShip in one go SuperTanker/UltraHeavyBooster isn't usefull to it, said SuperTanker/UltraHeavy/GSE/Pad will just be a very expensive one trick pony.
Wow! Arm waving for sure, but half the trick to arm waving is finding the right numbers to wave over. I think ya dun good.


IMO, once refueling is proven and reliable, and the need is established, a custom built depot makes sense. It allows a steady tempo of launches over time instead of a flurry all at once. Between a sun shade and PV powered cryo cooling losses should be acceptable. A steady tempo with dedicated SS based tankers takes a lot of stress off launch and supply operations.


Keeping a Depot orbit at the inclination of BC or Kennedy (only ~2deg different) gives an opportunity to refuel missions to any location beyond LEO and with the correct mission profile, at minimum dV for a plane change. All without a massive one off SuperDuper booster.


Once there is a depot established for general, rather than mission specific use, the
small size might become an issue. Figure on a mars mission with two crewed and 10 cargo ships and a lot of refueling has to happen. Throw in a few GEO and lunar launches and it's a problem.


Down the road (2-3 mars missions?) there might be reason for a custom 12m hammerhead tank perched atop an SH with only enough propellant to make LEO. Total mass the same as SS but with no cargo, heatshield, fins or legs. Figure a nominal 100 tons dry mass for a standard SS and a guesstimate of 190 tons savings that can become dry mass and that tank can hold quite a bit of propellant.


I have no pencil or paper handy so my arm waving does not cover structural feasibility through launch. Probably not very good. Still, it can shrink back to 9m, include a cryo cooler, PV and a sunshade and some starter propellant and still be quite large. And a lot less hassle than mega launchers. A total WAG, 2-3 SS refueling.


Phil


Edit: I was editing for clarity and sanity when by battery quit, so here goes.
- the 190t savings came from legs, heatshield and fins: 40t estimate. No cargo =150t. These numbers are provisional.
- following that number in the main body where I spoke of the depot holding quite a bit of propellant I meant once on orbit. It would launch dry except for propellant to reach LEO.
-Converting the 190t savings into 1.6t rings allows 118 additional 9m rings with no launch mass change. Give it 33% penalty for additional structure and it's 79 additional rings. Make the depot 12m and it gets 52 rings. No way this is going to happen.
- The current SS HAS 9+ propellant rings. For every nine+ additional rings an SS can be topped off. What might be the longest it can be  safely stretched? Even with reinforcing and maybe greater ullage pressure. Would 12m make it easier? I doubt the engineering effort is worth it.
- Even with any reasonable stretch, more than one would be needed to refuel a modest Mars fleet. Does a long term depots impact on launch cadence justify its existence?
Best bet is to remove the heat shield, legs, fins and header tanks. Launch that and if greater capacity is required launch more of them and link them together in orbit. Perhaps a multi tank docking node with sun shield and cryocooler could also be put up into orbit to round it off (although this would require a bit of additional work).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/18/2020 02:24 pm
Yep. I got the first 40t stripping that stuff off. Another 150t by ditching payload. You get a LOT of extra 1.6t rings for that savings.


ISTM that joining depots together just isn't worth the hassle. Just have more in the same local. Keep one for routine GEO and lunar stuff and the additional ones for surges like Mars. But this still begs the core question. Is creating depot capacity to meet surge needs worth doing? I'm looking at ~10 year window after refueling is proven.


Pro: A steady predictable tanker launch cadence instead of a flurry of launches competing for launch capacity. This assumes it will be quite a while before any launch site can support even one launch a day.


Con: A super stretch depot based on SS can probably add only ~10 rings (18m+) of propellant storage - a total top off capacity of 2 SS. This is based on unwieldiness at launch, not mass. An SS based depot without any tank stretch can top off 1 SS.


Is a halving the total number of depots for a mars fleet worth it? ISTM to depend on things we don't know and can't know until SS is much more mature.


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/18/2020 03:04 pm
Firstly, just to clarify, I have never "advocated" a 3 stage Star Ship, I personally think that is a expensive Bad Idea.
But...
I do think it makes more sense than making a dedicated 20-22m diameter 1200t capacity SuperTanker AND the UltraSuperHeavy booster needed to loft it AND the launch site/GSE for all of it (which would likely have to be far out at sea [is it still GSE when at sea, not on the ground?]). And all that for the sole purpose of reducing what, by then, should be standard, well practiced and reliable rendezvous/docking/fuel transfer events from 5 or 6 down to 1, about an 80% risk reduction to this one part of the overall mission. If instead, by making this system 23m diameter and treat a fully fueled StarShip as cargo you eliminate the need for them all-together, giving a 100% risk reduction? (Okay, maybe 99.9% due to added stage seperation event, but that has got to be a lot less risky than even one rendezvous/docking/fuel transfer/undocking event.)

As I said though, I do not view it as a good idea, just better than a dedicated single purpose SuperTanker/UltraSuperHeavy setup.

Warning and disclaimer - all numbers used in the rest of this post are derived by very detailed arm waving, and then simplified to nicely rounded numbers - this is fully justifiable as even the great Elon himself knows not yet what these the numbers may come to be ;)

I do not believe an 18m diameter tanker could get 1200t propellant to orbit. As is, Raptor has an almost unbeatable thrust to nozzle area ratio.  As has be discussed to death several times  across multiple NSF forums, doubling the diameter only increases the area four time, so with the same thrust per area, you are only getting 4 times the thrust and therefor can only lift 4 times as much meaning your payload only increases by ~4 times, all else being equal (ISP, full-burnout mass ratio, etc.).   StarShip as described needs ~1200t propellant to fully refuel when empty.  The cargo version is to have ~150t cargo capacity.  The Tanker version has to be able to carry over 200t propellant to fully refuel a StarShip in 6 tanker runs, and over 240t to do it in 5 runs - the over bit being whatever amount is needed to compensate for boil-off. At the best value of 240t, 4 times that is only 960t, 240t short of a full load, so you'll need a second tanker run and associated rendezvous/docking/transfer/undocking event to fully fuel, and unless you can use the excess ~720t left in that second tanker to dock with and top off two or three more Starships it is a waste and you are still doing multiple rendezvous/docking/fuel transfer/undocking events.  So you need a SuperTanker larger than 18m.  Over 20m diameter if the ratios of a 240+t 5 run tanker are scaled, and almost 22m diameter if the 200+t 6 run tanker ratios are scaled, in order for the SuperTanker to be able to refuel a 9 m StarShip in just one rendezvous/docking/fuel transfer/undocking event.  And unless you've developed a new 5-6 times bigger engine, you are talking 155-186 or more current Raptors on that booster alone  :P .

Musk has said that 9m StarShip is what he considers to be the minimum needed to start setting up and populating a sizable (10,000 person plus) Mars BASE, but to get to the 1 million person colony level WILL need a bigger vehicle.  Mars trips will be the vast majority of flights that will need fully refueled StarShips in LEO.  The realistic need for fully refueled starships for GTO/DirectGEO/Lunar/DeepSpace  trips is only going to be in the few tens per year at very best for the next decade plus.  So these possibly 22m SuperTankers and associated facilities are going to be super busy 3 or 4 out of every 26 month Mars synodoic period, then just sit around almost unused till the next synod?  This is going to jack the cost of StarShip flights thru the roof in order to cover the costs of developing and maintaining such massive and overall seldom used system (shades of SLS).

By the time 9m SS and SH can't keep up with trips to mars, new bigger StarShip and associated booster will be needed anyway, and it will be sized to its needs, whether 12, 18 or 20+m diameter.

And if this specifically designed just to fill a 9m StarShip in one go SuperTanker/UltraHeavyBooster isn't usefull to it, said SuperTanker/UltraHeavy/GSE/Pad will just be a very expensive one trick pony.
Wow! Arm waving for sure, but half the trick to arm waving is finding the right numbers to wave over. I think ya dun good.


IMO, once refueling is proven and reliable, and the need is established, a custom built depot makes sense. It allows a steady tempo of launches over time instead of a flurry all at once. Between a sun shade and PV powered cryo cooling losses should be acceptable. A steady tempo with dedicated SS based tankers takes a lot of stress off launch and supply operations.


Keeping a Depot orbit at the inclination of BC or Kennedy (only ~2deg different) gives an opportunity to refuel missions to any location beyond LEO and with the correct mission profile, at minimum dV for a plane change. All without a massive one off SuperDuper booster.


Once there is a depot established for general, rather than mission specific use, the
small size might become an issue. Figure on a mars mission with two crewed and 10 cargo ships and a lot of refueling has to happen. Throw in a few GEO and lunar launches and it's a problem.


Down the road (2-3 mars missions?) there might be reason for a custom 12m hammerhead tank perched atop an SH with only enough propellant to make LEO. Total mass the same as SS but with no cargo, heatshield, fins or legs. Figure a nominal 100 tons dry mass for a standard SS and a guesstimate of 190 tons savings that can become dry mass and that tank can hold quite a bit of propellant.


I have no pencil or paper handy so my arm waving does not cover structural feasibility through launch. Probably not very good. Still, it can shrink back to 9m, include a cryo cooler, PV and a sunshade and some starter propellant and still be quite large. And a lot less hassle than mega launchers. A total WAG, 2-3 SS refueling.


Phil


Edit: I was editing for clarity and sanity when by battery quit, so here goes.
- the 190t savings came from legs, heatshield and fins: 40t estimate. No cargo =150t. These numbers are provisional.
- following that number in the main body where I spoke of the depot holding quite a bit of propellant I meant once on orbit. It would launch dry except for propellant to reach LEO.
-Converting the 190t savings into 1.6t rings allows 118 additional 9m rings with no launch mass change. Give it 33% penalty for additional structure and it's 79 additional rings. Make the depot 12m and it gets 52 rings. No way this is going to happen.
- The current SS HAS 9+ propellant rings. For every nine+ additional rings an SS can be topped off. What might be the longest it can be  safely stretched? Even with reinforcing and maybe greater ullage pressure. Would 12m make it easier? I doubt the engineering effort is worth it.
- Even with any reasonable stretch, more than one would be needed to refuel a modest Mars fleet. Does a long term depots impact on launch cadence justify its existence?
Best bet is to remove the heat shield, legs, fins and header tanks. Launch that and if greater capacity is required launch more of them and link them together in orbit. Perhaps a multi tank docking node with sun shield and cryocooler could also be put up into orbit to round it off (although this would require a bit of additional work).

I do love the idea, ISTM more practical  in the mid term than the back to back SS refueling, we may also send this multi tank docking node with some sort of propulsion system to Mars orbit, this could reduce the requirement for back home journey and such allowing more cargo.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/18/2020 10:49 pm

You might be misunderstanding. There is nothing in a depot, one tank or multiple, that precludes butt to butt refueling. Two entirely separate issues.


As for sending a depot to Mars, there has been discussion in this. First you need to understand that it does not remove the SS need to refuel before leaving earth. It's an alternative to having ISRU propellant production on mars. However, it's always seen as a supplement for early missions until reliable ISRU is on line, and not a replacement. Mars will never be more than a base if it depends on earth for its very ability to return.


Phil

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: mikelepage on 08/19/2020 08:10 am
The orbital plane is relatively fixed as Earth moves around the sun, hence it will move in and out of sync with the ecliptic. So the orbital plane of a a particular LEO orbit will have two opportunities per year to launch into an interplanetary trajectory without a massive plane-change manoeuvre. If the target requires a transfer window outside of those two opportunities, you can't use that plane. You therefore need to pick a plane such that it will line up for the intended mission window.

It's not a huge burden, the mission window will last days (or rather, dozens of windows per day, for several days). But it means you aren't going to have a single LEO orbit that can be used as a staging ground for many different mission types and targets. Even for the same target, a plane that works for one window (eg, 2020 Mars window) won't work for the next (eg, 2022.) Each mission will need its own orbit (barring happy coincidences.)

Somewhat off-topic, but this particular quirk of orbital mechanics is why I doubt Phobos or Deimos will ever (?) be used as way-stations/depots on the way to Mars surface.  Only twice per Martian year would it be possible to arrive at Mars and capture into an orbital plane aligned with the moons... and how often would those arrival windows synchronise with the Earth-Mars synodic cycle?  You might be able to get around this by performing multiple plane change maneuvers whilst aero-capturing into Martian orbit, but there would have to be an upper limit to how great a plane change you could perform that way.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/19/2020 08:21 am

You might be misunderstanding. There is nothing in a depot, one tank or multiple, that precludes butt to butt refueling. Two entirely separate issues.


As for sending a depot to Mars, there has been discussion in this. First you need to understand that it does not remove the SS need to refuel before leaving earth. It's an alternative to having ISRU propellant production on mars. However, it's always seen as a supplement for early missions until reliable ISRU is on line, and not a replacement. Mars will never be more than a base if it depends on earth for its very ability to return.


Phil

I keep my opinion on  Mars colonisation prospect for me, but getting first human and first permanent settlement on the red planet will require a step by step approach. If we want to get a real plan to do so we can't bet on future developement of new technologies that does not exist yet. this is why interim solutions is required you improve the process while you achieve what can be achieved in a timely manner, the air travel didn't await for the modern jet reaction airliners to be a reality, it took years and years and many accidents to get to what is now air travel.

The above mentioned idea of a multiple disposable tanks that will be joint to a nodal comprtment as is the case for ISS, seems to me practicle for multiple reasons:

-No need for a fleet of tankers and subsequent logistic issues.
-No time waisted for SS departing to Mars.
-Depot could be positioned in any strategic position.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tyrred on 08/19/2020 09:44 am
There is a need for tankers. Logistics of tankers is a solved problem Earth-side, and must be solved space-side anyway.

Tanking has to be done before the optimal TMI window anyway, tanking operations just need to have enough certainty and margin built in to compensate for delays.

Depots can be placed strategically, but the strategies get very convoluted considering orbital mechanics.

Step by step approach should always consider reusability and sustainability.

Break away from the disposable past when building a new planet, for Bahphomet's sake.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/19/2020 10:22 am
There is a need for tankers. Logistics of tankers is a solved problem Earth-side, and must be solved space-side anyway.

Tanking has to be done before the optimal TMI window anyway, tanking operations just need to have enough certainty and margin built in to compensate for delays.

Depots can be placed strategically, but the strategies get very convoluted considering orbital mechanics.

Step by step approach should always consider reusability and sustainability.

Break away from the disposable past when building a new planet, for Bahphomet's sake.

Reusability is meant for keeping launching cost low, you reuse what is expensive, you dont reuse what is cheap or at least does not worth the reuse, if SH/SS building cost will get cheaper considering staineless steel use and others consideration, the reusability wont affect costs that much, so I think that going to Mars should considers this advices:

-Travel light.
-Have a plan B
-Simple is beautiful.
-Cheap is wonderful.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 08/19/2020 12:23 pm

Reusability is meant for keeping launching cost low, you reuse what is expensive, you dont reuse what is cheap or at least does not worth the reuse, if SH/SS building cost will get cheaper considering staineless steel use and others consideration, the reusability wont affect costs that much, so I think that going to Mars should considers this advices:

-Travel light.
-Have a plan B
-Simple is beautiful.
-Cheap is wonderful.
There are persistent, deeply unsolved problems with your proposal that you cannot seem to explain which contradict almost every point you make.  You still haven't explained how fully fueled large tanks arrive in LEO, how they're cheaper, how they're simpler, or how they enable you to 'travel light'.  The skepticism you're encountering is not 'the old guard versus John Houbolt, it's curiosity that's evolved into a little bit of exasperation because you seem to have either chosen not to answer or cannot answer the questions posed to you that seem to undercut the entire basis of your argument. 

And this is all before we get to the point where developing an entirely new, complicated hardware system with its own undefined capabilities and challenges is 'easier' than solving free-fall refueling of cryogenics. 

The skepticism is not assuaged by breezy, 'it's no big deal to disregard everything being done and start a brand new plan, my plan' summary like the bullet points you end with above.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/19/2020 01:03 pm

I keep my opinion on  Mars colonisation prospect for me, but getting first human and first permanent settlement on the red planet will require a step by step approach.
Who made that claim, and when do they expect to reach Mars?

Because under even the most pessimistic projections, SpaceX is going "before the decade is out," while if all goes well, they will be sending the first unmanned ships in 2 years.

You have repeatedly waved away the cheaper approach claiming that it can be made cheaper later. That's not how the world works. Time IS Money. Cheaper is faster.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/19/2020 01:06 pm

Reusability is meant for keeping launching cost low, you reuse what is expensive, you dont reuse what is cheap or at least does not worth the reuse, if SH/SS building cost will get cheaper considering staineless steel use and others consideration, the reusability wont affect costs that much, so I think that going to Mars should considers this advices:

-Travel light.
-Have a plan B
-Simple is beautiful.
-Cheap is wonderful.
There are persistent, deeply unsolved problems with your proposal that you cannot seem to explain which contradict almost every point you make.  You still haven't explained how fully fueled large tanks arrive in LEO, how they're cheaper, how they're simpler, or how they enable you to 'travel light'.  The skepticism you're encountering is not 'the old guard versus John Houbolt, it's curiosity that's evolved into a little bit of exasperation because you seem to have either chosen not to answer or cannot answer the questions posed to you that seem to undercut the entire basis of your argument. 

And this is all before we get to the point where developing an entirely new, complicated hardware system with its own undefined capabilities and challenges is 'easier' than solving free-fall refueling of cryogenics. 

The skepticism is not assuaged by breezy, 'it's no big deal to disregard everything being done and start a brand new plan, my plan' summary like the bullet points you end with above.

I thought it was obvious but let put thing clear:

-Send fuel tanks to LEO atop of a SH.
-Make tanks rendezvous with a multi tank docking node .
-When all tanks are in place send StarShip bound for Mars, make rendezvous transfert and launch.
-Travel lighter: if SS can get her fuel for the journey back home in Mars orbit it wont have to have the burden fuel for leaving Mars gravity.
-Cheap: because the switch to stainless steel will significantlly reduce the cost of the hardware (starship/SH), so even in the case of disposable vehicles that wont be affecting the costs that much.


I keep my opinion on  Mars colonisation prospect for me, but getting first human and first permanent settlement on the red planet will require a step by step approach.
Who made that claim, and when do they expect to reach Mars?

Because under even the most pessimistic projections, SpaceX is going "before the decade is out," while if all goes well, they will be sending the first unmanned ships in 2 years.

You have repeatedly waved away the cheaper approach claiming that it can be made cheaper later. That's not how the world works. Time IS Money. Cheaper is faster.

My point  is that a lot of enthousiasts here think that when StarShip will reach Mars, people will flock as in the west gold rush, IMHO things wont works like this, we could imagine NASA paying 15/20 billions a ride ticket for first step for man on the red planet, may be the fare will get lower for coming 10 to 20 years for scientists for a permanent base (200 to 400 millions ticket), for beyond I can't predict anything it's too far for me, its pure speculation.

But your unmanned SS for Mars in 2 years is overly optimistic.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Alvian@IDN on 08/19/2020 01:17 pm

Reusability is meant for keeping launching cost low, you reuse what is expensive, you dont reuse what is cheap or at least does not worth the reuse, if SH/SS building cost will get cheaper considering staineless steel use and others consideration, the reusability wont affect costs that much, so I think that going to Mars should considers this advices:

-Travel light.
-Have a plan B
-Simple is beautiful.
-Cheap is wonderful.
There are persistent, deeply unsolved problems with your proposal that you cannot seem to explain which contradict almost every point you make.  You still haven't explained how fully fueled large tanks arrive in LEO, how they're cheaper, how they're simpler, or how they enable you to 'travel light'.  The skepticism you're encountering is not 'the old guard versus John Houbolt, it's curiosity that's evolved into a little bit of exasperation because you seem to have either chosen not to answer or cannot answer the questions posed to you that seem to undercut the entire basis of your argument. 

And this is all before we get to the point where developing an entirely new, complicated hardware system with its own undefined capabilities and challenges is 'easier' than solving free-fall refueling of cryogenics. 

The skepticism is not assuaged by breezy, 'it's no big deal to disregard everything being done and start a brand new plan, my plan' summary like the bullet points you end with above.

I thought it was obvious but let put thing clear:

-Send fuel tanks to LEO atop of a SH.
-Make tanks rendezvous with a multi tank docking node .
-When all tanks are in place send StarShip bound for Mars, make rendezvous transfert and launch.
-Travel lighter: if SS can get her fuel for the journey back home in Mars orbit it wont have to have the burden fuel for leaving Mars gravity.
-Cheap: because the switch to stainless steel will significantlly reduce the cost of the hardware (starship/SH), so even in the case of disposable vehicles that wont be affecting the costs that much.
Did you already in your company's office doing the design work?

I imagine in your office there will be a poster saying "Beat SpaceX, because they didn't used or bother at all with my plan, which is so much better the world is super stupid for not implementing it. I'm a rocket engineering prophet!"
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 08/19/2020 01:33 pm

Reusability is meant for keeping launching cost low, you reuse what is expensive, you dont reuse what is cheap or at least does not worth the reuse, if SH/SS building cost will get cheaper considering staineless steel use and others consideration, the reusability wont affect costs that much, so I think that going to Mars should considers this advices:

-Travel light.
-Have a plan B
-Simple is beautiful.
-Cheap is wonderful.
There are persistent, deeply unsolved problems with your proposal that you cannot seem to explain which contradict almost every point you make.  You still haven't explained how fully fueled large tanks arrive in LEO, how they're cheaper, how they're simpler, or how they enable you to 'travel light'.  The skepticism you're encountering is not 'the old guard versus John Houbolt, it's curiosity that's evolved into a little bit of exasperation because you seem to have either chosen not to answer or cannot answer the questions posed to you that seem to undercut the entire basis of your argument. 

And this is all before we get to the point where developing an entirely new, complicated hardware system with its own undefined capabilities and challenges is 'easier' than solving free-fall refueling of cryogenics. 

The skepticism is not assuaged by breezy, 'it's no big deal to disregard everything being done and start a brand new plan, my plan' summary like the bullet points you end with above.

I thought it was obvious but let put thing clear:

-Send fuel tanks to LEO atop of a SH.
-Make tanks rendezvous with a multi tank docking node .
-When all tanks are in place send StarShip bound for Mars, make rendezvous transfert and launch.
-Travel lighter: if SS can get her fuel for the journey back home in Mars orbit it wont have to have the burden fuel for leaving Mars gravity.
-Cheap: because the switch to stainless steel will significantlly reduce the cost of the hardware (starship/SH), so even in the case of disposable vehicles that wont be affecting the costs that much.

[ ... snip ...]
This feels like "drop tanks" all over again ...

If you want to be taken seriously, you're going to have to back these notions up with numbers, because without that they don't seen to make much sense (to those who have looked at the numbers).

Here are some of the most obvious issues I see on casual examination of this plan ... I'm sure others here can (and have) picked this idea apart in somewhat more detail:

* How big are these fuel tanks you propose to launch?  How many will you need?  Do the math!
* Will these fuel tank be launched full (small tanks) or empty (allowing bigger tanks)?  If empty, you're going to need  ... tankers.
* How is you "multi tank docking node" simpler that tail-to-tail fuel transfer?  Be specific!
* If you're taking not just SS to mars but all the return fuel as well, your're going to need a MUCH more capable propulsion system to do it.  Please provide details.  Be specific, and do the math!

If you can't do this, then please, please try to listen and learn from those here who can.  If you can do this and just won't then you're just trolling, and I ask you to please stop.  If you can do this and are willing to put in the effort, then I think you will find this forum a wonderful place for fruitful discussions.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/19/2020 01:35 pm
-Send fuel tanks to LEO atop of a SH.
That doesnt work. Superheavy  cannot put any significant mass into orbit. If you add starship (and yes, reusable starship is significantly cheaper  per ton of fuel than expendable upperstage, even figuring in the reduced lift) then you have starship refueling starship, which is the program of record.
Quote
-Make tanks rendezvous with a multi tank docking node .
Instead of plumbing each tank through a heavy structure designed to keep multiple large tanks aligned  under thrust while rocket fuel goes through said plumbing... just plumb two starships together under 0g, transfer the fuel at a safe rate, and unplug.

Quote
-Travel lighter: if SS can get her fuel for the journey back home in Mars orbit it wont have to have the burden fuel for leaving Mars gravity.
a wagon train of return fuel coming from earth is not "Light travel" compared to going over the pass for supplies. when you need it. Starship is Spec'd for  single stage earth return, so adding a refueling stop there isnt saving anything.
Quote
-Cheap: because the switch to stainless steel will significantlly reduce the cost of the hardware (starship/SH), so even in the case of disposable vehicles that wont be affecting the costs that much.
Plastic knives are $3.50 for a box of 100. Stainless steel cutlery (40 pieces) is 30 dollars. Breakeven would be around 20-25 uses, and that's with a cheaper material being used.
Stainless vs stainless, expendable doesnt save you nearly enough to be worthwhile.

Quote

I keep my opinion on  Mars colonisation prospect for me, but getting first human and first permanent settlement on the red planet will require a step by step approach.
Who made that claim, and when do they expect to reach Mars?

Because under even the most pessimistic projections, SpaceX is going "before the decade is out," while if all goes well, they will be sending the first unmanned ships in 2 years.

You have repeatedly waved away the cheaper approach claiming that it can be made cheaper later. That's not how the world works. Time IS Money. Cheaper is faster.
(...)
But your unmanned SS for Mars in 2 years is overly optimistic.
Even if it's 4 years instead of 2, that's faster than a multi-billion nasa program can possibly be. Doing things "like NASA" would only slow them down.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 08/19/2020 01:36 pm
-Send fuel tanks to LEO atop of a SH.
SH can't bring anything to orbit, the 'tanks' would need to propel themselves the rest of the way after a couple minutes of boost.  So what you call 'tanks' are Starship tanker spacecraft without landing legs or heat shield but everything else.

-Make tanks rendezvous with a multi tank docking node .
-When all tanks are in place send StarShip bound for Mars, make rendezvous transfert and launch.
So they're docking with a depot instead of a mars-bound Starship?  Then the Starship docks with the depot?  If it is going to use the fuel from these mostly expended tanks (and all the deadweight from the tankage and engines needed to carry it to orbit after separating from the first stage) then they will need to transfer, all you're doing is switching this complicated cryogenic transfer to the worst possible method because you're transferring it during the trans-martian injection burn and carrying a hundred tons or more of extra tanks instead of doing the fuel transfer into your existing tank.

-Travel lighter: if SS can get her fuel for the journey back home in Mars orbit it wont have to have the burden fuel for leaving Mars gravity.
There is no source of fuel in Mars orbit unless you're ALSO shipping kilotons of propellant to Mars and elaborately and expensively aerobraking them into an orbit where they're out of reach of thirsty spacecraft stuck on the surface.

-Cheap: because the switch to stainless steel will significantlly reduce the cost of the hardware (starship/SH), so even in the case of disposable vehicles that wont be affecting the costs that much.
This is perplexing, the spacecraft are already stainless and nothing you do lowers costs; in fact you increase costs at every single possible step from the development of new spacecraft to requiring more fuel launches (to make up for all the extra disposable mass you're carrying during your TMI burn) to making everything disposable and throwing away raptors and avionics instead of re-using them....

There are deep, load bearing structural weaknesses in your proposal and that's usually fine, that's a normal part of spitballing different ideas and usually those are either worked out or used to decide 'hey, this won't work', but there's this element of repeating the same problems without recognizing them that probably some in your audience deeply skeptical about how much thought you've put into this and whether or not you're participating in a back and forth conversation or just repeating the same points without recognizing or addressing the problems others have identified.  I can't speak for anyone else, but this is where I am.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/19/2020 02:04 pm
Just to answer  : cdebuhr, rakaydos and Thunderscreech as your post are pretty the same:

-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? if you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.

-I evocked Stainless steel in term of cost reduction for hardware building costs not only in terms of overall material cost but also in term of building process simplification.

-Lowering costs with reusability will require using hardware multiple times and the later is not for near future so expandable option is best sueted coming years.

-No viable business plan for Mars bound missions will make you depend on governement spending and this one wont require large scale use planned by Elon Musk.


In the end there is still a lot of shadow area for EM Mars projections that makes his assumptions in difficult position.






Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/19/2020 02:11 pm
The orbital plane is relatively fixed as Earth moves around the sun, hence it will move in and out of sync with the ecliptic. So the orbital plane of a a particular LEO orbit will have two opportunities per year to launch into an interplanetary trajectory without a massive plane-change manoeuvre. If the target requires a transfer window outside of those two opportunities, you can't use that plane. You therefore need to pick a plane such that it will line up for the intended mission window.

It's not a huge burden, the mission window will last days (or rather, dozens of windows per day, for several days). But it means you aren't going to have a single LEO orbit that can be used as a staging ground for many different mission types and targets. Even for the same target, a plane that works for one window (eg, 2020 Mars window) won't work for the next (eg, 2022.) Each mission will need its own orbit (barring happy coincidences.)

Somewhat off-topic, but this particular quirk of orbital mechanics is why I doubt Phobos or Deimos will ever (?) be used as way-stations/depots on the way to Mars surface.  Only twice per Martian year would it be possible to arrive at Mars and capture into an orbital plane aligned with the moons... and how often would those arrival windows synchronise with the Earth-Mars synodic cycle?  You might be able to get around this by performing multiple plane change maneuvers whilst aero-capturing into Martian orbit, but there would have to be an upper limit to how great a plane change you could perform that way.
Well, I'm going to wander OT with you into orbital mechanics but for on topic reasons. This has very much to do with refueling strategies and budgets for plane changes.


Changing planes is considered to be a high energy maneuver and they can be, but they can also be done with much lower energy at the cost of time. Let's look at the worst possible example; turning around and going the other way.


Let's assume the universe is only earth and the orbiting ship and all maneuvers are instantaneous. The ship is in LEO at 7.75kps. It needs go in the opposite direction. The brute force approach is a retro burn of 7.75kps to bring it to a halt, then 7.75kps in the new direction. Total budget is 15.5kps.


Escape velocity is 11.2kps. So for a deltaV of under 3.45kps the ship can raise itself to an apogee of any arbitrary height and then upon return to perigee apply the same dV to circularize it's orbit. Total budget is 6.9kps.


At apogee the ship is moving slow and, using a highly technical term, one 'puff' will stop it dead in its tracks. A second puff will send it back the way it came. Total budget for the most radical possible plane change possible is 6.9kps +2 puffs. Total savings is 8.6kps less two puffs. The higher the apogee, the smaller the puffs.


This applies directly to refueling operations and mission profiles. Canaveral is at 28.5N latitude and the natural orbit inclination from there is the same. With a depot placed at this inclination and a lunar mission launched the same, refueling is, orbitally speaking, straight forward.


The refueled ship then does a burn for TLI at the same inclination and when passing through L1 (I think) will be at its lowest velocity. This is where a plane change is most economical. Want a polar orbit? This is where to do it.


The same principal applies to plane changes on a mars mission. The difference is that the ideal place to change planes will be the Mars-Sun L1 (I think). It might work out to be cheaper heading to lunar L1 and picking a plane change that slings the ship around Luna and off to Mars awaiting further refinement later on, but the numbers are beyond me.


The upshot is settling into co-orbit with Phobos or Demos should not be especially expensive. There will probably be a plane change of some sort anyway.


Phil



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/19/2020 02:32 pm
Just to answer  : cdebuhr, rakaydos and Thunderscreech as your post are pretty the same:

-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? if you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.
The number being thrown around in the Lunar Starship thread is 40 tons for reuse equipment. No idea how accurate that is, but it's a starting point. With the same tanks, engines, and booster, that would increase payload from 150 to 190 tons... Less than a 33% increase. A SINGLE tanker reuse more than breaks even, putting more fuel into orbit than an expendable stage.

Quote
-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
Why a fleet? Just use one or two tankers, and fully reuse, even if it isnt rapid yet? Expendable is NOT a better option.

Quote
-Lowering costs with reusability will require using hardware multiple times and the later is not for near future so expandable option is best sueted coming years.
Why is this "not for the near future?" Using hardware multiple times is already happening- hopper flew twice, SN5 and SN6 will be trading off flights, SN8 will bemaking multiple hops... Your premice is ungrounded in reality.

Quote
-No viable business plan for Mars bound missions will make you depend on governement spending and this one wont require large scale use planned by Elon Musk.
Only if it's too expensive for a billionare to self-fund. The wealth gap in America is unbelievable.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/19/2020 02:47 pm

You might be misunderstanding. There is nothing in a depot, one tank or multiple, that precludes butt to butt refueling. Two entirely separate issues.


As for sending a depot to Mars, there has been discussion in this. First you need to understand that it does not remove the SS need to refuel before leaving earth. It's an alternative to having ISRU propellant production on mars. However, it's always seen as a supplement for early missions until reliable ISRU is on line, and not a replacement. Mars will never be more than a base if it depends on earth for its very ability to return.


Phil

I keep my opinion on  Mars colonisation prospect for me, but getting first human and first permanent settlement on the red planet will require a step by step approach. If we want to get a real plan to do so we can't bet on future developement of new technologies that does not exist yet. this is why interim solutions is required you improve the process while you achieve what can be achieved in a timely manner, the air travel didn't await for the modern jet reaction airliners to be a reality, it took years and years and many accidents to get to what is now air travel.

The above mentioned idea of a multiple disposable tanks that will be joint to a nodal comprtment as is the case for ISS, seems to me practicle for multiple reasons:

-No need for a fleet of tankers and subsequent logistic issues.
-No time waisted for SS departing to Mars.
-Depot could be positioned in any strategic position.
Ahhh. I think I see the problem. It's a matter of religious doctrine and not amenable to reasoned argument - from either side of the question.


Stardust, from what I'm picking up, you are skeptical of the entire approach. This is observation, not criticism.


For the majority of us, we are skeptical of the traditional approach. Again, an observation and not an endorsement.


The two religions are so diametrically at odds that they can never be reconciled no matter how much technical argument is presented. So let's stop this bickering over detail and agree to disagree. THERE IS NOTHING HERE TO GET frakked ABOUT!!


I learned a long time ago that all logic based arguments have at least one predicate - an assumption. What we have here is a conflict of assumptions.


Stardust, the only fault I can assign you is that you have stepped into a den of new spacers and are espousing the doctrine of old space or something close to it. In another time that would get you burned at the stake or an invitation to make love with the Iron Maiden. I promise, we won't do that.


What I would suggest is that you start another discussion with those who share your beliefs. I'd like to believe that we are collectively mature enough to allow others to discuss that which we disagree with although we might want to draw the line at flat earthers 8)


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 08/19/2020 03:00 pm
Well, I'm going to wander OT with you into orbital mechanics but for on topic reasons. This has very much to do with refueling strategies and budgets for plane changes.


Changing planes is considered to be a high energy maneuver and they can be, but they can also be done with much lower energy at the cost of time. Let's look at the worst possible example; turning around and going the other way.


Let's assume the universe is only earth and the orbiting ship and all maneuvers are instantaneous. The ship is in LEO at 7.75kps. It needs go in the opposite direction. The brute force approach is a retro burn of 7.75kps to bring it to a halt, then 7.75kps in the new direction. Total budget is 15.5kps.


Escape velocity is 11.2kps. So for a deltaV of under 3.45kps the ship can raise itself to an apogee of any arbitrary height and then upon return to perigee apply the same dV to circularize it's orbit. Total budget is 6.9kps.


At apogee the ship is moving slow and, using a highly technical term, one 'puff' will stop it dead in its tracks. A second puff will send it back the way it came. Total budget for the most radical possible plane change possible is 6.9kps +2 puffs. Total savings is 8.6kps less two puffs. The higher the apogee, the smaller the puffs.


This applies directly to refueling operations and mission profiles. Canaveral is at 28.5N latitude and the natural orbit inclination from there is the same. With a depot placed at this inclination and a lunar mission launched the same, refueling is, orbitally speaking, straight forward.


The refueled ship then does a burn for TLI at the same inclination and when passing through L1 (I think) will be at its lowest velocity. This is where a plane change is most economical. Want a polar orbit? This is where to do it.


The same principal applies to plane changes on a mars mission. The difference is that the ideal place to change planes will be the Mars-Sun L1 (I think). It might work out to be cheaper heading to lunar L1 and picking a plane change that slings the ship around Luna and off to Mars awaiting further refinement later on, but the numbers are beyond me.


The upshot is settling into co-orbit with Phobos or Demos should not be especially expensive. There will probably be a plane change of some sort anyway.


Phil

Why bother with a plane change if you don't have to. The only orbital depot you need is the one in LEO with enough fuel for all the ships going to wherever. Fill them up when the plan is made.

There really isn't a plane change when going interplanetary. As you approach escape velocity the tangential  component drops to zero. The tangential component is what the "plane change" changes.

A depot is simply a collection of ships held together by positioning, ropes, struts whatever. Each individual depot ship will fill one destination ship with butt to butt refueling.

Now if a depot was located in a very high orbit. l1, l2, l3, l4, l5 etc. then the tangential component is low so any plane change is less expensive. But for mars and most destinations this high orbit refueling isn't needed.

EDIT: spelling
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/19/2020 03:19 pm

You might be misunderstanding. There is nothing in a depot, one tank or multiple, that precludes butt to butt refueling. Two entirely separate issues.


As for sending a depot to Mars, there has been discussion in this. First you need to understand that it does not remove the SS need to refuel before leaving earth. It's an alternative to having ISRU propellant production on mars. However, it's always seen as a supplement for early missions until reliable ISRU is on line, and not a replacement. Mars will never be more than a base if it depends on earth for its very ability to return.


Phil

I keep my opinion on  Mars colonisation prospect for me, but getting first human and first permanent settlement on the red planet will require a step by step approach. If we want to get a real plan to do so we can't bet on future developement of new technologies that does not exist yet. this is why interim solutions is required you improve the process while you achieve what can be achieved in a timely manner, the air travel didn't await for the modern jet reaction airliners to be a reality, it took years and years and many accidents to get to what is now air travel.

The above mentioned idea of a multiple disposable tanks that will be joint to a nodal comprtment as is the case for ISS, seems to me practicle for multiple reasons:

-No need for a fleet of tankers and subsequent logistic issues.
-No time waisted for SS departing to Mars.
-Depot could be positioned in any strategic position.
Ahhh. I think I see the problem. It's a matter of religious doctrine and not amenable to reasoned argument - from either side of the question.


Stardust, from what I'm picking up, you are skeptical of the entire approach. This is observation, not criticism.


For the majority of us, we are skeptical of the traditional approach. Again, an observation and not an endorsement.


The two religions are so diametrically at odds that they can never be reconciled no matter how much technical argument is presented. So let's stop this bickering over detail and agree to disagree. THERE IS NOTHING HERE TO GET frakked ABOUT!!


I learned a long time ago that all logic based arguments have at least one predicate - an assumption. What we have here is a conflict of assumptions.


Stardust, the only fault I can assign you is that you have stepped into a den of new spacers and are espousing the doctrine of old space or something close to it. In another time that would get you burned at the stake or an invitation to make love with the Iron Maiden. I promise, we won't do that.


What I would suggest is that you start another discussion with those who share your beliefs. I'd like to believe that we are collectively mature enough to allow others to discuss that which we disagree with although we might want to draw the line at flat earthers 8)


Phil

I got a lot of things that make me skeptical from low maturity technology to financial plan lacking realism and no sustainable viable business plan. :-\
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wes_wilson on 08/19/2020 03:35 pm

You might be misunderstanding. There is nothing in a depot, one tank or multiple, that precludes butt to butt refueling. Two entirely separate issues.


As for sending a depot to Mars, there has been discussion in this. First you need to understand that it does not remove the SS need to refuel before leaving earth. It's an alternative to having ISRU propellant production on mars. However, it's always seen as a supplement for early missions until reliable ISRU is on line, and not a replacement. Mars will never be more than a base if it depends on earth for its very ability to return.


Phil

I keep my opinion on  Mars colonisation prospect for me, but getting first human and first permanent settlement on the red planet will require a step by step approach. If we want to get a real plan to do so we can't bet on future developement of new technologies that does not exist yet. this is why interim solutions is required you improve the process while you achieve what can be achieved in a timely manner, the air travel didn't await for the modern jet reaction airliners to be a reality, it took years and years and many accidents to get to what is now air travel.

The above mentioned idea of a multiple disposable tanks that will be joint to a nodal comprtment as is the case for ISS, seems to me practicle for multiple reasons:

-No need for a fleet of tankers and subsequent logistic issues.
-No time waisted for SS departing to Mars.
-Depot could be positioned in any strategic position.
Ahhh. I think I see the problem. It's a matter of religious doctrine and not amenable to reasoned argument - from either side of the question.


Stardust, from what I'm picking up, you are skeptical of the entire approach. This is observation, not criticism.


For the majority of us, we are skeptical of the traditional approach. Again, an observation and not an endorsement.


The two religions are so diametrically at odds that they can never be reconciled no matter how much technical argument is presented. So let's stop this bickering over detail and agree to disagree. THERE IS NOTHING HERE TO GET frakked ABOUT!!


I learned a long time ago that all logic based arguments have at least one predicate - an assumption. What we have here is a conflict of assumptions.


Stardust, the only fault I can assign you is that you have stepped into a den of new spacers and are espousing the doctrine of old space or something close to it. In another time that would get you burned at the stake or an invitation to make love with the Iron Maiden. I promise, we won't do that.


What I would suggest is that you start another discussion with those who share your beliefs. I'd like to believe that we are collectively mature enough to allow others to discuss that which we disagree with although we might want to draw the line at flat earthers 8)


Phil

I got a lot of things that make me skeptical from low maturity technology to financial plan lacking realism and no sustainable viable business plan. :-\

That's glass half empty, or some people might say more realistic lol.  I prefer to think they're the team who finally have a real plan for doing this with re-use, powered landing, and funding from starlink.  If they've got it, they're building a new era in human travel.  If they fail, then they'll have advanced the state of the art a long way and eventually someone will take their successes and build on them. 

Either way, spending time on approaches they've clearly discarded is wasteful.  Same reason we're no longer dissecting the nuances of large scale composite construction; they discarded that approach.  Likewise, they long ago discarded the notion of expendable rocketry.  Right, wrong?  Only history will decide, but burning up the forum advocating SpaceX go disposable isn't a great use of page space.

If you really like the expendable idea you should explore the habitat threads.  There's a lot of supportive discussion for the  concept there in terms of leaving them at Mars and using as permanent habitats.  Musk's even hinted that may be the fate of early flights.  You're not wrong that there may be some expendable parts  in the architecture for some time; but you're really misplaced thinking it's going to be their plan in earth orbit since they've clearly stated the opposite.

Just my $0.02,  I think your idea would generate more healthy debate in the habitats thread where people are eager to only fly them once ;)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/19/2020 05:03 pm
Well, I'm going to wander OT with you into orbital mechanics but for on topic reasons. This has very much to do with refueling strategies and budgets for plane changes.


Changing planes is considered to be a high energy maneuver and they can be, but they can also be done with much lower energy at the cost of time. Let's look at the worst possible example; turning around and going the other way.


Let's assume the universe is only earth and the orbiting ship and all maneuvers are instantaneous. The ship is in LEO at 7.75kps. It needs go in the opposite direction. The brute force approach is a retro burn of 7.75kps to bring it to a halt, then 7.75kps in the new direction. Total budget is 15.5kps.


Escape velocity is 11.2kps. So for a deltaV of under 3.45kps the ship can raise itself to an apogee of any arbitrary height and then upon return to perigee apply the same dV to circularize it's orbit. Total budget is 6.9kps.


At apogee the ship is moving slow and, using a highly technical term, one 'puff' will stop it dead in its tracks. A second puff will send it back the way it came. Total budget for the most radical possible plane change possible is 6.9kps +2 puffs. Total savings is 8.6kps less two puffs. The higher the apogee, the smaller the puffs.


This applies directly to refueling operations and mission profiles. Canaveral is at 28.5N latitude and the natural orbit inclination from there is the same. With a depot placed at this inclination and a lunar mission launched the same, refueling is, orbitally speaking, straight forward.


The refueled ship then does a burn for TLI at the same inclination and when passing through L1 (I think) will be at its lowest velocity. This is where a plane change is most economical. Want a polar orbit? This is where to do it.


The same principal applies to plane changes on a mars mission. The difference is that the ideal place to change planes will be the Mars-Sun L1 (I think). It might work out to be cheaper heading to lunar L1 and picking a plane change that slings the ship around Luna and off to Mars awaiting further refinement later on, but the numbers are beyond me.


The upshot is settling into co-orbit with Phobos or Demos should not be especially expensive. There will probably be a plane change of some sort anyway.


Phil

Why bother with a plane change if you don't have to. The only orbital depot you need is the one in LEO with enough fuel for all the ships going to wherever. Fill them up when the plan is made.

There really isn't a plane change when going interplanetary. As you approach escape velocity the tangential  component drops to zero. The tangential component is what the "plane change" changes.

A depot is simply a collection of ships held together by positioning, ropes, struts whatever. Each individual depot ship will fill one destination ship with butt to butt refueling.

Now if a depot was located in a very high orbit. l1, l2, l3, l4, l5 etc. then the tangential component is low so any plane change is less expensive. But for mars and most destinations this high orbit refueling isn't needed.

EDIT: spelling
I see your point on interplanetary. Which also negates the concern about Phobos & Demos.


Plane change is still needed for lunar and GEO. My point that a general purpose depot doing fine in any convenient LEO still holds.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 08/19/2020 05:05 pm

I see your point on interplanetary. Which also negates the concern about Phobos & Demos.


Plane change is still needed for lunar and GEO. My point that a general purpose depot doing fine in any convenient LEO still holds.

Moon is basically/almost interplanetary.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AJW on 08/19/2020 11:32 pm
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? If you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
 

Star-Dust, you have aspirations of sending rockets to Mars, that is admirable, and if rockets were fueled with persistence, you might already be in orbit.  My unsolicited advice to you is that like early aviators, take one step at a time and develop a firm grasp of the fundamentals before taking that giant leap. 

A great example is your lack of understanding that SH can’t deliver your proposed depot to orbit.  This is an easy mistake to make, but as a result of this error, you have been now forced to fabricate this new expendable stage.  You deride the cost of maintaining tankers, but now somehow building expendable second stages comes without a tremendous cost?  That cost comes in many forms, not just in dollars, but also in manufacturing time and a workforce that is now dedicated to building hardware that will be thrown away after every flight.  This is very inefficient. 

If you have paid attention to the work at BC, you can see that constructing a SS is a multi-month process.  How is SS minus wings and the payload section any different from your proposed second stage?  Are you really willing to wait months between depot launches?  Have you done any analysis of the cost to dispose of five stages vs. maintenance of a single tanker?  Now work the other side of the problem.  You have your depot, but without the stage you just threw away, you can’t change orbits or dock with other depots.  You will need power to keep the fuels at the proper pressure.  Where will this come from?  Are you now going to add solar arrays and compressors that you also dispose of along with communications equipment, RCS, batteries, fuel pumps, etc.?  How will you de-orbit your empty depot?

My recommendation is to stick with this first step.  Perhaps spend some time studying the capabilities of SH and SS so you can avoid further simple errors.  Look at necessary capabilities for that second stage you are proposing compared to a SS Tanker/Depot concept. Again, before you can convince us that you have a viable vision for how to get to Mars and back, at least do the minimum legwork and develop a viable plan for how to get just a single Starship worth of fuel to LEO without one year required to build your five expendable second stages, along with 30 or more Raptors that you plan to dispose of.  Yes, this will take some discipline, and you just might succeed in this first step and either make a compelling case for your depots, or garner the knowledge and understanding as to why so many who have studied this problem, some for decades, disagree with you.  Best of luck.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: envy887 on 08/20/2020 12:51 am
...
Escape velocity is 11.2kps. So for a deltaV of under 3.45kps the ship can raise itself to an apogee of any arbitrary height and then upon return to perigee apply the same dV to circularize it's orbit. Total budget is 6.9kps.
There really isn't a plane change when going interplanetary. As you approach escape velocity the tangential  component drops to zero. The tangential component is what the "plane change" changes.
...

That's not accurate. All departure burns accelerate the vehicle prograde relative to Earth, increasing only the tangential component of the velocity. If your LEO orbit plane is parallel to the Sun-Earth line, you have a problem, because you need to go in the direction the Earth is travelling around the Sun, which is 90 degrees to the Earth-Sun line. That requires a plane change because at no point in the orbit does your velocity vector point in that direction.

But you don't need to raise the entire depot far from Earth to change planes. Just let nodal precession do that for you, changing altitude slightly to take advantage of different precession rates.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/20/2020 02:00 am
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? If you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
 

Star-Dust, you have aspirations of sending rockets to Mars, that is admirable, and if rockets were fueled with persistence, you might already be in orbit.  My unsolicited advice to you is that like early aviators, take one step at a time and develop a firm grasp of the fundamentals before taking that giant leap. 

A great example is your lack of understanding that SH can’t deliver your proposed depot to orbit.  This is an easy mistake to make, but as a result of this error, you have been now forced to fabricate this new expendable stage.  You deride the cost of maintaining tankers, but now somehow building expendable second stages comes without a tremendous cost?  That cost comes in many forms, not just in dollars, but also in manufacturing time and a workforce that is now dedicated to building hardware that will be thrown away after every flight.  This is very inefficient. 

If you have paid attention to the work at BC, you can see that constructing a SS is a multi-month process.  How is SS minus wings and the payload section any different from your proposed second stage?  Are you really willing to wait months between depot launches?  Have you done any analysis of the cost to dispose of five stages vs. maintenance of a single tanker?  Now work the other side of the problem.  You have your depot, but without the stage you just threw away, you can’t change orbits or dock with other depots.  You will need power to keep the fuels at the proper pressure.  Where will this come from?  Are you now going to add solar arrays and compressors that you also dispose of along with communications equipment, RCS, batteries, fuel pumps, etc.?  How will you de-orbit your empty depot?

My recommendation is to stick with this first step.  Perhaps spend some time studying the capabilities of SH and SS so you can avoid further simple errors.  Look at necessary capabilities for that second stage you are proposing compared to a SS Tanker/Depot concept. Again, before you can convince us that you have a viable vision for how to get to Mars and back, at least do the minimum legwork and develop a viable plan for how to get just a single Starship worth of fuel to LEO without one year required to build your five expendable second stages, along with 30 or more Raptors that you plan to dispose of.  Yes, this will take some discipline, and you just might succeed in this first step and either make a compelling case for your depots, or garner the knowledge and understanding as to why so many who have studied this problem, some for decades, disagree with you.  Best of luck.
This is good advice. The best way I can think of educating yourself is to go back to the earliest SS thread and start reading. Sounds like quite a task but it is only if you let it. Best damn Science Fiction Fact I've read in a long time.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Pahimarus on 08/20/2020 02:32 am
...
Escape velocity is 11.2kps. So for a deltaV of under 3.45kps the ship can raise itself to an apogee of any arbitrary height and then upon return to perigee apply the same dV to circularize it's orbit. Total budget is 6.9kps.
There really isn't a plane change when going interplanetary. As you approach escape velocity the tangential  component drops to zero. The tangential component is what the "plane change" changes.
...

That's not accurate. All departure burns accelerate the vehicle prograde relative to Earth, increasing only the tangential component of the velocity. If your LEO orbit plane is parallel to the Sun-Earth line, you have a problem, because you need to go in the direction the Earth is travelling around the Sun, which is 90 degrees to the Earth-Sun line. That requires a plane change because at no point in the orbit does your velocity vector point in that direction.

But you don't need to raise the entire depot far from Earth to change planes. Just let nodal precession do that for you, changing altitude slightly to take advantage of different precession rates.

So I've been lurking on this thread for awhile and this is a critical point that has come up several times around the depot concept.

The key question is does launching from an inclined orbit have a meaningful effect on interplanetary transfers?

If no, then you can have a general purpose depot in LEO. If yes, then you can only have mission specific depots. I'm not sure I've seen a clear consensus in this thread and I've been doing some noodling and reading so I'm going to take a stab at it.

Caveat: I am entirely kerbal trained plus I work in the soft sciences and math makes me slightly nauseated (insert Luke walking into the cantina gif here) so take this for what it is worth.

I think the answer is yes and no. Obviously you can launch to Mars from an inclined orbit. Three vehicles just did it last month. I highly doubt they spent the dv to plane change into equatorial orbits.

However I think your inclined orbit has to be pointed in the right direction (AN and DN aligned in a certain way relative to your intended ejection angle).

This is no big deal for a specific mission. Do some maths and launch your ships at the right time of day basically (though as previously discussed you'd have limited launch windows). But I really do think it means you can't easily have a depot in an inclined orbit to service multiple different interplanetary transfers.

You also want it in LEO for oberth.

However, despair not general purpose depot proponents.

It occurs to me that the problem fundamentally is inclined orbits which occur because KSC is not on the equator and can't easily be moved there.

But if one was an ambitious billionaire planning on turning an oil rig into a launch platform... Well now...<stokes beard thoughtfully>...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: armchairfan on 08/20/2020 02:38 am
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? if you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.
I think that you'll gain more insight if you run such what-if exercises yourself.

I suggest Silverbird Astronautics Launch Vehicle Performance Calculator (http://www.silverbirdastronautics.com/LVperform.html). Unfortunately, the SS/SH stack isn't listed so you'll have to use the "User-defined" option.

Here are some reasonable suggestions to get you going but they are by no means definitive.

Number of Stages: 2
Strap-on Boosters? No
Dry Mass (kg) 1st stage: 220000 , 2nd stage 135000 (both #s include landing propellant)
Propellant (kg) 1st stage 3400000 , 2nd stage 1200000
Thrust (KN) 1st stage 56000, 2nd stage 11000
Isp (s) 1st stage 330, 2nd stage 350
Default Propellant Residuals? Yes
Restartable Upper Stage? Yes
Payload Fairing: 0 and 0
Launch Site: Cape Canaveral (USA)
Destination:
Earth Orbit
Apogee (km): 300
Perigee (km): 300
Inclination (deg): 28.4
Trajectory: Optimal

"Calculate" yields an estimated payload of about 100 tonnes. SpaceX hopes to get that to 150 tonnes by various improvements (and unlike me, they don't have to guesstimate what the vehicle specs are). Leave off the second stage and see what SH can do by itself. You won't be impressed.

The web-page will even estimate the payload for an escape trajectory (try Moon @ C3=-2 or Mars @ C3=13 km^2/s^2). Spoiler alert: it's 0. As everyone knows, you need to refuel Starship to go beyond Earth orbit and that means lifting up to 1200 tonnes of propellant into LEO. So what's the best way to do that?

You could use expendable tankers which would gain you back the landing propellant (15 tonnes) and the various other items needed for landing (probably around 25 tonnes). So expendable tankers might carry 40% more propellant than reusable ones. Even ignoring the significant cost of expending the tankers, do you really think that you can build two expendable tankers in less time than you can launch a single tanker three times?

Quote
I got a lot of things that make me skeptical from low maturity technology to financial plan lacking realism and no sustainable viable business plan. :-\

[rant]
Fortunately for SpaceX, they don't have to convince you. They've already got investors who see the value in developing Starship and other technologies even if the motivation might not be for Mars. And, they've got the world's 4th richest person who calls the shots and is motivated to make it happen. Perhaps you've noticed that there's a lot going on in Boca Chica and elsewhere these days. Maybe cities on Mars don't pan out but Elon's giving it a shot and I for one wouldn't bet against him.

And as a Space advocate *and* US taxpayer, the bang for the buck of SpaceX's vision is off-scale high.
[/rant]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: armchairfan on 08/20/2020 05:08 am
But you don't need to raise the entire depot far from Earth to change planes. Just let nodal precession do that for you, changing altitude slightly to take advantage of different precession rates.

However I think your inclined orbit has to be pointed in the right direction (AN and DN aligned in a certain way relative to your intended ejection angle).

As an exercise and if I've done the math right -- a big ask -- a 340 km circular orbit at 28.4 inclination (Cape Canaveral) will precess at such a rate as to be aligned toward Mars every 26 months; i.e. when the next launch window rolls around. While that's a fairly low orbit, (I'm guessing that) you can probably keep it going with just the propellant that you'd otherwise lose to boil-off.

This probably only makes sense if you're going to use an entire synod sending tankers to a depot(s) in preparation for a vast fleet of starships leaving all at once.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/20/2020 09:28 am
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? If you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
 

Star-Dust, you have aspirations of sending rockets to Mars, that is admirable, and if rockets were fueled with persistence, you might already be in orbit.  My unsolicited advice to you is that like early aviators, take one step at a time and develop a firm grasp of the fundamentals before taking that giant leap. 

A great example is your lack of understanding that SH can’t deliver your proposed depot to orbit.  This is an easy mistake to make, but as a result of this error, you have been now forced to fabricate this new expendable stage.  You deride the cost of maintaining tankers, but now somehow building expendable second stages comes without a tremendous cost?  That cost comes in many forms, not just in dollars, but also in manufacturing time and a workforce that is now dedicated to building hardware that will be thrown away after every flight.  This is very inefficient. 

If you have paid attention to the work at BC, you can see that constructing a SS is a multi-month process.  How is SS minus wings and the payload section any different from your proposed second stage?  Are you really willing to wait months between depot launches?  Have you done any analysis of the cost to dispose of five stages vs. maintenance of a single tanker?  Now work the other side of the problem.  You have your depot, but without the stage you just threw away, you can’t change orbits or dock with other depots.  You will need power to keep the fuels at the proper pressure.  Where will this come from?  Are you now going to add solar arrays and compressors that you also dispose of along with communications equipment, RCS, batteries, fuel pumps, etc.?  How will you de-orbit your empty depot?

My recommendation is to stick with this first step.  Perhaps spend some time studying the capabilities of SH and SS so you can avoid further simple errors.  Look at necessary capabilities for that second stage you are proposing compared to a SS Tanker/Depot concept. Again, before you can convince us that you have a viable vision for how to get to Mars and back, at least do the minimum legwork and develop a viable plan for how to get just a single Starship worth of fuel to LEO without one year required to build your five expendable second stages, along with 30 or more Raptors that you plan to dispose of.  Yes, this will take some discipline, and you just might succeed in this first step and either make a compelling case for your depots, or garner the knowledge and understanding as to why so many who have studied this problem, some for decades, disagree with you.  Best of luck.
This is good advice. The best way I can think of educating yourself is to go back to the earliest SS thread and start reading. Sounds like quite a task but it is only if you let it. Best damn Science Fiction Fact I've read in a long time.
It is not immediately clear to me what Stardust meant here. Expandable or expendable?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: armchairfan on 08/20/2020 10:20 am
It is not immediately clear to me what Stardust meant here. Expandable or expendable?
Good point. I read it as "expendable" ("spell-checked" to "expandable") based on the earlier "drop-tank" idea. However, re-reading, I think that expandable actually is the intended meeting. So ...

If the Starship tanker uses the same outer mold line as the other Starship variants, it's pretty much expandable as is. Just relocate the tank bulkheads upward into the otherwise unused payload section. That might not even require more mass. Elon recently tweeted that there's 1000 cubic meters of useable volume in the fairing. That'd be 1000 tonnes of additional propellant capacity, which is *way* more than Starship could lift.

Or create a single tanker variant by removing three rings from the payload section and adding two of them to the LOX tank and one to the LCH4 tank. That's more than 300 tons of additional propellant capacity. Then only fill the tanks with as much propellant as you can take to the destination orbit. An optimized tanker, which Elon tweeted about a while back, would remove more rings from the nose section and look stubbier.

BTW, this isn't a new idea; it's been discussed many times. There's an incredible amount of information on NSF though it can take a while to wade through it if you're just getting started.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Pahimarus on 08/20/2020 11:57 am
Quote from: armchairfan

As an exercise and if I've done the math right -- a big ask -- a 340 km circular orbit at 28.4 inclination (Cape Canaveral) will precess at such a rate as to be aligned toward Mars every 26 months; i.e. when the next launch window rolls around. While that's a fairly low orbit, (I'm guessing that) you can probably keep it going with just the propellant that you'd otherwise lose to boil-off.

This probably only makes sense if you're going to use an entire synod sending tankers to a depot(s) in preparation for a vast fleet of starships leaving all at once.

Thanks. I was trying to understand if such an orbit existed or not.

This was discussed earlier in the thread and I realized that I missed the page where it came to a conclusion. So I won't continue necro'ing the discussion beyond to offer my own $0.02. Initially it will make more sense to just launch a train of tanker starships (say 1 every 12-24 hr based on your twice daily launch window and presumably some time to do refueling ops for each tanker) to rendezvous with the mission starship shortly before the intended launch. So you'd only spend a few days to a week on orbit refueling.

However, when sending many many starships that launch cadence doesn't seem sustainable to me over a short mars transfer window even with ksc, bc, and maybe an offshore platform or two pitching in. So an accumulation tanker in a carefully selected orbit like you calculated would make sense to allow you to preposition fuel for a launch window. By this time the use case will be existent not theoretical and launch costs would have been massively reduced so one could justify the added costs of on orbit construction for a dedicated facility.

That said it may prove out that a rapid reuse and launch cadence is easier to solve than in space depot construction. It will certainly need to be if point to point is going to be a serious proposition.

Either way holy smokes are you looking at a lot of launches for a mars bound fleet.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 08/20/2020 12:08 pm
...
Escape velocity is 11.2kps. So for a deltaV of under 3.45kps the ship can raise itself to an apogee of any arbitrary height and then upon return to perigee apply the same dV to circularize it's orbit. Total budget is 6.9kps.
There really isn't a plane change when going interplanetary. As you approach escape velocity the tangential  component drops to zero. The tangential component is what the "plane change" changes.
...

That's not accurate. All departure burns accelerate the vehicle prograde relative to Earth, increasing only the tangential component of the velocity. If your LEO orbit plane is parallel to the Sun-Earth line, you have a problem, because you need to go in the direction the Earth is travelling around the Sun, which is 90 degrees to the Earth-Sun line. That requires a plane change because at no point in the orbit does your velocity vector point in that direction.

But you don't need to raise the entire depot far from Earth to change planes. Just let nodal precession do that for you, changing altitude slightly to take advantage of different precession rates.

Trying to understand what you said...

I should have said as you leave the gravity well tangential velocity approaches zero.
What do you mean as parallel?
As I understand it the plane of LEO orbit should have intersecting nodes in the direction of the interplanetary object.
This is hard to explain!
We have the plane of LEO orbit.
We have the direction of where we want to go.
The direction we want to go should be in the plane of the LEO orbit.

EDIT:

Another way to understand this. Take an extreme example(not efficient).
A polar orbit done at the equinox. It is perpendicular to the sun earth plane. Twice per year the plane will align with where you want to go (the direction will be in the plane). So this plane will not be correct for many 2 year launch oppurtunities to mars because this plane is more or less fixed in space.

Now of course LEO orbits precess relative to the stars. What we want for mars is a orbit that precesses just enough every 2 years.

Or just launch the depot in the correct orbital plane for the upcoming departure window.   
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/20/2020 02:32 pm
This might be useful.

https://en.m.wikipedia.org/wiki/File:Orbit1.svg

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 08/20/2020 03:07 pm
This might be useful.

https://en.m.wikipedia.org/wiki/File:Orbit1.svg

Thats fine but I didn't want to use those terms because they are relative to earths orbit and the first point of ares.

Various destinations depart dramatically from this coordinate system.
I.E. moon, pluto
The rest of the them the destination direction pretty much resides in the ecliptic plane.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/20/2020 06:19 pm
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? if you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.
I think that you'll gain more insight if you run such what-if exercises yourself.

I suggest Silverbird Astronautics Launch Vehicle Performance Calculator (http://www.silverbirdastronautics.com/LVperform.html). Unfortunately, the SS/SH stack isn't listed so you'll have to use the "User-defined" option.

Here are some reasonable suggestions to get you going but they are by no means definitive.

Number of Stages: 2
Strap-on Boosters? No
Dry Mass (kg) 1st stage: 220000 , 2nd stage 135000 (both #s include landing propellant)
Propellant (kg) 1st stage 3400000 , 2nd stage 1200000
Thrust (KN) 1st stage 56000, 2nd stage 11000
Isp (s) 1st stage 330, 2nd stage 350
Default Propellant Residuals? Yes
Restartable Upper Stage? Yes
Payload Fairing: 0 and 0
Launch Site: Cape Canaveral (USA)
Destination:
Earth Orbit
Apogee (km): 300
Perigee (km): 300
Inclination (deg): 28.4
Trajectory: Optimal

"Calculate" yields an estimated payload of about 100 tonnes. SpaceX hopes to get that to 150 tonnes by various improvements (and unlike me, they don't have to guesstimate what the vehicle specs are). Leave off the second stage and see what SH can do by itself. You won't be impressed.

The web-page will even estimate the payload for an escape trajectory (try Moon @ C3=-2 or Mars @ C3=13 km^2/s^2). Spoiler alert: it's 0. As everyone knows, you need to refuel Starship to go beyond Earth orbit and that means lifting up to 1200 tonnes of propellant into LEO. So what's the best way to do that?

You could use expendable tankers which would gain you back the landing propellant (15 tonnes) and the various other items needed for landing (probably around 25 tonnes). So expendable tankers might carry 40% more propellant than reusable ones. Even ignoring the significant cost of expending the tankers, do you really think that you can build two expendable tankers in less time than you can launch a single tanker three times?

Quote
I got a lot of things that make me skeptical from low maturity technology to financial plan lacking realism and no sustainable viable business plan. :-\

[rant]
Fortunately for SpaceX, they don't have to convince you. They've already got investors who see the value in developing Starship and other technologies even if the motivation might not be for Mars. And, they've got the world's 4th richest person who calls the shots and is motivated to make it happen. Perhaps you've noticed that there's a lot going on in Boca Chica and elsewhere these days. Maybe cities on Mars don't pan out but Elon's giving it a shot and I for one wouldn't bet against him.

And as a Space advocate *and* US taxpayer, the bang for the buck of SpaceX's vision is off-scale high.
[/rant]
A useful metric I've come up with is the increase in tank volume allowed by one ton of mass savings. One ton of propellant, O2 and CH4 at 3.6:1, allows a consolidated tank stretch of 0.016m. Using this saves a lot of math.


So, when discussing a tanker intended to become a no return depot, every ton of heatshield, fins & mechanism and landing legs saved allow a 0.016m tank stretch while  allowing maximum propellant on board and keeping mass constant. Reducing propellant load allows greater depot tank volume at the cost of less propellant delivered on that launch.


All the other variables, max mass to LEO and  using different orbits being a couple biggies, are still there but this helps give an easily calculated apples to apples baseline.


I don't exactly hate math, but I do struggle.


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: yg1968 on 08/20/2020 06:22 pm
I am highly skeptical of this information but there is a guy on Twitter that says that lunar Starship requires 12 refueling flights in order to get to the NRHO. He says that he got this information from someone involved in the HLS program at MSFC.

I would expect only one or two refueling flights would be required, not 12. What is a realistic number of re-fueling missions required for lunar Starship to get to NRHO?

https://twitter.com/Kerb2024/status/1296485561306419203
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/20/2020 06:52 pm
there is a guy on Twitter that says that lunar Starship requires 12 refueling flights in order to get to the NRHO. He says that he got this information from someone involved in the HLS program at MSFC.

I assume they just went: Starship holds 1,200 tonnes of propellant. If one Starship can launch 100t into LEO, then 1200/100=12 launches. There's no analysis of NRHO, or delta-v, or anything deeper than that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/20/2020 06:55 pm
Quote from: armchairfan

As an exercise and if I've done the math right -- a big ask -- a 340 km circular orbit at 28.4 inclination (Cape Canaveral) will precess at such a rate as to be aligned toward Mars every 26 months; i.e. when the next launch window rolls around. While that's a fairly low orbit, (I'm guessing that) you can probably keep it going with just the propellant that you'd otherwise lose to boil-off.

This probably only makes sense if you're going to use an entire synod sending tankers to a depot(s) in preparation for a vast fleet of starships leaving all at once.

Thanks. I was trying to understand if such an orbit existed or not.

This was discussed earlier in the thread and I realized that I missed the page where it came to a conclusion. So I won't continue necro'ing the discussion beyond to offer my own $0.02. Initially it will make more sense to just launch a train of tanker starships (say 1 every 12-24 hr based on your twice daily launch window and presumably some time to do refueling ops for each tanker) to rendezvous with the mission starship shortly before the intended launch. So you'd only spend a few days to a week on orbit refueling.

However, when sending many many starships that launch cadence doesn't seem sustainable to me over a short mars transfer window even with ksc, bc, and maybe an offshore platform or two pitching in. So an accumulation tanker in a carefully selected orbit like you calculated would make sense to allow you to preposition fuel for a launch window. By this time the use case will be existent not theoretical and launch costs would have been massively reduced so one could justify the added costs of on orbit construction for a dedicated facility.

That said it may prove out that a rapid reuse and launch cadence is easier to solve than in space depot construction. It will certainly need to be if point to point is going to be a serious proposition.

Either way holy smokes are you looking at a lot of launches for a mars bound fleet.
Yup, a buttload of launches. But a depot(s) doesn't have to involve in space construction.


A standard SS in LEO for example, when filled by tankers would hold enough propellant to support one SS going to Mars. If the mars fleet is 10 ships, use 10 SS's accumulating fuel, which can then return to earth.


For another example, strip everything unnecessary and increase capacity for a permanent no return depot.  My very rough calculation shows this could fill 1.2 mars bound SS's. If it is structurally possible to use the three remaining rings as tankage and stretch an SS by an estimate five additional rings the depot could tank up two mars bound ships. Of course it could not launch with all this tankage filled with propellant.


This does not change the number of tankers to orbit but it does add safety by limiting each mars bound ship to one refueling and it avoids the difficulty of on orbit construction.


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 08/20/2020 07:17 pm
I am highly skeptical of this information but there is a guy on Twitter that says that lunar Starship requires 12 refueling flights in order to get to the NRHO. He says that he got this information from someone involved in the HLS program at MSFC.

I would expect only one or two refueling flights would be required, not 12. What is a realistic number of re-fueling missions required for lunar Starship to get to NRHO?

https://twitter.com/Kerb2024/status/1296485561306419203
Quick check on wikipedia: 3.2km/s TLI from LEO, 0.43km/s powered flyby and insertion into NRHO. Lets say standard Starship is 120t empty with 100t payload to LEO and that payload mass can be traded for residual fuel 1:1. Lets use 365 for average in space Isp.

Maths:
Empty Starship from LEO to Gateway and back to reentry/aerobreaking is 3.2+2*0.43=4.06km/s.
Rocket equation gives initial mass as 373t, so 253t of fuel or 100t residual from initial launch and two refuelings with 100t and 53t respectively. Full refueling of 200t to initial mass of 420t gives 15t of payload to Gateway and back or 17t payload and empty back.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/20/2020 07:56 pm
...
Escape velocity is 11.2kps. So for a deltaV of under 3.45kps the ship can raise itself to an apogee of any arbitrary height and then upon return to perigee apply the same dV to circularize it's orbit. Total budget is 6.9kps.
There really isn't a plane change when going interplanetary. As you approach escape velocity the tangential  component drops to zero. The tangential component is what the "plane change" changes.
...

That's not accurate. All departure burns accelerate the vehicle prograde relative to Earth, increasing only the tangential component of the velocity. If your LEO orbit plane is parallel to the Sun-Earth line, you have a problem, because you need to go in the direction the Earth is travelling around the Sun, which is 90 degrees to the Earth-Sun line. That requires a plane change because at no point in the orbit does your velocity vector point in that direction.

But you don't need to raise the entire depot far from Earth to change planes. Just let nodal precession do that for you, changing altitude slightly to take advantage of different precession rates.

So I've been lurking on this thread for awhile and this is a critical point that has come up several times around the depot concept.

The key question is does launching from an inclined orbit have a meaningful effect on interplanetary transfers?

If no, then you can have a general purpose depot in LEO. If yes, then you can only have mission specific depots. I'm not sure I've seen a clear consensus in this thread and I've been doing some noodling and reading so I'm going to take a stab at it.

Caveat: I am entirely kerbal trained plus I work in the soft sciences and math makes me slightly nauseated (insert Luke walking into the cantina gif here) so take this for what it is worth.

I think the answer is yes and no. Obviously you can launch to Mars from an inclined orbit. Three vehicles just did it last month. I highly doubt they spent the dv to plane change into equatorial orbits.

However I think your inclined orbit has to be pointed in the right direction (AN and DN aligned in a certain way relative to your intended ejection angle).

This is no big deal for a specific mission. Do some maths and launch your ships at the right time of day basically (though as previously discussed you'd have limited launch windows). But I really do think it means you can't easily have a depot in an inclined orbit to service multiple different interplanetary transfers.

You also want it in LEO for oberth.

However, despair not general purpose depot proponents.

It occurs to me that the problem fundamentally is inclined orbits which occur because KSC is not on the equator and can't easily be moved there.

But if one was an ambitious billionaire planning on turning an oil rig into a launch platform... Well now...<stokes beard thoughtfully>...
An extra point to consider. Earth is tipped 23.5 deg to our orbital plane so an equatorial launch faces the same limitations of node alignment with the ecliptic.


I don't think there is any place on earth without this problem. It's something we have to live with until we can afford to take it to the shop for an axis alignment


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/20/2020 08:04 pm
The orbital plane is relatively fixed as Earth moves around the sun, hence it will move in and out of sync with the ecliptic. So the orbital plane of a a particular LEO orbit will have two opportunities per year to launch into an interplanetary trajectory without a massive plane-change manoeuvre.
this particular quirk of orbital mechanics is why I doubt Phobos or Deimos will ever (?) be used as way-stations/depots on the way to Mars surface.  Only twice per Martian year would it be possible to arrive at Mars and capture into an orbital plane aligned with the moons... and how often would those arrival windows synchronise with the Earth-Mars synodic cycle?  You might be able to get around this by performing multiple plane change maneuvers whilst aero-capturing into Martian orbit, but there would have to be an upper limit to how great a plane change you could perform that way.

Just one plane change.

You aero-capture into a very high Mars orbit, putting apoapsis right on the edge of the Hill Sphere (or even fractionally beyond), then do a plane change at apoapsis to match the destination orbital plane. At this distance, the delta-v cost of even a 90 degree plane change is in the low hundreds of m/s. On the next pass, do a second smaller aero-capture to lower apoapsis to match the altitude of the target orbit. (Then do your circularisation burn when the orbital phasing catches up to the target.) Adds about two weeks or so of additional travel for the passengers. Not a deal-breaker if the ship has sustained months-long interplanetary flight.

You might be able to perform a plane change during the aero-capture, but it's not necessary. And with Mars' variable atmosphere, doing a high plane-change is probably less risky.

If Phobos or Deimos have tethers, you can get away with one aero-capture. There's a point on the tethers that matches hyperbolic velocity, so capturing at near-hyperbolic velocity is possible.

(Note that this also applied to docking with Mars space elevators, only with the added fun of oscillating the elevator to avoid Phobos crashing into it. The space elevator is in Mars' equatorial plane (roughly), which is at 25 degrees to its solar orbital plane. You need at least one plane change manoeuvre.)



You can do the same thing when returning to Earth. Do a burn from Phobos (say) to the edge of the Hill Sphere. Do a plane-change burn at apoapsis, where it costs virtually nothing. Then on periapsis, do the interplanetary burn. You preserve the velocity from your boost into high-orbit as excess velocity when you swing back near the planet, so Mr. Oberth is happy. And your interplanetary burn also happens while low, so he's even happier.

Earth orbit is a little more complex because you want to minimise passes through the Van Allen belts. But in theory, you can do the same when leaving Earth from a 28 degree inclination "universal depot" that is even 90 degrees out-of-sync with the exit plane. Costs you two extra passes through the belts, but you have a "storm shelter" on any interplanetary ship, use it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 08/20/2020 08:09 pm
Some smart people with names you might recognize have put some relevant musings on paper.

Selinanboondocks blog post. (https://selenianboondocks.com/2018/02/aas-paper-review-practical-methodologies-for-low-delta-v-penalty-on-time-departures-to-arbitrary-interplanetary-destinations-from-a-medium-inclination-low-earth-orbit-depot/)

https://selenianboondocks.com/wp-content/uploads/2018/02/AAS-17-696-1.pdf

https://selenianboondocks.com/wp-content/uploads/2018/09/RAAN-Agnostic_3-BurnPaperV2_AAS-18-447-1.pdf

:)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/20/2020 08:14 pm
However I think your inclined orbit has to be pointed in the right direction (AN and DN aligned in a certain way relative to your intended ejection angle).
This is no big deal for a specific mission. Do some maths and launch your ships at the right time of day basically (though as previously discussed you'd have limited launch windows). But I really do think it means you can't easily have a depot in an inclined orbit to service multiple different interplanetary transfers.
An extra point to consider. Earth is tipped 23.5 deg to our orbital plane so an equatorial launch faces the same limitations of node alignment with the ecliptic.
I don't think there is any place on earth without this problem. It's something we have to live with until we can afford to take it to the shop for an axis alignment

As Pahimarus correctly notes, a launch site effectively rotates its orbital plane 360 degrees every day. Therefore there's a launch window every day (or two) that can put you into an orbit that matches your desired exit plane.

The issue with permanent depots is changing that orbital plane after it's in orbit.

But if "propellant is cheap", then doing the low-high-low manoeuvre gives you a cheap plane-change, provided you can handle the two extra passes through the VA belts.

[edit: eriblo - Yeah, it was either Jon Goff or Hollister David where I first read about low-high-low plane-changes.]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: yg1968 on 08/20/2020 08:57 pm
I am highly skeptical of this information but there is a guy on Twitter that says that lunar Starship requires 12 refueling flights in order to get to the NRHO. He says that he got this information from someone involved in the HLS program at MSFC.

I would expect only one or two refueling flights would be required, not 12. What is a realistic number of re-fueling missions required for lunar Starship to get to NRHO?

https://twitter.com/Kerb2024/status/1296485561306419203
Quick check on wikipedia: 3.2km/s TLI from LEO, 0.43km/s powered flyby and insertion into NRHO. Lets say standard Starship is 120t empty with 100t payload to LEO and that payload mass can be traded for residual fuel 1:1. Lets use 365 for average in space Isp.

Maths:
Empty Starship from LEO to Gateway and back to reentry/aerobreaking is 3.2+2*0.43=4.06km/s.
Rocket equation gives initial mass as 373t, so 253t of fuel or 100t residual from initial launch and two refuelings with 100t and 53t respectively. Full refueling of 200t to initial mass of 420t gives 15t of payload to Gateway and back or 17t payload and empty back.

Thanks! I am not engineer, so I am not sure that I fully understand, you are saying that two refueling missions should be enough to deliver 15mt to the Moon and back (or 17mt one way to the Moon)?

Would it be possible for lunar Starship to get to the Moon (through NRHO) and back to LEO (through NRHO) without any re-fueling?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 08/20/2020 09:18 pm
If Phobos or Deimos have tethers, you can get away with one aero-capture. There's a point on the tethers that matches hyperbolic velocity, so capturing at near-hyperbolic velocity is possible.

Capture orbit lacks hyperbolic trajectory velocity, by definition.  State your goal, terms and steps more clearly, if you want to claim something.  You made an untrue tether claim (https://forum.nasaspaceflight.com/index.php?topic=51387.msg2120985#msg2120985) recently.

this also applied to docking with Mars space elevators, only with the added fun of oscillating the elevator to avoid Phobos crashing into it.

No, the MSE-oscillation notion is outdated.  As you've seen, Dr. Lades' Mars Lift (http://www.lakematthew.com/press/press-release-september-18-2017/) calculation demonstrated that an MSE can float off the equatorial plane, never touching it.  No oscillation is required, which is good, because no engineer knows how to manage the merely notional Phobos oscillation.  1 (https://forum.nasaspaceflight.com/index.php?topic=37667.msg1951870#msg1951870) 2 (https://forum.nasaspaceflight.com/index.php?topic=37667.msg1952008#msg1952008) 3 (https://forum.nasaspaceflight.com/index.php?topic=37667.msg1952021#msg1952021)

Image:  Numerical solutions for "Mars Lift" MSE tether curve.  High specific strength in green, low in purple.  Phobos orbit range in brown.  Dr. Martin Lades, ISEC.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 08/20/2020 11:28 pm
I am highly skeptical of this information but there is a guy on Twitter that says that lunar Starship requires 12 refueling flights in order to get to the NRHO. He says that he got this information from someone involved in the HLS program at MSFC.

I would expect only one or two refueling flights would be required, not 12. What is a realistic number of re-fueling missions required for lunar Starship to get to NRHO?

[tweet]
Quick check on wikipedia: 3.2km/s TLI from LEO, 0.43km/s powered flyby and insertion into NRHO. Lets say standard Starship is 120t empty with 100t payload to LEO and that payload mass can be traded for residual fuel 1:1. Lets use 365 for average in space Isp.

Maths:
Empty Starship from LEO to Gateway and back to reentry/aerobreaking is 3.2+2*0.43=4.06km/s.
Rocket equation gives initial mass as 373t, so 253t of fuel or 100t residual from initial launch and two refuelings with 100t and 53t respectively. Full refueling of 200t to initial mass of 420t gives 15t of payload to Gateway and back or 17t payload and empty back.

Thanks! I am not engineer, so I am not sure that I fully understand, you are saying that two refueling missions should be enough to deliver 15mt to the Moon and back (or 17mt one way to the Moon)?

Would it be possible for lunar Starship to get to the Moon (through NRHO) and back to LEO (through NRHO) without any re-fueling?
Not quite - you pass by the Moon going to NRHO but staying there (i.e. entering low lunar orbit) is harder and getting to the surface is much harder. I will also mention that the tweet said Moonship while I was talking about a standard Starship with a heat shield useful at lunar return velocities (and forgot about landing propellant, so payload a bit less). Moonship lacks heat shield and can not get back to LEO as easily.
To summarize for Starships/Moonships at 120t empty, up to 100t payload to LEO and 1200t of propellant capacity:
 
If capable of cislunar travel a standard Starship can service the Gateway from LEO with two refuelings and return.

A Moonship in LEO can not quite get to the Gateway with a single refueling (unless it is below 115t or so) but can easily get there with plenty of payload with two.

A fully refueled Moonship in LEO (i.e. the 12 refueling flights) should be able to get the full 100t payload to the Gateway :P and back to LEO propulsively (empty). With no payload it is right at the limit for getting to Gateway, down to the surface and back to the Gateway again.

This is all very simplistic, highly sensitive to performance numbers and totally ignores things like boil off or the feasibility of actually docking to the Gateway.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: CJ on 08/21/2020 12:06 am
Quote from: armchairfan

As an exercise and if I've done the math right -- a big ask -- a 340 km circular orbit at 28.4 inclination (Cape Canaveral) will precess at such a rate as to be aligned toward Mars every 26 months; i.e. when the next launch window rolls around. While that's a fairly low orbit, (I'm guessing that) you can probably keep it going with just the propellant that you'd otherwise lose to boil-off.

This probably only makes sense if you're going to use an entire synod sending tankers to a depot(s) in preparation for a vast fleet of starships leaving all at once.

Thanks. I was trying to understand if such an orbit existed or not.

This was discussed earlier in the thread and I realized that I missed the page where it came to a conclusion. So I won't continue necro'ing the discussion beyond to offer my own $0.02. Initially it will make more sense to just launch a train of tanker starships (say 1 every 12-24 hr based on your twice daily launch window and presumably some time to do refueling ops for each tanker) to rendezvous with the mission starship shortly before the intended launch. So you'd only spend a few days to a week on orbit refueling.

However, when sending many many starships that launch cadence doesn't seem sustainable to me over a short mars transfer window even with ksc, bc, and maybe an offshore platform or two pitching in. So an accumulation tanker in a carefully selected orbit like you calculated would make sense to allow you to preposition fuel for a launch window. By this time the use case will be existent not theoretical and launch costs would have been massively reduced so one could justify the added costs of on orbit construction for a dedicated facility.

That said it may prove out that a rapid reuse and launch cadence is easier to solve than in space depot construction. It will certainly need to be if point to point is going to be a serious proposition.

Either way holy smokes are you looking at a lot of launches for a mars bound fleet.
Yup, a buttload of launches. But a depot(s) doesn't have to involve in space construction.


A standard SS in LEO for example, when filled by tankers would hold enough propellant to support one SS going to Mars. If the mars fleet is 10 ships, use 10 SS's accumulating fuel, which can then return to earth.


For another example, strip everything unnecessary and increase capacity for a permanent no return depot.  My very rough calculation shows this could fill 1.2 mars bound SS's. If it is structurally possible to use the three remaining rings as tankage and stretch an SS by an estimate five additional rings the depot could tank up two mars bound ships. Of course it could not launch with all this tankage filled with propellant.


This does not change the number of tankers to orbit but it does add safety by limiting each mars bound ship to one refueling and it avoids the difficulty of on orbit construction.


Phil

I keep wondering if having a depot (enough to refuel, say, 3 Starships) might be a viable future development.

Would it be plausible (and viable) to use a Superheavy near the end of its service life as a depot? It'd need a few modifications, such as refueling compatibility with Starships (both ways, of course, seeing as the prop has to get there somehow). It would also need something akin to IVF (Integrated Vehicle Fluids, basically a tiny internal combustion engine to provide power for temp, pressurization, station keeping, etc).

This, of course, assumes you can get a stripped-down (no recovery hardware, no fins, fewer Raptors, etc) Superheavy to do SSTO. It appears possible, assuming an aggressive launch profile, and assuming (yeah, lots of assumptions here, not good) that projected dry mass is in the ballpark. It would need a nosecone of some sort, as light as possible, perhaps modeled on the FH side booster nosecones.

It would, I think, require a sunshade before adding prop. Seeing as this notional Superheavy would arrive in LEO with  no payload and essentially no remaining prop, the sunshade could be added later, by a Starship. For a sunshade, I'm thinking something roughly akin to the Echo2 satalite, which was essentially a giant Mylar spherical balloon (self-rigidizing, so no need to keep it pressurized). It would require very little gas to inflate it in orbit (perhaps tank residual?). I can't find a figure for Echo 2, except a mention of "a few pounds" of gas. Echo 2 was a 40 meter diameter sphere, weighing  256 kg (including other hardware, like the inflation tank). Deflated, it fit inside the payload fairing of a Thor-Agena (looks like a 65 inch diameter fairing). A couple of 40 meter spheres might be sub-optimum for a sunshade for a Superheavy, so why not make it a tube?  A tube with a 10 meter internal diameter and a 12 meter external diameter would fit around SH. Even if several times the mass of Echo2, this is still under a tonne. It would also provide micrometeorite and small orbital debris protection for the depot. (and, something similar but smaller might be useful for Starship on a Mars voyage). As for how to emplace it once inflated, I'm not sure (unless cargo Starships will have a RMS). Perhaps add a cold gas thruster pack from a F9 or a fairing, nudge it into position, then hold it in place with a few extendable rods (like old-style extending car radio antennas - maybe literally use those).   

Might this concept be a way to create an orbital depot (capable of holding 3400 tons of prop) if such a depot would be useful? Perhaps, if sufficient end-of-life Superheavies are available, deploy one in each of, say, 3 inclinations?

Many things could make this non-viable; cost, the refueling system or IVF system might require too many modifications to SH, SH's systems might be incompatible with this purpose, the "sunshade" might be inadequate to reduce thermal influx to manageable levels, etc, . (and, if a stripped down SH can't do SSTO). Or, of course, this whole concept might be more trouble and expense than just using Starships as temporary depots.  There's also the fact that it would require a Superheavy at the end of its service life, and around 20 Raptors likewise at the end of their service life. I have no doubt that I am failing to see  some other potential dealkillers.

A couple of links; Integrated Vehicle Fluids
https://arc.aiaa.org/doi/abs/10.2514/6.2012-5302
(I know this is for HydroLox, but a similar system for MethLox ought to be possible)

Echo 2
http://www.astronautix.com/e/echo2.html

     
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/21/2020 12:33 am
However I think your inclined orbit has to be pointed in the right direction (AN and DN aligned in a certain way relative to your intended ejection angle).
This is no big deal for a specific mission. Do some maths and launch your ships at the right time of day basically (though as previously discussed you'd have limited launch windows). But I really do think it means you can't easily have a depot in an inclined orbit to service multiple different interplanetary transfers.
An extra point to consider. Earth is tipped 23.5 deg to our orbital plane so an equatorial launch faces the same limitations of node alignment with the ecliptic.
I don't think there is any place on earth without this problem. It's something we have to live with until we can afford to take it to the shop for an axis alignment

As Pahimarus correctly notes, a launch site effectively rotates its orbital plane 360 degrees every day. Therefore there's a launch window every day (or two) that can put you into an orbit that matches your desired exit plane.

The issue with permanent depots is changing that orbital plane after it's in orbit.

But if "propellant is cheap", then doing the low-high-low manoeuvre gives you a cheap plane-change, provided you can handle the two extra passes through the VA belts.

[edit: eriblo - Yeah, it was either Jon Goff or Hollister David where I first read about low-high-low plane-changes.]
Agree. I was just pointing out that equatorial launch doesn't make much difference for this problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: yg1968 on 08/21/2020 12:56 am
I am highly skeptical of this information but there is a guy on Twitter that says that lunar Starship requires 12 refueling flights in order to get to the NRHO. He says that he got this information from someone involved in the HLS program at MSFC.

I would expect only one or two refueling flights would be required, not 12. What is a realistic number of re-fueling missions required for lunar Starship to get to NRHO?

[tweet]
Quick check on wikipedia: 3.2km/s TLI from LEO, 0.43km/s powered flyby and insertion into NRHO. Lets say standard Starship is 120t empty with 100t payload to LEO and that payload mass can be traded for residual fuel 1:1. Lets use 365 for average in space Isp.

Maths:
Empty Starship from LEO to Gateway and back to reentry/aerobreaking is 3.2+2*0.43=4.06km/s.
Rocket equation gives initial mass as 373t, so 253t of fuel or 100t residual from initial launch and two refuelings with 100t and 53t respectively. Full refueling of 200t to initial mass of 420t gives 15t of payload to Gateway and back or 17t payload and empty back.

Thanks! I am not engineer, so I am not sure that I fully understand, you are saying that two refueling missions should be enough to deliver 15mt to the Moon and back (or 17mt one way to the Moon)?

Would it be possible for lunar Starship to get to the Moon (through NRHO) and back to LEO (through NRHO) without any re-fueling?
Not quite - you pass by the Moon going to NRHO but staying there (i.e. entering low lunar orbit) is harder and getting to the surface is much harder. I will also mention that the tweet said Moonship while I was talking about a standard Starship with a heat shield useful at lunar return velocities (and forgot about landing propellant, so payload a bit less). Moonship lacks heat shield and can not get back to LEO as easily.
To summarize for Starships/Moonships at 120t empty, up to 100t payload to LEO and 1200t of propellant capacity:
 
If capable of cislunar travel a standard Starship can service the Gateway from LEO with two refuelings and return.

A Moonship in LEO can not quite get to the Gateway with a single refueling (unless it is below 115t or so) but can easily get there with plenty of payload with two.

A fully refueled Moonship in LEO (i.e. the 12 refueling flights) should be able to get the full 100t payload to the Gateway :P and back to LEO propulsively (empty). With no payload it is right at the limit for getting to Gateway, down to the surface and back to the Gateway again.

This is all very simplistic, highly sensitive to performance numbers and totally ignores things like boil off or the feasibility of actually docking to the Gateway.

I think that the tweet meant lunar Starship would pick up the astronauts at NRHO and then continue to the Moon since that is what lunar Starship is meant to do. Lunar Starship isn't meant for ferrying cargo to Gateway (that should be Dragon XL's job).

My question is what is the minimum amount of refueling missions which will be necessary in order for lunar Starhip to bring 2 astronauts to the Moon (through NRHO) with a minimum amount of cargo and then come back from the Moon to LEO (through NRHO).

I believe that is the itinerary for lunar Starship. It will travel from Earth to the Moon (through NRHO) and it will come back from the Moon to LEO through NRHO. My understanding of lunar Starship is that it will be re-fueled in LEO, not in NRHO. 

In any event, thanks for your answers. I have asked this question before but it's been hard to get an answer. It's possible that this question has been answered elsewhere but I haven't seen it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/21/2020 12:59 am
Quote from: armchairfan

As an exercise and if I've done the math right -- a big ask -- a 340 km circular orbit at 28.4 inclination (Cape Canaveral) will precess at such a rate as to be aligned toward Mars every 26 months; i.e. when the next launch window rolls around. While that's a fairly low orbit, (I'm guessing that) you can probably keep it going with just the propellant that you'd otherwise lose to boil-off.

This probably only makes sense if you're going to use an entire synod sending tankers to a depot(s) in preparation for a vast fleet of starships leaving all at once.

Thanks. I was trying to understand if such an orbit existed or not.

This was discussed earlier in the thread and I realized that I missed the page where it came to a conclusion. So I won't continue necro'ing the discussion beyond to offer my own $0.02. Initially it will make more sense to just launch a train of tanker starships (say 1 every 12-24 hr based on your twice daily launch window and presumably some time to do refueling ops for each tanker) to rendezvous with the mission starship shortly before the intended launch. So you'd only spend a few days to a week on orbit refueling.

However, when sending many many starships that launch cadence doesn't seem sustainable to me over a short mars transfer window even with ksc, bc, and maybe an offshore platform or two pitching in. So an accumulation tanker in a carefully selected orbit like you calculated would make sense to allow you to preposition fuel for a launch window. By this time the use case will be existent not theoretical and launch costs would have been massively reduced so one could justify the added costs of on orbit construction for a dedicated facility.

That said it may prove out that a rapid reuse and launch cadence is easier to solve than in space depot construction. It will certainly need to be if point to point is going to be a serious proposition.

Either way holy smokes are you looking at a lot of launches for a mars bound fleet.
Yup, a buttload of launches. But a depot(s) doesn't have to involve in space construction.


A standard SS in LEO for example, when filled by tankers would hold enough propellant to support one SS going to Mars. If the mars fleet is 10 ships, use 10 SS's accumulating fuel, which can then return to earth.


For another example, strip everything unnecessary and increase capacity for a permanent no return depot.  My very rough calculation shows this could fill 1.2 mars bound SS's. If it is structurally possible to use the three remaining rings as tankage and stretch an SS by an estimate five additional rings the depot could tank up two mars bound ships. Of course it could not launch with all this tankage filled with propellant.


This does not change the number of tankers to orbit but it does add safety by limiting each mars bound ship to one refueling and it avoids the difficulty of on orbit construction.


Phil

I keep wondering if having a depot (enough to refuel, say, 3 Starships) might be a viable future development.

Would it be plausible (and viable) to use a Superheavy near the end of its service life as a depot? It'd need a few modifications, such as refueling compatibility with Starships (both ways, of course, seeing as the prop has to get there somehow). It would also need something akin to IVF (Integrated Vehicle Fluids, basically a tiny internal combustion engine to provide power for temp, pressurization, station keeping, etc).

This, of course, assumes you can get a stripped-down (no recovery hardware, no fins, fewer Raptors, etc) Superheavy to do SSTO. It appears possible, assuming an aggressive launch profile, and assuming (yeah, lots of assumptions here, not good) that projected dry mass is in the ballpark. It would need a nosecone of some sort, as light as possible, perhaps modeled on the FH side booster nosecones.

It would, I think, require a sunshade before adding prop. Seeing as this notional Superheavy would arrive in LEO with  no payload and essentially no remaining prop, the sunshade could be added later, by a Starship. For a sunshade, I'm thinking something roughly akin to the Echo2 satalite, which was essentially a giant Mylar spherical balloon (self-rigidizing, so no need to keep it pressurized). It would require very little gas to inflate it in orbit (perhaps tank residual?). I can't find a figure for Echo 2, except a mention of "a few pounds" of gas. Echo 2 was a 40 meter diameter sphere, weighing  256 kg (including other hardware, like the inflation tank). Deflated, it fit inside the payload fairing of a Thor-Agena (looks like a 65 inch diameter fairing). A couple of 40 meter spheres might be sub-optimum for a sunshade for a Superheavy, so why not make it a tube?  A tube with a 10 meter internal diameter and a 12 meter external diameter would fit around SH. Even if several times the mass of Echo2, this is still under a tonne. It would also provide micrometeorite and small orbital debris protection for the depot. (and, something similar but smaller might be useful for Starship on a Mars voyage). As for how to emplace it once inflated, I'm not sure (unless cargo Starships will have a RMS). Perhaps add a cold gas thruster pack from a F9 or a fairing, nudge it into position, then hold it in place with a few extendable rods (like old-style extending car radio antennas - maybe literally use those).   

Might this concept be a way to create an orbital depot (capable of holding 3400 tons of prop) if such a depot would be useful? Perhaps, if sufficient end-of-life Superheavies are available, deploy one in each of, say, 3 inclinations?

Many things could make this non-viable; cost, the refueling system or IVF system might require too many modifications to SH, SH's systems might be incompatible with this purpose, the "sunshade" might be inadequate to reduce thermal influx to manageable levels, etc, . (and, if a stripped down SH can't do SSTO). Or, of course, this whole concept might be more trouble and expense than just using Starships as temporary depots.  There's also the fact that it would require a Superheavy at the end of its service life, and around 20 Raptors likewise at the end of their service life. I have no doubt that I am failing to see  some other potential dealkillers.

A couple of links; Integrated Vehicle Fluids
https://arc.aiaa.org/doi/abs/10.2514/6.2012-5302 (https://arc.aiaa.org/doi/abs/10.2514/6.2012-5302)
(I know this is for HydroLox, but a similar system for MethLox ought to be possible)

Echo 2
http://www.astronautix.com/e/echo2.html (http://www.astronautix.com/e/echo2.html)

     
I've been noodling an SSTO SH and don't know if it's possible, but if it is, a reasonable way to go - but not an early priority.


The SH has no heatshield or fins so that's a job that's not needed. It's already designed to interface with SS plumbing - the same plumbing SS will use to refuel. No extra work there. It will need a sun shade, maybe an earth shade, a fuel fridge and a radiator. The fuel fridge can be PV powered but I don't have a feel for power needs.


The big Q is getting SH to orbit SSTO. Beyond my skills.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: yg1968 on 08/21/2020 01:46 am
Here is another tweets that says lunar Starship needs a lot of re-fueling flights:

https://twitter.com/joe_mckirdy/status/1296563733519577090

Quote
Starship is estimated to have a dry mass of 100 metric tons. To refuel it is LEO to fly to the Gateway with its payload of 100 tons would require 300 tons of liquid oxygen and methane (LOx/CH4) propellant. This could be supported with three tanker Starship flights to LEO. But then to land on the moon, discharge its cargo, and return to the Gateway would require a further 400 tons of propellant delivered to the Gateway, or four trans-lunar tankers. Each of these would need to be enabled for flight from LEO to the gateway by three further tanker flights, for a total of 20 Starship launches for each piloted lunar mission. That doesn’t make sense.

https://spacenews.com/op-ed-toward-a-coherent-artemis-plan/
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/21/2020 01:51 am
It occurred to me that thermal shields will necessarily be reflective and might cause a reaction from the astronomical community. No nearly as numerous as Starlinks but the good neighbor thing...


The shields need to run interference from the sun and earth which implies constant repositioning of the shades and maybe the whole depot. the ISS used Control Moment Gyros (CMG) instead of reaction jets. Would they be justified for a depot?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: envy887 on 08/21/2020 02:09 am
Here is a couple of other tweets that says lunar Starship needs a lot of re-fueling flights:

Those tweets are complete garbageposts and demonstrate a total lack of even the most basic understanding of any of the mechanics needed to optimize a refueling solution.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: yg1968 on 08/21/2020 02:22 am
Here is a couple of other tweets that says lunar Starship needs a lot of re-fueling flights:

Those tweets are complete garbageposts and demonstrate a total lack of even the most basic understanding of any of the mechanics needed to optimize a refueling solution.

I think that they are based on Zubrin's article but his article assumed a 100mt payload to the Moon.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: yg1968 on 08/21/2020 03:17 am
I was reading comments to a couple of posts to some Space News articles and some of the posters say 8 to 12 missions for a lunar landing.

Quote from: gunsandrockets
To land that much cargo as a one way expendable cargo mission requires the support of 6 tanker Starship missions to LEO. To land that much cargo as a reusable lander, each HLS Starship mission will require the support of a minimum of 11 tanker Starship missions plus 1 cargo Starship mission, more likely 12 tankers plus 2 cargo Starships.

http://disq.us/p/292z8r9

Quote from: envy
A single Starship tanker would be refueled in LEO with a total of about 8 launches. This single tanker would make 1 flight to the Gateway and deliver about 400 tons of propellant to the HLS Starship lander, which is enough for the HLS Starship lander to make the round trip to the surface and return to Gateway. The tanker would return to Earth, and participate in the next set of about 8 launches needed to again send it to gateway to refill the HLS lander.

So only about 8 Starship/SuperHeavy launches, and not 20 (!!!), are needed for each landing.

http://disq.us/p/29c9ih2
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 08/21/2020 04:02 am
It will need a sun shade, maybe an earth shade, a fuel fridge and a radiator. The fuel fridge can be PV powered but I don't have a feel for power needs.

Gravlee et al. 2012 anticipates a 2 W Stirling cooler to refrigerate 71 kg of LH2.

Image:  CRYOTE Grande.  Gravlee et al. 2012, Fig. 10.

Refs.

Gravlee, M., Kutter, B., McLean, C. and Marquardt, J., 2012. Cryogenic Orbital Testbed (CRYOTE) Development Status. (https://www.ulalaunch.com/docs/default-source/extended-duration/cryogenic-orbital-test-bed-development.pdf) Cryogenics, 52(4-6), pp.231-235.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 08/21/2020 06:55 am

I keep wondering if having a depot (enough to refuel, say, 3 Starships) might be a viable future development.

Would it be plausible (and viable) to use a Superheavy near the end of its service life as a depot? It'd need a few modifications, such as refueling compatibility with Starships (both ways, of course, seeing as the prop has to get there somehow). It would also need something akin to IVF (Integrated Vehicle Fluids, basically a tiny internal combustion engine to provide power for temp, pressurization, station keeping, etc).
...
I've been noodling an SSTO SH and don't know if it's possible, but if it is, a reasonable way to go - but not an early priority.

The SH has no heatshield or fins so that's a job that's not needed. It's already designed to interface with SS plumbing - the same plumbing SS will use to refuel. No extra work there. It will need a sun shade, maybe an earth shade, a fuel fridge and a radiator. The fuel fridge can be PV powered but I don't have a feel for power needs.

The big Q is getting SH to orbit SSTO. Beyond my skills.

Why would you want a Super Heavy with all those extra engines as a depot?

A modified Starship with no legs, fins, heat-shield or header tanks, and the main tanks extended to almost full height should hold close to 2000 t of propellant in orbit. It would have to launch with partially filled tanks, getting to LEO with maybe 200 t remaining.
Or a little less if some of that mass is replaced by a cryo-cooler, solar panels and a sun-shield.

Other options include an extra long version, or maybe a 12 m diameter version, holding 3500+ t.  Not needing an EDL capability opens up a lot of options.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: SoTOP on 08/21/2020 07:29 am
Why would you want a Super Heavy with all those extra engines as a depot?

A modified Starship with no legs, fins, heat-shield or header tanks, and the main tanks extended to almost full height should hold close to 2000 t of propellant in orbit. It would have to launch with partially filled tanks, getting to LEO with maybe 200 t remaining.
Or a little less if some of that mass is replaced by a cryo-cooler, solar panels and a sun-shield.

Other options include an extra long version, or maybe a 12 m diameter version, holding 3500+ t.  Not needing an EDL capability opens up a lot of options.
You can modify SH the same way you suggest to modify Starship. For example, have just enough engines to get into orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: mikelepage on 08/21/2020 07:45 am
The orbital plane is relatively fixed as Earth moves around the sun, hence it will move in and out of sync with the ecliptic. So the orbital plane of a a particular LEO orbit will have two opportunities per year to launch into an interplanetary trajectory without a massive plane-change manoeuvre.
this particular quirk of orbital mechanics is why I doubt Phobos or Deimos will ever (?) be used as way-stations/depots on the way to Mars surface.  Only twice per Martian year would it be possible to arrive at Mars and capture into an orbital plane aligned with the moons... and how often would those arrival windows synchronise with the Earth-Mars synodic cycle?  You might be able to get around this by performing multiple plane change maneuvers whilst aero-capturing into Martian orbit, but there would have to be an upper limit to how great a plane change you could perform that way.

Just one plane change.

You aero-capture into a very high Mars orbit, putting apoapsis right on the edge of the Hill Sphere (or even fractionally beyond), then do a plane change at apoapsis to match the destination orbital plane. At this distance, the delta-v cost of even a 90 degree plane change is in the low hundreds of m/s. On the next pass, do a second smaller aero-capture to lower apoapsis to match the altitude of the target orbit. (Then do your circularisation burn when the orbital phasing catches up to the target.) Adds about two weeks or so of additional travel for the passengers. Not a deal-breaker if the ship has sustained months-long interplanetary flight.


I don't want to derail this thread any further, so I went back and found the "Station on Phobos" thread where Hop_David put the problem that I was getting at into technical terms.  Also helped me refresh my memory on how I came to that thinking.  Never got my final question answered though, so maybe those of us interested can continue the discussion there?

https://forum.nasaspaceflight.com/index.php?topic=40176.100

TLDR: Assuming you've aerobraked into high martian orbit, the plane change to become co-orbital with Phobos/Deimos is easy at apoapsis, *if* the longitude of ascending node of your HMO (which is constrained by your original approach vector/timing) is aligned with that of Phobos/Deimos equatorial orbit.  Also, because the moons' orbits are more-or-less equatorial, you would have to make yet another plane change maneuver to get to/from the moons to any higher inclination site on Mars' surface - hence why I said I doubt they will be used "way-stations" (though I could still see them eventually becoming destinations in their own right).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 08/21/2020 12:38 pm
You can modify SH the same way you suggest to modify Starship. For example, have just enough engines to get into orbit.
So you go from 31 engines to...  25 engines?  20 engines?  You're still going to be throwing away millions of dollars in engines to gain an unclear advantage over using recoverable (or even expendable) Starships in exchange for something like 20% more storage volume on-orbit and even then, you have razor thin margins for hardware like insulation & coolers to maintain the propellants so it's an expensive, capital-rich expenditure for a difficult to quantify benefit, isn't it?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: envy887 on 08/21/2020 03:07 pm
Here is a couple of other tweets that says lunar Starship needs a lot of re-fueling flights:

Those tweets are complete garbageposts and demonstrate a total lack of even the most basic understanding of any of the mechanics needed to optimize a refueling solution.

I think that they are based on Zubrin's article but his article assumed a 100mt payload to the Moon.

Unless we're landing hundreds of people at a time, there's no reason to assume that 100 tonnes come back. Coming back is the hard part.

For 100 t down and 5 t up, as one might expect for a heavy cargo lander with minimalist crew return, then the HLS Starship lander only needs about 400 t of propellant delivered to NRHO per landing. A single tanker can bring this from LEO.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 08/21/2020 03:29 pm
The orbital plane is relatively fixed as Earth moves around the sun, hence it will move in and out of sync with the ecliptic. So the orbital plane of a a particular LEO orbit will have two opportunities per year to launch into an interplanetary trajectory without a massive plane-change manoeuvre.
this particular quirk of orbital mechanics is why I doubt Phobos or Deimos will ever (?) be used as way-stations/depots on the way to Mars surface.  Only twice per Martian year would it be possible to arrive at Mars and capture into an orbital plane aligned with the moons... and how often would those arrival windows synchronise with the Earth-Mars synodic cycle?  You might be able to get around this by performing multiple plane change maneuvers whilst aero-capturing into Martian orbit, but there would have to be an upper limit to how great a plane change you could perform that way.

Just one plane change.

You aero-capture into a very high Mars orbit, putting apoapsis right on the edge of the Hill Sphere (or even fractionally beyond), then do a plane change at apoapsis to match the destination orbital plane. At this distance, the delta-v cost of even a 90 degree plane change is in the low hundreds of m/s. On the next pass, do a second smaller aero-capture to lower apoapsis to match the altitude of the target orbit. (Then do your circularisation burn when the orbital phasing catches up to the target.) Adds about two weeks or so of additional travel for the passengers. Not a deal-breaker if the ship has sustained months-long interplanetary flight.


I don't want to derail this thread any further, so I went back and found the "Station on Phobos" thread where Hop_David put the problem that I was getting at into technical terms.  Also helped me refresh my memory on how I came to that thinking.  Never got my final question answered though, so maybe those of us interested can continue the discussion there?

https://forum.nasaspaceflight.com/index.php?topic=40176.100

TLDR: Assuming you've aerobraked into high martian orbit, the plane change to become co-orbital with Phobos/Deimos is easy at apoapsis, *if* the longitude of ascending node of your HMO (which is constrained by your original approach vector/timing) is aligned with that of Phobos/Deimos equatorial orbit.  Also, because the moons' orbits are more-or-less equatorial, you would have to make yet another plane change maneuver to get to/from the moons to any higher inclination site on Mars' surface - hence why I said I doubt they will be used "way-stations" (though I could still see them eventually becoming destinations in their own right).

Just a note: equatorial orbit doesn't have a longtitude of ascending node. It's like dividing 0 by 0.

Phobos is not exactly equatorial, but it's inclination is so low (~1°) that even if you miss it's ascending node by 180° dV to it would be small.

I don't see a point of making Phobos a waystation, as getting to higher latitudes from it is costly, but capturing to its orbit is not the reason why not.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/21/2020 03:50 pm
I was reading comments to a couple of posts to some Space News articles and some of the posters say 8 to 12 missions for a lunar landing.

Quote from: gunsandrockets
To land that much cargo as a one way expendable cargo mission requires the support of 6 tanker Starship missions to LEO. To land that much cargo as a reusable lander, each HLS Starship mission will require the support of a minimum of 11 tanker Starship missions plus 1 cargo Starship mission, more likely 12 tankers plus 2 cargo Starships.

http://disq.us/p/292z8r9 (http://disq.us/p/292z8r9)

Quote from: envy
A single Starship tanker would be refueled in LEO with a total of about 8 launches. This single tanker would make 1 flight to the Gateway and deliver about 400 tons of propellant to the HLS Starship lander, which is enough for the HLS Starship lander to make the round trip to the surface and return to Gateway. The tanker would return to Earth, and participate in the next set of about 8 launches needed to again send it to gateway to refill the HLS lander.

So only about 8 Starship/SuperHeavy launches, and not 20 (!!!), are needed for each landing.

http://disq.us/p/29c9ih2 (http://disq.us/p/29c9ih2)
ISTM the number of tanker loads is irrelevant. Cost per mission or $/kilo delivered seems more to the point.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: brainbit on 08/21/2020 04:18 pm
Could not a fuel depot in space be used to store any excess fuel from a SH, for example if SH had to spend some time in orbit saving any boil off as am assuming depot has fuel management built in. Also for SH which has to have enough fuel to land it must save fuel to be able to unload this fuel onto depot and when ready to return to earth recover said fuel. When SH needs to orbit a heavy load and cannot carry enough fuel for a landing as well, then it should be able to use fuel depot to re-fuel enough for a landing, same for SH returning from higher orbits. I like the fuel depot.  ;D   
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: philw1776 on 08/21/2020 04:30 pm
Could not a fuel depot in space be used to store any excess fuel from a SH, for example if SH had to spend some time in orbit saving any boil off as am assuming depot has fuel management built in. Also for SH which has to have enough fuel to land it must save fuel to be able to unload this fuel onto depot and when ready to return to earth recover said fuel. When SH needs to orbit a heavy load and cannot carry enough fuel for a landing as well, then it should be able to use fuel depot to re-fuel enough for a landing, same for SH returning from higher orbits. I like the fuel depot.  ;D

A SH can't land period.  It has no TPS.  It cannot re-enter at orbital speeds.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 08/21/2020 04:39 pm
Could not a fuel depot in space be used to store any excess fuel from a SH, for example if SH had to spend some time in orbit saving any boil off as am assuming depot has fuel management built in. Also for SH which has to have enough fuel to land it must save fuel to be able to unload this fuel onto depot and when ready to return to earth recover said fuel. When SH needs to orbit a heavy load and cannot carry enough fuel for a landing as well, then it should be able to use fuel depot to re-fuel enough for a landing, same for SH returning from higher orbits. I like the fuel depot.  ;D

A SH can't land period.  It has no TPS.  It cannot re-enter at orbital speeds.
I think I know what you're getting at ... but for the benefit of brainbit, SH will be able to land - that's sort of the whole point of reusable rockets.  But it will not be able to:
1) Achieve anything even remotely close to orbital velocity, particularly when boosting SS.  This precludes entering orbit and rendezvousing with the notional depot.
2) As you point out, if it did somehow enter orbit, it would never be coming back.  At least not in one piece.  Without TPS, orbital re-entry would be possible, but very ... traumatic.  (Re-entry is always possible - you just might not like it very much!)

So the flight profile is like a big F9 (more or less) - boost, then re-enter and land, either RTLS, or downrange, presumably on a floating platform.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/21/2020 04:42 pm

I keep wondering if having a depot (enough to refuel, say, 3 Starships) might be a viable future development.

Would it be plausible (and viable) to use a Superheavy near the end of its service life as a depot? It'd need a few modifications, such as refueling compatibility with Starships (both ways, of course, seeing as the prop has to get there somehow). It would also need something akin to IVF (Integrated Vehicle Fluids, basically a tiny internal combustion engine to provide power for temp, pressurization, station keeping, etc).
...
I've been noodling an SSTO SH and don't know if it's possible, but if it is, a reasonable way to go - but not an early priority.

The SH has no heatshield or fins so that's a job that's not needed. It's already designed to interface with SS plumbing - the same plumbing SS will use to refuel. No extra work there. It will need a sun shade, maybe an earth shade, a fuel fridge and a radiator. The fuel fridge can be PV powered but I don't have a feel for power needs.

The big Q is getting SH to orbit SSTO. Beyond my skills.

Why would you want a Super Heavy with all those extra engines as a depot?

A modified Starship with no legs, fins, heat-shield or header tanks, and the main tanks extended to almost full height should hold close to 2000 t of propellant in orbit. It would have to launch with partially filled tanks, getting to LEO with maybe 200 t remaining.
Or a little less if some of that mass is replaced by a cryo-cooler, solar panels and a sun-shield.

Other options include an extra long version, or maybe a 12 m diameter version, holding 3500+ t.  Not needing an EDL capability opens up a lot of options.
I agree. The exact numbers can be quibbled but the concept is sound. There are unknowns like the exact mass of the removed items and the dry mass of SS. And how much of a stretch that can launch safely.


My crude BOE sez that adding 5, maybe 6 more rings and turning it all into tankage gives capacity to top off two SS's. Then there is the possibility of a 12m tank expansion but this would entail some serious engineering and R&D. My gut says this would be a late development and would be a precursor to the next generation of ships.


The loss of a set of raptors on each SSTO SH is not to be sneezed at but also has to be balanced against development costs for an outsized SS. A five ring stretch, IMO, would be minimal development cost and possibly the most cost effective solution in the mid term. One issue thrown into the trade off matrix is each SS based depot of any size eats six raptors.


Bigger is better. Ideally one depot would be adequate for one Mars fleet. Without numbers in front of me I guesstimate one SH worth of tankage could refuel 3-4 SS's. Good, but not ideal. I have no idea where this fits into the trade offs but it 'feels' like it's close enough to the ballpark that it deserves a close look before dismissal.


Unfortunately there are enough unknowns that the best we can do is look at possibilities and as the unknowns resolve, narrow things down a bit.


Phil





Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/21/2020 04:50 pm
The orbital plane is relatively fixed as Earth moves around the sun, hence it will move in and out of sync with the ecliptic. So the orbital plane of a a particular LEO orbit will have two opportunities per year to launch into an interplanetary trajectory without a massive plane-change manoeuvre.
this particular quirk of orbital mechanics is why I doubt Phobos or Deimos will ever (?) be used as way-stations/depots on the way to Mars surface.  Only twice per Martian year would it be possible to arrive at Mars and capture into an orbital plane aligned with the moons... and how often would those arrival windows synchronise with the Earth-Mars synodic cycle?  You might be able to get around this by performing multiple plane change maneuvers whilst aero-capturing into Martian orbit, but there would have to be an upper limit to how great a plane change you could perform that way.

Just one plane change.

You aero-capture into a very high Mars orbit, putting apoapsis right on the edge of the Hill Sphere (or even fractionally beyond), then do a plane change at apoapsis to match the destination orbital plane. At this distance, the delta-v cost of even a 90 degree plane change is in the low hundreds of m/s. On the next pass, do a second smaller aero-capture to lower apoapsis to match the altitude of the target orbit. (Then do your circularisation burn when the orbital phasing catches up to the target.) Adds about two weeks or so of additional travel for the passengers. Not a deal-breaker if the ship has sustained months-long interplanetary flight.


I don't want to derail this thread any further, so I went back and found the "Station on Phobos" thread where Hop_David put the problem that I was getting at into technical terms.  Also helped me refresh my memory on how I came to that thinking.  Never got my final question answered though, so maybe those of us interested can continue the discussion there?

https://forum.nasaspaceflight.com/index.php?topic=40176.100 (https://forum.nasaspaceflight.com/index.php?topic=40176.100)

TLDR: Assuming you've aerobraked into high martian orbit, the plane change to become co-orbital with Phobos/Deimos is easy at apoapsis, *if* the longitude of ascending node of your HMO (which is constrained by your original approach vector/timing) is aligned with that of Phobos/Deimos equatorial orbit.  Also, because the moons' orbits are more-or-less equatorial, you would have to make yet another plane change maneuver to get to/from the moons to any higher inclination site on Mars' surface - hence why I said I doubt they will be used "way-stations" (though I could still see them eventually becoming destinations in their own right).

Just a note: equatorial orbit doesn't have a longtitude of ascending node. It's like dividing 0 by 0.

Phobos is not exactly equatorial, but it's inclination is so low (~1°) that even if you miss it's ascending node by 180° dV to it would be small.

I don't see a point of making Phobos a waystation, as getting to higher latitudes from it is costly, but capturing to its orbit is not the reason why not.
Ignoring the purpose of the maneuver and looking only at the orbital mechanics; the mars axis of rotation is tilted to the plane of its orbit. An entry from the local ecliptic to an equatorial orbit needs a plane change.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/21/2020 05:08 pm
Could not a fuel depot in space be used to store any excess fuel from a SH, for example if SH had to spend some time in orbit saving any boil off as am assuming depot has fuel management built in. Also for SH which has to have enough fuel to land it must save fuel to be able to unload this fuel onto depot and when ready to return to earth recover said fuel. When SH needs to orbit a heavy load and cannot carry enough fuel for a landing as well, then it should be able to use fuel depot to re-fuel enough for a landing, same for SH returning from higher orbits. I like the fuel depot.  ;D
As philw1776 points out, SH can't do reentry from orbit. Indeed, the only reason it might ever go to orbit would be without SS and only to become a fuel depot - not a done deal.
As for a depot collecting residual propellant from other launches, the devil is in the details. The residuals would have to be more than what it takes to match orbit with the depot. If the SS is needed for a fast turnaround the lost time for orbit matching, approach and transfer might not be worth it.


All in all, probably better to keep tighter control of propellant loading unless it's a very light launch to a convenient orbit allowing enough residuals to be worthwhile.


As for how much residuals might be enough to make it worthwhile, I'm drawing a blank. Maybe 200 tons? Just a guess.


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/21/2020 05:36 pm
Super Heavy (with a nose fairing) would make a heck of a Mars launch vehicle. Like, for launching stuff to low Mars orbit. Thousands of tons to LMO per launch. Maybe around 10,000 tons if it were lengthened (lower Mars gravity would mean the same engine thrust could allow almost triple the take-off mass). Low Mars orbit reentry is pretty modest, 3.3-3.5km/s or so. Comparable to a Falcon Heavy center core booster reentry. Might not need much (any?) TPS since it's made of stainless.

This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ach1000 on 08/21/2020 05:59 pm
This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.

If I'm reading this chart correctly, deltaV requirement is 6.3 km/s for Mars to Earth orbit, v. 9.4 km/s for Earth surface to Earth orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 08/21/2020 08:21 pm
most of the propellant for trips to Mars might actually come *from* Mars.

If Deimos' very low density (https://solarsystem.nasa.gov/moons/mars-moons/deimos/by-the-numbers/) truly indicates volatiles, propellant for both Mars launch and LEO launch (https://forum.nasaspaceflight.com/index.php?topic=45299.msg1802366#msg1802366) could be produced more efficiently on Deimos than on Mars.  At scale, there'd be no need for Mars ISRU propellant (https://forum.nasaspaceflight.com/index.php?topic=45299.msg1802366#msg1802366), or for a tanker SH fleet - apart from the possibility of a small Earth-orbit depot fleet.

Various Earth-Moon-Mars propellant scenarios could be calculated with the network flow modeling method for optimized logistics in Ishimatsu et al. 2016.  Note their particular interest in Deimos:

Quote from: Ishimatsu et al. 2016
...it is interesting that part of the resources produced on Deimos are delivered back to GTO and LEO, and wait to be used for the crew and cargo outbound trip. Though it seems strange and non-intuitive at first, this is true at least computationally because in terms of ∆V, LEO is closer to Deimos than to the lunar surface and even Earth’s surface.

Image:  Network graph featuring Deimos ISRU propellant.  Ishimatsu et al. 2016, Fig. 10e.

Refs.

Ishimatsu, T., de Weck, O.L., Hoffman, J.A., Ohkami, Y. and Shishko, R., 2016. Generalized multicommodity network flow model for the earth–moon–mars logistics system. (https://dspace.mit.edu/bitstream/handle/1721.1/99360/JSR_Final_Manuscript_Ishimatsu.pdf?sequence=1&isAllowed=y) Journal of Spacecraft and Rockets, 53(1), pp.25-38.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/21/2020 08:35 pm
Super Heavy (with a nose fairing) would make a heck of a Mars launch vehicle. Like, for launching stuff to low Mars orbit. Thousands of tons to LMO per launch. Maybe around 10,000 tons if it were lengthened (lower Mars gravity would mean the same engine thrust could allow almost triple the take-off mass). Low Mars orbit reentry is pretty modest, 3.3-3.5km/s or so. Comparable to a Falcon Heavy center core booster reentry. Might not need much (any?) TPS since it's made of stainless.

This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.
Interesting concept. As a variation, might it be better to launch CO2 and water from Mars and process en route to take advantage of increasing solar power? A big argument against would be novel tank requirements.


On the pro side, fewer PV's to trip over on Mars. The return could be more about filling a pipeline than immediate deliverables so the slowest most economical return can be a great advantage in reducing the size of PV.


There are a class of trajectories that allow a low energy, long elapsed time, approach to mars orbit. The essence is aiming ahead of mars and slightly further outside mars orbit or behind mars and slightly inside mars orbit. In either case the craft is eventually swept up by mars into a highly elliptical orbit using aerobraking to circularize. The investigators claim up to 25% propellant savings.
https://www.scientificamerican.com/article/a-new-way-to-reach-mars-safely-anytime-and-on-the-cheap/ (https://www.scientificamerican.com/article/a-new-way-to-reach-mars-safely-anytime-and-on-the-cheap/)


I'm not sure but I think this might be adaptable to earth return of a depot or cargo.


So, once one of these depots returns to earth and refuels part of a mars fleet, what do you do with it? Could it carry enough dV for a useful propellant delivery AND a return to mars for another load? Oh, my brain hurts. So many variables.


Phil
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 08/21/2020 09:29 pm
once one of these depots returns to earth and refuels part of a mars fleet, what do you do with it? Could it carry enough dV for a useful propellant delivery AND a return to mars for another load?

Why build dedicated Earth-Mars tankers?  Just cycle the existing cargo ships, using them as tankers when inbound.  This saves all the cost of a redundant Earth-Mars tanker fleet, which is cut.  E.g., Omaha Trail (http://www.lakematthew.com/press/press-release-november-7-2017/) cargo.   
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/21/2020 09:53 pm
The biggest problem with filling LEO tankers with non-terran methane, is that departure burns from LEO put the exaust in a suborbital trajectory. And we have enough problems with co2 on earth without importing martian Co2.

Terran methane would just be recycled back to earth, but martian methane would be a pure import.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 08/22/2020 12:07 am
The biggest problem with filling LEO tankers with non-terran methane, is that departure burns from LEO put the exaust in a suborbital trajectory. And we have enough problems with co2 on earth without importing martian Co2.

Terran methane would just be recycled back to earth, but martian methane would be a pure import.
I take the problem of excess CO2 in Earth's atmosphere as seriously as anyone.  We must halt and begin to revers this sooner rhater than later.  That said ...

Compared to current and reasonably foreseeable terrestrial emissions, Star ship exhaust, even in full, heavy colonization mode, won''t even be a rounding error.   If it ever does become a concern, build a few more air-capture plants.  We're likely going to need plenty of those anyway.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: CJ on 08/22/2020 01:10 am
The biggest problem with filling LEO tankers with non-terran methane, is that departure burns from LEO put the exaust in a suborbital trajectory. And we have enough problems with co2 on earth without importing martian Co2.

Terran methane would just be recycled back to earth, but martian methane would be a pure import.

IMHO, you may be overlooking something - the net. Regardless of the source of the methane's combustion byproducts, it's going to enter Earth's carbon cycle if suborbital. Viewed alone, there's no difference. However, if Martian, it would be replacing Terran methane - and Terran methane has to get to orbit somehow. That somehow is going to use many times the orbital fuel's mass to get it to orbit. All of that will be going into the atmosphere. This is not true for Martian methane. So, utilizing Martian methane is actually a very large net reduction in byproducts ending up in the Earth's atmosphere. And that's not even counting the lower total delta/v to LEO from the surface of Mars vs. the surface of Earth. 

IMHO, that lower delta/v to LEO makes getting prop from Mars (or an asteroid, or a martian moon) for use in LEO very attractive, though of course very long-term. Likewise, other consumables for LEO infrastructure. 

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Pahimarus on 08/22/2020 01:45 am
Quote
An extra point to consider. Earth is tipped 23.5 deg to our orbital plane so an equatorial launch faces the same limitations of node alignment with the ecliptic.


I don't think there is any place on earth without this problem. It's something we have to live with until we can afford to take it to the shop for an axis alignment


Phil

You are right of course. I thought about that after I posted.

That is the problem with being kerbal trained without the realism mods.  ;) :-[

I do think the the offshore platforms could help by allowing you more launch sites within the desired latitude range but it would make more sense to align them with bc and ksc than to put them on the equator though you do get a touch more effiency with you launch.

The idea of using a modified or just empty ss as a depot makes sense. At least as you say you only have to tank once instead of 5 times with the mission starship which significantly reduces risk of a problem you don't have time to recover from during the transfer window. Really what you described with 10 ss depots is just a distributed depot. In fact as I think about it there are a number of advantages to doing it that way. Namely that an accident only takes out 1/10th of your depot. Unless there is some major advantage to sharing equipment across multiple, a larger mega depot like I was envisioning probably wouldn't make sense.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/22/2020 02:00 am
The biggest problem with filling LEO tankers with non-terran methane, is that departure burns from LEO put the exaust in a suborbital trajectory. And we have enough problems with co2 on earth without importing martian Co2.

Terran methane would just be recycled back to earth, but martian methane would be a pure import.

IMHO, you may be overlooking something - the net. Regardless of the source of the methane's combustion byproducts, it's going to enter Earth's carbon cycle if suborbital. Viewed alone, there's no difference. However, if Martian, it would be replacing Terran methane - and Terran methane has to get to orbit somehow. That somehow is going to use many times the orbital fuel's mass to get it to orbit. All of that will be going into the atmosphere. This is not true for Martian methane. So, utilizing Martian methane is actually a very large net reduction in byproducts ending up in the Earth's atmosphere. And that's not even counting the lower total delta/v to LEO from the surface of Mars vs. the surface of Earth. 

IMHO, that lower delta/v to LEO makes getting prop from Mars (or an asteroid, or a martian moon) for use in LEO very attractive, though of course very long-term. Likewise, other consumables for LEO infrastructure.
I am assuming all terran methane is made the same way martian methane is- with water, CO2 and electricity. That makes any and all terran methane net carbon neutral, for launch as well as for departure burns. Martian landing burns are an export of terran methane, landing burns of starships returning from mars are an import, but the balance is net earth export. (Pretty much all lunar operations are net export as well.)

But bringing martian methane to earth orbit to burn upsets that "trade balance."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Pahimarus on 08/22/2020 05:04 pm
Quote from: rakaydos
I am assuming all terran methane is made the same way martian methane is- with water, CO2 and electricity. That makes any and all terran methane net carbon neutral, for launch as well as for departure burns.

Except it isn't.

Terran methane is dug up from underground sequestered areas and burned.

As pointed out previously the amount of carbon is trivial compared to the amount in the Earth's overall carbon cycle. So what you really have to concern yourself with is the immediate emissions impact (if that is even significant which I doubt WAG each ss launch is 0.0000004% of annual emissions but for the sake of argument).

If that is the case you must consider the whole life cycle analysis of your methane/co2.

From earth's perspective martian sourced fuel appears in the atmosphere as CO2. So does Terran methane. If anything more emissions are generated drilling for and processing the earth methane than vice versa.

Until the majority of earth methane is synthesized from atmospheric CO2 using renewable power this will continue to be true so you can't write it off as net neutral.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/22/2020 05:12 pm
AIUI, gasses in the upper atmosphere get banged around quite a bit by UV. Is this enough to disassociate CH4? Any ideas on what the half life would be at different altitudes if this happens?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/22/2020 06:30 pm
Quote from: rakaydos
I am assuming all terran methane is made the same way martian methane is- with water, CO2 and electricity. That makes any and all terran methane net carbon neutral, for launch as well as for departure burns.
Until the majority of earth methane is synthesized from atmospheric CO2 using renewable power this will continue to be true so you can't write it off as net neutral.
Not to put too fine a point on it, but we dont have an orbital class methane rocket yet, let alone martian methane to fill it up with.

By the time a martian colony is developed enough to export methane, earth had BETTER have moved to non-fossil methane, or we will have bigger problems than supplying a martian base.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/22/2020 08:55 pm
The biggest problem with filling LEO tankers with non-terran methane, is that departure burns from LEO put the exaust in a suborbital trajectory. And we have enough problems with co2 on earth without importing martian Co2.

Terran methane would just be recycled back to earth, but martian methane would be a pure import.

IMHO, you may be overlooking something - the net. Regardless of the source of the methane's combustion byproducts, it's going to enter Earth's carbon cycle if suborbital. Viewed alone, there's no difference. However, if Martian, it would be replacing Terran methane - and Terran methane has to get to orbit somehow. That somehow is going to use many times the orbital fuel's mass to get it to orbit. All of that will be going into the atmosphere. This is not true for Martian methane. So, utilizing Martian methane is actually a very large net reduction in byproducts ending up in the Earth's atmosphere. And that's not even counting the lower total delta/v to LEO from the surface of Mars vs. the surface of Earth. 

IMHO, that lower delta/v to LEO makes getting prop from Mars (or an asteroid, or a martian moon) for use in LEO very attractive, though of course very long-term. Likewise, other consumables for LEO infrastructure.
I am assuming all terran methane is made the same way martian methane is- with water, CO2 and electricity. That makes any and all terran methane net carbon neutral, for launch as well as for departure burns. Martian landing burns are an export of terran methane, landing burns of starships returning from mars are an import, but the balance is net earth export. (Pretty much all lunar operations are net export as well.)

But bringing martian methane to earth orbit to burn upsets that "trade balance."
Unfortunately Terran methane is currently derived from fossil sources and likely to remain that way for many years . It will not be energy efficient to generate methane from the sabatier process on Earth until all fossil power stations have been replaced. Because it will always be more efficient to plug a new solar or wind farm into the grid and switch off a fossil fuel plant than using the solar/wind electricity to unscramble the CO2 egg after it has been burnt.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/22/2020 09:02 pm
AIUI, gasses in the upper atmosphere get banged around quite a bit by UV. Is this enough to disassociate CH4?

Burning it in oxygen sure would.

(Rockets are releasing combustion products, not raw fuel.)

[Also, what Rakaydos said. Should we be concerned about the accumulation of re-entry heat? Changes to Earth's rotational inertia due to mass loss to deep space missions? Yes. Yes we should. Once we're a K1+ civilisation, there are many such things we'll need to manage. Should we be concerned now? No.]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 08/22/2020 09:22 pm
For some of you guys information, the US has a 200 year supply of natural gas drilled, tapped, and capped, without drilling another well.  Producing methane from CO2 and water on earth costs more than just buying existing natural gas.  Unless Musk wants to make it anyways, the first methane rockets for several years, will be extracted from the existing 95% methane natural gas.  Coal, oil, and gasoline, produce far far more CO2 per year than natural gas.  Rockets will be a minor use, even at high flight rates. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: CJ on 08/22/2020 11:20 pm
Do we have a ballpark estimate of the G needed for Starship-to-Starship refueling, and how long it would need to be applied for?

As I recall, the plan was to use thrusters to impart some G to settle the prop in order to transfer it. I was wondering how much prop would need to be used for said thrusters for each transfer. (which could be calculated if we knew the G and the time).

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TomH on 08/23/2020 12:12 am
Do we have a ballpark estimate of the G needed for Starship-to-Starship refueling, and how long it would need to be applied for?

As I recall, the plan was to use thrusters to impart some G to settle the prop in order to transfer it. I was wondering how much prop would need to be used for said thrusters for each transfer. (which could be calculated if we knew the G and the time).

I don't think they intend for the flow to be induced by gravity. You need some G to keep the prop settled, but as soon as settling begins, pumps would be used to transfer the liquids. Waiting for artificial gravity alone to induce the transfer would take a long time. The transfer rate will depend upon pump capacity. Does anyone have any specs re. the pumps they intend to use? Pipe diameter and flow resistance related to coefficient of friction in relation to material type, temperature, etc. are also factors in the flow rate, but the pump capacity is the main thing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/23/2020 12:15 am
AIUI, gasses in the upper atmosphere get banged around quite a bit by UV. Is this enough to disassociate CH4?

Burning it in oxygen sure would.

(Rockets are releasing combustion products, not raw fuel.)

[Also, what Rakaydos said. Should we be concerned about the accumulation of re-entry heat? Changes to Earth's rotational inertia due to mass loss to deep space missions? Yes. Yes we should. Once we're a K1+ civilisation, there are many such things we'll need to manage. Should we be concerned now? No.]
My bad. Change the Q to CO2.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/23/2020 12:26 am
Do we have a ballpark estimate of the G needed for Starship-to-Starship refueling, and how long it would need to be applied for?

As I recall, the plan was to use thrusters to impart some G to settle the prop in order to transfer it. I was wondering how much prop would need to be used for said thrusters for each transfer. (which could be calculated if we knew the G and the time).
Literally micro G's. Think if it as the tank sliding 'up' around the propellant. It would take a while to settle. More G, faster settling but still probably a fraction of a G. Once settled, back to micro G.


Playing with this to find optimum will be high priority once they start doing it.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/23/2020 12:28 am
Do we have a ballpark estimate of the G needed for Starship-to-Starship refueling, and how long it would need to be applied for?

As I recall, the plan was to use thrusters to impart some G to settle the prop in order to transfer it. I was wondering how much prop would need to be used for said thrusters for each transfer. (which could be calculated if we knew the G and the time).

I don't think they intend for the flow to be induced by gravity. You need some G to keep the prop settled, but as soon as settling begins, pumps would be used to transfer the liquids. Waiting for artificial gravity alone to induce the transfer would take a long time. The transfer rate will depend upon pump capacity. Does anyone have any specs re. the pumps they intend to use? Pipe diameter and flow resistance related to coefficient of friction in relation to material type, temperature, etc. are also factors in the flow rate, but the pump capacity is the main thing.
Could it be done with pressure?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Redclaws on 08/23/2020 01:05 am
So if you pulled a mild vacuum in the empty tank, then topped the other up with ... whatever is going to fill it...  that might work.  I don’t think you could pull much vacuum, though, without collapsing the tank.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: SkyRate on 08/23/2020 01:20 am
So if you pulled a mild vacuum in the empty tank, then topped the other up with ... whatever is going to fill it...  that might work.  I don’t think you could pull much vacuum, though, without collapsing the tank.
What pressure would collapse the tank?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: envy887 on 08/23/2020 01:21 am
So if you pulled a mild vacuum in the empty tank, then topped the other up with ... whatever is going to fill it...  that might work.  I don’t think you could pull much vacuum, though, without collapsing the tank.

A ship in LEO is sitting in a hard vacuum. You don't need to "pull" to get a vacuum, and you certainly won't be collapsing anything.

They will probably pressurize the tanker and vent the tankee to make a pressure differential. If they pump anything, it will be compressing boiloff in COPVs to be used to press the tanker. This can be done with solar electric over a period of time between refueling flights, so it doesn't require high powered pumps.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ericgu on 08/23/2020 03:01 am
Quote from: rakaydos
I am assuming all terran methane is made the same way martian methane is- with water, CO2 and electricity. That makes any and all terran methane net carbon neutral, for launch as well as for departure burns.

Except it isn't.

Terran methane is dug up from underground sequestered areas and burned.


This is an interesting question; I found a paper https://www.sciencedirect.com/science/article/pii/S0306261919312681 (https://www.sciencedirect.com/science/article/pii/S0306261919312681) that looked at the economics.

There could be reasons that would push you towards wanting to create your own methane...

* Maybe there isn't a big enough pipeline the area where you want to operate.
* Maybe the refining is problematic  - you can't easily do it, you can't dispose of the fractions you don't want, etc.
* Maybe you have an excess of renewable power at certain times of the day (like Texas has too much wind power at night) that you can economically use.
* Maybe there are carbon taxes that make fossil fuel more expensive.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: livingjw on 08/23/2020 03:49 am
Do we have a ballpark estimate of the G needed for Starship-to-Starship refueling, and how long it would need to be applied for?

As I recall, the plan was to use thrusters to impart some G to settle the prop in order to transfer it. I was wondering how much prop would need to be used for said thrusters for each transfer. (which could be calculated if we knew the G and the time).

I don't think they intend for the flow to be induced by gravity. You need some G to keep the prop settled, but as soon as settling begins, pumps would be used to transfer the liquids. Waiting for artificial gravity alone to induce the transfer would take a long time. The transfer rate will depend upon pump capacity. Does anyone have any specs re. the pumps they intend to use? Pipe diameter and flow resistance related to coefficient of friction in relation to material type, temperature, etc. are also factors in the flow rate, but the pump capacity is the main thing.
Could it be done with pressure?

- Yes, you can do it with low pressure pumps and as a benefit, you don't have to vent any gases.

- You could also do it with pressure differential, but you would have to vent some gases.

John
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thrustpuzzle on 08/23/2020 07:18 am
- You could also do it with pressure differential, but you would have to vent some gases.
There may be little or no venting needed since we're dealing with cryogenic fluids, not water or kerosene. The receiving tanks will have their ullage space filled with gas, and the incoming liquids will displace some of that volume, compressing that gas to higher pressure, requiring venting. But these are cryo liquids, by definition at temperatures lower than (perhaps much lower than) the boiling point of the gas, so in fact that ullage gas will be cooled by the incoming cryogenic rain at the same time it's compressed. Depending on the temperature of the incoming liquid, and the efficiency of heat transfer between the rain and the gas, the ullage gas pressure could even lower as the tank is filled, increasing the pressure differential between source and destination tanks. It depends on the temperature balance and mixing efficiency.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/23/2020 08:08 am
Quote from: rakaydos
I am assuming all terran methane is made the same way martian methane is- with water, CO2 and electricity. That makes any and all terran methane net carbon neutral, for launch as well as for departure burns.

Except it isn't.

Terran methane is dug up from underground sequestered areas and burned.


This is an interesting question; I found a paper https://www.sciencedirect.com/science/article/pii/S0306261919312681 (https://www.sciencedirect.com/science/article/pii/S0306261919312681) that looked at the economics.

There could be reasons that would push you towards wanting to create your own methane...

* Maybe there isn't a big enough pipeline the area where you want to operate.
* Maybe the refining is problematic  - you can't easily do it, you can't dispose of the fractions you don't want, etc.
* Maybe you have an excess of renewable power at certain times of the day (like Texas has too much wind power at night) that you can economically use.
* Maybe there are carbon taxes that make fossil fuel more expensive.
This is a good point. There are certainly cases where you would want to create methane, if the nearest supply is far enough away to make delivery impractical and you need large quantities (and you really do need it) and you have a lot of renewable power available. Having an accessible source of CO2 would also be beneficial.

The most obvious place where these condition hold is Mars. It could happen on Earth but the conditions would have to be  very unusual.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 08/23/2020 10:11 am
[SNIP]

- Yes, you can do it with low pressure pumps and as a benefit, you don't have to vent any gases.

This.

You pump the gas from the empty ship into the tanker, this pushes the liquid into the empty ship.
(or the other way around if that's your preference ie, pump the liquid which pushes the gas)

It would be a roughly equal volumetric exchange so no need to vent or compress anything.  :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/23/2020 11:34 am
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? If you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
 

Star-Dust, you have aspirations of sending rockets to Mars, that is admirable, and if rockets were fueled with persistence, you might already be in orbit.  My unsolicited advice to you is that like early aviators, take one step at a time and develop a firm grasp of the fundamentals before taking that giant leap. 

A great example is your lack of understanding that SH can’t deliver your proposed depot to orbit.  This is an easy mistake to make, but as a result of this error, you have been now forced to fabricate this new expendable stage.  You deride the cost of maintaining tankers, but now somehow building expendable second stages comes without a tremendous cost?  That cost comes in many forms, not just in dollars, but also in manufacturing time and a workforce that is now dedicated to building hardware that will be thrown away after every flight.  This is very inefficient. 

If you have paid attention to the work at BC, you can see that constructing a SS is a multi-month process.  How is SS minus wings and the payload section any different from your proposed second stage?  Are you really willing to wait months between depot launches?  Have you done any analysis of the cost to dispose of five stages vs. maintenance of a single tanker?  Now work the other side of the problem.  You have your depot, but without the stage you just threw away, you can’t change orbits or dock with other depots.  You will need power to keep the fuels at the proper pressure.  Where will this come from?  Are you now going to add solar arrays and compressors that you also dispose of along with communications equipment, RCS, batteries, fuel pumps, etc.?  How will you de-orbit your empty depot?

My recommendation is to stick with this first step.  Perhaps spend some time studying the capabilities of SH and SS so you can avoid further simple errors.  Look at necessary capabilities for that second stage you are proposing compared to a SS Tanker/Depot concept. Again, before you can convince us that you have a viable vision for how to get to Mars and back, at least do the minimum legwork and develop a viable plan for how to get just a single Starship worth of fuel to LEO without one year required to build your five expendable second stages, along with 30 or more Raptors that you plan to dispose of.  Yes, this will take some discipline, and you just might succeed in this first step and either make a compelling case for your depots, or garner the knowledge and understanding as to why so many who have studied this problem, some for decades, disagree with you.  Best of luck.

1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH, this is if the figures available are correct twice as much as a reusable SS capability in term of payload (100 tons) , and by definition will require twice less the number of flights, no need to remind  that SH on each flight require a 3400 tons of fuel+ 1200 tons SS ( 2.5 millions$ saved on fuel rough calculation based on 600$/ton LCH4, 200$/ton LOX).

2-A second stage by definition is a basic rocket which mean engines + fuel tank, I dont think that a stainless steel tank coupled with six raptors engines will weigh that much in the overall balance of a $/per ton of payload (no heat shield, no legs/no extra fuel,.........etc.).

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

4-No extra spending on O&M for tankers and less expensive 2S than a tanker.

5- Fuel depot could be sent to Mars orbit.


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/23/2020 12:43 pm

You might be misunderstanding. There is nothing in a depot, one tank or multiple, that precludes butt to butt refueling. Two entirely separate issues.


As for sending a depot to Mars, there has been discussion in this. First you need to understand that it does not remove the SS need to refuel before leaving earth. It's an alternative to having ISRU propellant production on mars. However, it's always seen as a supplement for early missions until reliable ISRU is on line, and not a replacement. Mars will never be more than a base if it depends on earth for its very ability to return.


Phil

I keep my opinion on  Mars colonisation prospect for me, but getting first human and first permanent settlement on the red planet will require a step by step approach. If we want to get a real plan to do so we can't bet on future developement of new technologies that does not exist yet. this is why interim solutions is required you improve the process while you achieve what can be achieved in a timely manner, the air travel didn't await for the modern jet reaction airliners to be a reality, it took years and years and many accidents to get to what is now air travel.

The above mentioned idea of a multiple disposable tanks that will be joint to a nodal comprtment as is the case for ISS, seems to me practicle for multiple reasons:

-No need for a fleet of tankers and subsequent logistic issues.
-No time waisted for SS departing to Mars.
-Depot could be positioned in any strategic position.
Ahhh. I think I see the problem. It's a matter of religious doctrine and not amenable to reasoned argument - from either side of the question.


Stardust, from what I'm picking up, you are skeptical of the entire approach. This is observation, not criticism.


For the majority of us, we are skeptical of the traditional approach. Again, an observation and not an endorsement.


The two religions are so diametrically at odds that they can never be reconciled no matter how much technical argument is presented. So let's stop this bickering over detail and agree to disagree. THERE IS NOTHING HERE TO GET frakked ABOUT!!


I learned a long time ago that all logic based arguments have at least one predicate - an assumption. What we have here is a conflict of assumptions.


Stardust, the only fault I can assign you is that you have stepped into a den of new spacers and are espousing the doctrine of old space or something close to it. In another time that would get you burned at the stake or an invitation to make love with the Iron Maiden. I promise, we won't do that.


What I would suggest is that you start another discussion with those who share your beliefs. I'd like to believe that we are collectively mature enough to allow others to discuss that which we disagree with although we might want to draw the line at flat earthers 8)


Phil

I got a lot of things that make me skeptical from low maturity technology to financial plan lacking realism and no sustainable viable business plan. :-\

That's glass half empty, or some people might say more realistic lol.  I prefer to think they're the team who finally have a real plan for doing this with re-use, powered landing, and funding from starlink.  If they've got it, they're building a new era in human travel.  If they fail, then they'll have advanced the state of the art a long way and eventually someone will take their successes and build on them. 

Either way, spending time on approaches they've clearly discarded is wasteful.  Same reason we're no longer dissecting the nuances of large scale composite construction; they discarded that approach.  Likewise, they long ago discarded the notion of expendable rocketry.  Right, wrong?  Only history will decide, but burning up the forum advocating SpaceX go disposable isn't a great use of page space.

If you really like the expendable idea you should explore the habitat threads.  There's a lot of supportive discussion for the  concept there in terms of leaving them at Mars and using as permanent habitats.  Musk's even hinted that may be the fate of early flights.  You're not wrong that there may be some expendable parts  in the architecture for some time; but you're really misplaced thinking it's going to be their plan in earth orbit since they've clearly stated the opposite.

Just my $0.02,  I think your idea would generate more healthy debate in the habitats thread where people are eager to only fly them once ;)

For me it's more about technology maturation time requirement than affinity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: livingjw on 08/23/2020 12:55 pm
[SNIP]

- Yes, you can do it with low pressure pumps and as a benefit, you don't have to vent any gases.

This.

You pump the gas from the empty ship into the tanker, this pushes the liquid into the empty ship.
(or the other way around if that's your preference ie, pump the liquid which pushes the gas)

It would be a roughly equal volumetric exchange so no need to vent or compress anything.  :)

No, if you are using a pressure difference, the gas from the tank being filled is lower than the tank doing the filling, so the gas has to be either condensed or vented.

John
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Keldor on 08/23/2020 02:43 pm
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? If you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
 

Star-Dust, you have aspirations of sending rockets to Mars, that is admirable, and if rockets were fueled with persistence, you might already be in orbit.  My unsolicited advice to you is that like early aviators, take one step at a time and develop a firm grasp of the fundamentals before taking that giant leap. 

A great example is your lack of understanding that SH can’t deliver your proposed depot to orbit.  This is an easy mistake to make, but as a result of this error, you have been now forced to fabricate this new expendable stage.  You deride the cost of maintaining tankers, but now somehow building expendable second stages comes without a tremendous cost?  That cost comes in many forms, not just in dollars, but also in manufacturing time and a workforce that is now dedicated to building hardware that will be thrown away after every flight.  This is very inefficient. 

If you have paid attention to the work at BC, you can see that constructing a SS is a multi-month process.  How is SS minus wings and the payload section any different from your proposed second stage?  Are you really willing to wait months between depot launches?  Have you done any analysis of the cost to dispose of five stages vs. maintenance of a single tanker?  Now work the other side of the problem.  You have your depot, but without the stage you just threw away, you can’t change orbits or dock with other depots.  You will need power to keep the fuels at the proper pressure.  Where will this come from?  Are you now going to add solar arrays and compressors that you also dispose of along with communications equipment, RCS, batteries, fuel pumps, etc.?  How will you de-orbit your empty depot?

My recommendation is to stick with this first step.  Perhaps spend some time studying the capabilities of SH and SS so you can avoid further simple errors.  Look at necessary capabilities for that second stage you are proposing compared to a SS Tanker/Depot concept. Again, before you can convince us that you have a viable vision for how to get to Mars and back, at least do the minimum legwork and develop a viable plan for how to get just a single Starship worth of fuel to LEO without one year required to build your five expendable second stages, along with 30 or more Raptors that you plan to dispose of.  Yes, this will take some discipline, and you just might succeed in this first step and either make a compelling case for your depots, or garner the knowledge and understanding as to why so many who have studied this problem, some for decades, disagree with you.  Best of luck.

1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH, this is if the figures available are correct twice as much as a reusable SS capability in term of payload (100 tons) , and by definition will require twice less the number of flights, no need to remind  that SH on each flight require a 3400 tons of fuel+ 1200 tons SS ( 2.5 millions$ saved on fuel rough calculation based on 600$/ton LCH4, 200$/ton LOX).

2-A second stage by definition is a basic rocket which mean engines + fuel tank, I dont think that a stainless steel tank coupled with six raptors engines will weigh that much in the overall balance of a $/per ton of payload (no heat shield, no legs/no extra fuel,.........etc.).

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

4-No extra spending on O&M for tankers and less expensive 2S than a tanker.

5- Fuel depot could be sent to Mars orbit.

There are a few reasons I doubt the practicality of a fueling deopt.

First, orbital mechanics.  The problem is that orbits have to be properly aligned for the rocket to be able to rendevous with the depot and then proceed to Mars.  With a depot in low orbit, this might not be a problem, but something on a high eccentric orbit (such as needed for landing on the Moon) will rarely be in the right alignment.  You can imagine the oblong shape of the orbit being like an arrow, pointing the direction the vehicle has to depart in.  Most of the time, it will be pointing the wrong way.  Also, once the depot is in this orbit, it takes a lot of fuel and/or time to change this orbit, so you're largely stuck with whatever direction it happens to point.


Low orbits have a different problem, namely heat.  Cryogenic fuel must be kept very cold to stay liquid.  If it boils into a gas state, your tank will explode from the pressure much like a giant steam boiler that has been overheated.  But the environment in low earth orbit is very hot.  A good comparason would be to Death Valley or somewhere in the middle of the Sahara.  Except in orbit, there's no wind blowing in from cooler places to moderate temperatures.  The same problem happens in high orbits or interplanetary transit, but here the vehicle can do things like pointing the engines toward the sun to shade the rest of the vehicle.  But there's no escaping the reflection off Earth when you're in low orbit.  It's difficult to shed heat in space, since again, there's no air flow to conduct it away, it all has to leave in the form of infrared radiation.  This is a slow process, requiring huge radiators, which again, need to somehow be shaded from both the Earth and the Sun or else they'll become heaters.  Also, since they need to reject enough heat to cool the tanks down to cryogenic temperatures, they'll have to be well below the temperature of the propellant, or else you'll need a heat pump system much like air conditioning, but this generates a lot of waste heat which now must be dealt with as well...

It's simply very difficult to store propellant in orbit for long periods of time in any practical fashion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/23/2020 03:27 pm
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? If you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
 

Star-Dust, you have aspirations of sending rockets to Mars, that is admirable, and if rockets were fueled with persistence, you might already be in orbit.  My unsolicited advice to you is that like early aviators, take one step at a time and develop a firm grasp of the fundamentals before taking that giant leap. 

A great example is your lack of understanding that SH can’t deliver your proposed depot to orbit.  This is an easy mistake to make, but as a result of this error, you have been now forced to fabricate this new expendable stage.  You deride the cost of maintaining tankers, but now somehow building expendable second stages comes without a tremendous cost?  That cost comes in many forms, not just in dollars, but also in manufacturing time and a workforce that is now dedicated to building hardware that will be thrown away after every flight.  This is very inefficient. 

If you have paid attention to the work at BC, you can see that constructing a SS is a multi-month process.  How is SS minus wings and the payload section any different from your proposed second stage?  Are you really willing to wait months between depot launches?  Have you done any analysis of the cost to dispose of five stages vs. maintenance of a single tanker?  Now work the other side of the problem.  You have your depot, but without the stage you just threw away, you can’t change orbits or dock with other depots.  You will need power to keep the fuels at the proper pressure.  Where will this come from?  Are you now going to add solar arrays and compressors that you also dispose of along with communications equipment, RCS, batteries, fuel pumps, etc.?  How will you de-orbit your empty depot?

My recommendation is to stick with this first step.  Perhaps spend some time studying the capabilities of SH and SS so you can avoid further simple errors.  Look at necessary capabilities for that second stage you are proposing compared to a SS Tanker/Depot concept. Again, before you can convince us that you have a viable vision for how to get to Mars and back, at least do the minimum legwork and develop a viable plan for how to get just a single Starship worth of fuel to LEO without one year required to build your five expendable second stages, along with 30 or more Raptors that you plan to dispose of.  Yes, this will take some discipline, and you just might succeed in this first step and either make a compelling case for your depots, or garner the knowledge and understanding as to why so many who have studied this problem, some for decades, disagree with you.  Best of luck.

1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH, this is if the figures available are correct twice as much as a reusable SS capability in term of payload (100 tons) , and by definition will require twice less the number of flights, no need to remind  that SH on each flight require a 3400 tons of fuel+ 1200 tons SS ( 2.5 millions$ saved on fuel rough calculation based on 600$/ton LCH4, 200$/ton LOX).

2-A second stage by definition is a basic rocket which mean engines + fuel tank, I dont think that a stainless steel tank coupled with six raptors engines will weigh that much in the overall balance of a $/per ton of payload (no heat shield, no legs/no extra fuel,.........etc.).

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

4-No extra spending on O&M for tankers and less expensive 2S than a tanker.

5- Fuel depot could be sent to Mars orbit.

There are a few reasons I doubt the practicality of a fueling deopt.

First, orbital mechanics.  The problem is that orbits have to be properly aligned for the rocket to be able to rendevous with the depot and then proceed to Mars.  With a depot in low orbit, this might not be a problem, but something on a high eccentric orbit (such as needed for landing on the Moon) will rarely be in the right alignment.  You can imagine the oblong shape of the orbit being like an arrow, pointing the direction the vehicle has to depart in.  Most of the time, it will be pointing the wrong way.  Also, once the depot is in this orbit, it takes a lot of fuel and/or time to change this orbit, so you're largely stuck with whatever direction it happens to point.


Low orbits have a different problem, namely heat.  Cryogenic fuel must be kept very cold to stay liquid.  If it boils into a gas state, your tank will explode from the pressure much like a giant steam boiler that has been overheated.  But the environment in low earth orbit is very hot.  A good comparason would be to Death Valley or somewhere in the middle of the Sahara.  Except in orbit, there's no wind blowing in from cooler places to moderate temperatures.  The same problem happens in high orbits or interplanetary transit, but here the vehicle can do things like pointing the engines toward the sun to shade the rest of the vehicle.  But there's no escaping the reflection off Earth when you're in low orbit.  It's difficult to shed heat in space, since again, there's no air flow to conduct it away, it all has to leave in the form of infrared radiation.  This is a slow process, requiring huge radiators, which again, need to somehow be shaded from both the Earth and the Sun or else they'll become heaters.  Also, since they need to reject enough heat to cool the tanks down to cryogenic temperatures, they'll have to be well below the temperature of the propellant, or else you'll need a heat pump system much like air conditioning, but this generates a lot of waste heat which now must be dealt with as well...

It's simply very difficult to store propellant in orbit for long periods of time in any practical fashion.

Fuel will certainly be the biggest challenge for Mars missions.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/23/2020 04:52 pm
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? If you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
 

Star-Dust, you have aspirations of sending rockets to Mars, that is admirable, and if rockets were fueled with persistence, you might already be in orbit.  My unsolicited advice to you is that like early aviators, take one step at a time and develop a firm grasp of the fundamentals before taking that giant leap. 

A great example is your lack of understanding that SH can’t deliver your proposed depot to orbit.  This is an easy mistake to make, but as a result of this error, you have been now forced to fabricate this new expendable stage.  You deride the cost of maintaining tankers, but now somehow building expendable second stages comes without a tremendous cost?  That cost comes in many forms, not just in dollars, but also in manufacturing time and a workforce that is now dedicated to building hardware that will be thrown away after every flight.  This is very inefficient. 

If you have paid attention to the work at BC, you can see that constructing a SS is a multi-month process.  How is SS minus wings and the payload section any different from your proposed second stage?  Are you really willing to wait months between depot launches?  Have you done any analysis of the cost to dispose of five stages vs. maintenance of a single tanker?  Now work the other side of the problem.  You have your depot, but without the stage you just threw away, you can’t change orbits or dock with other depots.  You will need power to keep the fuels at the proper pressure.  Where will this come from?  Are you now going to add solar arrays and compressors that you also dispose of along with communications equipment, RCS, batteries, fuel pumps, etc.?  How will you de-orbit your empty depot?

My recommendation is to stick with this first step.  Perhaps spend some time studying the capabilities of SH and SS so you can avoid further simple errors.  Look at necessary capabilities for that second stage you are proposing compared to a SS Tanker/Depot concept. Again, before you can convince us that you have a viable vision for how to get to Mars and back, at least do the minimum legwork and develop a viable plan for how to get just a single Starship worth of fuel to LEO without one year required to build your five expendable second stages, along with 30 or more Raptors that you plan to dispose of.  Yes, this will take some discipline, and you just might succeed in this first step and either make a compelling case for your depots, or garner the knowledge and understanding as to why so many who have studied this problem, some for decades, disagree with you.  Best of luck.

1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH, this is if the figures available are correct twice as much as a reusable SS capability in term of payload (100 tons) , and by definition will require twice less the number of flights, no need to remind  that SH on each flight require a 3400 tons of fuel+ 1200 tons SS ( 2.5 millions$ saved on fuel rough calculation based on 600$/ton LCH4, 200$/ton LOX).

2-A second stage by definition is a basic rocket which mean engines + fuel tank, I dont think that a stainless steel tank coupled with six raptors engines will weigh that much in the overall balance of a $/per ton of payload (no heat shield, no legs/no extra fuel,.........etc.).

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

4-No extra spending on O&M for tankers and less expensive 2S than a tanker.

5- Fuel depot could be sent to Mars orbit.

There are a few reasons I doubt the practicality of a fueling deopt.

First, orbital mechanics.  The problem is that orbits have to be properly aligned for the rocket to be able to rendevous with the depot and then proceed to Mars.  With a depot in low orbit, this might not be a problem, but something on a high eccentric orbit (such as needed for landing on the Moon) will rarely be in the right alignment.  You can imagine the oblong shape of the orbit being like an arrow, pointing the direction the vehicle has to depart in.  Most of the time, it will be pointing the wrong way.  Also, once the depot is in this orbit, it takes a lot of fuel and/or time to change this orbit, so you're largely stuck with whatever direction it happens to point.


Low orbits have a different problem, namely heat.  Cryogenic fuel must be kept very cold to stay liquid.  If it boils into a gas state, your tank will explode from the pressure much like a giant steam boiler that has been overheated.  But the environment in low earth orbit is very hot.  A good comparason would be to Death Valley or somewhere in the middle of the Sahara.  Except in orbit, there's no wind blowing in from cooler places to moderate temperatures.  The same problem happens in high orbits or interplanetary transit, but here the vehicle can do things like pointing the engines toward the sun to shade the rest of the vehicle.  But there's no escaping the reflection off Earth when you're in low orbit.  It's difficult to shed heat in space, since again, there's no air flow to conduct it away, it all has to leave in the form of infrared radiation.  This is a slow process, requiring huge radiators, which again, need to somehow be shaded from both the Earth and the Sun or else they'll become heaters.  Also, since they need to reject enough heat to cool the tanks down to cryogenic temperatures, they'll have to be well below the temperature of the propellant, or else you'll need a heat pump system much like air conditioning, but this generates a lot of waste heat which now must be dealt with as well...

It's simply very difficult to store propellant in orbit for long periods of time in any practical fashion.

Fuel will certainly be the biggest challenge for Mars missions.
Not really. Not if you can get replacement fuel up cheaply enough. But it's still better to fill in LEO immediately before departure.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/23/2020 04:56 pm
[SNIP]

- Yes, you can do it with low pressure pumps and as a benefit, you don't have to vent any gases.

This.

You pump the gas from the empty ship into the tanker, this pushes the liquid into the empty ship.
(or the other way around if that's your preference ie, pump the liquid which pushes the gas)

It would be a roughly equal volumetric exchange so no need to vent or compress anything.  :)

No, if you are using a pressure difference, the gas from the tank being filled is lower than the tank doing the filling, so the gas has to be either condensed or vented.

John
With autogenous pressurization, conceivably the pressure in the receiving tanks could be tapered off during the last burn, within operational limits. There is also a reasonable chance that autogenous gasses will distribute to high pressure tanks before distribution to lower pressure systems.


The Depot will have its own gas storage system that could be kept charged from boiloff or, save the boiloff for cryo cooler condensation and allow the receiving ship to recharge the depot high pressure gas system for general thruster use and tank press during refueling.


This could face some refinement but at first blush it seems to replenish consumables on the depot while minimizing propellant transfer wastage, at the cost of two more quick disconnects.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/23/2020 05:39 pm
-I dont know what is SH/SS payload to LEO, but let imagine some sort of expandable second stage on top of SH what would be the payload? If you have an answer then that is the weight of one single fuel tank bound for the mechanism (depot) put in space.

-In terms of costs we must put the cost of building and maintaining a fleet of tankers into consideration, achieving full and rapid reusability will take some time so while awaiting for that moment expandable would be a better option as an interim solution.
 

Star-Dust, you have aspirations of sending rockets to Mars, that is admirable, and if rockets were fueled with persistence, you might already be in orbit.  My unsolicited advice to you is that like early aviators, take one step at a time and develop a firm grasp of the fundamentals before taking that giant leap. 

A great example is your lack of understanding that SH can’t deliver your proposed depot to orbit.  This is an easy mistake to make, but as a result of this error, you have been now forced to fabricate this new expendable stage.  You deride the cost of maintaining tankers, but now somehow building expendable second stages comes without a tremendous cost?  That cost comes in many forms, not just in dollars, but also in manufacturing time and a workforce that is now dedicated to building hardware that will be thrown away after every flight.  This is very inefficient. 

If you have paid attention to the work at BC, you can see that constructing a SS is a multi-month process.  How is SS minus wings and the payload section any different from your proposed second stage?  Are you really willing to wait months between depot launches?  Have you done any analysis of the cost to dispose of five stages vs. maintenance of a single tanker?  Now work the other side of the problem.  You have your depot, but without the stage you just threw away, you can’t change orbits or dock with other depots.  You will need power to keep the fuels at the proper pressure.  Where will this come from?  Are you now going to add solar arrays and compressors that you also dispose of along with communications equipment, RCS, batteries, fuel pumps, etc.?  How will you de-orbit your empty depot?

My recommendation is to stick with this first step.  Perhaps spend some time studying the capabilities of SH and SS so you can avoid further simple errors.  Look at necessary capabilities for that second stage you are proposing compared to a SS Tanker/Depot concept. Again, before you can convince us that you have a viable vision for how to get to Mars and back, at least do the minimum legwork and develop a viable plan for how to get just a single Starship worth of fuel to LEO without one year required to build your five expendable second stages, along with 30 or more Raptors that you plan to dispose of.  Yes, this will take some discipline, and you just might succeed in this first step and either make a compelling case for your depots, or garner the knowledge and understanding as to why so many who have studied this problem, some for decades, disagree with you.  Best of luck.

1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH, this is if the figures available are correct twice as much as a reusable SS capability in term of payload (100 tons) , and by definition will require twice less the number of flights, no need to remind  that SH on each flight require a 3400 tons of fuel+ 1200 tons SS ( 2.5 millions$ saved on fuel rough calculation based on 600$/ton LCH4, 200$/ton LOX).

2-A second stage by definition is a basic rocket which mean engines + fuel tank, I dont think that a stainless steel tank coupled with six raptors engines will weigh that much in the overall balance of a $/per ton of payload (no heat shield, no legs/no extra fuel,.........etc.).

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

4-No extra spending on O&M for tankers and less expensive 2S than a tanker.

5- Fuel depot could be sent to Mars orbit.

There are a few reasons I doubt the practicality of a fueling deopt.

First, orbital mechanics.  The problem is that orbits have to be properly aligned for the rocket to be able to rendevous with the depot and then proceed to Mars.  With a depot in low orbit, this might not be a problem, but something on a high eccentric orbit (such as needed for landing on the Moon) will rarely be in the right alignment.  You can imagine the oblong shape of the orbit being like an arrow, pointing the direction the vehicle has to depart in.  Most of the time, it will be pointing the wrong way.  Also, once the depot is in this orbit, it takes a lot of fuel and/or time to change this orbit, so you're largely stuck with whatever direction it happens to point.


Low orbits have a different problem, namely heat.  Cryogenic fuel must be kept very cold to stay liquid.  If it boils into a gas state, your tank will explode from the pressure much like a giant steam boiler that has been overheated.  But the environment in low earth orbit is very hot.  A good comparason would be to Death Valley or somewhere in the middle of the Sahara.  Except in orbit, there's no wind blowing in from cooler places to moderate temperatures.  The same problem happens in high orbits or interplanetary transit, but here the vehicle can do things like pointing the engines toward the sun to shade the rest of the vehicle.  But there's no escaping the reflection off Earth when you're in low orbit.  It's difficult to shed heat in space, since again, there's no air flow to conduct it away, it all has to leave in the form of infrared radiation.  This is a slow process, requiring huge radiators, which again, need to somehow be shaded from both the Earth and the Sun or else they'll become heaters.  Also, since they need to reject enough heat to cool the tanks down to cryogenic temperatures, they'll have to be well below the temperature of the propellant, or else you'll need a heat pump system much like air conditioning, but this generates a lot of waste heat which now must be dealt with as well...

It's simply very difficult to store propellant in orbit for long periods of time in any practical fashion.
Cryo cooling is hard - not impossible.


One strategy would be a small fixed sun shield over the nose and keep that end aimed at the sun. Include a larger mobile shade to shield earth. Radiator panels would extend anti-earthward with the edge towards the sun and the faces towards open space. Use a cryo cooler. Fit in solar panels where appropriate.


The biggest operational issue is propellant use for keeping everything aligned. ISS uses Control Moment Gyros for this making it a watt issue, not a consumption issue. The problems with this are power consumption, mechanical longevity and 'saturation'.


PV can be as big as needed. Mechanical longevity seems to have improved post ISS failure mode analysis, and (I'm on weak ground here) I think the CMB's can desaturate (unwind) during the The night period. Multiple small CMB's for redundancy would be smarter than a few big ones.


All in all, much more complex than a simple tank in orbit. The ultimate question is not if it's doable, but does it give enough advantage to be justified. My gut says keep things simple at first and move in this direction as flight ops mature. Hmmm. Sounds almost like the SX mantra.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/23/2020 07:47 pm
An interesting paper on liquification and zero boiloff of martian ISRU propellant. Because it assumes gravity it is not directly applicable to depot boiloff. Assumption on tank dimensions, temperature and pressure both ambient and of the feed stock, are also different.


It's value is in a methodology for a preliminary assessment of different approaches and numbers that give a rough feel for the magnitude of the job.


What was looked at for a given delivery volume was mass, power consumption, power rejection, cost, maintainability, scaleability, reliability and what they call Volume-ility - the physical size and how much the hardware lends itself to physical distribution.


A methodology was developed to assign a weighting to these characteristics, and systems compared.

https://www.sciencedirect.com/science/article/pii/S0011227517302187 (https://www.sciencedirect.com/science/article/pii/S0011227517302187)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: vholub on 08/23/2020 11:12 pm
Hi there!
I created a somewhat in depth video about orbital debris, MMOD protection and related potential Starship issue that bothered for a while. I would appreciate your thoughts on both the aforementioned problems and the video.

https://youtu.be/2FrBEuAMHzw

TLDR: Starship needs to be at least several days in orbit for refueling and has no protection against any debris larger than 1 mm. Potentially high risk of mission failure.

Dear mods, I really did not know where to place this post, so please put it anywhere you think it belongs, or erase it if it goes against any rules.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/23/2020 11:27 pm
ISTM that either SpaceX will quickly get up to speed and will be able to launch a lot of propellant in a short period of time so that only modest cooling or shading is needed OR they will have to develop a proper orbital depot for the Mars mission with a larger shade, electric power, attitude control, cryo-coolers and radiators etc we will see.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/24/2020 12:46 am
1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH

I bet you haven't, because...

I dont think that a stainless steel tank coupled with six raptors engines will weigh that much

...others have mentioned that the figures that estimate Starship's mass also estimate 40 tonnes for re-entry and landing. 25 tonnes for structure/heat-shield, and 15 tonnes for fuel.

Hence, if you're using a calculator that gives 100 tonnes payload (hence propellant) for the Starship tanker, then it will give 140 tonnes payload (prop) for the expendable version. If you got a higher figure, then you dropped the drymass to an unrealistic level.

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

As mentioned repeatedly: Everything you require for your expendable tanks are the same as the tanker. You need to launch on SH, you need to be able to do docking manoeuvres in space, you need to connect up propellant lines (and gas-return lines for tank pressurisation), you need to be able to transfer propellant from the tanks to the engines on the main vehicle.

In addition, because you seem to have (finally!) accepted that you can't launch a single expendable fully fuelled tank into orbit, you now invent another whole structure, the docking node that holds a bunch of tanks as a depot. The depot then has to transfer fuel to the main Mars vehicle... kind of like... you know... a tanker.

You have all the requirements of building tankers, plus you need a depot before it's usable. It's not an "interim" solution.

The only technology you are trying to skip is reusability, which only requires re-entry and landing. But if Starship can land on Mars, then the technology for reusable tankers has already been developed. If the technology to land tankers isn't ready, Starship can't go to Mars.

So again, there's no "interim" where the Mars Starship exists but the tankers don't. Either the landing technology is ready, making tankers viable, or you can't land on Mars.

5- Fuel depot could be sent to Mars orbit.

Having fuel in Mars orbit doesn't replace ISRU production for a manned Starship mission: Starship can't land while fully fuelled, hence can't launch back into Mars orbit without ISRU propellant on the ground. If they have that, then your proposal isn't needed. If they don't, then you can't send humans to Mars. Again, there's no "interim".
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 08/24/2020 02:22 am
[SNIP]

- Yes, you can do it with low pressure pumps and as a benefit, you don't have to vent any gases.

This.

You pump the gas from the empty ship into the tanker, this pushes the liquid into the empty ship.
(or the other way around if that's your preference ie, pump the liquid which pushes the gas)

It would be a roughly equal volumetric exchange so no need to vent or compress anything.  :)

No, if you are using a pressure difference, the gas from the tank being filled is lower than the tank doing the filling, so the gas has to be either condensed or vented.

John

Sorry, perhaps I wasn't being clear, I'm talking about displacement exchange to negate the need to compress or vent gas.

When liquid is transferred from the tanker to the recipient, the tanker pressure drops and the recipient pressure increases. The excess recipient pressure needs to be reduced (either by venting or compression) and the tanker pressure needs to be maintained to aid pumping.

I'm saying solve two problems at once by transferring the excess pressure, in the form of gas, from the recipient to the tanker.

The pumping can occur either by pumping gas into the tanker to push the liquid through or by pumping liquid into the recipient and allow excess pressure gas to flow into the tanker. Either way you're not wasting energy compressing gas or wasting gas by venting.

Microgravity can be used to assist the flow from one to the other.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/24/2020 02:51 am
This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.

It isn't just delta-v. Producing enough propellant to launch enough propellant as cargo to Earth, to fully fuel its own return plus one SS flight puts a burden on the ISRU production that puts disproportionate burdens on a Mars settlement that it doesn't on a few extra Earth launches.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/24/2020 02:51 am
The biggest problem with filling LEO tankers with non-terran methane, is that departure burns from LEO put the exaust in a suborbital trajectory. And we have enough problems with co2 on earth without importing martian Co2.

Terran methane would just be recycled back to earth, but martian methane would be a pure import.
This is an interesting idea. I don't think whether it's martian or terran methane makes a big difference, tho, to the outcome.

It's worth considering that long-lived water vapor is a really powerful greenhouse gas. So you want to minimize Earth launches and consider alternate propellants. Ironically, CO/O2 is a better propellant from this perspective as the CO2 is a MUCH less powerful greenhouse gas than water vapor and CO2 already has a really long life in the atmosphere (~200 years), so adding another 5 years or so makes little difference (just install more direct air capture capacity on the ground). Additionally, you could go for Solar Electric Propulsion and use inert (and infrared-transparent) gases like Argon (which Mars has plenty of) for Earth departure. SpaceX has been also proving super cheap SEP using Starlink. In either case, getting the propellant from Mars reduces the number of Earth launches required for a given Mars trip and would reduce the greenhouse gas impact. Also, it might be cheaper in the long-run.

And the flip side (to water vapor being a really powerful greenhouse gas if placed in the upper atmosphere where its lifetime is much greater) is that using Mars to launch a whole bunch of stuff to Earth would actually help *terraform* Mars as a side effect!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/24/2020 02:52 am
most of the propellant for trips to Mars might actually come *from* Mars.

If Deimos' very low density (https://solarsystem.nasa.gov/moons/mars-moons/deimos/by-the-numbers/) truly indicates volatiles, propellant for both Mars launch and LEO launch (https://forum.nasaspaceflight.com/index.php?topic=45299.msg1802366#msg1802366) could be produced more efficiently on Deimos than on Mars.  At scale, there'd be no need for Mars ISRU propellant (https://forum.nasaspaceflight.com/index.php?topic=45299.msg1802366#msg1802366), or for a tanker SH fleet - apart from the possibility of a small Earth-orbit depot fleet.

Various Earth-Moon-Mars propellant scenarios could be calculated with the network flow modeling method for optimized logistics in Ishimatsu et al. 2016.  Note their particular interest in Deimos:

Quote from: Ishimatsu et al. 2016
...it is interesting that part of the resources produced on Deimos are delivered back to GTO and LEO, and wait to be used for the crew and cargo outbound trip. Though it seems strange and non-intuitive at first, this is true at least computationally because in terms of ∆V, LEO is closer to Deimos than to the lunar surface and even Earth’s surface.

Image:  Network graph featuring Deimos ISRU propellant.  Ishimatsu et al. 2016, Fig. 10e.

Refs.

Ishimatsu, T., de Weck, O.L., Hoffman, J.A., Ohkami, Y. and Shishko, R., 2016. Generalized multicommodity network flow model for the earth–moon–mars logistics system. (https://dspace.mit.edu/bitstream/handle/1721.1/99360/JSR_Final_Manuscript_Ishimatsu.pdf?sequence=1&isAllowed=y) Journal of Spacecraft and Rockets, 53(1), pp.25-38.

I'm skeptical about the Martian moons for direct ISRU. But using them for a delta-v assist using a modest tether IS interesting and could help a lot without requiring magic materials (for Earth elevator) or super long transit times (in the case of a lunar elevator).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/24/2020 02:56 am
For some of you guys information, the US has a 200 year supply of natural gas drilled, tapped, and capped, without drilling another well.  Producing methane from CO2 and water on earth costs more than just buying existing natural gas.  Unless Musk wants to make it anyways, the first methane rockets for several years, will be extracted from the existing 95% methane natural gas.  Coal, oil, and gasoline, produce far far more CO2 per year than natural gas.  Rockets will be a minor use, even at high flight rates.
Do the math on the 100 Megatons-per-year tweet that Musk mentioned earlier. It's something like a third of the entire world's natural gas production. It's not a rounding error. Particularly since it's putting water vapor (a strong greenhous gas) in the stratosphere (and upward) where the lifetime is years, not days like it is in the troposphere. That's like 10 times a bigger problem than the CO2 alone.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/24/2020 02:59 am
ISTM that either SpaceX will quickly get up to speed and will be able to launch a lot of propellant in a short period of time so that only modest cooling or shading is needed OR they will have to develop a proper orbital depot for the Mars mission with a larger shade, electric power, attitude control, cryo-coolers and radiators etc we will see.
SpaceX already mentioned a depot variant of Starship for their lunar mission.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/24/2020 02:59 am
This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.

It isn't just delta-v. Producing enough propellant to launch enough propellant as cargo to Earth, to fully fuel its own return plus one SS flight puts a burden on the ISRU production that puts disproportionate burdens on a Mars settlement that it doesn't on a few extra Earth launches.
Moving that industrial burden off of the Earth and on to Mars is a net-plus for the biosphere of Earth, long-term. And gives a Mars settlement a legitimate revenue stream. And helps terraform Mars through side effects of a really high launch rate.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/24/2020 03:15 am
AIUI, gasses in the upper atmosphere get banged around quite a bit by UV. Is this enough to disassociate CH4?

Burning it in oxygen sure would.

(Rockets are releasing combustion products, not raw fuel.)

[Also, what Rakaydos said. Should we be concerned about the accumulation of re-entry heat? Changes to Earth's rotational inertia due to mass loss to deep space missions? Yes. Yes we should. Once we're a K1+ civilisation, there are many such things we'll need to manage. Should we be concerned now? No.]
Reentry heat is not a major concern (it's basically just waste heat, which is a factor of ~100 less important than anthropogenic greenhouse gases in human-caused warming of the atmosphere) directly. But it CAN cause nitrogen oxides, which ARE a significant problem for various reasons. Something like 5% of reentered mass from LEO is converted into nitrogen oxides. Although this is highly non-linear, so I think you can reduce it by doing some more propulsive reentry. Might tip the scales toward more solar electric propulsion. And that amount of nitrogen oxides becomes a real problem at the flight levels SpaceX wants. But we CAN engineer around it, like I mentioned.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/24/2020 03:39 am
Microgravity can be used to assist the flow from one to the other.

For ullage, "milli" not "micro".

Not sure if it's enough to act as a steep enough hill to not need a dedicated pump. With 150 tonnes cargo-propellant plus 15% landing-propellant, the initial "pressure" is 3kN per milligee. (Assuming 3 milli-gee (typical ISS thruster burn), a half metre sump, and completely ignoring the fuel/ox ratio, you get around 19kPa / 2.8psi, dropping to 10kPa/1.5psi. Is that enough?)



This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.
[...] And gives a Mars settlement a legitimate revenue stream.

It's not a revenue stream. It's just paying for the cost of supporting a Mars settlement in a different way.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 08/24/2020 03:39 am
1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH

I bet you haven't, because...

I dont think that a stainless steel tank coupled with six raptors engines will weigh that much

...others have mentioned that the figures that estimate Starship's mass also estimate 40 tonnes for re-entry and landing. 25 tonnes for structure/heat-shield, and 15 tonnes for fuel.

Hence, if you're using a calculator that gives 100 tonnes payload (hence propellant) for the Starship tanker, then it will give 140 tonnes payload (prop) for the expendable version. If you got a higher figure, then you dropped the drymass to an unrealistic level.


For the sake of argument, lets just assume that 200 tonnes payload second stage. It's really just an expendable Starship: no legs, fins, TPS & header tanks.  If it can do 200, a reusable tanker can do 150 tonnes.  Let's also ignore the mass of the drop tank, which would have to be close to 10 tonnes.

If a tanker Starship can be built for $5M (SpaceX stretch goal), then an expendable SS might be $3.5M and a drop tank $0.5M.  (Probably way more with all the extra fittings and equipment on the drop tank, but let's be extra charitable.)

To refill a Starship in orbit with 1200 tonnes of propellant, assuming only 8 lifetime uses of tanker, is: 8 x $2.5M + $5M = $25M

To launch the equivalent drop tanks using expendable stages is: 6 x ($2.5M + $4M) = $39M

If the number of reuses for a Starship goes up, the numbers get worse for drop tanks.
If the cost of a Starship is higher, the numbers get much worse for drop tanks.


Absurdly, on these numbers they could even launch completely expendable tanker Starships, carrying more payload propellant per launch than would be in drop tanks, for less cost than expendable stages with drop tanks:  6 x ($2.5M + $3.5M) = $36M
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/24/2020 03:45 am
...
This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.
[...] And gives a Mars settlement a legitimate revenue stream.

It's not a revenue stream. It's just paying for the cost of supporting a Mars settlement in a different way.
I imagine there are more users of cheap Mars volatiles in LEO than just the Mars settlement, but sure, valid. Point is that it gives the Mars settlement folks something to do while relieving Earth of some of the work. Exporting some of the economy and industrial capacity to Mars.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 08/24/2020 03:53 am
...
This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.
[...] And gives a Mars settlement a legitimate revenue stream.

It's not a revenue stream. It's just paying for the cost of supporting a Mars settlement in a different way.
I imagine there are more users of cheap Mars volatiles in LEO than just the Mars settlement, but sure, valid. Point is that it gives the Mars settlement folks something to do while relieving Earth of some of the work. Exporting some of the economy and industrial capacity to Mars.

I kind of think the Mars settlement folks will have quite a lot to do already.  Like trying to bootstrap an entire planetary industrialization.

Edit:  The cost of sending the resources to Mars to make the extra propellant to send back to LEO, would have to be less than the resources to just launch the prop to LEO in the first place.  What percentage of propellant made on Mars would reach LEO? 5%? 10%?

Edit 2:  I get 9% (112 tonnes) assuming aerocapture into LEO and no orbit adjustment burns. For a 120 tonne dry mass, 1200 tonne propellant, 370 ISP.  Of course that SS would then be stuck in LEO unless it reserved 15 tonnes for EDL, so 8%.   And how does the value of 120 tonnes of used SS on Mars compare to 97 tonnes of propellant in LEO?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: livingjw on 08/24/2020 04:13 am


No, if you are using a pressure difference, the gas from the tank being filled is lower than the tank doing the filling, so the gas has to be either condensed or vented.

John

Sorry, perhaps I wasn't being clear, I'm talking about displacement exchange to negate the need to compress or vent gas.

When liquid is transferred from the tanker to the recipient, the tanker pressure drops and the recipient pressure increases. The excess recipient pressure needs to be reduced (either by venting or compression) and the tanker pressure needs to be maintained to aid pumping.

I'm saying solve two problems at once by transferring the excess pressure, in the form of gas, from the recipient to the tanker.

The pumping can occur either by pumping gas into the tanker to push the liquid through or by pumping liquid into the recipient and allow excess pressure gas to flow into the tanker. Either way you're not wasting energy compressing gas or wasting gas by venting.

Microgravity can be used to assist the flow from one to the other.

- Pure displacement exchange will not be nearly fast enough without a pump to provide a few bar of pressure to speed it along.

- If you are using a pressure difference of a few bar, then you are introducing more gas into the tanks. For example:
 1) we have two tanks connected as shown, both at 1 bar, with all valves closed with propellants settled with micro g.
 2) if we open valves 1 and 5 on both vehicles fluid would slowly flow to the receiver while ullage gas flows to the tanker
 3) Not practical because it is too slow.
 4) If we introduce a 3-5 bar pump in the feed line we can speed up the transfer greatly and no venting required.
 5) Or we could provide a delta pressure using a high pressure gas source to introduce a 3-5 bar pressure difference but we would have to close off the receiving tank's valve 1. Fluid would flow until the ullage gas in the receiving tank reached a little over 3-5 bars, then no more flow. You can't finish transferring the fluid. You would have to vent the receiving tank using receiving valve 3.

John
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/24/2020 04:21 am
...
This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.
[...] And gives a Mars settlement a legitimate revenue stream.

It's not a revenue stream. It's just paying for the cost of supporting a Mars settlement in a different way.
I imagine there are more users of cheap Mars volatiles in LEO than just the Mars settlement, but sure, valid. Point is that it gives the Mars settlement folks something to do while relieving Earth of some of the work. Exporting some of the economy and industrial capacity to Mars.

I kind of think the Mars settlement folks will have quite a lot to do already.  Like trying to bootstrap an entire planetary industrialization.
I said at the very first it's a long-term thing. But rockets & propellant is something they better get REALLY good at REALLY fast.

Edit:  The cost of sending the resources to Mars to make the extra propellant to send back to LEO, would have to be less than the resources to just launch the prop to LEO in the first place.  What percentage of propellant made on Mars would reach LEO? 5%? 10%?

Edit 2:  I get 9% (112 tonnes) assuming aerocapture into LEO and no orbit adjustment burns. For a 120 tonne dry mass, 1200 tonne propellant, 370 ISP.  Of course that SS would then be stuck in LEO unless it reserved 15 tonnes for EDL, so 8%.   And how does the value of 120 tonnes of used SS on Mars compare to 97 tonnes of propellant in LEO?
About 30% using SEP and perhaps a Deimos tether (compare to about 3% for Earth's surface to LEO). But mass of propellant is less relevant than efficiency in this case. You could launch the propellant to LMO using CO/O2 rockets on Mars. The lower Isp is compensated nicely by the reduced energy cost per kg of propellant. Atmospheric ISRU on Mars may actually be easier than on Earth as direct air capture of CO2 is not cheap on Earth (but is easy on Mars).

Just the CH4 and O2 (or whatever propellant) would be needed to be delivered to LEO.

Also, you'd send it in large tanker variants of Starship fairly optimized for the trip. Not just regular versions of Starship. SpaceX already mentioned both tanker and depot variants of Starship. Most importantly: You wouldn't send the ship directly from mars to Earth but instead refuel a Starship tanker right at high Mars orbit, right before Earth injection burn. Dry mass-to-wet-mass would be reduced this way to around 0.1 instead of your 0.6 or so. (This makes an enormous difference.)

Secondly, you'd likely only put a small portion of the propellant in LEO. It makes sense to also refuel at HEO on the way to Mars, and if your propellant is coming from Mars anyway, might as well keep much of it in HEO to begin with as it gives you a greater benefit on the departure burn than the same amount of propellant in LEO. EDIT: Splitting the departure propellant between LEO and HEO like this almost halves the total amount of propellant needed for the same delta-v (say, 7.5km/s total starting in LEO), and the effect is significantly larger for faster transits than slower transit. I did NOT include that consideration below... If I did, the energy advantage for Mars over Earth would be about 15-20:1 instead of 10:1.

So again, I think you'd have roughly a 10:1 energy advantage (knocked down to, say, 4:1 by the lower sunlight intensity at Mars... which could be more-than-compensated-for, long-term, using areosynchronous orbital mirrors once you get to Terawatts of power... something you really shouldn't be doing at Earth... although this is all irrelevant if you use nuclear power which should be similar for Earth and Mars... tho the details get hairy).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/24/2020 04:38 am
The pumping can occur either by pumping gas into the tanker to push the liquid through or by pumping liquid into the recipient and allow excess pressure gas to flow into the tanker. Either way you're not wasting energy compressing gas or wasting gas by venting.
- Pure displacement exchange will not be nearly fast enough without a pump

You keep misreading what he is saying.

Neyyn72 didn't say "without a pump". Quite the opposite, his point was that pumping the gas from receiver to tanker provides both pressure reduction in the receiver, and increased pressurant in the tanker. Exactly as in your diagram. He was responding to people concerned about a) the increasing pressure in the receiver-tank as it fills, requiring venting or cryo-cooling, and b) a lack of pressure in the tanker, and was simply saying "Uh, guys, you realise that one solves the other?"

I'm the one who speculated that the whole thing might work without a pump using just the pressure created by milli-gee ullage thrust. You should get a few PSI fluid pressure at the sump.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 08/24/2020 05:37 am
About 30% using SEP and perhaps a Deimos tether (compare to about 3% for Earth's surface to LEO). But mass of propellant is less relevant than efficiency in this case. You could launch the propellant to LMO using CO/O2 rockets on Mars. The lower Isp is compensated nicely by the reduced energy cost per kg of propellant. Atmospheric ISRU on Mars may actually be easier than on Earth as direct air capture of CO2 is not cheap on Earth (but is easy on Mars).

Just the CH4 and O2 (or whatever propellant) would be needed to be delivered to LEO.

Also, you'd send it in large tanker variants of Starship fairly optimized for the trip. Not just regular versions of Starship. SpaceX already mentioned both tanker and depot variants of Starship. Most importantly: You wouldn't send the ship directly from mars to Earth but instead refuel a Starship tanker right at high Mars orbit, right before Earth injection burn. Dry mass-to-wet-mass would be reduced this way to around 0.1 instead of your 0.6 or so. (This makes an enormous difference.)

Secondly, you'd likely only put a small portion of the propellant in LEO. It makes sense to also refuel at HEO on the way to Mars, and if your propellant is coming from Mars anyway, might as well keep much of it in HEO to begin with as it gives you a greater benefit on the departure burn than the same amount of propellant in LEO. EDIT: Splitting the departure propellant between LEO and HEO like this almost halves the total amount of propellant needed for the same delta-v (say, 7.5km/s total starting in LEO), and the effect is significantly larger for faster transits than slower transit. I did NOT include that consideration below... If I did, the energy advantage for Mars over Earth would be about 15-20:1 instead of 10:1.

So again, I think you'd have roughly a 10:1 energy advantage (knocked down to, say, 4:1 by the lower sunlight intensity at Mars... which could be more-than-compensated-for, long-term, using areosynchronous orbital mirrors once you get to Terawatts of power... something you really shouldn't be doing at Earth... although this is all irrelevant if you use nuclear power which should be similar for Earth and Mars... tho the details get hairy).

Ok, for something quite a bit more technically complex, (than just refueling a Starship on Mars then launching it), this makes a bit more sense. 

I hadn't factored in the 'staging' like advantages of depots in High Mars/Earth Orbit at each end which do reduce the delta v of each leg, so increase the overall efficiency.  I already think Starship storage tankers (depots) are going to end up being used at High & Low, Earth & Mars Orbits.

On the solar PV efficiency question, it might make sense to ship the 'raw feed-stock' as far down the logistics pipe as you can, before converting it to usable propellant.  Solar PV is more efficient in HMO than on the surface, and HEO more so than HMO.  Also liquid CO2 and H2O should be easier to store than LCH4 & LOX, let alone LH2.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 08/24/2020 06:31 am
The pumping can occur either by pumping gas into the tanker to push the liquid through or by pumping liquid into the recipient and allow excess pressure gas to flow into the tanker. Either way you're not wasting energy compressing gas or wasting gas by venting.
- Pure displacement exchange will not be nearly fast enough without a pump

You keep misreading what he is saying.

Neyyn72 didn't say "without a pump". Quite the opposite, his point was that pumping the gas from receiver to tanker provides both pressure reduction in the receiver, and increased pressurant in the tanker. Exactly as in your diagram. He was responding to people concerned about a) the increasing pressure in the receiver-tank as it fills, requiring venting or cryo-cooling, and b) a lack of pressure in the tanker, and was simply saying "Uh, guys, you realise that one solves the other?"

I'm the one who speculated that the whole thing might work without a pump using just the pressure created by milli-gee ullage thrust. You should get a few PSI fluid pressure at the sump.

Exactly!  :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 08/24/2020 07:22 am
Just for fun, I ran the numbers for percentage of Mars produced propellant reaching High Earth Orbit, using just standard methalox Starships as tankers, and depots in LMO and HMO:

I get 13% using propulsive capture into HEO, and 15% using aero-capture.

Eliminating legs, fins and heat-shield for the LMO-HMO & HMO-HEO legs makes negligible difference. (Except on the return leg).

Switching to hydrolox for LMO-HMO & HMO-HEO gives 15% (assuming same MR, which would be worse in reality, due to larger tanks)

Switching to hydrolox for every leg gives 19% (assuming same MR, which would be a lot worse in reality, due to EDL for larger tanks)


Everything else being equal, which it isn't, fully Hydrolox makes 50% more propellant available in HEO, which because you need less, enables 75% more payload than methalox.  More like 40% and 70% with propellant density factored in, and who know with ISRU production & storage factored in.


Using methalox for Mars to LMO, then SEP, I got 21% to 24% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then SEP, I got 14% to 15% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then hydrolox, I got 10%.

Using CO2/LOX, for Mars to LMO, then methalox , I got 8%. Might as well just send a Starship direct from the surface.


NOTE: All these trips are one-way. When you need to send an empty tanker back to HMO or LMO, it requires about 16% of the propellant just delivered for methalox, and 21% for hydrolox (lower density -> higher empty mass). Around 12% for SEP.

Edit: fix return %
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/24/2020 10:31 am
1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH

I bet you haven't, because...

I dont think that a stainless steel tank coupled with six raptors engines will weigh that much

...others have mentioned that the figures that estimate Starship's mass also estimate 40 tonnes for re-entry and landing. 25 tonnes for structure/heat-shield, and 15 tonnes for fuel.

Hence, if you're using a calculator that gives 100 tonnes payload (hence propellant) for the Starship tanker, then it will give 140 tonnes payload (prop) for the expendable version. If you got a higher figure, then you dropped the drymass to an unrealistic level.

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

As mentioned repeatedly: Everything you require for your expendable tanks are the same as the tanker. You need to launch on SH, you need to be able to do docking manoeuvres in space, you need to connect up propellant lines (and gas-return lines for tank pressurisation), you need to be able to transfer propellant from the tanks to the engines on the main vehicle.

In addition, because you seem to have (finally!) accepted that you can't launch a single expendable fully fuelled tank into orbit, you now invent another whole structure, the docking node that holds a bunch of tanks as a depot. The depot then has to transfer fuel to the main Mars vehicle... kind of like... you know... a tanker.

You have all the requirements of building tankers, plus you need a depot before it's usable. It's not an "interim" solution.

The only technology you are trying to skip is reusability, which only requires re-entry and landing. But if Starship can land on Mars, then the technology for reusable tankers has already been developed. If the technology to land tankers isn't ready, Starship can't go to Mars.

So again, there's no "interim" where the Mars Starship exists but the tankers don't. Either the landing technology is ready, making tankers viable, or you can't land on Mars.

5- Fuel depot could be sent to Mars orbit.

Having fuel in Mars orbit doesn't replace ISRU production for a manned Starship mission: Starship can't land while fully fuelled, hence can't launch back into Mars orbit without ISRU propellant on the ground. If they have that, then your proposal isn't needed. If they don't, then you can't send humans to Mars. Again, there's no "interim".

Just some remarques:

-Please don't get my phrases out of context like;
 
I dont think that a stainless steel tank coupled with six raptors engines will weigh that much

I dont think that a stainless steel tank coupled with six raptors engines will weigh that much in the overall balance of a $/per ton of payload (no heat shield, no legs/no extra fuel,.........etc.).


Quote
In addition, because you seem to have (finally!) accepted that you can't launch a single expendable fully fuelled tank into orbit, you now invent another whole structure, the docking node that holds a bunch of tanks as a depot. The depot then has to transfer fuel to the main Mars vehicle... kind of like... you know... a tanker.

I didn't accepted anything it was just to stay in Elon Musk philosophy of one versatile spacecraft namely SS, if I was to send fuel to orbit I will do it in a space shuttle manner I will put engines to the main tanker and two boosters (reusable).

Quote
As mentioned repeatedly: Everything you require for your expendable tanks are the same as the tanker. You need to launch on SH, you need to be able to do docking manoeuvres in space, you need to connect up propellant lines (and gas-return lines for tank pressurisation), you need to be able to transfer propellant from the tanks to the engines on the main vehicle.

The difference is that your tanker will require months of checks and refrubishments to be ready for the next mission, I don't think SS in orbit will be able await for that time, of course until SX reach a rapid reuse but this will require some time and considering EM time frame of Mars missions this won't be possible.



Quote
Having fuel in Mars orbit doesn't replace ISRU production for a manned Starship mission: Starship can't land while fully fuelled, hence can't launch back into Mars orbit without ISRU propellant on the ground. If they have that, then your proposal isn't needed. If they don't, then you can't send humans to Mars. Again, there's no "interim".

ISRU is also a long term reach goal, but still if we want getting the red planet relatively in short time it could still be possible with some SS modifications (interim solution or plan B).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/24/2020 11:34 am
Quote
In addition, because you seem to have (finally!) accepted that you can't launch a single expendable fully fuelled tank into orbit, you now invent another whole structure, the docking node that holds a bunch of tanks as a depot. The depot then has to transfer fuel to the main Mars vehicle... kind of like... you know... a tanker.

I didn't accepted anything it was just to stay in Elon Musk philosophy of one versatile spacecraft namely SS, if I was to send fuel to orbit I will do it in a space shuttle manner I will put engines to the main tanker and two boosters (reusable).

Quote
As mentioned repeatedly: Everything you require for your expendable tanks are the same as the tanker. You need to launch on SH, you need to be able to do docking manoeuvres in space, you need to connect up propellant lines (and gas-return lines for tank pressurisation), you need to be able to transfer propellant from the tanks to the engines on the main vehicle.

The difference is that your tanker will require months of checks and refrubishments to be ready for the next mission, I don't think SS in orbit will be able await for that time, of course until SX reach a rapid reuse but this will require some time and considering EM time frame of Mars missions this won't be possible.

You dont seem to be making this connection. Shuttle took months of checks and refurbishments because it was built like the Shuttle- ie: by a committee in the 1970s.

Even oldspace tycoons like Tory Bruno believe a modern design can easilly reach fast turnaround times, though he was talking about the SSTO Venture Star.

https://twitter.com/torybruno/status/1296159004473937920?s=20

(discussion thread here, please read full tweetstorm before commenting: https://forum.nasaspaceflight.com/index.php?topic=51773.msg2122619#msg2122619)

Starship is being designed for fast turnaround, using lessons learned not just from the shuttle, but from SpaceX's own Falcon 9 program, which has surpassed the fastest ever shuttle turnaround in launch-to-launch. But not even the makers of the shuttle will touch a shuttle design anymore.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/24/2020 12:13 pm
Quote
In addition, because you seem to have (finally!) accepted that you can't launch a single expendable fully fuelled tank into orbit, you now invent another whole structure, the docking node that holds a bunch of tanks as a depot. The depot then has to transfer fuel to the main Mars vehicle... kind of like... you know... a tanker.

I didn't accepted anything it was just to stay in Elon Musk philosophy of one versatile spacecraft namely SS, if I was to send fuel to orbit I will do it in a space shuttle manner I will put engines to the main tanker and two boosters (reusable).

Quote
As mentioned repeatedly: Everything you require for your expendable tanks are the same as the tanker. You need to launch on SH, you need to be able to do docking manoeuvres in space, you need to connect up propellant lines (and gas-return lines for tank pressurisation), you need to be able to transfer propellant from the tanks to the engines on the main vehicle.

The difference is that your tanker will require months of checks and refrubishments to be ready for the next mission, I don't think SS in orbit will be able await for that time, of course until SX reach a rapid reuse but this will require some time and considering EM time frame of Mars missions this won't be possible.

You dont seem to be making this connection. Shuttle took months of checks and refurbishments because it was built like the Shuttle- ie: by a committee in the 1970s.

Even oldspace tycoons like Tory Bruno believe a modern design can easilly reach fast turnaround times, though he was talking about the SSTO Venture Star.

https://twitter.com/torybruno/status/1296159004473937920?s=20

(discussion thread here, please read full tweetstorm before commenting: https://forum.nasaspaceflight.com/index.php?topic=51773.msg2122619#msg2122619)

Starship is being designed for fast turnaround, using lessons learned not just from the shuttle, but from SpaceX's own Falcon 9 program, which has surpassed the fastest ever shuttle turnaround in launch-to-launch. But not even the makers of the shuttle will touch a shuttle design anymore.

I meant shuttle external tank (+ engines) + the two boosters without the orbiters.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/24/2020 12:24 pm
Quote
In addition, because you seem to have (finally!) accepted that you can't launch a single expendable fully fuelled tank into orbit, you now invent another whole structure, the docking node that holds a bunch of tanks as a depot. The depot then has to transfer fuel to the main Mars vehicle... kind of like... you know... a tanker.

I didn't accepted anything it was just to stay in Elon Musk philosophy of one versatile spacecraft namely SS, if I was to send fuel to orbit I will do it in a space shuttle manner I will put engines to the main tanker and two boosters (reusable).

Quote
As mentioned repeatedly: Everything you require for your expendable tanks are the same as the tanker. You need to launch on SH, you need to be able to do docking manoeuvres in space, you need to connect up propellant lines (and gas-return lines for tank pressurisation), you need to be able to transfer propellant from the tanks to the engines on the main vehicle.

The difference is that your tanker will require months of checks and refrubishments to be ready for the next mission, I don't think SS in orbit will be able await for that time, of course until SX reach a rapid reuse but this will require some time and considering EM time frame of Mars missions this won't be possible.

You dont seem to be making this connection. Shuttle took months of checks and refurbishments because it was built like the Shuttle- ie: by a committee in the 1970s.

Even oldspace tycoons like Tory Bruno believe a modern design can easilly reach fast turnaround times, though he was talking about the SSTO Venture Star.

https://twitter.com/torybruno/status/1296159004473937920?s=20

(discussion thread here, please read full tweetstorm before commenting: https://forum.nasaspaceflight.com/index.php?topic=51773.msg2122619#msg2122619)

Starship is being designed for fast turnaround, using lessons learned not just from the shuttle, but from SpaceX's own Falcon 9 program, which has surpassed the fastest ever shuttle turnaround in launch-to-launch. But not even the makers of the shuttle will touch a shuttle design anymore.

I meant shuttle external tank (+ engines) + the two boosters without the orbiters.
Exactly. Even oldspace thinks same day turnaround is possible if you get rid of that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/24/2020 12:32 pm
About 30% using SEP and perhaps a Deimos tether (compare to about 3% for Earth's surface to LEO). But mass of propellant is less relevant than efficiency in this case. You could launch the propellant to LMO using CO/O2 rockets on Mars. The lower Isp is compensated nicely by the reduced energy cost per kg of propellant. Atmospheric ISRU on Mars may actually be easier than on Earth as direct air capture of CO2 is not cheap on Earth (but is easy on Mars).

Just the CH4 and O2 (or whatever propellant) would be needed to be delivered to LEO.

Also, you'd send it in large tanker variants of Starship fairly optimized for the trip. Not just regular versions of Starship. SpaceX already mentioned both tanker and depot variants of Starship. Most importantly: You wouldn't send the ship directly from mars to Earth but instead refuel a Starship tanker right at high Mars orbit, right before Earth injection burn. Dry mass-to-wet-mass would be reduced this way to around 0.1 instead of your 0.6 or so. (This makes an enormous difference.)

Secondly, you'd likely only put a small portion of the propellant in LEO. It makes sense to also refuel at HEO on the way to Mars, and if your propellant is coming from Mars anyway, might as well keep much of it in HEO to begin with as it gives you a greater benefit on the departure burn than the same amount of propellant in LEO. EDIT: Splitting the departure propellant between LEO and HEO like this almost halves the total amount of propellant needed for the same delta-v (say, 7.5km/s total starting in LEO), and the effect is significantly larger for faster transits than slower transit. I did NOT include that consideration below... If I did, the energy advantage for Mars over Earth would be about 15-20:1 instead of 10:1.

So again, I think you'd have roughly a 10:1 energy advantage (knocked down to, say, 4:1 by the lower sunlight intensity at Mars... which could be more-than-compensated-for, long-term, using areosynchronous orbital mirrors once you get to Terawatts of power... something you really shouldn't be doing at Earth... although this is all irrelevant if you use nuclear power which should be similar for Earth and Mars... tho the details get hairy).

Ok, for something quite a bit more technically complex, (than just refueling a Starship on Mars then launching it), this makes a bit more sense. 

I hadn't factored in the 'staging' like advantages of depots in High Mars/Earth Orbit at each end which do reduce the delta v of each leg, so increase the overall efficiency.  I already think Starship storage tankers (depots) are going to end up being used at High & Low, Earth & Mars Orbits.

On the solar PV efficiency question, it might make sense to ship the 'raw feed-stock' as far down the logistics pipe as you can, before converting it to usable propellant.  Solar PV is more efficient in HMO than on the surface, and HEO more so than HMO.  Also liquid CO2 and H2O should be easier to store than LCH4 & LOX, let alone LH2.
I had never considered shipping water and CO2 from Mars into Mars orbit. It's a really neat idea that does help solve some problems, although I'm not exactly sure how practical it would be. It would need two dedicated tanks in Mars orbit one for water and CO2 plus one for methane and oxygen. There might also be issues with the water freezing. I suppose that could be solved by an electrically powered heater, but I'm not qualified to comment on the practicalities or trades of that arrangement.

One thing that does appeal is the thought of putting half the solar power in orbit where it could receive nearly double the energy from the sun and avoid dust storms and still have the other half on the surface for base operations. Unfortunately there would still need to be ISRU producing methalox on the surface in order to re-tank the ships heading up to the orbital facility. So its doable but I suspect the issues will prevent it's use until a lot further in the future when they're optimising the logistics chain.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/24/2020 03:53 pm
1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH

I bet you haven't, because...

I dont think that a stainless steel tank coupled with six raptors engines will weigh that much

...others have mentioned that the figures that estimate Starship's mass also estimate 40 tonnes for re-entry and landing. 25 tonnes for structure/heat-shield, and 15 tonnes for fuel.

Hence, if you're using a calculator that gives 100 tonnes payload (hence propellant) for the Starship tanker, then it will give 140 tonnes payload (prop) for the expendable version. If you got a higher figure, then you dropped the drymass to an unrealistic level.

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

As mentioned repeatedly: Everything you require for your expendable tanks are the same as the tanker. You need to launch on SH, you need to be able to do docking manoeuvres in space, you need to connect up propellant lines (and gas-return lines for tank pressurisation), you need to be able to transfer propellant from the tanks to the engines on the main vehicle.

In addition, because you seem to have (finally!) accepted that you can't launch a single expendable fully fuelled tank into orbit, you now invent another whole structure, the docking node that holds a bunch of tanks as a depot. The depot then has to transfer fuel to the main Mars vehicle... kind of like... you know... a tanker.

You have all the requirements of building tankers, plus you need a depot before it's usable. It's not an "interim" solution.

The only technology you are trying to skip is reusability, which only requires re-entry and landing. But if Starship can land on Mars, then the technology for reusable tankers has already been developed. If the technology to land tankers isn't ready, Starship can't go to Mars.

So again, there's no "interim" where the Mars Starship exists but the tankers don't. Either the landing technology is ready, making tankers viable, or you can't land on Mars.

5- Fuel depot could be sent to Mars orbit.

Having fuel in Mars orbit doesn't replace ISRU production for a manned Starship mission: Starship can't land while fully fuelled, hence can't launch back into Mars orbit without ISRU propellant on the ground. If they have that, then your proposal isn't needed. If they don't, then you can't send humans to Mars. Again, there's no "interim".
I can't speak for Stardust but I've considered a stripped SS without tank mods as a short term solution and an assembly capable of refueling an entire mars fleet as a long term solution. I see everything else as an interim solution. That said, your criticisms of Stardust's idea are on point.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/24/2020 04:34 pm
This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.

It isn't just delta-v. Producing enough propellant to launch enough propellant as cargo to Earth, to fully fuel its own return plus one SS flight puts a burden on the ISRU production that puts disproportionate burdens on a Mars settlement that it doesn't on a few extra Earth launches.
Thus, the suggestion for shipping CO2 & water and processing into propellant on the way back.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/24/2020 05:24 pm
Microgravity can be used to assist the flow from one to the other.

For ullage, "milli" not "micro".

Not sure if it's enough to act as a steep enough hill to not need a dedicated pump. With 150 tonnes cargo-propellant plus 15% landing-propellant, the initial "pressure" is 3kN per milligee. (Assuming 3 milli-gee (typical ISS thruster burn), a half metre sump, and completely ignoring the fuel/ox ratio, you get around 19kPa / 2.8psi, dropping to 10kPa/1.5psi. Is that enough?)



This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.
[...] And gives a Mars settlement a legitimate revenue stream.

It's not a revenue stream. It's just paying for the cost of supporting a Mars settlement in a different way.
Any G will settle props. The higher the G the faster they get to the bottom of the tank. The lower the G the less slosh at the cost of time. Solution: low G for initial settling, higher G to damp out slosh and entrained bubbles, back to low G for transfer. Allow the receiving ship to replenish the depots high pressure gas reserves which are used in turn to pressurize  the depots tanks to facilitate transfer.


As for characterizing as a cash flow or an offset to expenses, that's an artifact of how the books are set up. It's a positive on one side or a negative on the other side. Impact is the same either way. "That which we call a rose..."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/24/2020 08:26 pm
About 30% using SEP and perhaps a Deimos tether (compare to about 3% for Earth's surface to LEO). But mass of propellant is less relevant than efficiency in this case. You could launch the propellant to LMO using CO/O2 rockets on Mars. The lower Isp is compensated nicely by the reduced energy cost per kg of propellant. Atmospheric ISRU on Mars may actually be easier than on Earth as direct air capture of CO2 is not cheap on Earth (but is easy on Mars).

Just the CH4 and O2 (or whatever propellant) would be needed to be delivered to LEO.

Also, you'd send it in large tanker variants of Starship fairly optimized for the trip. Not just regular versions of Starship. SpaceX already mentioned both tanker and depot variants of Starship. Most importantly: You wouldn't send the ship directly from mars to Earth but instead refuel a Starship tanker right at high Mars orbit, right before Earth injection burn. Dry mass-to-wet-mass would be reduced this way to around 0.1 instead of your 0.6 or so. (This makes an enormous difference.)

Secondly, you'd likely only put a small portion of the propellant in LEO. It makes sense to also refuel at HEO on the way to Mars, and if your propellant is coming from Mars anyway, might as well keep much of it in HEO to begin with as it gives you a greater benefit on the departure burn than the same amount of propellant in LEO. EDIT: Splitting the departure propellant between LEO and HEO like this almost halves the total amount of propellant needed for the same delta-v (say, 7.5km/s total starting in LEO), and the effect is significantly larger for faster transits than slower transit. I did NOT include that consideration below... If I did, the energy advantage for Mars over Earth would be about 15-20:1 instead of 10:1.

So again, I think you'd have roughly a 10:1 energy advantage (knocked down to, say, 4:1 by the lower sunlight intensity at Mars... which could be more-than-compensated-for, long-term, using areosynchronous orbital mirrors once you get to Terawatts of power... something you really shouldn't be doing at Earth... although this is all irrelevant if you use nuclear power which should be similar for Earth and Mars... tho the details get hairy).

Ok, for something quite a bit more technically complex, (than just refueling a Starship on Mars then launching it), this makes a bit more sense. 

I hadn't factored in the 'staging' like advantages of depots in High Mars/Earth Orbit at each end which do reduce the delta v of each leg, so increase the overall efficiency.  I already think Starship storage tankers (depots) are going to end up being used at High & Low, Earth & Mars Orbits.

On the solar PV efficiency question, it might make sense to ship the 'raw feed-stock' as far down the logistics pipe as you can, before converting it to usable propellant.  Solar PV is more efficient in HMO than on the surface, and HEO more so than HMO.  Also liquid CO2 and H2O should be easier to store than LCH4 & LOX, let alone LH2.
Why wait to get close before starting feed stock conversion? It not like the PV is metered.


Every indication is that while transiting, freezing will be more of an issue than boiloff so storage shouldn't be a problem. Keeping the water liquid might be the hardest chore. Hmmm. Maybe this is a reason to not bother with conversion until further in.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/24/2020 09:48 pm
About 30% using SEP and perhaps a Deimos tether (compare to about 3% for Earth's surface to LEO). But mass of propellant is less relevant than efficiency in this case. You could launch the propellant to LMO using CO/O2 rockets on Mars. The lower Isp is compensated nicely by the reduced energy cost per kg of propellant. Atmospheric ISRU on Mars may actually be easier than on Earth as direct air capture of CO2 is not cheap on Earth (but is easy on Mars).

Just the CH4 and O2 (or whatever propellant) would be needed to be delivered to LEO.

Also, you'd send it in large tanker variants of Starship fairly optimized for the trip. Not just regular versions of Starship. SpaceX already mentioned both tanker and depot variants of Starship. Most importantly: You wouldn't send the ship directly from mars to Earth but instead refuel a Starship tanker right at high Mars orbit, right before Earth injection burn. Dry mass-to-wet-mass would be reduced this way to around 0.1 instead of your 0.6 or so. (This makes an enormous difference.)

Secondly, you'd likely only put a small portion of the propellant in LEO. It makes sense to also refuel at HEO on the way to Mars, and if your propellant is coming from Mars anyway, might as well keep much of it in HEO to begin with as it gives you a greater benefit on the departure burn than the same amount of propellant in LEO. EDIT: Splitting the departure propellant between LEO and HEO like this almost halves the total amount of propellant needed for the same delta-v (say, 7.5km/s total starting in LEO), and the effect is significantly larger for faster transits than slower transit. I did NOT include that consideration below... If I did, the energy advantage for Mars over Earth would be about 15-20:1 instead of 10:1.

So again, I think you'd have roughly a 10:1 energy advantage (knocked down to, say, 4:1 by the lower sunlight intensity at Mars... which could be more-than-compensated-for, long-term, using areosynchronous orbital mirrors once you get to Terawatts of power... something you really shouldn't be doing at Earth... although this is all irrelevant if you use nuclear power which should be similar for Earth and Mars... tho the details get hairy).

Ok, for something quite a bit more technically complex, (than just refueling a Starship on Mars then launching it), this makes a bit more sense. 

I hadn't factored in the 'staging' like advantages of depots in High Mars/Earth Orbit at each end which do reduce the delta v of each leg, so increase the overall efficiency.  I already think Starship storage tankers (depots) are going to end up being used at High & Low, Earth & Mars Orbits.

On the solar PV efficiency question, it might make sense to ship the 'raw feed-stock' as far down the logistics pipe as you can, before converting it to usable propellant.  Solar PV is more efficient in HMO than on the surface, and HEO more so than HMO.  Also liquid CO2 and H2O should be easier to store than LCH4 & LOX, let alone LH2.
I had never considered shipping water and CO2 from Mars into Mars orbit. It's a really neat idea that does help solve some problems, although I'm not exactly sure how practical it would be. It would need two dedicated tanks in Mars orbit one for water and CO2 plus one for methane and oxygen. There might also be issues with the water freezing. I suppose that could be solved by an electrically powered heater, but I'm not qualified to comment on the practicalities or trades of that arrangement.

One thing that does appeal is the thought of putting half the solar power in orbit where it could receive nearly double the energy from the sun and avoid dust storms and still have the other half on the surface for base operations. Unfortunately there would still need to be ISRU producing methalox on the surface in order to re-tank the ships heading up to the orbital facility. So its doable but I suspect the issues will prevent it's use until a lot further in the future when they're optimising the logistics chain.
I've been talking about sending raw feedstock back to earth and converting to props along the way but doing it LMO is attractive. It gives a natural growth path.


Sending props ISRU a synod before crew and depending on it to work is a high risk plan. A well discussed alternative is sending a depot from earth to LMO with return propellant and landing enough to get the crewed ships up to it. This takes the pressure off and gives a good opportunity for the crew to get ISRU working without depending on it.


Setting up an orbital propellant factory with feedstock supplied from Mars creates only an incremental logistics change, not a forklift. Coupled with lower capacity ISRU on the surface and at least one more synod of return propellant from earth creates a low risk operation with the terrible risk of, OMG, too much propellant on hand. What a problem!


Orbital ISRU seems to add another yet another layer of safety. If dust storms disrupt surface ISRU, a small reserve of propellant will enable an empty cargo ship to make orbit, tank up and return - as many times as necessary.


From there optimization for an HMO depot is yet again only an incremental change. Keep the systems overlapped until working smooth and the cost is having too much propellant. Follow this pattern and except for those dismal delivery numbers, Mars could deliver propellant to earth.


Which segues into another idea that has been beaten to death. Asteroid ISRU.


Let's not go too far back into the weeds but with a couple of depots in LMO and HMO, there is existing infrastructure if the asteroids pan out. /end weeds.


Freezing water. This problem is there whether making props in orbit or on the surface, so it pretty much cancels out. There will be handling issues in 0g but that shouldn't be a showstopper.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/24/2020 10:46 pm
Just for fun, I ran the numbers for percentage of Mars produced propellant reaching High Earth Orbit, using just standard methalox Starships as tankers, and depots in LMO and HMO:

I get 13% using propulsive capture into HEO, and 15% using aero-capture.

Eliminating legs, fins and heat-shield for the LMO-HMO & HMO-HEO legs makes negligible difference. (Except on the return leg).

Switching to hydrolox for LMO-HMO & HMO-HEO gives 15% (assuming same MR, which would be worse in reality, due to larger tanks)

Switching to hydrolox for every leg gives 19% (assuming same MR, which would be a lot worse in reality, due to EDL for larger tanks)


Everything else being equal, which it isn't, fully Hydrolox makes 50% more propellant available in HEO, which because you need less, enables 75% more payload than methalox.  More like 40% and 70% with propellant density factored in, and who know with ISRU production & storage factored in.


Using methalox for Mars to LMO, then SEP, I got 21% to 24% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then SEP, I got 14% to 15% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then hydrolox, I got 10%.

Using CO2/LOX, for Mars to LMO, then methalox , I got 8%. Might as well just send a Starship direct from the surface.


NOTE: All these trips are one-way. When you need to send an empty tanker back to HMO or LMO, it requires about 16% of the propellant just delivered for methalox, and 21% for hydrolox (lower density -> higher empty mass). Around 12% for SEP.

Edit: fix return %
This is really dismal. Especially since not going back for another load doesn't make sense.


What do the numbers look like comparing mars launch vs earth launch to deliver CO2 for CH4 production from lunar water? I thought about just delivering carbon but the handling difficulties just kept getting worse the more I thought about it.


With all the possible destinations in cislunar there are a lot of moving parts. If the numbers work Mars and Luna could conceivably become the go to for all cislunar refueling. If the numbers work.


To help the numbers it might work best as pure SEP and never touching the ground. Tankers bring CO2 up from Mars using whatever propellant works best. No raptors or legs. Fins and heatshield only if aerobraking. Return can be a low energy transfer to martian orbit. Slow, but it builds a pipeline over time and takes the load off of earth.


Not an overnight project.





Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 08/24/2020 11:37 pm
Starship needs to be at least several days in orbit for refueling and has no protection against any debris larger than 1 mm. Potentially high risk of mission failure.

Altitude is a big factor.  ISS orbits at ~ 409 km.  Starship propellant loading could occur at ~ 250 km.  Debris flux is roughly an order of magnitude lower at the lower altitude (red vs. green line).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: armchairfan on 08/25/2020 03:32 am
Do the math on the 100 Megatons-per-year tweet that Musk mentioned earlier.
If/when this becomes a thing, IMO we should seriously consider momentum exchange tethers at least for surface/LEO traffic and possibly interplanetary traffic too. The tether orbits would be carefully selected and maintained (mostly by nodal precession) to align with the synodic arrival and departure orbital planes. Just for scale, a 100km tether at 3 gees with starships attached tail-to-tail gives each ship a 2*sqrt(3*9.8*100000/2))/1000 = 2.4 km/s DV "for free". There are nuances but they're better discussed in a tethers thread.

Payload goes way up and propellant requirements go way down. Win/win!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 08/25/2020 03:47 am
To me it is simple.  Launch several tankers in orbit.  Refuel the tankers.  Then when the Mars synod comes up, launch as many Starships as you have tankers in orbit.  Refuel Starships with one docking then go on to Mars.  Tanker/depots, can carry refrigeration, solar panels and shielding to avoid boil-offs while being filled for the Mars Starship synod.  Tanker/depots can either stay in orbit, or return after a Mars fueling run to be checked out and refurbished if needed, then returned to orbit be refilled.  No need for any other developments.  Starships are versatile and modular and can be used for cargo, tanker, passenger/crews, fuel depot, or any combination.  Just outfit them for whatever you need. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 08/25/2020 04:02 am
Here's a different idea I don;t think I've seen here yet ... if we get to the scale of Starship operations envisioned, we're going to need more launch complexes.  Lots more.  And they're probably going to need multiple pads.  Lots and lots of pads.  So build a complex with a dozen pads.  When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.


I've made no effort so far to qualify this idea to graduate beyond the Crazy Ideas Division.  I'm not sure where to start.  But I'm sure someone here can point out all the reasons this is a dreadfully, catastrophically bad idea.   Or not.  I eagerly await well reasoned feedback!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/25/2020 04:47 am
Here's a different idea I don;t think I've seen here yet ... if we get to the scale of Starship operations envisioned, we're going to need more launch complexes.  Lots more.  And they're probably going to need multiple pads.  Lots and lots of pads.  So build a complex with a dozen pads.  When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.


I've made no effort so far to qualify this idea to graduate beyond the Crazy Ideas Division.  I'm not sure where to start.  But I'm sure someone here can point out all the reasons this is a dreadfully, catastrophically bad idea.   Or not.  I eagerly await well reasoned feedback!
To my eye the problem is massive over capacity of launch facilities and tankers. For a mars fleet of 10 ships, figure five tankers loading into the sixth for each Mars bound ship. That's 60 tankers, 60 SH and 60 pads that sit idle for most of a synod. There will be other refueling but this would be the big nut.


Operationally it's pretty intense. A large surge of personnel with little else to do when they're not launching. It wouldn't be 60 times what's needed for one launch but it would be up there. And think about that poor sucker that has to run from pad to pad lighting fuses. He's an interesting story. Got his start pulling the MGM lions tail.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/25/2020 05:01 am
When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.

Propellant boil-off isn't that bad. We're not using hydrogen. You can spread the fuelling of the accumulator-tanker out over a few months. At worst, you add one extra tanker mission a few days before the main Mars-SS launches.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/25/2020 05:02 am
This is why I wonder if, long-term, most of the propellant for trips to Mars might actually come *from* Mars.
It isn't just delta-v. Producing enough propellant to launch enough propellant as cargo to Earth, to fully fuel its own return plus one SS flight puts a burden on the ISRU production that puts disproportionate burdens on a Mars settlement that it doesn't on a few extra Earth launches.
Thus, the suggestion for shipping CO2 & water and processing into propellant on the way back.

You misunderstand, the "burden" isn't just producing the propellant that arrives in Earth orbit. It's the extra propellant needed to push that to Earth. Say SS has a return cargo mass of 50 tonnes and you need 600 tonnes for each SS you fuel in Earth orbit. (I haven't looked up the proper estimated numbers; for the purpose of this, it doesn't matter.) You need 12 returning ships for every one you can refuel in Earth orbit.

Alternatively, for each returning ship to supply enough fuel for itself to return, you'll have to launch the return ships to Mars orbit, then launch (via a tanker fleet on Mars) enough propellant to allow them to carry 600 tonnes of surplus propellant back to Earth. So even if you are only launching CO2+Water as cargo, you still have to produce a) the extra propellant needed to get those ships back to Earth with 600 tonnes of CO2+Water, and b) the extra propellant needed to launch the tankers to bring up that propellant into Mars orbit.

It's not just the cost of the cargo-propellant that ends up in Earth orbit. It's the propellant needed to fuel the ships that launch the propellant needed to fuel the ships that return the cargo-propellant to Earth. Earth surface to Earth orbit is usually a 5-7 times multiplier. (Ie, it takes 5-7kg of fuel to launch 1kg of payload.) I haven't worked out the ratio for Mars-surface to Earth-orbit, but it ain't zero.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/25/2020 05:06 am
Microgravity can be used to assist the flow from one to the other.
For ullage, "milli" not "micro".
Any G will settle props.

No it wont. Microgravity isn't enough to cause fluids to settle. Microgravity is what the ISS aims for in order to achieve high-quality science. Micro-g is functionally zero-g for the purposes of fluid flow.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/25/2020 05:12 am
Microgravity can be used to assist the flow from one to the other.
For ullage, "milli" not "micro".
Any G will settle props.

No it wont. Microgravity isn't enough to cause fluids to settle. Microgravity is what the ISS aims for in order to achieve high-quality science. Micro-g is functionally zero-g for the purposes of fluid flow.
Not true.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Star-Dust on 08/25/2020 08:31 am
1- I have calculated (online calculator) a 200 tons  payload to LEO with a second stage on top of SH

I bet you haven't, because...

I dont think that a stainless steel tank coupled with six raptors engines will weigh that much

...others have mentioned that the figures that estimate Starship's mass also estimate 40 tonnes for re-entry and landing. 25 tonnes for structure/heat-shield, and 15 tonnes for fuel.

Hence, if you're using a calculator that gives 100 tonnes payload (hence propellant) for the Starship tanker, then it will give 140 tonnes payload (prop) for the expendable version. If you got a higher figure, then you dropped the drymass to an unrealistic level.

3-The option of depot will mitigate the need for rapid  launch of tankers, SS bound for Mars will only be launched when all tanks are in place, this allowing time for further developement of rapid reuse of SH/SH while not delaying first mission to Mars (interim solution).

As mentioned repeatedly: Everything you require for your expendable tanks are the same as the tanker. You need to launch on SH, you need to be able to do docking manoeuvres in space, you need to connect up propellant lines (and gas-return lines for tank pressurisation), you need to be able to transfer propellant from the tanks to the engines on the main vehicle.

In addition, because you seem to have (finally!) accepted that you can't launch a single expendable fully fuelled tank into orbit, you now invent another whole structure, the docking node that holds a bunch of tanks as a depot. The depot then has to transfer fuel to the main Mars vehicle... kind of like... you know... a tanker.

You have all the requirements of building tankers, plus you need a depot before it's usable. It's not an "interim" solution.

The only technology you are trying to skip is reusability, which only requires re-entry and landing. But if Starship can land on Mars, then the technology for reusable tankers has already been developed. If the technology to land tankers isn't ready, Starship can't go to Mars.

So again, there's no "interim" where the Mars Starship exists but the tankers don't. Either the landing technology is ready, making tankers viable, or you can't land on Mars.

5- Fuel depot could be sent to Mars orbit.

Having fuel in Mars orbit doesn't replace ISRU production for a manned Starship mission: Starship can't land while fully fuelled, hence can't launch back into Mars orbit without ISRU propellant on the ground. If they have that, then your proposal isn't needed. If they don't, then you can't send humans to Mars. Again, there's no "interim".
I can't speak for Stardust but I've considered a stripped SS without tank mods as a short term solution and an assembly capable of refueling an entire mars fleet as a long term solution. I see everything else as an interim solution. That said, your criticisms of Stardust's idea are on point.

Don't be shy to swim against the tide, some time you should express what you think. ;D
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/25/2020 08:41 am
When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.

Propellant boil-off isn't that bad. We're not using hydrogen. You can spread the fuelling of the accumulator-tanker out over a few months. At worst, you add one extra tanker mission a few days before the main Mars-SS launches.
I would have thought that some form of large inflatable highly reflective screen could be used if there were problems. Does anyone have a handle on the scale of the boil off problem such as it is?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 08/25/2020 09:15 am
Just for fun, I ran the numbers for percentage of Mars produced propellant reaching High Earth Orbit, using just standard methalox Starships as tankers, and depots in LMO and HMO:

I get 13% using propulsive capture into HEO, and 15% using aero-capture.

Eliminating legs, fins and heat-shield for the LMO-HMO & HMO-HEO legs makes negligible difference. (Except on the return leg).

Switching to hydrolox for LMO-HMO & HMO-HEO gives 15% (assuming same MR, which would be worse in reality, due to larger tanks)

Switching to hydrolox for every leg gives 19% (assuming same MR, which would be a lot worse in reality, due to EDL for larger tanks)


Everything else being equal, which it isn't, fully Hydrolox makes 50% more propellant available in HEO, which because you need less, enables 75% more payload than methalox.  More like 40% and 70% with propellant density factored in, and who know with ISRU production & storage factored in.


Using methalox for Mars to LMO, then SEP, I got 21% to 24% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then SEP, I got 14% to 15% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then hydrolox, I got 10%.

Using CO2/LOX, for Mars to LMO, then methalox , I got 8%. Might as well just send a Starship direct from the surface.


NOTE: All these trips are one-way. When you need to send an empty tanker back to HMO or LMO, it requires about 16% of the propellant just delivered for methalox, and 21% for hydrolox (lower density -> higher empty mass). Around 12% for SEP.

Edit: fix return %
This is really dismal. Especially since not going back for another load doesn't make sense.

What do the numbers look like comparing mars launch vs earth launch to deliver CO2 for CH4 production from lunar water? I thought about just delivering carbon but the handling difficulties just kept getting worse the more I thought about it.

With all the possible destinations in cislunar there are a lot of moving parts. If the numbers work Mars and Luna could conceivably become the go to for all cislunar refueling. If the numbers work.

To help the numbers it might work best as pure SEP and never touching the ground. Tankers bring CO2 up from Mars using whatever propellant works best. No raptors or legs. Fins and heatshield only if aerobraking. Return can be a low energy transfer to martian orbit. Slow, but it builds a pipeline over time and takes the load off of earth.

Not an overnight project.


The good news is that the slow, synod limited part of the tanker run, HMO to HEO, is fairly efficient both ways, so one full SS tanker, probably delivers enough methalox to HEO for 2 x outbound cargo SS, or 1 x fast transit crew SS, plus it's own return.  Although I guess it has to do it the previous synod. But that one trip needs 8 x Surface to LMO flights, and a couple of LMO to HMO, before the synod Mars departure window.*

I haven't done the numbers for water from the Moon, but given 80% by mass of SS propellant is LOX, it seems to me the best options is probably CH4 made on Mars or in Mars Orbit, shipped to Earth-Moon L2. Then make LOX on the Moon, ship it to L2 and return CH4.

Unless there is carbon freely available on the Moon.  If there's lots of CO or CO2 in the polar ice-traps, that might be better.


* These numbers are approximate, the exact numbers depend on so many assumptions and quanta of tanks capacities.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/25/2020 01:12 pm
Microgravity can be used to assist the flow from one to the other.
For ullage, "milli" not "micro".
Any G will settle props.

No it wont. Microgravity isn't enough to cause fluids to settle. Microgravity is what the ISS aims for in order to achieve high-quality science. Micro-g is functionally zero-g for the purposes of fluid flow.
If the tank contains an amorphous glob of frothy propellant and the tank starts accelerating around it at 1mm/hour/hour (example only) in a consistent direction eventually all the propellant will be settled at one end.


The ISS would prefer true 0g but with orientation changing 360deg every orbit, the PV changing orientation constantly and people moving around it's unachievable. The max allowable is probably buried down in the specs somewhere but microgravity, like micro brewery, is also a generic term.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 08/25/2020 01:34 pm
Here's a different idea I don;t think I've seen here yet ... if we get to the scale of Starship operations envisioned, we're going to need more launch complexes.  Lots more.  And they're probably going to need multiple pads.  Lots and lots of pads.  So build a complex with a dozen pads.  When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.


I've made no effort so far to qualify this idea to graduate beyond the Crazy Ideas Division.  I'm not sure where to start.  But I'm sure someone here can point out all the reasons this is a dreadfully, catastrophically bad idea.   Or not.  I eagerly await well reasoned feedback!
To my eye the problem is massive over capacity of launch facilities and tankers. For a mars fleet of 10 ships, figure five tankers loading into the sixth for each Mars bound ship. That's 60 tankers, 60 SH and 60 pads that sit idle for most of a synod. There will be other refueling but this would be the big nut.


Operationally it's pretty intense. A large surge of personnel with little else to do when they're not launching. It wouldn't be 60 times what's needed for one launch but it would be up there. And think about that poor sucker that has to run from pad to pad lighting fuses. He's an interesting story. Got his start pulling the MGM lions tail.
Obviously my lunatic idea is just that in the short to medium term, probably more trouble than its worth to launch a few, or even a couple dozen ships now and then.  Though it wouldn't have to be just for Mars.  It could also serve lunar missions.  Heck, even placing LARGE payloads direct to GEO might be able to use a complex like this.  As for Mars, if you just wanted to launch a few starships every synod, this might be overkill, but I'm thinking a bit further out ... Musk wast to send these things by the thousand, eventually.  Then again, by the time we're doing that, we may have other means at our disposal, and the whole discussion will be moot.

One other consideration.  I'm not thinking about mega-launch complexes like a giant KSC with dozens of 39As.  More like a launch complex is the style of Elon.  Maximum automation, "the best part is no part", etc.  Just what's needed for the specified mission and no more.  To slightly modify a recent thread title from this section "What is SS/SH launch complexes are unreasonably cheap?".
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bob Shaw on 08/25/2020 01:38 pm
I wonder whether a cluster of 'parachutes' might work to both create a slight accelaration and to cool the Starship?

1. Set the Starship rotating once per orbit, pointing towards the Sun.

2. From the nose, eject one pilot parachute with subliming material inside strut balloons to form the shape, preferably going into nightside to avoid Solar thrust.

3. On the next orbit, eject six more as it passess into darkness to make the full umbrella.

4. The parachutes always point at the Sun, and suffer drag. The struts keep them stable. Over several orbits the Starship is protected from direct insolation and the fuel gradually sings to the bottom of the tank.

5. Some tanker LO is vented during docking and while the transfer continues.

Someone else can do the numbers - I'm trying to think of an almost passive way of doing all this!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bob Shaw on 08/25/2020 01:39 pm
I would have thought that some form of large inflatable highly reflective screen could be used if there were problems. Does anyone have a handle on the scale of the boil off problem such as it is?

See my post above for cooling and ullage!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/25/2020 01:45 pm
When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.

Propellant boil-off isn't that bad. We're not using hydrogen. You can spread the fuelling of the accumulator-tanker out over a few months. At worst, you add one extra tanker mission a few days before the main Mars-SS launches.
To amplify on this a bit, The first mission standard build SS accumulators can follow Paul's last minute top off, maybe with some sunshade gizmos. The cargo ships can launch early and make a slower transit which would also ease the load on a tanker surge strategy.


A tank stretch depot with full shades and cryo cooler can be introduced one off at any time and mix n match with standard build accumulator SS's. If it works out, build more.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/25/2020 03:04 pm
When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.

Propellant boil-off isn't that bad. We're not using hydrogen. You can spread the fuelling of the accumulator-tanker out over a few months. At worst, you add one extra tanker mission a few days before the main Mars-SS launches.
I would have thought that some form of large inflatable highly reflective screen could be used if there were problems. Does anyone have a handle on the scale of the boil off problem such as it is?
In LEO the big thermal input is the sun. Keeping a ship oriented with one end pointed sunward minimizes the input and a shade can be not much wider than the ships diameter. Because we're talking about SS refueling, shading the nose keeps the aft end clear for operations.


Reflection from earth is a large secondary input. I'd WAG up to 25% of solar input at some points.   Happily, the point of greatest input, local high noon, has the ship very roughly oriented tail earthward. Not dead on, but in that direction. Shading this is more complicated than shading from the sun because of constantly changing orientation.


I wish I could put numbers on all of this and can't, so instead I offer arm waving and generalizations. One number I did find is the solar thermal input at earth is 1370W/m^2.


A standard SS sans shields could work as an accumulator if time between first tanker load and refueling is short. Maybe on the order of a 2-3 weeks. Adding minimum shielding extends this and full shielding yet more, but in the end it all will boil off.


That's why, IMO, anything short of good shading and a PV based Zero Boil Off (ZBO) system can only be a stop gap. I'm writing this dodging into shade as the morning sun sweeps across my deck to remind me of the magnitude of the problem.


Orbital refueling is great leverage for space operations and one element of making operations inexpensive is making them routine. Using standard build SS's as accumulators will probably work but the tanker surge needed to make it work, while itself may become routine, will be a special operation that by definition is not routine. Ergo, full fledged depots are necessary.


Now my curiosity is up and I'll probably spend some time pinning down some numbers and agonizing over some assumptions to guesstimate thermal input, thermal rejection of sunshields and PV requirements for a cryo cooler. Mass should not be an issue if I work with a 6 ring stretch to a stripped SS. If launched empty it would have plenty of mass margin to LEO.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: envy887 on 08/25/2020 03:27 pm
That's why, IMO, anything short of good shading and a PV based Zero Boil Off (ZBO) system can only be a stop gap. I'm writing this dodging into shade as the morning sun sweeps across my deck to remind me of the magnitude of the problem.

The "and" there isn't strictly required. Either good shading or a good active ZBO system will work without the other. The best solution is probably some combination of both, although I think it will lean heavily towards the shading.

In Earth orbit, if they point the nose constantly at the Sun, then the Earth will circle the vehicle once per orbit. In a sun-synchronous dawn-dusk polar orbit, the vehicle can shade one side, and roll once per orbit so that the nose is always pointed at the Sun and the shade is always pointed at Earth. That way the shade (or shades, if the nose is shaded as well), are both fixed relative to the vehicle.

For other more accessible orbits this also works, but the Earth shade needs to be able to rotate or slide fore-aft as the vehicle rolls once per orbit to stay pointed at the Earth. Or the shade could be fixed, but then it needs to fully cover one side of the vehicle and will occasionally be blocking the view of cold deep space.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/25/2020 03:34 pm
I wonder whether a cluster of 'parachutes' might work to both create a slight accelaration and to cool the Starship?

1. Set the Starship rotating once per orbit, pointing towards the Sun.

2. From the nose, eject one pilot parachute with subliming material inside strut balloons to form the shape, preferably going into nightside to avoid Solar thrust.

3. On the next orbit, eject six more as it passess into darkness to make the full umbrella.

4. The parachutes always point at the Sun, and suffer drag. The struts keep them stable. Over several orbits the Starship is protected from direct insolation and the fuel gradually sings to the bottom of the tank.

5. Some tanker LO is vented during docking and while the transfer continues.

Someone else can do the numbers - I'm trying to think of an almost passive way of doing all this!
The thrust would truly me micro (or less) but the there are downsides.


The slight impacts of another ship hooking up would overwhelm the chute G and just when the props need to be settled, they're not. Waiting for this system to resettle props will  be a long wait.


Second problem is that a depot receiving propellant and providing props need to settle in opposite directions. This can be overcome with plumbing but still an issue.


The consumables needed for settling is not great and arriving tankers will be in position to replenish the depot.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/25/2020 04:11 pm
That's why, IMO, anything short of good shading and a PV based Zero Boil Off (ZBO) system can only be a stop gap. I'm writing this dodging into shade as the morning sun sweeps across my deck to remind me of the magnitude of the problem.

The "and" there isn't strictly required. Either good shading or a good active ZBO system will work without the other. The best solution is probably some combination of both, although I think it will lean heavily towards the shading.

In Earth orbit, if they point the nose constantly at the Sun, then the Earth will circle the vehicle once per orbit. In a sun-synchronous dawn-dusk polar orbit, the vehicle can shade one side, and roll once per orbit so that the nose is always pointed at the Sun and the shade is always pointed at Earth. That way the shade (or shades, if the nose is shaded as well), are both fixed relative to the vehicle.

For other more accessible orbits this also works, but the Earth shade needs to be able to rotate or slide fore-aft as the vehicle rolls once per orbit to stay pointed at the Earth. Or the shade could be fixed, but then it needs to fully cover one side of the vehicle and will occasionally be blocking the view of cold deep space.
Yes, a sun synchronous orbit reduces shade dynamics but at the cost of an orbit with reduced utility. The gyrations an earth shield will need are why I mentioned it being a more complex problem.


AIUI, even a highly reflective shield still heats up and this thermal load is re emitted from both sides so half goes towards the ship. A second layer will mostly reflect this back to the first shield where it continues the reflection emission process.


Upshot is that more than one layer doesn't add much. But, since the deployment and tracking gear is the bulk of the mass and complexity, more layers doesn't add much to the cost.


The advantage of a cryo cooler is heavily dependent on reliability. Once it breaks down it's dead mass. A quick google yesterday showed me that since it's never been done in space, everything is theoretical. Fill and forget is a great ideal but with little engineering legacy, only an ideal.


You're probably right about depending on shades. Any cryo cooler flown will be an experiment.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/25/2020 04:15 pm
It's been pointed out that HEO is more mass effect than LEO for mars mission refueling. It also has the advantage of less earth thermal input.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: mmeijeri on 08/25/2020 05:47 pm
The advantage of a cryo cooler is heavily dependent on reliability. Once it breaks down it's dead mass. A quick google yesterday showed me that since it's never been done in space, everything is theoretical. Fill and forget is a great ideal but with little engineering legacy, only an ideal.

Cryocoolers have flown in space before (on telescopes?), and IIRC an experiment was done with one on the ISS. It ended after a few months when the cryocooler failed and the methane it was supposed to cool had to be vented. Not proven technology, but not entirely unproven either.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/25/2020 06:21 pm
The advantage of a cryo cooler is heavily dependent on reliability. Once it breaks down it's dead mass. A quick google yesterday showed me that since it's never been done in space, everything is theoretical. Fill and forget is a great ideal but with little engineering legacy, only an ideal.

Cryocoolers have flown in space before (on telescopes?), and IIRC an experiment was done with one on the ISS. It ended after a few months when the cryocooler failed and the methane it was supposed to cool had to be vented. Not proven technology, but not entirely unproven either.
I could be batting 0 on this but I think telescopes use cryo fluids to chill the sensor. The James Webb will use a cryo cooler.


Breaking down during an experiment on ISS doesn't have me whaling that cryo cooling on orbit can't be done. I believe it can be. But it does tell me that more work is needed to make it bullet proof.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: mmeijeri on 08/25/2020 06:24 pm
Breaking down during an experiment on ISS doesn't have me whaling that cryo cooling on orbit can't be done. I believe it can be. But it does tell me that more work is needed to make it bullet proof.

Absolutely, and fortunately it looks as if SpaceX's part in Artemis will help with that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/25/2020 09:00 pm
Here's a different idea I don;t think I've seen here yet ... if we get to the scale of Starship operations envisioned, we're going to need more launch complexes.  Lots more.  And they're probably going to need multiple pads.  Lots and lots of pads.  So build a complex with a dozen pads.  When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.


I've made no effort so far to qualify this idea to graduate beyond the Crazy Ideas Division.  I'm not sure where to start.  But I'm sure someone here can point out all the reasons this is a dreadfully, catastrophically bad idea.   Or not.  I eagerly await well reasoned feedback!
To my eye the problem is massive over capacity of launch facilities and tankers. For a mars fleet of 10 ships, figure five tankers loading into the sixth for each Mars bound ship. That's 60 tankers, 60 SH and 60 pads that sit idle for most of a synod. There will be other refueling but this would be the big nut.

Except Elon isn't going for a fleet of 10 ships. He's going for a fleet of hundreds to thousands of ships.

https://twitter.com/elonmusk/status/1217990326867988480

Operationally it's pretty intense. A large surge of personnel with little else to do when they're not launching. It wouldn't be 60 times what's needed for one launch but it would be up there. And think about that poor sucker that has to run from pad to pad lighting fuses. He's an interesting story. Got his start pulling the MGM lions tail.

I don't think there will be as many personnel as you imagine. Quoting from a recent SpaceX job posting (https://www.jobilize.com/job/launch-engineer-control-flight-operations-starship-brownsville-spacex):

Quote
LAUNCH ENGINEER - LAUNCH CONTROL AND FLIGHT OPERATIONS (STARSHIP) RESPONSIBILITIES:
• Develop and execute launch, flight, on-orbit, and recovery operations streamlining future Starship launches
Drive development towards power on to launch in less than one hour with zero operators on console
• Help design automation back-end software (ControlCode, Stage0) to optimize how operators write and interact with automation
• Develop software and processes that automate how SpaceX tests and validates ground and vehicle software
• Support Falcon launch operations as needed

Obviously they won't climb that automation mountain in one day, but by the time they're launching hundreds of ships in one synod it should be a pretty well-oiled (and lightly staffed) machine.


When you're ready to go somewhere, launch a salvo of all the tankers you're going to need more or less at the same time.  On orbit, the tankers do the tanker dance and transfer all the prop (except for landing fuel) to one tanker.  24ish hours later (or whenever the launch complex aligns properly with the staging orbit). The empty tankers come home, and the outbound Starship launches, rendezvouses with the full tanker, transfers a full load of prop, and executes its departure burn as soon as it is convenient to do so.

Propellant boil-off isn't that bad. We're not using hydrogen. You can spread the fuelling of the accumulator-tanker out over a few months. At worst, you add one extra tanker mission a few days before the main Mars-SS launches.

The advantages aren't just propellant boil-off. It also reduces time spent rendevous-and-docking (increasing tanker utilization efficiency), it lets you combine airspace closure (NOTAM) windows, it lets you combine launch area personnel evacs and closures (streamlining operations vs. having a bunch of pads running on independent schedules), and it reduces closures of beaches and waterways. I would argue it also reduces the per-launch acoustic disruption to the local community, because by clustering the schedules it gives longer "quiet periods" between launch events.

Some may object that these costs aren't paid directly by SpaceX out-of-pocket, but they still incur political costs onto SpaceX. If you can't scale Starship (again, to thousands of launches per day) without completely disrupting modern air travel, then you can't scale Starship.

It also increases the density of launch pads per launch site. On the urbanized Earth, good candidate locations (and jurisdictions) for launch sites are relatively scarce, even after you add in the feasible off-shore launch sites. If you can fit 10x as many launch pads per launch site then that increases your maximum scale by 10x, while at the same time lowering your per-launch costs. Obviously the per-pad costs (eg fuel) don't go down, but the per-site costs (eg closures, security) do go down.

And your tanker use tempo increases rather dramatically.

And yeah, less boil-off.


I think cdebuhr's Crazy Idea is pretty brilliant, all things considered.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/25/2020 09:37 pm
If the tank contains an amorphous glob of frothy propellant and the tank starts accelerating around it at 1mm/hour/hour (example only) in a consistent direction eventually all the propellant will be settled at one end.

It won't overcome the "stickiness" of the froth, won't separate gas and fluid. Surface wetting and capillary effects dominate.

but microgravity, like micro brewery, is also a generic term.

As a technical term, microgravity means "millionths of a g". As a generic term, microgravity is a synonym for near-zero-g or orbital free-fall. Microgravity does not mean "anything lower than Earth gravity". It's weird to keep encountering the latter on a technical site. It's weirder that this is the only site where I've seen people do it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/25/2020 09:53 pm
Re: launching in waves

It also reduces time spent rendevous-and-docking (increasing tanker utilization efficiency),

How?

If there's five ships, each spends as long doing rendezvous and docking. Launching them together doesn't make them dock quicker. If anything, clustering means they're more likely to get in each others way. (Eg, a tanker that's taking longer to dock than expected delays the others.)

it lets you combine airspace closure (NOTAM) windows, it lets you combine launch area personnel evacs and closures (streamlining operations vs. having a bunch of pads running on independent schedules), and it reduces closures of beaches and waterways.

OTOH, it reduces site efficiency, since it spends most of its time unused. It reduces manpower efficiency, since you need more ops people to handle the peak, who are unneeded during the lulls. It seems to go against the way SpaceX is costing their systems.

I would argue it also reduces the per-launch acoustic disruption to the local community, because by clustering the schedules it gives longer "quiet periods" between launch events.

OTOH, it puts vehicles in each others acoustic environments during critical pre-launch periods.

It also increases the density of launch pads per launch site.

Even denser is to use fewer pads more often. The best part is no part.

And your tanker use tempo increases rather dramatically.

Having fewer tankers used more often is a more efficient than many tankers used rarely.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 08/25/2020 10:35 pm
If you place a tanker in orbit to be a fuel depot.  It will take 5 tanker flights to fill it.  One tanker per week in 5 weeks you have a full tanker in orbit ready to fuel up a Mars bound Starship.  In 50 weeks you could have 10 full tankers for 10 Mars bound Starships.  18 months between synods and you can have 15 Starships going to Mars each synod.  You only have to produce 15 in orbit fuel depot tankers and 15 Mars bound Starships per synod along with say two reusable tankers to fill the depot tankers.  This is a total of 32 Starships.  All this could be done from one launch pad with one flight per week plus a little at the end for the Starships leaving earth. 

Musk said he will eventually be able to produce 100 Starships per year.  This would trippld the number of Starships going to Mars to 45. 

Now if you get 2-3 flights per week, just from one launchpad, a flotilla of at least 100 Starships could go to Mars every synod.  2-3 flights per week is not impossible from one pad.  Now they are going to build a Florida pad also.  So that gets two pads.  Musk may build another pad offshore of Boca Chica and offshore Florida also and still use the same onshore facilities.  That gives about 4 pads. 

There is no reason why a launchpad can't handle one launch per day if need be.  One or two a week from 3-4 pads can fuel in orbit tankers, and fly 100's of Starships per synod to Mars. 

This is just SpaceX, now if others want to join the Mars colonization, there will be others possibly filling the tanker Starship/Depots, to get a ride.  Blue Origin may build New Armstrong by then. 

I believe Starship will be the Model T or the Volkswagon of the space age.  Low cost to orbit.  Reusable, and lower maintenance.  Others will have to follow.  Then larger Starships. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/25/2020 10:55 pm
Re: launching in waves

It also reduces time spent rendevous-and-docking (increasing tanker utilization efficiency),

How?

Because you eliminate phasing, and you eliminate the need to wait 12-24 hours for the launch site to rotate under the orbital plane. Every launch opportunity into the Mars departure parking orbit (1-2 per day) you can launch 10 tankers instead of 1.

If there's five ships, each spends as long doing rendezvous and docking.

Five ships seems a bit too small. You can't get a full tanker, for one.

Eight tankers would make good sense, because then you only need three rounds of rendezvous (comprising 7 rendezvous total) and you get a full tanker at the end (rather than ~half a tanker).

Launching them together doesn't make them dock quicker.

You're not waiting as long for phasing, and you don't have to wait for the launch site to line up again. That makes the rendezvous quicker. Naturally the propellant transfer takes the same amount of time, so that's no different either way.

If anything, clustering means they're more likely to get in each others way. (Eg, a tanker that's taking longer to dock than expected delays the others.)

That sounds like a "it hurts when I do this" problem. Solution: don't launch select parking orbits that are too close together. Work the problem like Goldilocks and choose a distance that's "just right:" not right on top of each-other, but it's not necessary to circle the Earth 16 times until your launch site is aligned either.

it lets you combine airspace closure (NOTAM) windows, it lets you combine launch area personnel evacs and closures (streamlining operations vs. having a bunch of pads running on independent schedules), and it reduces closures of beaches and waterways.

OTOH, it reduces site efficiency, since it spends most of its time unused.

You're conflating two separate issues: number of launch pads per site, and number of launch pads on Earth total. You're raising a problem that occurs if the latter number gets too high, but I'm advocating increasing the former number, not necessarily the latter (though the latter must increase anyway to reach SpaceX's goal of thousands of launches per day).

Due to the Mars synodic cycle, low utilization might be unavoidable. If so, pick your poison. Better to have unused pads than unused launch sites.

It reduces manpower efficiency, since you need more ops people to handle the peak, who are unneeded during the lulls. It seems to go against the way SpaceX is costing their systems.

Again, probably unavoidable. Take up your complaint with the laws of orbital mechanics. ;)

The more highly automated your launches are, the smaller this problem becomes. Hence SpaceX's desire to go "power on to launch in less than one hour with zero operators on console."

I would argue it also reduces the per-launch acoustic disruption to the local community, because by clustering the schedules it gives longer "quiet periods" between launch events.

OTOH, it puts vehicles in each others acoustic environments during critical pre-launch periods.

Fortunately that's no problem. Each launch pad would have to be far enough away to avoid fratricide in case of a RUD, so the nominal launch acoustics should be easily tolerated.

It also increases the density of launch pads per launch site.

Even denser is to use fewer pads more often. The best part is no part.

Again conflating. I'm not saying they need more pads, I'm saying they need leaner pads (ie the same number of pads at fewer unique launch sites).

Either way, SpaceX needs hundreds of launch pads globally. Would you rather they each have their own dedicated 30 km keep-out circle, or would it perhaps make sense to combine a few of them? That's the fundamental question.

https://twitter.com/elonmusk/status/1298019842356703232

And your tanker use tempo increases rather dramatically.

Having fewer tankers used more often is a more efficient than many tankers used rarely.

I'm not advocating increasing the tanker fleet size beyond the optimum (thus hurting utilization), I'm advocating using the tanker fleet efficiently.

That's why a system where tankers have a high flights-per-day utilization, as cdebuhr describes, makes sense.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 08/25/2020 11:50 pm
Re: launching in waves

It also reduces time spent rendevous-and-docking (increasing tanker utilization efficiency),

How?

Because you eliminate phasing, and you eliminate the need to wait 12-24 hours for the launch site to rotate under the orbital plane. Every launch opportunity into the Mars departure parking orbit (1-2 per day) you can launch 10 tankers instead of 1.

If there's five ships, each spends as long doing rendezvous and docking.

Five ships seems a bit too small. You can't get a full tanker, for one.

Eight tankers would make good sense, because then you only need three rounds of rendezvous (comprising 7 rendezvous total) and you get a full tanker at the end (rather than ~half a tanker).

Launching them together doesn't make them dock quicker.

You're not waiting as long for phasing, and you don't have to wait for the launch site to line up again. That makes the rendezvous quicker. Naturally the propellant transfer takes the same amount of time, so that's no different either way.

If anything, clustering means they're more likely to get in each others way. (Eg, a tanker that's taking longer to dock than expected delays the others.)

That sounds like a "it hurts when I do this" problem. Solution: don't launch select parking orbits that are too close together. Work the problem like Goldilocks and choose a distance that's "just right:" not right on top of each-other, but it's not necessary to circle the Earth 16 times until your launch site is aligned either.

it lets you combine airspace closure (NOTAM) windows, it lets you combine launch area personnel evacs and closures (streamlining operations vs. having a bunch of pads running on independent schedules), and it reduces closures of beaches and waterways.

OTOH, it reduces site efficiency, since it spends most of its time unused.

You're conflating two separate issues: number of launch pads per site, and number of launch pads on Earth total. You're raising a problem that occurs if the latter number gets too high, but I'm advocating increasing the former number, not necessarily the latter (though the latter must increase anyway to reach SpaceX's goal of thousands of launches per day).

Due to the Mars synodic cycle, low utilization might be unavoidable. If so, pick your poison. Better to have unused pads than unused launch sites.

It reduces manpower efficiency, since you need more ops people to handle the peak, who are unneeded during the lulls. It seems to go against the way SpaceX is costing their systems.

Again, probably unavoidable. Take up your complaint with the laws of orbital mechanics. ;)

The more highly automated your launches are, the smaller this problem becomes. Hence SpaceX's desire to go "power on to launch in less than one hour with zero operators on console."

I would argue it also reduces the per-launch acoustic disruption to the local community, because by clustering the schedules it gives longer "quiet periods" between launch events.

OTOH, it puts vehicles in each others acoustic environments during critical pre-launch periods.

Fortunately that's no problem. Each launch pad would have to be far enough away to avoid fratricide in case of a RUD, so the nominal launch acoustics should be easily tolerated.

It also increases the density of launch pads per launch site.

Even denser is to use fewer pads more often. The best part is no part.

Again conflating. I'm not saying they need more pads, I'm saying they need leaner pads (ie the same number of pads at fewer unique launch sites).

Either way, SpaceX needs hundreds of launch pads globally. Would you rather they each have their own dedicated 30 km keep-out circle, or would it perhaps make sense to combine a few of them? That's the fundamental question.

https://twitter.com/elonmusk/status/1298019842356703232

And your tanker use tempo increases rather dramatically.

Having fewer tankers used more often is a more efficient than many tankers used rarely.

I'm not advocating increasing the tanker fleet size beyond the optimum (thus hurting utilization), I'm advocating using the tanker fleet efficiently.

That's why a system where tankers have a high flights-per-day utilization, as cdebuhr describes, makes sense.
Well, I'm glad to see that my Crazy Idea has at least stimulated an interesting discussion!  For what its worth, I only see this idea as being even remotely viable if ground operations can be streamlined and automated to a degree that is almost unthinkable today, but which will be needed anyway (IMO anyway) to reach the scale of operations Musk envisions.  A few more points submitted for your collective consideration:

1) I know that on-orbit prop boil-off is not an insurmountable problem but ... "the best problem is no problem".  If you can get from first tanker launch to departure burn in 24-48 hours, the problem is very much reduced if not eliminated.

2) "The best part is no part" ... This is based solely on my gut feeling - counter arguments welcome: When trying to turn "parts" into "no parts", eliminating unique parts and subsystems is much more significant than just reducing the count of otherwise identical systems.  Getting rid of or reducing the need for cryocoolers/sunshades/prop storage heroics for the expense of needing a few extra tankers seems like a win to me.  Also relevant to this is the discussion in another thread: "What if Starship is unreasonably cheap".

3) One serious issue that seems to keep coming up whenever depots get discussed is the pesky problem of orbital mechanics - in short, plane changes suck.  I don't think I spelled it out in my original post, but the complete avoidance of this issue is one of the key benefits I see in tanker salvos.

4) I don't see this notion as applying solely to Mars.  Indeed, Mars may be one case where you've got so many craft going in one direction that large orbital depots may make more sense (eventually ... not anytime soon IMHO).   The advantage I see to a tanker salvo is that it allows you to put a Starship on just about any orbit or departure trajectory you wish quickly without having to worry too much about boil-off, excessive exposure to MMOD risk, etc.

Edit: I really need to proofread more carefully ...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: yg1968 on 08/26/2020 03:06 am
Quote from: Tim Dodd
[Question:] How will lunar starship refuel? It needs to go back to Earth orbit, and then get refueled by multiple tankers to do that round trip each time right? Can’t make methane on the moon sans a large cow farm, right?

Quote from: Elon Musk
[Answer:] Starship propellant is ~78% oxygen, so an O2 plant on the moon would be enough. Otherwise, we could brute-force it with tankers to low Earth orbit. That’s probably faster.

https://twitter.com/elonmusk/status/1298426245991063554
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/26/2020 03:38 am
If the tank contains an amorphous glob of frothy propellant and the tank starts accelerating around it at 1mm/hour/hour (example only) in a consistent direction eventually all the propellant will be settled at one end.

It won't overcome the "stickiness" of the froth, won't separate gas and fluid. Surface wetting and capillary effects dominate.

but microgravity, like micro brewery, is also a generic term.

As a technical term, microgravity means "millionths of a g". As a generic term, microgravity is a synonym for near-zero-g or orbital free-fall. Microgravity does not mean "anything lower than Earth gravity". It's weird to keep encountering the latter on a technical site. It's weirder that this is the only site where I've seen people do it.
That 1mm/hour/hour was not an operational suggestion. It was meant to illustrate the props migrating to one end of the tank. Yes higher G would be needed. In an earlier post I suggested low G initially, higher G to get the props settled down, then back to a lower G while transferring. No exact definitions on what high and low G mean here, but not very much. Deci-G's would be overkill.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/26/2020 04:00 am
[edit: For those getting sick of this, I'll let it drop after this post.]

Re: launching in waves

It also reduces time spent rendevous-and-docking (increasing tanker utilization efficiency),
How?
Because you eliminate phasing, and you eliminate the need to wait 12-24 hours for the launch site to rotate under the orbital plane. Every launch opportunity into the Mars departure parking orbit (1-2 per day) you can launch 10 tankers instead of 1.

That's just rephrasing the original concept. It doesn't reduce time spent in rendezvous and docking manoeuvres.

Unless you are creating the strawman alternative that the launches in both scenarios have to be over the same period (within a day). In which case, no. That's a strawman. Don't do that.

One launch a day is fine. If the pads can operate more often than that, then they are still free for other customers. You squeeze the tanker flights in between paying jobs.

If there's five ships, each spends as long doing rendezvous and docking.
Five ships seems a bit too small. You can't get a full tanker, for one.

{Shrug} I just used the figure from Musk's presentation. It's "however many refuelling flights are required per Mars Starship."

If anything, clustering means they're more likely to get in each others way. (Eg, a tanker that's taking longer to dock than expected delays the others.)
Solution: don't launch select parking orbits that are too close together.

You misread what I wrote. If there is an issue with a tanker docking and transferring propellant, it holds up the whole fleet. It's less efficient to throw them all at the same target at the same time, rather than bring them in over a more natural pace.

You're conflating two separate issues: number of launch pads per site, and number of launch pads on Earth total. You're raising a problem that occurs if the latter number gets too high, but I'm advocating increasing the former number, not necessarily the latter (though the latter must increase anyway to reach SpaceX's goal of thousands of launches per day).

None of that has anything to do with anything I wrote.

Due to the Mars synodic cycle, low utilization might be unavoidable.

Making it foolish to try to jam as many launches into as short a period as possible if you don't have to.

And I don't believe you have to. Boil-off isn't that big an issue for an accumulation-tanker, so it makes more sense to spread the refuelling out as much as possible in order to operate fewer fixed resources more often.

It reduces manpower efficiency, since you need more ops people to handle the peak, who are unneeded during the lulls. It seems to go against the way SpaceX is costing their systems.
Again, probably unavoidable. Take up your complaint with the laws of orbital mechanics.

"This is an issue. I know, let's make it worse!"

It also increases the density of launch pads per launch site.
Even denser is to use fewer pads more often. The best part is no part.
Again conflating. I'm not saying they need more pads,

Then you misunderstood cdebuhr's proposal.

Either way, SpaceX needs hundreds of launch pads globally. Would you rather they each have their own dedicated 30 km keep-out circle, or would it perhaps make sense to combine a few of them? That's the fundamental question.

One that has nothing to do with the topic.

And your tanker use tempo increases rather dramatically.
Having fewer tankers used more often is a more efficient than many tankers used rarely.
I'm not advocating increasing the tanker fleet size beyond the optimum (thus hurting utilization), I'm advocating using the tanker fleet efficiently.
That's why a system where tankers have a high flights-per-day utilization, as cdebuhr describes, makes sense.

Again, I think you misunderstand cdebuhr's proposal.

He wants more tankers, more launch pads, with all the tankers launching at once to ensure the refuelling and departure process happens in one hit. That is the opposite of increasing the per-tanker flightrate.

Either that or I misunderstood him.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/26/2020 05:26 am
Just for fun, I ran the numbers for percentage of Mars produced propellant reaching High Earth Orbit, using just standard methalox Starships as tankers, and depots in LMO and HMO:

I get 13% using propulsive capture into HEO, and 15% using aero-capture.

Eliminating legs, fins and heat-shield for the LMO-HMO & HMO-HEO legs makes negligible difference. (Except on the return leg).

Switching to hydrolox for LMO-HMO & HMO-HEO gives 15% (assuming same MR, which would be worse in reality, due to larger tanks)

Switching to hydrolox for every leg gives 19% (assuming same MR, which would be a lot worse in reality, due to EDL for larger tanks)


Everything else being equal, which it isn't, fully Hydrolox makes 50% more propellant available in HEO, which because you need less, enables 75% more payload than methalox.  More like 40% and 70% with propellant density factored in, and who know with ISRU production & storage factored in.


Using methalox for Mars to LMO, then SEP, I got 21% to 24% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then SEP, I got 14% to 15% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then hydrolox, I got 10%.

Using CO2/LOX, for Mars to LMO, then methalox , I got 8%. Might as well just send a Starship direct from the surface.


NOTE: All these trips are one-way. When you need to send an empty tanker back to HMO or LMO, it requires about 16% of the propellant just delivered for methalox, and 21% for hydrolox (lower density -> higher empty mass). Around 12% for SEP.

Edit: fix return %
This is really dismal....
Not at all! I think the numbers are pretty good, especially with SEP. (Or a Phobos tether, not listed.) We're about 10x that of Earth, particularly with the higher orbit taken into consideration.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Kansan52 on 08/26/2020 04:15 pm
So, once every Mars synod, 700 ships must launch so 100 ships can go to Mars. Seems like a good time for creation of several fuel depots so the tankers can launch before the synod reducing that launch strain.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 08/26/2020 07:35 pm
An in space accumulation tanker is the fuel depot.  A Starship can launch say 150 tons to orbit.  So an accumlation Starship will have maybe 150 tons of sun shielding, solar panels, and refrigeration equipment to deploy.  Then each tanker flight will dock with the accumulation Tanker/fuel depot.  Each tanker can deliver 150 tons of metholox.  Mass wise mostly lox, then return to earth for another fill up and relaunch.  The accumulation Starship/Tankers/fuel depot will then fill up a Starship going to Mars when the synod comes around.  You don't need 700 ships to supply 100 ships going to Mars.  You need maybe 220.  100 going to Mars, 100 in space accumulation tankers, and maybe 20 tankers filling those 100 with 5-6 flights each during the off synod.  Fewer tankers making 2-3 flights a week to fill the accumulation tankers is not insurmountable.   
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/27/2020 12:54 am
[edit: For those getting sick of this, I'll let it drop after this post.]

Suck-up. ::)

Re: launching in waves

It also reduces time spent rendevous-and-docking (increasing tanker utilization efficiency),
How?
Because you eliminate phasing, and you eliminate the need to wait 12-24 hours for the launch site to rotate under the orbital plane. Every launch opportunity into the Mars departure parking orbit (1-2 per day) you can launch 10 tankers instead of 1.

That's just rephrasing the original concept. It doesn't reduce time spent in rendezvous and docking manoeuvres.

Let's be clear about our terms. "Rendezvous" is the process of performing orbital maneuvers that bring two orbiting objects close together at ~0 relative velocity. So if you're launching all your vehicles into the same orbit, you definitely do "reduce time spent in rendezvous... manoeuvers."

And yes, the docking and fuel transfer is the same. It's the same in either case, so it doesn't provide an argument either way.

Unless you are creating the strawman alternative that the launches in both scenarios have to be over the same period (within a day). In which case, no. That's a strawman. Don't do that.

Don't worry, I'm not. But unless you're designing a system that can launch thousands of flights per year, you're not designing the same type of system that SpaceX is.

One launch a day is fine.

Maybe so, maybe not. Maybe this winds up being the " I think there is a world market for maybe five computers" of our time. But it makes no difference to what I'm saying -- you could easily limit each pad to one flight/day and still have multiple pads per site launching simultaneously. Again you're conflating two different issues.

Elon Musk's stated goal is 3 flights per day per vehicle, for reference. Unless you're suggesting he'd have more launch pads than vehicles, that implies more than one flight per day per launch pad.


If the pads can operate more often than that, then they are still free for other customers. You squeeze the tanker flights in between paying jobs.

Again, that's fine for low volume. But when you're launching thousands of flights per year, you're going to have many pads regardless. The cheaper those pads are, the better.


If there's five ships, each spends as long doing rendezvous and docking.
Five ships seems a bit too small. You can't get a full tanker, for one.

{Shrug} I just used the figure from Musk's presentation. It's "however many refuelling flights are required per Mars Starship."

And I just used the numbers from Musk's presentation. 1100 tonnes (now 1200 tonnes) divided by 150 tonnes per launch equals 8 refueling flights per Mars Starship. If we assume 100 tonnes per launch, it's twelve flights.

If anything, clustering means they're more likely to get in each others way. (Eg, a tanker that's taking longer to dock than expected delays the others.)
Solution: don't launch select parking orbits that are too close together.

You misread what I wrote. If there is an issue with a tanker docking and transferring propellant, it holds up the whole fleet.

Not at all. It doesn't hold up the whole fleet, it just holds up those two tankers. There's no reason the other tankers have to stop. You just get 6 launches worth of propellant into one tanker, rather than 8.

It's less efficient to throw them all at the same target at the same time, rather than bring them in over a more natural pace.

Two problems:

1. That's just re-phrasing your opinion without providing any additional reasoning, and

2. It's pretty clear that you call a "natural pace" is what Elon would call "insufficiently low volume."


it lets you combine airspace closure (NOTAM) windows, it lets you combine launch area personnel evacs and closures (streamlining operations vs. having a bunch of pads running on independent schedules), and it reduces closures of beaches and waterways.

OTOH, it reduces site efficiency, since it spends most of its time unused.
You're conflating two separate issues: number of launch pads per site, and number of launch pads on Earth total. You're raising a problem that occurs if the latter number gets too high, but I'm advocating increasing the former number, not necessarily the latter (though the latter must increase anyway to reach SpaceX's goal of thousands of launches per day).

None of that has anything to do with anything I wrote.

Don't worry, I'll explain it to you. I restored the quote above, too.

You're suggesting that the pads would go underutilized, because there would be an oversupply of launch pads. But that's a function of the total number of launch pads on Earth divided by the number of launches per year.

You seem to be thinking that you'd take each launch site and add N pads, increasing the total number of pads on Earth by a factor of (N+1) times. But instead I'm picturing grouping the same number of pads into a smaller number of launch sites.

The problem of course is schedule conflicts, but remember that tanker flights can go from idle to fueled in 1 hour and requires no payload processing. So tanker launch campaigns should be easy to "fit in" to the rest of the schedule.


Due to the Mars synodic cycle, low utilization might be unavoidable.

Making it foolish to try to jam as many launches into as short a period as possible if you don't have to.

You're misunderstanding.

You don't want to jam many launches into a short period. Ideally you want to have lots of launches all the time.

Maybe that's not possible, due to the Mars cycle. But that's a totally separate and unrelated question.

And I don't believe you have to. Boil-off isn't that big an issue for an accumulation-tanker, so it makes more sense to spread the refuelling out as much as possible in order to operate fewer fixed resources more often.

Once again, I'm not. The idea isn't to build an oversupply of pads and bunch up the launch schedule. You keep conflating these two things, but they're very different.

It reduces manpower efficiency, since you need more ops people to handle the peak, who are unneeded during the lulls. It seems to go against the way SpaceX is costing their systems.
Again, probably unavoidable. Take up your complaint with the laws of orbital mechanics.

"This is an issue. I know, let's make it worse!"

Except it doesn't make it worse unless you're oversupplying pads. And you're not really continuing to conflate those two, are you?

It also increases the density of launch pads per launch site.
Even denser is to use fewer pads more often. The best part is no part.
Again conflating. I'm not saying they need more pads,

Then you misunderstood cdebuhr's proposal.

I'll let cdebuhr be the judge of that. I'd be happy to accept authorship if necessary.

But if so, perhaps I improved it. :)

Either way, SpaceX needs hundreds of launch pads globally. Would you rather they each have their own dedicated 30 km keep-out circle, or would it perhaps make sense to combine a few of them? That's the fundamental question.

One that has nothing to do with the topic.

On the contrary, that's exactly the heart of the topic. You keep missing the point though.

And your tanker use tempo increases rather dramatically.
Having fewer tankers used more often is a more efficient than many tankers used rarely.
I'm not advocating increasing the tanker fleet size beyond the optimum (thus hurting utilization), I'm advocating using the tanker fleet efficiently.
That's why a system where tankers have a high flights-per-day utilization, as cdebuhr describes, makes sense.

Again, I think you misunderstand cdebuhr's proposal.

He wants more tankers, more launch pads, with all the tankers launching at once to ensure the refuelling and departure process happens in one hit. That is the opposite of increasing the per-tanker flightrate.

Either that or I misunderstood him.

Maybe. Again, I don't really care who gets credit for what idea.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/27/2020 02:30 am
Just for fun, I ran the numbers for percentage of Mars produced propellant reaching High Earth Orbit, using just standard methalox Starships as tankers, and depots in LMO and HMO:

I get 13% using propulsive capture into HEO, and 15% using aero-capture.

Eliminating legs, fins and heat-shield for the LMO-HMO & HMO-HEO legs makes negligible difference. (Except on the return leg).

Switching to hydrolox for LMO-HMO & HMO-HEO gives 15% (assuming same MR, which would be worse in reality, due to larger tanks)

Switching to hydrolox for every leg gives 19% (assuming same MR, which would be a lot worse in reality, due to EDL for larger tanks)


Everything else being equal, which it isn't, fully Hydrolox makes 50% more propellant available in HEO, which because you need less, enables 75% more payload than methalox.  More like 40% and 70% with propellant density factored in, and who know with ISRU production & storage factored in.


Using methalox for Mars to LMO, then SEP, I got 21% to 24% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then SEP, I got 14% to 15% (1800 to 3000 Isp).

Using CO2/LOX, for Mars to LMO, then hydrolox, I got 10%.

Using CO2/LOX, for Mars to LMO, then methalox , I got 8%. Might as well just send a Starship direct from the surface.


NOTE: All these trips are one-way. When you need to send an empty tanker back to HMO or LMO, it requires about 16% of the propellant just delivered for methalox, and 21% for hydrolox (lower density -> higher empty mass). Around 12% for SEP.

Edit: fix return %
This is really dismal....
Not at all! I think the numbers are pretty good, especially with SEP. (Or a Phobos tether, not listed.) We're about 10x that of Earth, particularly with the higher orbit taken into consideration.
I see your point. Didn't think of it in that way.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 08/27/2020 03:13 am
RE: Salvo launching of tankers.

Wow!  It's always nice to have a constructive comment or two, but I wan't expecting to provoke this much discussion - very interesting to see everyone's perspectives.  Back when I was in lock-down I had more time for this, but now I can hardly keep up (especially with Twark_Main and Paul451).  I apologize for not quoting prior posts and responding point by point, but I can't keep up with you two!

In case anyone want to catch up with my recent comments on the topic:
Original post: https://forum.nasaspaceflight.com/index.php?topic=50157.msg2124099#msg2124099
Reply to comments: https://forum.nasaspaceflight.com/index.php?topic=50157.msg2124225#msg2124225
Reply to comments: https://forum.nasaspaceflight.com/index.php?topic=50157.msg2124507#msg2124507

With respect to the ongoing discussion, if I must be the judge so be it!  Between the two main combatants, Twark_Main seems to have immediately grasped the essence of what I was suggesting, and pointed out a few additional advantages I hadn't considered.  Paul451, you seem to either not quite get what I'm proposing, or just don't agree with it.  That's OK ... thank you for making Twark_Main explain my idea for me!  Maybe you'll come around ...

I hope I'm not being too repetitive with prior posts here, but to summarize:

Key advantages:
1) gets past most of that orbital mechanical unpleasantness - no plane changes, minimal wait for phasing, etc.
1b) gets past all the problems of #1 when applied to the question of where to put a notional prop depot, for those thinking along those lines (this is really what gave me the idea in the first place).
2) gets tankers up and down quickly, allowing them the be re-tasked more quickly (meaning this idea might not actually require that many more tankers in the fleet), and minimizing the MMOD risk in the relative shooting gallery of LEO (I'm not really qualified to have a real opinion on how serious a risk this is, but I know there are those here who find in concerning).
3) minimize prop boil-off using only "standard" tankers, without need for special boil-off mitigation measures ("the best part is no part")
4) allows for clustering of acoustic disturbances and extreme coordination of NOTAMs (thanks for that one Twark_Main!)

A few additional comments:
1) This idea isn't sized for current or nearish-future operations - its assuming the fulfillment of Musk's vision.  Definitely doesn't make sense in the next few Mars synods, but then again Elon does seem to be in a bit of a hurry, so it might become appropriate sooner than any of us think.
2) As far as operational complexity goes, I don't see it as being any more difficult on orbit.  The tankers pair off and one half the tankers transfer all but EDL fuel to the other half.  Repeat until one tanker remains.  Now empty tankers are go for EDL at the next opportunity, and departure vessel is go for launch.  On the ground, to make Elon's vision work at all will require a massive streamlining and automation of ground operations, so that gets done either way.
3) I don't see this as needing massively more pads and tankers than the "standard" version that's been bandied about up to now.  We're going to need oodles of pads any way we do it.  Just a difference in how the pads get clustered and how they and the tankers are used.  As soon as the tankers are down, they can be getting ready for the next salvo.  It also occurs to me that you might not actually need to cluster the pads if you don't want to, you just need to mind orbital phasing and launch windows.  Instead of a salvo it would then be more like the orbital refueling version of a time-on-target artillery barrage.  It's simpler though if you do it all from one multi-pad complex.
4) This isn't just for Mars.  It also works for any mission where you need orbital refueling to make it work.  Maybe the DOD has a 50 ton payload they want to place direct to HEO/GEO? (No I have not done the math ... I've got no idea how much mass a fully fueled SS could take to GEO.)  Want to send a fully fueled tanker on its way lunar orbit?  This works for these missions too.

Anyway, thank you for the interesting discussion.  Keep it up if you wish, and I'll chime in when I can if I have something worthwhile to add.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/27/2020 03:28 am
In reference to a surge refueling fleet:


There is a great theoretical ideal of the SS system being hands off as much as possible but ground support is more than an ops center with an 'on' switch.


In the real world of real machines, there are always glitches. Somebody has to fix real broken or misadjusted things. Ignoring the ship itself, cranes, an incredible sized tank farm with pumps, cryo coolers, valves, etc.


In theory, X man hours maintenance/repair per launch doesn't care how the hours are distributed as long as they're available. In reality, that launch stand and crane that was signed off on 20 months ago has built up 20 months of bird poop in some critical place. And it hasn't been looked at since because the other gear has to be gone through for the upcoming surge. Realistically, there has to be enough manpower to service everything in some reasonable time period leading up to the surge.


Then, what happens if there's a RUD? Yup, spread em out. OK, so is all of KSC available for rent? That won't be enough so maybe we can set something up in Kazakhstan. Oh, ITAR! Hawaii? NIMBY! Ocean platforms? Now this might just work out. Yup, well set up a couple-three dozen offshore platforms. But only use em every 26 months except for a couple of em. Doesn't matter, well mass produce them and get the price down to $100-150M each - before emplacement.


This surge tanking thing has efficiency in some areas but with concurrent loss of greater efficiency in others. One ocean platform launching a tanker an arbitrary every three days will launch enough to refuel 40+ Mars ships over one synod. The downside is that the depots must have ZBO. That looks like a no brainer to me.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 08/27/2020 03:38 am
Not at all! I think the numbers are pretty good, especially with SEP. (Or a Phobos tether, not listed.) We're about 10x that of Earth, particularly with the higher orbit taken into consideration.
[/quote]

It's not terrible, but the ships still have to get from LEO to HEO, which requires about 3 Earth launched tankers each.

So for slower (but not minimal dv) cargo transits, I estimate (per cargo Starship):

5 x Earth launched tankers, with TMI directly from LEO.
3 x Earth launched tankers and 4 x Mars launched tankers, with 0.5 x methalox Mar-Earth tanker transit.
3 x Earth launched tankers and 2.5 x Mars launched tankers, with 0.5 x SEP Mar-Earth tanker transit.
3 x Earth launched tankers, 0.5 x Moon LOX tanker, and 2 x Mars launched tankers, with 0.4 x methalox Mar-Earth CH4 tanker transit.
3 x Earth launched tankers, 0.5 x Moon LOX tanker, and 1 x Mars launched tankers, with 0.2 x SEP Mar-Earth CH4 tanker transit.

For fast crew transits, I estimate (per crew Starship):

8 x Earth launched tankers, with TMI directly from LEO.
3 x Earth launched tankers and 8 x Mars launched tankers, with 1 x methalox Mar-Earth tanker transit.
3 x Earth launched tankers and 5 x Mars launched tankers, with 1 x SEP Mar-Earth tanker transit.
3 x Earth launched tankers, 1 x Moon LOX tanker, and 4 x Mars launched tankers, with 0.8 x methalox Mar-Earth CH4 tanker transit.
3 x Earth launched tankers, 1 x Moon LOX tanker, and 2 x Mars launched tankers, with 0.4 x SEP Mar-Earth CH4 tanker transit.

NB:  You need more CH4 from Mars if using Lunar LOX, because the lunar tanker needs to be re-fueled with martian CH4. If the lunar LOX tankers used hydrolox, the Mars tankers would be about halved.

To me, it doesn't make a lot of sense unless you use Lunar LOX and SEP, and even then, it's probably simpler to just bring up the CH4 from Earth to HEO (L2).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/27/2020 04:09 am
...
It's not terrible, but the ships still have to get from LEO to HEO, which requires about 3 Earth launched tankers each.
...
My idea was to send about half the propellant to LEO and half to HEO, not to just send it all the HEO.

I think a lot about this because I think for various reasons Earth launch may become limited for legitimate environmental reasons (in the long term). Additionally, by off-loading stuff to Mars, you're building up the industrial capacity there. And it might even save costs eventually.

I mean, to establish a self-sustaining architecture on Mars probably requires on the order of 100,000 tons IMLEO every year. But SpaceX is shooting for 100 megatons/year according to Musk. Eventually, if you add enough zeros to launch capacity, it becomes a serious problem. (But probably not at the 100,000tons/year level.)

Eventually you'd hope that only about 100kg per passenger is required. You fill about 1000 passengers in a Starship vehicle optimized for transit from Earth to LEO, then they transfer to Starships optimized for long distance travel. These have room for 100 people apiece, and they are filled with water, food, and other consumables stocked from Mars. They're also serviced and maybe even partially manufactured on Mars (hey, if they can do it in a field in south Texas!). All propellant comes from Mars (or maybe the Moon or asteroids). So instead of needing almost 10 tons IMLEO per passenger to Mars (for your fast transit with 8 Earth launched tankers), it could be as low as 100kg.

To put it another way, for any given constraint on Earth launch capacity, you can send nearly 100 times as many people to Mars. If the limit is 1million tons IMLEO per 2 years, you can get either 100,000 people to Mars or 10 million people to Mars. Or to put it another way, if Earth launch is capped at $100/kg IMLEO, the Earth-launch portion of a the price per ticket of a trip to Mars goes from $1 million to a mere $10,000.

But I digress.


And I think SEP may be a big part of it, but a Phobos (or perhaps Deimos) tether could make a big difference and be cheaper.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 08/27/2020 04:22 am
...
It's not terrible, but the ships still have to get from LEO to HEO, which requires about 3 Earth launched tankers each.
...
My idea was to send about half the propellant to LEO and half to HEO, not to just send it all the HEO.

I think a lot about this because I think for various reasons Earth launch may become limited for legitimate environmental reasons. Additionally, by off-loading stuff to Mars, you're building up the industrial capacity there.

If you send it to LEO, it has to be via Starships with heat-shields using the Earth's atmosphere to brake.  So you can't get them back to Mars.  All they can do is land on Earth afterwards.  But I guess there should be plenty of spare cargo SS on Mars.

However, the number of Mars launched tankers will more than double. Probably fairly cheap to send lunar LOX to LEO too, but I'm not as sure how many spare SS would be available in cis-lunar space.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/27/2020 05:31 am
...
It's not terrible, but the ships still have to get from LEO to HEO, which requires about 3 Earth launched tankers each.
...
My idea was to send about half the propellant to LEO and half to HEO, not to just send it all the HEO.

I think a lot about this because I think for various reasons Earth launch may become limited for legitimate environmental reasons. Additionally, by off-loading stuff to Mars, you're building up the industrial capacity there.

If you send it to LEO, it has to be via Starships with heat-shields using the Earth's atmosphere to brake.  So you can't get them back to Mars.  All they can do is land on Earth afterwards.  ...
Not at all. You can do slower aerobraking and then return via SEP.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/27/2020 03:44 pm
...
It's not terrible, but the ships still have to get from LEO to HEO, which requires about 3 Earth launched tankers each.
...
My idea was to send about half the propellant to LEO and half to HEO, not to just send it all the HEO.

I think a lot about this because I think for various reasons Earth launch may become limited for legitimate environmental reasons. Additionally, by off-loading stuff to Mars, you're building up the industrial capacity there.

If you send it to LEO, it has to be via Starships with heat-shields using the Earth's atmosphere to brake.  So you can't get them back to Mars.  All they can do is land on Earth afterwards.  But I guess there should be plenty of spare cargo SS on Mars.

However, the number of Mars launched tankers will more than double. Probably fairly cheap to send lunar LOX to LEO too, but I'm not as sure how many spare SS would be available in cis-lunar space.
The numbers are beyond me but AIUI, dV lunar surface to LEO is about the same as earth surface to LEO. With water mining at the poles there is ample power nearby for rail gun launch. Propellant would still be needed for insertion and return but not nearly as much.


Insertion can be a mix of thrust and aero breaking. The lunar origin tanker need not be an SS derivative. If launched magnetically 304 SS would be a hindrance. It's non magnetic. The job is so specialized a clean sheet design suggests itself. Offshoot question: Is the magnetic dipole of water strong enough that it's practical for the rail gun to interact with it directly? If so, a design trade space gets very wide very fast.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Vanspace on 08/27/2020 05:45 pm
In reference to surge/salvo refuelling:

It seems to me that mass launch to a single destination looks exactly the same as mass launch to multiple destinations. Which looks exactly the same as normal operations. Thought experiment: If you wanted to get 10 planes to the same destination at the same time from ten different airports, do the airports do anything different than if it was ten planes to ten different destinations?

This is late stage starship with multiple launch sites launching multiple times per day, say 4 pads launching 3 times per day. So if the 12 flights all go to the same destination or to 12 different locations, it all is the same the only difference is the schedule.

I would expect that surge capability will exist, but mostly be used to unscrew schedule disruptions from things like weather. Just like an airport closed due to weather will need a surge of planes when it reopens. If a pad is closed for a couple days due to hurricane/RUD/Godzilla, then a surge-like flight pattern will occur to get things back on schedule. Because disruptions will occur, ISTM that the normal schedule will leave lots of room for it.

The main advantages of salvoing as described are reduced time between starting the refuelling and mission launch. Sure, you can launch for the Moon from scratch on 24 hour notice using salvos. But how often will a BEO mission have less than a weeks notice? If it has more than a weeks notice, then scheduling 8 of the 96 flights for that week seems like no big deal.

The rest of the advantages don't seem like advantages so much as mission planning options. Yes, they can do half tank transfers until only one remains but the Why for doing it needs to come from the mission. Eliminating the need for ZBO is again mission planning options.

Clustering flights from the same pad into a short daily operations window seems likely to me particularly in BC. This is not from operational demands but because it will get imposed. Much like airports often have restricted hours of operation to reduced annoyance of the locals. I don't think 2AM launches are going to be allowed anywhere except out to sea.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 08/27/2020 07:53 pm
The numbers are beyond me but AIUI, dV lunar surface to LEO is about the same as earth surface to LEO.

Not by far.  The delta-v map I have, says that it takes 5.66 km/s to go from LEO to the Moon and land, and a bit more than 2.54 km/s to go from the surface of the Moon and back to LEO (assuming aerobraking from trans-Earth injection to LEO).  Delta-v from LEO to the Moon and back, though, is almost as much as from Earth surface to LEO (which is around 9 km/s).

I'm attaching the map I use.  Unfortunately I don't remember where I found it, so I can't give proper credit for it. :(

Note also that the exact delta-v depends on where on the Moon you land, what inclination your starting LEO orbit has, if the Moon is at apogee or perigee, and more.  The map assumes best (easiest) case, so use it just as a rough guide.

Quote from: OTV Booster
With water mining at the poles there is ample power nearby for rail gun launch. Propellant would still be needed for insertion and return but not nearly as much.

Insertion can be a mix of thrust and aero breaking. The lunar origin tanker need not be an SS derivative. If launched magnetically 304 SS would be a hindrance. It's non magnetic. The job is so specialized a clean sheet design suggests itself. Offshoot question: Is the magnetic dipole of water strong enough that it's practical for the rail gun to interact with it directly? If so, a design trade space gets very wide very fast.

Mass drivers are more suitable for launching raw material, not complicated and fragile stuff like Starship.  Most proposals I have seen have given accelerations of 10-100 g to the payloads (the mass driver prototype Gerard O'Neill and his students built in the 1970s, Mass Driver 1 (https://en.wikipedia.org/wiki/Mass_Driver_1), achieved 30 g).  Mass drivers are also typically designed to launch payloads of a few tonnes each, not the 120+ tonnes of a Starship (or 1000+ tonnes of a Starship filled with water).

I think it makes most sense to launch entirely passive cargo pods when using mass drivers, and have an active "catcher" somewhere in cis-lunar space.  That catcher would have the propulsion to counteract the impulse when they catch the cargo.  Then you don't need to send the (comparatively expensive) propulsion units back to the Moon, and the pods will be just a simple shell.  (Depending on the cargo, you might not even need a container.)

Realistically, though, mining propellant on the Moon and sending it off the Moon using a mass driver, is probably 20-30 years into the future, and by that time Starship is likely retired, so it might be a bit off topic for this thread and board...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/27/2020 10:35 pm
This is late stage starship with multiple launch sites launching multiple times per day, say 4 pads launching 3 times per day. So if the 12 flights all go to the same destination or to 12 different locations, it all is the same the only difference is the schedule.

You're still not thinking big enough. The scale Elon is proposing is truly brain-stretching.

And schedule can still impact cost.

The main advantages of salvoing as described are reduced time between starting the refuelling and mission launch. Sure, you can launch for the Moon from scratch on 24 hour notice using salvos. But how often will a BEO mission have less than a weeks notice? If it has more than a weeks notice, then scheduling 8 of the 96 flights for that week seems like no big deal.

But you only get 3 flights per 24 hour period. So you're tying up the accumulation tanker for at least 3 days, vs one day. At large scale this starts to matter, but you don't really see the advantages at such a small scale.



Clustering flights from the same pad into a short daily operations window seems likely to me particularly in BC. This is not from operational demands but because it will get imposed. Much like airports often have restricted hours of operation to reduced annoyance of the locals. I don't think 2AM launches are going to be allowed anywhere except out to sea.

Agreed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/27/2020 10:37 pm
The numbers are beyond me but AIUI, dV lunar surface to LEO is about the same as earth surface to LEO.

Not by far.  The delta-v map I have, says that it takes 5.66 km/s to go from LEO to the Moon and land, and a bit more than 2.54 km/s to go from the surface of the Moon and back to LEO (assuming aerobraking from trans-Earth injection to LEO).  Delta-v from LEO to the Moon and back, though, is almost as much as from Earth surface to LEO (which is around 9 km/s).

I'm attaching the map I use.  Unfortunately I don't remember where I found it, so I can't give proper credit for it. :(

Note also that the exact delta-v depends on where on the Moon you land, what inclination your starting LEO orbit has, if the Moon is at apogee or perigee, and more.  The map assumes best (easiest) case, so use it just as a rough guide.

Quote from: OTV Booster
With water mining at the poles there is ample power nearby for rail gun launch. Propellant would still be needed for insertion and return but not nearly as much.

Insertion can be a mix of thrust and aero breaking. The lunar origin tanker need not be an SS derivative. If launched magnetically 304 SS would be a hindrance. It's non magnetic. The job is so specialized a clean sheet design suggests itself. Offshoot question: Is the magnetic dipole of water strong enough that it's practical for the rail gun to interact with it directly? If so, a design trade space gets very wide very fast.

Mass drivers are more suitable for launching raw material, not complicated and fragile stuff like Starship.  Most proposals I have seen have given accelerations of 10-100 g to the payloads (the mass driver prototype Gerard O'Neill and his students built in the 1970s, Mass Driver 1 (https://en.wikipedia.org/wiki/Mass_Driver_1), achieved 30 g).  Mass drivers are also typically designed to launch payloads of a few tonnes each, not the 120+ tonnes of a Starship (or 1000+ tonnes of a Starship filled with water).

I think it makes most sense to launch entirely passive cargo pods when using mass drivers, and have an active "catcher" somewhere in cis-lunar space.  That catcher would have the propulsion to counteract the impulse when they catch the cargo.  Then you don't need to send the (comparatively expensive) propulsion units back to the Moon, and the pods will be just a simple shell.  (Depending on the cargo, you might not even need a container.)

Realistically, though, mining propellant on the Moon and sending it off the Moon using a mass driver, is probably 20-30 years into the future, and by that time Starship is likely retired, so it might be a bit off topic for this thread and board...


Thanks for the chart and insights. Gotta download that chart. Can't read it on my phone.


The discussion of dV to LEO involved a physics grad student back in the 80's and IIRC was about propulsive insertion.


I agree on mass driver for raw material. The little gizmo that grabs it would be, wait for it... an OTV. this is why I suggested something other than SS. If a slug of water can be frozen with all molecules aligned, and if the mass driver can grab this as is, all else needed would be a shiny Mylar cover and maybe some inexpensive homing electronics. No deposit, no return. Well, there were only two 'if's' there.


Lunar ISRU is too convenient for cislunar operations to ignore getting it where it's needed as economically as possible. A mass driver can be a rail gun or a retro steam catapult. Steam cats are proven technology albeit, not in a lunar environment. The latest aircraft carrier is cutting its teeth on a troublesome EM cat. Methinks the first small scale experimental units will show up much quicker than you suggest.


Perhaps the biggest advantage of inexpensive lift is not the direct saving but the opportunity it offers to build out the infrastructure that drives costs down further. If Eisenhower has done his famous cross country convoy with modern SUV's but still the infrastructure of the day, it wouldn't have been much easier. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TomH on 08/27/2020 11:22 pm
Initially, I thought an 18m tanker was not a good idea. I still don't think it makes any sense for the immediate future. By the time things reach the numbers being bandied about in this thread, however, it may become more efficient than so many 9m prop launches.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/27/2020 11:47 pm
Re: Delta-v maps.

Be careful using someone else's delta-v maps.

In order to work out the delta-v between two orbits in a single injection burn, you can't linearly add up the delta-v's between all the incremental orbits in-between. You can use those delta-v's, but you have to add the energies (square of velocity), then sqrt the total to get the actual delta-v for the whole injection burn.

You can create a delta-v map that does that for you, so you can just add the figures. But that map only works from a single origin point (usually Earth). It won't be correct for any other manoeuvre between other points.

Unfortunately, delta-v map creators never clarify which type it is, making it impossible to use someone else's delta-v map unless you manually work out some of the values so you can check.

This isn't a trivial thing: For some mission profiles, it can result in an order-of-magnitude propellant difference.

(There are other issues. Are they using Hohmann orbits assuming a co-planar solar-system? Are they giving the average delta-v or the optimum one? Etc. But those are more marginal.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/27/2020 11:55 pm
Initially, I thought an 18m tanker was not a good idea. I still don't think it makes any sense for the immediate future. By the time things reach the numbers being bandied about in this thread, however, it may become more efficient than so many 9m prop launches.

It still doesn't make sense, IMO. If you have an 18m tanker, then you've developed an 18m Super-Starship and Super-Superheavy. At that point, you are better off upgrading the Mars fleet to that. Lets you increase the cargo/passengers to Mars by 4-fold, or reduce the number of ships to something more manageable.

There will be a small window during the development of the 18m version where you use them as tankers for the 9m Mars-SS fleet as nearly expendable test missions for the 18's. You need to test them, you might as well get some benefits out of each successful test; similar to how the F9 landing was tested on operational launches for "free". But you aren't building your fuelling eco-system around that, the moment you are comfortable with the 18's, you start switching your entire launch infrastructure over.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/28/2020 12:14 am
You seem to be thinking that you'd take each launch site and add N pads, increasing the total number of pads on Earth by a factor of (N+1) times.

I'm not. Not even remotely. If that's what you were arguing against, then that explains why your arguments were way over there when I'm over here

Due to the Mars synodic cycle, low utilization might be unavoidable.
Making it foolish to try to jam as many launches into as short a period as possible if you don't have to.
You're misunderstanding.
You don't want to jam many launches into a short period. Ideally you want to have lots of launches all the time.
Maybe that's not possible, due to the Mars cycle. But that's a totally separate and unrelated question.

[emphasis mine]

And again, if you think that's a "separate question" then you misunderstood my original response to cdebuhr's all-at-once proposal.

The innately intermittent nature of the Mars launch window was the core, central issue behind my reply. You don't build infrastructure (ground facilities, people, tankers, superheavies, etc) that are mostly going to sit unused for 2.1yrs every 2.2yrs because you've so exceeded the ongoing launch demand between Mars synods.

You develop architecture to ensure your Mars propellant launch demand is roughly the same as your ongoing non-Mars launch demand. Even if there's vastly more demand in Earth orbit, that still means spreading out the propellant launches as much as possible.

This mostly applies after the number of Starships going to Mars increases. During the early days, you'll have extra pads for basic redundancy; extra tankers because you've multiple versions of SS in various stages of testing and you might as well use them as tankers; and extra people because you are still in rapid development mode. But when you are launching 100 Starships to Mars each synod, let alone 1000, you don't want multiple times that many pads/people/tankers that get used once per synod in order to launch all the tankers at once in order to avoid 2%-per-week boil-off, or having to spend 12hrs doing orbital phasing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/28/2020 12:41 am
it lets you combine airspace closure (NOTAM) windows, it lets you combine launch area personnel evacs and closures (streamlining operations vs. having a bunch of pads running on independent schedules), and it reduces closures of beaches and waterways.

OTOH, it reduces site efficiency, since it spends most of its time unused.

You seem to be thinking that you'd take each launch site and add N pads, increasing the total number of pads on Earth by a factor of (N+1) times.

I'm not. Not even remotely.

So then why are you worried about reduced site efficiency? You're only going to have reduced launch site efficiency if you have an inefficiently large supply of pads.

Due to the Mars synodic cycle, low utilization might be unavoidable.
Making it foolish to try to jam as many launches into as short a period as possible if you don't have to.
You're misunderstanding.
You don't want to jam many launches into a short period. Ideally you want to have lots of launches all the time.
Maybe that's not possible, due to the Mars cycle. But that's a totally separate and unrelated question.

[emphasis mine]

And again, if you think that's a "separate question" then you misunderstood my original response to cdebuhr's all-at-once proposal.

The innately intermittent nature of the Mars launch window was the core, central issue behind my reply. You don't build infrastructure (ground facilities, people, tankers, superheavies, etc) that are mostly going to sit unused for 2.1yrs every 2.2yrs because you've so exceeded the ongoing launch demand between Mars synods.

And I would agree. That's the aforementioned "inefficiently large supply of pads."

You develop architecture to ensure your Mars propellant launch demand is roughly the same as your ongoing non-Mars launch demand. Even if there's vastly more demand in Earth orbit, that still means spreading out the propellant launches as much as possible.

Again, agreed. "Inefficient oversupply," etc.

This mostly applies after the number of Starships going to Mars increases. During the early days, you'll have extra pads for basic redundancy; extra tankers because you've multiple versions of SS in various stages of testing and you might as well use them as tankers; and extra people because you are still in rapid development mode. But when you are launching 100 Starships to Mars each synod, let alone 1000, you don't want multiple times that many pads/people/tankers
[/quote]

Yep, all agreed.

that get used once per synod in order to launch all the tankers at once in order to avoid 2%-per-week boil-off, or having to spend 12hrs doing orbital phasing.

Ahh, finally we come to the misunderstanding. See, I knew you were conflating this idea with constructing an inefficient oversupply of pads!

These pads wouldn't be used intermittently. They'd be used at high utilization all synod long.s
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 08/28/2020 07:42 am
In reference to surge/salvo refuelling:

It seems to me that mass launch to a single destination looks exactly the same as mass launch to multiple destinations. Which looks exactly the same as normal operations. Thought experiment: If you wanted to get 10 planes to the same destination at the same time from ten different airports, do the airports do anything different than if it was ten planes to ten different destinations?

This is late stage starship with multiple launch sites launching multiple times per day, say 4 pads launching 3 times per day. So if the 12 flights all go to the same destination or to 12 different locations, it all is the same the only difference is the schedule.

I would expect that surge capability will exist, but mostly be used to unscrew schedule disruptions from things like weather. Just like an airport closed due to weather will need a surge of planes when it reopens. If a pad is closed for a couple days due to hurricane/RUD/Godzilla, then a surge-like flight pattern will occur to get things back on schedule. Because disruptions will occur, ISTM that the normal schedule will leave lots of room for it.

The main advantages of salvoing as described are reduced time between starting the refuelling and mission launch. Sure, you can launch for the Moon from scratch on 24 hour notice using salvos. But how often will a BEO mission have less than a weeks notice? If it has more than a weeks notice, then scheduling 8 of the 96 flights for that week seems like no big deal.

The rest of the advantages don't seem like advantages so much as mission planning options. Yes, they can do half tank transfers until only one remains but the Why for doing it needs to come from the mission. Eliminating the need for ZBO is again mission planning options.

Clustering flights from the same pad into a short daily operations window seems likely to me particularly in BC. This is not from operational demands but because it will get imposed. Much like airports often have restricted hours of operation to reduced annoyance of the locals. I don't think 2AM launches are going to be allowed anywhere except out to sea.
I think Musk said that they would not be flying out of BC on a daily basis due to noise. So will be opting for an off shore platform
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Vanspace on 08/28/2020 04:26 pm
This is late stage starship with multiple launch sites launching multiple times per day, say 4 pads launching 3 times per day. So if the 12 flights all go to the same destination or to 12 different locations, it all is the same the only difference is the schedule.

You're still not thinking big enough. The scale Elon is proposing is truly brain-stretching.

And schedule can still impact cost.
You are missing the point and the scale. Any cost difference is due to the schedule requested by the customer. I chose 4 pads/3flights/day to show that even at a fairly small scale, any request can be accommodated so the reason to do salvos must come from the customers mission planning.


The main advantages of salvoing as described are reduced time between starting the refuelling and mission launch. Sure, you can launch for the Moon from scratch on 24 hour notice using salvos. But how often will a BEO mission have less than a weeks notice? If it has more than a weeks notice, then scheduling 8 of the 96 flights for that week seems like no big deal.

But you only get 3 flights per 24 hour period. So you're tying up the accumulation tanker for at least 3 days, vs one day. At large scale this starts to matter, but you don't really see the advantages at such a small scale.
Again, you are missing the scale. Does it matter if the accumulation tanker is used for 3 days versus one? If you are short of tankers for your needs, order more, they come off the assembly line every couple days. Again, at that scale, the only reason for an accumulation tanker to come off orbit is maintenance and repair. It makes no sense to land the accumulator between every fill-up, leave it there and fill it for the next mission. So it really doesn't matter how long it takes to fill the tanker, so long as it is filled before the next customer needs it.

Clustering flights from the same pad into a short daily operations window seems likely to me particularly in BC. This is not from operational demands but because it will get imposed. Much like airports often have restricted hours of operation to reduced annoyance of the locals. I don't think 2AM launches are going to be allowed anywhere except out to sea.

Agreed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/29/2020 03:03 am

I think a lot about this because I think for various reasons Earth launch may become limited for legitimate environmental reasons (in the long term). Additionally, by off-loading stuff to Mars, you're building up the industrial capacity there. And it might even save costs eventually.

I'm not saying that well meaning individuals wouldnt mandate offworld propellant, but that is NOT the enviromentally sound approach.

The correct apprach, which elon has even aluded to, is to mandate 100% renewable sabatier-originated natural gas- not just for space launch, but for all applications where electricity doesnt beat combustion. Yes, there's a hundred years reserve of fossil natural gas. But using it  is piling more sweaters on our already overheated planet. Sabatier methane is more expensive, but is inherently carbon neutral- even if you burn it on an escape burn, the exaust will be suborbital, returning to the earthly carbon cycle.

Martian methane for earth departure is the wrong answer because the exaust will also fall to earth- but it's not from earth, it's purely addition. Technically not a fossil fuel, but a carbon import all the same, when we want to be exporting carbon as much as possible. And that's ignoring as irrelivant aerocapture atmospheric heating and the depletion of ion propellants, which are energetic enough NOT to fall in a suborbital path back to the parent body.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/29/2020 06:51 pm

I think a lot about this because I think for various reasons Earth launch may become limited for legitimate environmental reasons (in the long term). Additionally, by off-loading stuff to Mars, you're building up the industrial capacity there. And it might even save costs eventually.

I'm not saying that well meaning individuals wouldnt mandate offworld propellant, but that is NOT the enviromentally sound approach.

The correct apprach, which elon has even aluded to, is to mandate 100% renewable sabatier-originated natural gas- not just for space launch, but for all applications where electricity doesnt beat combustion. Yes, there's a hundred years reserve of fossil natural gas. But using it  is piling more sweaters on our already overheated planet. Sabatier methane is more expensive, but is inherently carbon neutral- even if you burn it on an escape burn, the exaust will be suborbital, returning to the earthly carbon cycle.

Martian methane for earth departure is the wrong answer because the exaust will also fall to earth- but it's not from earth, it's purely addition. Technically not a fossil fuel, but a carbon import all the same, when we want to be exporting carbon as much as possible. And that's ignoring as irrelivant aerocapture atmospheric heating and the depletion of ion propellants, which are energetic enough NOT to fall in a suborbital path back to the parent body.
All correct in a 'purest' sense but in a practical sense? There are many unknowns here.


I don't think anybody is suggesting exoterran propellants for use for launch from earth. Someone, maybe you, ran some numbers showing rocket exhaust from near earth operations staying with earth. I question this. Some will stay for sure but the exhaust vectors are many and the exhaust velocity is high. It will be CO2 and water vapor in any case. Not methane. Not good, but enough to amount to anything at even the most optimistic traffic projections?


Ignoring earth launch and focusing on exoterran propellants in near Earth use, and assuming 75% is captured, how does this compare to the overall the man made CO2 excesses? Among the unknowns here are future space traffic density, actual recapture rates and future Earth based CO2 production rates. Might as well add that we know little about the impact of CO2 and H2O at extreme altitudes.


The thermal impact of EDL is non zero. But with solar input of what, 1.2kW/m^2 for visible light only, unless spaceships were as common as cars this wouldn't amount to a bubble of flatulence in a cyclonic disturbance. Actually I don't have any numbers. If you have a BOE, (I'm not sure how to myself), it would interesting no matter what it shows.


Depletion of ionization gasses is a concern. If life were fair this would gracefully become a market influence and the transition to alternatives would be a painless and economically efficient affair. Do I hear laughter out there?


These things may come to pass. Or, assuming we finally have a permanent presence in space, technology may transcend these problems with propulsion alternatives we can only speculate about. They'll give us a whole new set of headaches.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/29/2020 10:04 pm
exoterran propellants

"Extraterrestrial?"  :-\

The thermal impact of EDL is non zero. But with solar input of what, 1.2kW/m^2 for visible light only, unless spaceships were as common as cars this wouldn't amount to a bubble of flatulence in a cyclonic disturbance. Actually I don't have any numbers. If you have a BOE, (I'm not sure how to myself), it would interesting no matter what it shows.

I think the real concern isn't thermal heating, but production of NOx on reentry, which is both ozone depleting and a super greenhouse gas.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 08/30/2020 02:11 am
Ozone is replaced by lightning.  Lightning happens all the time around the world. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 08/30/2020 02:17 am
Ozone is replaced by lightning.  Lightning happens all the time around the world.
While that is technically correct .... it's irrelevant.  We don't need ozone in the troposphere (where lightning produces it in minor amounts), and in fact surface level ozone is harmful to us meatbags.  It's stratospheric ozone, and the depletion of same, that's the subject here.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/30/2020 02:55 am

I think a lot about this because I think for various reasons Earth launch may become limited for legitimate environmental reasons (in the long term). Additionally, by off-loading stuff to Mars, you're building up the industrial capacity there. And it might even save costs eventually.

I'm not saying that well meaning individuals wouldnt mandate offworld propellant, but that is NOT the enviromentally sound approach.

The correct apprach, which elon has even aluded to, is to mandate 100% renewable sabatier-originated natural gas- not just for space launch, but for all applications where electricity doesnt beat combustion. Yes, there's a hundred years reserve of fossil natural gas. But using it  is piling more sweaters on our already overheated planet. Sabatier methane is more expensive, but is inherently carbon neutral- even if you burn it on an escape burn, the exaust will be suborbital, returning to the earthly carbon cycle.

Martian methane for earth departure is the wrong answer because the exaust will also fall to earth- but it's not from earth, it's purely addition. Technically not a fossil fuel, but a carbon import all the same, when we want to be exporting carbon as much as possible. And that's ignoring as irrelivant aerocapture atmospheric heating and the depletion of ion propellants, which are energetic enough NOT to fall in a suborbital path back to the parent body.
All correct in a 'purest' sense but in a practical sense? There are many unknowns here.


I don't think anybody is suggesting exoterran propellants for use for launch from earth. Someone, maybe you, ran some numbers showing rocket exhaust from near earth operations staying with earth. I question this. Some will stay for sure but the exhaust vectors are many and the exhaust velocity is high. It will be CO2 and water vapor in any case. Not methane. Not good, but enough to amount to anything at even the most optimistic traffic projections?
The effective exaust velocity is the ISP times the Standard Gravity, 9.80665 m/s^2
Even a raptor vac has an ISP of only 380 seconds
That's an effective velocity of 3.724 kilomiters a second.

LEO is 7.8 km/s. Earth escape velocity is 11.190 km/s. So if a raptor in LEO fired  in the direction of motion, the exact opposite of what a departing starship would do, propellant would reach escape velocity until the starship has spent 334 m/s of delta V

Noone's suggested propulsive braking at earth orbit, so all propellant spent will be traped in earth's SoI, much of it  in suborbital or surface-intercepting elliptical orbits. what is there to question?

Quote
Ignoring earth launch and focusing on exoterran propellants in near Earth use, and assuming 75% 100% is captured, how does this compare to the overall the man made CO2 excesses? Among the unknowns here are future space traffic density, actual recapture rates and future Earth based CO2 production rates. Might as well add that we know little about the impact of CO2 and H2O at extreme altitudes.
We are not talking about now. We are talking about a future in which mars has sufficent propellant production capability  to export back to earth orbit. A future, meanwhile, wherein earth has been ravanged by centuries  of industrial revolution, resulting in events that make the current california wildfires and the american south's double hurricane look like a cloudy day.

If, in the future when mars is capable of exporting propellant, earth industrial CO2 isnt Net Zero or lower, we deserve the venesian helscape it will be delivering us.

Quote
These things may come to pass. Or, assuming we finally have a permanent presence in space, technology may transcend these problems with propulsion alternatives we can only speculate about. They'll give us a whole new set of headaches.

None of which is defending the proposal of importing greenhouse gasses, no matter how small an absolute quantity, in a time of enviromental crisis. 1200 tons of CO2+H2O per ship's worth of fuel... and for what? to avoid building on earth the same fuel production facilities you have already built up on mars? Why is that a goal?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: FutureSpaceTourist on 10/02/2020 06:44 am
In context of more powerful Raptor reducing number on Super Heavy:

twitter.com/ppathole/status/1311893344297996289

Quote
And ultimately this upgrade will result into cutting the number of refuelling flights to 4 instead of 8 which would be a huge improvement!

https://twitter.com/elonmusk/status/1311907493182926849

Quote
Probably 5 or 6 with an optimized tanker, although filling up the ship in orbit isn’t required for Mars, so 4 is possible
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sdsds on 10/02/2020 07:51 am
Has someone presented a plausible calculation of how many refueling flights would be required to take the maximum Starship LEO payload onward to GEO?

Specifically suppose a Starship launches with the most massive payload it can carry to LEO. Does refilling from a single tanker give the Starship enough propellant to bring its payload to GEO? (The delta-v from LEO to GEO is maybe half that of Earth to LEO, but sheesh, there are logarithms involved in the calculation.... ;-) )
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 10/06/2020 03:15 am
Has someone presented a plausible calculation of how many refueling flights would be required to take the maximum Starship LEO payload onward to GEO?

Specifically suppose a Starship launches with the most massive payload it can carry to LEO. Does refilling from a single tanker give the Starship enough propellant to bring its payload to GEO? (The delta-v from LEO to GEO is maybe half that of Earth to LEO, but sheesh, there are logarithms involved in the calculation.... ;-) )

Just GEO?  GEO and return the SS to Earth?  GTO? (and then which GTO? e.g. -1500 m/s, -1800 m/s, etc.)

As a rough estimate:

Just GEO with 150t payload and 120t SS, from equatorial LEO, 370 Isp, I get 520t propellant.  Maybe 600 from higher inclination LEO. Call it 3 optimized (200t payload) tankers.

To also take enough extra propellant (approx.100t) to deorbit the empty SS and land from GEO, another 190t or a 4th optimized tanker.

If the max payload can use it's own propulsion to insert from GTO, then about 1.5 optimized tankers, with SS returning to Earth.


I also get the max payload sent to GTO on one optimized tanker load is about 100t. (Assuming 2500 m/s from LEO to GTO, you basically need 1 ton tanker propellant for each ton of SS, return prop (30t?) and payload. That's assuming the SS has it's own return prop already, plus about 50t unused prop from the lighter payload.


Edit:  Should be about 3 optimized tanker loads to send 150t payload to Mars in a minimum delta v trajectory, including landing propellant.  So I expect Elon is still planning on faster trips since he said 4.




Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sdsds on 10/06/2020 07:21 am
Thanks — having estimates for a variety of mission profiles really helps clarify things! I found this observations particularly ingriguing:

[...] I also get the max payload sent to GTO on one optimized tanker load is about 100t. (Assuming 2500 m/s from LEO to GTO, you basically need 1 ton tanker propellant for each ton of SS, return prop (30t?) and payload. That's assuming the SS has it's own return prop already, plus about 50t unused prop from the lighter payload.

Just one tanker transfer might put a 100t payload onto a GTO trajectory. That's maybe a 10-fold increase over what Ariane 5 can do, but fully reusable? Wow.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ricardoholanda511 on 10/06/2020 03:56 pm
Thanks — having estimates for a variety of mission profiles really helps clarify things! I found this observations particularly ingriguing:

[...] I also get the max payload sent to GTO on one optimized tanker load is about 100t. (Assuming 2500 m/s from LEO to GTO, you basically need 1 ton tanker propellant for each ton of SS, return prop (30t?) and payload. That's assuming the SS has it's own return prop already, plus about 50t unused prop from the lighter payload.

Just one tanker transfer might put a 100t payload onto a GTO trajectory. That's maybe a 10-fold increase over what Ariane 5 can do, but fully reusable? Wow.


I've been lurking this site for more than three years now and this time I just couldn't control my curiosity. I haven't seen any discussion on the possibility of a Martian orbital refueling in order to reduce the risks of  a hot atmospheric breaking. 

Since Phobos and Deimos look like good candidates for a LOX factory, couldn't they also be used as a tanker base to send intercepting tankers  in a long orbit  to couple with incoming starships, giving them enough fuel to reduce delta-v and avoid the dangers of a fully atmospheric breaking?

I apologise in advance if this question is stupid but I don't know how to use KSP in order to make a simulation like that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 10/06/2020 04:09 pm
Thanks — having estimates for a variety of mission profiles really helps clarify things! I found this observations particularly ingriguing:

[...] I also get the max payload sent to GTO on one optimized tanker load is about 100t. (Assuming 2500 m/s from LEO to GTO, you basically need 1 ton tanker propellant for each ton of SS, return prop (30t?) and payload. That's assuming the SS has it's own return prop already, plus about 50t unused prop from the lighter payload.

Just one tanker transfer might put a 100t payload onto a GTO trajectory. That's maybe a 10-fold increase over what Ariane 5 can do, but fully reusable? Wow.


I've been lurking this site for more than three years now and this time I just couldn't control my curiosity. I haven't seen any discussion on the possibility of a Martian orbital refueling in order to reduce the risks of  a hot atmospheric breaking. 

Since Phobos and Deimos look like good candidates for a LOX factory, couldn't they also be used as a tanker base to send intercepting tankers  in a long orbit  to couple with incoming starships, giving them enough fuel to reduce delta-v and avoid the dangers of a fully atmospheric breaking?

I apologise in advance if this question is stupid but I don't know how to use KSP in order to make a simulation like that.
Welcome to the forum!

You'd need to carry lots of prop from earth for a propulsive MOI burn, otherwise you're doing aerocapture into Mars orbit anyway, which, I've been led to understand here, is actually at least as rough as EDL is going to be.  Nice to have a Phobos depot, but if you can't get into Martiam orbit, it may as well be in another galaxy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 10/06/2020 06:42 pm
I've been lurking this site for more than three years now and this time I just couldn't control my curiosity. I haven't seen any discussion on the possibility of a Martian orbital refueling in order to reduce the risks of  a hot atmospheric breaking. 

Since Phobos and Deimos look like good candidates for a LOX factory, couldn't they also be used as a tanker base to send intercepting tankers  in a long orbit  to couple with incoming starships, giving them enough fuel to reduce delta-v and avoid the dangers of a fully atmospheric breaking?

I apologise in advance if this question is stupid but I don't know how to use KSP in order to make a simulation like that.

From Deimos to Mars' surface, entirely propulsively, is about 5.35 km/s Δv.  For a Starship with empty mass 120 tonnes, cargo of 50 tonnes, and effective exhaust velocity of 3.7 km/s (Isp=377 s), that requires (120t+50t)*(e5.35km/s / 3.7km/s - 1) ≈ 550 tonnes of propellant.

A dedicated tanker Starship, where the propellant tanks take up the entire ship and can be filled with 2000 tonnes of propellant in LEO, will have about 385 tonnes of propellant left when it reaches Deimos (Δv = 5.3 km/s from LEO to Deimos).  Assuming a "slow" (8-9 months) transfer.  So you need three such tankers for every two ships going to Mars' surface.  Assuming a tanker from Earth's surface can bring 150 tonnes of propellant to LEO, that's another 40 tanker launches to LEO.

By stopping at Deimos, you are also adding almost 1.7 km/s of Δv for the ships going to the surface before they can be refuelled at Deimos.  (Normally they would spend almost no Δv between trans-Mars injection, TMI, and landing, but now they need to spend 1.7 km/s to reach Deimos.)  They will thus need to be filled up more in LEO, or they need to go slower than the 5 months or so Elon wants.

You have thus gone from somewhere between 10 and 15 launches from Earth in order to send two ships to Mars' surface, to something like 60 launches.  Alternatively, you could have used those 45+ extra launches to send 8-10 more cargo ships, trying to land on Mars using aerobraking.  Doing that would give you extra practice in aerobraking, and with a little bit of luck making it safe.  And those ships that don't crash or burn during that testing, will have brought extra cargo to Mars, making the manned stays safer and more comfortable.

It is absolutely possible to do what you suggest, but I very much doubt that it will be worth it.

EDIT to add:

It's worth noting that stopping at Phobos or Deimos incurs a noticable delta-v cost.  If instead you refuel during the coasting phase towards Mars, the tankers save almost 1.7 km/s Δv, and can deliver 670 tonnes of propellant each, instead of 385 tonnes.  The surface-destined ships save 1.3 km/s by not stopping at Deimos.  They will now need 625 tonnes to capture and land, which is slightly less than what one tanker could deliver.  So then you would be down to 28 extra launches from Earth per two ships going to Mars' surface.  Still a fairly large overhead.

(625 tonnes is more than the 550 tonnes I got earlier, but that's because those 550 tonnes only go from Deimos to Mars' surface, while the 625 tonnes here goes from TMI to Mars' surface.  The ships that refuelled at Deimos instead needed to have ca 100 tonnes of propellant left over after TMI in order to stop at Deimos.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: JaimeZX on 10/06/2020 08:49 pm
I don't think (edit: ricardoholanda511) was suggesting Starship stop at Diemos, but rather sending tankers from Deimos to intercept the inbound ship on arrival.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 10/06/2020 09:45 pm
I don't think cdebuhr was suggesting Starship stop at Diemos, but rather sending tankers from Deimos to intercept the inbound ship on arrival.

(It was ricardoholanda511 who suggested it, not cdebuhr.)

Ah, you're probably right.  Upon re-reading again, I think I misread Ricardo's proposal.  (I actually read it that way first, then upon re-reading I managed to interpret it differently...)

But in that case, I have other objections. :)

The atmospheric braking will likely only be dangerous in the early days of Mars missions.  After a few tries of landing unmanned ships, SpaceX will hopefully have made that maneuver safe for Starship.  And then the Phobos or Deimos propellant factory will no longer be needed (at least not for this purpose).

But, that propellant factory would need to be built early.  Before any manned flights to Mars.  And importantly, you would need to develop equipment for mining Phobos/Deimos crust and processing it into propellant.  Equipment that would need to work in almost no gravity (less than 1/1000th of Earth gravity).  That's probably technology readiness level 3, or possibly 4, so it will take quite a lot of time, effort, and money, before it can be deployed for real and be relied upon.

You probably also first need to do some more surveying of Phobos and/or Deimos to verify if and how they can be used for making propellant.

I believe it will be significantly faster and cheaper to just send unmanned Starships to try out aerobraking at Mars until it has become safe enough for humans.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ricardoholanda511 on 10/06/2020 10:22 pm
Many thanks to cdebuhr, JaimeZX e tbellman for the assessment of my question.

Tbellman's second answer really went to the point and made me realise that Elon Musk's self imposed (rightly so) time and technology constraints pretty much exclude some possibilities.

What made me worry about the atmospheric braking only option was also the impossibility of delay, in case  Mars' climate conditions go awry and pose some unknown danger to the ship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: dgmckenzie on 10/06/2020 10:25 pm
In the Sands of Mars by Arthur C. Clarke the liner parked off Phobos, I don't think it was Deimos.
They pulled in in and then the passengers transshipped to a Moon to Mars shuttle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Norm38 on 10/07/2020 04:54 pm
Quote
Probably 5 or 6 with an optimized tanker, although filling up the ship in orbit isn’t required for Mars, so 4 is possible


What does Musk mean my this, that filling up the ship in orbit (LEO?) isn't required for Mars?  He's saying Starship doesn't need full tanks to get to Mars?  But if it did have full tanks, couldn't it make the trip faster?
Or speed is limited by aerobraking heating and coming in faster isn't possible, so only need to fill up as full as it takes to get to that max speed?

Has this been discussed already and I missed it?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: xvel on 10/07/2020 05:12 pm
Quote
Probably 5 or 6 with an optimized tanker, although filling up the ship in orbit isn’t required for Mars, so 4 is possible


What does Musk mean my this, that filling up the ship in orbit (LEO?) isn't required for Mars?  He's saying Starship doesn't need full tanks to get to Mars?  But if it did have full tanks, couldn't it make the trip faster?
Or speed is limited by aerobraking heating and coming in faster isn't possible, so only need to fill up as full as it takes to get to that max speed?

Has this been discussed already and I missed it?

1. Coming in faster is possible but will probably need also propulsive braking
2. Each time you refuel, the delta v increase is smaller and smaller, example:

starship dry mass 120t, 100t payload, ISP 370s

propload dV
200t   - 2346 m/s
400t   - 3759 m/s
600t   - 4774 m/s
800t   - 5566 m/s
1000t  - 6215 m/s
1200t  - 6766 m/s


you don't get a lot on the last two refuellings
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Simbulation on 10/07/2020 09:23 pm
I am not sure if this has been asked yet, but I could not find this talked about in this thread. Sorry if this has been already asked.

Does anybody have any idea about the technical details of actually doing In-Orbit refueling? How do you go about physically getting propellants in 0g to transfer from one vehicle to another? SpaceX renders show SS docking with a tanker from the aft end, does this make sense?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: docmordrid on 10/07/2020 09:36 pm
Quote
Probably 5 or 6 with an optimized tanker, although filling up the ship in orbit isn’t required for Mars, so 4 is possible

What does Musk mean my this, that filling up the ship in orbit (LEO?) isn't required for Mars?  He's saying Starship doesn't need full tanks to get to Mars?
>

That's how I took it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/07/2020 10:19 pm
I am not sure if this has been asked yet, but I could not find this talked about in this thread. Sorry if this has been already asked.

Does anybody have any idea about the technical details of actually doing In-Orbit refueling? How do you go about physically getting propellants in 0g to transfer from one vehicle to another? SpaceX renders show SS docking with a tanker from the aft end, does this make sense?
As I understand it...

SS is designed to be fuelled via feed lines from SH, through its skirt. To refuel in orbit you dock the two of them skirt-to-skirt and connect the same feed lines. Then they use RCS to accelerate slightly to settle the propellant, and pump it from the tanker to the other SS.

If you look at images/video of the refuelling you'll see that the two SSs are rotated 180 degrees with respect to each other, so the LOX feed line of one matches up with the LOX vent line of the other (and the same for Methane).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: guckyfan on 10/08/2020 06:04 am
Quote
Probably 5 or 6 with an optimized tanker, although filling up the ship in orbit isn’t required for Mars, so 4 is possible

What does Musk mean my this, that filling up the ship in orbit (LEO?) isn't required for Mars?  He's saying Starship doesn't need full tanks to get to Mars?
>

That's how I took it.
s

The travel time has increased somewhat. The fuel needs have come down. Looks to me like they need to reduce the load on the heat shield on EDL. Maybe with improvements of the heat shield they can go faster, or they can add payload.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 10/08/2020 03:32 pm
I am not sure if this has been asked yet, but I could not find this talked about in this thread. Sorry if this has been already asked.

Does anybody have any idea about the technical details of actually doing In-Orbit refueling? How do you go about physically getting propellants in 0g to transfer from one vehicle to another? SpaceX renders show SS docking with a tanker from the aft end, does this make sense?

The consensus is that after docking reaction control thrusters produce small acceleration (about a milli-gee) to settle the propellants, then:

* receiving tank is vented to very low pressure (well below atmospheric)
* valves are opened and pressure difference pushes the propellants to the receiving tanks

Lowering pressure has one additional advantage: it ensures the remaining liquid is superchilled (lowering pressure lowers boiling point and boiloff then cools the stuff)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Simbulation on 10/08/2020 06:59 pm
I am not sure if this has been asked yet, but I could not find this talked about in this thread. Sorry if this has been already asked.

Does anybody have any idea about the technical details of actually doing In-Orbit refueling? How do you go about physically getting propellants in 0g to transfer from one vehicle to another? SpaceX renders show SS docking with a tanker from the aft end, does this make sense?

The consensus is that after docking reaction control thrusters produce small acceleration (about a milli-gee) to settle the propellants, then:

* receiving tank is vented to very low pressure (well below atmospheric)
* valves are opened and pressure difference pushes the propellants to the receiving tanks

Lowering pressure has one additional advantage: it ensures the remaining liquid is superchilled (lowering pressure lowers boiling point and boiloff then cools the stuff)
I am not sure if this has been asked yet, but I could not find this talked about in this thread. Sorry if this has been already asked.

Does anybody have any idea about the technical details of actually doing In-Orbit refueling? How do you go about physically getting propellants in 0g to transfer from one vehicle to another? SpaceX renders show SS docking with a tanker from the aft end, does this make sense?
As I understand it...

SS is designed to be fuelled via feed lines from SH, through its skirt. To refuel in orbit you dock the two of them skirt-to-skirt and connect the same feed lines. Then they use RCS to accelerate slightly to settle the propellant, and pump it from the tanker to the other SS.

If you look at images/video of the refuelling you'll see that the two SSs are rotated 180 degrees with respect to each other, so the LOX feed line of one matches up with the LOX vent line of the other (and the same for Methane).
thank you very much for the answers

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Norm38 on 10/08/2020 09:38 pm
Is there a need on orbit for the fuel to be superchilled?  If the tanks aren't full, then the higher density doesn't matter. 

And for lunar missions, assuming full tanks are needed, then how is the fuel maintained as superchilled?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 10/09/2020 01:10 am
Is there a need on orbit for the fuel to be superchilled?  If the tanks aren't full, then the higher density doesn't matter. 

And for lunar missions, assuming full tanks are needed, then how is the fuel maintained as superchilled?

For a lunar mission from NRHO to surface and back, full tanks aren't needed.

Assuming an HLS SS with crew, cabin and all the equipment is 150 t and it takes 50 t cargo to the surface and only a few hundred kg of rocks back, it would need just over 600 t of propellant. Roughly 70% for the descent & landing and 30% for the ascent. And in this scenario the propellant need decreases by about 1.1 t for each 1 t decrease in cargo.

Of course, with or without cargo, a 150 t HLS SS can't get from LEO to NHRO with 600+ t of propellant remaining. It will need at one tanker flight for the first landing, and two for each subsequent landing.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/09/2020 04:45 pm
I am not sure if this has been asked yet, but I could not find this talked about in this thread. Sorry if this has been already asked.

Does anybody have any idea about the technical details of actually doing In-Orbit refueling? How do you go about physically getting propellants in 0g to transfer from one vehicle to another? SpaceX renders show SS docking with a tanker from the aft end, does this make sense?

The consensus is that after docking reaction control thrusters produce small acceleration (about a milli-gee) to settle the propellants, then:

* receiving tank is vented to very low pressure (well below atmospheric)
* valves are opened and pressure difference pushes the propellants to the receiving tanks

Lowering pressure has one additional advantage: it ensures the remaining liquid is superchilled (lowering pressure lowers boiling point and boiloff then cools the stuff)

The advantage of sub-chilling is that it gets more propellant in the tank. But venting would do the opposite. I expect they'll just pump liquid from A to B, while simultaneously letting ullage gas flow from B to A.

If this sub-chilling is intended to reduce boil-off on orbit, then it's counter-productive. You'd lose less propellant if you simply have an always-on "thermostat" that vents the tanker when it's over a set pressure (technically a barostat). This sort of "bang bang" pressure control is already used on Falcon 9, and presumably Starship. Standard PMDs allow gas venting without an ullage burn.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 10/09/2020 11:58 pm
I am not sure if this has been asked yet, but I could not find this talked about in this thread. Sorry if this has been already asked.

Does anybody have any idea about the technical details of actually doing In-Orbit refueling? How do you go about physically getting propellants in 0g to transfer from one vehicle to another? SpaceX renders show SS docking with a tanker from the aft end, does this make sense?

The consensus is that after docking reaction control thrusters produce small acceleration (about a milli-gee) to settle the propellants, then:

* receiving tank is vented to very low pressure (well below atmospheric)
* valves are opened and pressure difference pushes the propellants to the receiving tanks

Lowering pressure has one additional advantage: it ensures the remaining liquid is superchilled (lowering pressure lowers boiling point and boiloff then cools the stuff)

The advantage of sub-chilling is that it gets more propellant in the tank. But venting would do the opposite. I expect they'll just pump liquid from A to B, while simultaneously letting ullage gas flow from B to A.

If this sub-chilling is intended to reduce boil-off on orbit, then it's counter-productive. You'd lose less propellant if you simply have an always-on "thermostat" that vents the tanker when it's over a set pressure (technically a barostat). This sort of "bang bang" pressure control is already used on Falcon 9, and presumably Starship. Standard PMDs allow gas venting without an ullage burn.
Sub-chilling would still get more propellant into the tanks, it is just a question of whether it will be needed. If you need 5.x tanker loads to fully refuel in a specific orbit you get 1-x loads of propellant from tanker 6 to play with. You can use it to either raise the orbit (dragging the extra tanker mass up there as well) or you can use it to sub-chill the loaded propellant and squeeze in that last drop... It might even be that the Raptors needs/prefers sub-chilled propellants, I seem to recall that they have aborted static fires due to the propellant getting to hot.

BTW, sub-chilled technically just means below boiling point at the current pressure, the temperature can be both lower and higher than the boiling point at 1 atm. It is worth noting that the common bulkhead means that the methane will tend to always be sub-chilled and even then there will be a heat flow from the LCH4 tank to the LOX as the vapor pressure of oxygen at 90.86 K (the freezing point of methane) is still 1.07 bar (attempting to invert the bulkhead).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/10/2020 10:20 pm
I am not sure if this has been asked yet, but I could not find this talked about in this thread. Sorry if this has been already asked.

Does anybody have any idea about the technical details of actually doing In-Orbit refueling? How do you go about physically getting propellants in 0g to transfer from one vehicle to another? SpaceX renders show SS docking with a tanker from the aft end, does this make sense?

The consensus is that after docking reaction control thrusters produce small acceleration (about a milli-gee) to settle the propellants, then:

* receiving tank is vented to very low pressure (well below atmospheric)
* valves are opened and pressure difference pushes the propellants to the receiving tanks

Lowering pressure has one additional advantage: it ensures the remaining liquid is superchilled (lowering pressure lowers boiling point and boiloff then cools the stuff)

The advantage of sub-chilling is that it gets more propellant in the tank. But venting would do the opposite. I expect they'll just pump liquid from A to B, while simultaneously letting ullage gas flow from B to A.

If this sub-chilling is intended to reduce boil-off on orbit, then it's counter-productive. You'd lose less propellant if you simply have an always-on "thermostat" that vents the tanker when it's over a set pressure (technically a barostat). This sort of "bang bang" pressure control is already used on Falcon 9, and presumably Starship. Standard PMDs allow gas venting without an ullage burn.
Sub-chilling would still get more propellant into the tanks, it is just a question of whether it will be needed. If you need 5.x tanker loads to fully refuel in a specific orbit you get 1-x loads of propellant from tanker 6 to play with. You can use it to either raise the orbit (dragging the extra tanker mass up there as well) or you can use it to sub-chill the loaded propellant and squeeze in that last drop... It might even be that the Raptors needs/prefers sub-chilled propellants, I seem to recall that they have aborted static fires due to the propellant getting to hot.

BTW, sub-chilled technically just means below boiling point at the current pressure, the temperature can be both lower and higher than the boiling point at 1 atm. It is worth noting that the common bulkhead means that the methane will tend to always be sub-chilled and even then there will be a heat flow from the LCH4 tank to the LOX as the vapor pressure of oxygen at 90.86 K (the freezing point of methane) is still 1.07 bar (attempting to invert the bulkhead).

Good point. In the special case where it's the very last tanker load (which won't be used for Mars flights, per Elon), it might make sense to subchill the propellants. You gain between 0% and 7% vs raising orbit, depending on how much prop is left over in the tanker.

But I don't think they'll tailor the in-space refueling method for every flight just based on that one special case. Since Starship already has vents they can sub-chill in that special case (if the numbers shake out) without adding any additional hardware. But if venting is the only way to do on-orbit refueling, then they're forced to lose propellant always, whether or not it's the last tanker load.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thrustpuzzle on 10/10/2020 10:57 pm
Remember subchilling doesn't just increase storage capacity of tanks (by densification), it also increases both engine thrust and ISP by allowing the turbopumps to inject more mass per second at higher pressure. So even if the Starship doesn't need to subchill for extra capacity, it might still be worthwhile for performance, especially during a max-thrust injection burn.

With regards to in-orbit subchilling the cryogens by venting, I kind of cringe when it's proposed to subchill the cryogens by venting some into space to chill the remainder. Sure, that works, but you're literally throwing away precious mass. What are the energy economics comparing a solar-powered Stirling cryocooler versus venting?  I guess you'd start with computing how many kg of LOX or LCH4 you must vent to decrease temperature of the whole tank by say 20C, then compare to the wattage and mass of a Stirling cooler system (with solar and batteries) that could do the same thing in say 12 hours.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: FutureSpaceTourist on 10/14/2020 04:38 pm
https://twitter.com/spacex/status/1316417597257129985

Quote
NASA has selected Starship for a propellant transfer demonstration! Combining Starship’s rapid reusability with orbital refilling is critical to economically transporting large numbers of crew and cargo to the Moon and Mars go.nasa.gov/3jWLKpA

Quote
SpaceX of Hawthorne, California, $53.2 million
Large-scale flight demonstration to transfer 10 metric tons of cryogenic propellant, specifically liquid oxygen, between tanks on a Starship vehicle. SpaceX will collaborate with Glenn and Marshall.

https://www.nasa.gov/directorates/spacetech/solicitations/tipping_points/2020_selections/
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: enzo on 10/14/2020 05:30 pm
The way it's worded, "between tanks on a Starship", is presumably a mistake, as there should be two Starships involved? Otherwise it would not be a "large-scale" test but a small one.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: maquinsa on 10/14/2020 05:57 pm
The way it's worded, "between tanks on a Starship", is presumably a mistake, as there should be two Starships involved? Otherwise it would not be a "large-scale" test but a small one.

It might be a fuel transfer between main LOX tank and header tank to test it, 10 tons is within the capacity of headers.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: RonM on 10/14/2020 06:20 pm
The way it's worded, "between tanks on a Starship", is presumably a mistake, as there should be two Starships involved? Otherwise it would not be a "large-scale" test but a small one.

It might be a fuel transfer between main LOX tank and header tank to test it, 10 tons is within the capacity of headers.

It would be a good idea to test the transfer method between tanks on a ship before testing transfer between two ships.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 10/14/2020 07:59 pm
The way it's worded, "between tanks on a Starship", is presumably a mistake, as there should be two Starships involved? Otherwise it would not be a "large-scale" test but a small one.

It might be a fuel transfer between main LOX tank and header tank to test it, 10 tons is within the capacity of headers.

It would be a good idea to test the transfer method between tanks on a ship before testing transfer between two ships.
And once you have docked the difference between two tanks on a single Starship or two different Starships is just slightly longer pipes and perhaps an extra valve or two.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/14/2020 08:52 pm
The way it's worded, "between tanks on a Starship", is presumably a mistake, as there should be two Starships involved? Otherwise it would not be a "large-scale" test but a small one.

It might be a fuel transfer between main LOX tank and header tank to test it, 10 tons is within the capacity of headers.

It would be a good idea to test the transfer method between tanks on a ship before testing transfer between two ships.
And once you have docked the difference between two tanks on a single Starship or two different Starships is just slightly longer pipes and perhaps an extra valve or two.
Even if the difference was that minor (which I don't think it is), there's the small matter of "once you have docked" :-)

SpaceX need to figure out the details of reliably transferring cryo propellants in micro-g, and starting with everything plumbed in together and shown to be working on the ground is a good first step.

That programme can happen in parallel with orbital rendezvous work, which would lead to docking, which would lead to connecting up the pipes and testing the valves, and then finally the two streams come together with actual orbital refuelling.

The award is likely just NASA paying SpaceX to do something that was an early step in their plan anyway, in order to share information / data.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 10/14/2020 09:11 pm
Cargo SS flies to orbit, opens chomper and drops a 'tank'.

Circle around and 'dock' with the tank and transfer fuel.

When finished, transfer fuel back and discard the 'tank' to burn up on re-entry.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/14/2020 09:35 pm
Cargo SS flies to orbit, opens chomper and drops a 'tank'.

Circle around and 'dock' with the tank and transfer fuel.

When finished, transfer fuel back and discard the 'tank' to burn up on re-entry.
But why?

It's practicing something they'll never need to do.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 10/14/2020 09:38 pm
Is it just me. It seems that inorbit fueling is pretty simple. No new physics. Nothing pushing physics to the limit.

Its just plumbing.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/14/2020 09:51 pm
Is it just me. It seems that inorbit fueling is pretty simple. No new physics. Nothing pushing physics to the limit.

Its just plumbing.
Safely, reliably and robotically transferring cryogenic liquids in microgravity while dealing with large changes in heat (sunlight/shadow) is presumably not trivial.

That said, these awards are for "tipping point" technologies that are on the brink of being considered viable, to nudge them across the line. NASA is quite a conservative organisation these days, and often needs to see something done before they will incorporate it into their plans.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Jimmy Murdok on 10/14/2020 10:17 pm
NASA is covering the cost of launching a Starhip and testing the "baby steps" of cryogenic transfer. In large quantities. This is great news!

A sphere of 2.6m of diameter would have 10 tones of liquid.The header tanks fit the bill.
https://pbs.twimg.com/media/EOZKIjwUcAA96zr?format=jpg&name=small (https://pbs.twimg.com/media/EOZKIjwUcAA96zr?format=jpg&name=small)

I don´t think it´s between 2 Starships, that will come later. Neither there is need to shape it like a specialized payload.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oiorionsbelt on 10/14/2020 10:29 pm
Two starships would the quickest easiest way to transfer 10 tons of LOX from one tank to another.
 Starships by design, plan to have everything needed to accomplish the task and are currently under construction.
Edit: Nope, it says specifically :) 
Quote
between tanks on a Starship vehicle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/14/2020 11:00 pm
Two starships would the quickest easiest way to transfer 10 tons of LOX from one tank to another.
 Starships by design, plan to have everything needed to accomplish the task and are currently under construction.
Edit: Nope, it says specifically :) 
Quote
between tanks on a Starship vehicle.
Yep.

And as recently noted, 10 tons is about the capacity of a header tank.

I think NASA asked for mission concepts for demonstrating orbital cryogenic refuelling and SpaceX responded with "how about we show you a transfer of LOX between a Starship's main tank and its header tank?" NASA responded with "sounds good, how much will it cost?" so SpaceX produced a bill-of-materials for getting a Starship into orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cferreir on 10/15/2020 03:45 am
It would be awesome if someone could post:

1- How much propellant is transferred by the Progress to the ISS?
2- How do they transfer the two different propellants?
3- How does the hypergolic propellant transfer relate to cryo propellant in terms of temp/pressure?

I tried to search around but have still not found info on it. Anyone?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/15/2020 11:09 am
It would be awesome if someone could post:

1- How much propellant is transferred by the Progress to the ISS?
2- How do they transfer the two different propellants?
3- How does the hypergolic propellant transfer relate to cryo propellant in terms of temp/pressure?

I tried to search around but have still not found info on it. Anyone?
I found a figure of 650kg of propellant transferred at http://russianspaceweb.com/progress-ms-13.html

There are some additional details under "refuelling module" at https://spaceflight101.com/spacecraft/progress-ms/

I also found this paper on the subject: https://www.researchgate.net/publication/255579266_Technologies_for_Refueling_Spacecraft_On-Orbit

"To avoid the problems of phase
separation a flexible membrane separates the liquid
from the pressurant gas. Then the liquid can be
transferred by pressurizing the tank without worrying
about ingesting vapor. Drawbacks of this system
include life of the membrane, weight and an inability
to deal with vapor evolved from the bulk liquid.
Nevertheless the Progress module includes resupply
tanks holding about 870 kg of propellant (two tanks
of nitrogen tetroxide and two of UDMH hydrazine).
High-pressure nitrogen is used as the pressurant. A
compressor is used to lower pressure in the receiver
tank by transferring nitrogen back into high-pressure
storage bottles. After the lines have been leak checked
the fuel then oxidizer are transferred one at a time to the
station. Separate transfer for each reduces the hazard in
case of a leak. The process can be controlled either by a
ground station or the space station crew."

Also...

"Systems for cryogen propellants such as liquid hydrogen and
liquid oxygen have unique challenges. The large scale of
the systems for which these propellants are attractive
makes any in-tank structure large and complex. No
membrane material that can be used at cryogenic
temperatures has been found. Elastomeric membranes
have poor cycle life in liquid oxygen and hydrogen
diffuses through at an unacceptable rate 43 . At these low
temperatures metal membranes suffer from poor
flexibility and limited life due to cracking. "

[edit: fix quote from paper]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: volker2020 on 10/15/2020 12:11 pm
SpaceX is planing to accelerate the ship, before pumping, which means they don't need a membrane to separate the fluids from the gas.

I think it might be hard, to do the same with the ISS.

I have to admit though, that I don't understand one part of the problem, to me pumping fuel into a engine, is actually the same as pumping it into another tank. We have seen various time, that the process of starting and restarting an engine in space is possible, why should the tanking be any different? I even say it would be easier, because you don't need to pump as much, and bubbles of gas pose no problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ZChris13 on 10/15/2020 02:33 pm
SpaceX is planing to accelerate the ship, before pumping, which means they don't need a membrane to separate the fluids from the gas.

I think it might be hard, to do the same with the ISS.

I have to admit though, that I don't understand one part of the problem, to me pumping fuel into a engine, is actually the same as pumping it into another tank. We have seen various time, that the process of starting and restarting an engine in space is possible, why should the tanking be any different? I even say it would be easier, because you don't need to pump as much, and bubbles of gas pose no problem.
The main difference is that once your engine is started it produces thrust, which will continually settle your tanks. No such easy mechanism exists when you're capturing the propellant in another tank once it passes through your pump.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: r8ix on 10/15/2020 03:08 pm
SpaceX is planing to accelerate the ship, before pumping, which means they don't need a membrane to separate the fluids from the gas.

I think it might be hard, to do the same with the ISS.

I have to admit though, that I don't understand one part of the problem, to me pumping fuel into a engine, is actually the same as pumping it into another tank. We have seen various time, that the process of starting and restarting an engine in space is possible, why should the tanking be any different? I even say it would be easier, because you don't need to pump as much, and bubbles of gas pose no problem.
The main difference is that once your engine is started it produces thrust, which will continually settle your tanks. No such easy mechanism exists when you're capturing the propellant in another tank once it passes through your pump.
AIUI, the plan has always been to use slight continuous acceleration (~0.1g?) to keep the propellant settled for the duration of the transfer...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: aero on 10/15/2020 03:10 pm
Would it be possible to just blow the LOX into the other tank? That is, fan gaseous O2 from the full tank to the empty with globs of LOX entrained, then return the gaseous O2 via a filter that mostly keeps the globs of LOX in the tank being filled.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/15/2020 06:31 pm
SpaceX is planing to accelerate the ship, before pumping, which means they don't need a membrane to separate the fluids from the gas.

I think it might be hard, to do the same with the ISS.

I have to admit though, that I don't understand one part of the problem, to me pumping fuel into a engine, is actually the same as pumping it into another tank. We have seen various time, that the process of starting and restarting an engine in space is possible, why should the tanking be any different? I even say it would be easier, because you don't need to pump as much, and bubbles of gas pose no problem.
The main difference is that once your engine is started it produces thrust, which will continually settle your tanks. No such easy mechanism exists when you're capturing the propellant in another tank once it passes through your pump.
AIUI, the plan has always been to use slight continuous acceleration (~0.1g?) to keep the propellant settled for the duration of the transfer...
That's my understanding, yes, though they can potentially stop the acceleration one the propellant is settled and just use pressure & surface tension to keep it that way.

This is one of the benefits of using the primary propellants in your RCS system though; you don't have to worry about running out of RCS fuel from running them for a long time.

I suspect (based on nothing at all) that they'll use electric pumps to speed up the transfer as well.


So no fundamental issues to overcome, but you are working in micro-gravity with cryogenic propellants going through a full day/night cycle every 90 minutes, while trying to minimise losses. No-one has done that before, so you can't just dismiss it as trivial.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 10/15/2020 07:21 pm
AIUI, the plan has always been to use slight continuous acceleration (~0.1g?) to keep the propellant settled for the duration of the transfer...

More like 0.001 g ("milli-g acceleration" is the words they have used), but otherwise correct.  At 0.1 g, they would very quickly leave LEO...  (And of course, quickly consume the propellant.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 10/15/2020 08:18 pm
SpaceX is planing to accelerate the ship, before pumping, which means they don't need a membrane to separate the fluids from the gas.

I think it might be hard, to do the same with the ISS.

I have to admit though, that I don't understand one part of the problem, to me pumping fuel into a engine, is actually the same as pumping it into another tank. We have seen various time, that the process of starting and restarting an engine in space is possible, why should the tanking be any different? I even say it would be easier, because you don't need to pump as much, and bubbles of gas pose no problem.
The main difference is that once your engine is started it produces thrust, which will continually settle your tanks. No such easy mechanism exists when you're capturing the propellant in another tank once it passes through your pump.
AIUI, the plan has always been to use slight continuous acceleration (~0.1g?) to keep the propellant settled for the duration of the transfer...
That's my understanding, yes, though they can potentially stop the acceleration one the propellant is settled and just use pressure & surface tension to keep it that way.

This is one of the benefits of using the primary propellants in your RCS system though; you don't have to worry about running out of RCS fuel from running them for a long time.

I suspect (based on nothing at all) that they'll use electric pumps to speed up the transfer as well.


So no fundamental issues to overcome, but you are working in micro-gravity with cryogenic propellants going through a full day/night cycle every 90 minutes, while trying to minimise losses. No-one has done that before, so you can't just dismiss it as trivial.

Except just about every time a second stage has been relight.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Stan-1967 on 10/15/2020 09:04 pm
AIUI, the plan has always been to use slight continuous acceleration (~0.1g?) to keep the propellant settled for the duration of the transfer...

More like 0.001 g ("milli-g acceleration" is the words they have used), but otherwise correct.  At 0.1 g, they would very quickly leave LEO...  (And of course, quickly consume the propellant.)
Some quick calculations suggest a 250ton SS ( 100 tons propellant) would need around a 2.5kN ullage thruster for .001g acceleration.  If the thruster had a  ISP of around 280 (i.e hypergolic level ISP) it would only consume around 1kg/s.  If the thruster could maintain that .001g for 3600 seconds ( 1 hour), the mass penalty for the burn is then 3.6tons.  Seems reasonable.

As to leaving LEO, you don't have to burn in a prograde direction.  Do the burn 90 degrees from prograde and so as to not change perigee or apogee so that you only create slight inclination change.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/15/2020 09:11 pm
That's my understanding, yes, though they can potentially stop the acceleration one the propellant is settled and just use pressure & surface tension to keep it that way.

This is one of the benefits of using the primary propellants in your RCS system though; you don't have to worry about running out of RCS fuel from running them for a long time.

I suspect (based on nothing at all) that they'll use electric pumps to speed up the transfer as well.


So no fundamental issues to overcome, but you are working in micro-gravity with cryogenic propellants going through a full day/night cycle every 90 minutes, while trying to minimise losses. No-one has done that before, so you can't just dismiss it as trivial.

Except just about every time a second stage has been relight.
So what's your take on this then? NASA are simply so totally incompetent that they have to pay SpaceX tens of millions to do something trivial?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 10/15/2020 11:44 pm
Some quick calculations suggest a 250ton SS ( 100 tons propellant) would need around a 2.5kN ullage thruster for .001g acceleration.  If the thruster had a  ISP of around 280 (i.e hypergolic level ISP) it would only consume around 1kg/s.  If the thruster could maintain that .001g for 3600 seconds ( 1 hour), the mass penalty for the burn is then 3.6tons.  Seems reasonable.

At a minimum, you will be pushing 340 tonnes: two Starships (120t each) docked together, and 100 tonnes of propellant.  3.3 kN needed, using 1.2 kg/s (assuming your Isp).

But you will also need to be able to push ~1600 tonnes: two Starships, 150 tonnes of cargo, 1200 tonnes of propellant (from a full accumulation tanker to an empty payload ship, or the last transfer from a tanker to the accumulation tanker), plus ~10 tonnes of propellant for the tanker to land.  In this case, 15.7 kN thrust is needed to achieve 9.8 mm/s2, using 5.7 kg/s, or roughly 20 tonnes/hour.

Then it depends on how long a propellant transfer takes (I'm guessing they will aim for around 30-90 minutes for a full 1200 tonne transfer), and the specific impulse of the RCS thrusters.  And "milli-g" is propably just order-of-magnitude correct, so our calculations could be off by a factor 3, or even a factor 5, in either direction...

Quote
As to leaving LEO, you don't have to burn in a prograde direction.  Do the burn 90 degrees from prograde and so as to not change perigee or apogee so that you only create slight inclination change.

My point with that remark was just that 0.1 g of settling acceleration would not be reasonable to use.  Just 10 minutes of thrust at ~100 cm/s2 gives 600 m/s Δv, which would cause a 4.4° inclination change (if I'm calculating that right), or raise the orbit by several hundred kilometers.  It would also use somewhere between 72 and 340 tonnes of propellant (depending on if tanks are almost empty or almost full).

At ~1 cm/s2 (0.001 g) on the other hand, a full hour of thrusting is just 36 m/s Δv.  You definitely need to take heed of it in your orbital calculations, but it's not going to give a huge change in either altitude or inclination.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 10/16/2020 12:37 am
Then it depends on how long a propellant transfer takes (I'm guessing they will aim for around 30-90 minutes for a full 1200 tonne transfer), and the specific impulse of the RCS thrusters.  And "milli-g" is propably just order-of-magnitude correct, so our calculations could be off by a factor 3, or even a factor 5, in either direction...
A quick search shows that 1e-4 g and possibly as low as 1e-5 g is considered enough for propellant settling. Centaur apparently does 8e-5 g for longer coasts to keep the propellant settled to simplify handling and lower boil-off. I am guessing that one of the major points of interest will be what transfer rates require what settling accelerations. As has been pointed out settling of propellants is standard but large volume transfers without an running engine is not.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Morgun on 10/16/2020 02:23 am
Is there any reason they couldn't just spin the starships and use centripetal acceleration instead of continuous thrust?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: jabe on 10/16/2020 02:31 am
Is there any reason they couldn't just spin the starships and use centripetal acceleration instead of continuous thrust?
direction of fluid transfer would  go away from middle ..want from "left to right"
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 10/16/2020 12:26 pm
Is there any reason they couldn't just spin the starships and use centripetal acceleration instead of continuous thrust?
It's been my painful experience that almost any time I use the word 'just' to describe a solution to a problem that experts see as complicated, there ends up being a reason why they didn't do the thing that seemed obvious and simple to me.

It's possible there will be propellant transfer solutions at some point that involve spinning a pair of docked vehicles along one axis or another, but the latest planned method they've shared with us for refueling Starship involves a very modest ullage burn so I have to assume the mechanical/procedural challenges for that are less than they'd be for something involving spinning, at least right now.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 10/16/2020 12:47 pm
That's my understanding, yes, though they can potentially stop the acceleration one the propellant is settled and just use pressure & surface tension to keep it that way.

This is one of the benefits of using the primary propellants in your RCS system though; you don't have to worry about running out of RCS fuel from running them for a long time.

I suspect (based on nothing at all) that they'll use electric pumps to speed up the transfer as well.


So no fundamental issues to overcome, but you are working in micro-gravity with cryogenic propellants going through a full day/night cycle every 90 minutes, while trying to minimise losses. No-one has done that before, so you can't just dismiss it as trivial.

Except just about every time a second stage has been relight.
So what's your take on this then? NASA are simply so totally incompetent that they have to pay SpaceX tens of millions to do something trivial?

Well there is still docking 2 SS's together solidly and hooking up the LCH4 and LOX leaklessly and robotically. Doesn't sound too hard.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 10/16/2020 06:33 pm
That's my understanding, yes, though they can potentially stop the acceleration one the propellant is settled and just use pressure & surface tension to keep it that way.

This is one of the benefits of using the primary propellants in your RCS system though; you don't have to worry about running out of RCS fuel from running them for a long time.

I suspect (based on nothing at all) that they'll use electric pumps to speed up the transfer as well.


So no fundamental issues to overcome, but you are working in micro-gravity with cryogenic propellants going through a full day/night cycle every 90 minutes, while trying to minimise losses. No-one has done that before, so you can't just dismiss it as trivial.

Except just about every time a second stage has been relight.
So what's your take on this then? NASA are simply so totally incompetent that they have to pay SpaceX tens of millions to do something trivial?

Well there is still docking 2 SS's together solidly and hooking up the LCH4 and LOX leaklessly and robotically. Doesn't sound too hard.
But that wouldn't match the wording of the NASA award, which was to "transfer 10 metric tons of cryogenic propellant, specifically liquid oxygen, between tanks on a Starship vehicle".

I fully agree that autonomously docking a pair of Starships and transferring hundreds of tons of propellant is a lot harder, but I don't think that's what this award is for.

Like I said up-thread, I think this is NASA giving some money to SpaceX to get access to the data from some of the tests they were planning anyway. And I think they are doing that because they don't have this kind of data yet, and there are a bunch of things they consider still risky about it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cuddihy on 10/17/2020 09:21 pm
So going to try some physics here, don't crucify me if I dropped a unit:

 if you're using N2 cold gas thrusters for settling (Isp =73 s), and you really need a =0.1 g (=0.981 m/s) of accel to get proper settling & transfer, for a Starship of 140 t (w/ landing propellant aboard), you're looking at a huge amount of thrust (F=ma=137 kN). No way the current RCS can provide that. At an ISP of only 73 sec you're looking at a mass flow rate of F/(Isp*g0) = 137,000N/(73s*9.81m/s2) = 191 kg/s or ~ 10 t over the course of a minute!

Clearly either the transfer needs to be much done faster, at a much lower settling Gs, and with a much higher ISP thruster. Or maybe my math is wildly off.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thunderscreech on 10/17/2020 10:01 pm
.1G sounds like a lot, what's the source of that?  1/10 of that should be sufficient, and then once it's settled against the inlet , a pressure differential should move the liquid with alacricity, right? 

If .01G (.098 m/s^2) is the acceleration and the LOX tank is 20 meters tall, that's just over 20 seconds to settle the tank.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/17/2020 10:41 pm
Yes, .1g is a lot. The S-IVB had a mass at stage sep of around 650 tonnes and used 2 x 15.1 kN solid ullage motors (reduced from 3x on early test flights), ie a settling acceleration of ~0.005g.

Now granted the S-IVB isn't in free-fall very long, and the tank is almost full. But if anything I think that calls for lower levels of acceleration. Remember in microgravity you only need to overcome the liquid's surface tension. You don't want your liquid to slosh around after hitting the bottom dome, so ideally the propellant will fall nice-and-slow so that surface tension effects can damp the residual sloshing. If it falls quickly (ie high settling acceleration) then the surface tension effects will be swamped.

Without the influence of surface tension it doesn't really matter what your settling acceleration is. The liquid will behave exactly the same except in "slow motion," and since fuel use is acceleration * time it becomes invariant. If you have half the acceleration and thrust then you need to burn for twice as long, so it always takes the same amount of fuel.

Surface tension breaks this mathematical invariant, since it means that propellant no longer settles in exactly the same way (except for being "time stretched") under different acceleration levels. If surface tension is relatively weak, this effect goes to zero. This is what leads me to hypothesize that the only way to "cheat the invariant" will be to use very low ullage acceleration, such that ullage forces and surface tension forces become within the same order-of-magnitude.

I'm presuming here that these non-linearities (ie those caused by operating in the low-thrust, surface tension-dominated regime) will be beneficial, but perhaps not. If these non-linearities turn out to be detrimental instead than you still only need sufficient acceleration to avoid that regime, and beyond that you again run into the same invariant where it doesn't matter how big your ullage thruster is. At that point you're just balancing thruster mass with the amount of time it takes for on-orbit refueling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Pueo on 10/17/2020 11:08 pm
A quick search shows that 1e-4 g and possibly as low as 1e-5 g is considered enough for propellant settling. Centaur apparently does 8e-5 g for longer coasts to keep the propellant settled to simplify handling and lower boil-off. I am guessing that one of the major points of interest will be what transfer rates require what settling accelerations. As has been pointed out settling of propellants is standard but large volume transfers without an running engine is not.

Why does settled propellant have a lower boil-off rate than non-settled propellant?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/17/2020 11:37 pm
A quick search shows that 1e-4 g and possibly as low as 1e-5 g is considered enough for propellant settling. Centaur apparently does 8e-5 g for longer coasts to keep the propellant settled to simplify handling and lower boil-off. I am guessing that one of the major points of interest will be what transfer rates require what settling accelerations. As has been pointed out settling of propellants is standard but large volume transfers without an running engine is not.

Why does settled propellant have a lower boil-off rate than non-settled propellant?

I wondered about this too--which doesn't prevent me from making a guess:  Settled propellant has a smaller surface area.  We have basically three cases:

1) Prop in contact with a cold (non-illuminated) tank wall.
2) Prop in contact with a warm (illuminated) tank wall.
3) Prop floating around in a blob, not in contact with any walls.

You obviously want to avoid #2 as much as possible, but what about #3?  A blob is not getting any conductive heat transfer, but it is getting radiative heat transfer from warm walls.  If the blob is is in the path of that radiation, it'll absorb it, and the average extensive heat of the propellant increases, which is what you want to avoid.

There are limits to how much blob reduction should help you.  For example, any radiation from a warm wall that impinges on the "top" of the settled prop will be absorbed.  But some of it will just go wall-to-wall, which will heat the walls somewhat, but presumably they're pretty much in equilibrium.

Seems like minimizing surface area as much as possible and putting as much of that surface area as possible in contact with cold walls is the optimal strategy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/18/2020 12:23 am
I haven't seen anybody discussing the mechanics of tail-to-tail docking on this thread.  IMO, the actual prop transfer is trivial compared to the RPOD.  Nobody's even come close to doing precision alignment and hard dock of two 9m-diameter surfaces before.  And then there's the not inconsiderable issue that minor errors can lean to pranging engine bells and fueling lines.

Three geometrical possibilities:

1) A NDS-like soft/hard capture system that's somehow located on the center axis and magically extends through the small gap at the center of the three SL engines on the thrust puck.  I don't think there's room to do this but, even if you could, the mechanical forces on a single-axis support column that had to deal with two Starships, one of them possibly full to the gills with prop, are likely huge.

2) Put a 5m-ish diameter extensible ring around the outside of the 3-SL engine cluster.  Looking at Rafael's CAD drawings, it looks--at least in theory--like there's room to do this, with the struts to extend the ring just outside the anchors for the gimbals, and just inside the bells of the RVacs.  If all this ring did was soft capture, that might work.  It would then be up to some kind of latch system around the outer mould to complete hard dock.  Biggest problem:  You're going to need a lot of play in a 5m-wide soft capture system unless you can substantially reduce the translation and rotation errors that are the current state of the art.

3) Instead of One Soft Capture Ring to Rule Them All, deploy, say, three separate rings, spaced 120º apart near the outer mould line, and use software to coordinate the capture and damping process.  ISTM that this would be more error-tolerant than option #2, but that's more of a hunch than something on which I've done the math.  At the very least, it reduces the soft capture problem to something IDSS-like, and maybe you could even use the existing implementations with a layer of software to get the three rings to cooperate.

One other thing relevant to this discussion:  You have to engage the fueling lines reliably.  I assume that this occurs after hard dock, but maybe there's something to be said for that being part of the hard dock process itself.

After all of that, transferring prop seems like it would be pretty boring.  Even undocking is scarier than the prop transfer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Redclaws on 10/18/2020 12:40 am
Boring is not the same as trivial.  It’s still a real problem that has to be solved.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wes_wilson on 10/18/2020 02:46 am
I haven't seen anybody discussing the mechanics of tail-to-tail docking on this thread.  IMO, the actual prop transfer is trivial compared to the RPOD.  Nobody's even come close to doing precision alignment and hard dock of two 9m-diameter surfaces before.  And then there's the not inconsiderable issue that minor errors can lean to pranging engine bells and fueling lines.

Three geometrical possibilities:

1) A NDS-like soft/hard capture system that's somehow located on the center axis and magically extends through the small gap at the center of the three SL engines on the thrust puck.  I don't think there's room to do this but, even if you could, the mechanical forces on a single-axis support column that had to deal with two Starships, one of them possibly full to the gills with prop, are likely huge.

2) Put a 5m-ish diameter extensible ring around the outside of the 3-SL engine cluster.  Looking at Rafael's CAD drawings, it looks--at least in theory--like there's room to do this, with the struts to extend the ring just outside the anchors for the gimbals, and just inside the bells of the RVacs.  If all this ring did was soft capture, that might work.  It would then be up to some kind of latch system around the outer mould to complete hard dock.  Biggest problem:  You're going to need a lot of play in a 5m-wide soft capture system unless you can substantially reduce the translation and rotation errors that are the current state of the art.

3) Instead of One Soft Capture Ring to Rule Them All, deploy, say, three separate rings, spaced 120º apart near the outer mould line, and use software to coordinate the capture and damping process.  ISTM that this would be more error-tolerant than option #2, but that's more of a hunch than something on which I've done the math.  At the very least, it reduces the soft capture problem to something IDSS-like, and maybe you could even use the existing implementations with a layer of software to get the three rings to cooperate.

One other thing relevant to this discussion:  You have to engage the fueling lines reliably.  I assume that this occurs after hard dock, but maybe there's something to be said for that being part of the hard dock process itself.

After all of that, transferring prop seems like it would be pretty boring.  Even undocking is scarier than the prop transfer.

I always naively assumed that they'd extend the landing legs on both machines and use gear on the legs to dock "feet to feet". 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/18/2020 04:37 am
I always naively assumed that they'd extend the landing legs on both machines and use gear on the legs to dock "feet to feet".

That's an interesting idea.  How would the feet soak up the residual motions that weren't straight down the x-axis?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: John Santos on 10/18/2020 04:56 am
I always naively assumed that they'd extend the landing legs on both machines and use gear on the legs to dock "feet to feet".

That's an interesting idea.  How would the feet soak up the residual motions that weren't straight down the x-axis?
Agreed, but I think the feet would get pretty grotty after even a single landing, and any docking mechanism on the bottom of the feet would get trashed, so they would have to be replaced or refurbished after each flight.

On the other hand, the legs would have to be pretty strong, extendible, able to handle a fair amount of off-axis force and able to quickly damp unwanted motion, so maybe the docking mechanism could be attached to the side of the legs, next to the feet.  During a fuel-transfer docking, the feet would never make contact.  The legs would extend, and then a secondary docking grapple would extend from near the bottom of each leg and mate with the same on the other Starship.

Or a second set of legs specifically for docking mounted next to the landing legs, if there is room between the vacuum Raptor nozzles.  (By mounting the docking mechanism on the same legs as the landing feet, maybe they could save some mass and space at the cost of additional complexity by sharing the motors, hydraulics or whatever is used to extend and retract them.  SpaceX often seems willing to trade mass for reduced complexity, but on the other hand, super-cooled propellant...) 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: yg1968 on 10/18/2020 04:58 am
On the topic of in-orbit re-fueling, see below:

At 33 minutes of [the video below], Jim Bridenstine made some interesting comments about Starship and more generally about refueling. He said that Starship heavily relies on the ability to transfer cryogenics in LEO for the purpose of taking a system all the way to the Moon. He said that SpaceX's Starship system does not require a fuel depot around the Moon, it requires a depot in orbit around the Earth.

A few minutes later, he said that NASA wants HLS to be sustainable by 2028, and by sustainable he means that it would be reusable and could be refueled. He added that the refueling could be in orbit around the earth or in orbit around the moon (that would be up to the private sector).


https://youtu.be/3VrdzbBoi5U
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TrevorMonty on 10/18/2020 10:51 am
If flight rate is higher enough SpaceX may eventually move to fuel depots, these may not be much more than SS tanker with better insulation or cryocooler.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 10/18/2020 01:07 pm
I always naively assumed that they'd extend the landing legs on both machines and use gear on the legs to dock "feet to feet".

That's an interesting idea.  How would the feet soak up the residual motions that weren't straight down the x-axis?

This seems like a great idea.
Legs extend and have shock absorption. Line up the legs and an additional latch added to the legs would grab the other leg on the other ship and hold them in position.

Have robotic "snakes" to mate the propellant. Tesla already has snakes for charging.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: meekGee on 10/18/2020 03:22 pm
I always naively assumed that they'd extend the landing legs on both machines and use gear on the legs to dock "feet to feet".

That's an interesting idea.  How would the feet soak up the residual motions that weren't straight down the x-axis?

This seems like a great idea.
Legs extend and have shock absorption. Line up the legs and an additional latch added to the legs would grab the other leg on the other ship and hold them in position.

Have robotic "snakes" to mate the propellant. Tesla already has snakes for charging.
Agreed the legs are a brilliant idea.

Once docked though, no need for snakes.  The spatial relationship is known, so a straight tube will do the job, even if there's a tiny bit of misalignment left.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: libra on 10/18/2020 03:30 pm
Some quick calculations suggest a 250ton SS ( 100 tons propellant) would need around a 2.5kN ullage thruster for .001g acceleration.  If the thruster had a  ISP of around 280 (i.e hypergolic level ISP) it would only consume around 1kg/s.  If the thruster could maintain that .001g for 3600 seconds ( 1 hour), the mass penalty for the burn is then 3.6tons.  Seems reasonable.

At a minimum, you will be pushing 340 tonnes: two Starships (120t each) docked together, and 100 tonnes of propellant.  3.3 kN needed, using 1.2 kg/s (assuming your Isp).

But you will also need to be able to push ~1600 tonnes: two Starships, 150 tonnes of cargo, 1200 tonnes of propellant (from a full accumulation tanker to an empty payload ship, or the last transfer from a tanker to the accumulation tanker), plus ~10 tonnes of propellant for the tanker to land.  In this case, 15.7 kN thrust is needed to achieve 9.8 mm/s2, using 5.7 kg/s, or roughly 20 tonnes/hour.

Then it depends on how long a propellant transfer takes (I'm guessing they will aim for around 30-90 minutes for a full 1200 tonne transfer), and the specific impulse of the RCS thrusters.  And "milli-g" is propably just order-of-magnitude correct, so our calculations could be off by a factor 3, or even a factor 5, in either direction...

Quote
As to leaving LEO, you don't have to burn in a prograde direction.  Do the burn 90 degrees from prograde and so as to not change perigee or apogee so that you only create slight inclination change.

My point with that remark was just that 0.1 g of settling acceleration would not be reasonable to use.  Just 10 minutes of thrust at ~100 cm/s2 gives 600 m/s Δv, which would cause a 4.4° inclination change (if I'm calculating that right), or raise the orbit by several hundred kilometers.  It would also use somewhere between 72 and 340 tonnes of propellant (depending on if tanks are almost empty or almost full).

At ~1 cm/s2 (0.001 g) on the other hand, a full hour of thrusting is just 36 m/s Δv.  You definitely need to take heed of it in your orbital calculations, but it's not going to give a huge change in either altitude or inclination.

just for the sake of comparison... I recently checked "On the shoulders of Titans" to see how much delta-v did the Gemini-Agena "high rides" needed.

https://history.nasa.gov/SP-4203/ch15-3.htm

In the case of Gemini 11, the  standard orbit was 300 km. They pushed the apogee to nearly 1400 km and then back to 300 km.
To achieve that: 280 m/s "up" and 280 m/s "down", total 560 m/s.

Calculator there. https://www.satsig.net/orbit-research/delta-v-geo-injection-calculator.htm
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 10/18/2020 04:59 pm
I haven't seen anybody discussing the mechanics of tail-to-tail docking on this thread.  IMO, the actual prop transfer is trivial compared to the RPOD.  Nobody's even come close to doing precision alignment and hard dock of two 9m-diameter surfaces before.  And then there's the not inconsiderable issue that minor errors can lean to pranging engine bells and fueling lines.

(tbellman: Quote snipped)

I always naively assumed that they'd extend the landing legs on both machines and use gear on the legs to dock "feet to feet".

I don't remember if it has been in this thread, but there have been discussions about the propellant transfer docking on these forums.

SuperHeavy and Starship will have some way of mating and holding on to each other during ascent.  Most likely, in my opinion, is for that mechanism to be used during tail-to-tail docking of two Starships as well.  The mechanisms on SuperHeavy and Starship would be identical and symmetric, and can then be used for both purposes.  (In theory, you could then also put a SuperHeavy upside down on top of another SuperHeavy and have them latch on to each other, but why one would do that is beyond me...)  As for exactly how that mechanism will work, I don't think we have any good information.  But I don't think the feet will be involved.

And some kind of quick disconnect on the propellant tanking lines.  I suspect those will connect separately, after the docking.  I.e, they would still be separated by a few decimeters, and then lowered towards each other until the QDs connect.  The QDs are probably much more sensitive to mechanical damage than the docking mechanism, so you want to avoid them touching each other during the docking maneuver, as a slight misalignment between the ships could otherwise damage the QDs.

Likewise, I don't think we have anything concrete about the details of how those QDs will work, or how the tanking lines will be lowered towards each other.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cuddihy on 10/18/2020 05:45 pm
.1G sounds like a lot, what's the source of that?  1/10 of that should be sufficient, and then once it's settled against the inlet , a pressure differential should move the liquid with alacricity, right? 

If .01G (.098 m/s^2) is the acceleration and the LOX tank is 20 meters tall, that's just over 20 seconds to settle the tank.

I actually got the 0.1G from higher up the thread.

With my numbers above, but 0.01G, that's still 14 kN for 20 seconds, which is 14,000N/(73s*9.81m/s2)*20 sec ~ 390 kg of nitrogen settling gas. That's not peanuts either, but definitely doable, and probably lighter than developing a biprop RCS thruster for the purpose.

But you still have to pressurize the tank at that point, so some kind of heater / engine is probably needed, even for the 10 ton test that NASA is paying for.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/18/2020 07:49 pm
I don't remember if it has been in this thread, but there have been discussions about the propellant transfer docking on these forums.

SuperHeavy and Starship will have some way of mating and holding on to each other during ascent.  Most likely, in my opinion, is for that mechanism to be used during tail-to-tail docking of two Starships as well.  The mechanisms on SuperHeavy and Starship would be identical and symmetric, and can then be used for both purposes.  (In theory, you could then also put a SuperHeavy upside down on top of another SuperHeavy and have them latch on to each other, but why one would do that is beyond me...)  As for exactly how that mechanism will work, I don't think we have any good information.  But I don't think the feet will be involved.

And some kind of quick disconnect on the propellant tanking lines.  I suspect those will connect separately, after the docking.  I.e, they would still be separated by a few decimeters, and then lowered towards each other until the QDs connect.  The QDs are probably much more sensitive to mechanical damage than the docking mechanism, so you want to avoid them touching each other during the docking maneuver, as a slight misalignment between the ships could otherwise damage the QDs.

Likewise, I don't think we have anything concrete about the details of how those QDs will work, or how the tanking lines will be lowered towards each other.

I tend to agree that the feet are going to be dicey to use for this.  I also agree that repurposing the QDs for use as the hard-dock mechanism makes sense, especially since the hard-dock doesn't need to hold pressure (except in the fueling lines themselves, of course).

The real issue is soft capture.  That's not only a question of 6 degree-of-freedom shock absorption but also of 6 degree-of-freedom position correction.  We have two general classes of systems that have been developed: non-androgynous probe-and-drogue-like systems (pretty much everything up through Apollo-Soyuz), and androgynous systems with a standoff soft-capture ring (APAS-xx, IDSS).  There is of course no reason why SpaceX can't invent a new way of doing things, but this is a hard problem with a lot of fairly well-developed technology behind it.  I suspect that they'll want to repurpose some of that tech.

Probe-and-drogue systems aren't androgynous, which would be a real problem for a docking between two generic tankers (which is likely to be much more common during prop accumulation than docking between a tanker and a payload Starship).  So we're left with something IDSS-like.

I don't think that Starship needs 1m-wide soft capture systems if it's using 3 of them spaced out along the circumference of the skirt, but it likely will need the standoff capability (perhaps with an even longer standoff to ensure that all 3 pairs are soft-latched before they begin the damping process).

After soft-capture, you have to damp out the residual motions.  We're dealing with much larger vehicles, with even larger moments of inertia, so the loads are going to be quite a bit higher.  On the other hand, if you're got 3 soft capture systems, they can split the load to some extent.

Retraction should be straightforward.  I'd expect to see the use of IDSS-like guide pins to do fine alignment.  Only then will the quick disconnect system become a quick reconnect.  Once all of that is done, then the fueling connectors can extend and engage.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/18/2020 08:18 pm
.1G sounds like a lot, what's the source of that?  1/10 of that should be sufficient, and then once it's settled against the inlet , a pressure differential should move the liquid with alacricity, right? 

If .01G (.098 m/s^2) is the acceleration and the LOX tank is 20 meters tall, that's just over 20 seconds to settle the tank.

I actually got the 0.1G from higher up the thread.

With my numbers above, but 0.01G, that's still 14 kN for 20 seconds, which is 14,000N/(73s*9.81m/s2)*20 sec ~ 390 kg of nitrogen settling gas. That's not peanuts either, but definitely doable, and probably lighter than developing a biprop RCS thruster for the purpose.

But you still have to pressurize the tank at that point, so some kind of heater / engine is probably needed, even for the 10 ton test that NASA is paying for.

The 1E-4 to 1E-5 m/s² range, per the Centaur cites above, sounds more reasonable. Settling time isn't only a function of how long it takes for the "highest" blob to "land", but also how much of a splash it makes.  Then you need to wait for the sloshing to abate as well.  Lower accelerations = lower impact velocities = less energy dissipation.  For a 20m tank, the free-flight of the highest blob will take 10.5 minutes, but it'll only impact at 64 mm/s.

For a top-off with a max payload, that's 240t + 100t + 1200t = 1540t of mass, so 1E-4 m/s² needs no more than 154N of thrust, so the cold-gas system has to provide 0.22kg/s

Note that you have to maintain the settling thrust throughout the entire prop transfer.  I have no clue what the fill rate should be.  As a SWAG, I'm going to use 500kg/s, so a worst-case 1200t takes 40 minutes.  So:

1) Settling time is small compared to prop transfer time.
2) You need roughly 700kg of cold gas.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 10/18/2020 10:25 pm
For the capture and docking process you need somewhere to put the mechanism.

You're going to have 6 dirty great cargo pods inside the skirt area.....
Why not use part of that volume for the docking mechanism?

With six equally spaced around the circumference you can have 3 'male' and 3 'female' mechanisms.
When 2 SS's are tail to tail this gives you the 180o relative position we've seen in all renders for fueling so far.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: docmordrid on 10/19/2020 03:40 am
I can't help but think SpaceX may leverage Tesla's "snake" combined with a quick connect/disconnect....

https://youtu.be/uMM0lRfX6YI
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/19/2020 06:08 am
For the capture and docking process you need somewhere to put the mechanism.

You're going to have 6 dirty great cargo pods inside the skirt area.....
Why not use part of that volume for the docking mechanism?

With six equally spaced around the circumference you can have 3 'male' and 3 'female' mechanisms.
When 2 SS's are tail to tail this gives you the 180o relative position we've seen in all renders for fueling so far.

There are only 3 cargo pods, if they still exist at all.  The other 3 spots have RVacs in them.  But less is more: you only need enough to be able to stabilize and control both Starships in all six degrees of freedom.

Three capture devices work fine, with the arrangement shown, assuming that there are both fill and drain lines for both LCH4 and LOX:
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: paulthew on 10/19/2020 08:46 am
Has anybody proposed using a heat pump to make fuel transfers in space happen?  If two tanks in zero gravity are connected and they are different temperatures, all of the fuel should end up in the cooler tank.  The greater the heat difference, the quicker the transfer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: SkyRate on 10/19/2020 12:08 pm
Has anybody proposed using a heat pump to make fuel transfers in space happen?  If two tanks in zero gravity are connected and they are different temperatures, all of the fuel should end up in the cooler tank.  The greater the heat difference, the quicker the transfer.
Two big assumptions: That you can heat it a lot and that you can contain the pressure that that creates. Venting the gas out of the target tank + optionally pumping is way safer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: livingjw on 10/19/2020 01:18 pm
Has anybody proposed using a heat pump to make fuel transfers in space happen?  If two tanks in zero gravity are connected and they are different temperatures, all of the fuel should end up in the cooler tank.  The greater the heat difference, the quicker the transfer.
Two big assumptions: That you can heat it a lot and that you can contain the pressure that that creates. Venting the gas out of the target tank + optionally pumping is way safer.

No heating or venting is required. Using a low pressure electric pump allows for reasonable transfer rates without venting.

John
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 10/19/2020 01:22 pm
I tend to agree that the feet are going to be dicey to use for this.  I also agree that repurposing the QDs for use as the hard-dock mechanism makes sense, especially since the hard-dock doesn't need to hold pressure (except in the fueling lines themselves, of course).

Just to make sure we are not talking past each other:  In my post, in the parts where I used the words "quick disconnect" (or "QD"), I was specifically talking about connectors for the propellant lines.  The mating mechanism for holding together SuperHeavy and Starship during launch are of course also a kind of quick disconnect, and I should probably have been clearer in my wording and used "prop QD" or something like that when I was talking about the propellant lines.

And I think it is very unlikely that the propellant line QDs will be used for holding the ships together.  Rather, they would only connect and engage after the two ships are securely clamped together by the docking mechanisms, and all relative motions have been cancelled.  They would also be designed to disconnect, close their valves and retract when subject to fairly small forces, so as to minimize risk of rupturing the propellant lines.

I suspect that in your post, when you were talking about repurposing the QDs for hard docking, you were talking about the mechanisms for securely holding together SH and SS, right?

As for the soft capture part of docking, I don't have any insights or good guesses on how that will be handled.  Other than that it could probably be useful during the stacking of a Starship on top of SuperHeavy on the ground as well; they also need to be aligned properly.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 10/19/2020 01:23 pm
Has anybody proposed using a heat pump to make fuel transfers in space happen?  If two tanks in zero gravity are connected and they are different temperatures, all of the fuel should end up in the cooler tank.  The greater the heat difference, the quicker the transfer.
Two big assumptions: That you can heat it a lot and that you can contain the pressure that that creates. Venting the gas out of the target tank + optionally pumping is way safer.
Doesnt that requir a return line? I thought SpaceX was aiming to just have 1 transfer line (also used for initial fueling) and a vent line, to save weight
No heating or venting is required. Using a low pressure electric pump allows for reasonable transfer rates without venting.

John
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 10/19/2020 01:33 pm
For the capture and docking process you need somewhere to put the mechanism.

You're going to have 6 dirty great cargo pods inside the skirt area.....
Why not use part of that volume for the docking mechanism?

With six equally spaced around the circumference you can have 3 'male' and 3 'female' mechanisms.
When 2 SS's are tail to tail this gives you the 180o relative position we've seen in all renders for fueling so far.

There are only 3 cargo pods, if they still exist at all.  The other 3 spots have RVacs in them.  But less is more: you only need enough to be able to stabilize and control both Starships in all six degrees of freedom.

Three capture devices work fine, with the arrangement shown, assuming that there are both fill and drain lines for both LCH4 and LOX:
I don't think this will work. They need to have bilateral symmetry. Starships will approach each out back to back and inverted so the heatshields of each will be on different sides
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 10/19/2020 01:35 pm
For the capture and docking process you need somewhere to put the mechanism.

You're going to have 6 dirty great cargo pods inside the skirt area.....
Why not use part of that volume for the docking mechanism?

With six equally spaced around the circumference you can have 3 'male' and 3 'female' mechanisms.
When 2 SS's are tail to tail this gives you the 180o relative position we've seen in all renders for fueling so far.

There are only 3 cargo pods, if they still exist at all.  The other 3 spots have RVacs in them.  But less is more: you only need enough to be able to stabilize and control both Starships in all six degrees of freedom.

Three capture devices work fine, with the arrangement shown, assuming that there are both fill and drain lines for both LCH4 and LOX:
I don't think this will work. They need to have bilateral symmetry. Starships will approach each out back to back and inverted so the heatshields of each will be on different sides
Why inverted? Back to back is enough to flip connection points, with, say, a male connector on the left and a female connector on the right.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Pete on 10/19/2020 01:51 pm
About propellant needs for settling the prop and transfer.
Surely once you are actually pumping the liquids across, the rather massive shift in center of mass will remove the need for any further sustained ullage thrust? You are moving half a ton of mass over a distance of more than 30 meters, *per second*. This will impart a very significant apparent "thrust" to the tank being pumped out, resulting in all the settling you could possibly need.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 10/19/2020 02:13 pm
For the capture and docking process you need somewhere to put the mechanism.

You're going to have 6 dirty great cargo pods inside the skirt area.....
Why not use part of that volume for the docking mechanism?

With six equally spaced around the circumference you can have 3 'male' and 3 'female' mechanisms.
When 2 SS's are tail to tail this gives you the 180o relative position we've seen in all renders for fueling so far.

There are only 3 cargo pods, if they still exist at all.  The other 3 spots have RVacs in them.  But less is more: you only need enough to be able to stabilize and control both Starships in all six degrees of freedom.

Three capture devices work fine, with the arrangement shown, assuming that there are both fill and drain lines for both LCH4 and LOX:
I don't think this will work. They need to have bilateral symmetry. Starships will approach each out back to back and inverted so the heatshields of each will be on different sides
Why inverted? Back to back is enough to flip connection points, with, say, a male connector on the left and a female connector on the right.

Maybe because you need to mate CH4 to CH4 and LOX to LOX? Which maybe on opposite sides of the ship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 10/19/2020 04:25 pm
About propellant needs for settling the prop and transfer.
Surely once you are actually pumping the liquids across, the rather massive shift in center of mass will remove the need for any further sustained ullage thrust? You are moving half a ton of mass over a distance of more than 30 meters, *per second*. This will impart a very significant apparent "thrust" to the tank being pumped out, resulting in all the settling you could possibly need.
The problem is each of those half ton masses is going to impart and acceleration forward, but then it will hit the reverse bulkhead or liquid surface and impart an equal but opposite acceleration in the opposite direction. If they could get any useful acceleration from just pumping liquid around a ship they would use it to propel the ship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 10/19/2020 05:39 pm
About propellant needs for settling the prop and transfer.
Surely once you are actually pumping the liquids across, the rather massive shift in center of mass will remove the need for any further sustained ullage thrust? You are moving half a ton of mass over a distance of more than 30 meters, *per second*. This will impart a very significant apparent "thrust" to the tank being pumped out, resulting in all the settling you could possibly need.
The problem is each of those half ton masses is going to impart and acceleration forward, but then it will hit the reverse bulkhead or liquid surface and impart an equal but opposite acceleration in the opposite direction. If they could get any useful acceleration from just pumping liquid around a ship they would use it to propel the ship.

No, Pete is correct.  As the propellant is moved (pumped) in one direction, everything else, i.e. the ships themselves, must shift in the opposite direction.  Centre of mass of a system (in this case ships+propellant) cannot move, unless there is some external force acting on it.

As soon as you stop pumping, though, the ships will also stop moving.  The end result will then be that the propellant has moved some distance in the -X direction (relative to the tanker) while the ships' hulls have moved some distance in the +X direction.  The centre of mass for the entire system will not have moved at all, however, which is why you can't use pumping for propulsion.

(And movement here is relative to an observer in the same trajectory as the ships.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: cdebuhr on 10/19/2020 06:08 pm
About propellant needs for settling the prop and transfer.
Surely once you are actually pumping the liquids across, the rather massive shift in center of mass will remove the need for any further sustained ullage thrust? You are moving half a ton of mass over a distance of more than 30 meters, *per second*. This will impart a very significant apparent "thrust" to the tank being pumped out, resulting in all the settling you could possibly need.
The problem is each of those half ton masses is going to impart and acceleration forward, but then it will hit the reverse bulkhead or liquid surface and impart an equal but opposite acceleration in the opposite direction. If they could get any useful acceleration from just pumping liquid around a ship they would use it to propel the ship.

No, Pete is correct.  As the propellant is moved (pumped) in one direction, everything else, i.e. the ships themselves, must shift in the opposite direction.  Centre of mass of a system (in this case ships+propellant) cannot move, unless there is some external force acting on it.

As soon as you stop pumping, though, the ships will also stop moving.  The end result will then be that the propellant has moved some distance in the -X direction (relative to the tanker) while the ships' hulls have moved some distance in the +X direction.  The centre of mass for the entire system will not have moved at all, however, which is why you can't use pumping for propulsion.

(And movement here is relative to an observer in the same trajectory as the ships.)
I had Pete's very same tough yesterday, and nearly posted about it, but I couldn't shake the feeling that I was missing something important.  Now I think I see the problem.  While its true that moving prop will shift the COG, resulting in overall movement of the mated ships, that alone isn't good enough.  To ensure that fluids stay settled, the ships need to be continuously accelerating, otherwise any deviation from a perfectly smooth transfer risks bumping the fluids away from the lower bulkhead.  Unless you're actually ramping the pumping rate over time, this requirement would not be met.  Now I suppose you could ramp the pumping rate - whether or not that would be enough to make this work, I won't comment on, as I've not done the math, but it's slightly more complicated than what was initially proposed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/19/2020 07:33 pm
I don't think this will work. They need to have bilateral symmetry. Starships will approach each out back to back and inverted so the heatshields of each will be on different sides

It does have bilateral symmetry--about the y-axis.  Note that the second image hasn't been folded to be back-to-back yet.  When you do that, everything matches up.

But the reason I had separate fill and drain lines is that I couldn't get a double rotation-invariant version to work without them.  My problem was that I hadn't put them on the axis of symmetry.  When I do that, I get this:

Update:  Note that there's no particular reason whey the soft captures can't hug the inner mould line and have the LCH4 line be coaxial with it, other than I suspect you really want that sucker welded to the wall.  If that's the case, than the capture device needs to be mounted inboard of it.  I suspect either way can be made to work.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Apollo-phill on 10/19/2020 07:45 pm
Just side tracking but didn't Soviets/Russians undertake propellant transfer during some early dual flight Soyuz spacecraft  missions ?

And, if so, any published data/results on what was achieved and, any relevance to this topic ?

Also, may have been noted earlier - why transfer propellant from ship to ship ?
Why not undock crew/payload stage/module and just dock with the  upcoming "propellant 'tanker' " which then becomes the "celestial target bound" starship ? The now , payload less , empty starship returns to landing site. Just a thought.

Phill


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/19/2020 08:02 pm
Just side tracking but didn't Soviets/Russians undertake propellant transfer during some early dual flight Soyuz spacecraft  missions ?

And, if so, any published data/results on what was achieved and, any relevance to this topic ?

Also, may have been noted earlier - why transfer propellant from ship to ship ?
Why not undock crew/payload stage/module and just dock with the  upcoming "propellant 'tanker' " which then becomes the "celestial target bound" starship ? The now , payload less , empty starship returns to landing site. Just a thought.

Phill

The Russians transfer hydrazine and NTO using a bladder that lines the inside of the tank.  As the bladder contracts, it provides the pressure to move the prop from one tank to another.

For many Starship missions, you need more propellant than a tanker can haul up in a single launch, so swapping payload sections doesn't get the job done.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: RDoc on 10/19/2020 08:03 pm
Just side tracking but didn't Soviets/Russians undertake propellant transfer during some early dual flight Soyuz spacecraft  missions ?

And, if so, any published data/results on what was achieved and, any relevance to this topic ?

Also, may have been noted earlier - why transfer propellant from ship to ship ?
Why not undock crew/payload stage/module and just dock with the  upcoming "propellant 'tanker' " which then becomes the "celestial target bound" starship ? The now , payload less , empty starship returns to landing site. Just a thought.

Phill
My understanding is that there will be several tanker loads per flight and Starship doesn't really have a separate payload stage, it's all one piece.
Transferring the payload(s) in orbit sounds pretty tricky compared to pumping liquid in any case.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DistantTemple on 10/19/2020 08:07 pm
Just side tracking but didn't Soviets/Russians undertake propellant transfer during some early dual flight Soyuz spacecraft  missions ?

And, if so, any published data/results on what was achieved and, any relevance to this topic ?

Also, may have been noted earlier - why transfer propellant from ship to ship ?
Why not undock crew/payload stage/module and just dock with the  upcoming "propellant 'tanker' " which then becomes the "celestial target bound" starship ? The now , payload less , empty starship returns to landing site. Just a thought.

Phill
It takes several tankers to "fill" an SS sufficiently for all expected tasks... except LEO perhaps. So orbiting and lift tankers (same or different) would still have to transfer fuel to prepare one, as you say "celestial target bound" starship"
Edit; great this all answering at once! ;-(
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 10/19/2020 08:23 pm
The problem with using tranfereed fuel CoM shift as ullage, is that the COM movement is measured in M/s, and ullage is measured in M/s/s. Once the fuel is being transfered at a steady rate, you no longer get any acceleration, just a steady shift in CoM.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/19/2020 09:21 pm
The problem with using tranfereed fuel CoM shift as ullage, is that the COM movement is measured in M/s, and ullage is measured in M/s/s. Once the fuel is being transfered at a steady rate, you no longer get any acceleration, just a steady shift in CoM.

ISTM the real issue is whether you can reduce ullage thrust (and therefore acceleration) during the transfer, after settling is complete.  You're going to go from free-floating blobs, to splashes as the blobs impact the main body of prop along the back wall, to lots of sloshing, to a relatively smooth, stable surface.  I think that more acceleration will shorten this dynamic period.

Once the stable mode has been achieved, pumping (or pushing, via ullage pressure) prop out shouldn't make a huge change in stability.  It's quite possible that stability can be maintained with ullage acceleration reduced down into the 1E-5m/s² range, which will reduce the amount of cold gas or methalox required to maintain it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: livingjw on 10/20/2020 03:14 am
I don't think this will work. They need to have bilateral symmetry. Starships will approach each out back to back and inverted so the heatshields of each will be on different sides

It does have bilateral symmetry--about the y-axis.  Note that the second image hasn't been folded to be back-to-back yet.  When you do that, everything matches up.

But the reason I had separate fill and drain lines is that I couldn't get a double rotation-invariant version to work without them.  My problem was that I hadn't put them on the axis of symmetry.  When I do that, I get this:

Update:  Note that there's no particular reason whey the soft captures can't hug the inner mould line and have the LCH4 line be coaxial with it, other than I suspect you really want that sucker welded to the wall.  If that's the case, than the capture device needs to be mounted inboard of it.  I suspect either way can be made to work.

For this diagram,  I assume you are venting the receiving SS's ullage gas?

John
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/20/2020 04:39 am
I don't think this will work. They need to have bilateral symmetry. Starships will approach each out back to back and inverted so the heatshields of each will be on different sides

It does have bilateral symmetry--about the y-axis.  Note that the second image hasn't been folded to be back-to-back yet.  When you do that, everything matches up.

But the reason I had separate fill and drain lines is that I couldn't get a double rotation-invariant version to work without them.  My problem was that I hadn't put them on the axis of symmetry.  When I do that, I get this:

Update:  Note that there's no particular reason whey the soft captures can't hug the inner mould line and have the LCH4 line be coaxial with it, other than I suspect you really want that sucker welded to the wall.  If that's the case, than the capture device needs to be mounted inboard of it.  I suspect either way can be made to work.

For this diagram,  I assume you are venting the receiving SS's ullage gas?

John

I would assume so.  I'd further assume that the vents were an open system and therefore don't have to match up during the docking operations.  Is there any reason why the the regular pressure-relief vents wouldn't work fine for keeping the pressure low enough on the receiving tanks?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: kkattula on 10/20/2020 06:32 am
...
For this diagram,  I assume you are venting the receiving SS's ullage gas?

John

I would assume so.  I'd further assume that the vents were an open system and therefore don't have to match up during the docking operations.  Is there any reason why the the regular pressure-relief vents wouldn't work fine for keeping the pressure low enough on the receiving tanks?

If you are venting the ullage gas, it would make sense to do it through nozzles to provide the settling acceleration. 

However, since the tanker needs to replace the volume of liquid with gas, a vent pipe in the opposite direction seems reasonable.


It occurs to me that a Starship would not have dedicated vent and fill lines. Rather it would have a line to the top of each tank and one to the bottom of each, which would act as the fill/drain or vent/gas-fill, depending on whether the SS was the tanker or tankee.

These lines only need to match up to their opposite, of the same propellant type, when docked. i.e. Bottom LOX on the tanker to Top LOX on the tankee, and vice versa:
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Overtone on 10/20/2020 10:26 am
But the reason I had separate fill and drain lines is that I couldn't get a double rotation-invariant version to work without them.  My problem was that I hadn't put them on the axis of symmetry.  When I do that, I get this:

Looks to me like any axis of symmetry is going to have the bell of a vacuum raptor at one end or the other.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/20/2020 07:35 pm
But the reason I had separate fill and drain lines is that I couldn't get a double rotation-invariant version to work without them.  My problem was that I hadn't put them on the axis of symmetry.  When I do that, I get this:

Looks to me like any axis of symmetry is going to have the bell of a vacuum raptor at one end or the other.

Yup, that is definitely a problem.  Two possible solutions:

1) Define an axis of docking/fueling symmetry separate from the rest of the vehicle.  Kinda weird looking, and likely requires moving various stuff (like landing legs and the cargo pods) around, but I'd be surprised if that weren't in the offing before Starship takes on its final orbital form.

2) Move the fueling lines medial, so they're in the gap between RaptorSLs and the RaptorVacs.  However, this seems structurally iffy, and a huge pain for stacking the Starship on the SuperHeavy.

Both options attached, using Rafael's CAD drawing as a starting place.  Rafael, how sure are you about your x- and y-axis placement?  Also, are the circles around the mould line just the guide pins that show up in some of the photos, or are the existing fuel and vent lines in there somewhere?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/20/2020 08:10 pm
If you are venting the ullage gas, it would make sense to do it through nozzles to provide the settling acceleration. 

However, since the tanker needs to replace the volume of liquid with gas, a vent pipe in the opposite direction seems reasonable.


It occurs to me that a Starship would not have dedicated vent and fill lines. Rather it would have a line to the top of each tank and one to the bottom of each, which would act as the fill/drain or vent/gas-fill, depending on whether the SS was the tanker or tankee.

These lines only need to match up to their opposite, of the same propellant type, when docked. i.e. Bottom LOX on the tanker to Top LOX on the tankee, and vice versa:

That's pretty good.  Some comments:

1) If you're hooking a bottom drain to a top fill, then that pipe is acting as a vent part of the time and as a prop transport line at other times.  That requires both gas and liquid valves at the tank head, with the gas valve usually being a relief valve.  That's probably doable but it's a bit complex.  On the other hand, the receiving tank is going to have the prop settled to the top, so having each opening at both the top and bottom doing double duty is elegant.

2) Good point about using the ullage gas as reaction mass for settling/stability during stability.  But that means that you'd need to make gas to keep the ullage volume at constant volume and pressure (which it would be, if you're just transferring prop from one volume to the other).  Once you've gone to the trouble of an autogenous (presumably electric) heater, maybe it's simpler just to keep the whole venting system distinct from the fill/drain system?

3) This suffers from the same deficiency as my previous attempt, in that there's one pair of lines that's where an RVac is.  I think you can solve that simply by spacing them wider, so they're outside the engine bell.

4) However, note that docking system and the prop lines have to share the same axis of symmetry.  So, per #3, if you have one pair of lines hugging each side of an engine bell, you have to find somewhere along that axis to put your soft capture, and it would have to go in the gap between the SL's and Vacs.  That'll give less damping control during the capture process, because there's less leverage.  Another alternative is simply to rotate the docking axis of symmetry away from the engine axis.  See the picture just up-thread for how that would work, although I only have a single fill/drain line in for each fluid.

5) This of course requires double the lines, and double the couplers at docking.  There's a weight and complexity penalty there, which might be less with just a simple vent manifold that handles both tank ends as needed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 10/20/2020 08:39 pm
If you move the fueling lines inboard (medial) then aren't they likely to interfere with SL Raptor gimbaling?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 10/20/2020 08:50 pm
I just had a thought, currently GSE connections appear to be through a single, multi-piped, quick connect location on the skirt.

If the final version of SS retains this single fueling location then that would explain why the two SS need to be inverted by 180o when tail to tail.

Now if this requires 4 pipes, a fill and vent for each tank then this could work.

When a tanker is delivering fuel under micro-acceleration then the fuel would be exiting from the 'vent' pipe in the top of the tank. It then enters the 'fill' pipe in the bottom of the receiving SS.

It's giving me a headache but is there a 4 pipe arrangement, when mirrored in a docking situation, that would have the 4 pipes align so that the vent and fill pipes for each fuel type line up in an alternating arrangement?
That is, tanker vents to recipient fill pipes?

Damned if I can find the pictures but I seem to recall that the GSE connection currently used had more then two pipes....
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/20/2020 09:04 pm
I just had a thought, currently GSE connections appear to be through a single, multi-piped, quick connect location on the skirt.

If the final version of SS retains this single fueling location then that would explain why the two SS need to be inverted by 180o when tail to tail.

Now if this requires 4 pipes, a fill and vent for each tank then this could work.

When a tanker is delivering fuel under micro-acceleration then the fuel would be exiting from the 'vent' pipe in the top of the tank. It then enters the 'fill' pipe in the bottom of the receiving SS.

It's giving me a headache but is there a 4 pipe arrangement, when mirrored in a docking situation, that would have the 4 pipes align so that the vent and fill pipes for each fuel type line up in an alternating arrangement?
That is, tanker vents to recipient fill pipes?

Damned if I can find the pictures but I seem to recall that the GSE connection currently used had more then two pipes....

I think there are both fill/drain pipes and vents for the two fluids, which makes four lines.

I'm really bad at the whole "do two rotations in your head and describe what happened" thing, but from the pictures I did up-thread I think I convinced myself that rotating 180º about both the x- and z-axes doesn't give you a workable system unless there are lines at both ends of the axis of symmetry.  However, if you have four lines, one in/out pair at one end, and another in/out pair at the other, then the double rotation hooks the "out" of one end to the "in" of the other--which is pretty much what you want.

I'm really liking kkattula's combined lines that can either vent gas or fill/drain fluid.  That requires two pairs of lines as well.

Update:  OK, here's a 4-line version, similar to my first version, but with all the prop and gas flows mapped out.  NOTE:  The y- and z-axes shown here are with respect to the docking and fueling lines, not necessarily the pitch and yaw axes of the vehicle.  They may be but they may not.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 10/20/2020 11:17 pm
I just had a thought, currently GSE connections appear to be through a single, multi-piped, quick connect location on the skirt.

If the final version of SS retains this single fueling location then that would explain why the two SS need to be inverted by 180o when tail to tail.

Now if this requires 4 pipes, a fill and vent for each tank then this could work.

When a tanker is delivering fuel under micro-acceleration then the fuel would be exiting from the 'vent' pipe in the top of the tank. It then enters the 'fill' pipe in the bottom of the receiving SS.

It's giving me a headache but is there a 4 pipe arrangement, when mirrored in a docking situation, that would have the 4 pipes align so that the vent and fill pipes for each fuel type line up in an alternating arrangement?
That is, tanker vents to recipient fill pipes?

Damned if I can find the pictures but I seem to recall that the GSE connection currently used had more then two pipes....

I think there are both fill/drain pipes and vents for the two fluids, which makes four lines.

I'm really bad at the whole "do two rotations in your head and describe what happened" thing, but from the pictures I did up-thread I think I convinced myself that rotating 180º about both the x- and z-axes doesn't give you a workable system unless there are lines at both ends of the axis of symmetry.  However, if you have four lines, one in/out pair at one end, and another in/out pair at the other, then the double rotation hooks the "out" of one end to the "in" of the other--which is pretty much what you want.

I'm really liking kkattula's combined lines that can either vent gas or fill/drain fluid.  That requires two pairs of lines as well.

Update:  OK, here's a 4-line version, similar to my first version, but with all the prop and gas flows mapped out.  NOTE:  The y- and z-axes shown here are with respect to the docking and fueling lines, not necessarily the pitch and yaw axes of the vehicle.  They may be but they may not.

Nice, that's how I envisaged the flows.

Now put the pipe connections in a single location like we currently see on the prototype SS and test stands.

I apologise for mangling your diagrams but below is sort of what I have in mind....
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: livingjw on 10/21/2020 02:42 am
https://forum.nasaspaceflight.com/index.php?topic=50157.msg2123644#msg2123644

John
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/21/2020 03:12 am
I haven't seen anybody discussing the mechanics of tail-to-tail docking on this thread.  IMO, the actual prop transfer is trivial compared to the RPOD.  Nobody's even come close to doing precision alignment and hard dock of two 9m-diameter surfaces before.  And then there's the not inconsiderable issue that minor errors can lean to pranging engine bells and fueling lines.

Three geometrical possibilities:

1) A NDS-like soft/hard capture system that's somehow located on the center axis and magically extends through the small gap at the center of the three SL engines on the thrust puck.  I don't think there's room to do this but, even if you could, the mechanical forces on a single-axis support column that had to deal with two Starships, one of them possibly full to the gills with prop, are likely huge.

2) Put a 5m-ish diameter extensible ring around the outside of the 3-SL engine cluster.  Looking at Rafael's CAD drawings, it looks--at least in theory--like there's room to do this, with the struts to extend the ring just outside the anchors for the gimbals, and just inside the bells of the RVacs.  If all this ring did was soft capture, that might work.  It would then be up to some kind of latch system around the outer mould to complete hard dock.  Biggest problem:  You're going to need a lot of play in a 5m-wide soft capture system unless you can substantially reduce the translation and rotation errors that are the current state of the art.

3) Instead of One Soft Capture Ring to Rule Them All, deploy, say, three separate rings, spaced 120º apart near the outer mould line, and use software to coordinate the capture and damping process.  ISTM that this would be more error-tolerant than option #2, but that's more of a hunch than something on which I've done the math.  At the very least, it reduces the soft capture problem to something IDSS-like, and maybe you could even use the existing implementations with a layer of software to get the three rings to cooperate.

One other thing relevant to this discussion:  You have to engage the fueling lines reliably.  I assume that this occurs after hard dock, but maybe there's something to be said for that being part of the hard dock process itself.

After all of that, transferring prop seems like it would be pretty boring.  Even undocking is scarier than the prop transfer.

#3, except they're not really "rings," just self-aligning latches w integrated soft capture. These would be incorporated into the standard interstage latching mechanism.

Whether N = 3 (as you speculate) or some other number, I'm not sure. F9 originally (v1.0) had 9x latches and 3x pneumatic pushers, while F9 v1.1 switched to 3x combination latches / pushers. I suspect Starship will use N = 6, but it's just a hunch.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/21/2020 03:22 am
I always naively assumed that they'd extend the landing legs on both machines and use gear on the legs to dock "feet to feet".

That's an interesting idea.  How would the feet soak up the residual motions that weren't straight down the x-axis?
Agreed, but I think the feet would get pretty grotty after even a single landing, and any docking mechanism on the bottom of the feet would get trashed, so they would have to be replaced or refurbished after each flight.

On the other hand, the legs would have to be pretty strong, extendible, able to handle a fair amount of off-axis force and able to quickly damp unwanted motion, so maybe the docking mechanism could be attached to the side of the legs, next to the feet.  During a fuel-transfer docking, the feet would never make contact.  The legs would extend, and then a secondary docking grapple would extend from near the bottom of each leg and mate with the same on the other Starship.

Yeah, the legs and the docking mechanism should logically share a single mechanical hardpoint, but not necessarily be the same actual mechanism.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/21/2020 03:28 am
The real issue is soft capture.  That's not only a question of 6 degree-of-freedom shock absorption but also of 6 degree-of-freedom position correction.  We have two general classes of systems that have been developed: non-androgynous probe-and-drogue-like systems (pretty much everything up through Apollo-Soyuz), and androgynous systems with a standoff soft-capture ring (APAS-xx, IDSS).  There is of course no reason why SpaceX can't invent a new way of doing things, but this is a hard problem with a lot of fairly well-developed technology behind it.  I suspect that they'll want to repurpose some of that tech.

Probe-and-drogue systems aren't androgynous, which would be a real problem for a docking between two generic tankers (which is likely to be much more common during prop accumulation than docking between a tanker and a payload Starship).  So we're left with something IDSS-like.

It's perfectly possible to create an androgenous system using probe-and-drogue. You just arrange the probes and drogues on both vehicles such that they mate with a rotated copy of themselves.

Ultimately both concepts rely on the simple idea of an inclined plane / ramp. A drogue cone is just a ramp revolved 360 degrees. The IDSS-alike systems have six ramps arranged in an equilateral triangle (the "pedals").
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/21/2020 03:44 am
But the reason I had separate fill and drain lines is that I couldn't get a double rotation-invariant version to work without them.  My problem was that I hadn't put them on the axis of symmetry.  When I do that, I get this:

Looks to me like any axis of symmetry is going to have the bell of a vacuum raptor at one end or the other.

Yup, that is definitely a problem.  Two possible solutions:

1) Define an axis of docking/fueling symmetry separate from the rest of the vehicle.  Kinda weird looking, and likely requires moving various stuff (like landing legs and the cargo pods) around, but I'd be surprised if that weren't in the offing before Starship takes on its final orbital form.

No stranger to SpaceX. Check out the crazy angle of the F9 v1.0 fairing split line, for instance.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/21/2020 03:52 am
I just had a thought, currently GSE connections appear to be through a single, multi-piped, quick connect location on the skirt.

If the final version of SS retains this single fueling location then that would explain why the two SS need to be inverted by 180o when tail to tail.

Now if this requires 4 pipes, a fill and vent for each tank then this could work.

When a tanker is delivering fuel under micro-acceleration then the fuel would be exiting from the 'vent' pipe in the top of the tank. It then enters the 'fill' pipe in the bottom of the receiving SS.

It's giving me a headache but is there a 4 pipe arrangement, when mirrored in a docking situation, that would have the 4 pipes align so that the vent and fill pipes for each fuel type line up in an alternating arrangement?
That is, tanker vents to recipient fill pipes?

Damned if I can find the pictures but I seem to recall that the GSE connection currently used had more then two pipes....

I think there are both fill/drain pipes and vents for the two fluids, which makes four lines.

I'm really bad at the whole "do two rotations in your head and describe what happened" thing, but from the pictures I did up-thread I think I convinced myself that rotating 180º about both the x- and z-axes doesn't give you a workable system unless there are lines at both ends of the axis of symmetry.  However, if you have four lines, one in/out pair at one end, and another in/out pair at the other, then the double rotation hooks the "out" of one end to the "in" of the other--which is pretty much what you want.

I'm really liking kkattula's combined lines that can either vent gas or fill/drain fluid.  That requires two pairs of lines as well.

Update:  OK, here's a 4-line version, similar to my first version, but with all the prop and gas flows mapped out.  NOTE:  The y- and z-axes shown here are with respect to the docking and fueling lines, not necessarily the pitch and yaw axes of the vehicle.  They may be but they may not.

Nice, that's how I envisaged the flows.

Now put the pipe connections in a single location like we currently see on the prototype SS and test stands.

I apologise for mangling your diagrams but below is sort of what I have in mind....

Note that on Falcon 9 they intentionally put the fuel and oxidizer GSE connections on opposite sides of the vehicle, for obvious safety reasons. IMO Starship will not "un-learn" this safety feature.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/21/2020 05:29 am
Nice, that's how I envisaged the flows.

Now put the pipe connections in a single location like we currently see on the prototype SS and test stands.

I apologise for mangling your diagrams but below is sort of what I have in mind....

That'll work too, but you have to put a soft capture on the axis somewhere.  Probably just spread the two pairs out and for the put the soft capture between them.  (The soft capture is obviously a bone of contention, but I don't think you can manage with only two captures.  You need to be able to apply enough torque to get the 6-degree-of-freedom stabilization.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/21/2020 05:41 am
#3, except they're not really "rings," just self-aligning latches w integrated soft capture. These would be incorporated into the standard interstage latching mechanism.

Whether N = 3 (as you speculate) or some other number, I'm not sure. F9 originally (v1.0) had 9x latches and 3x pneumatic pushers, while F9 v1.1 switched to 3x combination latches / pushers. I suspect Starship will use N = 6, but it's just a hunch.

What do you mean by "integrated soft capture"?

Starship has the same basic problem as any other docking spacecraft.  The reason for the active soft capture rings is because the hard latches aren't self-aligning.  They need the soft system to restrict the problem enough that the guide pins will engage.

I have no clue about the actual hard latches.  I'd think that the staging quick-releases could be engineered to act as hard latches as well, but you want really, really, really reliable staging latches (failure to stage is an LOC event, when/if they get to launching people), so maybe a separate set of IDSS-like hard latches will turn out to be better.  Either way, I don't think we're talking about massive loads on the latches on-orbit; they just need to be strong enough to stabilize the fueling lines in the even that there's some kind of wacky torque on the stack (e.g. a transiently firing thruster).  And they obviously don't need to hold pressure.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/21/2020 05:48 am
Note that on Falcon 9 they intentionally put the fuel and oxidizer GSE connections on opposite sides of the vehicle, for obvious safety reasons. IMO Starship will not "un-learn" this safety feature.

I've been thinking about that issue and it's... not easy to solve.  I can't come up with an arrangement where each guzouta mates up with a guzinta after a y- and z-rotation that doesn't have fuel and oxidizer lines next to each other.  I'm bad at this stuff, so I'm willing to believe I've had a failure of imagination.  You should play with it and see if you can figure it out.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/21/2020 05:56 am
https://forum.nasaspaceflight.com/index.php?topic=50157.msg2123644#msg2123644

John

Yup, after I answered your question up-thread, I figured out what you were driving at.  Looks right, and I agree that you probably need some modest pumps.

There's a tradeoff between reaction mass for settling and how hefty the mass flow has to be through the transfer system.  Assuming they want to get rid of cold gas thrusters and go with methox only (presumably big ones and little ones) you probably want to be able to do a max transfer (approaching 1200t) in enough time that you don't have to stop and repressurize the COPVs.

Question for you:  Do you think the ullage thrust has to be continuous, or can they get away with a series of sufficiently small-impulse-bit burns?  Getting methox to burn reliably at the kind of thrusts needed for settling sounds dicey.  For that matter, could they manage with electric thrusters for settling?  (Because nothing bad ever happens when you do fuel transfers near giant electric fields...)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/21/2020 06:50 am
A silly back-of-napkin:

Prop for both tanker and tankee shouldn't be more than about 1500t.
120t dry mass * 2 = 240t.
150t payload.
Total mass of docked system: 1890t

Thrust required for 1E-4m/s² acceleration:  189N.

Hall Effect thrusters generally run about 60mN/kW.
So you'd need 3.15MW of Hall thrusters to do settling.
Even 315kW for 1E-5m/s² settling acceleration is pretty iffy.

Nope.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sevenperforce on 10/21/2020 09:14 pm
It's possible to have an androgynous mating with just two probe-and-drogue connectors. There is space at the base of the the chine structure that encloses the aft flaps. Put a probe on one side and a drogue on the other.

RCS is used by both vehicles to line up tail to tail, but NOT inverted as early mockups have shown. This allows the port side of one vehicle to align with the starboard side of the other. Once aligned, both probes are mechanically extended. Both vehicles fire their forward RCS to "reverse" and push the probes into the corresponding drogues, which damps most degrees of freedom. The drogues grip the probes to soft dock, and the probes mechanically retract to pull the ships back together and force the hard dock. Because the connecting points are as far lateral as possible, you end up with much less torque on the docking points.

How will the tanks be repressurized? Obviously, while the Raptors are firing the autogenous press gas lines will be replenishing ullage. But if the gas-gas thrusters are fired by straight tapoff of the tanks, then you end up with reduced ullage and a low-pressure combustion. It would make more sense for the Raptors to push their ultra-high-pressure autogenous press gases into accumulators, which then use a valve-controlled release to maintain a constant head pressure. The accumulators can also feed the gas-gas-thrusters in series for much higher efficiency.

Rather than being located within the probe-and-drogue system, the propellant lines can be located above and below the thrust puck -- methane on top, LOX below. You only need the two because the vehicles will be aligned properly, and the lines don't even have to be placed axisymmetrically. The propellant lines connect and then RCS on both vehicles is fired for settling.

As soon as the propellant begins to flow down both lines simultaneously, the receiving vehicle's tanks are allowed to propulsively vent, maintaining ullage, while the tanker's tanks are continuously repressurized from the accumulators on it. This continues until transfer is complete.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/21/2020 10:14 pm
It's possible to have an androgynous mating with just two probe-and-drogue connectors. There is space at the base of the the chine structure that encloses the aft flaps. Put a probe on one side and a drogue on the other.

RCS is used by both vehicles to line up tail to tail, but NOT inverted as early mockups have shown. This allows the port side of one vehicle to align with the starboard side of the other. Once aligned, both probes are mechanically extended. Both vehicles fire their forward RCS to "reverse" and push the probes into the corresponding drogues, which damps most degrees of freedom. The drogues grip the probes to soft dock, and the probes mechanically retract to pull the ships back together and force the hard dock. Because the connecting points are as far lateral as possible, you end up with much less torque on the docking points.

The problem with just a pair opposed soft points is that they form a hinge.  They're fine managing forces along their axis, but any translations or rotations perpendicular to that axis can't be controlled.  It's not a good idea to have the edges of two Starships grinding into one another, especially with engine bells nearby.  The only way I can think of to mitigate that hinge effect is with three capture points.  Once you do that, the geometry requires flipping the vehicles around 180º, and then the symmetry of the lines becomes an issue.

The other problem, specific to probe and drogue, is that the drogues are fixed.  That means that the moment one probe strikes the drogue cone, the vehicles are pushed perpendicular to the center axis.  If one probe strikes first, that motion can potentially cause the other drogue to miss completely, and that's a mess.

The advantage of the soft-capture rings is that they can lunge with very little perpendicular motion applied to the vehicles; most of the errors are soaked up by the floating struts that connect the ring to the vehicle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/22/2020 12:06 am
#3, except they're not really "rings," just self-aligning latches w integrated soft capture. These would be incorporated into the standard interstage latching mechanism.

Whether N = 3 (as you speculate) or some other number, I'm not sure. F9 originally (v1.0) had 9x latches and 3x pneumatic pushers, while F9 v1.1 switched to 3x combination latches / pushers. I suspect Starship will use N = 6, but it's just a hunch.

What do you mean by "integrated soft capture"?

The self-aligning latches will need some sort of damped suspension system to soak up the residual kinetic energy. This soft-capture mechanism has to be attached to the self-aligning latches for the obvious reason that the latches are the first part that gets attached, hence "integrated."

Starship has the same basic problem as any other docking spacecraft.  The reason for the active soft capture rings is because the hard latches aren't self-aligning.  They need the soft system to restrict the problem enough that the guide pins will engage.

Again, my whole point was that there's no reason these peripheral mechanisms need to be "rings." They don't need to accommodate a pressurized tunnel inside like IDSS, so a simple and lightweight probe-and-drogue system should work fine.

I have no clue about the actual hard latches.  I'd think that the staging quick-releases could be engineered to act as hard latches as well, but you want really, really, really reliable staging latches (failure to stage is an LOC event, when/if they get to launching people), so maybe a separate set of IDSS-like hard latches will turn out to be better.  Either way, I don't think we're talking about massive loads on the latches on-orbit; they just need to be strong enough to stabilize the fueling lines in the even that there's some kind of wacky torque on the stack (e.g. a transiently firing thruster).  And they obviously don't need to hold pressure.

I think the high reliability latch is a feature not a bug. You "simply" engineer your soft capture and retraction mechanism to achieve the same tolerance and positive engagement force as the ground stacking spec. Let's not forget that a failure to disengage after refilling would be a LOM event, and possible LOC barring some questionable rescue mission architecture.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/22/2020 12:20 am
Note that on Falcon 9 they intentionally put the fuel and oxidizer GSE connections on opposite sides of the vehicle, for obvious safety reasons. IMO Starship will not "un-learn" this safety feature.

I've been thinking about that issue and it's... not easy to solve.  I can't come up with an arrangement where each guzouta mates up with a guzinta after a y- and z-rotation that doesn't have fuel and oxidizer lines next to each other.  I'm bad at this stuff, so I'm willing to believe I've had a failure of imagination.  You should play with it and see if you can figure it out.

In order to separate the fuel and oxidizer by exactly 180 degrees, the propellant connectors should be located along the rotation axis.  If you're flipping along Y (pitch flip, heat shields on opposite sides), the connectors would be located left/right. If you're flipping along Z (yaw flip, heat shields on the same side), the connectors would be located top/bottom.

Since we want to connect fill/drain on A to vent on B and vent on A to fill/drain on B, the connectors should be located symmetrically on each side of the rotation axis. As long as the connectors are the same distance from the imaginary "flip plane," they should align perfectly after the 180 degree rotation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/22/2020 12:33 am
It's possible to have an androgynous mating with just two probe-and-drogue connectors. There is space at the base of the the chine structure that encloses the aft flaps. Put a probe on one side and a drogue on the other.

RCS is used by both vehicles to line up tail to tail, but NOT inverted as early mockups have shown. This allows the port side of one vehicle to align with the starboard side of the other. Once aligned, both probes are mechanically extended. Both vehicles fire their forward RCS to "reverse" and push the probes into the corresponding drogues, which damps most degrees of freedom. The drogues grip the probes to soft dock, and the probes mechanically retract to pull the ships back together and force the hard dock. Because the connecting points are as far lateral as possible, you end up with much less torque on the docking points.

The problem with just a pair opposed soft points is that they form a hinge.  They're fine managing forces along their axis, but any translations or rotations perpendicular to that axis can't be controlled.  It's not a good idea to have the edges of two Starships grinding into one another, especially with engine bells nearby.  The only way I can think of to mitigate that hinge effect is with three capture points.  Once you do that, the geometry requires flipping the vehicles around 180º, and then the symmetry of the lines becomes an issue.

Any even number of probes and drogues works. You could have four or six.

Again, just arrange them so each probe has a mirrored drogue across the flip plane. The flip plane can be vertical or horizontal or at some crazy angle, entirely your choice (subject to other design constraints like engine clearance).


The other problem, specific to probe and drogue, is that the drogues are fixed.  That means that the moment one probe strikes the drogue cone, the vehicles are pushed perpendicular to the center axis.  If one probe strikes first, that motion can potentially cause the other drogue to miss completely, and that's a mess.

This can only happen if one port is "off" in one direction, and the other is "off" in the opposite direction.

I think the real answer is that this just feeds into your accuracy requirements. You just set your allowable position / angle / velocity / angular velocity error so that it can't do that.

The advantage of the soft-capture rings is that they can lunge with very little perpendicular motion applied to the vehicles; most of the errors are soaked up by the floating struts that connect the ring to the vehicle.

Probes can be mounted on floating struts too, and for a smaller mass penalty.  Each probe only needs three struts, half the number of struts required for a soft-capture ring.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/22/2020 05:31 am
The self-aligning latches will need some sort of damped suspension system to soak up the residual kinetic energy. This soft-capture mechanism has to be attached to the self-aligning latches for the obvious reason that the latches are the first part that gets attached, hence "integrated."

I don't think that the geometry favors actual integration of the soft and hard capture systems.  As you say below, there's no pressurized tunnel.  But the 9m diameter of the contact surface is big enough that the hard latches are likely more numerous than the soft capture systems, and minimizing those soft systems is important for docking reliability: the more soft systems you have, the more likely one of them will get missed or even fouled during the capture process.  So the real question is: what's the minimum you need?  I'm pretty sure that the answer to that is 3.

As for whether the soft systems are "rings" or not:  Meh, no clue.  They certainly don't have to be.  Could they be probe-and-drogue?  I guess so... buy why? The existing IDSS systems are already developed, and they accommodate a lot more error in position and rates than P&D did.  My guess is that SpaceX'll just adapt what they have and optimize it later.

Quote
I think the high reliability latch is a feature not a bug. You "simply" engineer your soft capture and retraction mechanism to achieve the same tolerance and positive engagement force as the ground stacking spec. Let's not forget that a failure to disengage after refilling would be a LOM event, and possible LOC barring some questionable rescue mission architecture.

IDSS has soft and hard capture systems coaxial because the tunnel is the contact surface.  That's not going to be true with Starship.  That's why I think the soft and hard systems aren't co-located.

I agree that the hard latches for docking have to be as high reliability as the quick-release latches for staging, but they aren't nearly as time-sensitive in release.  If  you can avoid compromising quick-release performance with a hard latch system that'll serve both purposes, that's clearly a superior design.  But if you can't, then it's not a big deal to make the docking latches separate from the staging latches.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/22/2020 06:13 am
In order to separate the fuel and oxidizer by exactly 180 degrees, the propellant connectors should be located along the rotation axis.  If you're flipping along Y (pitch flip, heat shields on opposite sides), the connectors would be located left/right. If you're flipping along Z (yaw flip, heat shields on the same side), the connectors would be located top/bottom.

Since we want to connect fill/drain on A to vent on B and vent on A to fill/drain on B, the connectors should be located symmetrically on each side of the rotation axis. As long as the connectors are the same distance from the imaginary "flip plane," they should align perfectly after the 180 degree rotation.

If you're willing to go to 4 soft-captures, you have some more leeway, but then you've got more complexity and more chance of a failure on capture, and the whole Starship is tri-laterally symmetric.  Cramming 4 devices with the three engines in there is gonna be weird.

But you're right that if I arrange the lines along the axis of symmetry, things'll match up.  See below.

However, there are some real estate problems:

1) Note that, as I have them, one set of lines has to live in fairly close proximity to an RVac nozzle, and you can't space them further away without running into a landing leg.   ISTM that the RVac is a greater risk than pairing fuel with oxidizer.

2) The other alternative is to place the lines more medially, outside of the thrust puck (far enough away to avoid the gimbaling center engines), but inside the RVac nozzles.  I'm not sure that this is better than #1 (shown below), and it's certainly a lot more iffy structurally.

Going to more that one axis of symmetry (e.g. with 4 soft captures) doesn't help a lot because of the positioning of the RVacs and the legs.

One modification to the scheme I had here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2144577#msg2144577) would be simply to space the pairs of LCH4 and LOX further away from the capture points, i.e., move each one closer to the leg--and in this case there's more room to spread out.  It doesn't give you 180º, but it gives you >3m between the each line--which is pretty much what F9 has.

Frankly, I don't think there's much you can do if there's a leak.  Unlike RP-1, LCH4 is volatile.  If there's a leak, it'll fill the entire interstage space almost instantly, and spacing really isn't going to help.  Note that this is just as much of a problem on the ground and during SuperHeavy boost as it is during RPOD for refueling.

Moral of the story:  Really, really don't allow leaks.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: armchairfan on 10/22/2020 07:00 am
However, there are some real estate problems:

I think that you're trying to be overly symmetric. You just need bilateral symmetry along a single axis (y in your diagram). Adapting your diagram, just replace the (say) LOX locations with ones further from the vacuum raptor. (Big B and T.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Pete on 10/22/2020 07:01 am
Is the a good reason why you need the liquid and gas-return plumbing to be separate?

Why not just have a very small gas return pipe in/directly alongside the liquid pipe.
If you *must* have the gas pipe vent at the top of the tank, just continue the gas return pipe along the tank wall to the top.

The gas return is similar volume to the liquid pumped, but being a gas it is ridiculously easier to pipe, thus requiring a ridiculously smaller pipe diameter.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/22/2020 07:17 am
However, there are some real estate problems:

I think that you're trying to be overly symmetric. You just need bilateral symmetry along a single axis (y in your diagram). Adapting your diagram, just replace the (say) LOX locations with ones further from the vacuum raptor. (Big B and T.)

That's a good idea.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/22/2020 07:22 am
Is the a good reason why you need the liquid and gas-return plumbing to be separate?

Why not just have a very small gas return pipe in/directly alongside the liquid pipe.
If you *must* have the gas pipe vent at the top of the tank, just continue the gas return pipe along the tank wall to the top.

The gas return is similar volume to the liquid pumped, but being a gas it is ridiculously easier to pipe, thus requiring a ridiculously smaller pipe diameter.

You can do that, but then you have to vent ullage gases as the tank fills on one side, and make new ullage gas as the tank drains on the other side.  Note also that, under settling thrust, the liquid in the "backward" Starship is up against the top of the tanks, so what would normally be the vent line doesn't have any gas near it.

So, instead, if you connect all the "bottom" lines to their corresponding "top" lines on the other Starship, then the ullage gases just move in the opposite direction from the way that the liquid is being moved.

There are obviously complexities in making both sets of lines do double duty as both drain/fill and vent lines, but it's very flexible.  Any Starship can fulfill the role of either tanker or tankee.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: armchairfan on 10/22/2020 08:19 am
Is the a good reason why you need the liquid and gas-return plumbing to be separate?
A picture may be worth a thousand words. Shamelessly adapting John's diagram (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2123644#msg2123644).

Of historical note, after previously lurking for years, a few pages of similar back and forth a couple of years ago about the need for either androgynous connections or different tanker vs ship mating hardware a couple of years ago finally got me to register just so that I could post (https://forum.nasaspaceflight.com/index.php?topic=46395.msg1862227#msg1862227) that you don't need either. In fact, the SpaceX IAC 2017 presentation showed four refueling lines with two male and two female. The symmetry plane changed from 2017 to 2018 necessitating an additional 180-degree roll when refueling, which remains to this day.

SpaceX has lots of challenges on the road to Mars but I think that ship to ship docking is one of their easier ones.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 10/22/2020 02:41 pm

I don't think that the geometry favors actual integration of the soft and hard capture systems.  As you say below, there's no pressurized tunnel.  But the 9m diameter of the contact surface is big enough that the hard latches are likely more numerous than the soft capture systems, and minimizing those soft systems is important for docking reliability: the more soft systems you have, the more likely one of them will get missed or even fouled during the capture process.  So the real question is: what's the minimum you need?  I'm pretty sure that the answer to that is 3.


I think 2 is the minimum as long as they are along the skirt.
Remember we have a large flat 9m ring defined by the bottom of the skirt. All you have to do is hold them together and all forces are transmitted through the skirt.

If you wanted 1 then it would have to be in the middle between the 3 sea level raptor. But this wouldn't control the rotation alignment. But it would be easier for first contact.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/22/2020 09:09 pm
I think 2 is the minimum as long as they are along the skirt.
Remember we have a large flat 9m ring defined by the bottom of the skirt. All you have to do is hold them together and all forces are transmitted through the skirt.

If you wanted 1 then it would have to be in the middle between the 3 sea level raptor. But this wouldn't control the rotation alignment. But it would be easier for first contact.

I agree that the hard contact surface makes the number of hard latches fairly small (maybe 2, although that seems kinda scary from a fault-tolerance standpoint), but soft capture is actually more demanding.

The problem is that both Starships are offset, misaligned, translating, and rotating with respect to each other at soft capture.  Those motions have to be damped out, and they have to be damped out before the edge of one Starship digs into someplace on the other Starship that isn't its hard contact point (like an engine nozzle, a prop line, or a landing leg).

You can't do that damping with two soft capture spots.  All the motions off-axis can't be adequately controlled.  With 3, though, there's leverage in all the degrees of freedom.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 10/22/2020 09:16 pm
To have the ability for any two identical SS to dock tail to tail with a 180o rotation you would need 4 equally spaced around the skirt and 'clocked' to the same reference location.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/22/2020 09:45 pm
To have the ability for any two identical SS to dock tail to tail with a 180o rotation you would need 4 equally spaced around the skirt and 'clocked' to the same reference location.

No, you don't.  You simply need an axis of symmetry.  Remember, there are two 180º rotations, about both the x- and z-axes. x rolls one over, z puts the two tail-to-tail.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/23/2020 03:19 am
The self-aligning latches will need some sort of damped suspension system to soak up the residual kinetic energy. This soft-capture mechanism has to be attached to the self-aligning latches for the obvious reason that the latches are the first part that gets attached, hence "integrated."

I don't think that the geometry favors actual integration of the soft and hard capture systems.

Er... I didn't say that. :D

I said integration of the soft-capture mechanism and the self-aligning latches (needed because... well, how else does your soft capture mechanism work?), not the hard capture latches. Hopefully you can can now see why I was very confused by your question. ;)

But since you brought it up, I do think the soft capture and hard capture system will share hardpoint locations on the skirt. Co-location, but not necessarily integration (though integration is probably perferred, eg like how the latches and pushers were integrated during the transition from F9 v1.0 to F9 v1.1).

As you say below, there's no pressurized tunnel.  But the 9m diameter of the contact surface is big enough that the hard latches are likely more numerous than the soft capture systems, and minimizing those soft systems is important for docking reliability: the more soft systems you have, the more likely one of them will get missed or even fouled during the capture process.

That's one possibility. The other possibility is that the more numerous the soft capture system, the more likely that retraction will "click" the unlatched ports in place.

If you're trying to do that with three latches, it won't work (no leverage). Four is the minimum number of soft capture mechanisms that could be expected to do that, though.

As for whether the soft systems are "rings" or not:  Meh, no clue.  They certainly don't have to be.  Could they be probe-and-drogue?  I guess so... buy why? The existing IDSS systems are already developed, and they accommodate a lot more error in position and rates than P&D did.  My guess is that SpaceX'll just adapt what they have and optimize it later.

I think you're confusing implementation details for fundamentals. There's no reason you couldn't make a P&D which accepts more error, there just wasn't a reason to do so for Apollo/Soyuz.

And also, are we expecting a lot of error? Dragon docks with centimeter precision, and I don't see any good reasons why Starship would be any different. Clearly you can scale the RCS bit to achieve whatever precision you need.

I think the high reliability latch is a feature not a bug. You "simply" engineer your soft capture and retraction mechanism to achieve the same tolerance and positive engagement force as the ground stacking spec. Let's not forget that a failure to disengage after refilling would be a LOM event, and possible LOC barring some questionable rescue mission architecture.

IDSS has soft and hard capture systems coaxial because the tunnel is the contact surface.  That's not going to be true with Starship.  That's why I think the soft and hard systems aren't co-located.

This from the guy who was just saying that they'll use IDSS because it's "what they have..." :P

The soft capture systems need to deal with dynamic load and the hard capture / latches needs to deal with both static and dynamic loads. So again I think they'll probably share a hardpoint, rather than structurally reinforcing two separate points on the vehicle.

I agree that the hard latches for docking have to be as high reliability as the quick-release latches for staging, but they aren't nearly as time-sensitive in release.  If  you can avoid compromising quick-release performance with a hard latch system that'll serve both purposes, that's clearly a superior design.  But if you can't, then it's not a big deal to make the docking latches separate from the staging latches.

Yes, my baseline assumption is that SpaceX is good enough to hit the "superior design" in this case.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/23/2020 03:44 am
In order to separate the fuel and oxidizer by exactly 180 degrees, the propellant connectors should be located along the rotation axis.  If you're flipping along Y (pitch flip, heat shields on opposite sides), the connectors would be located left/right. If you're flipping along Z (yaw flip, heat shields on the same side), the connectors would be located top/bottom.

Since we want to connect fill/drain on A to vent on B and vent on A to fill/drain on B, the connectors should be located symmetrically on each side of the rotation axis. As long as the connectors are the same distance from the imaginary "flip plane," they should align perfectly after the 180 degree rotation.

If you're willing to go to 4 soft-captures, you have some more leeway, but then you've got more complexity and more chance of a failure on capture, and the whole Starship is tri-laterally symmetric.  Cramming 4 devices with the three engines in there is gonna be weird.

Again, SpaceX is no stranger to weird [a]symmetry. The Falcon 9 is mostly 4-symmetric (legs, grid fins), but the docking latches are 3-symmetric, and the RCS thrusters are 2-symmetric.

But you're right that if I arrange the lines along the axis of symmetry, things'll match up.  See below.

However, there are some real estate problems:

1) Note that, as I have them, one set of lines has to live in fairly close proximity to an RVac nozzle, and you can't space them further away without running into a landing leg.   ISTM that the RVac is a greater risk than pairing fuel with oxidizer.

"Running into a landing leg" isn't a big problem, because the landing legs are folded during refilling. Surely SpaceX is capable of designing a docking latch that doesn't physically intersect the landing leg or its deployment arc.

This is a "my arm hurts when I do this" problem. If you know from the start that your docking mechanism and your landing legs need to avoid interfering with each-other, you just design them with that constraint in mind.

And again, the landing legs need a structural hardpoint. So combining the soft latch, hard latch, and landing legs into a single unified mechanism saves mass, and this is likely the path SpaceX will take IMO. SpaceX has never skirked from an engineering challenge, and unlike most "why doesn't SpaceX..." proposals this would actually simplify things overall (as you're discovering with your current "Tetris all the parts together" exercise).

2) The other alternative is to place the lines more medially, outside of the thrust puck (far enough away to avoid the gimbaling center engines), but inside the RVac nozzles.  I'm not sure that this is better than #1 (shown below), and it's certainly a lot more iffy structurally.

Going to more that one axis of symmetry (e.g. with 4 soft captures) doesn't help a lot because of the positioning of the RVacs and the legs.

Yeah, I think the real answer is you just orient the axis of symmetry up/down instead of left/right. This way the refueling lines are as far apart from each-other as possible and as far apart from the Raptor engines as possible.

Yes, they co-locate with the legs. But that's a feature, not a bug.

Fuel lines + oxidizer lines = bad synergy (boom).

Fuel/oxidizer lines + engines = bad synergy (boom).

Fuel/oxidizer lines + hard latch = good synergy (tight alignment)

Hard latch + landing legs = good synergy (sharing a structural hardpoint).

One modification to the scheme I had here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2144577#msg2144577) would be simply to space the pairs of LCH4 and LOX further away from the capture points, i.e., move each one closer to the leg--and in this case there's more room to spread out.  It doesn't give you 180º, but it gives you >3m between the each line--which is pretty much what F9 has.

Frankly, I don't think there's much you can do if there's a leak.  Unlike RP-1, LCH4 is volatile.  If there's a leak, it'll fill the entire interstage space almost instantly, and spacing really isn't going to help.  Note that this is just as much of a problem on the ground and during SuperHeavy boost as it is during RPOD for refueling.

Moral of the story:  Really, really don't allow leaks.

"Really isn't going to help" is too pessimistic an assessment. The interstage isn't airtight, so there's still a non-zero advantage.

I think this one falls into the "don't do nuthin' dumb" category. :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/23/2020 05:58 am
Er... I didn't say that. :D

I said integration of the soft-capture mechanism and the self-aligning latches (needed because... well, how else does your soft capture mechanism work?), not the hard capture latches. Hopefully you can can now see why I was very confused by your question. ;)

But since you brought it up, I do think the soft capture and hard capture system will share hardpoint locations on the skirt. Co-location, but not necessarily integration (though integration is probably perferred, eg like how the latches and pushers were integrated during the transition from F9 v1.0 to F9 v1.1).

"Self-aligning" is not necessarily a modifier for "latches".  All you need is a guide pin.  But I agree that there's nothing preventing colocation of the soft and hard systems.  A lot probably depends on whether the staging latches and the hard capture latches are one and the same.

I suspect we've gotten wrapped around the axle in terms of what "integrated" means.  ISTM that you're asserting that integration of the soft and hard systems per se isn't required.  I agree with that.

Quote
That's one possibility. The other possibility is that the more numerous the soft capture system, the more likely that retraction will "click" the unlatched ports in place.

If you're trying to do that with three latches, it won't work (no leverage). Four is the minimum number of soft capture mechanisms that could be expected to do that, though.

Not sure I'm following you here. Most alignment is done during stabilization, and 3 points will give you all the leverage you need there.  (Note that the soft system has to be able to pull and push, but that's current state of the art.)

If you're saying that retraction causes subsequent misalignments... I'm not sure I buy that.  That seems to be a "Doctor, Doctor, it hurts when I do this" kind of problem.  A soft system that can damp out misalignments can surely prevent them from creeping in during retraction.

Quote
And also, are we expecting a lot of error? Dragon docks with centimeter precision, and I don't see any good reasons why Starship would be any different. Clearly you can scale the RCS bit to achieve whatever precision you need.

I don't think that scaling is linear, because the moments of inertia are in some cases 2-3 orders of magnitude larger with Starship.

I'll also be interested to see what kind of minimum impulse bit the Starship RCS system comes up with.  It may turn out that there's a little RCS and a big RCS, but that's pretty complicated.  If they go with only big RCS, things are going to be interesting.

Quote
Yes, my baseline assumption is that SpaceX is good enough to hit the "superior design" in this case.

Even superior designs have trade spaces.  In the grand scheme of things, saving a couple hundred kgs in optimized latches probably isn't exactly low-hanging fruit in terms of mass reduction.  SpaceX likes "superior", but they often like "cheap" even more.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/23/2020 06:29 am
Again, SpaceX is no stranger to weird [a]symmetry. The Falcon 9 is mostly 4-symmetric (legs, grid fins), but the docking latches are 3-symmetric, and the RCS thrusters are 2-symmetric.

I have no problem with weird asymmetry, but there are real estate constraints around the skirt.  If you can figure out how to put a soft capture system covering up a landing leg, then things get a little less crowded.  Otherwise, if you have a 0.5-1.0m capture target, it takes up some space.

There's plenty of real estate open on the skirt, but there's considerably less if you try to put two things on the same symmetry axis, because the most open spaces on the skirt have an RVac on the opposite side.

Quote
"Running into a landing leg" isn't a big problem, because the landing legs are folded during refilling. Surely SpaceX is capable of designing a docking latch that doesn't physically intersect the landing leg or its deployment arc.

Of course they are.  But that trims the amount of real estate down by quite a bit.  If you then further constrain the problem by insisting on opposing latches (i.e., any kind of even symmetry), then things are getting pretty cozy.

Quote
And again, the landing legs need a structural hardpoint. So combining the soft latch, hard latch, and landing legs into a single unified mechanism saves mass, and this is likely the path SpaceX will take IMO.

I'm completely prepared to believe that your average SpaceX ME is smarter than I am--especially about ME problems.  I'm just having a failure of imagination in how a soft capture anchor can get out of the way of a landing leg without some sort of whiz-bang way of folding it out of the way, post-Mr. Fusion DeLorean-style, to make way for the leg--and then it's not really sharing the same hard point.

Quote
Yeah, I think the real answer is you just orient the axis of symmetry up/down instead of left/right. This way the refueling lines are as far apart from each-other as possible and as far apart from the Raptor engines as possible.

That's an invariant change, unless the size and number of RVacs changes.

Two things:

1) If you're still worried about keeping max distance between fuel and oxidizer lines, then something like armchairfan's suggestion above solves that problem completely without doing anything exotic.

2) Note that all of the legs are next to RVac nozzles.  Again, if you have magic fold-away soft targets, that's no big deal.  Otherwise, it's not an ideal environment for a piece of hardware designed to lunge smoothly through a decent-sized range of positions and velocities.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tyrred on 10/23/2020 07:04 am
I have nothing to add, but just want to break in and note that I'm really enjoying this civil discourse. Exemplary of the NSF way. Carry on!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Keldor on 10/23/2020 08:41 am
Is there any particular reason a solid attachment is needed?  Couldn't you use a mechanism similar to what aircraft use for inflight refueling, and basically just have a hose (well, two hoses) running from the tanker to the receiving vehicle?

As far as using microaccelleration to settle fuel, both vehicles have their own RCS systems and are perfectly capable of staying in formation and accellerating, provided some care is taken to make sure they never blast each other with exhaust plumes.

Another thing is that you don't necesarily need a gas return line at all.  It's quite possible that just venting it and letting suction help bring propellant over is a win.  Likely the tank pressurization alone wouldn't be enough to push the propellant through the pipes in a reasonable amount of time, given that you have to be continuously accellerating to keep the propellant settled against the outlet, but we may also be greatly overestimating how much accelleration is needed to keep the propellant settled.  After all, there's not really any other force that would be working against it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: JustDAppie on 10/23/2020 08:55 am
Something that has occurred to me, and why not throw it out here as a first timer.

Why not have a ‘service station’ which handles the docking and transfer of fuel from one vehicle to another?

Such a ‘service station’ could have multiple docking stations for different storage tanks, and even multiple docking stations for vehicles to be filled from.

‘Storage tanks’ could include:
•   Starships, ‘ordinary’ or ‘stripped down’ without anything required for re-entry, allowing them to be simple ‘dumb’ tanks (launched empty or full)
•   Falcon second stages, why dump them, get them to the station and use their tanks for storage
•   Anything else that may yet even be developed…
The point is the ‘service station’ has the appropriate docking and fuel connections for what ever comes to it. Develop and new rocket? No problem, just send up a new docking adapter and it can be used by the ‘service station’, no need for symmetry!

Similarly, any kind of vehicle could come to be refueled, it just connects to it’s specific docking and fuel transfer connection. Maybe, someday in the future these things will become more standard, like the USB charging for mobile phones. (Who still has a nest of charging adapters in a drawer somewhere?)

The 'service station' would handle all the power, propulsion, plumbing, and pumping required. It could even handle different fuels. Transfer from the 'storage tanks' could be done by 'sucking' the fuel from the 'bottom' or 'pumping' gas into the 'top' to force the fuel out the bottom. Transfer to the vehicle to be refueled would be done either way. This could be built into the main body of the 'service station' or into each adapter that is connected. More adapters could be added either way to increase the stations capacity.

We have already had two of the three ‘R’s of sustainability:
•   Reduce – make them rockets lighter!
•   Reuse – Falcon 9 anyone?
Now we should get onto the third, Repurpose. Rather than throw away those empty second stages, repurpose them! Photon anyone?

This could even be an opportunity for a third party company to get involved, to deal with all that messy fuel stuff. Just pull up your vehicle and fill-er-up (for a fee of course).

I include a huge amount of quotes and hand waving, even introduce a layer of complexity to try to simplify elsewhere, so fully expect some robust feedback.
Enjoy!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TrevorMonty on 10/23/2020 06:03 pm
Something that has occurred to me, and why not throw it out here as a first timer.

Why not have a ‘service station’ which handles the docking and transfer of fuel from one vehicle to another?

Such a ‘service station’ could have multiple docking stations for different storage tanks, and even multiple docking stations for vehicles to be filled from.

‘Storage tanks’ could include:
•Starships, ‘ordinary’ or ‘stripped down’ without anything required for re-entry, allowing them to be simple ‘dumb’ tanks (launched empty or full)
•Falcon second stages, why dump them, get them to the station and use their tanks for storage
•Anything else that may yet even be developed…
The point is the ‘service station’ has the appropriate docking and fuel connections for what ever comes to it. Develop and new rocket? No problem, just send up a new docking adapter and it can be used by the ‘service station’, no need for symmetry!

Similarly, any kind of vehicle could come to be refueled, it just connects to it’s specific docking and fuel transfer connection. Maybe, someday in the future these things will become more standard, like the USB charging for mobile phones. (Who still has a nest of charging adapters in a drawer somewhere?)

The 'service station' would handle all the power, propulsion, plumbing, and pumping required. It could even handle different fuels. Transfer from the 'storage tanks' could be done by 'sucking' the fuel from the 'bottom' or 'pumping' gas into the 'top' to force the fuel out the bottom. Transfer to the vehicle to be refueled would be done either way. This could be built into the main body of the 'service station' or into each adapter that is connected. More adapters could be added either way to increase the stations capacity.

We have already had two of the three ‘R’s of sustainability:
•Reduce – make them rockets lighter!
•Reuse – Falcon 9 anyone?
Now we should get onto the third, Repurpose. Rather than throw away those empty second stages, repurpose them! Photon anyone?

This could even be an opportunity for a third party company to get involved, to deal with all that messy fuel stuff. Just pull up your vehicle and fill-er-up (for a fee of course).

I include a huge amount of quotes and hand waving, even introduce a layer of complexity to try to simplify elsewhere, so fully expect some robust feedback.
Enjoy!
Distributed launch has some advantages over fuel depots. The tankers can launched to optimum refuelling orbit for mission. Only need to in orbit long enough to support mission after that return it to earth.  Depots or service stations are better for servicing high flight rate destinations.

Much same as services stations and fuel dumps on earth. Plus side of space fuel depots is ability to move them if there current location isn't economic unlike earth's abandoned service stations.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 10/23/2020 06:29 pm
Hey a new job:

Maybe they will put one crew on the tankers to handle unforeseen situations. Like when auto docking doesn't work exactly right.

I would sign up for a job like that.
2 runs per day.
Space trucker.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/23/2020 08:04 pm
Is there any particular reason a solid attachment is needed?  Couldn't you use a mechanism similar to what aircraft use for inflight refueling, and basically just have a hose (well, two hoses) running from the tanker to the receiving vehicle?

As far as using microaccelleration to settle fuel, both vehicles have their own RCS systems and are perfectly capable of staying in formation and accellerating, provided some care is taken to make sure they never blast each other with exhaust plumes.

Another thing is that you don't necesarily need a gas return line at all.  It's quite possible that just venting it and letting suction help bring propellant over is a win.  Likely the tank pressurization alone wouldn't be enough to push the propellant through the pipes in a reasonable amount of time, given that you have to be continuously accellerating to keep the propellant settled against the outlet, but we may also be greatly overestimating how much accelleration is needed to keep the propellant settled.  After all, there's not really any other force that would be working against it.

There's a fair amount of discussion on this topic, either up-thread or on some other thread.  Three things:

1) It doesn't appear to be what SpaceX has in mind.

2) Getting fuel lines to snake from one Starship to another is no picnic.  Jon Goff and Altius have a concept (or maybe more than a concept now) for a "sticky boom" that might fill the bill.  Basically, you still have the same docking problem, with the same stabilization stresses.  Only now you have less leverage over the thing you're trying to stabilize.

3) None of the symmetry requirements magically go away with stand-off fueling.

The acceleration for settling is really small.  Somewhere up-thread, there's a citation that the Centaur people estimated somewhere between 1E-4 and 1E-5 m/s² was adequate.  (Not sure if that was settling for ignition, prop transfer, or consolidation for low-boiloff storage.)

The question of whether there are one or two lines for each tank is strongly affected by the attitude of the Starships during fueling.  We know they both start out on the pad in gravity, where a bottom fill/drain is pretty much essential.  But if they're fueling back-to-back on-orbit, then the prop in one of them will settle at the top of the tank, so you need lines up there.  Then you might as well manage the ullage gas through whichever line doesn't have the prop covering it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/23/2020 08:14 pm
Something that has occurred to me, and why not throw it out here as a first timer.

Why not have a ‘service station’ which handles the docking and transfer of fuel from one vehicle to another?

Such a ‘service station’ could have multiple docking stations for different storage tanks, and even multiple docking stations for vehicles to be filled from.

‘Storage tanks’ could include:
•   Starships, ‘ordinary’ or ‘stripped down’ without anything required for re-entry, allowing them to be simple ‘dumb’ tanks (launched empty or full)
•   Falcon second stages, why dump them, get them to the station and use their tanks for storage
•   Anything else that may yet even be developed…
The point is the ‘service station’ has the appropriate docking and fuel connections for what ever comes to it. Develop and new rocket? No problem, just send up a new docking adapter and it can be used by the ‘service station’, no need for symmetry!

Similarly, any kind of vehicle could come to be refueled, it just connects to it’s specific docking and fuel transfer connection. Maybe, someday in the future these things will become more standard, like the USB charging for mobile phones. (Who still has a nest of charging adapters in a drawer somewhere?)

The 'service station' would handle all the power, propulsion, plumbing, and pumping required. It could even handle different fuels. Transfer from the 'storage tanks' could be done by 'sucking' the fuel from the 'bottom' or 'pumping' gas into the 'top' to force the fuel out the bottom. Transfer to the vehicle to be refueled would be done either way. This could be built into the main body of the 'service station' or into each adapter that is connected. More adapters could be added either way to increase the stations capacity.

We have already had two of the three ‘R’s of sustainability:
•   Reduce – make them rockets lighter!
•   Reuse – Falcon 9 anyone?
Now we should get onto the third, Repurpose. Rather than throw away those empty second stages, repurpose them! Photon anyone?

This could even be an opportunity for a third party company to get involved, to deal with all that messy fuel stuff. Just pull up your vehicle and fill-er-up (for a fee of course).

I include a huge amount of quotes and hand waving, even introduce a layer of complexity to try to simplify elsewhere, so fully expect some robust feedback.
Enjoy!

Depots with specialized facilities make more sense when you have lots of missions with very similar orbital elements, which won't be the case for SpaceX for a long time, if ever.  There are two other things to think about:

1) Refueling is most efficient when it occurs at the lowest feasible orbital energy, while depots, because they need to service a lot of different orbits from one place, make more sense in high-energy orbits, where making inclination, RAAN and argument of perigee changes is cheap.

2) Depots and higher-orbit tankers have a non-trivial debris risk associated with them if there's an accident.

Both of these militate toward starting out refueling in VLEO, with "tankers" that are simply ordinary Starships with no payload.  You launch a series of "lift tankers" to meet the needs of the mission you're mounting.  The first lift tanker stays in VLEO and becomes the "accumulation tanker".  The remaining lift tankers fuel the accumulation tanker until enough prop has accumulated to satisfy the mission requirement.  Only then does the payload Starship launch and rendezvous with the accumulation tanker, so there's only one docking and refueling operation.  After that, the accumulation tanker goes home to get ready for the next mission.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 10/24/2020 12:02 am
If I may, let's re-cap on what we currently know has been suggested and used for fueling by SpaceX.

All renders of SS on-orbit refueling with versions similar to what we now see have the craft tail to tail and inverted relative to eachother. See the first image from Neopork below.

However earlier renders, while still tail to tail are not inverted, see second image below (source unknown, Google image search)

Finally they are currently using a single GSE connection location with SN prototypes on the dorsal (Z-?) part of the skirt as shown (red circle) in the excellent aerial photo of SN8 by RGV shown third below.

What does this mean?
- tail to tail so far is consistent.
- 180o inversion appears negotiable over time.
- current location of fuel hardware is not consistent with 180o rotation when tail to tail, but is with early renders.

So ultimately the question is how close is the current fuel hardware location to the final location?
Everything else depends on this question.....
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/24/2020 02:58 am
Er... I didn't say that. :D

I said integration of the soft-capture mechanism and the self-aligning latches (needed because... well, how else does your soft capture mechanism work?), not the hard capture latches. Hopefully you can can now see why I was very confused by your question. ;)

But since you brought it up, I do think the soft capture and hard capture system will share hardpoint locations on the skirt. Co-location, but not necessarily integration (though integration is probably perferred, eg like how the latches and pushers were integrated during the transition from F9 v1.0 to F9 v1.1).

"Self-aligning" is not necessarily a modifier for "latches".  All you need is a guide pin.  But I agree that there's nothing preventing colocation of the soft and hard systems.  A lot probably depends on whether the staging latches and the hard capture latches are one and the same.

I suspect we've gotten wrapped around the axle in terms of what "integrated" means.  ISTM that you're asserting that integration of the soft and hard systems per se isn't required.  I agree with that.

Yep, agreed. I think we were just talking past each-other.

That's one possibility. The other possibility is that the more numerous the soft capture system, the more likely that retraction will "click" the unlatched ports in place.

If you're trying to do that with three latches, it won't work (no leverage). Four is the minimum number of soft capture mechanisms that could be expected to do that, though.

Not sure I'm following you here. Most alignment is done during stabilization, and 3 points will give you all the leverage you need there.  (Note that the soft system has to be able to pull and push, but that's current state of the art.)

If you're saying that retraction causes subsequent misalignments... I'm not sure I buy that.  That seems to be a "Doctor, Doctor, it hurts when I do this" kind of problem.  A soft system that can damp out misalignments can surely prevent them from creeping in during retraction.

Yeah, sorry about the confusion. I'm not saying that.

I was responding to the concerns you expressed earlier about multiple soft capture systems increasing the chances of a capture failure, to whit:

If you're willing to go to 4 soft-captures, you have some more leeway, but then you've got more complexity and more chance of a failure on capture


And also, are we expecting a lot of error? Dragon docks with centimeter precision, and I don't see any good reasons why Starship would be any different. Clearly you can scale the RCS bit to achieve whatever precision you need.

I don't think that scaling is linear, because the moments of inertia are in some cases 2-3 orders of magnitude larger with Starship.

If anything that makes alignment easier, since the same impulse results in a smaller rotation rate.

I'll also be interested to see what kind of minimum impulse bit the Starship RCS system comes up with.  It may turn out that there's a little RCS and a big RCS, but that's pretty complicated.  If they go with only big RCS, things are going to be interesting.

Yeah, personally I do expect coarse and vernier RCS. It just makes far too much sense.


Yes, my baseline assumption is that SpaceX is good enough to hit the "superior design" in this case.

Even superior designs have trade spaces.  In the grand scheme of things, saving a couple hundred kgs in optimized latches probably isn't exactly low-hanging fruit in terms of mass reduction.  SpaceX likes "superior", but they often like "cheap" even more.

Not just the mass reduction. Slapping a few IDSSs on the bottom of Starship is a big packaging penalty. As you're demonstrating, there's not a lot of room to put extra stuff down there.

But yes, I do expect SpaceX will work this problem From First Principles rather than practicing catalog engineering. I think this expectation is amply supported by our historical experience with SpaceX.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/24/2020 03:13 am
Again, SpaceX is no stranger to weird [a]symmetry. The Falcon 9 is mostly 4-symmetric (legs, grid fins), but the docking latches are 3-symmetric, and the RCS thrusters are 2-symmetric.

I have no problem with weird asymmetry, but there are real estate constraints around the skirt.  If you can figure out how to put a soft capture system covering up a landing leg, then things get a little less crowded.  Otherwise, if you have a 0.5-1.0m capture target, it takes up some space.

Agreed. One option is alignment guides on either side of the legs, rather than a single capture target in the middle.


There's plenty of real estate open on the skirt, but there's considerably less if you try to put two things on the same symmetry axis, because the most open spaces on the skirt have an RVac on the opposite side.

I'm not sure why we're treating the cargo pod area as "open real estate." It has to hold... cargo pods, right? Those vehicles require refueling too.

"Running into a landing leg" isn't a big problem, because the landing legs are folded during refilling. Surely SpaceX is capable of designing a docking latch that doesn't physically intersect the landing leg or its deployment arc.

Of course they are.  But that trims the amount of real estate down by quite a bit.  If you then further constrain the problem by insisting on opposing latches (i.e., any kind of even symmetry), then things are getting pretty cozy.

If you're not interfering with the cargo pods, that constraint doesn't make it any more cozy.

]And again, the landing legs need a structural hardpoint. So combining the soft latch, hard latch, and landing legs into a single unified mechanism saves mass, and this is likely the path SpaceX will take IMO.

I'm completely prepared to believe that your average SpaceX ME is smarter than I am--especially about ME problems.  I'm just having a failure of imagination in how a soft capture anchor can get out of the way of a landing leg without some sort of whiz-bang way of folding it out of the way, post-Mr. Fusion DeLorean-style, to make way for the leg--and then it's not really sharing the same hard point.

Agreed, something "fold-away" wouldn't be very elegant. It will be interesting to see what they come up with.

Yeah, I think the real answer is you just orient the axis of symmetry up/down instead of left/right. This way the refueling lines are as far apart from each-other as possible and as far apart from the Raptor engines as possible.

That's an invariant change, unless the size and number of RVacs changes.

I'm afraid I don't follow. If you look back at the geometry, the change I described definitely doesn't result in an identical situation.

Two things:

1) If you're still worried about keeping max distance between fuel and oxidizer lines, then something like armchairfan's suggestion above solves that problem completely without doing anything exotic.

Ditto, cargo pods.

2) Note that all of the legs are next to RVac nozzles.  Again, if you have magic fold-away soft targets, that's no big deal.  Otherwise, it's not an ideal environment for a piece of hardware designed to lunge smoothly through a decent-sized range of positions and velocities.

The legs are positioned 1/4 and 3/4 of the way between R-Vacs. They're half-way between the half-way  :D(which, again, the half-way itself is blocked by pods).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/24/2020 03:17 am
Is there any particular reason a solid attachment is needed?  Couldn't you use a mechanism similar to what aircraft use for inflight refueling, and basically just have a hose (well, two hoses) running from the tanker to the receiving vehicle?

Do we have flexible hose material that's compatible with cryogenic liquids?

Also, it would need a robot arm or some other way to "fly" the connectors together. At that point you might as well simplify it as much as possible, and you're back to a simple extending mechanism.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/24/2020 04:24 am
I'm not sure why we're treating the cargo pod area as "open real estate." It has to hold... cargo pods, right? Those vehicles require refueling too.

Yes, I've been studiously ignoring the cargo pods.  Three options:

1) They're there all the time, in which case mould-line real estate is really tight.
2) They're removed for missions requiring refueling.
3) They don't exist any more.

Assuming they exist, the lids could be structural, with the capture devices mounted on them.

I've kinda been wondering if SpaceX had a plan for the LSS/HLS to receive cargo by pushing cargo pods from a (mostly) tanker to the LSS.  That obviously can't happen if the cargo pods are integral  to the docking system.  But it looks to me that NASA simply isn't that interested in landing cargo with crew, at least in the early days of Artemis.

Quote

Quote
Yeah, I think the real answer is you just orient the axis of symmetry up/down instead of left/right. This way the refueling lines are as far apart from each-other as possible and as far apart from the Raptor engines as possible.

That's an invariant change, unless the size and number of RVacs changes.

I'm afraid I don't follow. If you look back at the geometry, the change I described definitely doesn't result in an identical situation.

You're right; it's not invariant.  But it also doesn't lead to the fueling lines being far apart from the nozzles.  (I'll refrain from harping about the landing legs...)

I still don't see why armchairfan's suggestion doesn't allay your fears.  I attached it below.  I think it's pretty trivial to put the lines against the skirt and not interfere with the cargo pods.  This version also assumes that the capture devices are part of the lids of the pods.

Another thing:  If you go with a single line for LOX and one for LCH4, then the lines must be on the axis of symmetry, or you don't get them to line up after a two-axis rotation.  On the other hand, if you go with the top/bottom pair for each of LOX and LCH4, any symmetrical arrangement will always result in tops being paired with bottoms, just as you'd like.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/24/2020 05:05 am
If I may, let's re-cap on what we currently know has been suggested and used for fueling by SpaceX.

All renders of SS on-orbit refueling with versions similar to what we now see have the craft tail to tail and inverted relative to eachother. See the first image from Neopork below.

However earlier renders, while still tail to tail are not inverted, see second image below (source unknown, Google image search)

Finally they are currently using a single GSE connection location with SN prototypes on the dorsal (Z-?) part of the skirt as shown (red circle) in the excellent aerial photo of SN8 by RGV shown third below.

What does this mean?
- tail to tail so far is consistent.
- 180o inversion appears negotiable over time.
- current location of fuel hardware is not consistent with 180o rotation when tail to tail, but is with early renders.

So ultimately the question is how close is the current fuel hardware location to the final location?
Everything else depends on this question.....

To further add to the confusion, we have this from the Sept. 2019 update preso:

[zubenelgenubi: Attach image files. Do not embed them.]

Note that this has a different (presumed) y-axis than Rafael has been using.  Also note that this picture still has exterior chines/housings for the landing legs.  I'd assume that they want the mould line to be pretty well set before beginning belly-flop and swoop-n-slam tests, so I'd assume that this is at least somewhat out of date.

I've been making a couple of assumptions which might be incorrect:

1) That the external placement of the lines is temporary.  Unless they plan on running an LCH4 line through the LOX tank (which is a really bad idea), at least the LCH4 line(s) have to be external, and they likely have to put them into the flap chines (which would be consistent with the picture I included).

2) When you add in the RVacs, you want the lines as far away from them as possible.  The picture doesn't appear to be very concerned with that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 10/24/2020 08:45 am
I'm not sure why we're treating the cargo pod area as "open real estate." It has to hold... cargo pods, right? Those vehicles require refueling too.
Yes, I've been studiously ignoring the cargo pods.  Three options:
1) They're there all the time, in which case mould-line real estate is really tight.
2) They're removed for missions requiring refueling.
3) They don't exist any more.

I wonder if the diagrams (example below) have been misleading people.

Do you really need three 800mm docking rings? Are the fuel-lines really 300mm across each?

The entire current fuel+ox assembly on the test SS's, including all the surrounds and stand-offs, fits within the area people are showing for a single fuel connector. The connectors in the SpaceX render would fit between the cargo-pods and landing-gear, even if the landing gear was the current fold-out type. With room to spare.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/24/2020 10:22 am
I'm not sure why we're treating the cargo pod area as "open real estate." It has to hold... cargo pods, right? Those vehicles require refueling too.
Yes, I've been studiously ignoring the cargo pods.  Three options:
1) They're there all the time, in which case mould-line real estate is really tight.
2) They're removed for missions requiring refueling.
3) They don't exist any more.

I wonder if the diagrams (example below) have been misleading people.

Do you really need three 800mm docking rings? Are the fuel-lines really 300mm across each?

The entire current fuel+ox assembly on the test SS's, including all the surrounds and stand-offs, fits within the area people are showing for a single fuel connector. The connectors in the SpaceX render would fit between the cargo-pods and landing-gear, even if the landing gear was the current fold-out type. With room to spare.

Fair point.  Scale is driven more by the size that a legible label will fit in, rather than... you know... reality.

I think the biggest mistake, though, is thinking that the lines ran on the inside.  That may change a lot of things.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: livingjw on 10/24/2020 02:09 pm

.... Unless they plan on running an LCH4 line through the LOX tank (which is a really bad idea), at least the LCH4 line(s) have to be external, and they likely have to put them into the flap chines (which would be consistent with the picture I included).
.....

I'm confused. The LCH4 main feed lines have always run through the LOX tank. Are you talking about the vent line?

John
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 10/24/2020 02:54 pm
Yes, I've been studiously ignoring the cargo pods.  Three options:
1) They're there all the time, in which case mould-line real estate is really tight.
2) They're removed for missions requiring refueling.
3) They don't exist any more.
I wonder if the diagrams (example below) have been misleading people.
Do you really need three 800mm docking rings? Are the fuel-lines really 300mm across each?
Fair point.  Scale is driven more by the size that a legible label will fit in, rather than... you know... reality.

But then people, yourself included, are treating those label sizes as if they are real. "The mould-line real estate is really tight". The map is not the territory.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/24/2020 09:08 pm

.... Unless they plan on running an LCH4 line through the LOX tank (which is a really bad idea), at least the LCH4 line(s) have to be external, and they likely have to put them into the flap chines (which would be consistent with the picture I included).
.....

I'm confused. The LCH4 main feed lines have always run through the LOX tank. Are you talking about the vent line?

John

You're not confused, I am.  I fell into a risk hole.  Piping fuel through oxidizer and vice versa is pretty much the cost of doing business, isn't it?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/24/2020 09:16 pm
Fair point.  Scale is driven more by the size that a legible label will fit in, rather than... you know... reality.

But then people, yourself included, are treating those label sizes as if they are real. "The mould-line real estate is really tight". The map is not the territory.

Well...  The prop lines are clearly too big.  As for the soft captures, I'm not sure.

The main reason the current IDSS soft capture rings are the size they are is because they need to keep the pressure tunnel clear, once the petals have been removed.  That's obviously not a requirement here.

But there's another requirement, which is simply that they be big enough to accommodate the expected positioning and velocity errors that occur during docking.  Since we don't know what those errors are likely to be for two Starships, it's hard to say how big the captures have to be.  I'd kinda guess that they don't need to be 1m wide, but I wouldn't be surprised if they had to be half a meter.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 11/10/2020 02:09 am
A small correction change is needed in this image from previous posts. The right hand diagram is the left hand diagram rotated 180 degrees. Note that the northwest cargo bin and propellant juncture is closer to one Rvac than the other one. After rotation, they end up in the southeast position, as they should. But when the second diagram is folded back onto the first diagram using the suggested fold line, these propellant junctures do not line up with their corresponding junctures on the first SS.

Propellant lines will only line up if they are located exactly 120 degrees apart (trilateral symmetry). Perhaps color code the cargo bins to trace their rotated positions?

Added: This is a nice image and worth making applicable in multiple contexts.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: baddux on 11/10/2020 11:11 pm
Both of these militate toward starting out refueling in VLEO, with "tankers" that are simply ordinary Starships with no payload.  You launch a series of "lift tankers" to meet the needs of the mission you're mounting.  The first lift tanker stays in VLEO and becomes the "accumulation tanker".  The remaining lift tankers fuel the accumulation tanker until enough prop has accumulated to satisfy the mission requirement.  Only then does the payload Starship launch and rendezvous with the accumulation tanker, so there's only one docking and refueling operation.  After that, the accumulation tanker goes home to get ready for the next mission.

I came here to ask exactly this thing. Is there any reason why for example Mars mission is not launched this way? So the "single use fuel depo" tanker is sent to orbit first, it is filled with ~5 times and only when its ready the actual payload Starship is launched. It is refueled only once with the tanker which is already waiting and it does not have to wait for 5 launches. This is simpler and safer mission profile than the one where the payload Starship waits for 5 refuelings.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 11/11/2020 01:25 am
Is there any reason why for example Mars mission is not launched this way? So the "single use fuel depo" tanker ...

In this thread, that's what people are talking about when they say "accumulation tanker".

... is sent to orbit first, it is filled with ~5 times and only when its ready the actual payload Starship is launched. It is refueled only once with the tanker which is already waiting and it does not have to wait for 5 launches. This is simpler and safer mission profile than the one where the payload Starship waits for 5 refuelings.

I suspect Musk's reasoning is two-fold:

1) If Starship is as rapid turn-around as SpaceX wants (or at least Musk wants), then having the Mars crew waiting a few days in LEO isn't a great burden when they are planning a 3+ month trip. And it's time to check out the vehicle on-orbit, let everyone get their spacelegs, etc.

2) OTOH, if refuelling takes longer, then when flying as many Starships to Mars as Musk has talked about in each synod, you don't want one dedicated accumulation tanker having to sit in orbit for each Mars Starship. In effect, you're doubling the number of ships that are effectively out-of-service. OTOH, use the Mars Starship to free up the accumulation tanker to repeatedly launch propellant up.

But I think most people in this thread are assuming either an accumulation tanker or an actual specialised orbital depot.

[edit: words.]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/15/2020 06:25 pm
Propellant lines will only line up if they are located exactly 120 degrees apart (trilateral symmetry).

As long as their position are mirrored across the flip plane, you can use any angle you want. This diagram happens to use 120 degrees and (almost) trilateral symmetry, but it's not necessary.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/15/2020 06:30 pm
you don't want one dedicated accumulation tanker having to sit in orbit for each Mars Starship. In effect, you're doubling the number of ships

That's why you accumulate in the Mars Starships, and late-load passengers using a high capacity P2P-derived short-haul taxi. Solves both problems at once.

The "downside" is that your Mars ship needs a docking tunnel, but honestly that's not a bad thing if you need to evacuate to another Starship in the fleet during transit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: FutureSpaceTourist on 11/24/2020 07:20 pm
https://twitter.com/starshipfairing/status/1307788619642482692

Quote
Starship Tanker v2

With the Superheavy booster, this will bring approximately 50% more fuel to orbit than the default Starship.

@elonmusk, what do you think? Please RT so he can see it!

Summary, uses, and specs in the comments - go read it!

https://twitter.com/elonmusk/status/1331310252927676416

Quote
Not bad! We will definitely need more engines if we make the cargo bay all propellant, but it’s probably smarter than a whole new shorter external hull.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/24/2020 11:27 pm
https://twitter.com/starshipfairing/status/1307788619642482692

Quote
Starship Tanker v2

With the Superheavy booster, this will bring approximately 50% more fuel to orbit than the default Starship.

@elonmusk, what do you think? Please RT so he can see it!

Summary, uses, and specs in the comments - go read it!

https://twitter.com/elonmusk/status/1331310252927676416

Quote
Not bad! We will definitely need more engines if we make the cargo bay all propellant, but it’s probably smarter than a whole new shorter external hull.

Makes me wonder if it would make sense to just launch partly filled at first, with the "evolution path" being to simply increase the prop load as you add / uprate engines...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: gin455res on 03/20/2021 07:42 pm
I've seen it posted that falcon 9 gets 22 tons to orbit with a 4 ton upperstage. So roughly 2:13 upperstage to payload.


whereas starship is 120:120 ish.


Would it be worth eventually designing a tanker StarShip that has drop tanks. So that it doesn't lug so much empty tankage to orbit every time.


I wonder how much extra fuel would get to orbit if the tanker expended 2/3 of its tankage on the way. And how much would that expended tankage cost as a fraction of marginal launch cost.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AC in NC on 03/20/2021 08:40 pm
I've seen it posted that falcon 9 gets 22 tons to orbit with a 4 ton upperstage. So roughly 2:13 upperstage to payload whereas starship is 120:120 ish.

Would it be worth eventually designing a tanker StarShip that has drop tanks. So that it doesn't lug so much empty tankage to orbit every time.

No.  A fully reusable system is so cheap that you don't care about the payload fraction comparisons to non-reusable systems.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: vaporcobra on 03/20/2021 09:28 pm
I don't remember if it has been in this thread, but there have been discussions about the propellant transfer docking on these forums.

SuperHeavy and Starship will have some way of mating and holding on to each other during ascent.  Most likely, in my opinion, is for that mechanism to be used during tail-to-tail docking of two Starships as well.  The mechanisms on SuperHeavy and Starship would be identical and symmetric, and can then be used for both purposes.  (In theory, you could then also put a SuperHeavy upside down on top of another SuperHeavy and have them latch on to each other, but why one would do that is beyond me...)  As for exactly how that mechanism will work, I don't think we have any good information.  But I don't think the feet will be involved.

And some kind of quick disconnect on the propellant tanking lines.  I suspect those will connect separately, after the docking.  I.e, they would still be separated by a few decimeters, and then lowered towards each other until the QDs connect.  The QDs are probably much more sensitive to mechanical damage than the docking mechanism, so you want to avoid them touching each other during the docking maneuver, as a slight misalignment between the ships could otherwise damage the QDs.

Likewise, I don't think we have anything concrete about the details of how those QDs will work, or how the tanking lines will be lowered towards each other.

FWIW, NASA has already awarded SpaceX $3M to develop "cryogenic fluid coupling" prototypes for Starship with MSFC in 2019. Not unreasonable to assume that the umbilical panels we've seen on Starship prototypes are also prototypes (to some degree) of the QD mechanisms needed for hands-free SS->SH and SS->SH mating and prop transfer.

https://www.nasa.gov/press-release/nasa-announces-new-tipping-point-partnerships-for-moon-and-mars-technologies
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 03/20/2021 09:32 pm
This is what should be built.  This is awesome. 

https://youtu.be/T9EFqPcoTwU


Then build this.

https://twitter.com/SpacexVision/status/1372939067130793986/photo/1?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1372939067130793986%7Ctwgr%5E%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Fforum.nasaspaceflight.com%2Findex.php%3Ftopic%3D47144.1040
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: vaporcobra on 03/20/2021 10:00 pm
This is what should be built.  This is awesome. 

https://youtu.be/T9EFqPcoTwU

It's hard to imagine why SpaceX would go for such a complex, custom solution when they already intend to build tanker Starships, though. Remove recovery hardware, add insulation, and develop a node you can dock those lightly modified Starships to and you end up with a modular, cheap(ish) orbital tank farm that uses the same production line you've already built.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Slarty1080 on 03/20/2021 10:34 pm
This is what should be built.  This is awesome. 

https://youtu.be/T9EFqPcoTwU

It's hard to imagine why SpaceX would go for such a complex, custom solution when they already intend to build tanker Starships, though. Remove recovery hardware, add insulation, and develop a node you can dock those lightly modified Starships to and you end up with a modular, cheap(ish) orbital tank farm that uses the same production line you've already built.
The trouble is the tanker is only good for re-tanking in the orbit it is launched into (or very similar). Anything that needs a different inclination orbit wouldn't work. Also using a Starship as a tanker provides as much propellant as you will need for any Starship mission.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 03/21/2021 01:32 am
Did anyone watch the video? 

This 13m orbital depot can hold launched with a Superheavy booster and hold more fuel than a Starship.  It also has 3 engines on it to move it anywhere it is needed.  One in orbit for outgoing moon Starships, one in outgoing Mars Starships.  One could be sent to Mars orbit and refueled with Mars fuel from Mars Starships.  One could be sent to the Artemis orbit for moon lander Starships to refuel for another landing.  The fuel depot would hold enough fuel to refill two Starships.  It is larger than a Starship but would not have to deorbit and could be used in space only.   

One the second idea is to use one of these large depots as a huge space tug to have 10 docking ports for outgoing cargo on a 9 month journey to Mars.  There is another idea from this same website to use one as a large space station that would be larger than ISS in one launch. 

Since this is to be "in space only" it is to be built from composites, less mass to orbit, but a larger diameter to hold more fuel or become a tug or space station. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AC in NC on 03/21/2021 01:40 am
Did anyone watch the video? 

This 13m orbital depot can [be] launched with a Superheavy booster and hold more fuel than a Starship.  It also has 3 engines on it to move it anywhere it is needed. One in orbit for outgoing moon Starships, one in outgoing Mars Starships.  One could be sent to Mars orbit and refueled with Mars fuel from Mars Starships.  One could be sent to the Artemis orbit for moon lander Starships to refuel for another landing.  The fuel depot would hold enough fuel to refill two Starships.  It is larger than a Starship but would not have to deorbit and could be used in space only.   

One the second idea is to use one of these large depots as a huge space tug to have 10 docking ports for outgoing cargo on a 9 month journey to Mars.  There is another idea from this same website to use one as a large space station that would be larger than ISS in one launch. 

Since this is to be "in space only" it is to be built from composites, less mass to orbit, but a larger diameter to hold more fuel or become a tug or space station.

Yes.  Watched the video.  Doesn't seem very compelling.

The strikethrough marks things that are no different than a "standard" SS Tanker so aren't really advantages. 
The underlined, Composites, cargo docking ports, and franken-13m-Tanker-Space-Station just seem like solutions looking for problems and a lot of unnecessary distracting re-engineering.

The proposal seems that it more-or-less boils down to an expensive variant to provide 2x fuel capacity vs. just using 2+ Accumulation Tankers.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: vaporcobra on 03/21/2021 05:28 am
Did anyone watch the video? 

This 13m orbital depot can [be] launched with a Superheavy booster and hold more fuel than a Starship.  It also has 3 engines on it to move it anywhere it is needed. One in orbit for outgoing moon Starships, one in outgoing Mars Starships.  One could be sent to Mars orbit and refueled with Mars fuel from Mars Starships.  One could be sent to the Artemis orbit for moon lander Starships to refuel for another landing.  The fuel depot would hold enough fuel to refill two Starships.  It is larger than a Starship but would not have to deorbit and could be used in space only.   

One the second idea is to use one of these large depots as a huge space tug to have 10 docking ports for outgoing cargo on a 9 month journey to Mars.  There is another idea from this same website to use one as a large space station that would be larger than ISS in one launch. 

Since this is to be "in space only" it is to be built from composites, less mass to orbit, but a larger diameter to hold more fuel or become a tug or space station.

Yes.  Watched the video.  Doesn't seem very compelling.

The strikethrough marks things that are no different than a "standard" SS Tanker so aren't really advantages. 
The underlined, Composites, cargo docking ports, and franken-13m-Tanker-Space-Station just seem like solutions looking for problems and a lot of unnecessary distracting re-engineering.

The proposal seems that it more-or-less boils down to an expensive variant to provide 2x fuel capacity vs. just using 2+ Accumulation Tankers.

100%. For the space station bit in particular, a single off-the-shelf crewed Starship will already have a greater pressurized volume than the ISS by default. If, for some inconceivable reason, SpaceX had a need for more than that, docking two or three or four Starships would be a FAR easier engineering challenge than building a new spacecraft from scratch.

But this is mostly off-topic for this thread. As far as refueling and orbital propellant depots go, it's just hard to imagine a near-term need to develop a completely new 'solution' when something as simple as staging an existing tanker (or even just unmodified Starships) with tacked-on insulation in orbit and preemptively refueling them would be almost equally effective. Launch your temporary tanker-turned-depot, fully refuel it in orbit, launch the applicable cargo or crew Starship after the orbiting tanker is full, dock and refuel in one fell swoop, and head on your way. Recover the tanker or leave it in orbit and rinse and repeat. No having to wait hours to days in orbit, no need to stack risks w/ 4+ consecutively successful time-sensitive refuelings, and no need to re-engineer the wheel to get those results.

One big drawback of the proposed 13m vehicle we're discussing is that it's only maybe 2-3 times more capable than a 9m tanker, which is marginal at best. At that scale, the benefits are irrelevant when we're talking about a handful of Starships sent per Hohmann transfer. If/when that jumps to 10-100+, the 13m tanker's benefits are so marginal that it doesn't really solve anything, as a single 13m wide cylinder is probably one or two magnitudes too small to make a dent. Then you still end up having build tons of them and the end result is just the same thing you'd get by using existing 9m vehicles and tooling but with slightly different building blocks.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: libra on 03/21/2021 06:12 am
There is a "clash of visions" here. From my humble POV.

Launching lots of big tanks and assembling them to create a depot is kind of the old NASA way of doing things: this is how von Braun Mars 1969 would have done it. And this is how they do it in Stephen Baxter "Voyage".
...
With expendable Saturn V !  And there it is justified, because the "tanker rocket" has to be rebuild each time, and so the infrastructure allows to "wait".

Seems Elon wants to do it a very different way: no orbital infrastructure. So one partially fueled Starship - bound for Mars later - would be refueled "in serie" and pretty fast by ten launches; ten more Starships each with 150 mt of propellants. The Starships would go into orbit, offload their propellants, return, and start all over again.
Because the vehicle is reusable and with fast turnaround, there is no point for a massive orbital infrastructure to store the fuel: the Mars-bound-Starship IS the depot all by itself.

(https://upload.wikimedia.org/wikipedia/commons/c/c1/Nuclear_Shuttle_fuel_loading.jpg)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 03/21/2021 01:20 pm
One word.
Inclination change is expensive.
Every moon/mars opportinity will require different inclinations.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: _MECO on 03/21/2021 02:04 pm
Did anyone watch the video? 

This 13m orbital depot can [be] launched with a Superheavy booster and hold more fuel than a Starship.  It also has 3 engines on it to move it anywhere it is needed. One in orbit for outgoing moon Starships, one in outgoing Mars Starships.  One could be sent to Mars orbit and refueled with Mars fuel from Mars Starships.  One could be sent to the Artemis orbit for moon lander Starships to refuel for another landing.  The fuel depot would hold enough fuel to refill two Starships.  It is larger than a Starship but would not have to deorbit and could be used in space only.   

One the second idea is to use one of these large depots as a huge space tug to have 10 docking ports for outgoing cargo on a 9 month journey to Mars.  There is another idea from this same website to use one as a large space station that would be larger than ISS in one launch. 

Since this is to be "in space only" it is to be built from composites, less mass to orbit, but a larger diameter to hold more fuel or become a tug or space station.

Yes.  Watched the video.  Doesn't seem very compelling.

The strikethrough marks things that are no different than a "standard" SS Tanker so aren't really advantages. 
The underlined, Composites, cargo docking ports, and franken-13m-Tanker-Space-Station just seem like solutions looking for problems and a lot of unnecessary distracting re-engineering.

The proposal seems that it more-or-less boils down to an expensive variant to provide 2x fuel capacity vs. just using 2+ Accumulation Tankers.

100%. For the space station bit in particular, a single off-the-shelf crewed Starship will already have a greater pressurized volume than the ISS by default. If, for some inconceivable reason, SpaceX had a need for more than that, docking two or three or four Starships would be a FAR easier engineering challenge than building a new spacecraft from scratch.

But this is mostly off-topic for this thread. As far as refueling and orbital propellant depots go, it's just hard to imagine a near-term need to develop a completely new 'solution' when something as simple as staging an existing tanker (or even just unmodified Starships) with tacked-on insulation in orbit and preemptively refueling them would be almost equally effective. Launch your temporary tanker-turned-depot, fully refuel it in orbit, launch the applicable cargo or crew Starship after the orbiting tanker is full, dock and refuel in one fell swoop, and head on your way. Recover the tanker or leave it in orbit and rinse and repeat. No having to wait hours to days in orbit, no need to stack risks w/ 4+ consecutively successful time-sensitive refuelings, and no need to re-engineer the wheel to get those results.

One big drawback of the proposed 13m vehicle we're discussing is that it's only maybe 2-3 times more capable than a 9m tanker, which is marginal at best. At that scale, the benefits are irrelevant when we're talking about a handful of Starships sent per Hohmann transfer. If/when that jumps to 10-100+, the 13m tanker's benefits are so marginal that it doesn't really solve anything, as a single 13m wide cylinder is probably one or two magnitudes too small to make a dent. Then you still end up having build tons of them and the end result is just the same thing you'd get by using existing 9m vehicles and tooling but with slightly different building blocks.

Yeah, manufacturing commonality is the word of God here. You know how SpaceX is turning what is essentially partially completed Starships into ground service equipment over there? It's because they're all tooled up to produce those 9 meter tanks and now that they're good at it, making 9 meter tanks for whatever application is a breeze.

So in my mind, if I were SpaceX and I wanted to place an orbital propellant depot up there here's what I would do:

1. Build several "tanker" variant Starships like mentioned earlier but without the flaps, TPS, extra engines, landing legs, and header tanks. Instead, install docking hardware on the sides of the vehicle as well as most of the plumbing for fuel transfer. Solar arrays and insulative coatings may be added to the outside of the vehicle as needed.

2. Launch these "depot Starships" into orbit one at a time empty, with no additional payload. Depending on how this works out it may be possible to get them there on three sea level engines alone.

3. As each depot Starship gets placed into orbit, have it rendezvous with each previous depot Starship until there is a line of two or three or four daisy-chained together, side to side.

4. Finally, a series of crewed Starships equipped with service teams will ascend in order to begin converting the assembly into a depot. The teams will go on EVA and manually connect the plumbing between each of the depot ships, weld the docking hardware into a more sturdy and permanent configuration, and install the final parts of cryogenic refrigeration systems. At the end of this procedure, the crews will uninstall the raptors from the depot ships and stow them aboard their own ship before deorbiting and coming home.

Obviously this requires the development of tons of brand new technologies and procedures. It will require an unprecedented amount of EVA construction, but it completely avoids the need to sink huge amounts of research and development funds into creating something larger than 9 meters that launches once.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Redclaws on 03/21/2021 02:59 pm
There is a "clash of visions" here. From my humble POV.

Launching lots of big tanks and assembling them to create a depot is kind of the old NASA way of doing things: this is how von Braun Mars 1969 would have done it. And this is how they do it in Stephen Baxter "Voyage".
...
With expendable Saturn V !  And there it is justified, because the "tanker rocket" has to be rebuild each time, and so the infrastructure allows to "wait".

Seems Elon wants to do it a very different way: no orbital infrastructure. So one partially fueled Starship - bound for Mars later - would be refueled "in serie" and pretty fast by ten launches; ten more Starships each with 150 mt of propellants. The Starships would go into orbit, offload their propellants, return, and start all over again.
Because the vehicle is reusable and with fast turnaround, there is no point for a massive orbital infrastructure to store the fuel: the Mars-bound-Starship IS the depot all by itself.

(https://upload.wikimedia.org/wikipedia/commons/c/c1/Nuclear_Shuttle_fuel_loading.jpg)

I think Libra nails it.  Musk has - to my knowledge - never suggested starship would involve any form of orbital depot.  He’s repeatedly described what Libra just did; launch rapidly and accumulate fuel in the departing ship.  The only variant of this I see being terribly likely is filling *one* ship on orbit to fill a departure ship, so humans don’t have to wait around/experience 10 (or whatever) refuelings while onboard.  This requires almost no specific technology or engineering at all - you’re already doing refueling and long duration hold of cryogenics (though not in LEO and maybe not in the main tanks).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 03/21/2021 03:17 pm
I think Libra nails it.  Musk has - to my knowledge - never suggested starship would involve any form of orbital depot.
Ironically, Musk tweeted this yesterday...

Quote
An orbital propellant depot optimized for cryogenic storage probably makes sense long-term
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ludus on 03/21/2021 03:35 pm
Much of the disagreement seems to stem from whether the application is a large synodic fleet going to Mars together or steady state operations in earth orbit. For a big fleet, that necessarily is waiting around for a long time anyway, it doesn’t make sense to require a parallel fleet of accumulation tankers.

For day to day steady operations, it doesn’t make sense to make a priority mission wait around and dock 6-10 times with tankers when you could assign one tanker to accumulate and simplify the logistical planning. For steady operations it’s a major benefit to be able to just keep launching tankers 24/7 all the time rather than starting only when a mission has been launched and is waiting. It makes risk mitigation sense to only dock once with higher value Starship with cargo aboard.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: joek on 03/21/2021 04:53 pm
I think Libra nails it.  Musk has - to my knowledge - never suggested starship would involve any form of orbital depot.
Ironically, Musk tweeted this yesterday...
Quote
An orbital propellant depot optimized for cryogenic storage probably makes sense long-term

One reason everyone was stepping carefully around the word "depot" (this goes back a ways, circa 2019 and earlier):
Quote
Senator Shelby called NASA and said if he hears one more word about propellant depots he’s going to cancel the space technology program.
Thus IMO why Musk studiously avoided "depot"; instead "tankers".  Things have loosened up a bit since.  Why the change?  Longer conversation and OT for this thread (Shelby, SLS, Artemis, SpaceX-Starship, depots, ...  makes for a potentially toxic discussion).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Redclaws on 03/21/2021 07:00 pm
I think Libra nails it.  Musk has - to my knowledge - never suggested starship would involve any form of orbital depot.
Ironically, Musk tweeted this yesterday...

Quote
An orbital propellant depot optimized for cryogenic storage probably makes sense long-term

Hah, bit of egg on my face then.  Interesting.  I think it's an interesting discussion anyway, but I still don't think it's likely as an early phase.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Redclaws on 03/21/2021 07:03 pm
I think Libra nails it.  Musk has - to my knowledge - never suggested starship would involve any form of orbital depot.
Ironically, Musk tweeted this yesterday...
Quote
An orbital propellant depot optimized for cryogenic storage probably makes sense long-term

One reason everyone was stepping carefully around the word "depot" (this goes back a ways, circa 2019 and earlier):
Quote
Senator Shelby called NASA and said if he hears one more word about propellant depots he’s going to cancel the space technology program.
Thus IMO why Musk studiously avoided "depot"; instead "tankers".  Things have loosened up a bit since.  Why the change?  Longer conversation and OT for this thread (Shelby, SLS, Artemis, SpaceX-Starship, depots, ...  makes for a potentially toxic discussion).

There's obviously a lot of ugly context around "depots" for NASA, etc, but Musk didn't just "say" tankers - He showed (consistently across the various ITS, BFR, SS/SH presentations) a bunch of ships launching to fill one ship.

But, I suppose if you put one ship up there and fill it, then it fills the departing ship, well, if you do that a few times with one of them...  It does start to look pretty depot-ish.  I just think it's highly likely that the early stuff will be just be a standard or *lightly* modified starship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: alastairmayer on 03/22/2021 03:27 am
One word.
Inclination change is expensive.
Every moon/mars opportinity will require different inclinations.

That's four words (twelve, if you count the next sentence).

But it's most expensive in LEO. The higher (further) you're going, the easier it gets. Worst case, add another tanker load of fuel.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: mikelepage on 03/22/2021 07:23 am
I don't think this will work. They need to have bilateral symmetry. Starships will approach each out back to back and inverted so the heatshields of each will be on different sides
Why inverted? Back to back is enough to flip connection points, with, say, a male connector on the left and a female connector on the right.

Exactly. I must admit I've been confused by the need for inversion every time I've seen it depicted, because it seems to me your heat management processes will be served better by having both heat shields point in the same direction, and your electricity generation helped by having your solar panels pointing in the same direction. Is inversion necessary just to put the CH4 and O2 lines on opposite sides of the craft? If so it seems like quite a tradeoff. Maybe I'm missing something?

Also, I tend to find many of the 2D depictions in this thread somewhat hard to interpret, so hoping my putting my 3D models into this will help visualisation. I'm still working on my next "Starship Design speculation" video, and we now know from SN8-11 that the orientation of the raptors is two sea-level raptors on the "ventral" side and two vac raptors on the "dorsal" side.

Thank you to whoever it was who said that the prop lines would make most sense if they ran along the inside edge of the skirt (that's one reason I've made the landing leg design as it is - more on that in the video). Also I prefer the idea from this thread  that each tank would have two connectors that would be male and female - one routed to the "top" of the tank and one to the "bottom" (or what would be top and bottom if the Starship was standing with the pointy end up). Here I've depicted the CH4 lines in red and the O2 lines in blue.

I've done a second version with separate gas connection pipes (8-pipe version) that would connect to the tanks separately from the liquid fill lines? although I'm not sure whether the ability to handle the gas and liquid propellants separately would be worth the extra mass and docking complexity. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BZHSpace on 03/22/2021 02:29 pm
It is correct that Superheavy/Starship will be fuel by the bottom of the booster ? So maybe to stack Starship and Superheavy they will need an inner ring to join the CH4/LOX pipe that could be use also for Starship/Starship orbital refueling operation ?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: steveleach on 03/22/2021 04:05 pm
It is correct that Superheavy/Starship will be fuel by the bottom of the booster ? So maybe to stack Starship and Superheavy they will need an inner ring to join the CH4/LOX pipe that could be use also for Starship/Starship orbital refueling operation ?
Using the same plumbing for fuelling the stack and orbital refuelling is exactly their plan, and the reason why they are fuelling from the bottom.

I'm not sure what you mean by "inner ring" though.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BZHSpace on 03/22/2021 04:43 pm
It is correct that Superheavy/Starship will be fuel by the bottom of the booster ? So maybe to stack Starship and Superheavy they will need an inner ring to join the CH4/LOX pipe that could be use also for Starship/Starship orbital refueling operation ?
Using the same plumbing for fuelling the stack and orbital refuelling is exactly their plan, and the reason why they are fuelling from the bottom.

I'm not sure what you mean by "inner ring" though.

Inner ring means : ring welded inside the skirt as current landing legs supports.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: libra on 03/23/2021 11:00 am
Quote
Hah, bit of egg on my face then.

Egg on mine, too  ;D
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Rossco on 03/23/2021 03:03 pm
Rather than butt2butt refilling why not side to side?

My thinking is you could either have two tankers fuelling a single SS at the same time or maybe leaving one/two attached for the entire trip to mars to act as a fuel tank for the main SS - doing this could require less engines on the tankers, thus saving weight?
Once in orbit at Mars they disconnect ready to be collected by a returning SS?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AC in NC on 03/23/2021 03:14 pm
Rather than butt2butt refilling why not side to side?

My thinking is you could either have two tankers fuelling a single SS at the same time or maybe leaving one/two attached for the entire trip to mars to act as a fuel tank for the main SS - doing this could require less engines on the tankers, thus saving weight?
Once in orbit at Mars they disconnect ready to be collected by a returning SS?

- Because there are no GSE attachments on the side
- Because there's no need to have two tankers refueling at the same time
- Because leaving them attached requires a redesign to the thrust structure
- No, less engines wouldn't be the result
- Collected how?

No.  Just no.  There's no problem being solved here and rockets aren't LegosTM.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: joek on 03/23/2021 03:32 pm
Rather than butt2butt refilling why not side to side?

My thinking is you could either have two tankers fuelling a single SS at the same time or maybe leaving one/two attached for the entire trip to mars to act as a fuel tank for the main SS - doing this could require less engines on the tankers, thus saving weight?
Once in orbit at Mars they disconnect ready to be collected by a returning SS?

- Because there are no GSE attachments on the side
- Because there's no need to have two tankers refueling at the same time
- Because leaving them attached requires a redesign to the thrust structure
- No, less engines wouldn't be the result
- Collected how?

No.  Just no.  There's no problem being solved here and rockets aren't LegosTM.

To expand on that... "Once in orbit at Mars..." Not going to happen until Mars orbit aerocapture is proven; a significant challenge.  For the foreseeable future, it's going to be direct Mars entry per current SpaceX (as best we can tell) conops.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/28/2021 08:55 pm
Trying to shed a little light on the subject of Starship refueling ladders, I made up a spreadsheet to calculate optimal elliptical refueling orbits. I call it my Big Starship Refueling Spreadsheet of DoomTM.

The idea here is to start with N ships (1 "mission" Starship and N-1 tankers) in LEO. The whole fleet raises into an elliptical orbit, and using on-orbit refueling one tanker empties into the remaining fleet, filling all the rest of the ships completely (the math trick here is that these two constraints, put together, allow you to solve for the optimum elliptical orbit apogee). Repeat with N-1 tankers until only the "mission" Starship remains.

This avoids the need to develop new hardware (nuclear rockets, stretched tankers, stripped tankers, etc), which is a huge advantage. It also prevents the problem where a depot gets stuck in the "wrong orbit."

The only disadvantage for manned missions is that it requires multiple passes through the radiation belt. This can be minimized with trajectory planning (https://science.thewire.in/the-sciences/apollo-11-van-allen-radiation-belts-translunar-injection/), or simply by discharging the radiation belts temporarily (https://spectrum.ieee.org/aerospace/astrophysics/hacking-the-van-allen-belts).

Using the spreadsheet we can check the math on Casey Handmer's excellent blog post on lunar Starship (https://caseyhandmer.wordpress.com/2021/03/26/lunar-starship-and-unnecessary-operational-complexity/). We can calculate that it's possible to perform the same two-way cargo transport of 192 tonnes using only 29 flights (down from 31), raising the payload per launch from 6.2 tonnes per launch to 6.6 tonnes per launch. That's a huge win.

If anyone is interested in putting forth a serious proposal for an orbital depot, this is the "business as usual" plan that they've gotta beat. :D

Corrections, suggestions, and feedback welcome.

Edit:

Instructions
On the spreadsheet red highlighted cells indicate invalid values. such as
  • insufficient delta-v in LEO to complete the mission
  • an elliptical orbit gets too high and escapes into solar orbit
  • insufficient aggregate tank volume in LEO for that number of refueling flights

Typical usage is to set your mission- and vehicle-related numbers, then look at the first block ("LEO Direct") and find the first row without any red cells. Proceed left-to-right across blocks until finding a valid row.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: superg05 on 05/26/2021 10:00 pm
can a StarShip Superheavy booters IN RAPTOR HEAVY STYLE CONFIGURATION get a fully fuelled starship to orbit? if so it can skip all the refueling of multiple tankers



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AC in NC on 05/26/2021 10:02 pm
can a StarShip Superheavy booters IN RAPTOR HEAVY STYLE CONFIGURATION get a fully fuelled starship to orbit? if so it can skip all the refueling of multiple tankers

No.  And if it could you'd be throwing away one or three SH Boosters.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thrustpuzzle on 05/26/2021 11:35 pm
The idea here is to start with N ships (1 "mission" Starship and N-1 tankers) in LEO. The whole fleet raises into an elliptical orbit, and using on-orbit refueling one tanker empties into the remaining fleet, filling all the rest of the ships completely (the math trick here is that these two constraints, put together, allow you to solve for the optimum elliptical orbit apogee). Repeat with N-1 tankers until only the "mission" Starship remains.
These tools are always fun to play with! Thanks for the model.

Wouldn't the strategy be optimized by first consolidating all fresh tanker launches (each holding 100t)  in LEO to form fully-fueled tankers holding 1200t, then laddering those up with the (similarly fully fueled) mission starship  into higher orbits? Every time a non-full tanker raises its orbit to PO1, its wasting propellant lifting the tanker's dry mass. You want to amortize that dry mass by having as large a fuel loading as possible.

It's most likely I'm misunderstanding the spreadsheet and this is already the model, but it looks like the 100t tankers are doing their first fuel transfers at higher orbit PO1 without an initial LEO consolidation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 06/04/2021 10:37 pm
The idea here is to start with N ships (1 "mission" Starship and N-1 tankers) in LEO. The whole fleet raises into an elliptical orbit, and using on-orbit refueling one tanker empties into the remaining fleet, filling all the rest of the ships completely (the math trick here is that these two constraints, put together, allow you to solve for the optimum elliptical orbit apogee). Repeat with N-1 tankers until only the "mission" Starship remains.
These tools are always fun to play with! Thanks for the model.

Wouldn't the strategy be optimized by first consolidating all fresh tanker launches (each holding 100t)  in LEO to form fully-fueled tankers holding 1200t, then laddering those up with the (similarly fully fueled) mission starship  into higher orbits?

Yes! That's exactly what my model assumes. :) Sorry for not making that clear.


This spreadsheet lets you run the numbers for partially-fueled tankers too (say, two fully-fueled and one half-fueled Starship), depending on exactly how much delta-v your mission requires. But you're exactly right, the mathematical "optimum" (most delta-v for the least launch mass) is to fully fuel all vehicles in LEO, aka PO0.

Problem is, how do you deal with the "leftover fuel" on that last refueling flight? It's rather unlikely that your propellant mass will work out exactly an integer number of 100 t refuelings.

Typically you'll need to tweak the payload mass to achieve exactly that optimum you're describing. Scroll down until you find the blue cell in the "Prop" column. Blue in that column indicates that the last refueling tanker launched into LEO has extra (wasted) fuel left over. Then use the Goal Seek... tool to set the blue cell to the value of the total tank capacity by changing the outbound payload mass (cell B7). Eg if you're using one Tanker and one Mission Starship, you'd Goal Seek to set the blue cell to 1200+1200 = 2400 by changing B7. If you're using three Tankers and one Mission Starship, you'd Goal Seek to set the blue cell to 4800.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/07/2021 11:04 am
Yeah, manufacturing commonality is the word of God here. You know how SpaceX is turning what is essentially partially completed Starships into ground service equipment over there? It's because they're all tooled up to produce those 9 meter tanks and now that they're good at it, making 9 meter tanks for whatever application is a breeze.
To crack open the worm can for a moment: SpaceX also have a production line for 12m ring segments and domes, capable of holding at least a minimum pressure of 3 Bar (70ft high water tank at atmospheric pressure).

None of SpaceX's proposed CONOPS would need a wider diameter tanker, but if they or someone else can think up a mission that does, they do have tooling to repurpose.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: davis_404 on 06/07/2021 06:17 pm
Seems to me, if you can get a Super Heavy into orbit - I'm just assuming a SH can make it with a cone on top - that would make a great "accumulation tanker" for quickfilling a Starship. Multiple Starships.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Alberto-Girardi on 06/07/2021 06:28 pm
Seems to me, if you can get a Super Heavy into orbit - I'm just assuming a SH can make it with a cone on top - that would make a great "accumulation tanker" for quickfilling a Starship. Multiple Starships.

If it is possible theoretically it shoud work, but probably it is simpler to use a SS because you don't have to re-engineer solar panels, bateries, different avionics and the docking mechanism.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: su27k on 04/13/2022 02:04 am
Saving this from the SLS thread:

Won't Starship require a dozen refueling launches before its even ready to head to lunar orbit?

Six refuelings is currently being baselined for each of the uncrewed and crewed demo missions. And those refuelings are for the depot ship. The HLS Starship refuels only ONCE: from the depot ship.
And that number drops to just four refuelings once SpaceX gets the stretched tanker version of Starship going.
The 12 - 14 number is what Blue Origin ran with based on a highly conservative NASA estimate. It never was an official SpaceX number.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: warp99 on 04/14/2022 06:21 am
Saving this from the SLS thread:

Won't Starship require a dozen refueling launches before its even ready to head to lunar orbit?

Six refuelings is currently being baselined for each of the uncrewed and crewed demo missions. And those refuelings are for the depot ship. The HLS Starship refuels only ONCE: from the depot ship.
And that number drops to just four refuelings once SpaceX gets the stretched tanker version of Starship going.
The 12 - 14 number is what Blue Origin ran with based on a highly conservative NASA estimate. It never was an official SpaceX number.
Blue Origin were quoting directly from the NASA evaluation document which in turn was generated directly from the SpaceX proposal.  Nothing was calculated by NASA - conservatively or otherwise. 

To be clear it was *up to* 14 flights which is one HLS launch, one depot launch and up to 12 tanker launches.  At 100 tonnes minimum tanker payload and 150 tonnes for the depot this implies 1350 tonnes of propellant for HLS which seems about right. 

Of course the design goal is 150 tonnes of propellant per tanker which is a more manageable 10 total launches which SpaceX were not willing to commit to 100%. 

The extreme stretch goal is 200 tonnes of propellant per tanker which is 6 tanker flights.  There is no reusable option for four tankers which would be 300 tonnes.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 04/14/2022 10:10 am
To be clear it was *up to* 14 flights which is one HLS launch, one depot launch and up to 12 tanker launches.  At 100 tonnes minimum tanker payload and 150 tonnes for the depot this implies 1350 tonnes of propellant for HLS which seems about right. 

No, it was actually 14 tanker flights, plus one depot launch and one HLS lander, for a total of sixteen launches.  (Page 27 of the GAO decision (https://www.gao.gov/assets/b-419783.pdf), footnote 13.)

(But I agree with the general gist of your comment; I'm just nitpicking a detail.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 04/14/2022 11:15 am
The number of tanker flights depends not only on max tank capacity but on the actual amount of propellant needed. And the later depends on the landed payload and the dry mass of the vehicle in general.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Steve D on 04/14/2022 03:20 pm
I haven't read the whole thread so my apologies if this has been addressed already but how do you measure the amount of fuel and LOX that is in your tank in freefall? If you only need a partial fill how do you measure it?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BT52 on 04/14/2022 04:12 pm
Pressure and temperature maybe?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: randomly on 04/14/2022 04:17 pm
You need to settle the fuel in the tanks with thrusters for transfer, so you can use the differential pressure from the bottom to the top of the tank coupled with the acceleration of the ship to calculate the fuel mass in the tank.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sbuffet on 04/14/2022 04:28 pm
I haven't read the whole thread so my apologies if this has been addressed already but how do you measure the amount of fuel and LOX that is in your tank in freefall? If you only need a partial fill how do you measure it?

I don't know what they will use but I know what I would use. And it would be one of those 80 GHz radars used to measure distance hence the volume left. Obviously would work only when the LOX or the LCH4 are pushed in the bottom of the tank on earth or during the refiling phase under micro g accelerations.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 04/14/2022 04:57 pm
I haven't read the whole thread so my apologies if this has been addressed already but how do you measure the amount of fuel and LOX that is in your tank in freefall? If you only need a partial fill how do you measure it?
Addition and subtraction.
You start with a known propellant mass and volume at launch. As you ascend and burn that propellant, you are measuring the mas-flow to the engines, the temperature of the flow (which gets you volumetric flow) and the acceleration of the vehicle and the thrust of the engines, so you have two independent different what's-my-current-mas measurements up until you reach orbit and cut thrust. That gives you current the mass of propellant in orbit.
From there, you then add the mass or volume (depending on measurement method) of propellant transferred to get your final mass, by measuring the flow rate (mass or volumetric, plus temperature to get the other value whichever you measure).
If there is sufficient boiloff that the tanks ned to be vented, then you may need to add some estimation: current pressure and temperature in the tank head (to give you gas density) plus a model of mass flow rate based on behaviour of your pressure release value to give total mass ejected. This mass should be fairly small compared to overall propellant mass anyway.

If a final confirmation is needed, an RCS burn of known thrust for a short duration (only needs to be enough to settle any fluids if you have not already been under microgee settling thrust) and a measurement of vehicle acceleration would give total mass (with propellant temperature giving you volume) via Newton's 2nd law. If you're using microgee settling thrust then it's even easier, a load cell on your thruster is enough to weigh the vehicle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oiorionsbelt on 04/14/2022 06:04 pm
Or cameras in the tanks?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 04/14/2022 06:17 pm
Or cameras in the tanks?
Photogrammetry of a transparent liquid (possibly with droplets, condensate, multiple free-floating blobs, etc) against a highly reflective background is a very nontrivial machine vision challenge.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 04/14/2022 06:49 pm
Or cameras in the tanks?
Photogrammetry of a transparent liquid (possibly with droplets, condensate, multiple free-floating blobs, etc) against a highly reflective background is a very nontrivial machine vision challenge.
Considering that Tesla and SpaceX are at the cutting edge in visual pattern recognition technology. I don't think it would be to difficult for them to solve.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: alugobi on 04/14/2022 07:00 pm
Quote
cutting edge in visual pattern recognition technology
Tesla is still crashing cars with that tech once in a while. 

Wouldn't bet the Starship farm on it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oiorionsbelt on 04/14/2022 09:30 pm
In currently available videos of the inside of the F9 tanks the level is fairly obvious.

https://www.youtube.com/watch?v=gyNKuHh0ZWY
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Steve D on 04/14/2022 10:40 pm
I haven't read the whole thread so my apologies if this has been addressed already but how do you measure the amount of fuel and LOX that is in your tank in freefall? If you only need a partial fill how do you measure it?
Addition and subtraction.
You start with a known propellant mass and volume at launch. As you ascend and burn that propellant, you are measuring the mas-flow to the engines, the temperature of the flow (which gets you volumetric flow) and the acceleration of the vehicle and the thrust of the engines, so you have two independent different what's-my-current-mas measurements up until you reach orbit and cut thrust. That gives you current the mass of propellant in orbit.
From there, you then add the mass or volume (depending on measurement method) of propellant transferred to get your final mass, by measuring the flow rate (mass or volumetric, plus temperature to get the other value whichever you measure).
If there is sufficient boiloff that the tanks ned to be vented, then you may need to add some estimation: current pressure and temperature in the tank head (to give you gas density) plus a model of mass flow rate based on behaviour of your pressure release value to give total mass ejected. This mass should be fairly small compared to overall propellant mass anyway.

If a final confirmation is needed, an RCS burn of known thrust for a short duration (only needs to be enough to settle any fluids if you have not already been under microgee settling thrust) and a measurement of vehicle acceleration would give total mass (with propellant temperature giving you volume) via Newton's 2nd law. If you're using microgee settling thrust then it's even easier, a load cell on your thruster is enough to weigh the vehicle.

I would think the errors would start to add up after awhile especially when you have to account for boil off.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/19/2022 05:13 am
If a final confirmation is needed, an RCS burn of known thrust for a short duration (only needs to be enough to settle any fluids if you have not already been under microgee settling thrust) and a measurement of vehicle acceleration would give total mass (with propellant temperature giving you volume) via Newton's 2nd law. If you're using microgee settling thrust then it's even easier, a load cell on your thruster is enough to weigh the vehicle.

This last part confuses me. Surely the load cell will merely measure the thrust force of the thruster.

Technically there will be some miniscule non-linearity that doesn't cancel out, but it will be vanishingly small: on the order of the ratio between the "unsprung" mass of the thruster divided by the total mass of Starship.

Load cells on the ullage thrusters aren't a bad idea for providing a ground-truth measurements of thrust, but using that sensor alone to measure propellant load seems unworkable.

The Real AnswerTM is RF level sensors plus pressure transducers. Let's not overcomplicate it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 04/19/2022 06:08 am
Load cell plus accelerometer and/or position tracking would do it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: John Santos on 04/19/2022 06:13 am
If a final confirmation is needed, an RCS burn of known thrust for a short duration (only needs to be enough to settle any fluids if you have not already been under microgee settling thrust) and a measurement of vehicle acceleration would give total mass (with propellant temperature giving you volume) via Newton's 2nd law. If you're using microgee settling thrust then it's even easier, a load cell on your thruster is enough to weigh the vehicle.

This last part confuses me. Surely the load cell will merely measure the thrust force of the thruster.

Technically there will be some miniscule non-linearity that doesn't cancel out, but it will be vanishingly small: on the order of the ratio between the "unsprung" mass of the thruster divided by the total mass of Starship.

Load cells on the ullage thrusters aren't a bad idea for providing a ground-truth measurements of thrust, but using that sensor alone to measure propellant load seems unworkable.

The Real AnswerTM is RF level sensors plus pressure transducers. Let's not overcomplicate it.
You need to measure both the force of the ullage thrusters and the acceleration of the ship and then use F=ma to get the mass.  Easy-peasy Physics 1, if you have an accurate force transducer and accelerometer.

Argh, Robotbeat beat me to it...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/19/2022 06:14 am
Load cell plus accelerometer and/or position tracking would do it.

Agreed. That's the non-bold part I quoted. I understood that bit.

The bold part is supposed to be a way that's "even easier" than that, ie dispensing with the accelerometer and just "weigh[ing] the vehicle." That's the part I can't figure out.



Edit: d'oh, is it simply the case that edzieba momentarily forgot a "scale" needs constant acceleration (eg gravity) and thrusters provide constant force? Or is there some deep physics voodoo going on here and I'm late to the party?  ???
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 04/19/2022 09:42 am
Load cell plus accelerometer and/or position tracking would do it.

Agreed. That's the non-bold part I quoted. I understood that bit.

The bold part is supposed to be a way that's "even easier" than that, ie dispensing with the accelerometer and just "weigh[ing] the vehicle." That's the part I can't figure out.



Edit: d'oh, is it simply the case that edzieba momentarily forgot a "scale" needs constant acceleration (eg gravity) and thrusters provide constant force? Or is there some deep physics voodoo going on here and I'm late to the party?  ???
No, just that a settling thrust means you are already under thrust (so have your thrust value plus acceleration value) and in a nice settled state, whereas if you've been floating in microgravity you need to both apply some impulse and do so for long enough for all the sloshing to settle down (no good just briefly pushing the tank body if the propellant is floating in blob that isn't touching the walls.).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 04/21/2022 06:36 pm
ISTR that the MacGregor tanks are suspended with a strain gauge to allow calculating weight before and after a test. Maybe during a test. This would be the consumption number used in calculating any number of operating parameters.


On orbit no one method of measuring propellant is dead on. There's easily half a dozen ways to work back from the MacGregor consumption calculations so why not carefully choose three or more with known error bars and accept a value that they all agree on?


They'll never get it exact so why not settle for some redundancy and good 'nuff?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: brainbit on 04/21/2022 09:02 pm
Could you not use GPS and strain gauge to measure force of thrusters verses change of speed to calculate mass? Make sence to me. Don't know how accurate you could get though would probably depend on your orbit. Hmmm
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/22/2022 09:25 am
Load cell plus accelerometer and/or position tracking would do it.

Agreed. That's the non-bold part I quoted. I understood that bit.

The bold part is supposed to be a way that's "even easier" than that, ie dispensing with the accelerometer and just "weigh[ing] the vehicle." That's the part I can't figure out.



Edit: d'oh, is it simply the case that edzieba momentarily forgot a "scale" needs constant acceleration (eg gravity) and thrusters provide constant force? Or is there some deep physics voodoo going on here and I'm late to the party?  ???
No, just that a settling thrust means you are already under thrust (so have your thrust value plus acceleration value) and in a nice settled state, whereas if you've been floating in microgravity you need to both apply some impulse and do so for long enough for all the sloshing to settle down (no good just briefly pushing the tank body if the propellant is floating in blob that isn't touching the walls.).

Which part, specifically, are you saying "no" to?

I'm trying to figure out how you can measure propellant mass with just a strain gauge. "Weigh the vehicle," like you said.

If everyone else also agrees that this is impossible, then mystery solved.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 04/22/2022 09:49 am
Which part, specifically, are you saying "no" to?
The idea of 'dispensing with the accelerometer' that you've introduced. F=MA is classical mechanics, so you need both the F and A terms to get the M that you want.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/22/2022 11:08 am
Which part, specifically, are you saying "no" to?
The idea of 'dispensing with the accelerometer' that you've introduced.

[sic]

F=MA is classical mechanics, so you need both the F and A terms to get the M that you want.

Ok, so your...

even easier [vs. "an RCS burn of known thrust and a measurement of vehicle acceleration"], a load cell on your thruster is enough to weigh the vehicle.

...idea is out. A load cell is clearly not enough.

I think the extra confusing part is where you said "if you're using microgee settling thrusters" then suddenly something changes and you're now considering a load cell. Surely if we can assume that the RCS thrusters have a "known thrust," we can equally assume that for the settling thrusters, too.

So glad to know I'm not going crazy! I couldn't see how that idea could possibly work.  ???  Turns out edzieba has not invented new physics.

Mystery solved. Carry on.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 04/22/2022 12:17 pm
It helps to read posts in their entirety, rather than picking just a sentence fragment and ignoring everything else around it.
I think the extra confusing part is where you said "if you're using microgee settling thrusters" then suddenly something changes and you're considering a load cell.
Everyone else picked up that the difference when under settling thrust is that you do not need to perform a burn of sufficient time to allow settling of fluid to a steady state, as you are already doing so. I explained in this post here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2361198#msg2361198) if you're still having trouble.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/22/2022 12:36 pm
It helps to read posts in their entirety, rather than picking just a sentence fragment and ignoring everything else around it.

::) I wasn't.

I was legitimately trying to understand your entire post, and had to rely on context clues and inference. It helps if you write posts that aren't confusing. :P

I think the extra confusing part is where you said "if you're using microgee settling thrusters" then suddenly something changes and you're considering a load cell.
Everyone else picked up on that

You should be thanking me. If one kid in the class is brave enough to raise their hand, it means that 10 others sit in confused silence.

the difference is that when under settling thrust you do not need to perform a burn of sufficient time to allow settling of fluid to a steady state, as you are already doing so.

But how did that effect the decision to add a load cell or not? Those two technical decisions seemed completely unrelated. That's the part that was throwing me off.

If adding a load cell is a good idea, it's a good idea either way. If adding a load cell is a bad idea, it's a bad idea either way.

You seemed to be saying the two decisions were contingent on each-other ("if"). Hence my bewilderment.

Water under the bridge. Turns out it was just a confusing paragraph.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Steve D on 04/25/2022 06:25 pm
Pressure sensors on top and bottom of the tanks. Accelerate the vehicle measuring the acceleration then getting the pressure readings top and bottom of tank. You should be able to calculate the pressure differential and using the acceleration reading calculate the amount in the tank. More liquid mass the higher the pressure differential.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Jimmy Murdok on 05/10/2022 10:11 pm
Been thinking (again) about the best option to settle the liquids and pump them.

I realised that it should be by connecting both spacecrafts through the nosecone and rotate them so both liquids stay at the bottom of each tank.

Why?
1- Because once once you are rotating you will keep fuel in place without requiring extra acceleration.
2- Because then the orbital refueling is a stepping stone for human artificial gravity. The ship measures 50m, 24 of which are habitat.  https://www.artificial-gravity.com/sw/SpinCalc/ (https://www.artificial-gravity.com/sw/SpinCalc/) show us that 3.34 rpm give us 0.3 g´s at the basement floor. Which according to Scott Manley is at the limit where non trained people will start getting dizzy without any adaptation.
 https://youtu.be/nxeMoaxUpWk (https://youtu.be/nxeMoaxUpWk)

On top of that, the nosecone is in a place where you probably want to put a hatch no matter what.

Between the simplicity of the implementation and all the advantages and coolness of having the first real badass artificial gravity system online I say that this is the best option for refueling.

Edti:typo
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: whitelancer64 on 05/10/2022 11:11 pm
Been thinking (again) about the best option to settle the liquids and pump them.

I realised that it should be by connecting both spacecrafts through the nosecone and rotate them so both liquids stay at the bottom of each tank.

Why?
1- Because once once you are rotating you will keep fuel in place without requiring extra acceleration.
2- Because then the orbital refueling is a stepping stone for human artificial gravity. The ship measures 50m, 24 of which are habitat.  https://www.artificial-gravity.com/sw/SpinCalc/ (https://www.artificial-gravity.com/sw/SpinCalc/) show us that 3.34 rpm give us 0.3 g´s at the basement floor. Which according to Scott Manley is at the limit where non trained people will start getting dizzy without any adaptation.
*snip youtube link*

On top of that, the nosecone is in a place where you probably want to put a hatch no matter what.

Between the simplicity of the implementation and all the advantages and coolness of having the first real badass artificial gravity system online I say that this is the best option for refueling.

Edti:typo

The original concept for refueling was mating the two Starships tail-to-tail. That concept has been discarded, the current plan is to mate side to side, with both noses and tails in the same direction. This has the advantage of being able to use the same connections for loading fuel on the launch pad, to load fuel into the mission Starship.

Nose to nose fuel transfer would require additional plumbing and connections on the nose to move the fuel.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 05/11/2022 12:56 am
Been thinking (again) about the best option to settle the liquids and pump them.

I realised that it should be by connecting both spacecrafts through the nosecone and rotate them so both liquids stay at the bottom of each tank.

Why?
1- Because once once you are rotating you will keep fuel in place without requiring extra acceleration.
2- Because then the orbital refueling is a stepping stone for human artificial gravity. The ship measures 50m, 24 of which are habitat.  https://www.artificial-gravity.com/sw/SpinCalc/ (https://www.artificial-gravity.com/sw/SpinCalc/) show us that 3.34 rpm give us 0.3 g´s at the basement floor. Which according to Scott Manley is at the limit where non trained people will start getting dizzy without any adaptation.
*snip youtube link*

On top of that, the nosecone is in a place where you probably want to put a hatch no matter what.

Between the simplicity of the implementation and all the advantages and coolness of having the first real badass artificial gravity system online I say that this is the best option for refueling.

Edti:typo

The original concept for refueling was mating the two Starships tail-to-tail. That concept has been discarded, the current plan is to mate side to side, with both noses and tails in the same direction. This has the advantage of being able to use the same connections for loading fuel on the launch pad, to load fuel into the mission Starship.

Nose to nose fuel transfer would require additional plumbing and connections on the nose to move the fuel.

Wouldn't at least some of those plumbing connections already be there with the nose mounted header tanks?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/11/2022 08:01 am
Been thinking (again) about the best option to settle the liquids and pump them.

I realised that it should be by connecting both spacecrafts through the nosecone and rotate them so both liquids stay at the bottom of each tank.

Why?
1- Because once once you are rotating you will keep fuel in place without requiring extra acceleration.
2- Because then the orbital refueling is a stepping stone for human artificial gravity. The ship measures 50m, 24 of which are habitat.  https://www.artificial-gravity.com/sw/SpinCalc/ (https://www.artificial-gravity.com/sw/SpinCalc/) show us that 3.34 rpm give us 0.3 g´s at the basement floor. Which according to Scott Manley is at the limit where non trained people will start getting dizzy without any adaptation.
 https://youtu.be/nxeMoaxUpWk (https://youtu.be/nxeMoaxUpWk)

On top of that, the nosecone is in a place where you probably want to put a hatch no matter what.

Between the simplicity of the implementation and all the advantages and coolness of having the first real badass artificial gravity system online I say that this is the best option for refueling.

Edti:typo
That means you need to pump the entire propellant volume 'uphill' and then waste that energy dropping the propellant back downhill on the other side of the spin axis. You can use less energy with a low (microgee) settling thrust - as already used for long coast upper stages - and a low pressure differential between the sending and receiving tanks to move propellants. You can eliminate the transfer pumps entirely.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 05/11/2022 12:47 pm
Arguably spinning for transfer could be very very slow, producing like 1 milligee or so. And then use pressure difference like with any other type of docking.

But I agree that putting extra plumbing plus working around nose heatshield is an unnecessary complication. Just dock side by side or bottom to bottom and use whatever pressure difference is needed to push the liquids around.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 05/11/2022 03:54 pm
They should just use thrusters for settling. They could even just use that thrust to raise the orbit of the depot slowly, so it’s not wasted propellant.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/11/2022 04:00 pm
They should just use thrusters for settling. They could even just use that thrust to raise the orbit of the depot slowly, so it’s not wasted propellant.
Use of thrusters during side-by-side fuel transfer is "interesting". I think the ships need to be rigidly connected if thrusters are to be used, to avoid tourqueing the transfer lines.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/11/2022 04:26 pm
They should just use thrusters for settling. They could even just use that thrust to raise the orbit of the depot slowly, so it’s not wasted propellant.
Use of thrusters during side-by-side fuel transfer is "interesting". I think the ships need to be rigidly connected if thrusters are to be used, to avoid tourqueing the transfer lines.
For a given definition of rigid. A 1 microgee settling thrust applied to two Starships (with 100mT dry mass and 150mT payload each, or 500mT total) is ~ 5N.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 05/11/2022 07:27 pm
For a given definition of rigid. A 1 microgee settling thrust applied to two Starships (with 100mT dry mass and 150mT payload each, or 500mT total) is ~ 5N.
How much thrust is needed?  Do we have any idea how bulk liquids behave under 1 microgee?

Under a microgee it takes 25 minutes to fall 10m.  My gut assumption would be that settling would take a long time and you'd have to handle the liquids very gently to avoid stirring them up.  Things like surface tension, various types of viscosity, thermal conductivity and even compressibility would be non-negligible.  You might even get weird quantum effects.

This is not quite the same problem as starting a rocket stage.  To start a rocket you only need to get a small amount of liquid to fill the engine inlet, once the main rocket starts you have plenty of g to make sure the bulk fluid behaves as expected.

This would be worth testing, but there should be a plan B because some tests fail.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/11/2022 07:43 pm
For a given definition of rigid. A 1 microgee settling thrust applied to two Starships (with 100mT dry mass and 150mT payload each, or 500mT total) is ~ 5N.
How much thrust is needed?  Do we have any idea how bulk liquids behave under 1 microgee?

Under a microgee it takes 25 minutes to fall 10m.  My gut assumption would be that settling would take a long time and you'd have to handle the liquids very gently to avoid stirring them up.  Things like surface tension, various types of viscosity, thermal conductivity and even compressibility would be non-negligible.  You might even get weird quantum effects.

This is not quite the same problem as starting a rocket stage.  To start a rocket you only need to get a small amount of liquid to fill the engine inlet, once the main rocket starts you have plenty of g to make sure the bulk fluid behaves as expected.

This would be worth testing, but there should be a plan B because some tests fail.
For side-by-side transfer presumably the transfer connection is near the stern and can be reasonabley rigid, and differential thrust would apply torque over a long moment arm. However, you would not need much structure to connect the nose ends of the ships to solve this, probably just a flexible tube that can be connected and then made rigid. Automating this to allow an uncrewed tanker to dock to an uncrewed depot is a little bit interesting.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 05/11/2022 08:17 pm
For a given definition of rigid. A 1 microgee settling thrust applied to two Starships (with 100mT dry mass and 150mT payload each, or 500mT total) is ~ 5N.
How much thrust is needed?  Do we have any idea how bulk liquids behave under 1 microgee?

Under a microgee it takes 25 minutes to fall 10m.  My gut assumption would be that settling would take a long time and you'd have to handle the liquids very gently to avoid stirring them up.  Things like surface tension, various types of viscosity, thermal conductivity and even compressibility would be non-negligible.  You might even get weird quantum effects.

This is not quite the same problem as starting a rocket stage.  To start a rocket you only need to get a small amount of liquid to fill the engine inlet, once the main rocket starts you have plenty of g to make sure the bulk fluid behaves as expected.

This would be worth testing, but there should be a plan B because some tests fail.
For side-by-side transfer presumably the transfer connection is near the stern and can be reasonabley rigid, and differential thrust would apply torque over a long moment arm. However, you would not need much structure to connect the nose ends of the ships to solve this, probably just a flexible tube that can be connected and then made rigid. Automating this to allow an uncrewed tanker to dock to an uncrewed depot is a little bit interesting.
 

Whats wrong with flexible hoses and flying in formation? I see the thunderbirds/blue angels do it all the time.
Computers should be real good at this.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 05/11/2022 08:25 pm
For a given definition of rigid. A 1 microgee settling thrust applied to two Starships (with 100mT dry mass and 150mT payload each, or 500mT total) is ~ 5N.
How much thrust is needed?  Do we have any idea how bulk liquids behave under 1 microgee?

Under a microgee it takes 25 minutes to fall 10m.  My gut assumption would be that settling would take a long time and you'd have to handle the liquids very gently to avoid stirring them up.  Things like surface tension, various types of viscosity, thermal conductivity and even compressibility would be non-negligible.  You might even get weird quantum effects.

This is not quite the same problem as starting a rocket stage.  To start a rocket you only need to get a small amount of liquid to fill the engine inlet, once the main rocket starts you have plenty of g to make sure the bulk fluid behaves as expected.

This would be worth testing, but there should be a plan B because some tests fail.
For side-by-side transfer presumably the transfer connection is near the stern and can be reasonabley rigid, and differential thrust would apply torque over a long moment arm. However, you would not need much structure to connect the nose ends of the ships to solve this, probably just a flexible tube that can be connected and then made rigid. Automating this to allow an uncrewed tanker to dock to an uncrewed depot is a little bit interesting.
You could also just use attitude control and just fly the ships in formation (we don't need a rigid coupling for air-to-air or sea-to-sea refueling).  If you are using 5N thrust for settling the attitude control could use mN and the control rules might be interesting.

But that was not my main point.  Tonnes of fluid under microgee is several orders of magnitude outside our experience.  I expect some surprising behaviour, and possibly some failed tests while they learn things.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 05/11/2022 08:39 pm
For a given definition of rigid. A 1 microgee settling thrust applied to two Starships (with 100mT dry mass and 150mT payload each, or 500mT total) is ~ 5N.
How much thrust is needed?  Do we have any idea how bulk liquids behave under 1 microgee?

Under a microgee it takes 25 minutes to fall 10m.  My gut assumption would be that settling would take a long time and you'd have to handle the liquids very gently to avoid stirring them up.  Things like surface tension, various types of viscosity, thermal conductivity and even compressibility would be non-negligible.  You might even get weird quantum effects.

This is not quite the same problem as starting a rocket stage.  To start a rocket you only need to get a small amount of liquid to fill the engine inlet, once the main rocket starts you have plenty of g to make sure the bulk fluid behaves as expected.

This would be worth testing, but there should be a plan B because some tests fail.
For side-by-side transfer presumably the transfer connection is near the stern and can be reasonabley rigid, and differential thrust would apply torque over a long moment arm. However, you would not need much structure to connect the nose ends of the ships to solve this, probably just a flexible tube that can be connected and then made rigid. Automating this to allow an uncrewed tanker to dock to an uncrewed depot is a little bit interesting.
 

Whats wrong with flexible hoses and flying in formation? I see the thunderbirds/blue angels do it all the time.
Computers should be real good at this.
Why not hard dock so you don’t rely on synchronized computer systems fighting perturbations with active thruster firings? I see no advantage to formation flying with a flexible hose instead of a hard dock.

Hard dock would also let you use the depot as a stage…
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 05/11/2022 09:09 pm
Why not hard dock so you don’t rely on synchronized computer systems fighting perturbations with active thruster firings? I see no advantage to formation flying with a flexible hose instead of a hard dock.

Hard dock would also let you use the depot as a stage…
The advantage of not hard docking is that you don't need to hard dock.

I've worked on many projects where there was unwanted flex.  There are usually two approaches.  The old mechanical engineers think that if you add enough steel everything is approximately rigid.  Problem solved.  The control engineers work to model the flex and live with it.  Generally the control engineers win, because 1) it's lighter, cheaper and quicker; and 2) When you make some part rigid it flexes somewhere else and now you have a new problem.

Soft docking does not preclude the use of the depot as a stage.  The very purpose of in orbit refueling is to use multiple ships as "stages".
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ETurner on 05/11/2022 10:09 pm
The required delta-V for moving fluid to the bottom of a tank will be on the order of (tank length)/(acceptable time), maybe something like 100 m / 100 s = 1 m/s. Think of the vehicle moving and the fluid more-or-less staying in place.

The fluid will slosh, of course (this time, think of a fluid blob suddenly “dropping to the bottom of the tank”), but cranking up acceleration more slowly as the fluid settles will tend to force sloshing to a lower amplitude, even before baffles dissipate sloshing into turbulence.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 05/12/2022 12:17 am
Why not hard dock so you don’t rely on synchronized computer systems fighting perturbations with active thruster firings? I see no advantage to formation flying with a flexible hose instead of a hard dock.

Hard dock would also let you use the depot as a stage…
The advantage of not hard docking is that you don't need to hard dock.

I've worked on many projects where there was unwanted flex.  There are usually two approaches.  The old mechanical engineers think that if you add enough steel everything is approximately rigid.  Problem solved.  The control engineers work to model the flex and live with it.  Generally the control engineers win, because 1) it's lighter, cheaper and quicker; and 2) When you make some part rigid it flexes somewhere else and now you have a new problem.

Soft docking does not preclude the use of the depot as a stage.  The very purpose of in orbit refueling is to use multiple ships as "stages".
Hard docking is better than not hard docking but formation flying. The control engineers would hate you for trying to do something needlessly complex.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: IRobot on 05/12/2022 08:23 am
I've worked on a special oil & gas ship where gas/oil was transfered between ship by long flexible hoses, with massive floating buoy holding it above water.
So yes, it works even in complicated weather, although there was a fatality during very high seas, as the hose got tangled in a buoy and send it flying to the deck.
I can't see how it wouldn't work in much calmer earth orbit. The only big downside is hose extra weight, but on the other hand, you can save weight by not having docking mechanism.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 05/12/2022 11:57 am
I've worked on a special oil & gas ship where gas/oil was transfered between ship by long flexible hoses, with massive floating buoy holding it above water.
So yes, it works even in complicated weather, although there was a fatality during very high seas, as the hose got tangled in a buoy and send it flying to the deck.
I can't see how it wouldn't work in much calmer earth orbit. The only big downside is hose extra weight, but on the other hand, you can save weight by not having docking mechanism.

The primary issue is that you're transferring cryogens and the choice of materials which remain flexible in that temperature is somewhat limited.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 05/12/2022 12:01 pm
They should just use thrusters for settling. They could even just use that thrust to raise the orbit of the depot slowly, so it’s not wasted propellant.
Use of thrusters during side-by-side fuel transfer is "interesting". I think the ships need to be rigidly connected if thrusters are to be used, to avoid tourqueing the transfer lines.
For a given definition of rigid. A 1 microgee settling thrust applied to two Starships (with 100mT dry mass and 150mT payload each, or 500mT total) is ~ 5N.

I think you're off by a couple orders of magnitude. One microgee is around g-loads on ISS.

Propellant settling is in the ballpark of milligee. Maybe 0.1 milligee (i.e. 1 mm/s^2) would be OK. That would be 500 to 5000N thrust.

So you'd need a rigid connection, but nothing outwards crazy rigid.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 05/12/2022 12:21 pm
They should just use thrusters for settling. They could even just use that thrust to raise the orbit of the depot slowly, so it’s not wasted propellant.
Use of thrusters during side-by-side fuel transfer is "interesting". I think the ships need to be rigidly connected if thrusters are to be used, to avoid tourqueing the transfer lines.
For a given definition of rigid. A 1 microgee settling thrust applied to two Starships (with 100mT dry mass and 150mT payload each, or 500mT total) is ~ 5N.

I think you're off by a couple orders of magnitude. One microgee is around g-loads on ISS.

Propellant settling is in the ballpark of milligee. Maybe 0.1 milligee (i.e. 1 mm/s^2) would be OK. That would be 500 to 5000N thrust.

So you'd need a rigid connection, but nothing outwards crazy rigid.
I think you initially want to have a higher acceleration if you need to "collect" free falling propellant in a specific spot but once there you can lower the acceleration and maintain the settled state. ULA talks about going down to 10 microg on Centaur:
https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf

EDIT: It is worth mentioning that the above is with regard to long term propellant management for depots and interplanetary cruise. I can certainly understand the concerns about propellant use if the assumption is millig level acceleration for thousands of seconds using cold gas thrusters...

I think that SpaceX will want to go as fast as possible for orbital rendezvous, docking and propellant transfer and I suspect/hope that the propellant mass needed for the transfer itself will be a relatively small contribution. I am however still curious about the details with regard to docking two large objects where only a minority of the mass is rigidly attached to the thrusters...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wannamoonbase on 05/12/2022 02:24 pm
Regarding LEO propellant transfer, I'm curious about the interface plates between the two Starships. 

Many coupling systems have male and female components. 
If there is a male and female coupling then any tanker on orbit will have to have the opposite coupler to what it will fill. So would there be a change to the ground side equipment when launching a tanker versus a non tanker?

Of course this is predicated on the idea that the same piping will be used for transfer as ground loading.

Anyway, I'm curious about this part.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/12/2022 03:00 pm
It would be easier to add a second dedicated coupler to the tanker depot than try and make a single coupler androgynous.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/12/2022 03:43 pm
It would be easier to add a second dedicated coupler to the tanker than try and make a single coupler androgynous.
We do not know yet which Starships can do transfers. For the HLS mission, only the depot needs to couple to tankers and HLS. The depot needs to connect to the other ship's QD port, but tankers to not need to connect to HLS or each other.

To generalize, maybe it's not a just depot. Maybe it should be called a propellant transfer system (PTS).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: whitelancer64 on 05/12/2022 03:58 pm
It would be easier to add a second dedicated coupler to the tanker than try and make a single coupler androgynous.

If it's designed from the outset to be androgynous, then that's never a problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/12/2022 09:05 pm
It would be easier to add a second dedicated coupler to the tanker than try and make a single coupler androgynous.

If it's designed from the outset to be androgynous, then that's never a problem.
The current coupler is not androgynous. It's not L/R symmetric (or you'd be shoving LOX in the LCH4 hole), not U/D symmetric (high pressure port arrangement) and none of the ports themselves are androgynous.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: jimvela on 05/12/2022 09:48 pm
It would be easier to add a second dedicated coupler to the tanker than try and make a single coupler androgynous.

If it's designed from the outset to be androgynous, then that's never a problem.
The current coupler is not androgynous. It's not L/R symmetric (or you'd be shoving LOX in the LCH4 hole), not U/D symmetric (high pressure port arrangement) and none of the ports themselves are androgynous.

What if one or both had the ability to rotate 180 degrees? 
I've designed (electrical) (ground) (test) interfaces that were deliberately able to mate after flipping one side 180degrees.  Lots of good reasons one might choose to do that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/12/2022 10:12 pm
I'm confused about side-to-side docking.  How do you avoid the two ships' elonerons getting tangled up?  I've been assuming that dorsal-to-dorsal makes more sense than side-to-side. 

In re. hard-dock:  Altius had the "sticky boom" concept fairly well developed.  That's considerably less rigid than a hard dock, but it's more rigid than just dragging a Starship around by a couple of hose lines.

What's the current thinking on auxiliary lines?  At the very least, it seems like you need a gas line paired with each cryo liquid line, so that ullage gas that's displaced as a tank fills can be moved over to the donor ship to supplement its ullage gas as the tank empties.

One last thing:  I've been fooling around with the idea of a third-party Option B HLS provider (Dynetics, NorGrumm, maybe LockMart, maybe Sierra) building a bigger descent-ascent module and refueling it using a Starship tanker.  That would put SpaceX in the third-party refueling business.  Seems to me that Space Force would love for this to happen, as would a variety of other providers.  But that would imply that SpaceX would want to design the refueling plates with an eye to eventually issuing it as a standard, or at least as a proprietary interface that could be licensed out to third parties.

I think that might argue pretty heavily for a sticky boom or some other kind of stand-off soft dock for the coupling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: whitelancer64 on 05/12/2022 10:20 pm
I'm confused about side-to-side docking.  How do you avoid the two ships' elonerons getting tangled up?  I've been assuming that dorsal-to-dorsal makes more sense than side-to-side. 

In re. hard-dock:  Altius had the "sticky boom" concept fairly well developed.  That's considerably less rigid than a hard dock, but it's more rigid than just dragging a Starship around by a couple of hose lines.

What's the current thinking on auxiliary lines?  At the very least, it seems like you need a gas line paired with each cryo liquid line, so that ullage gas that's displaced as a tank fills can be moved over to the donor ship to supplement its ullage gas as the tank empties.

One last thing:  I've been fooling around with the idea of a third-party Option B HLS provider (Dynetics, NorGrumm, maybe LockMart, maybe Sierra) building a bigger descent-ascent module and refueling it using a Starship tanker.  That would put SpaceX in the third-party refueling business.  Seems to me that Space Force would love for this to happen, as would a variety of other providers.  But that would imply that SpaceX would want to design the refueling plates with an eye to eventually issuing it as a standard, or at least as a proprietary interface that could be licensed out to third parties.

I think that might argue pretty heavily for a sticky boom or some other kind of stand-off soft dock for the coupling.

Tankers have been shown dorsal-to-dorsal in renders.

"Side-by-Side" is technically incorrect, I only used the term to distinguish from nose-to-nose and tail-to-tail.

The Dynetics lander is methane fueled and requires fueling in the current design (IIRC it's launched empty, or burns fuel for TLI and needs more to land, one of the two), no need for a new, bigger design. It could take advantage of a Starship fuel depot or Tanker for refueling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/12/2022 11:03 pm
Many coupling systems have male and female components. 
If there is a male and female coupling then any tanker on orbit will have to have the opposite coupler to what it will fill. So would there be a change to the ground side equipment when launching a tanker versus a non tanker?

It's worth remembering that you aren't connecting like-to-like. You are connecting the fuel & oxy discharge lines on the tanker to the fuel/oxy supply lines on the depot; the vent lines on the depot to the pressurisation lines on the tanker.

And likewise, since lining up either dorsal-to-dorsal or belly-to-belly means you are connecting left-side to right-side & right-to-left, it should be relatively simple* to allocate male/female connectors in a consistent way that also works for GSE.

*(as these things go.)

This also means consistency and compatibility across vehicles. There isn't a "tanker" layout and a "depot" layout and a "ship" layout. It's the same layout on all vehicles, and any vehicle can connect to any other. Tanker-to-ship, tanker-to-tanker, etc.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/13/2022 03:04 am
Some thoughts. They've all been discussed at one time or another but putting them together suggests some mutual solutions. Propellant connection gender and polarity, commercial propellant services and keeping depots on orbit and time needed to transfer propellant from tanker to depot.

Assuming one launch site, the depot will phase every 12 hours. This is the theoretical target. Approach, hookup, transfer, disconnect, and getting out of the way has to happen in 12 hours to make way for the next tanker. It will be some time before this tempo can be reached but SX works hard to make things look so easy.

If the depot is never intended to return it looses the fins and heat shield. No fins eases proximity concerns. No heat shield means the side opposite GSE is available for use. Put a second set of connectors on what would normally be the ventral side. Let them duplicate the launch mount GSE and they can connect to both the tanker and the SS/LSS GSE with minimum plumbing runs.

For commercial propellant services there is nothing on the horizon that will need the SS GSE flow capacity. At least not for methalox. (I just can't see Bezos filling up at the Musk pump) Forcing a tiny methalox lander to use SS scale connectors makes no sense.

Smaller capacity connectors is where standards will do best. When there's something out there that needs SS scale of flow, will be the time to have a large connector standard. Whatever the SX system has morphed into by then may become the standard. 'Til then, the lack of a heat shield leaves plenty of room for another bank of smaller connections.

No one depot orbit meets all mission needs but one or two depots with relatively modest dV (I'm going out on a limb here) should be able to do LEO service for all lunar missions. Leave it up there until it breaks, then burn it up or keep it up there for parts when it looks like the LEO infrastructure is ready to support that.

Any one orbits suitability for multiple missions has been a point of contention with the propellant cost of repositioning being the main criticism. There are cases where this is true but IMO, the arguments place too much emphasis on mission optimization and not enough emphasis on the system level optimization where SX shines

If things go the way Musk wants, flinging an extra tanker or two so the depot can reposition shouldn't make all that much difference.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/13/2022 03:58 am
It's worth remembering that you aren't connecting like-to-like. You are connecting the fuel & oxy discharge lines on the tanker to the fuel/oxy supply lines on the depot; the vent lines on the depot to the pressurisation lines on the tanker.

Why are there separate fill and drain lines?  Seems like you ought to be able to make do with one.  Cryo liquid drains from one tank and fills the other.  Which one is which is a matter of which tank is being pressurized by pumping gas into it.  Presumably, that gas is ullage that you're removing from the tank being filled.

Quote
And likewise, since lining up either dorsal-to-dorsal or belly-to-belly means you are connecting left-side to right-side & right-to-left, it should be relatively simple* to allocate male/female connectors in a consistent way that also works for GSE.

*(as these things go.)

If you really do have common fill-drain lines and a ambi-directional pump for each gas line, then the easiest geometry for the connector plate is simply all four lines along the (vertical, x) axis.  No handedness at all.

Quote
This also means consistency and compatibility across vehicles. There isn't a "tanker" layout and a "depot" layout and a "ship" layout. It's the same layout on all vehicles, and any vehicle can connect to any other. Tanker-to-ship, tanker-to-tanker, etc.

The real question then is how to dock the connectors together.  I'm kinda thinking that dual soft-capture systems wouldn't be a terrible idea.  They probably each have to extend a couple of meters, but they can hard-connect after the soft-dock.

The biggest thing to figure out with dorsal-to-dorsal docking is how to minimize total torque.  Is one soft-dock enough to take all the torques out of the system, and manage them during the low-gee ullage burns that are needed for transfer?  Or do they need two different docks, one near the top and one near the bottom?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: greybeardengineer on 05/13/2022 05:31 am
The biggest thing to figure out with dorsal-to-dorsal docking is how to minimize total torque.  Is one soft-dock enough to take all the torques out of the system, and manage them during the low-gee ullage burns that are needed for transfer?  Or do they need two different docks, one near the top and one near the bottom?

For crewed SS a dorsal access port/docking fixture on the upper half will eventually be needed for transferring personnel and cargo between two SS and between SS and a space station or other craft. It would likely be immediately above the propulsion section to minimize interference with the payload door on cargo SS (the docking fixture on cargo SS wouldn't have the access tunnel/port, just the mating hardware to help secure the SS with a tanker).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: volker2020 on 05/13/2022 06:00 am
I wonder how much mass a fuel docking adapter requires, if it sits just between the two tanks. If the mass is acceptable, you could just have 1 adapters on each side. This would solve the issue of male/female coupling, you could connect more than 2 ships and even doing fuel and cargo transfers at the same time.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nomadd on 05/13/2022 06:58 am
Assuming one launch site, the depot will phase every 12 hours. This is the theoretical target. Approach, hookup, transfer, disconnect, and getting out of the way has to happen in 12 hours to make way for the next tanker. It will be some time before this tempo can be reached but SX works hard to make things look so easy.
Maybe not. It depends on how long the tanker can wait to offload.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: volker2020 on 05/13/2022 08:56 am
After thinking about it again. I guess I would go for 2 adapters on the top side. One on top of the tanks for receiving fuel, one on the bottom of the tank for providing fuel.
This way, you may end up without even needing pumps, just letting gravity inertia doing it's job.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 05/13/2022 10:36 am
[...]
Assuming one launch site, the depot will phase every 12 hours. This is the theoretical target. Approach, hookup, transfer, disconnect, and getting out of the way has to happen in 12 hours to make way for the next tanker. It will be some time before this tempo can be reached but SX works hard to make things look so easy.
[...]
Note that unless you are launching very close to the equator the only orbit that has evenly 12 h spaced ascending and descending nodes are polar orbits...

So far the only thing we have seen from SpaceX are mission specific tankers/depots that would then have to move/reenter and launch again to be used for another mission (or simply disposed of).

If your desire is simply to collect as much propellant as possible on orbit the ideal orbit would be a synchronous orbit that is as low and as close to the launch site inclination as possible (the reason you want synchronous orbits is because there is little room or time for phasing orbits).

As can be seen in the attached plot the primary options for Texas/Florida launches are a ~300 km orbit that repeats every 2 days or a ~475 km orbit that repeats every day. For each you can choose the exact inclination/altitude combination so that the depot passes the launch site on the descending pass after 2, 3, 4 and so on orbits.

For example, if I have done my math right a depot in a 34.3 degree 319 km circular orbit has a 31:2 repeat cycle and would allow a tanker to launch on an ascending pass from KSC and land on the 4th orbit descending pass just under 5 h later. With reasonably quick turn around it can do 3 such missions per day so a single tanker can service 6 depots in equally spaced orbits every 2 days.

Whether such a scheme is worth it would of course depend on how close these orbits are to the desired destination :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/13/2022 10:56 am
Many coupling systems have male and female components. 
If there is a male and female coupling then any tanker on orbit will have to have the opposite coupler to what it will fill. So would there be a change to the ground side equipment when launching a tanker versus a non tanker?

It's worth remembering that you aren't connecting like-to-like. You are connecting the fuel & oxy discharge lines on the tanker to the fuel/oxy supply lines on the depot; the vent lines on the depot to the pressurisation lines on the tanker.

And likewise, since lining up either dorsal-to-dorsal or belly-to-belly means you are connecting left-side to right-side & right-to-left, it should be relatively simple* to allocate male/female connectors in a consistent way that also works for GSE.

*(as these things go.)

This also means consistency and compatibility across vehicles. There isn't a "tanker" layout and a "depot" layout and a "ship" layout. It's the same layout on all vehicles, and any vehicle can connect to any other. Tanker-to-ship, tanker-to-tanker, etc.
You can look at photos of the connectors to see why this does not work: There is one connector for each fill/drain connection (not separate) and they are side by side. Connecting these directly facing each other results in plugging LOX in LCH4 and vice versa (and none of the other connectors line up, due to the lack of mirror symmetry). And the QD plate is not rotationally symmetric, so flipping 180 means your other support connectors (power, data, gas lines, etc) are flapping in the complete lack of breeze. And the plumbing connectors themselves are not androgynous, and I can't recall ever seeing an androgynous QD fluid connector (let alone a cryogenic one, let alone a cryogenic vacuum-compatible one) that are not composite connectors (two paired gendered connectors with 180° symmetry).

SpaceX started with an androgynous plumbing connector layout and switched to a dedicated QD plate that precludes that. As TheRadicalModerate mentioned, a vertical layout would allow the L/R symmetry for face to face connection, but that design was clearly not chosen. It was also not chosen when the QD plate was refreshed for Ship 25, so SpaceX clearly do place any priority on universal connector mating. "We have a matching connector, just mount that on a ship" seems to be the current thinking.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsnellenberger on 05/13/2022 01:51 pm
With regard to docking a tanker to Starship, I'd think that those four stabilization sockets used by the chopsticks might be useful if they can come up with some docking arms.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 05/13/2022 03:23 pm
With regard to docking a tanker to Starship, I'd think that those four stabilization sockets used by the chopsticks might be useful if they can come up with some docking arms.
Yeah, very true. I suspect they’ll using the fueling setup on Stage Zero to at least inform the in-orbit refueling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 05/13/2022 03:24 pm
Does anyone know any projected timeline for first attempted orbital refueling between two Starships?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/13/2022 03:38 pm
Does anyone know any projected timeline for first attempted orbital refueling between two Starships?
Clearly before the HLS uncrewed demo, so we can make some inferences. My guess: they will build and launch depot first and then build and launch tanker, so the first propellant transfer will be from tanker to depot.  Working backwards from Artemis III on 2025, I think the propellant transfer demo needed to have happened in Q1 2022  :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/13/2022 06:02 pm
SpaceX started with an androgynous plumbing connector layout and switched to a dedicated QD plate that precludes that. As TheRadicalModerate mentioned, a vertical layout would allow the L/R symmetry for face to face connection, but that design was clearly not chosen. It was also not chosen when the QD plate was refreshed for Ship 25, so SpaceX clearly do place any priority on universal connector mating. "We have a matching connector, just mount that on a ship" seems to be the current thinking.

Or maybe they've just decided that the GSE and on-orbit systems are fundamentally different, and they plan to add the on-orbit one later on.  This seems more likely to me, because the soft-dock and torque-control requirements swamp the details of the plumbing.  My guess is that they'll just graft in another access point to the fill/drain system and have done with it.  The plumbing is easy.  Reducing RPOD risk, minimizing moments of inertia, and being able to securely fly with tandem ullage burns are all hard.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/13/2022 06:08 pm
Does anyone know any projected timeline for first attempted orbital refueling between two Starships?
Clearly before the HLS uncrewed demo, so we can make some inferences. My guess: they will build and launch depot first and then build and launch tanker, so the first propellant transfer will be from tanker to depot.  Working backwards from Artemis III on 2025, I think the propellant transfer demo needed to have happened in Q1 2022  :)

Four things you can do with every single Starlink mission:

1) Launch to orbit.
2) Deploy Starlinks.
3) Attempt a refueling RPOD.
4) Attempt an EDL.

My guess is that, once the refueling connectors are finalized, they'll try to do all four, in this order, every single mission.  In many respects, getting refueling to work is more important than getting EDL recovery to work.  You can implement Option A without reusability.  But you can't do it without refueling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/13/2022 06:19 pm
Assuming one launch site, the depot will phase every 12 hours. This is the theoretical target. Approach, hookup, transfer, disconnect, and getting out of the way has to happen in 12 hours to make way for the next tanker. It will be some time before this tempo can be reached but SX works hard to make things look so easy.
Maybe not. It depends on how long the tanker can wait to offload.
Balanced by how long the depot can hang around waiting to unload. The depot is, IMO, the logical place for solar shielding and/or active cooling. This implies unloading the tankers as fast as possible leaving the more capable depot to wait around for a customer.


Working this fast for an Artemus launch would be overkill. Getting a fleet ready for mars is where it would pay off. 





Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/13/2022 07:03 pm
Does anyone know any projected timeline for first attempted orbital refueling between two Starships?
Clearly before the HLS uncrewed demo, so we can make some inferences. My guess: they will build and launch depot first and then build and launch tanker, so the first propellant transfer will be from tanker to depot.  Working backwards from Artemis III on 2025, I think the propellant transfer demo needed to have happened in Q1 2022  :)

Four things you can do with every single Starlink mission:

1) Launch to orbit.
2) Deploy Starlinks.
3) Attempt a refueling RPOD.
4) Attempt an EDL.

My guess is that, once the refueling connectors are finalized, they'll try to do all four, in this order, every single mission.  In many respects, getting refueling to work is more important than getting EDL recovery to work.  You can implement Option A without reusability.  But you can't do it without refueling.
Are you assuming that a tanker can refuel a starlink-dispensing cargo Starship directly? The HLS mission requires a tanker to transfer propellant to a depot and requires the depot to transfer propellant to the HLS. This means that the actual transfer hardware might only be present on the depot and not on the other types. If a tanker can transfer propellant to a Starlink-dispensing Starship, then the transfer hardware must either survive EDL or must remain in space or be expended.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/13/2022 07:20 pm
Many coupling systems have male and female components. 
If there is a male and female coupling then any tanker on orbit will have to have the opposite coupler to what it will fill. So would there be a change to the ground side equipment when launching a tanker versus a non tanker?

It's worth remembering that you aren't connecting like-to-like. You are connecting the fuel & oxy discharge lines on the tanker to the fuel/oxy supply lines on the depot; the vent lines on the depot to the pressurisation lines on the tanker.

And likewise, since lining up either dorsal-to-dorsal or belly-to-belly means you are connecting left-side to right-side & right-to-left, it should be relatively simple* to allocate male/female connectors in a consistent way that also works for GSE.

*(as these things go.)

This also means consistency and compatibility across vehicles. There isn't a "tanker" layout and a "depot" layout and a "ship" layout. It's the same layout on all vehicles, and any vehicle can connect to any other. Tanker-to-ship, tanker-to-tanker, etc.
You can look at photos of the connectors to see why this does not work: There is one connector for each fill/drain connection (not separate) and they are side by side. Connecting these directly facing each other results in plugging LOX in LCH4 and vice versa (and none of the other connectors line up, due to the lack of mirror symmetry). And the QD plate is not rotationally symmetric, so flipping 180 means your other support connectors (power, data, gas lines, etc) are flapping in the complete lack of breeze. And the plumbing connectors themselves are not androgynous, and I can't recall ever seeing an androgynous QD fluid connector (let alone a cryogenic one, let alone a cryogenic vacuum-compatible one) that are not composite connectors (two paired gendered connectors with 180° symmetry).

SpaceX started with an androgynous plumbing connector layout and switched to a dedicated QD plate that precludes that. As TheRadicalModerate mentioned, a vertical layout would allow the L/R symmetry for face to face connection, but that design was clearly not chosen. It was also not chosen when the QD plate was refreshed for Ship 25, so SpaceX clearly do place any priority on universal connector mating. "We have a matching connector, just mount that on a ship" seems to be the current thinking.
I think it's safe to say that the QD plates are a work in progress. As long as no article is launched more than once they can be changed at will with little impact on ops. No options have been closed off so it could go either way, Rad Mods vertical layout or a second connector bank.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/13/2022 07:32 pm
Four things you can do with every single Starlink mission:

1) Launch to orbit.
2) Deploy Starlinks.
3) Attempt a refueling RPOD.
4) Attempt an EDL.

My guess is that, once the refueling connectors are finalized, they'll try to do all four, in this order, every single mission.  In many respects, getting refueling to work is more important than getting EDL recovery to work.  You can implement Option A without reusability.  But you can't do it without refueling.
Are you assuming that a tanker can refuel a starlink-dispensing cargo Starship directly? The HLS mission requires a tanker to transfer propellant to a depot and requires the depot to transfer propellant to the HLS. This means that the actual transfer hardware might only be present on the depot and not on the other types. If a tanker can transfer propellant to a Starlink-dispensing Starship, then the transfer hardware must either survive EDL or must remain in space or be expended.

I'm assuming that a depot isn't always necessary.  The transfer hardware needs to be able to survive EDL for the lift tanker if it's to be reused.  And I can't imagine the hardware not being androgynous.

Depots are necessary when lift tanker cadence is low, or if the gap between completion of prop accumulation on-orbit and the launch of the payload mission could be long.  Given that the LSS launch commit process is likely to be fairly involved, that gap could be long, and SpaceX will obviously be fairly low on the learning curve wrt tanker cadence.  But as everybody gets the ops situation wrung out, a naked tanker acting as an accumulator has lots of advantages, especially if it's sent to NRHO, since it can do a direct EDL to return.  A depot can't do that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 05/13/2022 07:36 pm

Four things you can do with every single Starlink mission:

1) Launch to orbit.
2) Deploy Starlinks.
3) Attempt a refueling RPOD.
4) Attempt an EDL.

My guess is that, once the refueling connectors are finalized, they'll try to do all four, in this order, every single mission.  In many respects, getting refueling to work is more important than getting EDL recovery to work.  You can implement Option A without reusability.  But you can't do it without refueling.

I agree except I'd expect them to start trying things long before anything is finalized.  The equivalent of early F9 parachute recover and ocean landing tests.

For example you can test aspects of bulk handling of liquids by sloshing the ship or dumping fuel to space.  If something turns out to be harder than anticipated there could be a lot of experiments with different ullage thrust or tank baffles or whatever.  These would not need a rendezvous or refueling connectors.

The early connectors might not work or might not work well.  Version 2.0 will be in the pipeline.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/13/2022 07:38 pm
I think it's safe to say that the QD plates are a work in progress. As long as no article is launched more than once they can be changed at will with little impact on ops. No options have been closed off so it could go either way, Rad Mods vertical layout or a second connector bank.

I'd bet that the overarching QD requirement is that it's at a predictable spot, low down on the Starship, to minimize GSE complexity and allow Starships to be stretched to various lengths without forcing GSE rework.  But the refueling connector has different requirements:  It should be androgynous if at all possible, and it has to minimize RPOD risks.  Those two sets of requirements are sufficiently divergent that they probably need two different solutions.  The best part may be no part, but that's only true if you don't really need the part.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/13/2022 07:43 pm

Four things you can do with every single Starlink mission:

1) Launch to orbit.
2) Deploy Starlinks.
3) Attempt a refueling RPOD.
4) Attempt an EDL.

My guess is that, once the refueling connectors are finalized, they'll try to do all four, in this order, every single mission.  In many respects, getting refueling to work is more important than getting EDL recovery to work.  You can implement Option A without reusability.  But you can't do it without refueling.

I agree except I'd expect them to start trying things long before anything is finalized.  The equivalent of early F9 parachute recover and ocean landing tests.

For example you can test aspects of bulk handling of liquids by sloshing the ship or dumping fuel to space.  If something turns out to be harder than anticipated there could be a lot of experiments with different ullage thrust or tank baffles or whatever.  These would not need a rendezvous or refueling connectors.

The early connectors might not work or might not work well.  Version 2.0 will be in the pipeline.

They can do whatever they want to a Starship after it has deployed its payload, just as they could do whatever they wanted to F9 cores after they staged.  The scope of the experimentation is limited only by what's ready to be incorporated in a fashion that doesn't risk the payload.

The big decision here will always be how to prioritize refueling experiments vs. EDL experiments.  A refueling experiment that has a high chance of destructive failure precludes an EDL attempt.  But I'd guess that most refueling failures will be RPOD wave-offs, which still allow the EDL.  Still, there's some risk assessment involved.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 05/13/2022 08:20 pm
It's worth remembering that you aren't connecting like-to-like. You are connecting the fuel & oxy discharge lines on the tanker to the fuel/oxy supply lines on the depot; the vent lines on the depot to the pressurisation lines on the tanker.

Why are there separate fill and drain lines?  Seems like you ought to be able to make do with one.  Cryo liquid drains from one tank and fills the other.  Which one is which is a matter of which tank is being pressurized by pumping gas into it.  Presumably, that gas is ullage that you're removing from the tank being filled.

Quote
And likewise, since lining up either dorsal-to-dorsal or belly-to-belly means you are connecting left-side to right-side & right-to-left, it should be relatively simple* to allocate male/female connectors in a consistent way that also works for GSE.

*(as these things go.)

If you really do have common fill-drain lines and a ambi-directional pump for each gas line, then the easiest geometry for the connector plate is simply all four lines along the (vertical, x) axis.  No handedness at all.
e top and one near the bottom?

In the same vein as no separate fill & drain lines, why pump gas at all? Pumping gas is less efficient than pumping liquid. But I wouldn't pump either. I'd just vent the ullage on the receiving side. As a side effect by lowering pressure you are lowering the boiling point so you could get superchilled propellant as a side effect (at a cost of boiling off some, of course).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/13/2022 09:06 pm
In the same vein as no separate fill & drain lines, why pump gas at all? Pumping gas is less efficient than pumping liquid. But I wouldn't pump either. I'd just vent the ullage on the receiving side. As a side effect by lowering pressure you are lowering the boiling point so you could get superchilled propellant as a side effect (at a cost of boiling off some, of course).

Somebody's going to be pumping something, or you can't fill the target tank to a level higher than that of the source tank.  Pumping liquid is certainly a possibility, in which case you can vent the target tank.  However, if you do that, then you have to create new ullage gas for the source tank.  Note that the mass of the ullage gas is non-trivial; by venting, you're losing multiple tonnes of prop.

One of the things that worries me about pumping liquid is that your inlet pressures are low with only ullage thrust.  Cavitation isn't a big a problem when you're just transferring stuff, but you don't want to get into a position where slosh is uncovering the inlet.  My prejudice for making the system pressure-fed is that the slosh dynamics are a lot simpler.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 05/13/2022 10:16 pm
There is no "liquid level" that is relevant for propellant transfer, any thrust induced pressure differences are miniscule compared to the likely pressures in the tanks.

Remember that autogenous pressurization means that the tank pressure is self regulating to the vapor pressure except for short time scales:

Increase the temperature in the delivering tank and/or decrease it in the receiving one (through attitude changes end/or electrical heating), settle the propellant and open the valves.

The higher vapor pressure in the warmer tank will push liquid into the colder tank. As the pressure starts to lower in the warmer tank it will be maintained by the liquid evaporating and the pressure increase in the colder tank will be counteracted by gas condensing into the liquid. As long as original temperature difference is large enough to compensate for the lowering temperature in the warm tank (from evaporation and expansion) and increasing temperature in the colder tank (from condensation, compression and the additional warmer propellant) all of the liquid will be transferred. The ullage gas will in effect be transfered in the opposite direction through the liquid.

This situation is the opposite to launch: it is not desirable to develop temperature gradients between the liquid and gas (which can be reduced by spraying, bubbling or other agitation) and the occasional transfer of a gas bubble is unlikely to be an issue.

If a the temperature difference is insufficient by itself it can be supplemented by selective heating/venting of the gas.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/13/2022 10:34 pm
There is no "liquid level" that is relevant for propellant transfer, any thrust induced pressure differences are miniscule compared to the likely pressures in the tanks.

I was just arguing that work of some sort had to be done.  Tanks don't spontaneously transfer liquid beyond the point where they're in hydrostatic equilibrium.

Quote
Remember that autogenous pressurization means that the tank pressure is self regulating to the vapor pressure except for short time scales:

Autogenous pressurization when the engines aren't running is considerably less efficient.  Could you do the transfer work by heating one side and venting the other?  Sure.  But why would you bother?  An electric gas pump is going to be scads more efficient than dumping a bunch of enthalpy of vaporization into one side and wasting ullage gas on the other.

Quote
The higher vapor pressure in the warmer tank will push liquid into the colder tank. As the pressure starts to lower in the warmer tank it will be maintained by the liquid evaporating and the pressure increase in the colder tank will be counteracted by gas condensing into the liquid. As long as original temperature difference is large enough to compensate for the lowering temperature in the warm tank (from evaporation and expansion) and increasing temperature in the colder tank (from condensation, compression and the additional warmer propellant) all of the liquid will be transferred. The ullage gas will in effect be transfered in the opposite direction through the liquid.

This sounds wildly optimistic to me, especially if you're counting on condensation collapse in the target tank to make room.  I don't think that happens without extracting heat from the target tank which is, again, a lot more complicated than a simple gas pump.

The thing that seems to be bothering you is the step of connecting the ullage spaces between the source and target tanks with a pump.  This just isn't rocket science.  It conserves all of the mass, doesn't require heaters or heat pumps, and uses ridiculously high-TRL parts.  Your pump power is dependent solely on the desired transfer rate.  It's easy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BT52 on 05/13/2022 10:43 pm
Where energy for pump comes from? I would wildly guess Fuel itself is nice. Then just use raptors turbopumps. Hehe

Yeah anyway does anybody know how much energy would u need for full transfer? Mine hunch is that solar array with battery pack could still take lots of time.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 05/14/2022 12:54 am
Remember that autogenous pressurization means that the tank pressure is self regulating to the vapor pressure except for short time scales:
Autogenous pressurization when the engines aren't running is considerably less efficient.  Could you do the transfer work by heating one side and venting the other?  Sure.  But why would you bother?  An electric gas pump is going to be scads more efficient than dumping a bunch of enthalpy of vaporization into one side and wasting ullage gas on the other.
In most cases (delivery tankers) you are going to start out with a tank mostly full of hot high pressure ullage and have on the order of 100 kW of solar heating available during much of the coast. The problem will be to to keep the propellants cold enough...

In other cases you have relatively long periods of time to achieve the desired temperature by adjusting your orientation.

BTW, what is the reason you are suggesting pumping the gas and letting the pressure difference push the liquid instead of vice versa? Cavitation is only an issue if you have low tank pressure to start with and then there is not much gas to bother about anyway...
Quote
Quote
The higher vapor pressure in the warmer tank will push liquid into the colder tank. As the pressure starts to lower in the warmer tank it will be maintained by the liquid evaporating and the pressure increase in the colder tank will be counteracted by gas condensing into the liquid. As long as original temperature difference is large enough to compensate for the lowering temperature in the warm tank (from evaporation and expansion) and increasing temperature in the colder tank (from condensation, compression and the additional warmer propellant) all of the liquid will be transferred. The ullage gas will in effect be transfered in the opposite direction through the liquid.
This sounds wildly optimistic to me, especially if you're counting on condensation collapse in the target tank to make room.  I don't think that happens without extracting heat from the target tank which is, again, a lot more complicated than a simple gas pump.

The thing that seems to be bothering you is the step of connecting the ullage spaces between the source and target tanks with a pump.  This just isn't rocket science.  It conserves all of the mass, doesn't require heaters or heat pumps, and uses ridiculously high-TRL parts. Your pump power is dependent solely on the desired transfer rate.  It's easy.
Well, I was suggesting that if this approach is workable the best case scenario is no extra hardware and even a small liquid pump, line and sprinkler completely contained in each tank sounds simpler than a blower, separate gas line and  ship to ship coupling. If you want to you could just take the same blower and gas line and direct it into the liquid instead of to the other ship...

There is no need to look at the details of the gas-liquid interactions for most transfers, the temperature management is about controlling the pressure. I.e. if you can maintain the depot at 1 bar vapor pressure and the tankers at 2 bar you can fill it close to 50% purely pneumatically. You then wait for the liquid and gas to equalize (likely much quicker then the time before the next tanker arrives anyway) so you can fill it to 75% and so on...

The exception is transferring a propellant load into a "hot" Starship, i.e. the propellants are at the max pressure boiling point so that the depot/tanker can not increase the pressure further. In this case you might have to vent up to a few tonnes from the target Starship (likely on the order of mass needed for settling).

There might need to be dedicated pumps but I would think that would be due to details like quicker response time or a lack of independent control of the temperatures in the LOX and LCH4 tanks.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/14/2022 04:45 am
Remember that autogenous pressurization means that the tank pressure is self regulating to the vapor pressure except for short time scales:
Autogenous pressurization when the engines aren't running is considerably less efficient.  Could you do the transfer work by heating one side and venting the other?  Sure.  But why would you bother?  An electric gas pump is going to be scads more efficient than dumping a bunch of enthalpy of vaporization into one side and wasting ullage gas on the other.
In most cases (delivery tankers) you are going to start out with a tank mostly full of hot high pressure ullage and have on the order of 100 kW of solar heating available during much of the coast. The problem will be to to keep the propellants cold enough...

In other cases you have relatively long periods of time to achieve the desired temperature by adjusting your orientation.

BTW, what is the reason you are suggesting pumping the gas and letting the pressure difference push the liquid instead of vice versa? Cavitation is only an issue if you have low tank pressure to start with and then there is not much gas to bother about anyway...
Quote
Quote
The higher vapor pressure in the warmer tank will push liquid into the colder tank. As the pressure starts to lower in the warmer tank it will be maintained by the liquid evaporating and the pressure increase in the colder tank will be counteracted by gas condensing into the liquid. As long as original temperature difference is large enough to compensate for the lowering temperature in the warm tank (from evaporation and expansion) and increasing temperature in the colder tank (from condensation, compression and the additional warmer propellant) all of the liquid will be transferred. The ullage gas will in effect be transfered in the opposite direction through the liquid.
This sounds wildly optimistic to me, especially if you're counting on condensation collapse in the target tank to make room.  I don't think that happens without extracting heat from the target tank which is, again, a lot more complicated than a simple gas pump.

The thing that seems to be bothering you is the step of connecting the ullage spaces between the source and target tanks with a pump.  This just isn't rocket science.  It conserves all of the mass, doesn't require heaters or heat pumps, and uses ridiculously high-TRL parts. Your pump power is dependent solely on the desired transfer rate.  It's easy.
Well, I was suggesting that if this approach is workable the best case scenario is no extra hardware and even a small liquid pump, line and sprinkler completely contained in each tank sounds simpler than a blower, separate gas line and  ship to ship coupling. If you want to you could just take the same blower and gas line and direct it into the liquid instead of to the other ship...

There is no need to look at the details of the gas-liquid interactions for most transfers, the temperature management is about controlling the pressure. I.e. if you can maintain the depot at 1 bar vapor pressure and the tankers at 2 bar you can fill it close to 50% purely pneumatically. You then wait for the liquid and gas to equalize (likely much quicker then the time before the next tanker arrives anyway) so you can fill it to 75% and so on...

The exception is transferring a propellant load into a "hot" Starship, i.e. the propellants are at the max pressure boiling point so that the depot/tanker can not increase the pressure further. In this case you might have to vent up to a few tonnes from the target Starship (likely on the order of mass needed for settling).

There might need to be dedicated pumps but I would think that would be due to details like quicker response time or a lack of independent control of the temperatures in the LOX and LCH4 tanks.

I'm not wedded to pumping gas instead of pumping liquid.  The important part is equalizing the gas pressures.

Let's figure out what we're talking about here.  At O:F=3.5 and a 1200t tanker, we have:

LOX:  933t, density=1142kg/m³, volume=817m³, dynamic viscosity=5.0E-5Pa-s
LCH4: 267t, density=422.6kg/m³, volume=631m³, dynamic viscosity=1.1E-5Pa-s

From the Poiseuille equation for required pressure drop across the two ends of a pipe:

Δp = 8μLQ/(πR⁴)

where:
μ: dynamic viscosity
L: length of the pipe (let's use 50m between the two in/outlets of the tanks)
Q: volumetric flow rate.  For a 2000s fill time, that's 0.41m³/s for LOX, 0.32m³/s for LCH4.
R: radius of the pipe.  10cm?

So:
LOX Δp = 26Pa
LCH4 Δp = 4.5Pa

Multiply both of these by the volumetric rate to get the mechanical power:

Plox = 10.7W
Plch4 = 1.4W

These omit the hydrostatic head, but that's about 150Pa if ullage acceleration is 0.01m/s² for the LOX, and less for the LCH4.  So that's 90W for the LOX pump and less than that for the LCH4 pump.  Maybe 200W total?  You can do the whole job with a 112Wh battery.

This little exercise has convinced me that it is indeed better to pump the liquid.  So all you need to do is equalize gas pressure between the source and target tanks and you're good to go.

BTW:  The hot lift tanker fresh into orbit is not the interesting case, because it only has to pump about 150t of prop.  The interesting case is the depot tanker, which may be pumping up to 1500t of prop into the target LSS or Starship, and it'll be pretty much in thermal equilibrium.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 05/14/2022 06:17 pm
While not covered in the recent EDA interview/tour, I would suspect that the following would be near-optimum.

Tanks on starship are already set up for cold gas RCS using ullage propellants. Both tankers and mission craft would have one gender of connector, the tanker would have the other gender, so every transfer outside special mission craft would use a depot. By connecting a tanker to a depot, or a depot to a mission craft, and firing the thrusters on the receiving craft to accelerate toward the filling craft, it creates both a pressure drop in the receiving tank to suck dense liquid propellants in AND microacceleration to settle propellant for draining into feed lines.

All pumping is done using hardware already required for normal operation. "The best part is no part."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/14/2022 06:25 pm
Four things you can do with every single Starlink mission:

1) Launch to orbit.
2) Deploy Starlinks.
3) Attempt a refueling RPOD.
4) Attempt an EDL.

My guess is that, once the refueling connectors are finalized, they'll try to do all four, in this order, every single mission.  In many respects, getting refueling to work is more important than getting EDL recovery to work.  You can implement Option A without reusability.  But you can't do it without refueling.
Are you assuming that a tanker can refuel a starlink-dispensing cargo Starship directly? The HLS mission requires a tanker to transfer propellant to a depot and requires the depot to transfer propellant to the HLS. This means that the actual transfer hardware might only be present on the depot and not on the other types. If a tanker can transfer propellant to a Starlink-dispensing Starship, then the transfer hardware must either survive EDL or must remain in space or be expended.

I'm assuming that a depot isn't always necessary.  The transfer hardware needs to be able to survive EDL for the lift tanker if it's to be reused.  And I can't imagine the hardware not being androgynous.

Depots are necessary when lift tanker cadence is low, or if the gap between completion of prop accumulation on-orbit and the launch of the payload mission could be long.  Given that the LSS launch commit process is likely to be fairly involved, that gap could be long, and SpaceX will obviously be fairly low on the learning curve wrt tanker cadence.  But as everybody gets the ops situation wrung out, a naked tanker acting as an accumulator has lots of advantages, especially if it's sent to NRHO, since it can do a direct EDL to return.  A depot can't do that.
By naked tanker do you mean no heat shield? It's been discussed and works if it's worth it. If it's worth it is an open question in my mind. This use moves it a bit in that direction.


I've worked with many types of androgynous electrical and data connections but remember seeing only one androgynous fluid coupling. That's the glad hand connector on Westinghouse air brakes. Do they exist for cryo liquids? It'd be great if they exist but if they don't, it doesn't look like a trivial design problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/14/2022 06:46 pm
By naked tanker do you mean no heat shield? It's been discussed and works if it's worth it. If it's worth it is an open question in my mind. This use moves it a bit in that direction.

No, that was a bad description.  I was referring to a non-depot tanker, with no insulation or active cryocooling.  Sorry for the confusion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/14/2022 07:06 pm
Remember that autogenous pressurization means that the tank pressure is self regulating to the vapor pressure except for short time scales:
Autogenous pressurization when the engines aren't running is considerably less efficient.  Could you do the transfer work by heating one side and venting the other?  Sure.  But why would you bother?  An electric gas pump is going to be scads more efficient than dumping a bunch of enthalpy of vaporization into one side and wasting ullage gas on the other.
In most cases (delivery tankers) you are going to start out with a tank mostly full of hot high pressure ullage and have on the order of 100 kW of solar heating available during much of the coast. The problem will be to to keep the propellants cold enough...

In other cases you have relatively long periods of time to achieve the desired temperature by adjusting your orientation.

BTW, what is the reason you are suggesting pumping the gas and letting the pressure difference push the liquid instead of vice versa? Cavitation is only an issue if you have low tank pressure to start with and then there is not much gas to bother about anyway...
Quote
Quote
The higher vapor pressure in the warmer tank will push liquid into the colder tank. As the pressure starts to lower in the warmer tank it will be maintained by the liquid evaporating and the pressure increase in the colder tank will be counteracted by gas condensing into the liquid. As long as original temperature difference is large enough to compensate for the lowering temperature in the warm tank (from evaporation and expansion) and increasing temperature in the colder tank (from condensation, compression and the additional warmer propellant) all of the liquid will be transferred. The ullage gas will in effect be transfered in the opposite direction through the liquid.
This sounds wildly optimistic to me, especially if you're counting on condensation collapse in the target tank to make room.  I don't think that happens without extracting heat from the target tank which is, again, a lot more complicated than a simple gas pump.

The thing that seems to be bothering you is the step of connecting the ullage spaces between the source and target tanks with a pump.  This just isn't rocket science.  It conserves all of the mass, doesn't require heaters or heat pumps, and uses ridiculously high-TRL parts. Your pump power is dependent solely on the desired transfer rate.  It's easy.
Well, I was suggesting that if this approach is workable the best case scenario is no extra hardware and even a small liquid pump, line and sprinkler completely contained in each tank sounds simpler than a blower, separate gas line and  ship to ship coupling. If you want to you could just take the same blower and gas line and direct it into the liquid instead of to the other ship...

There is no need to look at the details of the gas-liquid interactions for most transfers, the temperature management is about controlling the pressure. I.e. if you can maintain the depot at 1 bar vapor pressure and the tankers at 2 bar you can fill it close to 50% purely pneumatically. You then wait for the liquid and gas to equalize (likely much quicker then the time before the next tanker arrives anyway) so you can fill it to 75% and so on...

The exception is transferring a propellant load into a "hot" Starship, i.e. the propellants are at the max pressure boiling point so that the depot/tanker can not increase the pressure further. In this case you might have to vent up to a few tonnes from the target Starship (likely on the order of mass needed for settling).

There might need to be dedicated pumps but I would think that would be due to details like quicker response time or a lack of independent control of the temperatures in the LOX and LCH4 tanks.
There is the seed of a nifty idea here. With the two ships (tanker and depot or depot and SS) along side each other they can orient to allow the supplying ship to shade the receiving ship. The 100kW thermal may not be enough to do the whole job in a reasonable time but it moves things in the right direction.


What we don't have, and I'm incapable of providing, is numbers. One that I can supply is the pressure differential. Six bar max. If one of you kind Sirs who are facile with numbers could noodle a BOE, it would be great. Maybe assume a clear connector aperture equivalent of 250, 500 and 750 sq cm and some reasonable pressure loss across the connector and plumbing. That should be enough to calculate a rough transfer time at six bar. Then we'd have some rough idea of how much assist, of any kind, is needed.


Hmmmm. If the depot needs a sun shield, could it be reoriented and work as a reflector to add heat to the delivering vessel? Probably not a trivial problem. If the transfer can be powered thermally it's more about procedures than gizmos. (except maybe moving the sun shield around)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/14/2022 07:09 pm
Remember that autogenous pressurization means that the tank pressure is self regulating to the vapor pressure except for short time scales:
Autogenous pressurization when the engines aren't running is considerably less efficient.  Could you do the transfer work by heating one side and venting the other?  Sure.  But why would you bother?  An electric gas pump is going to be scads more efficient than dumping a bunch of enthalpy of vaporization into one side and wasting ullage gas on the other.
In most cases (delivery tankers) you are going to start out with a tank mostly full of hot high pressure ullage and have on the order of 100 kW of solar heating available during much of the coast. The problem will be to to keep the propellants cold enough...

In other cases you have relatively long periods of time to achieve the desired temperature by adjusting your orientation.

BTW, what is the reason you are suggesting pumping the gas and letting the pressure difference push the liquid instead of vice versa? Cavitation is only an issue if you have low tank pressure to start with and then there is not much gas to bother about anyway...
Quote
Quote
The higher vapor pressure in the warmer tank will push liquid into the colder tank. As the pressure starts to lower in the warmer tank it will be maintained by the liquid evaporating and the pressure increase in the colder tank will be counteracted by gas condensing into the liquid. As long as original temperature difference is large enough to compensate for the lowering temperature in the warm tank (from evaporation and expansion) and increasing temperature in the colder tank (from condensation, compression and the additional warmer propellant) all of the liquid will be transferred. The ullage gas will in effect be transfered in the opposite direction through the liquid.
This sounds wildly optimistic to me, especially if you're counting on condensation collapse in the target tank to make room.  I don't think that happens without extracting heat from the target tank which is, again, a lot more complicated than a simple gas pump.

The thing that seems to be bothering you is the step of connecting the ullage spaces between the source and target tanks with a pump.  This just isn't rocket science.  It conserves all of the mass, doesn't require heaters or heat pumps, and uses ridiculously high-TRL parts. Your pump power is dependent solely on the desired transfer rate.  It's easy.
Well, I was suggesting that if this approach is workable the best case scenario is no extra hardware and even a small liquid pump, line and sprinkler completely contained in each tank sounds simpler than a blower, separate gas line and  ship to ship coupling. If you want to you could just take the same blower and gas line and direct it into the liquid instead of to the other ship...

There is no need to look at the details of the gas-liquid interactions for most transfers, the temperature management is about controlling the pressure. I.e. if you can maintain the depot at 1 bar vapor pressure and the tankers at 2 bar you can fill it close to 50% purely pneumatically. You then wait for the liquid and gas to equalize (likely much quicker then the time before the next tanker arrives anyway) so you can fill it to 75% and so on...

The exception is transferring a propellant load into a "hot" Starship, i.e. the propellants are at the max pressure boiling point so that the depot/tanker can not increase the pressure further. In this case you might have to vent up to a few tonnes from the target Starship (likely on the order of mass needed for settling).

There might need to be dedicated pumps but I would think that would be due to details like quicker response time or a lack of independent control of the temperatures in the LOX and LCH4 tanks.

I'm not wedded to pumping gas instead of pumping liquid.  The important part is equalizing the gas pressures.

Let's figure out what we're talking about here.  At O:F=3.5 and a 1200t tanker, we have:

LOX:  933t, density=1142kg/m³, volume=817m³, dynamic viscosity=5.0E-5Pa-s
LCH4: 267t, density=422.6kg/m³, volume=631m³, dynamic viscosity=1.1E-5Pa-s

From the Poiseuille equation for required pressure drop across the two ends of a pipe:

Δp = 8μLQ/(πR⁴)

where:
μ: dynamic viscosity
L: length of the pipe (let's use 50m between the two in/outlets of the tanks)
Q: volumetric flow rate.  For a 2000s fill time, that's 0.41m³/s for LOX, 0.32m³/s for LCH4.
R: radius of the pipe.  10cm?

So:
LOX Δp = 26Pa
LCH4 Δp = 4.5Pa

Multiply both of these by the volumetric rate to get the mechanical power:

Plox = 10.7W
Plch4 = 1.4W

These omit the hydrostatic head, but that's about 150Pa if ullage acceleration is 0.01m/s² for the LOX, and less for the LCH4.  So that's 90W for the LOX pump and less than that for the LCH4 pump.  Maybe 200W total?  You can do the whole job with a 112Wh battery.

This little exercise has convinced me that it is indeed better to pump the liquid.  So all you need to do is equalize gas pressure between the source and target tanks and you're good to go.

BTW:  The hot lift tanker fresh into orbit is not the interesting case, because it only has to pump about 150t of prop.  The interesting case is the depot tanker, which may be pumping up to 1500t of prop into the target LSS or Starship, and it'll be pretty much in thermal equilibrium.
Ah kind sir. One step ahead of me. Do your numbers say anything about transfer time?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/14/2022 07:12 pm
While not covered in the recent EDA interview/tour, I would suspect that the following would be near-optimum.

Tanks on starship are already set up for cold gas RCS using ullage propellants. Both tankers and mission craft would have one gender of connector, the tanker would have the other gender, so every transfer outside special mission craft would use a depot. By connecting a tanker to a depot, or a depot to a mission craft, and firing the thrusters on the receiving craft to accelerate toward the filling craft, it creates both a pressure drop in the receiving tank to suck dense liquid propellants in AND microacceleration to settle propellant for draining into feed lines.

All pumping is done using hardware already required for normal operation. "The best part is no part."

IMO, it would be a mistake to intermediate everything through a depot.  Depots are either going to be stuck at the highest orbital energy they attain, or they're going to have to spend a huge amount propellant to propulsively return to LEO.  Lift tankers don't have that problem; they can either aerobrake or do direct EDLs for very little return prop.

Note that a lift tanker makes a perfectly good "accumulation tanker", just like a depot does.  The difference is how long it can loiter without a lot of boil-off.  But if you assume that an NRHO refueling mission is a fairly rapid turnaround affair, it probably makes sense for the sustainable LSS missions to use the following conops:

1) Depot is in LEO to start.
2) Launch one fewer lift tankers than are necessary to fill the depot.  (You need a full 1500t of prop, which is likely 10 LTs.)
3) Depot loiters until SLS/Orion launch commit.
4) Launch the last LT.  It receives all prop from the depot, then boosts for NRHO.  (LEO-NRHO for a 1500t LT takes about 1025t of prop.)
5) Upon arrival in NRHO, immediately transfer all available prop (about 450t) to the LSS.
6) After verification of the transfer, launch SLS/Orion or other cislunar transit mission to NRHO ASAP, to minimize boiloff.
7) LT in NRHO returns directly to EDL (takes about 26t of prop).

The trick is to minimize the time between when the LT takes the prop out of the depot and when the Orion transfers the crew to the LSS.

Just FYI, to get a depot back from NRHO to a propulsive LEO insertion (because there's no heat shield or elonerons) takes about 205t of prop.  If this is to be carried by the depot itself, you'll need to be able to carry about 2200t of prop from LEO, which increases your prop to LEO by almost 50%.

Making an androgynous, bi-directional refueling dock/connector doesn't seem like an insurmountable problem to me.  And it provides a lot more mission flexibility.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/14/2022 07:14 pm
Ah kind sir. One step ahead of me. Do your numbers say anything about transfer time?

I used 2000 seconds (33 mins 20 secs).  That number was retrieved using the time-honored "rectal extraction" technique.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/14/2022 07:20 pm
With the two ships (tanker and depot or depot and SS) along side each other they can orient to allow the supplying ship to shade the receiving ship.

Not in LEO you can't.  I'd have to work out the penumbra from the ship closer to the daylit side of the Earth, but you basically have a 2π steradian plane illuminating the system.  And there's, you know, the Sun on the other side.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 05/14/2022 07:44 pm
While not covered in the recent EDA interview/tour, I would suspect that the following would be near-optimum.

Tanks on starship are already set up for cold gas RCS using ullage propellants. Both tankers and mission craft would have one gender of connector, the tanker would have the other gender, so every transfer outside special mission craft would use a depot. By connecting a tanker to a depot, or a depot to a mission craft, and firing the thrusters on the receiving craft to accelerate toward the filling craft, it creates both a pressure drop in the receiving tank to suck dense liquid propellants in AND microacceleration to settle propellant for draining into feed lines.

All pumping is done using hardware already required for normal operation. "The best part is no part."

IMO, it would be a mistake to intermediate everything through a depot.  Depots are either going to be stuck at the highest orbital energy they attain, or they're going to have to spend a huge amount propellant to propulsively return to LEO.  Lift tankers don't have that problem; they can either aerobrake or do direct EDLs for very little return prop.
(trimmed)
Why is this a problem? Are you assuming depots are going to be rare? Put one in every starlink plane to accumulate propellant on routine flights.

My assumtion is that building more depots is cheaper than designing a new connector that doesnt reduce overall vehical performance.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/14/2022 08:39 pm
While not covered in the recent EDA interview/tour, I would suspect that the following would be near-optimum.

Tanks on starship are already set up for cold gas RCS using ullage propellants. Both tankers and mission craft would have one gender of connector, the tanker would have the other gender, so every transfer outside special mission craft would use a depot. By connecting a tanker to a depot, or a depot to a mission craft, and firing the thrusters on the receiving craft to accelerate toward the filling craft, it creates both a pressure drop in the receiving tank to suck dense liquid propellants in AND microacceleration to settle propellant for draining into feed lines.

All pumping is done using hardware already required for normal operation. "The best part is no part."

IMO, it would be a mistake to intermediate everything through a depot.  Depots are either going to be stuck at the highest orbital energy they attain, or they're going to have to spend a huge amount propellant to propulsively return to LEO.  Lift tankers don't have that problem; they can either aerobrake or do direct EDLs for very little return prop.
(trimmed)
Why is this a problem? Are you assuming depots are going to be rare? Put one in every starlink plane to accumulate propellant on routine flights.

My assumtion is that building more depots is cheaper than designing a new connector that doesnt reduce overall vehical performance.

I'm assuming that depots can't be moved cheaply, while lift tankers can.

This is not an argument against depots.  Depots are necessary when prop needs to be stored for a long time or when (more likely) it takes a long time to accumulate the amount of prop necessary for a mission.  Both of those things are likely to be true in LEO.  But they're silly if neither of those two missions requirements exists.  As a practical matter, I don't think that either of them will exist in NRHO or any other cislunar orbit, or if they do exist, they won't as operational confidence increases.  And lift tankers will always be the correct way to move prop from LEO to cislunar, if for no other reason than it's a one-way trip for depots, while the lift tankers can return to EDL very cheaply.

The other reason why depots are unlikely to be common is that they can't be sent to arbitrary, idiosyncratic orbits.  Consider, for example a deep space expendable version of Starship, which needs to generate massive C3 or haul an extremely large payload (possibly >500t, assembled in LEO) to a more modest C3.  That will require sending a tanker to a high-energy HEEO with very specific orbital parameters, allowing the deep space Starship to gain a lot of Oberth effect by doing a perigee burn with full tanks.

A depot is effectively expendable for this mission.  Reducing its energy enough for re-use, or moving its inclination and argument of perigee, will cost well more than its value.  On the other hand, a lift tanker can boost to the proper HEEO, fill the target Starship, and do an EDL maneuver for almost nothing.  The depot simply doesn't make sense.  But that requires that a target Starship be able to do a refueling RPOD directly with the lift tanker.

Why would a universal connector reduce vehicle performance?  No matter what, the lift tanker and payload versions of the connectors are going to have to be able to withstand EDL.  What secret sauce is there in a universal version that prevents that from happening?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/14/2022 09:50 pm
A glad hand is two parallel faces, each faced with a thin rubber plate. There are matching holes in the middle of each face and rubber to pass compressed air. The latching mechanism is around the outer edge most easily described as something kinda sorta like the lid interface of a pressure cooker.


Face the two connectors together and rotate ~60 deg to lock. The rubber faces scuff against each other while locking. Maybe this can be adapted to cryo, maybe not. I have doubts. Maybe a heater coil under the rubber/Teflon gasket to aid separation while cold. No other style androgynous fluid connector comes to mind.


Image courtesy of [size=78%]https://www.anythingtruck.com/product/810-035042.html (https://www.anythingtruck.com/product/810-035042.html)[/size]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/14/2022 10:16 pm
The core differences between a depot and a tanker, in my mind, is the tanker has all the EDL hardware and the depot has passive and/or active cooling. The depot will need PV and maybe reaction wheels/CMB's to keep tight control on orientation for thermal and, now that I think of it, while moving propellant. Let it be the active member controlling fine attitude while transferring props.


This starts looking like something you think twice about when considering expendability. Cost is only one thing looked at. When production is settled in the depot will be a highly customized variant and will take up a largish hole in the production pipeline. That might or might not be an important issue. Gotta wait and see.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/14/2022 10:27 pm
With the two ships (tanker and depot or depot and SS) along side each other they can orient to allow the supplying ship to shade the receiving ship.

Not in LEO you can't.  I'd have to work out the penumbra from the ship closer to the daylit side of the Earth, but you basically have a 2π steradian plane illuminating the system.  And there's, you know, the Sun on the other side.
I dun forgot about that hot planet. Still, you can point the pointy end towards the sun and use the shield to block the earth heat - I think. It's a problem whether some well targeted heat would be helpful or not helpful.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/15/2022 04:49 am
A glad hand is two parallel faces, each faced with a thin rubber plate. [...] No other style androgynous fluid connector comes to mind.

There's a handful of unisex hose couplings, such as the barcelona coupling. They all use a similar principle, two o-rings clamped together via a symmetrical set of clamps/receivers.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/15/2022 08:55 am
And they all rely on operation at standard temperature and pressure. Once you start demanding compliant materials that operate at cryogenic temperatures (plus tolerate a few hundred kelvin swing in temperature), in a vacuum, you start to find your list of available elastomers is coming up empty.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: chopsticks on 05/15/2022 12:34 pm
And they all rely on operation at standard temperature and pressure. Once you start demanding compliant materials that operate at cryogenic temperatures (plus tolerate a few hundred kelvin swing in temperature), in a vacuum, you start to find your list of available elastomers is coming up empty.
What compliant materials are used in cryogenic environments?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/15/2022 06:22 pm
And they all rely on operation at standard temperature and pressure. Once you start demanding compliant materials that operate at cryogenic temperatures (plus tolerate a few hundred kelvin swing in temperature), in a vacuum, you start to find your list of available elastomers is coming up empty.
What compliant materials are used in cryogenic environments?
Personal experience: small scale, low pressure LN2 distribution from a dewar through a ball valve with Teflon seals. A very different environment than an androgynous coupling. Don't remember the exact pressure but doubt it was over .3-.4 bar.


I wonder what the BC delivery trucks use.


Here's the problem as I see it. An androgynous coupler would have to seal face to face. That throws some sharp limits on the design. Non androgynous QD's have all sorts of opportunities for tapered sealing surfaces that offer possibilities for getting around expansion/contraction issues. I can think of other design features that look like they offer advantages but that's  moving away from personal experience into a theoretical gee wiz space.


OMG. Couplings causing a gee wiz? Elons gotta launch something soon or we'll all go bonkers.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/15/2022 06:35 pm
And they all rely on operation at standard temperature and pressure. Once you start demanding compliant materials that operate at cryogenic temperatures (plus tolerate a few hundred kelvin swing in temperature), in a vacuum, you start to find your list of available elastomers is coming up empty.
What compliant materials are used in cryogenic environments?
Personal experience: small scale, low pressure LN2 distribution from a dewar through a ball valve with Teflon seals. A very different environment than an androgynous coupling. Don't remember the exact pressure but doubt it was over .3-.4 bar.


I wonder what the BC delivery trucks use.


Here's the problem as I see it. An androgynous coupler would have to seal face to face. That throws some sharp limits on the design. Non androgynous QD's have all sorts of opportunities for tapered sealing surfaces that offer possibilities for getting around expansion/contraction issues. I can think of other design features that look like they offer advantages but that's  moving away from personal experience into a theoretical gee wiz space.


OMG. Couplings causing a gee wiz? Elons gotta launch something soon or we'll all go bonkers.
You do not need an androgynous coupler to create an androgynous interface. Divide each flow within the interface into two flows and put a connector of each sex on each side. Done.
  https://commons.wikimedia.org/wiki/File:SAE_Connector.png
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/15/2022 07:19 pm
And they all rely on operation at standard temperature and pressure. Once you start demanding compliant materials that operate at cryogenic temperatures (plus tolerate a few hundred kelvin swing in temperature), in a vacuum, you start to find your list of available elastomers is coming up empty.
What compliant materials are used in cryogenic environments?
Personal experience: small scale, low pressure LN2 distribution from a dewar through a ball valve with Teflon seals. A very different environment than an androgynous coupling. Don't remember the exact pressure but doubt it was over .3-.4 bar.


I wonder what the BC delivery trucks use.


Here's the problem as I see it. An androgynous coupler would have to seal face to face. That throws some sharp limits on the design. Non androgynous QD's have all sorts of opportunities for tapered sealing surfaces that offer possibilities for getting around expansion/contraction issues. I can think of other design features that look like they offer advantages but that's  moving away from personal experience into a theoretical gee wiz space.


OMG. Couplings causing a gee wiz? Elons gotta launch something soon or we'll all go bonkers.
You do not need an androgynous coupler to create an androgynous interface. Divide each flow within the interface into two flows and put a connector of each sex on each side. Done.
  https://commons.wikimedia.org/wiki/File:SAE_Connector.png (https://commons.wikimedia.org/wiki/File:SAE_Connector.png)
Hmmm. Would work but it does double the seal failure points. Maybe acceptable if robust enough.


More properly, a hermaphrodite connector. Has characteristics of both genders, as opposed to no gender.


Thinking it through, it's still the polarity problem at a smaller scale, which a true androgynous solution would fix. It doesn't fix the need for clunky interface adapters or a slightly less clunky second QD plate somewhere, probably on the depot.


Edit: minor wording.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: frith01 on 05/15/2022 10:32 pm
Why not use 2 double walled tubes.  Inside tube is always methane (or whichever side has more volume), outside tube is always oxygen.  At launch use Inside tube A,  Outside Tube B .  For Sending fuel, Use Inside Tube B, Outside Tube A.

 QD Arm / SUPPLY Ship Connection                       Starship Receive Connection
   Tube B                                                             Tube A
=========                                                  ==========
Oxygen  --||  Closed                                            ||--  Closed Oxy
----------
Methane Out -->> OPEN                                       -->>. Methane IN
----------
Oxygen  --||                                                       || --
=========                                                ==========


  Tube A                                                        Tube B
=========                                              ===============                         
Oxygen Out  -->> OPEN                                -->> Oxy In 
----------                                                      ------------------
Methane --|| Closed                                        || Meth closed
----------                                                      ------------------
Oxygen   -->> OPEN                                     -->>
=========                                           =================



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/16/2022 05:30 am
If truly androgynous cryo connectors are a problem, hermaphroditic ones won't be.  Here's a simple example, with hermaprhoditic cyro and androgynous ullage gas.  (Hermaphoditic ullage is also easy, but probably unnecessary.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/16/2022 08:53 am
Why not use 2 double walled tubes.  Inside tube is always methane (or whichever side has more volume), outside tube is always oxygen.  At launch use Inside tube A,  Outside Tube B .  For Sending fuel, Use Inside Tube B, Outside Tube A.
Because you're one unsealed mating face away from mixing fuel and oxidiser in high volumes.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/16/2022 03:15 pm
Why not use 2 double walled tubes.  Inside tube is always methane (or whichever side has more volume), outside tube is always oxygen.  At launch use Inside tube A,  Outside Tube B .  For Sending fuel, Use Inside Tube B, Outside Tube A.

 QD Arm / SUPPLY Ship Connection                       Starship Receive Connection
   Tube B                                                             Tube A
=========                                                  ==========
Oxygen  --||  Closed                                            ||--  Closed Oxy
----------
Methane Out -->> OPEN                                       -->>. Methane IN
----------
Oxygen  --||                                                       || --
=========                                                ==========


  Tube A                                                        Tube B
=========                                              ===============                         
Oxygen Out  -->> OPEN                                -->> Oxy In 
----------                                                      ------------------
Methane --|| Closed                                        || Meth closed
----------                                                      ------------------
Oxygen   -->> OPEN                                     -->>
=========                                           =================
Even a little leak at the seals=a big problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/16/2022 03:25 pm
If truly androgynous cryo connectors are a problem, hermaphroditic ones won't be.  Here's a simple example, with hermaprhoditic cyro and androgynous ullage gas.  (Hermaphoditic ullage is also easy, but probably unnecessary.)
This interface cannot connect to itself. Try it yourself. To self-mate, it needs to be congruent with a rotation of itself.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 05/16/2022 03:57 pm
If truly androgynous cryo connectors are a problem, hermaphroditic ones won't be.  Here's a simple example, with hermaprhoditic cyro and androgynous ullage gas.  (Hermaphoditic ullage is also easy, but probably unnecessary.)
This interface cannot connect to itself. Try it yourself. To self-mate, it needs to be congruent with a rotation of itself.
Rotate around the y-axis (i.e. the line on the screen through the two ullage connectors.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/16/2022 04:17 pm
If truly androgynous cryo connectors are a problem, hermaphroditic ones won't be.  Here's a simple example, with hermaprhoditic cyro and androgynous ullage gas.  (Hermaphoditic ullage is also easy, but probably unnecessary.)
If the goal is to have only one standard shipboard QD plate, androgynous connectors will work. If androgynous connectors can not seal properly or show reasonable MTBF, a single QD plate design has polarity problems and appears undoable. 


I need to dig out the tinker toys to double check this. If the above is correct, the QD connectors will need to be augmented by a mirror copy somewhere in the system.


The depot seems the place for mirror connections. For LEO transfers this looks to be the most mass efficient. For higher energy orbits its less clear but keeping the mirror on the depot keeps over all system complexity down.


If a mission profile calls for direct tanker - receiving ship transfer, it will not work short of adding an extra QD plate to the tanker and creating a tanker/depot hybrid. Not that outrageous an idea in the long term. Short term - every situation is a situation.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Joffan on 05/16/2022 05:13 pm
If truly androgynous cryo connectors are a problem, hermaphroditic ones won't be.  Here's a simple example, with hermaprhoditic cyro and androgynous ullage gas.  (Hermaphoditic ullage is also easy, but probably unnecessary.)
This interface cannot connect to itself. Try it yourself. To self-mate, it needs to be congruent with a rotation of itself.
Sorry Dan, this is wrong and the design is fine because the two connectors will be facing in opposite directions. Imagine two copies of the interface and fold onto each other:
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/16/2022 06:03 pm
If truly androgynous cryo connectors are a problem, hermaphroditic ones won't be.  Here's a simple example, with hermaprhoditic cyro and androgynous ullage gas.  (Hermaphoditic ullage is also easy, but probably unnecessary.)
This interface cannot connect to itself. Try it yourself. To self-mate, it needs to be congruent with a rotation of itself.
Sorry Dan, this is wrong and the design is fine because the two connectors will be facing in opposite directions. Imagine two copies of the interface and fold onto each other:
Hmm. Maybe I need to rethink too. Where are those damned tinker toys?


Edit to add: androgynous ullage connectors might work at low pressure to help with transfer but 6 bar might be difficult.  Split the lines behind the QD plate and use the two connection hermaphrodite solution given earlier. Especially if the depot mission evolves to where it's required to top off the receiver's high pressure ullage COPV's. Probably a useless feature. Real experience will be our guide.


A high pressure androgynous QD would be a great thing to have. So would a Lear Jet and a hundred grand. I'm not holding my breath for either but I can dream.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: frith01 on 05/16/2022 10:23 pm
Why not use 2 double walled tubes.  Inside tube is always methane (or whichever side has more volume), outside tube is always oxygen.  At launch use Inside tube A,  Outside Tube B .  For Sending fuel, Use Inside Tube B, Outside Tube A.
Because you're one unsealed mating face away from mixing fuel and oxidiser in high volumes.

At the time of use, one pipe would only be used for 1 fluid.  vent the other side to vacuum or open air after split to 2 single pipes. Overlap of fluid use in the double pipes only needs to occur for 1-2 feet from the interface. Separate the double pipes by whatever comfortable safety distance is needed.

So Double pipe, Y into 2 single pipes , vent valve, tank valve.  They seem ok with Down-comers of 2 different gases in more extreme volumes. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/16/2022 11:45 pm
Why not use 2 double walled tubes.  Inside tube is always methane (or whichever side has more volume), outside tube is always oxygen.  At launch use Inside tube A,  Outside Tube B .  For Sending fuel, Use Inside Tube B, Outside Tube A.
Because you're one unsealed mating face away from mixing fuel and oxidiser in high volumes.

At the time of use, one pipe would only be used for 1 fluid.  vent the other side to vacuum or open air after split to 2 single pipes. Overlap of fluid use in the double pipes only needs to occur for 1-2 feet from the interface. Separate the double pipes by whatever comfortable safety distance is needed.

So Double pipe, Y into 2 single pipes , vent valve, tank valve.  They seem ok with Down-comers of 2 different gases in more extreme volumes.
I'm having a hard time following exactly what you mean. It's unclear where the overlap is. Could you do a sketch? Doesn't have to be pretty.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Llian Rhydderch on 05/16/2022 11:50 pm
... projected timeline for first attempted orbital refueling between two Starships?
...
Four things you can do with every single Starlink mission:

1) Launch to orbit.
2) Deploy Starlinks.
3) Attempt a refueling RPOD.
4) Attempt an EDL.
...

Could you kindly expand the acronym RPOD?

Many pop in on these threads less regularly than the heavy discussants.  Including the real mean of an esoteric acronym from time to time is super helpful.   :)

(and Wikipedia doesn't have this acronym; I checked)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: MikeWest on 05/17/2022 12:22 am
One possible answer: Rendezvous, proximity operations and docking (RPOD)
 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 05/17/2022 12:32 am
Every time someone unnecessarily uses a vague acronym, the get a demerit. :P
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tj on 05/17/2022 12:44 am
Remember DOD Boeing Orbital Express. LEOSat SV refueling. Planned. Successful! more than ten years ago. So generally state of industry practice by now or very close to it. Boeings fuel truck / repair Sat even had their reaction wheels running backwards and had to have the to be refueled Sat take over unti contro Sat wheels were dealt with!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: frith01 on 05/17/2022 02:36 am
Why not use 2 double walled tubes.  Inside tube is always methane (or whichever side has more volume), outside tube is always oxygen.  At launch use Inside tube A,  Outside Tube B .  For Sending fuel, Use Inside Tube B, Outside Tube A.
Because you're one unsealed mating face away from mixing fuel and oxidiser in high volumes.

At the time of use, one pipe would only be used for 1 fluid.  vent the other side to vacuum or open air after split to 2 single pipes. Overlap of fluid use in the double pipes only needs to occur for 1-2 feet from the interface. Separate the double pipes by whatever comfortable safety distance is needed.

So Double pipe, Y into 2 single pipes , vent valve, tank valve.  They seem ok with Down-comers of 2 different gases in more extreme volumes.

Rough Draft, but it gets the idea across. 

Add another pair  of double wall pipes for Pressure gas exchange.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 05/17/2022 03:44 am
Whatever the design will be, there's a due date in about 1.4 years due to Artemis contract:

https://forum.nasaspaceflight.com/index.php?topic=50806.msg2369207#msg2369207
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ehb on 05/17/2022 05:27 am
Whatever the design will be, there's a due date in about 1.4 years due to Artemis contract:

https://forum.nasaspaceflight.com/index.php?topic=50806.msg2369207#msg2369207
perhaps not ship to ship if milestone is this https://forum.nasaspaceflight.com/index.php?topic=53849.0
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/17/2022 08:13 am
Remember with paired-androgeny connectors, you are still adding 2x the number of plumbing connections but shoving them all into the same QD plate installed on every Starship.  If you take the 'extra' connectors and instead move them to a dedicated 'anti' QD plate and additional plumbing and valving, then only the depot variant needs to carry that extra mass to orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: frith01 on 05/17/2022 09:40 am
Remember with paired-androgeny connectors, you are still adding 2x the number of plumbing connections but shoving them all into the same QD plate installed on every Starship.  If you take the 'extra' connectors and instead move them to a dedicated 'anti' QD plate and additional plumbing and valving, then only the depot variant needs to carry that extra mass to orbit.

Couple hundred kg at most, makes all starships cross-loadable.  good trade-off if you ask me.  Mission flexibility, standardization, reliability, and most importantly:  Quicker to Moon/Mars , due to non-zero time to develop/test tanker variant
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 05/17/2022 10:17 am
How hard would it be for later models (androgonus) to carry an adaptor for the simpler early model connection (Single pipe active connector on an old tanker)?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/17/2022 01:44 pm
My big problem with all of these interconnects is how the interfaces will be brought together. I seems to me that one or both of the two ship will need to extend the interface on some sort of extrusion or boom. If that boom is on an EDL-capable ship, you have an "interesting" engineering problem. If this connection is the ship's only propellant transfer interface, also connects to the QD arm on the tower.

The complexity of this boom is the reason I think the interface will not be androgynous. A ship that just stays in space can have an elaborate boom structure because it does not EDL, and the EDL-capable ship will not need the boom and therefore does not need a way to make it aerodynamic.  Thus, I think they will just put this boom on the depot and nowhere else.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/17/2022 03:00 pm
Remember with paired-androgeny connectors, you are still adding 2x the number of plumbing connections but shoving them all into the same QD plate installed on every Starship.  If you take the 'extra' connectors and instead move them to a dedicated 'anti' QD plate and additional plumbing and valving, then only the depot variant needs to carry that extra mass to orbit.

Couple hundred kg at most, makes all starships cross-loadable.  good trade-off if you ask me.  Mission flexibility, standardization, reliability, and most importantly:  Quicker to Moon/Mars , due to non-zero time to develop/test tanker variant
The universal cross loading is a good thing but the coupler complexity might be more than you expect. With that reliability goes down. No matter how carefully the refueling hookup is handled there will be repeated mild impacts and micro misalignments. A coax coupler would be devilishly difficult to make robust. Not impossible, just hard. Hard = expensive.

Coax plumbing is difficult to fabricate to withstand pressure. Not impossible, just not as straight forward simple plumbing runs. The inner tube needs to pass through the outer tube at some point and this is a place where stress will collect.

Dual use runs speculatively might need an N2 purge at switchover. Again, not impossible, but added physical and operational complexity (= expensive).

A coax system would probably mass less than a second mirror QD while being more expensive, less robust and more complex. Mass is one thing that impacts the SS system less than is usual in this wacky business.


Edit to add: testing the coax system might actually take longer to test and debug than a mirrored QD plate. The mirror plate would be using couplers that are already in use.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/17/2022 03:55 pm
How hard would it be for later models (androgonus) to carry an adaptor for the simpler early model connection (Single pipe active connector on an old tanker)?
Staying away from androgynous specifically and considering any changes to the QD that create mix n match problems, finding a way to deploy the adapter would be a major problem.


Let's say they change the connections on the SS and its GSE. Any depots on orbit would be unable to service it. This makes an adapter look appealing. The easiest solution would be: DON'T DO THIS. If the QD absolutely has to be changed the whole system would need eventual upgrade and two parallel systems run while this happens. Shudder.


The upgrade could be done in stages. Before a new style LSS (for example) could be launched, any and all depots it would depend on would need a new style depot and a fleet of new style tankers. If both old and new style LSS's would be plying this route the new depot(s) would be companions to the old until all LSS's go new style. There would also have to be both old and new launch mounts. It's messy, so an adapter looks like a great solution until actually deploying it is considered.


Crack the deployment nut and it works. It would (opinion) be easier to swap out the depots and tankers and be done with it (how many, and what is their life expectancy?).


In the long run there will be general purpose service ships with arms for just this type of adapter install mission but we'll probably be in mars before they show up.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/17/2022 05:04 pm
There seems to be an assumption that on-orbit refueling will be done via the QD that talks to the GSE system.  I don't think this is going to work, because the QD is near the tail, where torques from docking errors will be near their maximum.  I think you need something close to a hard-dock midships to reduce the torques.

I'm also not understanding why anybody's freaked out about an androgynous system.  It's just not that hard, nor does it weigh very much.  Hermaphroditic connectors incur the mass of two connectors, but not two sets of plumbing.  Just hook both of 'em up to the same manifold, with one pipe coming off the manifold, and you're done.  It might be an extra 100kg, but I doubt it.

I still think that the docking mechanism and the plumbing connection mechanism are largely independent of one another.  If two Starships can soft-capture using an active-active configuration, then they can be brought arbitrarily close to one another during retraction.  That reduces the extension that one or both ships need to do to engage the refueling plumbing down to centimeters.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/17/2022 05:52 pm
There seems to be an assumption that on-orbit refueling will be done via the QD that talks to the GSE system.  I don't think this is going to work, because the QD is near the tail, where torques from docking errors will be near their maximum.  I think you need something close to a hard-dock midships to reduce the torques.

I'm also not understanding why anybody's freaked out about an androgynous system.  It's just not that hard, nor does it weigh very much.  Hermaphroditic connectors incur the mass of two connectors, but not two sets of plumbing.  Just hook both of 'em up to the same manifold, with one pipe coming off the manifold, and you're done.  It might be an extra 100kg, but I doubt it.

I still think that the docking mechanism and the plumbing connection mechanism are largely independent of one another.  If two Starships can soft-capture using an active-active configuration, then they can be brought arbitrarily close to one another during retraction.  That reduces the extension that one or both ships need to do to engage the refueling plumbing down to centimeters.
Why the QD connector for the non-depot ship? Because that ship already had the QD connector, and "the best part is no part", so adding anything else adds mass and complexity. On teh other hand the depot is a specialized ship whose function is to transfer propellant, so you can add all needed hardware to this one ship instead of adding ti to every ship.

But that adds the requirement to stabilize the two ships when they are connected. Again this stabilization hardware can be part of the depot and does not need to be added to every Starship, except for an interface of some sort. Let's for the heck of it assume the Depot's transfer interface is near the stern and connects to the other ship's QD interface, and further assume it is designed to provide some rigidity. further assume the ships will both face "forward" and use a small amount of thrust to settle the propellant. It seems to me that the ships will need a rigid connection near the bow to keep this mess together. The non-depot has chopstick lift points, so again "the best part is no part" and we do not need another interface. The depot can be designed with a rigid pair of articulated grabber arms to connect to those lift points.   All of this adds to the complexity of the depot (now better named the propellant transfer ship or PTS), but it means that no other variant needs any additional hardware.

Yes, I'm sorry to lose a more generalized buddy transfer capability, but I see the cost in mass, volume, and complexity as being too high. If there is ever a need for propellant transfer at a location (NRHO, LLO, GTO, Mars orbit...), deploy another PTS: they are fairly cheap after all.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/17/2022 10:54 pm
Why the QD connector for the non-depot ship? Because that ship already had the QD connector, and "the best part is no part", so adding anything else adds mass and complexity. On teh other hand the depot is a specialized ship whose function is to transfer propellant, so you can add all needed hardware to this one ship instead of adding ti to every ship.

I'm skeptical that a QD that's designed to provide reliable disconnect, but which relies on gravity and friction to connect in the first place, will be a sound basis for a microgravity docking system.  And a docking system is required.  It probably won't be IDSS-compliant.  But it has the same basic requirements:  soft capture, damping out positioning errors, the retract to some hard-ish dock so that the fueling connectors can link up.  As I said up-thread, the best part is no part if the other parts can do double duty.  Sometimes they can't.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/18/2022 11:01 am
but which relies on gravity and friction to connect in the first place
As we can see from the current QD plates, alignment is driven by spherical ended pins, mating is driven by actuators, and latching is driven by actuated latches. Sealing is accomplished by mating conical surfaces for fluid connectors. None of these require gravity, and do not require external atmospheric pressure (in some ways that simplifies things due to the auto-purging behaviour). There do nor appear to be any elastomeric components involved in either side of the QD plate, possibly barring internal O-rings (that only need to act as static seals rather than dynamic). I'm not sure what you mean by 'no fraction', friction still occurs in a vacuum & microgravity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/18/2022 03:24 pm
There seems to be an assumption that on-orbit refueling will be done via the QD that talks to the GSE system.  I don't think this is going to work, because the QD is near the tail, where torques from docking errors will be near their maximum.  I think you need something close to a hard-dock midships to reduce the torques.

I'm also not understanding why anybody's freaked out about an androgynous system.  It's just not that hard, nor does it weigh very much.  Hermaphroditic connectors incur the mass of two connectors, but not two sets of plumbing.  Just hook both of 'em up to the same manifold, with one pipe coming off the manifold, and you're done.  It might be an extra 100kg, but I doubt it.

I still think that the docking mechanism and the plumbing connection mechanism are largely independent of one another.  If two Starships can soft-capture using an active-active configuration, then they can be brought arbitrarily close to one another during retraction.  That reduces the extension that one or both ships need to do to engage the refueling plumbing down to centimeters.
Why the QD connector for the non-depot ship? Because that ship already had the QD connector, and "the best part is no part", so adding anything else adds mass and complexity. On teh other hand the depot is a specialized ship whose function is to transfer propellant, so you can add all needed hardware to this one ship instead of adding ti to every ship.

But that adds the requirement to stabilize the two ships when they are connected. Again this stabilization hardware can be part of the depot and does not need to be added to every Starship, except for an interface of some sort. Let's for the heck of it assume the Depot's transfer interface is near the stern and connects to the other ship's QD interface, and further assume it is designed to provide some rigidity. further assume the ships will both face "forward" and use a small amount of thrust to settle the propellant. It seems to me that the ships will need a rigid connection near the bow to keep this mess together. The non-depot has chopstick lift points, so again "the best part is no part" and we do not need another interface. The depot can be designed with a rigid pair of articulated grabber arms to connect to those lift points.   All of this adds to the complexity of the depot (now better named the propellant transfer ship or PTS), but it means that no other variant needs any additional hardware.

Yes, I'm sorry to lose a more generalized buddy transfer capability, but I see the cost in mass, volume, and complexity as being too high. If there is ever a need for propellant transfer at a location (NRHO, LLO, GTO, Mars orbit...), deploy another PTS: they are fairly cheap after all.
To amplify a bit, if the depot doesn't return, it has no fins. The fin reinforcement structure forms a natural base for any connection and alignment thingies. Reach out for the chopstick interface up high and find or introduce something aft. Pare back the depots fin reinforcement as much as possible but don't throw it away.


So far we've been talking about propellant transfer only. The goal is for those to be the only fluids but odds are there will be N2 in some form so why not include that in the service package. If there are other fluids (hopefully not) include them. It will be a service station and not just a 'gas' station so calling it a depot works. When they put a C store in it we can rethink the name.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/18/2022 04:27 pm
There seems to be an assumption that on-orbit refueling will be done via the QD that talks to the GSE system.  I don't think this is going to work, because the QD is near the tail, where torques from docking errors will be near their maximum.  I think you need something close to a hard-dock midships to reduce the torques.

I'm also not understanding why anybody's freaked out about an androgynous system.  It's just not that hard, nor does it weigh very much.  Hermaphroditic connectors incur the mass of two connectors, but not two sets of plumbing.  Just hook both of 'em up to the same manifold, with one pipe coming off the manifold, and you're done.  It might be an extra 100kg, but I doubt it.

I still think that the docking mechanism and the plumbing connection mechanism are largely independent of one another.  If two Starships can soft-capture using an active-active configuration, then they can be brought arbitrarily close to one another during retraction.  That reduces the extension that one or both ships need to do to engage the refueling plumbing down to centimeters.
Personally, I'm not freaked by androgynous connectors. I just see them as a techno hump that might be higher than you realize and that might call for too many compromises.


Mating is a specialized operation that is part of the core purpose of the depot so give it the lions share of the job and leave the tanker and receiving ships in as passive a role as possible. Once the tanker or receiving ship has cancelled out as much yaw, roll and pitch as it is capable of, let specialized hardware on the depot finish it off. The residual torques should not be much of a problem.


And yes, connecting the two ships and connecting the plumbing are two different problems although if the connector sealing faces are tapers without elastomers, they need to be held together with some force. This can be whatever the QD system uses now. If the QD plate can take on the role of aft soft docking it's a twofer. If it can do a hard dock too, that's a good thing. The system also needs to hook up towards the fore end.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/18/2022 05:55 pm
I just don't see much benefit for androgynous connections. It adds mass to every vehicle, without any practical real-world capability. And you need to pay to carry that extra mass to orbit with extra propellant, for every single launch over the life of that Starship [1].

As an example:
- If you can launch on a single Starship with no prop transfer, the androgynous connector and plumbing are superfluous.
- If you need a single tanker launch to refill a Starship to get the payload where it needs to go, you can instead of launching a tanker launch a single depot. The extra mass of the 'supplier' QD subtracts from the propellant it could lift, but the extra mass of the androgynous QD and plumbing (on both Starships) would also subtract from the propellant available and add to the dry mass of the payload Starship.
- If you need more than one tanker launch to re-fill the Starship, you need the depot anyway to allow the payload carrying Starship to launch last instead of first (launching first front-loads all of your risk and results in the payload stuck in a fairing for several days without ground support, which is more of a pain than it sounds).


[1] e.g. if the extra plumbing and enlarged and more complex QD plate add an extra ~1 tonne of mass, and Starship can manage ~100 flights before retirement, and you manage to get rough cost/kg down to ~$100/kg, you pay ~$10mn over the life of that Starship for that idle capability as total Starship launch cost trends towards propellant cost in the long run.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/18/2022 07:15 pm
I just don't see much benefit for androgynous connections. It adds mass to every vehicle, without any practical real-world capability. And you need to pay to carry that extra mass to orbit with extra propellant, for every single launch over the life of that Starship [1].

As an example:
- If you can launch on a single Starship with no prop transfer, the androgynous connector and plumbing are superfluous.
- If you need a single tanker launch to refill a Starship to get the payload where it needs to go, you can instead of launching a tanker launch a single depot. The extra mass of the 'supplier' QD subtracts from the propellant it could lift, but the extra mass of the androgynous QD and plumbing (on both Starships) would also subtract from the propellant available and add to the dry mass of the payload Starship.
- If you need more than one tanker launch to re-fill the Starship, you need the depot anyway to allow the payload carrying Starship to launch last instead of first (launching first front-loads all of your risk and results in the payload stuck in a fairing for several days without ground support, which is more of a pain than it sounds).


[1] e.g. if the extra plumbing and enlarged and more complex QD plate add an extra ~1 tonne of mass, and Starship can manage ~100 flights before retirement, and you manage to get rough cost/kg down to ~$100/kg, you pay ~$10mn over the life of that Starship for that idle capability as total Starship launch cost trends towards propellant cost in the long run.
I just don't see how you can engineer a peer transfer apparatus  with a mass of one tonne or less. It's not just the connector, it's the whole apparatus: the connector itself plus whatever boom or extruder or arm plus whatever stabilization scheme is needed, and all of this mess must be capable of EDL on an EDL-capable ship, including on a tanker. Yes, the velocities and accelerations needed for docking and for subsequent propellant settling are small, and yes sophisticated software controlling the RCS on both ships can help, but this is a connection between two masses each in excess of 200 tonnes. F=ma.   0.01 m/s2 * 200 t = 2000 N, which is about 450 pounds force. You need to build this strong conformable structure so it can be deployed when needed and then tucked away for re-entry.  docking will need to occur when the peer ships are somewhere between 2 m and 10 m apart, flying in formation dorsal-to-dorsal, so each peer's mating boom is between 1 m and 5 m long in the dorsal direction plus whatever tolerances you need in the fore-and-aft and port-and-starboard directions, and at least a little bit of pitch, roll, and yaw. This is all in addition to whatever peer stabilization you need at the bows.

So don't build a peer transfer apparatus. Put all the mass in the depot.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 05/18/2022 07:21 pm
None of these design details are that important compared to getting something in the way of orbital refuelling working.  Make it work first, then make it work better.  There are plenty of different ways of doing this without an ideal initial design.

Early on, and perhaps for a long time, the entire refueling chain will be planned ahead.  If you have a few female and male ships just make sure you dispatch the correct gender for each mission.  Maybe you'd occasionally want to swap the gender of a ship the same way you convert an F9 into a FH side booster.  You don't need an androgenous connector, a hermaphrodite ship, or a specially configured depot to make this work

If you can't use the GSE connector for orbital refuelling it might work the other way around.  Once you have a working orbital connector revamp the GSE and delete the old GSE connector from each ship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 05/18/2022 07:23 pm
I just don't see much benefit for androgynous connections. It adds mass to every vehicle, without any practical real-world capability. And you need to pay to carry that extra mass to orbit with extra propellant, for every single launch over the life of that Starship [1].

As an example:
- If you can launch on a single Starship with no prop transfer, the androgynous connector and plumbing are superfluous.
- If you need a single tanker launch to refill a Starship to get the payload where it needs to go, you can instead of launching a tanker launch a single depot. The extra mass of the 'supplier' QD subtracts from the propellant it could lift, but the extra mass of the androgynous QD and plumbing (on both Starships) would also subtract from the propellant available and add to the dry mass of the payload Starship.
- If you need more than one tanker launch to re-fill the Starship, you need the depot anyway to allow the payload carrying Starship to launch last instead of first (launching first front-loads all of your risk and results in the payload stuck in a fairing for several days without ground support, which is more of a pain than it sounds).

One thing to note (without declaring which side of this discussion is more right): depots as currently planned are incapable of EDL.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/18/2022 07:25 pm
I just don't see much benefit for androgynous connections. It adds mass to every vehicle, without any practical real-world capability. And you need to pay to carry that extra mass to orbit with extra propellant, for every single launch over the life of that Starship [1].

Two things:

1) If you don't need to refuel, don't install the hardware.  Maintaining a fleet of "optimized for LEO or GTO" Starships doesn't sound very difficult.

2) There are likely to be some sump losses with every prop transfer.  If you introduce an intermediary in all cases, those sump losses will always be larger than if you did a peer-to-peer transfer.  I'd be pretty surprised if the mass of the androgynous (or hermaphroditic) hardware was larger than the sump losses.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 05/18/2022 07:35 pm
2) There are likely to be some sump losses with every prop transfer.  If you introduce an intermediary in all cases, those sump losses will always be larger than if you did a peer-to-peer transfer.  I'd be pretty surprised if the mass of the androgynous (or hermaphroditic) hardware was larger than the sump losses.
You get a sump loss for each ship not for each transfer.  Multiple transfers don't hurt.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 05/18/2022 07:44 pm
2) There are likely to be some sump losses with every prop transfer.  If you introduce an intermediary in all cases, those sump losses will always be larger than if you did a peer-to-peer transfer.  I'd be pretty surprised if the mass of the androgynous (or hermaphroditic) hardware was larger than the sump losses.
You get a sump loss for each ship not for each transfer.  Multiple transfers don't hurt.
The propellant needed for settling thrust during each transfer is far larger than any reasonable estimate for the extra transfer hardware.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/18/2022 11:27 pm
2) There are likely to be some sump losses with every prop transfer.  If you introduce an intermediary in all cases, those sump losses will always be larger than if you did a peer-to-peer transfer.  I'd be pretty surprised if the mass of the androgynous (or hermaphroditic) hardware was larger than the sump losses.
You get a sump loss for each ship not for each transfer.  Multiple transfers don't hurt.

It's for each transfer if you're trying to use all the prop in the depot.  Whether that's a big deal or not depends on how much prop is in the depot and how much is needed.  If you have a mission that needs 7.2 lift tankers and it's just as cheap to get 8 get delivered, then the extra transfer out of the depot isn't a big deal.  But if it needs 7.95 lift tankers' worth, then the transfer out of the depot could force a ninth launch.

I doubt this is a problem for HLS, because the 9 tankers required to do LEO-NRHO-LS-NRHO is so much prop that the depot will be a net positive to curtail boil-off.  But this might not be the case for an LEO-NRHO-LEO transit mission (5 tankers) or for a direct insertion and return to GEO (slightly less than 3 tankers).

And you still don't want to use a depot in NRHO if at all possible, because it can't be returned cheaply.  It's far better to use a lift tanker, even if the prop needs to be stockpiled in a depot before being transferred to the last lift tanker.  That's the only way to recover the tanker.

The propellant needed for settling thrust during each transfer is far larger than any reasonable estimate for the extra transfer hardware.

Let's figure this out.  Suppose you have 1500t of prop in a depot that's 105t dry, and a 95t LSS with 30t of crew module and cargo.  Ullage gas thrusters will get... Isp= 70s = 686m/s?  And we'll need 0.01m/s² of settling acceleration for let's say 30 min (1800s).  That's 18m/s of delta-v.

We can use the aggregate mass of the two Starships (1605t + 125t = 1730t) as the inert mass for the rocket equation.  Then propRequired = (exp(18/686)-1)*1730t = 46t.

Yup, the hardware is going to be insignificant. 

Note, however, that this is going to be substantially more mass than there will be for warm ullage gas @ 6bar.  That implies that this new scheme of using only ullage pressure for RCS won't work for settling.  Maybe methox thrusters on the depot will still be required?  That would be something that would be a lot harder to replicate on all ships in the fleet than an androgynous/hermaphroditic prop connector.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/18/2022 11:29 pm
I just don't see much benefit for androgynous connections. It adds mass to every vehicle, without any practical real-world capability. And you need to pay to carry that extra mass to orbit with extra propellant, for every single launch over the life of that Starship [1].

Two things:

1) If you don't need to refuel, don't install the hardware.  Maintaining a fleet of "optimized for LEO or GTO" Starships doesn't sound very difficult.
That's another point against an androgynous QD plate: you don't have the option not to install it, as it's the only QD plate you have. You're stuck with it on every vehicle or they will be incompatible with your GSE.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/19/2022 12:27 am
I just don't see much benefit for androgynous connections. It adds mass to every vehicle, without any practical real-world capability. And you need to pay to carry that extra mass to orbit with extra propellant, for every single launch over the life of that Starship [1].

Two things:

1) If you don't need to refuel, don't install the hardware.  Maintaining a fleet of "optimized for LEO or GTO" Starships doesn't sound very difficult.
That's another point against an androgynous QD plate: you don't have the option not to install it, as it's the only QD plate you have. You're stuck with it on every vehicle or they will be incompatible with your GSE.
You're assuming that the QD and the refueling mechanism are one and the same.  I think that's unlikely.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Overtone on 05/19/2022 02:08 am
Let's figure this out.  Suppose you have 1500t of prop in a depot that's 105t dry, and a 95t LSS with 30t of crew module and cargo.  Ullage gas thrusters will get... Isp= 70s = 686m/s?  And we'll need 0.01m/s² of settling acceleration for let's say 30 min (1800s).  That's 18m/s of delta-v.

We can use the aggregate mass of the two Starships (1605t + 125t = 1730t) as the inert mass for the rocket equation.  Then propRequired = (exp(18/686)-1)*1730t = 46t.

Wow, that's a lot of prop mass for ullage.  That opens the question: why not use mechanical means to settle the propellant in the tanks, and do the propellant transfer in free fall?   I appreciate feedback or a pointer to previous discussion.

I imagine a plate inside the tank, riding on a long screw turned by a motor.  Only low pressure is required to get the propellant to the pump intakes, so the motor doesn't have to be big, the seals between the plate and the tank walls can be simple and leaky, and the plate can be thin and flexible hence lightweight.  It's going to weigh something noticeable but gut feeling says a lot less than 40t.

Lunar- and mars-starships wouldn't need the propellant compression mechanism.  Tankers need the mechanism, but if there are separate payload propellant tanks in the nosecone barrel, the tanks will be (comparatively) low diameter so weight of plates and motor for each tank seems low.  Weight of the propellant compression mechanism in the depot is higher due to high-diameter tanks involved, but the depot is launch-once and never lands so maybe this weight is acceptable to save the big ullage mass losses that otherwise hit every time you refill a starship.

As a side benefit, you simplify the challenge of the docking mechanism.  When using ullage to settle, the dock has to withstand the shear and bending stress associated with any mismatch in the acceleration vectors of the two vehicles, which are basically flying in close formation while trying to avoid hitting each other with their attitude control thrusters.

As another side benefit, mission planning gets simpler.  There's complexity associated with predicting, and adapting tanker launch timing to match, the delta-V's that add up on the depot during a sequence of long ullage burns. And as LEO gets more crowded, there's complexity associated with avoiding collision risks, which is much easier if the depot goes into one orbit and just stays there for a long time.

As a side complexity, any collection of pipes or wires running through a tank with a compression mechanism needs a fairing that the (leaky) compression plate seals can ride against.   Wouldn't expect any penetrations if dedicated payload tanks are used in in the tanker, but they will exist everywhere else.

Thoughts?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/19/2022 07:12 am
I just don't see much benefit for androgynous connections. It adds mass to every vehicle, without any practical real-world capability. And you need to pay to carry that extra mass to orbit with extra propellant, for every single launch over the life of that Starship [1].

Two things:

1) If you don't need to refuel, don't install the hardware.  Maintaining a fleet of "optimized for LEO or GTO" Starships doesn't sound very difficult.
That's another point against an androgynous QD plate: you don't have the option not to install it, as it's the only QD plate you have. You're stuck with it on every vehicle or they will be incompatible with your GSE.
You're assuming that the QD and the refueling mechanism are one and the same.  I think that's unlikely.
That was exactly my point: an androgynous QD used for propellant transfer does not make sense in terms of dry mass or CONOPs, and it is better to have a separate system for inter-vehicle transfer fitted only to those vehicles that need it ('depots'/'transfer-tankers'/whatever-you-call-them).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/19/2022 10:42 am
Ullage gas thrusters will get... Isp= 70s = 686m/s?

AIUI, Starship will have hot-gas thrusters. Isp=~300s.

= 46t.

Or else about 30kg.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 05/19/2022 11:05 am
Ullage gas thrusters will get... Isp= 70s = 686m/s?

AIUI, Starship will have hot-gas thrusters. Isp=~300s.

= 46t.

Or else about 30kg.

No. Just check out the part 1 of the latest EDA's - Elon interview/walkthrough.

It would be ullage gas which is likely around 400-500K. And would likely use methane which is also better than nitrogen. Elevated temperature and lighter molecular weight should combine to give ~120s ISP. But this is  far cry form 300s.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/19/2022 04:06 pm
Let's figure this out.  Suppose you have 1500t of prop in a depot that's 105t dry, and a 95t LSS with 30t of crew module and cargo.  Ullage gas thrusters will get... Isp= 70s = 686m/s?  And we'll need 0.01m/s² of settling acceleration for let's say 30 min (1800s).  That's 18m/s of delta-v.

We can use the aggregate mass of the two Starships (1605t + 125t = 1730t) as the inert mass for the rocket equation.  Then propRequired = (exp(18/686)-1)*1730t = 46t.

Wow, that's a lot of prop mass for ullage.  That opens the question: why not use mechanical means to settle the propellant in the tanks, and do the propellant transfer in free fall?   I appreciate feedback or a pointer to previous discussion.

I imagine a plate inside the tank, riding on a long screw turned by a motor.  Only low pressure is required to get the propellant to the pump intakes, so the motor doesn't have to be big, the seals between the plate and the tank walls can be simple and leaky, and the plate can be thin and flexible hence lightweight.  It's going to weigh something noticeable but gut feeling says a lot less than 40t.

Lunar- and mars-starships wouldn't need the propellant compression mechanism.  Tankers need the mechanism, but if there are separate payload propellant tanks in the nosecone barrel, the tanks will be (comparatively) low diameter so weight of plates and motor for each tank seems low.  Weight of the propellant compression mechanism in the depot is higher due to high-diameter tanks involved, but the depot is launch-once and never lands so maybe this weight is acceptable to save the big ullage mass losses that otherwise hit every time you refill a starship.

As a side benefit, you simplify the challenge of the docking mechanism.  When using ullage to settle, the dock has to withstand the shear and bending stress associated with any mismatch in the acceleration vectors of the two vehicles, which are basically flying in close formation while trying to avoid hitting each other with their attitude control thrusters.

As another side benefit, mission planning gets simpler.  There's complexity associated with predicting, and adapting tanker launch timing to match, the delta-V's that add up on the depot during a sequence of long ullage burns. And as LEO gets more crowded, there's complexity associated with avoiding collision risks, which is much easier if the depot goes into one orbit and just stays there for a long time.

As a side complexity, any collection of pipes or wires running through a tank with a compression mechanism needs a fairing that the (leaky) compression plate seals can ride against.   Wouldn't expect any penetrations if dedicated payload tanks are used in in the tanker, but they will exist everywhere else.

Thoughts?
IIRC the initial failure delivering the last Russian module to the ISS was a leak in an internal bellows system designed to do just this. It's complex but so is the alternative.


I think the key is designing this type of displacement from the beginning. To modify to it on even a mid term prototype like we have, would probably call for a program pause while a lot of fiddly things are tested.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/19/2022 04:20 pm
In reading the last day of posts I find myself going cross eyed. Part of it is terminology. We seem to be using different words to say the same thing, and maybe the same words to say different things. May I humbly propose the following:


Receiver, or Receiver Ship: any vessel receiving propellant from a Depot:


Depot: a vessel that accumulates propellant for transfer to a Receiver.


Tanker: a vessel that delivers propellant to LEO or higher.


Tanker/Accumulator: a vessel that combines the functions of both Depot and Tanker.


Androgynous Connector: a connector that has no identifiable gender.


Hermaphrodite connector: a connector that has elements of both genders.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/19/2022 06:02 pm
In reading the last day of posts I find myself going cross eyed. Part of it is terminology. We seem to be using different words to say the same thing, and maybe the same words to say different things. May I humbly propose the following:


Receiver, or Receiver Ship: any vessel receiving propellant from a Depot:


Depot: a vessel that accumulates propellant for transfer to a Receiver.


Tanker: a vessel that delivers propellant to LEO or higher.


Tanker/Accumulator: a vessel that combines the functions of both Depot and Tanker.


Androgynous Connector: a connector that has no identifiable gender.


Hermaphrodite connector: a connector that has elements of both genders.
Tightening up the ref's def's.

- Receiver, or Receiver Ship: any vessel receiving propellant from a Depot or a Tanker/Accumulator.
- Depot: a vessel that accumulates propellant for transfer to a Receiver.
- Tanker/Accumulator: a vessel that combines the functions of both Depot and Tanker. AKA Accumulation Tanker.
- Androgynous Connector: a connector that has no identifiable gender. Is capable of connecting to another connector of the same type.
- Hermaphrodite connector: a connector that has elements of both genders. Is capable of connecting to a connector of the same type.
- Mirror QD Plate: A QD Plate with connection genders and locations such that it is capable of connecting to a standard ships QD Plate.

There are probably a few more that will reduce confusion. The specific name is less important than the names be agreed on. These definitions are offered in the spirit of not somehow defining things into a preordained solution. Hopefully successfully.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/19/2022 06:08 pm
There seems to be an assumption that an androgynous connector will have a significant mass penalty. Where does this come from? It may be correct but I've seen no evidence.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/19/2022 06:38 pm
There seems to be an assumption that an androgynous connector will have a significant mass penalty. Where does this come from? It may be correct but I've seen no evidence.

I think the connector itself would incur a minor mass penalty. The mass is in the full connection system, not the actual connector. the connection system is like a small QD arm. If the ships are peers (as implied by an androgynous connector) then both ships have such an arm. If the ships are asymmetric, then only one ship has this extra mass.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/19/2022 07:03 pm
I just don't see much benefit for androgynous connections. It adds mass to every vehicle, without any practical real-world capability. And you need to pay to carry that extra mass to orbit with extra propellant, for every single launch over the life of that Starship [1].

Two things:

1) If you don't need to refuel, don't install the hardware.  Maintaining a fleet of "optimized for LEO or GTO" Starships doesn't sound very difficult.

2) There are likely to be some sump losses with every prop transfer.  If you introduce an intermediary in all cases, those sump losses will always be larger than if you did a peer-to-peer transfer.  I'd be pretty surprised if the mass of the androgynous (or hermaphroditic) hardware was larger than the sump losses.
If the depot has a mirror QD plate the Starships need no additional hardware.


Each tanker filling a depot or a receiver will have sump losses no matter what types of connector or transfer scheme is used. If a depot is in the middle it will also have sump losses x1 and only this can be counted against the specific transfer method.


If the depot has a mirror QD plate this will count 100% as a mass penalty. BUT, it only launches once. It will also penalize any orbit changing maneuvers. Assuming the mirror QD and supporting plumbing exactly mirrors the standard plate it will mass exactly the same. It will have to add the vent connectors, it may loose the power connectors. Anything else? Any guesstimates on current mass with supporting plumbing?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/19/2022 07:42 pm
2) There are likely to be some sump losses with every prop transfer.  If you introduce an intermediary in all cases, those sump losses will always be larger than if you did a peer-to-peer transfer.  I'd be pretty surprised if the mass of the androgynous (or hermaphroditic) hardware was larger than the sump losses.
You get a sump loss for each ship not for each transfer.  Multiple transfers don't hurt.

It's for each transfer if you're trying to use all the prop in the depot.  Whether that's a big deal or not depends on how much prop is in the depot and how much is needed.  If you have a mission that needs 7.2 lift tankers and it's just as cheap to get 8 get delivered, then the extra transfer out of the depot isn't a big deal.  But if it needs 7.95 lift tankers' worth, then the transfer out of the depot could force a ninth launch.

I doubt this is a problem for HLS, because the 9 tankers required to do LEO-NRHO-LS-NRHO is so much prop that the depot will be a net positive to curtail boil-off.  But this might not be the case for an LEO-NRHO-LEO transit mission (5 tankers) or for a direct insertion and return to GEO (slightly less than 3 tankers).

And you still don't want to use a depot in NRHO if at all possible, because it can't be returned cheaply.  It's far better to use a lift tanker, even if the prop needs to be stockpiled in a depot before being transferred to the last lift tanker.  That's the only way to recover the tanker.

The propellant needed for settling thrust during each transfer is far larger than any reasonable estimate for the extra transfer hardware.

Let's figure this out.  Suppose you have 1500t of prop in a depot that's 105t dry, and a 95t LSS with 30t of crew module and cargo.  Ullage gas thrusters will get... Isp= 70s = 686m/s?  And we'll need 0.01m/s² of settling acceleration for let's say 30 min (1800s).  That's 18m/s of delta-v.

We can use the aggregate mass of the two Starships (1605t + 125t = 1730t) as the inert mass for the rocket equation.  Then propRequired = (exp(18/686)-1)*1730t = 46t.

Yup, the hardware is going to be insignificant. 

Note, however, that this is going to be substantially more mass than there will be for warm ullage gas @ 6bar.  That implies that this new scheme of using only ullage pressure for RCS won't work for settling.  Maybe methox thrusters on the depot will still be required?  That would be something that would be a lot harder to replicate on all ships in the fleet than an androgynous/hermaphroditic prop connector.
If the only purpose for methalox thrusters is ullage settling for transfer, they need only be on the depot. Things change so fast I've lost track if methalox is still planned on SS upper stage. From what you say above, it appears not.


If methalox thrusters are on the depot only, it would be off axis thrust with a constantly changing CoM. Variable vector RCS? An RCS array? The ship opposite the depot using cold gas to adjust the resultant thrust vector?


It would also be a semi complex hardware addition on a limited number of vessels. There's no solution for this other than not doing it. If cold gas isn't adequate what other options are there?


Side thought. The thrust needed for timely settling is more than that needed to maintain settling. Any advantage to be found in this?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/19/2022 09:07 pm
There seems to be an assumption that an androgynous connector will have a significant mass penalty. Where does this come from? It may be correct but I've seen no evidence.

I think the connector itself would incur a minor mass penalty. The mass is in the full connection system, not the actual connector. the connection system is like a small QD arm. If the ships are peers (as implied by an androgynous connector) then both ships have such an arm. If the ships are asymmetric, then only one ship has this extra mass.
If a depot is always between the tanker and the receiver, only it needs the extension arm(s). If there are some circumstances where a tanker needs to interface directly to a receiver, one or the other will need the arms.

Since so far this looks to be a small percentage of the envisioned transfer scenarios. For this, special tanker builds. The Tanker/Accumulator, or maybe a Depot Light or Tanker Heavy.

All in all, the idea of androgynous connectors is appealing despite this minor kink they throw into ops. Now, if we only had them.

This issue with extension arms highlights how special the depot build is going to be.
- it will will most probably be stretched.
- no fins or heat shield.
- It may have unique RCS
- it may have CMB's or reaction wheels for fine attitude control.
- May have a second mirror type QD plate with extension mechanisim.
- PV's, battery's, and associated, appropriate for the mission.
This configuration is about what's needed for the basic depot design mission. Specific details can and will change but big picture, this is what it will be. The only mission that this design works against, as best I can tell, is tanker direct to Receiver.

I've also convinced myself at a very gut level that the propellant depot is the key to everything SX wants to do. I expect SX will give this a sharper focus than even EDL. The depot May end up as the iconic image the next 20 years of space flight.


Edit: changed a word
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/19/2022 10:15 pm
The best depot is no depot.  I'd expect development to go something like this:

1) Development of a minimum viable docking and fueling connector.  Of these two functions, I suspect that docking will be much more challenging than the fueling connector.  Androgynous, hermaphroditic--makes no difference.  But like-to-like docking and fueling just isn't that hard.  I continue to think that overloading the function of the QD is full-blown bat-guano crazy.

2) Early Starship-to-Starship refueling experiments, with the docking/fueling connector installed on Starlink missions.  Starship #1 stays in orbit after deployment.  Starship #2 deploys its Starlinks, then practices RPOD and refueling with a few tens of tonnes of prop.  After refueling, both Starship attempt EDL.  Perfecting EDL is more important than refueling.  If EDL is going well, Starship #1 can stay in orbit and act as an experimental accumulator, gathering boil-off data.

3) Build basic tanker prop tankage configuration.  I'd expect the prop domes and bulkheads to be moved forward so that it can hold 1500-1700t of prop, eating into the payload bay.  It probably can't launch full, but it can accumulate that much prop on-orbit.

4) Now's it's time to evaluate expected boil-off and launch cadence in advance of Option A test flight #1.  If the combination of boil-off and cadence results in acceptable loss of propellant during accumulation, they're done.

5) If boil-off and cadence can't get the job done, it's time for a first attempt at a depot:  A plain ol' tanker with a cryocooler.  If that manages the boil-off, they're done.

(A question:  anybody have any guesses for how much power the Starship solar arrays will generate?  A cryocooler-only solution will only work if the solar arrays can generate about double the absorbed heat flux.  However, you can throw tonnes of batteries at the problem if boil-off during eclipse is an issue.)

6) If the cryocooler-only version doesn't work, only then will they proceed with an MLI-covered ship, which presumably precludes EDL.

(Another question:  What happens if they paint the dorsal surface of the Starship solar white?  Is that enough to reduce the absorbed flux to a manageable level?  Even if the paint cooked on EDL, it might still be cheaper than committing to the development of a more exotic system.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 05/19/2022 11:17 pm
MLI covering size and positioning/orientation might be troublesome if you are expecting side-to-side docking. The easiest MLI solution is a generic inflatable mylar sheath (AKA the the giant silver condom) that can be fitted over the upper structure, but it obviously can not extend all the way to the end near the thrust structure as that is where all your various PV/radiators/receptacles/RCS/dangly bits are likely hanging out. Working out keep out zones may be troublesome as well.

Though covering the outside of the sheath with flexible PV (similar to some current solar sail work), could cover power and insulation at once. You could even use an outer flexible layer near the edges, hosting ammonia channels to run a radiator, which doubles as an inflation force for the struts to deploy such a structure.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/19/2022 11:44 pm
The best depot is no depot.  I'd expect development to go something like this:

1) Development of a minimum viable docking and fueling connector.  Of these two functions, I suspect that docking will be much more challenging than the fueling connector.  Androgynous, hermaphroditic--makes no difference.  But like-to-like docking and fueling just isn't that hard.  I continue to think that overloading the function of the QD is full-blown bat-guano crazy.

2) Early Starship-to-Starship refueling experiments, with the docking/fueling connector installed on Starlink missions.  Starship #1 stays in orbit after deployment.  Starship #2 deploys its Starlinks, then practices RPOD and refueling with a few tens of tonnes of prop.  After refueling, both Starship attempt EDL.  Perfecting EDL is more important than refueling.  If EDL is going well, Starship #1 can stay in orbit and act as an experimental accumulator, gathering boil-off data.

3) Build basic tanker prop tankage configuration.  I'd expect the prop domes and bulkheads to be moved forward so that it can hold 1500-1700t of prop, eating into the payload bay.  It probably can't launch full, but it can accumulate that much prop on-orbit.

4) Now's it's time to evaluate expected boil-off and launch cadence in advance of Option A test flight #1.  If the combination of boil-off and cadence results in acceptable loss of propellant during accumulation, they're done.

5) If boil-off and cadence can't get the job done, it's time for a first attempt at a depot:  A plain ol' tanker with a cryocooler.  If that manages the boil-off, they're done.

(A question:  anybody have any guesses for how much power the Starship solar arrays will generate?  A cryocooler-only solution will only work if the solar arrays can generate about double the absorbed heat flux.  However, you can throw tonnes of batteries at the problem if boil-off during eclipse is an issue.)

6) If the cryocooler-only version doesn't work, only then will they proceed with an MLI-covered ship, which presumably precludes EDL.

(Another question:  What happens if they paint the dorsal surface of the Starship solar white?  Is that enough to reduce the absorbed flux to a manageable level?  Even if the paint cooked on EDL, it might still be cheaper than committing to the development of a more exotic system.)
Methinks ye be both right and wrong. The development path is fine. But the 'no part is the best part' is, I think, misdirected. What you see as "overloading the function of the QD" is where I see a possible place for a twofer, and no part...yada yada.

Doing a mirror QD, or a subset of it is, from the development point of view, the path of least resistance in doing a minimal design. It must either live it's life extended, or it must extend. It must hard mate at some point. This is what's needed for that first hack. The structural hard docking can be elsewhere, but at some point this will get a hard look and if a twofer can be worked in...

If there is an androgynous cryo connector it would make refueling hookups much simpler. SX probably has, had or will have, somebody looking at this. If they come up with it there is a good possibility they can make a Direct QD to QD connection. I've enough experience with industrial plumbing to have very conservative attitude. Rockets are easy. Plumbing is hard. Unfortunately, rockets are mostly plumbing.

Bat guano crazy or not, all the incremental changes I think will be needed to make a tanker work as a stand in for a depot will result in a ... depot. Once the basic kinks are worked out each refueling campaign will be a one of a kind operation. The goal is to make it as routine as launching an F9 and retrieving the booster.


If there are lanes that are used frequently, IMO the economics make a complex and expensive depot worthwhile. One off launches may work out with ad hoc Tanker/Accumulators but circumstances will dictate.

As the infrastructure builds out and launch experience grows it may be found most economical to spot depots at various orbits of convenience and route missions to take non optimal advantage of the service at the cost of delivered mass. The economics will drive where the engineering goes
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 05/20/2022 12:28 am
I expect SpaceX to do kind of the minimum necessary to make it work well. Maybe eventually they'll use an androngynous connection, but I kind of expect they'll do a more straightforward connection like what they use for the QD.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/20/2022 03:01 am
Doing a mirror QD, or a subset of it is, from the development point of view, the path of least resistance in doing a minimal design. It must either live it's life extended, or it must extend. It must hard mate at some point. This is what's needed for that first hack. The structural hard docking can be elsewhere, but at some point this will get a hard look and if a twofer can be worked in...

A QD is not a docking system.  The docking is much, much harder than the plumbing.  A mirror QD won't do it, because whatever it is will need bellows plumbing or some other kind of extendable system to make contact.  Two Starship hulls are not going to be grinding against each other.

It would be interesting to figure out how much separation they're planning.  I'd guess that it has to be at least a couple of meters to let them null out bending torques at docking before the two vehicles collide.  Again, that problem is made considerably worse if you dock at the tails, where the QD is located.

There is a possibility that they're looking at a super-soft dock, i.e., that the plumbing snakes out to connect with some kind of fueling plate, with the control system being accurate enough to keep any stresses off of them.  That would indeed be a non-androgynous system that would always require a depot to intermediate.  But I have no idea what kind of booms could be used to make cryo-safe connections in that scenario.  That's why I think you're much more likely to see a variant of an existing soft-dock system that nulls out and aligns the two vehicles, with short plumbing booms that are extensible but otherwise rigid.  Then it's a two phase process:  first dock, then connect the plumbing.

Quote
If there are lanes that are used frequently, IMO the economics make a complex and expensive depot worthwhile. One off launches may work out with ad hoc Tanker/Accumulators but circumstances will dictate.

As the infrastructure builds out and launch experience grows it may be found most economical to spot depots at various orbits of convenience and route missions to take non optimal advantage of the service at the cost of delivered mass. The economics will drive where the engineering goes

That all sounds reasonable.  It's especially reasonable if the convolution of cadence and boil-off management isn't good enough for plain ol' tankers to act as depots.  But there simply aren't going to be any cases that need depots to hold non-integral loads of prop between missions until you have BEO cadences that are much higher than anything envisioned by Artemis.  So if there's a real depot, it's a last resort.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/20/2022 07:20 am
The docking is much, much harder than the plumbing.  A mirror QD won't do it, because whatever it is will need bellows plumbing or some other kind of extendable system to make contact.  Two Starship hulls are not going to be grinding against each other.
Proximity operations are hard. The actual contactor is not as hard as often made out, and the majority of docking systems in active use are rather simple:
+ a fixed structure on one vehicle, a structure on another vehicle that is sock-absorbed but not actively controlled
+ a set of alignment surfaces (either a cone, or a set of conic sections)
+ a set of spring-loaded latches for soft-dock
+ actuators and powered latches for hard-dock
This applies to probe & drogue, APAS, IDS, CDM, the MEVs bell-grappler, etc. Berthing systems can skip the alignment structures and soft-dock systems, but move that hardware to a separate capture mechanism for no net gain (just a shuffling of hardware locations).

The current QD system - composed of the Ship QD plate and the GSE QD assembly - have:
+ a fixed structure (on the ship) and a moving structure (on the GSE)
+ alignment features (alignment cones) and shock absorbers
- no apparent soft-latching mechanism (may be present but not visible inside the sphere/cone mating assembly e.g. as a annular detent, or may not be present at all)
+ actuators for powered latches and GSE QD plate positioning

Remember that the QD connector is under not insignificant loads during propellant transfer, the same pressure used to force fluids up several tens of metres of fluid head are also acting to push the QD plates apart again due to the cryogenic conic mating surfaces (the couplers themselves are not latched, the QD plate provides the latching).
Adding the entire GSE arm to the depot is likely not viable simply due to packaging issues, but it provides the majority of the specialised hardware needed.

Adding arms to reach out to the ship lift points may or may not be a useful addition since those lift points are already there, but beefing up the QD plate and using acceptably low settling thrust may prove sufficient. Unlike the ISS, contact force tolerance is likely a lot higher due to Starship being monolithic and already designed for high g tolerance (ascend and EDL) rather than a series of connected modules only intended to stay mated in microgravity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/20/2022 04:52 pm
Doing a mirror QD, or a subset of it is, from the development point of view, the path of least resistance in doing a minimal design. It must either live it's life extended, or it must extend. It must hard mate at some point. This is what's needed for that first hack. The structural hard docking can be elsewhere, but at some point this will get a hard look and if a twofer can be worked in...

A QD is not a docking system.  The docking is much, much harder than the plumbing.  A mirror QD won't do it, because whatever it is will need bellows plumbing or some other kind of extendable system to make contact.  Two Starship hulls are not going to be grinding against each other.

It would be interesting to figure out how much separation they're planning.  I'd guess that it has to be at least a couple of meters to let them null out bending torques at docking before the two vehicles collide.  Again, that problem is made considerably worse if you dock at the tails, where the QD is located.

There is a possibility that they're looking at a super-soft dock, i.e., that the plumbing snakes out to connect with some kind of fueling plate, with the control system being accurate enough to keep any stresses off of them.  That would indeed be a non-androgynous system that would always require a depot to intermediate.  But I have no idea what kind of booms could be used to make cryo-safe connections in that scenario.  That's why I think you're much more likely to see a variant of an existing soft-dock system that nulls out and aligns the two vehicles, with short plumbing booms that are extensible but otherwise rigid.  Then it's a two phase process:  first dock, then connect the plumbing.

Quote
If there are lanes that are used frequently, IMO the economics make a complex and expensive depot worthwhile. One off launches may work out with ad hoc Tanker/Accumulators but circumstances will dictate.

As the infrastructure builds out and launch experience grows it may be found most economical to spot depots at various orbits of convenience and route missions to take non optimal advantage of the service at the cost of delivered mass. The economics will drive where the engineering goes

That all sounds reasonable.  It's especially reasonable if the convolution of cadence and boil-off management isn't good enough for plain ol' tankers to act as depots.  But there simply aren't going to be any cases that need depots to hold non-integral loads of prop between missions until you have BEO cadences that are much higher than anything envisioned by Artemis.  So if there's a real depot, it's a last resort.

Quote
A QD is not a docking system.  The docking is much, much harder than the plumbing.  A mirror QD won't do it, because whatever it is will need bellows plumbing or some other kind of extendable system to make contact.  Two Starship hulls are not going to be grinding against each other.
That's why I said it would live extended or be able to extend, and yes, if extendable the lines must deploy too. A pantograph with some slop could extend the plate and have enough slop to allow the plate to translate and rotate a bit on the X and Y axis. This allows QD mating even if the two ships are not dead parallel.

Terminology becomes a bit of a problem here. The plates will have to "hard dock" to work but with some play behind the depot plate this does not mean the ships are hard docked. That has to be done elsewhere. Logic-ing through this and your arguments convince me of this. If nothing else, thermal issues seem to make it advisable to have some 'float' in the transfer plumbing.

All our experience with space docking is small to small or small to big. Props transfer will be big to big and calls for different techniques, or more correctly, differently scaled techniques. When the Gerald Ford docks it has lines fore and aft. We should expect no less in space.

A no return depot has no fins but the basic design has reinforcement for them. Up top there are the chopstick hard points begging for use on the tanker or receiver. On the depot there is the space where the fins and fin mechanism would be. Down low there is nothing specific to grab on the tanker or receiver but the area is reinforcement rich. The depot has the fin challenged space for a mechanism.

An idea for a docking mechanism I've been noodling might be called Canada arm light. Four pneumatically extendable arms with cable retraction. Two high, two low. Motors on the base for aiming and a minimal end effector for the grab. All four are free to extend except for the tension applied by the retraction cable. They are also free to compress and act as actively damped shock absorbers.

If the residual torque is low enough these arms would do that last tiny bit of torque transfer. The propellant settling acceleration load would most likely determine how strong they need be. A design like this blurs the distinction between hard and soft dock. It starts out soft, then between pneumatic pressure, cable tension and motor brakes, becomes hard. It also allows initial docking to stand off 2-3 meters then be drawn in to an arbitrarily close position with enough positioning control to ensure QD alignment. No, not quite belly to belly but close.

This has possibilities for a twofer. PV and sun shade both need some deployment love.

Quote
But there simply aren't going to be any cases that need depots to hold non-integral loads of prop between missions until you have BEO cadences that are much higher than anything envisioned by Artemis.

Agreed. But then, to paraphrase Freewheelin Franklin, "Resources in time of no need is better than need in time of no resources." A small but non-zero chance it would be of any use. At minimum, it's there, it has everything needed to manage boiloff, and for my own brother I couldn't make a better deal for 50 tons of methalox...  8)

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 05/20/2022 05:23 pm
Revisiting my earlier "adaptor" idea...

Consider a free-flier, with small methane tanks, TWO docking arms, and TWO mirror QDs, about the size of a Starlink v2.

All starships have "female" connectors, but anything the free-flier docks to becomes effectively male. A permamant depot would have the free-flier's systems built in, but any mission could bring along this... ahem, strap-on solution, if it needs non-depot fueling for it's mission.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 05/20/2022 05:37 pm
Revisiting my earlier "adaptor" idea...

Consider a free-flier, with small methane tanks, TWO docking arms, and TWO mirror QDs, about the size of a Starlink v2.

All starships have "female" connectors, but anything the free-flier docks to becomes effectively male. A permamant depot would have the free-flier's systems built in, but any mission could bring along this... ahem, strap-on solution, if it needs non-depot fueling for it's mission.
--You need a place to stow your device.
 --You still need to connect the two ships at the bow in addition to the device's stern connection
 --I think it needs to be a bit bigger than that to supply both a sufficient standoff space and the needed strength and rigidity while still providing the conformability in six dimensions need for docking.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 05/20/2022 06:12 pm

All our experience with space docking is small to small or small to big. Props transfer will be big to big and calls for different techniques, or more correctly, differently scaled techniques. When the Gerald Ford docks it has lines fore and aft. We should expect no less in space.


Not true at all. Shuttle docking to ISS was big to big. ISS from between like 20 tons and 400 tons (and everything in between) and Shuttle about 100 tons. Just used the APAS adapter, which is what the docking adapter Dragon uses was derived from, LIDS, which SpaceX has extensive experience with.

Same with Shuttle and Mir, of roughly comparable “big” mass, as well.

Very comparable in terms of mass and size.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/20/2022 06:39 pm
Random thought:  Is there any way to use the trailing edges of the elonerons to grab hard points on the body of the other Starship to stabilize a single docking interface?  You'd obviously have to have one ship offset forward of the other, which definitely would the system not able to do any-to-any refueling, but the elonerons would be very tolerant of alignment and rate errors.

The obvious problem is that you can't exceed the structural tolerances of the tiles, which probably makes this a dumb idea.  And I'm not sure that the elonerons fold up far enough to be helpful.  Still, something to (briefly) think about.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/20/2022 07:00 pm
The docking is much, much harder than the plumbing.  A mirror QD won't do it, because whatever it is will need bellows plumbing or some other kind of extendable system to make contact.  Two Starship hulls are not going to be grinding against each other.
Proximity operations are hard. The actual contactor is not as hard as often made out, and the majority of docking systems in active use are rather simple:
+ a fixed structure on one vehicle, a structure on another vehicle that is sock-absorbed but not actively controlled

IDSS is actively controlled.  The active side lunges to make initial contact, and then the struts have actuators and a whole bunch of processing to damp out rate error.  This all happens before hard-dock retraction and latching.

Quote
+ a set of alignment surfaces (either a cone, or a set of conic sections)
+ a set of spring-loaded latches for soft-dock
+ actuators and powered latches for hard-dock
This applies to probe & drogue, APAS, IDS, CDM, the MEVs bell-grappler, etc. Berthing systems can skip the alignment structures and soft-dock systems, but move that hardware to a separate capture mechanism for no net gain (just a shuffling of hardware locations).

I think we can safely rule out berthing solutions.

Remember that we're dealing with bigger torques than on any previous system.  Even the shuttle, which was pretty big, didn't have hundreds to thousands of tonnes of potentially sloshing prop in it during docking.  With Starship refueling, that's the whole purpose.  Presumably you'd make the heavier of the two ships the passive side, just to simplify the proximity ops, but you're still dealing with an extra 150-200t of payload and prop for the active docker.

Quote
The current QD system - composed of the Ship QD plate and the GSE QD assembly - have:
+ a fixed structure (on the ship) and a moving structure (on the GSE)
+ alignment features (alignment cones) and shock absorbers

That's a whole different animal.  Even in high winds, the base of the thing the QD arm is approaching is fixed, and the QD arm itself acts as the stabilizer to damp out rate errors before the two QD plates attach and retract.  It's the difference between having to take out millimeters of error vs. possibly tens of centimeters on-orbit.

Quote
- no apparent soft-latching mechanism (may be present but not visible inside the sphere/cone mating assembly e.g. as a annular detent, or may not be present at all)
+ actuators for powered latches and GSE QD plate positioning

Remember that the QD connector is under not insignificant loads during propellant transfer, the same pressure used to force fluids up several tens of metres of fluid head are also acting to push the QD plates apart again due to the cryogenic conic mating surfaces (the couplers themselves are not latched, the QD plate provides the latching).

I suspect that the hard-latch for the QD would work OK.  But that's like 10% of the difficulty of the soft-latch problem.

Quote
Adding the entire GSE arm to the depot is likely not viable simply due to packaging issues, but it provides the majority of the specialised hardware needed.

Adding arms to reach out to the ship lift points may or may not be a useful addition since those lift points are already there, but beefing up the QD plate and using acceptably low settling thrust may prove sufficient. Unlike the ISS, contact force tolerance is likely a lot higher due to Starship being monolithic and already designed for high g tolerance (ascend and EDL) rather than a series of connected modules only intended to stay mated in microgravity.

See my stupid idea just up-thread.  If there's some way to give the trailing edge of the elonerons a hard point, they might be able to do double duty as a stabilization mechanism.  But I still think that a single, fairly hefty, dorsal dock/connector midships  can be engineered to handle the torques, especially if it's near the center of mass of the lighter of the two ships.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 05/20/2022 09:53 pm
Revisiting my earlier "adaptor" idea...

Consider a free-flier, with small methane tanks, TWO docking arms, and TWO mirror QDs, about the size of a Starlink v2.

All starships have "female" connectors, but anything the free-flier docks to becomes effectively male. A permamant depot would have the free-flier's systems built in, but any mission could bring along this... ahem, strap-on solution, if it needs non-depot fueling for it's mission.
--You need a place to stow your device.
 --You still need to connect the two ships at the bow in addition to the device's stern connection
 --I think it needs to be a bit bigger than that to supply both a sufficient standoff space and the needed strength and rigidity while still providing the conformability in six dimensions need for docking.
1) the free-flier is already hooked into the refueling archetecture, and compared to a full tanker is trivial to reposition to a new orbit
2+)The Starlink form factor is only for deployment (and "Rear cargo slot" form factor also works, if that still exists for non-crew starships), it can unform into arms and clamps after deployment.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 05/21/2022 01:13 pm

Let's figure this out.  Suppose you have 1500t of prop in a depot that's 105t dry, and a 95t LSS with 30t of crew module and cargo.  Ullage gas thrusters will get... Isp= 70s = 686m/s?  And we'll need 0.01m/s² of settling acceleration for let's say 30 min (1800s).  That's 18m/s of delta-v.

We can use the aggregate mass of the two Starships (1605t + 125t = 1730t) as the inert mass for the rocket equation.  Then propRequired = (exp(18/686)-1)*1730t = 46t.

Yup, the hardware is going to be insignificant. 

so I had to check the calcs above for myself...
Here is how I went about it using just F=ma
I also used more optimistic numbers like .001m/s^2 and hot gas thrusters 300s.
I think a settling burst at the beginning and very small maintenance thrust might work.

      force for settling ma = F
      1730t*.001m/s^2 = 1730N
      exhvec
      300s*force = 2941m/s
      force/exhvec = kg/s F/a = m
      1730N/(2941m/s) = .58kg/s
      kg/s * 30min
      .57kg/s*30min = 1026kg
      all together
      1730t*.001m/s^2/(300s*force)*30min = 1058kg

I also note there is no exp() in there.
Might help some people with the math.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/21/2022 06:50 pm
Random thought:  Is there any way to use the trailing edges of the elonerons to grab hard points on the body of the other Starship to stabilize a single docking interface?  You'd obviously have to have one ship offset forward of the other, which definitely would the system not able to do any-to-any refueling, but the elonerons would be very tolerant of alignment and rate errors.

The obvious problem is that you can't exceed the structural tolerances of the tiles, which probably makes this a dumb idea.  And I'm not sure that the elonerons fold up far enough to be helpful.  Still, something to (briefly) think about.
Hmmm. Probably right about the tiles being an issue but I've got to wonder about a temporary hookup using the elonerons as the motive force to draw the ships together close enough for something else to do the grab. That leaves the problem of making that first hookup to fins. (Damn it. Fins has fewer letters) probably not worth the hassle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/21/2022 07:11 pm
so I had to check the calcs above for myself...
Here is how I went about it using just F=ma
I also used more optimistic numbers like .001m/s^2 and hot gas thrusters 300s.
I think a settling burst at the beginning and very small maintenance thrust might work.

      force for settling ma = F
      1730t*.001m/s^2 = 1730N
      exhvec
      300s*force = 2941m/s
      force/exhvec = kg/s F/a = m
      1730N/(2941m/s) = .58kg/s
      kg/s * 30min
      .57kg/s*30min = 1026kg
      all together
      1730t*.001m/s^2/(300s*force)*30min = 1058kg

I also note there is no exp() in there.
Might help some people with the math.

If you use 1mm/s² acceleration and 300s Isp, my calculation jibes with yours.

It's a good point to start with a fairly high acceleration to get things settled, then ramp down.  Whether you can get down to an average of 1mm/s² or not, I have no idea.  Assuming that the transfer is pump-driven and not pressure-driven, briefly uncovering the source tank's inlets is probably OK, as long as your flow rate is fairly modest.  That might in turn require a longer transfer time, which would increase the total ullage "burn" Δv, but let's stick with the 30min transfer time for now, remembering that it's a SWAG.

Isp=300s seems awfully aggressive, especially since they're now talking about using warm ullage gas for thrusters, not methox.  For cold-gas (roughly 30ºC) methane alone, I get Isp=115s.

Let's stick with your a=1mm/s².  That'll reduce Δv to 1.8m/s, which then requires 2.8t of prop for the ullage acceleration over 30mins at Isp=115s. 

If you go to hot-gas methox, I can believe you'll get Isp=320s.  That would reduce the reaction mass to 1.0t.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/21/2022 07:23 pm
Hmmm. Probably right about the tiles being an issue but I've got to wonder about a temporary hookup using the elonerons as the motive force to draw the ships together close enough for something else to do the grab. That leaves the problem of making that first hookup to fins. (Damn it. Fins has fewer letters) probably not worth the hassle.

After fooling with this a bit, I'm inclined to think that this was a stupid idea.  You'd probably have to have doors in the eloneron tips to make this work, which would be very complex.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/21/2022 08:40 pm
so I had to check the calcs above for myself...
Here is how I went about it using just F=ma
I also used more optimistic numbers like .001m/s^2 and hot gas thrusters 300s.
I think a settling burst at the beginning and very small maintenance thrust might work.

      force for settling ma = F
      1730t*.001m/s^2 = 1730N
      exhvec
      300s*force = 2941m/s
      force/exhvec = kg/s F/a = m
      1730N/(2941m/s) = .58kg/s
      kg/s * 30min
      .57kg/s*30min = 1026kg
      all together
      1730t*.001m/s^2/(300s*force)*30min = 1058kg

I also note there is no exp() in there.
Might help some people with the math.

If you use 1mm/s² acceleration and 300s Isp, my calculation jibes with yours.

It's a good point to start with a fairly high acceleration to get things settled, then ramp down.  Whether you can get down to an average of 1mm/s² or not, I have no idea.  Assuming that the transfer is pump-driven and not pressure-driven, briefly uncovering the source tank's inlets is probably OK, as long as your flow rate is fairly modest.  That might in turn require a longer transfer time, which would increase the total ullage "burn" Δv, but let's stick with the 30min transfer time for now, remembering that it's a SWAG.

Isp=300s seems awfully aggressive, especially since they're now talking about using warm ullage gas for thrusters, not methox.  For cold-gas (roughly 30ºC) methane alone, I get Isp=115s.

Let's stick with your a=1mm/s².  That'll reduce Δv to 1.8m/s, which then requires 2.8t of prop for the ullage acceleration over 30mins at Isp=115s. 

If you go to hot-gas methox, I can believe you'll get Isp=320s.  That would reduce the reaction mass to 1.0t.
Depending on ullage gas being warm sounds dicy. If there's a significant delay between engine firing and transfer the gas will have cooled down. Best to calculate for worst case.


Before settling on a minimum settling acceleration would it make sense to look at possible propellant disruptions that would set lower bounds? Slosh caused by the transfer itself might be one. Random pops and groans from thermal might be another.


The half hour transfer time seems aggressive taken by itself. There's probably a relationship between transfer rate and the slosh it produces that would dictate a minimum acceleration needed to overcome that slosh. Then theres the issue of minimizing expenditure of gas used for dV. I'd guess there's a sweet spot for total dV somewhere.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sevenperforce on 05/21/2022 09:38 pm
Depending on ullage gas being warm sounds dicy. If there's a significant delay between engine firing and transfer the gas will have cooled down. Best to calculate for worst case.

The half hour transfer time seems aggressive taken by itself. There's probably a relationship between transfer rate and the slosh it produces that would dictate a minimum acceleration needed to overcome that slosh. Then theres the issue of minimizing expenditure of gas used for dV. I'd guess there's a sweet spot for total dV somewhere.
If they're using ullage gas directly from the tanks, is there a risk of liquid ingestion into the ullage thruster lines? Or is that a non-issue?

They could also use electric resistance heaters to ensure that ullage gas stays warm (or even to rapidly boil off more propellant to maintain ullage). I believe SpaceX knows someone who has high-capacity batteries.

If liquid ingestion into the thruster lines is an issue, they can use an accumulator COPV to start the ullage "burn" and then open the valves between the tank and the accumulator to continue the flow. Resistance heaters, if needed, could be placed around/in the accumulators. That's a simple way to increase pressure in the accumulator too.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/21/2022 11:45 pm
Depending on ullage gas being warm sounds dicy. If there's a significant delay between engine firing and transfer the gas will have cooled down. Best to calculate for worst case.

The half hour transfer time seems aggressive taken by itself. There's probably a relationship between transfer rate and the slosh it produces that would dictate a minimum acceleration needed to overcome that slosh. Then theres the issue of minimizing expenditure of gas used for dV. I'd guess there's a sweet spot for total dV somewhere.
If they're using ullage gas directly from the tanks, is there a risk of liquid ingestion into the ullage thruster lines? Or is that a non-issue?

They could also use electric resistance heaters to ensure that ullage gas stays warm (or even to rapidly boil off more propellant to maintain ullage). I believe SpaceX knows someone who has high-capacity batteries.

If liquid ingestion into the thruster lines is an issue, they can use an accumulator COPV to start the ullage "burn" and then open the valves between the tank and the accumulator to continue the flow. Resistance heaters, if needed, could be placed around/in the accumulators. That's a simple way to increase pressure in the accumulator too.
AIUI, the autogenous tapoffs are close to or direct from the pumps. Could somebody fact check this? If correct, the COPV could be pressurized crazy high and be as big as needed up to the limits of what can be spared from the MCC. This would be logical source for reaction thrusters.


I think one of the number wizards looked at the power required for a heater scheme like this but don't remember the conclusions. If on a dedicated depot the PV and battery would have to be sized for ullage heaters. It would be up for awhile. Tankers otoh, would most probably be power constrained. Receiver ships intended for long duration missions would have robust PV and others might be mixed.


This sounds like an argument for the depot to have capability to supply power to whatever ship it is mated to. And another argument for the depot to eventually be much more than just a tanker on temporary accumulator duty.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sevenperforce on 05/22/2022 01:19 am
If they're using ullage gas directly from the tanks, is there a risk of liquid ingestion into the ullage thruster lines? Or is that a non-issue?

They could also use electric resistance heaters to ensure that ullage gas stays warm (or even to rapidly boil off more propellant to maintain ullage). I believe SpaceX knows someone who has high-capacity batteries.

If liquid ingestion into the thruster lines is an issue, they can use an accumulator COPV to start the ullage "burn" and then open the valves between the tank and the accumulator to continue the flow. Resistance heaters, if needed, could be placed around/in the accumulators. That's a simple way to increase pressure in the accumulator too.
AIUI, the autogenous tapoffs are close to or direct from the pumps. Could somebody fact check this? If correct, the COPV could be pressurized crazy high and be as big as needed up to the limits of what can be spared from the MCC. This would be logical source for reaction thrusters.
According to this well-known diagram (https://www.reddit.com/r/spacex/comments/cxkrtb/detailed_diagram_of_the_raptor_engine_er26_gimbal/), yes. There is cold gaseous methane and cold gaseous oxygen coming into the engine from an outside source to operate the torch igniters in the two preburners and the combustion chamber, and there is hot gaseous methane and hot gaseous oxygen coming out of the engine to pressurize the tanks. There is an oxygen valve tapoff along the high-pressure feed line that runs to the fuel-rich preburner; the tapoff, which is already supercritical because it has been pressurized high enough to run into the fuel-rich preburner, runs through a heat exchanger in the oxygen-rich preburner exhaust to be heated so that it will exhaust into the pressurant line as hot gas. There is a methane valve tapoff along the main flow line from the engine and nozzle cooling jacket just before it enters the fuel-rich preburner; it runs through a heat exhanger in the preburner exhaust.

This has the advantage of sending the methane through a fuel-rich heat exchanger and sending the oxygen through an oxygen-rich heat exchanger, so there was a leak in the heat exchangers nothing goes boom. It also has the advantage that both pressurant gases with similar energy characteristics since both would otherwise be pressurized to head into the fuel-rich preburner. The fuel-rich preburner operates at 512 bar so the pressurant gas can be coming out at some pretty spectacular pressures, if they want it to.

Quote
I think one of the number wizards looked at the power required for a heater scheme like this but don't remember the conclusions. If on a dedicated depot the PV and battery would have to be sized for ullage heaters. It would be up for awhile. Tankers otoh, would most probably be power constrained. Receiver ships intended for long duration missions would have robust PV and others might be mixed.
One of the problems is that for pressurant gas you want it to come out as hot as possible so you need less of it, while for a COPV accumulator you want it to be reasonably cold so that your COPV can hold more gas at higher pressure.

I suppose you could have your pressurant line go through a split valve, with one branch simply exhausting into tank ullage at high temperature and the other end flowing through a cooling heat exchanger (perhaps the downstream propellant feed line?) to flow to accumulator COPVs at high pressure and low temperature. Of course the computer would control the valves to maintain the correct ullage pressure and and shut off once the COPVs were full. Those COPVs in turn would be tapped to augment ullage and provide thrust, although they would need a resistance heater in their downstream for maximum efficiency. It's possible that the utility of a resistance heater is not high enough to outweigh the added complexity.

Of course all these considerations are only for Starship itself. Superheavy can rely entirely on directly vented tank ullage.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/22/2022 10:55 am
Part of it is terminology. May I humbly propose the following:
Tanker/Accumulator: a vessel that combines the functions of both Depot and Tanker.

Small quibble, I always use tanker/accumulator to mean as close as possible to a regular off-the-shelf tanker acting as an ad-hoc depot. Not that it "combines the functions".



Speaking of terminology.... "Docking".

A QD is not a docking system.  The docking is much, much harder than the plumbing.

Yes, "docking" is hard.

What the tanker and depot are doing is something else. I don't know if it has a common name, which is why everyone defaults to "docking". But the requirements are very different to those that lead to the complexity of IDAs/etc. "Docking" is hard. "Grabbing" is not. Docking requires making a pressure-tight connection in a circular area where both sides can be opened up into the spacecraft.

Grabbing just requires something to hold onto.

For example, when the original Cargo Dragon needed to dock with ISS, it first hung back and presented a fold-out grapple-fixture. The Canadarm reached out to grab it. That process might have been called "grappling" or similar, but the Dragon was drifting free from the ISS and the Canadarm had to approach and "hard-dock" to it.

Grapple fixtures are simple. (See below.) The end effector on the Canadarm is not complex. (See belower.) (Actually it's hideously complex, but it also needs to be able to latch the four power/data connectors).

So: A larger version of a fold-out grapple-fixture at the top and bottom of each tanker. An end-effector that deploys and extends straight out of two similar spots on the depot. Once solidly attached, you winch in the end-effectors until the fuel connectors mate. (If you want asexuality, you could offset them and have androgynous pairs. Ie, one grapple-fixture and one end-effector next to each other at each spot on each vehicle (dorsal/dorsal matches male/female & female/male in each pair.) I'd say "see belowest", but I think it exceeds my drawing ability.)

It would be interesting to figure out how much separation they're planning.  I'd guess that it has to be at least a couple of meters to let them null out bending torques at docking before the two vehicles collide.

I'd guess virtually nothing. As close to grinding hulls without actually grinding hulls. The best separation is no separation?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bob Shaw on 05/22/2022 11:38 am
So, Gemini didn’t dock with Agena? Or early Soyuz with early Soyuz?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/22/2022 01:00 pm
Docking requires making a pressure-tight connection in a circular area where both sides can be opened up into the spacecraft.

Seems like you’ve very accurately described the process of interconnecting two pressurized cryogenic plumbing systems.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/22/2022 01:02 pm
So, Gemini didn’t dock with Agena? Or early Soyuz with early Soyuz?

Well, they were practicing docking, so they probably get in on a technicality. (Interestingly, some of the hardware on Agena was called "mooring" rather than "docking".)

IMO, we either need another word for vehicles grabbing each other when there's no pressurised pass-through, or we need an extra word for for docking when there is the massive complexity of a pressurised pass-through.



Docking requires making a pressure-tight connection in a circular area where both sides can be opened up into the spacecraft.
Seems like you’ve very accurately described the process of interconnecting two pressurized cryogenic plumbing systems.

No. Very explicitly separating the grapple mechanism from the fuel connections. Combining them into one mechanism is, IMO, begging for unnecessary complexity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/22/2022 01:17 pm
Docking requires making a pressure-tight connection in a circular area where both sides can be opened up into the spacecraft.
Seems like you’ve very accurately described the process of interconnecting two pressurized cryogenic plumbing systems.

No. Very explicitly separating the grapple mechanism from the fuel connections. Combining them into one mechanism is, IMO, begging for unnecessary complexity.

Grappling is soft dock. Forming the connections is hard dock. The sequencing and topology are the same. The only difference is the diameter and number of the surfaces being hard-docked.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/22/2022 02:45 pm
Grappling is soft dock.

No, the grapple capture system has its own equivalent of hard-dock.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/22/2022 05:08 pm
Grappling is soft dock.

No, the grapple capture system has its own equivalent of hard-dock.

You can do that if you want to make it more complicated, but you're still going to have to "hard-dock" the two ends of the fill/drain pipes and gas equalization lines.  Either way, the the term "docking" is completely appropriate, because it accurately describes the process.

Personally, I think that leaving the grappling system with some degrees of freedom wouldn't be a terrible idea.  If you have a coupled system under ullage acceleration, being able to damp out certain kinds of oscillation might be necessary.

A big chunk of this depends on whether you're dealing with one docking point or two.  I've been imagining something very similar to an existing dock, which comprises:

1) A pair of soft-capture rings, one of which is active and one of which is latched.  (If you really want to rule out like-to-like refueling, you could go true active-passive and save a couple hundred kg, maybe.)  Probably doesn't need to be more than 1m wide, if that.

2) A set of hard points to latch after the soft system has stabilized and retracted.  Also <1m wide.

3) Once the hard points are latched, you extend the connectors to latch and pressurize.  You can do this either with a plate or individual connectors.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: RonM on 05/22/2022 07:57 pm
The terminology is based on seagoing vessels. Docking means a ship maneuvers itself to the wharf. Berthing means a tug maneuvers the ship to the wharf. When applied to ISS it depends if the spacecraft is grappled or not. The original cargo Dragon was grappled and berthed to the station. Dragon 2 docks itself. Note that in the case of ISS different interfaces are used to connect the vehicles.

So, if a Starship tanker is grappled by a robotic arm on the depot, the tanker has berthed. If the Starship tanker maneuvers itself to the depot it has docked.

Soft dock, hard dock, etc. are details of how the spacecraft-to-spacecraft interface works, not how the spacecraft got there.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/22/2022 10:25 pm
So, Gemini didn’t dock with Agena? Or early Soyuz with early Soyuz?

Well, they were practicing docking, so they probably get in on a technicality. (Interestingly, some of the hardware on Agena was called "mooring" rather than "docking".)

IMO, we either need another word for vehicles grabbing each other when there's no pressurised pass-through, or we need an extra word for for docking when there is the massive complexity of a pressurised pass-through.



Docking requires making a pressure-tight connection in a circular area where both sides can be opened up into the spacecraft.
Seems like you’ve very accurately described the process of interconnecting two pressurized cryogenic plumbing systems.

No. Very explicitly separating the grapple mechanism from the fuel connections. Combining them into one mechanism is, IMO, begging for unnecessary complexity.
Grabbing without passthrough: hugging?
With passthrough: mating? I think that one's already taken.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/22/2022 10:31 pm
Part of it is terminology. May I humbly propose the following:
Tanker/Accumulator: a vessel that combines the functions of both Depot and Tanker.

Small quibble, I always use tanker/accumulator to mean as close as possible to a regular off-the-shelf tanker acting as an ad-hoc depot. Not that it "combines the functions".



Speaking of terminology.... "Docking".

A QD is not a docking system.  The docking is much, much harder than the plumbing.

Yes, "docking" is hard.

What the tanker and depot are doing is something else. I don't know if it has a common name, which is why everyone defaults to "docking". But the requirements are very different to those that lead to the complexity of IDAs/etc. "Docking" is hard. "Grabbing" is not. Docking requires making a pressure-tight connection in a circular area where both sides can be opened up into the spacecraft.

Grabbing just requires something to hold onto.

For example, when the original Cargo Dragon needed to dock with ISS, it first hung back and presented a fold-out grapple-fixture. The Canadarm reached out to grab it. That process might have been called "grappling" or similar, but the Dragon was drifting free from the ISS and the Canadarm had to approach and "hard-dock" to it.

Grapple fixtures are simple. (See below.) The end effector on the Canadarm is not complex. (See belower.) (Actually it's hideously complex, but it also needs to be able to latch the four power/data connectors).

So: A larger version of a fold-out grapple-fixture at the top and bottom of each tanker. An end-effector that deploys and extends straight out of two similar spots on the depot. Once solidly attached, you winch in the end-effectors until the fuel connectors mate. (If you want asexuality, you could offset them and have androgynous pairs. Ie, one grapple-fixture and one end-effector next to each other at each spot on each vehicle (dorsal/dorsal matches male/female & female/male in each pair.) I'd say "see belowest", but I think it exceeds my drawing ability.)

It would be interesting to figure out how much separation they're planning.  I'd guess that it has to be at least a couple of meters to let them null out bending torques at docking before the two vehicles collide.

I'd guess virtually nothing. As close to grinding hulls without actually grinding hulls. The best separation is no separation?
Re: quibble. I meant it as you took it. Was staying away from specific details to avoid baking in unwarranted assumptions on what additional depot like features might be included.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 05/23/2022 01:56 am
I would basically suggest a 3 point grapple captured stance. The term captured was used by Shuttle program to indicate full control over the item grappled by the SSRMS (CANADARM).

Also note the SSRMS could manipulate a captured mass of 100t. With 3 of them a prop loaded Tanker being Captured by a Depot would be totally under control to the point the the two SS would act like a single vehicle A Tanker mass of 300t. Another note here is that these MDA derived grapple and capture systems would mass at less than 400kg each. By having no elbows and only a shoulder and wrist joint as well as a extensible tube section. The Tanker would approach the Depot where the Depot then reaches out and grabs then puts the Tanker in a Captured mode with one arm. Then using the shoulder and wrist joint motors manipulate the vehicles alignment to then grapple with the other 2 arms and bring to Captured mode. Now with all three holding the two vehicles steady the arms manipulate and draw the two vehicles closer. The rest of the actions are to do the fine alignment while drawing closer to do the QD interface "docking".

The reason I suggest 3 is that it not only gives a highly rigid but still flexible alignment of two very large and massive bodies in a zero G environment. The other item is that with such also having only 1 vehicle doing the low thrust prop settling is possible with this setup. Another is that only 2 of the 3 is absolutely needed giving a small level of redundancy. But after a Depot gets to this point it would probably time to retire it or send up a cargo SS that has a CANADARM2 that can then perform a R&R of the failed RMS on the Depot. That goes to another item and that is having standard RMS grapple sites on SS that use the CANADARM2 standards give a lot of flexibility in the future including out at the Gateway. The Standards include a hard-line comm and power hookup. Then there is a final item and that these designs are already space qualified and significantly long life with low failure rates. I am sure that SpaceX could probably get good cost per unit on orders of 60 to 120 once BEO missions to the Moon and Mars ramps up. Since grapple targets are extremely simple. SpaceX would like make their own eventually in the 1,000s.


And now here is one of those added benifits of having standard CANADARM2 grapple targets on SS. By sending up a CANADARM2 type of double end-effector arm. The arm reaches out of the cargo bay opening and then grapples and captures the target on the SS. It then releases it's capture of it's internal to the SS grapple point. Now it is fully outside of the vehicle and ready for any on orbit heavy duty or even fine detail work needed in orbit anywhere. When finished it crawls back inside for it's trip back to Earth. SpaceX does not need to design it. It already exists.


ADDED: A small item here is that in 1984-85 I work on the SSRMS avionics engineering from the government side. If anyone is interested in some of the interesting stories about things I learned from the developemnt guys of the CANADARM send me a message.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/23/2022 04:12 am
Either way, the the term "docking" is completely appropriate, because it accurately describes the process.
[...]
1) A pair of soft-capture rings

This is why I think we need a different term. The moment you say "docking", something about it immediately makes you restrict your design-space to designs similar to existing human docking systems.

If you stop thinking of it as "docking" but "capturing", "grabbing", "mating", "connecting", etc, then you start to widen the possible design-space.

Note all the hand-wringing previously in the thread over how to attach "docking" systems to the outside of the vehicle if you want it to be launchable but to allow any Starship-to-Starship fuel transfer. Grapple fixtures can just* fold out, low mass, minimal internal volume to allow it to fold away for launch/EDL. An end-effector can push straight out from behind a small hatch, heavier (but not as much as a full docking system), and deeper volume into the ship, but easily* tucked away during launch (and even EDL). It doesn't need a robot arm. Neither side has to carry cryo liquids, so there's no issue with trying to make flexible pipes. After the vehicles are stabilised, mating the cryo-plates is just* a matter of bringing them together, it is not another "docking", it's just* plugging them in.


* just, easily, just...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/23/2022 04:21 am
I would basically suggest a 3 point grapple captured stance.

I avoided 3 because I figured you are lining up two curved surfaces, the points of contact are in a single vertical line. So to have three grapple fixtures in a triangle, two of them have to be "around the corner", making the problem harder. And the closer you bring those two together, the less effective the "triangle" is.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/23/2022 04:29 am
Grabbing without passthrough: hugging?

I'd save "hugging" for paired arms like the "chopsticks".
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sevenperforce on 05/23/2022 02:00 pm
So, Gemini didn’t dock with Agena? Or early Soyuz with early Soyuz?

Well, they were practicing docking, so they probably get in on a technicality. (Interestingly, some of the hardware on Agena was called "mooring" rather than "docking".)

IMO, we either need another word for vehicles grabbing each other when there's no pressurised pass-through, or we need an extra word for for docking when there is the massive complexity of a pressurised pass-through.
I think four words are sufficient: grappling, berthing, docking, and mating.

Grappling is a primary vehicle grabbing a secondary vehicle. Berthing is when a primary vehicle grapples a secondary vehicle and establishes a hard capture. Docking is when two independent vehicles establish a hard capture.

Mating is when two berthed/docked vehicles exchange fluids.

So Gemini-Agena and LK-LOK docked but didn't mate.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/26/2022 05:22 am
Either way, the the term "docking" is completely appropriate, because it accurately describes the process.
[...]
1) A pair of soft-capture rings

This is why I think we need a different term. The moment you say "docking", something about it immediately makes you restrict your design-space to designs similar to existing human docking systems.

If you stop thinking of it as "docking" but "capturing", "grabbing", "mating", "connecting", etc, then you start to widen the possible design-space.

Note all the hand-wringing previously in the thread over how to attach "docking" systems to the outside of the vehicle if you want it to be launchable but to allow any Starship-to-Starship fuel transfer. Grapple fixtures can just* fold out, low mass, minimal internal volume to allow it to fold away for launch/EDL. An end-effector can push straight out from behind a small hatch, heavier (but not as much as a full docking system), and deeper volume into the ship, but easily* tucked away during launch (and even EDL). It doesn't need a robot arm. Neither side has to carry cryo liquids, so there's no issue with trying to make flexible pipes. After the vehicles are stabilised, mating the cryo-plates is just* a matter of bringing them together, it is not another "docking", it's just* plugging them in.


* just, easily, just...

The key difference between "soft docking" and "grappling" is that soft docking is designed to work when two vehicles maneuver into contact with one another, while grappling implies that one vehicle actively reaches out and grabs the other.  Note that docking doesn't require the establishment of an 80cm transfer tunnel; it merely requires that one vehicle latches onto another using guidance instead of remote manipulation.  You could implement a soft dock that was only 20-30cm across.

I agree that either will work.

But you still have to eliminate the alignment and rate errors, retract one spacecraft to hard-dock with the other, and ensure that the resulting more-or-less rigid system will survive whatever weird ullage acceleration dynamics are imposed on them.  Only then will it be safe to extend the fill/drain/vent plumbing and begin transfer.

In making the decision whether grappling or soft-dock is preferable, here are some questions that need to be answered:

1) Can you use a single grapple to null out rate and alignment errors between two Starships loaded with prop?  Can you use a single soft-dock?  (I suspect that a single soft-dock has a better chance of doing this.)

2) Which method is better suited to autonomous control?  (I don't think that autonomous grappling is a thing yet, is it?)

3) Can any conceivable hard dock scheme rely on a single region of attachment, or do you need at least two to ensure adequate rigidity?  If you need two, does that imply two grapples or soft-docks?  If you need two, which method is least likely to introduce errors when the first one makes contact before the second one?  (I suspect that grappling will win on this last point--if two are needed.)

4) What are the door/hatch profiles for grapples vs. soft docks?  (I'm not sure which one wins here.  If you're talking about an IDS-like soft dock, the grapple will win handily.  But that would be a silly design for a soft dock.  On the other hand, I'm sure that there's some minimum diameter required to handle the rate torques with a soft dock--but that's also true of grapples.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Rossco on 05/26/2022 07:33 am
wouldn't a version of the universal docking adapter work?
its already been designed, the software has been written and the hardware exists
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 05/26/2022 03:57 pm
Has everybody given up on the lightest in terms of hardware of precision parallel flying? Both vehicles need different ullage thrust.

Fluid moving from vehicle to vehicle would impart a thrust? Which would have to be countered?
Not sure if this true?
Fluid is moving at constant velocity within the hose connecting them. You only feel thrust from a hose where it exits the hose.
Since the net change is moving 1200t from one starship to the other it would seem that is the impulse that needs to be countered.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/26/2022 05:03 pm
Either way, the the term "docking" is completely appropriate, because it accurately describes the process.
[...]
1) A pair of soft-capture rings

This is why I think we need a different term. The moment you say "docking", something about it immediately makes you restrict your design-space to designs similar to existing human docking systems.

If you stop thinking of it as "docking" but "capturing", "grabbing", "mating", "connecting", etc, then you start to widen the possible design-space.

Note all the hand-wringing previously in the thread over how to attach "docking" systems to the outside of the vehicle if you want it to be launchable but to allow any Starship-to-Starship fuel transfer. Grapple fixtures can just* fold out, low mass, minimal internal volume to allow it to fold away for launch/EDL. An end-effector can push straight out from behind a small hatch, heavier (but not as much as a full docking system), and deeper volume into the ship, but easily* tucked away during launch (and even EDL). It doesn't need a robot arm. Neither side has to carry cryo liquids, so there's no issue with trying to make flexible pipes. After the vehicles are stabilised, mating the cryo-plates is just* a matter of bringing them together, it is not another "docking", it's just* plugging them in.


* just, easily, just...

The key difference between "soft docking" and "grappling" is that soft docking is designed to work when two vehicles maneuver into contact with one another, while grappling implies that one vehicle actively reaches out and grabs the other.  Note that docking doesn't require the establishment of an 80cm transfer tunnel; it merely requires that one vehicle latches onto another using guidance instead of remote manipulation.  You could implement a soft dock that was only 20-30cm across.

I agree that either will work.

But you still have to eliminate the alignment and rate errors, retract one spacecraft to hard-dock with the other, and ensure that the resulting more-or-less rigid system will survive whatever weird ullage acceleration dynamics are imposed on them.  Only then will it be safe to extend the fill/drain/vent plumbing and begin transfer.

In making the decision whether grappling or soft-dock is preferable, here are some questions that need to be answered:

1) Can you use a single grapple to null out rate and alignment errors between two Starships loaded with prop?  Can you use a single soft-dock?  (I suspect that a single soft-dock has a better chance of doing this.)

2) Which method is better suited to autonomous control?  (I don't think that autonomous grappling is a thing yet, is it?)

3) Can any conceivable hard dock scheme rely on a single region of attachment, or do you need at least two to ensure adequate rigidity?  If you need two, does that imply two grapples or soft-docks?  If you need two, which method is least likely to introduce errors when the first one makes contact before the second one?  (I suspect that grappling will win on this last point--if two are needed.)

4) What are the door/hatch profiles for grapples vs. soft docks?  (I'm not sure which one wins here.  If you're talking about an IDS-like soft dock, the grapple will win handily.  But that would be a silly design for a soft dock.  On the other hand, I'm sure that there's some minimum diameter required to handle the rate torques with a soft dock--but that's also true of grapples.)
Rad Mod. I think you're falling into a terminology rabbit hole. It isn't a matter of doing X to get something like a hard/soft grapple/hug/berthing/docking. It's doing whatever is necessary to get propellant from one ship to the other. Sometimes words (secondary symbols) get in the way of understanding (primary symbols). Let's understand the problem then figure out what to call it.


Peace bro.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/26/2022 05:58 pm
Has everybody given up on the lightest in terms of hardware of precision parallel flying? Both vehicles need different ullage thrust.

Fluid moving from vehicle to vehicle would impart a thrust? Which would have to be countered?
Not sure if this true?
Fluid is moving at constant velocity within the hose connecting them. You only feel thrust from a hose where it exits the hose.
Since the net change is moving 1200t from one starship to the other it would seem that is the impulse that needs to be countered.

I don't think the guidance and control issues are the problem here.  But, absent a rigid connection between the two spacecraft, you need a flexible set of cryogenic transfer lines, and a way of extending those lines from one ship to the other and connecting them securely.  The low temperature of the fluids is going to put some extreme materials constraints on flexible lines.  I'm inclined to believe that the challenges of a rigid connection are a lot easier to overcome.

Rad Mod. I think you're falling into a terminology rabbit hole. It isn't a matter of doing X to get something like a hard/soft grapple/hug/berthing/docking. It's doing whatever is necessary to get propellant from one ship to the other. Sometimes words (secondary symbols) get in the way of understanding (primary symbols). Let's understand the problem then figure out what to call it.

Climbing out of the terminology hole was what I was trying to do.  Apparently I failed.

It seems to me that the irreducible minimum is to establish a rigid connection between the two vehicles (assuming that flexible transfer lines are really hard), then extend and connect rigid transfer lines.

We know how to establish rigid connections:  all "docking", "grappling", and "berthing" technology--and its associated terminology--works fine for that.  So the problem can really be reduced to how you get inflexible transfer lines extended and reliably connected, then maintain those connections during however you decide to do the ullage acceleration.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 05/26/2022 06:38 pm

I don't think the guidance and control issues are the problem here.  But, absent a rigid connection between the two spacecraft, you need a flexible set of cryogenic transfer lines, and a way of extending those lines from one ship to the other and connecting them securely.  The low temperature of the fluids is going to put some extreme materials constraints on flexible lines.  I'm inclined to believe that the challenges of a rigid connection are a lot easier to overcome.



Why do they use flexible hoses on the ground gse? They don't seem to have a cold/flexibility problem.
The ground isn't that much different thermally than LEO.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 05/26/2022 07:20 pm
Has everybody given up on the lightest in terms of hardware of precision parallel flying? Both vehicles need different ullage thrust.

Fluid moving from vehicle to vehicle would impart a thrust? Which would have to be countered?
Not sure if this true?
Fluid is moving at constant velocity within the hose connecting them. You only feel thrust from a hose where it exits the hose.
Since the net change is moving 1200t from one starship to the other it would seem that is the impulse that needs to be countered.
Initiating the transfer will impart an impulse that will start moving the ships and ending it will impart the opposite impulse bringing everything to rest. Remember that the center of mass of the whole system will remain at rest.

The simplest thought experiment is two massless tanks next to each other, one full and one empty. Transferring the propellant at a constant rate would move the tanks linearly so that the receiving tank replaces the delivering tank and the propellant is stationary. If the propellant is settled to the "bottom" the tanks will first move "up" and then "down" as well.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/26/2022 08:48 pm
Why do they use flexible hoses on the ground gse? They don't seem to have a cold/flexibility problem.
The ground isn't that much different thermally than LEO.

I just went through some video of the QD arm moving and everything I saw looked like rigid plumbing to me.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 05/26/2022 10:15 pm
Why do they use flexible hoses on the ground gse? They don't seem to have a cold/flexibility problem.
The ground isn't that much different thermally than LEO.

I just went through some video of the QD arm moving and everything I saw looked like rigid plumbing to me.
It is basically the same type of plumbing used in the gimbaling engines. Two or more rotating joints allowing for a limited amount of motion. Remember this piping and connectors have to work for a few dozen times without failing.

As a hedge against the depot QD failure. Have a normal looking tanker face 180 degrees around the vehicle from the depot face. Such that for launch the tanker face is what connects to the QD on the tower. Then that interface is not used on orbit unless the depot face fails and the prop needs to be transferred to a new depot. Interestingly is that the depots then are the definition of a depot tanker having both interfaces.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 05/27/2022 04:15 am
Why do they use flexible hoses on the ground gse? They don't seem to have a cold/flexibility problem.
The ground isn't that much different thermally than LEO.

I just went through some video of the QD arm moving and everything I saw looked like rigid plumbing to me.
The ship QD doing it's best dead bobbit worm impression during installation did not look particularly rigid to me...
(At 5m3s)
https://youtu.be/sK44dhBTYqA?t=5m3s
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: CameronD on 05/27/2022 04:21 am
Why do they use flexible hoses on the ground gse? They don't seem to have a cold/flexibility problem.
The ground isn't that much different thermally than LEO.

I just went through some video of the QD arm moving and everything I saw looked like rigid plumbing to me.
The ship QD doing it's best dead bobbit worm impression during installation did not look particularly rigid to me...
(At 5m3s)

You do realise that's time-lapse (not real time) video??

Besides, cryo pipe has to have at least some amount of flexibility built-in to cater for expansion and contraction of the metal at such crazy low temperatures and high pressures.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/27/2022 10:50 am
The cryogen hoses are flexible. They are not a series of rotary joints on rigid sections, they are instead a series of bellows, enclosed in a steel mesh sleeve to prevent expansion (same bellows are used without the restraint as expansion joints. They can be formed on an expanding mandrel from a continuous tube or by stacking a series of alternate rings and welding.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/27/2022 06:18 pm
The cryogen hoses are flexible. They are not a series of rotary joints on rigid sections, they are instead a series of bellows, enclosed in a steel mesh sleeve to prevent expansion (same bellows are used without the restraint as expansion joints. They can be formed on an expanding mandrel from a continuous tube or by stacking a series of alternate rings and welding.

A bellows system sounds like it's not going to work for extending via multi-degree-of-freedom steering from one Starship to another.  However, that's OK, as long as the two Starships are latched in fairly rigid alignment with respect to one another.  Then there only needs to be a one-axis extension to make the connection.

Percentage-wise, how much extension can you get with a bellows?  And how many extension/retraction cycles can it survive before it needs to be replaced?  I figure they'll need at least a meter of extension, probably more.  And if this is the mechanism built into a depot, the number of cycles will control the reusability of the depot.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/28/2022 10:23 am
The cryogen hoses are flexible. They are not a series of rotary joints on rigid sections, they are instead a series of bellows, enclosed in a steel mesh sleeve to prevent expansion (same bellows are used without the restraint as expansion joints. They can be formed on an expanding mandrel from a continuous tube or by stacking a series of alternate rings and welding.

A bellows system sounds like it's not going to work for extending via multi-degree-of-freedom steering from one Starship to another.  However, that's OK, as long as the two Starships are latched in fairly rigid alignment with respect to one another.  Then there only needs to be a one-axis extension to make the connection.
It's literally the system used for the QD arm's QD plate. We've already seen it in action, it can move at least 1m laterally and several meters towards the ship QD plate from its resting position.
Cryogenic flex hoses are an off-the-shelf item. A solved problem.
And as if on schedule, here's a video of it in action:
https://twitter.com/CSI_Starbase/status/1530366511784898562

Quote
Hey #StarshipAddicts, check out this test of the new #Starship Quick Disconnect fueling arm. This is the first time we have seen this in action seen since it was replaced to fit the updated vehicle design.

🎥:
@StarshipGazer LIVE!

https://youtube.com/watch?v=C7cl7-Ykh_w
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/28/2022 07:52 pm
The cryogen hoses are flexible. They are not a series of rotary joints on rigid sections, they are instead a series of bellows, enclosed in a steel mesh sleeve to prevent expansion (same bellows are used without the restraint as expansion joints. They can be formed on an expanding mandrel from a continuous tube or by stacking a series of alternate rings and welding.

A bellows system sounds like it's not going to work for extending via multi-degree-of-freedom steering from one Starship to another.  However, that's OK, as long as the two Starships are latched in fairly rigid alignment with respect to one another.  Then there only needs to be a one-axis extension to make the connection.
It's literally the system used for the QD arm's QD plate. We've already seen it in action, it can move at least 1m laterally and several meters towards the ship QD plate from its resting position.
Cryogenic flex hoses are an off-the-shelf item. A solved problem.
And as if on schedule, here's a video of it in action:

Eyeballing this, it looks like about 7m of piping to get maybe 2.5m of travel.  So maybe 35% expansion?

But note two things here:

1) It's not really full expansion.  Instead, the piping is arranged to allow a fairly small amount of bending along its length.

2) This is very much a one-degree-of-freedom system, with all of the bending occurring in one plane.  I believe that it's probably easy to make it two degrees of freedom (bending in a combination of two planes).

The question I was trying to get answered is whether that kind of piping could be used as the sole connection between a couple of on-orbit Starships.  That would make it more like a grapple than a soft-capture.  Grapples require a lot of degrees of freedom, so they can accommodate rotational and translational misalignments and rate errors.  I can't see that happening with a bellows-based set of piping.

The other question is how you'd pack a set of piping so that it could extend enough to get the job done.  7m will fit in the payload bay, but it's a pretty weird configuration--and a horribly complex one.  And it would clearly rule out androgynous/hermaphroditic configurations between tankers and payload Starships (although it might work between tankers and depots).

But if you have a rigid "dock" (let's assume we all know what this means, rather than diving back into the terminology), then there are much easier ways of making an extendable connection of 2ish meters.  How 'bout a simple telescoping connector, as shown in the attached?

(Note that my crude artwork isn't showing any detail of how the telescoping flanges latch to one another to form a seal.  I'm assuming that this is a solvable problem.)

This is a gendered version, but you could make a hermaphroditic version simply by pairing two versions of the connectors for each fill/drain system (i.e., four cryo connectors, two for LOX and two for LCH4).

Note that this plumbing can be as low pressure as you want.  You get lower boil-off losses if you keep the system at 4-6bar during transfer, but you can drop the pressure lower if that increases the reliability of the telescoping connections.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/29/2022 11:54 am
That's still searching for new solutions to an already solved problem. Trying to make a rigid pipe flexible in order to have it be used as a rigid connection, vs. using the flexible pipe already in use and an articulated structure already in use, means a whole bunch of unnecessary R&D and testing.

Eyeballing this, it looks like about 7m of piping to get maybe 2.5m of travel.  So maybe 35% expansion?
No expansion. It just flexes, like flexible pipe does. Because its flexible pipe. Visible MBR is sub-metre, but that depends mostly on pipe diameter.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 05/29/2022 12:19 pm
https://twitter.com/CSI_Starbase/status/1530366511784898562

Did anyone else wonder what mechanism was pivoting that small light around, then realised it was a person and instantly had the scale of the whole thing snap out an order of magnitude?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/29/2022 05:13 pm
The cryogen hoses are flexible. They are not a series of rotary joints on rigid sections, they are instead a series of bellows, enclosed in a steel mesh sleeve to prevent expansion (same bellows are used without the restraint as expansion joints. They can be formed on an expanding mandrel from a continuous tube or by stacking a series of alternate rings and welding.

A bellows system sounds like it's not going to work for extending via multi-degree-of-freedom steering from one Starship to another.  However, that's OK, as long as the two Starships are latched in fairly rigid alignment with respect to one another.  Then there only needs to be a one-axis extension to make the connection.
It's literally the system used for the QD arm's QD plate. We've already seen it in action, it can move at least 1m laterally and several meters towards the ship QD plate from its resting position.
Cryogenic flex hoses are an off-the-shelf item. A solved problem.
And as if on schedule, here's a video of it in action:

Eyeballing this, it looks like about 7m of piping to get maybe 2.5m of travel.  So maybe 35% expansion?

But note two things here:

1) It's not really full expansion.  Instead, the piping is arranged to allow a fairly small amount of bending along its length.

2) This is very much a one-degree-of-freedom system, with all of the bending occurring in one plane.  I believe that it's probably easy to make it two degrees of freedom (bending in a combination of two planes).

The question I was trying to get answered is whether that kind of piping could be used as the sole connection between a couple of on-orbit Starships.  That would make it more like a grapple than a soft-capture.  Grapples require a lot of degrees of freedom, so they can accommodate rotational and translational misalignments and rate errors.  I can't see that happening with a bellows-based set of piping.

The other question is how you'd pack a set of piping so that it could extend enough to get the job done.  7m will fit in the payload bay, but it's a pretty weird configuration--and a horribly complex one.  And it would clearly rule out androgynous/hermaphroditic configurations between tankers and payload Starships (although it might work between tankers and depots).

But if you have a rigid "dock" (let's assume we all know what this means, rather than diving back into the terminology), then there are much easier ways of making an extendable connection of 2ish meters.  How 'bout a simple telescoping connector, as shown in the attached?

(Note that my crude artwork isn't showing any detail of how the telescoping flanges latch to one another to form a seal.  I'm assuming that this is a solvable problem.)

This is a gendered version, but you could make a hermaphroditic version simply by pairing two versions of the connectors for each fill/drain system (i.e., four cryo connectors, two for LOX and two for LCH4).

Note that this plumbing can be as low pressure as you want.  You get lower boil-off losses if you keep the system at 4-6bar during transfer, but you can drop the pressure lower if that increases the reliability of the telescoping connections.
Plane old flexible hose works. It becomes ridged when cryo flows and thaws afterword. Electric heater bands if needed for thaw. I've done it with a non rigid connections and expect careful attention to coefficient of expansion would make it work. Or bellows as  edzeiba suggests. The plumbing need not be the same diameter as the connector. If smaller, the packing problem changes.


One idea that sticks in my fertile (fevered?) mind is a blister on the depot that holds the transfer plumbing. A half meter deep or thereabouts. This assumes a mirror QD plate. A half meter is not insignificant, especially during MaxQ, but should be doable.


This raises the question: If there is a three point connection between the depot and the other ship, is this half meter an acceptable separation. This assumes the three point connection is not through the QD plate. If so, no extension mechanism needed. No part, Yada yada.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/29/2022 05:46 pm
That's still searching for new solutions to an already solved problem. Trying to make a rigid pipe flexible in order to have it be used as a rigid connection, vs. using the flexible pipe already in use and an articulated structure already in use, means a whole bunch of unnecessary R&D and testing.

Eyeballing this, it looks like about 7m of piping to get maybe 2.5m of travel.  So maybe 35% expansion?

No expansion. It just flexes, like flexible pipe does. Because its flexible pipe. Visible MBR is sub-metre, but that depends mostly on pipe diameter.

It's clearly expanding in the QD, because the outlet is closer to the inlet after the QD retracts than it was before.  In the QD's case, that's caused by flexion, with the bending causing the loop of pipe to translate down, making way for the end of the pipe to translate right.

This is not what you'd want for a refueling connection, because it's volumetrically bulky.  If you can get the bellows to flex so that a straight segment of pipe can expand purely axially, then bellows is the way to go.  If not, then it's not really a solved problem.

But even then, you're still going to have to have a rigid connection that's aligned the connectors ahead of time, because there's no way that you're going to get multi-degree-of-freedom precision from the bellows sufficient to deal with alignment and rotational errors on approach.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/29/2022 06:55 pm
...The video is right there.
Cryogneic flexible hoses: Exist. Available off-the-shelf. Do not stretch (due to wire sheath). Currently used for literally loading and offloading propellants to Starship.
QD arm's actuated QD plate: Exists. Can translate in X, Y, and Z as well as roll/pitch/yaw to align with Starship's QD plate - actuators to do so are visible attached to the rear. Actuation is powered and actively controlled (with alignment confirmation cameras visible on the QD plate). Currently used for literally loading and offloading propellants to Starship.

What exactly is a complex system of concentric tubes - with a new sealing problem (siding surfaces, in a vacuum, and cryogen compatible), vulnerability to any sort of imparted bending, no capacity for alignment without an additional rotating fluid coupler at the end, and entirely bespoke - adding over the system already sitting on site and that has already successfully loaded cryogenic propellants onto Starship and already has the capacity to provide a fluid coupling to a QD plate with 6-axis alignment capability?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: CorvusCorax on 05/29/2022 07:43 pm
More important than the question of androgynous vs male/female is the question of active vs passive.

One side, one the ground the qd arm, needs all the actuators not just for alignment but also for hard mating, fastening the,seals, and opening everything again once done. The otherside doesn't.

Even on technically androgynous docking adapters, ( see ISS) you typically have only one active side. As thats the side with the majority of complexity. For prop transfer add purge systems and valves, automated leak checks, etc...

All this makes the qd arm a multi ton behemoth vs a simple mostly passive plate on the ship.

It would probably not make sense to equip every starship with all of this extra hardware. But the moment they don't, and you have any difference, its typically easyest to also make the physical connection asymetric and say, for example, the active side is male and the passive side is female.

A ship acting as a depot must be,active, for then all other ships don't have to be.

A hard mate in more than 1 spot between Starships would still be a good thing IMHO, for then you can use centrifugal microgravity for prop settling instead of continuous propellant wastage.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/29/2022 08:53 pm
...The video is right there.
Cryogneic flexible hoses: Exist. Available off-the-shelf. Do not stretch (due to wire sheath). Currently used for literally loading and offloading propellants to Starship.
QD arm's actuated QD plate: Exists.

I think where our fundamental disagreement lies is that you think that plumbing connections can be made without prior establishment of a rigid connection between the two Starship.  I don't think that's possible.

I understand what you're saying about how the piping flexes.  You seem to be answering my question about whether it can extend (stretch) in the negative--which is what I expected.  In the arm--right there in video--there's a big, curved piece of piping that extends down in a 'U' shape, which is how there's enough flexion to do the actual retraction of the QD.  What I don't understand is how you'd put that kind of geometry into a connector with 12 degrees of freedom (6 translation/rotation positions, 6 translation/rotation velocities), and how that amount of plumbing could be packed into a tanker.  Could you provide a sketch?[/quote]

Quote
What exactly is a complex system of concentric tubes - with a new sealing problem (siding surfaces, in a vacuum, and cryogen compatible), vulnerability to any sort of imparted bending, no capacity for alignment without an additional rotating fluid coupler at the end, and entirely bespoke - adding over the system already sitting on site and that has already successfully loaded cryogenic propellants onto Starship and already has the capacity to provide a fluid coupling to a QD plate with 6-axis alignment capability?

I'm not wedded to telescoping plumbing.  I just don't see how bellows-based plumbing is going to be compact enough.  I also don't see how it's going to be accurate enough to do capture on the passive Starship.  The QD system used in the GSE has no more than a few centimeters of translational error, no more than a few minutes or arc of rotational error, and virtually no velocity error at all.  Those errors are going to be at least a factor of 10 larger on orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 05/29/2022 09:26 pm
...The video is right there.
Cryogneic flexible hoses: Exist. Available off-the-shelf. Do not stretch (due to wire sheath). Currently used for literally loading and offloading propellants to Starship.
QD arm's actuated QD plate: Exists. Can translate in X, Y, and Z as well as roll/pitch/yaw to align with Starship's QD plate - actuators to do so are visible attached to the rear. Actuation is powered and actively controlled (with alignment confirmation cameras visible on the QD plate). Currently used for literally loading and offloading propellants to Starship.

What exactly is a complex system of concentric tubes - with a new sealing problem (siding surfaces, in a vacuum, and cryogen compatible), vulnerability to any sort of imparted bending, no capacity for alignment without an additional rotating fluid coupler at the end, and entirely bespoke - adding over the system already sitting on site and that has already successfully loaded cryogenic propellants onto Starship and already has the capacity to provide a fluid coupling to a QD plate with 6-axis alignment capability?

Does the current QD setup load propellant at the full 6 bar? If so, operating at 1 less bar ambient shouldn be an issue. If it loads at ambient pressure, however, it might need some redesigning to handle vacuum operation.

I still stan the docking drone, however. Fold out arms that grapple the starship's tower catch points and the QD mount of both ships.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/30/2022 08:34 am
...The video is right there.
Cryogneic flexible hoses: Exist. Available off-the-shelf. Do not stretch (due to wire sheath). Currently used for literally loading and offloading propellants to Starship.
QD arm's actuated QD plate: Exists.

I think where our fundamental disagreement lies is that you think that plumbing connections can be made without prior establishment of a rigid connection between the two Starship.  I don't think that's possible.
I do, because it's how plumbing connections between space vehicles have been made for more than 40 years (since Progress 1).
Quote
What I don't understand is how you'd put that kind of geometry into a connector with 12 degrees of freedom (6 translation/rotation positions, 6 translation/rotation velocities), and how that amount of plumbing could be packed into a tanker.  Could you provide a sketch?
No sketch needed, you can watch the system in action. The hose flexes in any direction, because its a hose. You can play with your vacuum cleaner hose, it's a very similar concertina bellows construction.
Quote
I just don't see how bellows-based plumbing is going to be compact enough.
Is 'compact' a critical design criteria, and if so how compact is required? Is Starship payload volume constrained enough that the packed volume cannot be accommodated? For the initial Starship depot variant in particular, you have a whole fairing volume you can pack your system in that's otherwise going to waste, and with no need to retract the system or re-pressurise the nose volume for EDL. Later variants can worry about making the system smaller if that becomes necessary, but no need to overcomplicate things from the start with unnecessary requirements.
Quote
I also don't see how it's going to be accurate enough to do capture on the passive Starship.  The QD system used in the GSE has no more than a few centimeters of translational error, no more than a few minutes or arc of rotational error, and virtually no velocity error at all.  Those errors are going to be at least a factor of 10 larger on orbit.
The STS orbiter managed to successfully align with docking connectors without much issue, with low velocity approaches made to sub-centimetre accuracy. The 'factor of 10' is more of a factor of 1.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/30/2022 10:44 am
I do, because it's how plumbing connections between space vehicles have been made for more than 40 years (since Progress 1).

Progress hard-docks, then transfers fuel.  The plumbing isn't doing the docking.

Quote
No sketch needed, you can watch the system in action. The hose flexes in any direction, because its a hose. You can play with your vacuum cleaner hose, it's a very similar concertina bellows construction.

It's pretty clearly not flexing in any direction.  It's flexing in one plane only.

Quote
Is 'compact' a critical design criteria, and if so how compact is required? Is Starship payload volume constrained enough that the packed volume cannot be accommodated? For the initial Starship depot variant in particular, you have a whole fairing volume you can pack your system in that's otherwise going to waste, and with no need to retract the system or re-pressurise the nose volume for EDL.

Depots' payload bays will mostly be consumed by tankage, at least up to the ogive.  How are you going to fit a big door into the ogive?  More importantly, how are you going to dock with the ogive section?

Quote
The STS orbiter managed to successfully align with docking connectors without much issue, with low velocity approaches made to sub-centimetre accuracy. The 'factor of 10' is more of a factor of 1.

Nope.  Again, that used APAS--a docking system that was designed to handle 12 degrees of freedom, via a soft capture device.  How are you going to implement that with a hose tricked out with a bellows?

FWIW, here are the limits for the NDS block 1:
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 05/30/2022 11:04 am
...The video is right there.
Cryogneic flexible hoses: Exist. Available off-the-shelf. Do not stretch (due to wire sheath). Currently used for literally loading and offloading propellants to Starship.
QD arm's actuated QD plate: Exists. Can translate in X, Y, and Z as well as roll/pitch/yaw to align with Starship's QD plate - actuators to do so are visible attached to the rear. Actuation is powered and actively controlled (with alignment confirmation cameras visible on the QD plate). Currently used for literally loading and offloading propellants to Starship.

What exactly is a complex system of concentric tubes - with a new sealing problem (siding surfaces, in a vacuum, and cryogen compatible), vulnerability to any sort of imparted bending, no capacity for alignment without an additional rotating fluid coupler at the end, and entirely bespoke - adding over the system already sitting on site and that has already successfully loaded cryogenic propellants onto Starship and already has the capacity to provide a fluid coupling to a QD plate with 6-axis alignment capability?

Does the current QD setup load propellant at the full 6 bar? If so, operating at 1 less bar ambient shouldn be an issue. If it loads at ambient pressure, however, it might need some redesigning to handle vacuum operation.

I still stan the docking drone, however. Fold out arms that grapple the starship's tower catch points and the QD mount of both ships.
Pushing LOX from the QD interface to the top of the header tank requires 5-6 bar.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/30/2022 11:45 am
I do, because it's how plumbing connections between space vehicles have been made for more than 40 years (since Progress 1).

Progress hard-docks, then transfers fuel.  The plumbing isn't doing the docking.

Quote
No sketch needed, you can watch the system in action. The hose flexes in any direction, because its a hose. You can play with your vacuum cleaner hose, it's a very similar concertina bellows construction.

It's pretty clearly not flexing in any direction.  It's flexing in one plane only.

Quote
Is 'compact' a critical design criteria, and if so how compact is required? Is Starship payload volume constrained enough that the packed volume cannot be accommodated? For the initial Starship depot variant in particular, you have a whole fairing volume you can pack your system in that's otherwise going to waste, and with no need to retract the system or re-pressurise the nose volume for EDL.

Depots' payload bays will mostly be consumed by tankage, at least up to the ogive.  How are you going to fit a big door into the ogive?  More importantly, how are you going to dock with the ogive section?

Quote
The STS orbiter managed to successfully align with docking connectors without much issue, with low velocity approaches made to sub-centimetre accuracy. The 'factor of 10' is more of a factor of 1.

Nope.  Again, that used APAS--a docking system that was designed to handle 12 degrees of freedom, via a soft capture device.  How are you going to implement that with a hose tricked out with a bellows?

FWIW, here are the limits for the NDS block 1:
Attached is a photo of the QD plate. Note the alignment features (ball-ended pins and shock-absorbing 'bumpers'), and the latches. Used to align the QD plate, and to latch the QD plate.

Arguing that the 'plumbing' is not docking is missing the solution already implemented. No need to re-invent the wheel, particularly when said wheel is sitting out in the open, has been publicly photographed, and has been used to successfully perform propellant loading operations with attach and detach cycles.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: JamesH65 on 05/30/2022 02:59 pm
I do, because it's how plumbing connections between space vehicles have been made for more than 40 years (since Progress 1).

Progress hard-docks, then transfers fuel.  The plumbing isn't doing the docking.


So hard dock, which can be done irrespective of any fuel connection, then extend a fuelling arm as per the current system.

Why does this seem to be such a problem?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 05/30/2022 06:41 pm
So hard dock, which can be done irrespective of any fuel connection, then extend a fuelling arm as per the current system.

Why does this seem to be such a problem?

That's pretty much what I've been proposing.  I just don't think the fueling arm that's part of the GSE makes a lot of sense for on-orbit operations.  And even if it does, it still needs a rigid system to work.

Attached is a photo of the QD plate. Note the alignment features (ball-ended pins and shock-absorbing 'bumpers'), and the latches. Used to align the QD plate, and to latch the QD plate.

Arguing that the 'plumbing' is not docking is missing the solution already implemented. No need to re-invent the wheel, particularly when said wheel is sitting out in the open, has been publicly photographed, and has been used to successfully perform propellant loading operations with attach and detach cycles.

You don't need docking on the ground.  At the very worst, the Starship is dangling in a one gee field, held by the chopsticks.  More likely, it's also held rigidly in place by the SuperHeavy connections and the other supports on the QD arm before the plate is connected.  In addition, latching the QD plate in place is a quasi-static process: If there are misalignments, the ground crew will detect them and take corrective action.  That's easy to do because the Starship isn't moving.  If it's windy enough that the entire mechazilla/chopsticks/starship/superheavy assemblage is moving, you can simply wait for the wind to die down.  This is fundamentally different than two decoupled Starships floating around in microgravity.

It looks like a very fine wheel they've invented.  That doesn't help very much if what they really need is something closer to ice skates.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/30/2022 09:00 pm
This is fundamentally different than two decoupled Starships floating around in microgravity.
Yes: in microgravity, there is no wind to disturb you, and precise alignment of pairs of several hundred tonne objects has been demonstrated to work in practice (STS rendezvous with the ISS).
We can suppose that Starship will lack an effective RCS system unlike every space vehicle thus far including SpaceX's own prior two vehicles, but that seems like an exceptionally poor supposition.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/31/2022 01:05 am
Wow, this is really going around the bend.


Ignoring all other issues, an extension mechanism using bellows plumbing is almost trivial. The transfer plumbing need not be the same diameter piping as that used on the GSE. With a smaller diameter it packs more easily and wastage on disconnect will not even be a rounding error.


A pigtail or S layout gives all the flexibility needed. A standard solution to a common problem.


Smaller lines would have an impact on transfer times. That's what tradeoffs are all about.


Bringing the idea up again, would a half or 3/4 meter separation between ships be acceptable? Using a blister that depth on the depot with a mirror QD plate on its face would allow hookup without an extension mechanism, assuming another mechanism is used to draw the ships together and hold them.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 05/31/2022 03:39 am
After looking at the QD on the OLIT arm. It looks like a separation of 3/4 m should do it. But before the QD plate on the depot moves out to mate another system needs to have already established a separate rigid body grapple/capture. From a practical standpoint thing of a some sort of short grapple arm capable of snagging the tanker and then able to bring the two to an almost perfect alignment at a tank to tank separation of 3/4m. Perfect alignment is a mostly a tongue in check term for close enough and fairly stable relationship between 2 bodies in space. Think of the berthing procedure vs that of docking. Then the depot plat is pushed into the tanker plate QD to establish mating.

It should not be that hard. Almost all of this hardware already exits. Just not all put together to do this specific implementation. As well as most of it having been used in space already but not specifically for this type implementation.

From an engineering standpoint it is the adaptation of existing technology used in space that have been around for 40 years or more to end with a system that has a capability that has yet to be achieved. Just because the pieces have yet to be put together this way to achieve this goal.

(Sorry my engineering background as well as 11 years of large space vehicles experience [Shuttle] in the ~1980's, real hands on hardware stuff, came out.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 05/31/2022 07:15 am
Berthing is an example of why an additional arm is likely not necessary if the QD arm can move like the QD arm currently moves (6 axis motion with Z having an extended range):
- During ISS berthing, a vehicle actively stationkeeps in a fixed-relative-to-the-host position
- The host extends an arm to the 'stationary' vehicle and optically aligns with the capture point on the vehicle
- The arm moves the last few cm and attaches to the 'stationary' vehicle
- Arm drives clamps, vehicle is caught
- Arm continues to drive clamps, vehicle rigid mate is achieved and electrical connections made
- Arm repositions vehicle to empty CBM, realigns, and moves vehicle to mate

Add "and plumbing" between the word 'electrical' and 'connections' and delete the last step and you have the sequence needed for linking two Starships for propellant transfer. While Canadarm2 has more dexterity than the QD arm, the QD arm also does not need to perform the range of dextrous manipulation tasks that Canadarm2 does, nor does it need to manoeuvre a vehicle after connection is made. The two Starships can align themselves in the 'final' position before starting the sequence, so the arm only needs to reach across the separation gap rather than perform any manipulation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 05/31/2022 05:15 pm
The CANADARM2 has dexterity of 360 degree of manipulation of rotations in all 3 dimensions and the 10 or more meter of displacement in 3 dimensions. A QD with cryo piping will do good to have a 10 degree rotations in 2 dimensions as well as displacements capability of at best <5m in 1 dimension and significantly less than 5 m, more likely less than 2 m in the other 2 dimensions. A QD arm would work only when things are close to perfect between 2 free flyer multi 100t objects.

You want a system that can handle off nominal situations and still reach useful results. This is to be a system that has an ability to work 99+% of the time not <85% of the time. A higher dexterity arm without the cryo piping is a solution that increases the reliability of the cryo prop transfer success. It is because of the flexibility of the cryo piping that is the problem, and solely the problem that has to be solved when added to a mating plate that reaches out to attach to the Tanker's QD plate. Even though this QD plate on Tanker could be a special plate with smaller connectors and the resulting smaller piping it still causes high movement restrictions in the needed 6 DOF to perform a hard capture to be able to do the prop transfer without most of cryo leaking out of the connectors when the orientation between the 2 vehicles moves around during the low accel needed to do prop settling to be successful in transferring any prop. The forces needed by the structural components of the capture mechanism and prop arm would have to be able to exert multiple 10s of tons metric between the 2 vehicles. A simple note is that CANADARM and CANADARM2 were designed to be able to accomplish these force levels.

Another item here is that the CANADRAM(and 2 as well) does not latch onto a target object but is more appropriate to categorize it as lassoing the target. Then pulling on the "rope" to establish a tight interface of two shaped faces to create a situation of high forces capable capture due to the very high fit without having to have any latches at all. Since there are no latches and the lasso is manipulateable such that it can go from tight to totally loose. It is a very ingenious system with very very high ability to handle situations that are far from ideal. Such as lassoing a significant moving and rotating target. Try that with a arm with cryo piping.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/31/2022 07:51 pm
After looking at the QD on the OLIT arm. It looks like a separation of 3/4 m should do it. But before the QD plate on the depot moves out to mate another system needs to have already established a separate rigid body grapple/capture. From a practical standpoint thing of a some sort of short grapple arm capable of snagging the tanker and then able to bring the two to an almost perfect alignment at a tank to tank separation of 3/4m. Perfect alignment is a mostly a tongue in check term for close enough and fairly stable relationship between 2 bodies in space. Think of the berthing procedure vs that of docking. Then the depot plat is pushed into the tanker plate QD to establish mating.

It should not be that hard. Almost all of this hardware already exits. Just not all put together to do this specific implementation. As well as most of it having been used in space already but not specifically for this type implementation.

From an engineering standpoint it is the adaptation of existing technology used in space that have been around for 40 years or more to end with a system that has a capability that has yet to be achieved. Just because the pieces have yet to be put together this way to achieve this goal.

(Sorry my engineering background as well as 11 years of large space vehicles experience [Shuttle] in the ~1980's, real hands on hardware stuff, came out.)
I think you missed one of my points. I've been noodling back and forth on both extendable and non extendable QD plates so the confusion is on me.


The blister idea could be adaptable to either approach but the non extendable is what really interests me. The depth of the blister would be the spacing between the ships. I've no objection to the QD plate extending if it must, but if the ships can safely snuggle up that close it can be done away with.


This idea is independent of how the ships get snuggled up and latched together. Got to admit I like the Canadarm family and have other reasons to want it on the depot but if the lash up optimizes to another solution it's best that 'really neat' doesn't get in the way of 'good enough'.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/31/2022 08:42 pm
Going off topic a bit but directly related to extendable & maneuverable QD vs Canadarm.


Looking at the depot as being strictly a propellant depot I can go either way. If the ships can by some method (a Canadarm or similar probably) be brought close enough for a blister w/mirror QD plate to kiss the the other ships QD plate, the complexity of the extension mechanism would be avoided. KISS.


OTOH, a Canadarm adds it's own complexity. To counter this, it is proven technology and its perfectly adequate for rasslin them two big ships together. Without fins, the depot has a couple very nice places where an arm can stow for liftoff.


Now, here's another reason to consider an arm. This is looking maybe 10 years out so it's not a burning issue if for some reason an arm isn't needed currently. Throughout these forums we see ideas for orbital repairs, salvaging engines and the like. What lacks is a viable infrastructure to support these ideas.


Over time various families of orbits will find themselves used more often than others. This is where depots will naturally congregate. And this is where a maintenance facility will have enough trade to justify itself. A depot would already have a lot of what a maintenance facility needs and an arm or two takes it closer. Think of it as a full service Service Station. You younger readers out there might not know what this is. Ask an old fart.


On orbit assembly, as the Shuttle did for the ISS, is a capability we no longer have. A depot with arms can do this. Not perfect but good enough until something better is needed.


As I said earlier, if lashing the two ships for propellant transfer optimizes in a direction different than Canadarm, so be it. If it's a close call...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 05/31/2022 09:03 pm
Its beginning to sound like a small free flying universal adapter might be the way to go. Grappling stuff(arms) and 2 QD male versions. Maybe make it telescoping for the upper and lower grapple points.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 05/31/2022 09:33 pm
Thinking about it the combo of a blister QD where the QD on the blister can "wiggle". That movement is very limited to a few CM in position and rotation angles so that the arm is the one the is the primary controller of the alignment and position to the point that the blister and the Tanker QD "bump" and capture.

Arms mass about 460kg for a Shuttle sized arm. The CANADARM2 has been in orbit for years with a very low failure rate. It may take a while to find out when it failed last. If it has at all. There has been some upgrades on the CANADARM2 so that clouds the issue. A 1t mass budget for a pair of arms .5t each including the mounts and covers on depot. Only one of the two have to work so it is redundancy at the macro level. Should be able to provide sufficient capability to use the depot for as long as the QD plate does not fail by leaking too much during transfers.

If I remember correctly the end effector on the CANADARM was less than a meter in length itself. Such that 2 blisters one with the QD and another with the EE mounted on a driven plate that can swivel 3 dimensions of rotations of a few degrees and a few 10s of CMs in 2 dimensions for translation position to take care of misalignment caused by expansion contractions making the 2 40+m long SS when they "get ready to "bump" together may not be exactly the same length at that point. And also during prop transfer the lengths of each could shift some more. So no arm just and EE to grab and manipulate the fine 6D relationship to get the 2 QD plates to capture and mate.

The design space for workable solutions is not small. Which is an encouraging point by itself. That the on orbit prop transfer is not something as difficult as most make it. And primarily a study of the possible solutions and evaluations of them for which will best satisfy the SpaceX goals whatever they may end up being. For prop transfer and on orbit short term storage which is likely less than a year period and possible even significantly less than 6 months.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/31/2022 09:55 pm
Its beginning to sound like a small free flying universal adapter might be the way to go. Grappling stuff(arms) and 2 QD male versions. Maybe make it telescoping for the upper and lower grapple points.
Hmmm. Once the depot has an arm, an adapter plate does become doable and would mass less. It also becomes a sacrificial wear point and can be replaced - if it can fit in a cargo ships hold. I like it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 05/31/2022 10:44 pm
Thinking about it the combo of a blister QD where the QD on the blister can "wiggle". That movement is very limited to a few CM in position and rotation angles so that the arm is the one the is the primary controller of the alignment and position to the point that the blister and the Tanker QD "bump" and capture.

Arms mass about 460kg for a Shuttle sized arm. The CANADARM2 has been in orbit for years with a very low failure rate. It may take a while to find out when it failed last. If it has at all. There has been some upgrades on the CANADARM2 so that clouds the issue. A 1t mass budget for a pair of arms .5t each including the mounts and covers on depot. Only one of the two have to work so it is redundancy at the macro level. Should be able to provide sufficient capability to use the depot for as long as the QD plate does not fail by leaking too much during transfers.

If I remember correctly the end effector on the CANADARM was less than a meter in length itself. Such that 2 blisters one with the QD and another with the EE mounted on a driven plate that can swivel 3 dimensions of rotations of a few degrees and a few 10s of CMs in 2 dimensions for translation position to take care of misalignment caused by expansion contractions making the 2 40+m long SS when they "get ready to "bump" together may not be exactly the same length at that point. And also during prop transfer the lengths of each could shift some more. So no arm just and EE to grab and manipulate the fine 6D relationship to get the 2 QD plates to capture and mate.

The design space for workable solutions is not small. Which is an encouraging point by itself. That the on orbit prop transfer is not something as difficult as most make it. And primarily a study of the possible solutions and evaluations of them for which will best satisfy the SpaceX goals whatever they may end up being. For prop transfer and on orbit short term storage which is likely less than a year period and possible even significantly less than 6 months.
Interesting. How would the end effectors deploy? MaxQ would be rough without protection. An arm can inchworm to where it needs to be.


I was thinking an arm would deploy struts between the ships. Put a jack screw in each strut and pick the connection points carefully and they could compensate for thermal creep - I think.


Noodling that last, one strut to each side of the QD plate one quarter circumference out and two struts up forward, both connecting to one point on one of the ships. These last two would align fore and aft so that one jacking out while the other jacks in would do for thermal creep. The aft struts wouldn't need jack screws. The struts would need to be strong enough to withstand expected forces while holding but would not need to be stout enough to horse a ship around while hooking up.


With arms an adapter plate becomes practical saving the mass of a mirror plate. It also does away with the blister as the adapter can be as thick as necessary - as long as it will fit into what little cargo space is available.


This is going to be one heavy mother by the time we're through. Lol.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 05/31/2022 11:16 pm
In a little research looking for the failures over 20 years of a very high operations activity rate. Over the past 10 years it has been a rate that the arm was active 50% of the time. The work the system so hard that they have to every few years replace the end effectors because they wear the mechanical parts out. The joints are more robust and only a few joint fixes have occurred. The first failures in the first year of operations was related to the computer and not the mechanical hardware. One of the repairs was due to a hit by space debris causing damage.

The following has a table comparing features and capabilities between the 3 CANADARM versions but unfortunately no reliability numbers : https://www.asc-csa.gc.ca/eng/iss/canadarm2/canadarm-canadarm2-canadarm3-comparative-table.asp (https://www.asc-csa.gc.ca/eng/iss/canadarm2/canadarm-canadarm2-canadarm3-comparative-table.asp)

An interesting item is that a situation where you had two CANADARM3's they could with a handful of replacement parts keep each other fully functional without having any humans on site or in local control as in low latency com link distance. The arms can operate full auto mode. CANADARM3's are robots not arms.

The life of the EE's looks to be about 4,000 captures. That equates to being able to do enough tanking prop transfers ops for 500 Lunar or Mars missions.

If you go with a just blister with EE and blister with QD plate to establish a 2 point axial linkage capable of handling a low accel for prop settling. SpaceX could implement their own implementation of the EE and manufacture the total passive targets for adding onto the Tanker and the other BEO SS. Mass penalties for the 2 blisters may be under 2t and possibly under 1t. It will require significantly more mass for the conformal solar arrays and heat rejection panels.

It may be possible to design the depot SS system design after a few iterative versions to last longer than 10 years of near continuous operations every day -> 365 tanker prop transfer ops per year. Which equals 45 BEO mission in 1 year with 1 Depot.

Like I mentioned earlier. There may be too many options for us to determine how SpaceX would decided to solve the problem until we see actual Depot SS rolling out of one of the High Bays.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/01/2022 07:25 am
The CANADARM2 has dexterity of 360 degree of manipulation of rotations in all 3 dimensions and the 10 or more meter of displacement in 3 dimensions. A QD with cryo piping will do good to have a 10 degree rotations in 2 dimensions as well as displacements capability of at best <5m in 1 dimension and significantly less than 5 m, more likely less than 2 m in the other 2 dimensions. A QD arm would work only when things are close to perfect between 2 free flyer multi 100t objects.

You want a system that can handle off nominal situations and still reach useful results. This is to be a system that has an ability to work 99+% of the time not <85% of the time. A higher dexterity arm without the cryo piping is a solution that increases the reliability of the cryo prop transfer success. It is because of the flexibility of the cryo piping that is the problem, and solely the problem that has to be solved when added to a mating plate that reaches out to attach to the Tanker's QD plate. Even though this QD plate on Tanker could be a special plate with smaller connectors and the resulting smaller piping it still causes high movement restrictions in the needed 6 DOF to perform a hard capture to be able to do the prop transfer without most of cryo leaking out of the connectors when the orientation between the 2 vehicles moves around during the low accel needed to do prop settling to be successful in transferring any prop. The forces needed by the structural components of the capture mechanism and prop arm would have to be able to exert multiple 10s of tons metric between the 2 vehicles. A simple note is that CANADARM and CANADARM2 were designed to be able to accomplish these force levels.

Another item here is that the CANADRAM(and 2 as well) does not latch onto a target object but is more appropriate to categorize it as lassoing the target. Then pulling on the "rope" to establish a tight interface of two shaped faces to create a situation of high forces capable capture due to the very high fit without having to have any latches at all. Since there are no latches and the lasso is manipulateable such that it can go from tight to totally loose. It is a very ingenious system with very very high ability to handle situations that are far from ideal. Such as lassoing a significant moving and rotating target. Try that with a arm with cryo piping.
Before adding a whole new arm system, the question needs to be asked if it is solving an actual problem. For a high range of motion arm to be necessary over a small range of motion arm (the QD arm), there would need to be a situation where Starship could make with the high range arm but not the low range one. That would be a failure mode where:

- both Starships have sufficient manoeuvring capability to rendezvous
- both Starships have sufficient manoeuvring capability to stationkeep relative to each other in close enough proximity for the arm to extend between them
- both Starships lack sufficient manoeuvring capability to stationkeep relative to each other in the desired orientation

That seems like an extremely unlikely failure mode: if you have a functioning RCs system able to stationkeep in a poor orientation, you have a functioning RCS capable of stationkeeping in the correct orientation.

Canandarm2 is dextrous because it needs to perform a whole range of other tasks beyond berthing, but it does not require that flexibility for berthing.

And the end effectors does achieve a hard mate. That's how the electrical connections are made between the arm and the target. The same system that drives the wires to catch the 'bobble' on the target continue to drive to retract those wires down into the effector. This pulls the tip of the effector up against the base of the target, which contains alignment features. These features ensure that the end effector at the end of its drive is in an exact alignment with the target, at which point the hard mating features (the connector boxes) can then extend and latch without any additional alignment movements. The attached GIF shows this in action without a target present (so the features can be seen in motion).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: JamesH65 on 06/01/2022 11:59 am
If the two SS's are linked together (I imagine extending devices at top and bottom to they can link top to top), surely just a simple hydraulic ram system (space rated) pushing a QD plate from one ship to the other requires very little displacement capability. A couple of meters of throw means a pipe that extend that distance in to the depot, but simple (!) O rings should keep things liquid tight. The rams could even be the fuel pipe, using the fuel as hydraulic fluid to push the pipe out, and suck it back again.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 06/01/2022 03:00 pm
With cryogenic couplings, you don’t really use o-rings but soft metal seals that deform under pressure slightly to keep it sealed. Same idea basically.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/01/2022 06:17 pm
That seems like an extremely unlikely failure mode: if you have a functioning RCs system able to stationkeep in a poor orientation, you have a functioning RCS capable of stationkeeping in the correct orientation.

Two things:

1) You're dealing with a very wide range of total mass and moments of inertia, anywhere from a nearly empty payload Starship that should arrive in LEO with about 250t of wet mass, up to a fully-loaded depot with ~1600t of wet mass.  There's nothing to say that SpaceX can't put multiple kinds of thrusters with different minimum impulse bits on the spacecraft, or zillions of small thrusters so the impulses scale smoothly, but that adds complexity.  At the very least, there are some RCS design challenges here that don't exist on your ordinary spacecraft.

2) The most likely failure mode for the system is after it's mated/docked/whatever, and there's an ullage thruster failure on one of the two ships.  It's a small acceleration (1cm/s²?) but the coupling needs to support whatever torques are generated for however long it takes to recover from the failure.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/01/2022 06:53 pm
If the two SS's are linked together (I imagine extending devices at top and bottom to they can link top to top), surely just a simple hydraulic ram system (space rated) pushing a QD plate from one ship to the other requires very little displacement capability. A couple of meters of throw means a pipe that extend that distance in to the depot, but simple (!) O rings should keep things liquid tight. The rams could even be the fuel pipe, using the fuel as hydraulic fluid to push the pipe out, and suck it back again.
With a long throw the corrugated plumbing faces challenges. If it's designed to be stowed compressed and stretched out for deployment a large diameter works best. If stowed as a pigtail or an S shape it works best with a small diameter. There's some relationship between diameter and stretch ability and between diameter and radius of bend. A fast google shows promise but no hits yet.


Why a full 2m throw? I can see some small 3 axis translation and rotation for that final bit of alignment. That only calls for small forces on the bellows plumbing with alignment pins providing motive force. A blister or thick gender bending adapter on the depot offers promise of doing away with extension.


With careful positioning of connectors on the QD plate there is a possibility of using fluid pressure to extend the plate. I like twofers. Unfortunately it can't retract using this method. In terms of pressure, space sucks, so suction won't retract anything. Space is sucking just as hard in the other direction.


I was a fan of a depot side QD plate that roughly mimics the GSE hookup but a trip or two down in the weeds and extending the plate started looking like a bad idea if there's any way around it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/01/2022 07:49 pm
That seems like an extremely unlikely failure mode: if you have a functioning RCs system able to stationkeep in a poor orientation, you have a functioning RCS capable of stationkeeping in the correct orientation.

Two things:

1) You're dealing with a very wide range of total mass and moments of inertia, anywhere from a nearly empty payload Starship that should arrive in LEO with about 250t of wet mass, up to a fully-loaded depot with ~1600t of wet mass.  There's nothing to say that SpaceX can't put multiple kinds of thrusters with different minimum impulse bits on the spacecraft, or zillions of small thrusters so the impulses scale smoothly, but that adds complexity.  At the very least, there are some RCS design challenges here that don't exist on your ordinary spacecraft.

2) The most likely failure mode for the system is after it's mated/docked/whatever, and there's an ullage thruster failure on one of the two ships.  It's a small acceleration (1cm/s²?) but the coupling needs to support whatever torques are generated for however long it takes to recover from the failure.
Sounds like you're talking about the QD plate being a load bearing part of the lashup. You argued so eloquently against that you convinced me. It's hard enough changing my mind once, now I gotta do it again?  :(
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/01/2022 09:36 pm
Sounds like you're talking about the QD plate being a load bearing part of the lashup. You argued so eloquently against that you convinced me. It's hard enough changing my mind once, now I gotta do it again?  :(

No, that wasn't what I was arguing.  I was arguing that simple grappling isn't enough.

I think that I'm convinced that bellows plumbing will work fine, although I still don't know where you'd put it, and whether it implies some kind of active/passive dichotomy.  I am definitely not convinced that it can do anything other than be plumbing.  I still think you need a rigid connection to support the plumbing.

I still have a prejudice for a soft dock to initiate the formation of that rigid connection, but I guess grappling is a possibility as well.  But snaking some kind of QD-seeking gizmo out to form the soft connection still strikes me as full-blown bat-guano crazy, as does transmitting any loads through it at all.

One other thing about making the QD to double duty:  It's down in the skirt, right?  So if you're going to hook something to it from another ship, that something also has to be in the skirt, nestled amongst all the other stuff in the thrust structure and engine plumbing.  That seems... unlikely to me.

I guess we should take the pictures that show two Starship aligned back-to-back with multiple grains of salt.  Tankers/depots could have docking and plumbing hardware mounted above the LCH4 tank (i.e., in the lower part of the payload bay), which could still reach out to grab stuff around the tail-mounted QD on another ship.  That could still allow androgynous active/passive connections between tankers, but payload Starships would be gendered.  Given that payload Starships (either EDL-capable or LSS) will likely have less mass and lower y-z-axis moments of inertia, maybe a rigid connection around/near the QD wouldn't be so bad.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/01/2022 11:39 pm
Sounds like you're talking about the QD plate being a load bearing part of the lashup. You argued so eloquently against that you convinced me. It's hard enough changing my mind once, now I gotta do it again?  :(

No, that wasn't what I was arguing.  I was arguing that simple grappling isn't enough.

I think that I'm convinced that bellows plumbing will work fine, although I still don't know where you'd put it, and whether it implies some kind of active/passive dichotomy.  I am definitely not convinced that it can do anything other than be plumbing.  I still think you need a rigid connection to support the plumbing.

I still have a prejudice for a soft dock to initiate the formation of that rigid connection, but I guess grappling is a possibility as well.  But snaking some kind of QD-seeking gizmo out to form the soft connection still strikes me as full-blown bat-guano crazy, as does transmitting any loads through it at all.

One other thing about making the QD to double duty:  It's down in the skirt, right?  So if you're going to hook something to it from another ship, that something also has to be in the skirt, nestled amongst all the other stuff in the thrust structure and engine plumbing.  That seems... unlikely to me.

I guess we should take the pictures that show two Starship aligned back-to-back with multiple grains of salt.  Tankers/depots could have docking and plumbing hardware mounted above the LCH4 tank (i.e., in the lower part of the payload bay), which could still reach out to grab stuff around the tail-mounted QD on another ship.  That could still allow androgynous active/passive connections between tankers, but payload Starships would be gendered.  Given that payload Starships (either EDL-capable or LSS) will likely have less mass and lower y-z-axis moments of inertia, maybe a rigid connection around/near the QD wouldn't be so bad.
I think we're seeing the same problems and coming up with completely different approaches. ISTM you're also more focused on near term and I'm looking further out and envisioning where I think the depot will, through force of circumstances, naturally evolve to. A phrase that drives me nutso, but seems to fit: Its all good.


When Old Atlas Eguy brought up Canadarm, a lot of things dropped into place for me. The two ships really need to be firmly connected, which means making highly controlled contact in some manner. That's exactly what the arm is designed for. The arm sounds like an extravagance at first hearing but some variant on it offers so many advantages it's an eye opener.


An arm can; deploy a reflector, deploy PV and manhandle the ships into alignment under a wide range of off optimal conditions. Musk's development process looks for OTS solutions whenever possible and the arm, while not exactly OTS, is developed technology. He'll probably roll his own but it's too handy to ignore. He'll also not do it right off but will look for alternatives and eventually realize it actually simplified things. Boy, am I going out on a limb here.


As Old Atlas said, it has to work 99+% or better. IMO the depot is THE key piece for all trans GEO traffic and looking at everything it will have to do to fulfill its mission calls for a ship rivaling a HSF variant in cost and complexity. That calls for it to be robust enough for long term use. I just can't see it as a cheap throwaway.


An arm also gives so many options for firmly connecting the ships that firm connection can be taken as a given with specific details TBD by playing with it.


Back to the QD plate, an arm offers a way to easily deploy a gender bender adapter. All ships would have the standard QD plate and using the vertical layout that (I think) you suggested, the depot could deploy the gender bender stowed in its meager cargo space and present the same interface to all ships that they expect from GSE.


The gender bender need not be just a flat plate. It can be a flat plate atop a plumbing stack of whatever depth is dictated by safe separation of the two ships. The 'base' QD plate on the depot, or possibly the gender bender on the extension, would probably have a bit of float to allow that last bit of alignment pin movement. Wherever the 'float' is, it will be backed by a bellows that will never face extreme movement. The fluid couplings themselves would be would be squeezed tight and the hard connection(s) between ships will be elsewhere.


Woff! I need to catch my breath.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 06/02/2022 02:50 am
I like it!!

I see several advantages as well.

If a gender bender fails it can be swapped with a second one and everything then proceeds again. Redundancy through on orbit R&R.

The arm deploys out of a larger enough opening that it and the gender benders will fit through. Which could be setup so that it could all be returned back into the bay so that the Depot could do an EDL. This would occur when the depot can no longer support its primary function. Which would then turn the depot into looking like a Tanker so another depot just arrived can then receive the excess prop prior to the old Depot does the EDL.

Thanks this gives quite a few advantages with one being that the Depot is mostly that of a Tanker with possibly stretched tanks and a small cargo door. Other items are conformal radiators and solar arrays. The other options is having installable roll out solar array and heat radiator making the vehicle even more generic like a Tanker.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: groknull on 06/02/2022 05:05 am
The adapter plate and arms need not return to Earth.

If a depot tanker needs to do EDL, the plate and arms could be handed off to any available Starship, and in turn handed off to the next available depot tanker.

Adding some Starlink derived tech (solar arrays, batteries, thrusters, control systems, and basic communications) could turn the plate into a free flyer.  If a depot tanker needs to do EDL, the free flyer adapter plate does station keeping until the next depot tanker arrives (with another adapter plate and arms).

This functionality could be iterated.  Initial versions being "dumb" plates with arms.  Later versions could add large solar arrays and radiators.  This would allow operations without every ship needing to deploy and stow solar arrays and radiators.

Later versions could also be used for BEO operations, providing redundant power and cooling for transiting Starships.  At the destination, the free flyer plate could detach and go into orbit for other uses.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 02:42 pm
Why would the depot do EDL? Shedding the tiles, fins and header tanks would be the only simplification it would face. Groknull, one thing a free flying 'adapter' pack would need that you didn't mention is tankage for repositioning propellant. Alternatively, it could have electric thrusters. Still, I don't see the point.


If it can support two or more transfer campaigns it will probably be a break even. Support three or more and it's a win. As Old Atlas pointed out the potentially high wear gender bender plate can be replaceable. The arms end effector is a mechanical weak point that begs for an 'easy replace' design. If the arm can deploy PV, radiators and sunshield, it can replace them. If batteries, plumbing or wiring go south, it's dead. The free flyer would have to duplicate many system common to all spacecraft while adding operational complexity. What advantage does it give?


With a clutch of depots scattered about at a useful inclination they can precess same as StarLink to cluster for a multi ship mars mission, for example. Then scatter for less demanding business. With electric thrusters they can do plane change. None of this repositioning would be fast. Planning ahead is SOP for orbital ops. Worst case, launch more depots. They'll be useful at some point.


With a depot rich LEO no one depot would have a lot of work in any time frame but chances are that if one or more is needed, one or more will be handy.


This is obviously long term thinking. It points up the advantage of designing and using the depot as a long term facility. Short term, I expect to see a tanker with minimum extra to support one propellant transfer as proof of concept. After that each try it will become more sophisticated and complex until we see something along the lines I envision.


Of course those wily engineering devils at SX might find an approach that no human could possibly foresee. They seem to do that with great regularity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 06/02/2022 03:42 pm
So when propellant is transfered between ships it forces the fixed stuff(the 2 ships) the other direction. So this is like a settling thruster.

So putting ships butt to butt or butt to nose or etc. would allow the movement of propellant to supply the needed settling thrust. As long as the ship losing propellant is "in front of" the propellant movement.

Does this work?
Sorry for the poor explanation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 03:55 pm
I like it!!

I see several advantages as well.

If a gender bender fails it can be swapped with a second one and everything then proceeds again. Redundancy through on orbit R&R.

The arm deploys out of a larger enough opening that it and the gender benders will fit through. Which could be setup so that it could all be returned back into the bay so that the Depot could do an EDL. This would occur when the depot can no longer support its primary function. Which would then turn the depot into looking like a Tanker so another depot just arrived can then receive the excess prop prior to the old Depot does the EDL.

Thanks this gives quite a few advantages with one being that the Depot is mostly that of a Tanker with possibly stretched tanks and a small cargo door. Other items are conformal radiators and solar arrays. The other options is having installable roll out solar array and heat radiator making the vehicle even more generic like a Tanker.
I'm not sure why EDL is important. Think of a depot as a facility, not a ship. With no EDL enabling fins, there's a natural stowage space for an arm. If there's room for a gender bender and stand-off too, great. If not, store it up above the top dome like cargo or in a surface blister if the linear connector array is narrow enough. Some type of inter-ship struts need kind of skinny storage too, if used. PV, and radiators can be conformal but I suspect conformal will not be adequate if active cooling is needed. Sunshade, if needed can't be conformal.


Early marks may get away without all the deployed panel type stuff and active cooling but ISTM it will all be needed for the long term storage Artemus calls for. This would be the model for a true facility. The only wastage would be more raptors than needed.


In the end ISTM that a depot will need too many doodads; special mounting points, PV, arm grapple points, etc, that a plane Jane tanker wouldn't have, and while possible to trick out all tankers with this stuff, it doesn't seem practical.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 04:12 pm
So when propellant is transfered between ships it forces the fixed stuff(the 2 ships) the other direction. So this is like a settling thruster.

So putting ships butt to butt or butt to nose or etc. would allow the movement of propellant to supply the needed settling thrust. As long as the ship losing propellant is "in front of" the propellant movement.

Does this work?
Sorry for the poor explanation.
No, no. your explanation is clear. ISTM, with absolutely no experience transferring cryo fluids in microgravity, that it would work up to the point the transferred propellant hits the bottom of the receiving tank. At that point it would exactly counter whatever impulse it got from the pumping. I think.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/02/2022 04:24 pm
So when propellant is transfered between ships it forces the fixed stuff(the 2 ships) the other direction. So this is like a settling thruster.

So putting ships butt to butt or butt to nose or etc. would allow the movement of propellant to supply the needed settling thrust. As long as the ship losing propellant is "in front of" the propellant movement.

Does this work?
Sorry for the poor explanation.
No, no. your explanation is clear. ISTM, with absolutely no experience transferring cryo fluids in microgravity, that it would work up to the point the transferred propellant hits the bottom of the receiving tank. At that point it would exactly counter whatever impulse it got from the pumping. I think.
The center of mass of the system as a whole does not change. Therefore, if the propellant moves in one direction, the non-propellant portion of the mass (the ships) must move in the opposite direction. This movement will start with acceleration, which might act to settle the propellant.. However it will end with deceleration as the motion eventually stops with no net change in the velocity of the center of mass. This deceleration is not helping to settle the propellant unless you somehow flip the system end-for-end, which does not seem practical.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/02/2022 06:53 pm
With electric thrusters they can do plane change. None of this repositioning would be fast.

Update:  I dropped a factor of 1000 in my mass flow computation.  The times are unacceptably long.

I hadn't thought about electric propulsion on a depot.  Let's crank that one through.

Figure an average 5º plane change for each lunar mission.  Δv=2*v*sin(Δi/2).  With 500km orbital speed at 7610m/s, that's Δv=664m/s.

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.000041kg/s, and burn time becomes 37.3hours 1553days.

You're probably in eclipse 40% of the time, so that's really 62hours 2588days.  It's not terrible.  Better than I thought it would be.

Worst-case is probably a 20º plane change, which would be Δv=2640m/s, krypton=23.6t, and 268h 6662d burn time.  Eleven days still isn't terrible.  [Yes it is.]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/02/2022 07:51 pm

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.041kg/s, and burn time becomes 37.3hours.

Something is wrong with your numbers.
ISP of 1500s is exhaust velocity of 14715m/s.  With 5500kg of prop 1/2mv2 gives  6E11 J.  10kW for 37.3 hours is 1.3E9 J, short by several orders of magnitude.

Quote
Better than I thought it would be.
If it seems too good to be true, it usually is.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/02/2022 07:57 pm
I'm warming to this free-flyer, especially in Old Atlas's incarnation where it's essentially a kit that turns an ordinary EDL-capable lift tanker into a depot.  But let's put everything that's not needed by a lift tanker on the free-flyer.  It would need:

1) A few kW of solar panels.

2) A krypton tank with enough capacity to manage 3000m/s of delta-v for the free-flyer only.  Assuming the free-flyer has a dry mass of 20t, that would be 4.5t of krypton.

3) Enough electric thrusters to move the free-flyer from orbit to orbit in a few days.  Note that these could also be used to provide ullage thrust to the attached Starships much more efficiently than the Starship RCS thrusters.  (This also allows the minimum impulse bit for the Starship RCS thrusters to be substantially larger, which is good for everything except ullage thrust.)

4) Grappling arm(s) to initiate soft capture.

5) Hard-dock points to glue it onto both the depot (semi-permanently) and the target Starship (during prop transfer). Note that this makes the hard points on the Starships passive, and tiny.

6) A male-male pair of QD plates, allowing connection to the two female QD plates on the Starships.

7) Put the cryocoolers on the free flyer, pumping prop through the cryocooler and returning it, subcooled, to the main tanks of the depot.

8) Radiators for the cryocooler.

9) Sun shade (probably not for LEO, but handy in some cislunar orbits).

10) Extra credit:  You can build a version of this with an IDSS-compliant active-active docking adapter, for use on the Gateway or even between two passive lunar Starship hatches.  This would also allow you to put the docking port somewhere other than the nose, which will be required when you want to have crews launching and returning via EDL, where the nose LOX header tank is needed.

Assuming that you make all your lift tankers with 1500-1600t prop capacity (by moving the domes forward to eat the entire cylindrical payload bay), now any lift tanker can be turned into a depot for as long as you want, for as many missions as you want.  You can launch an LT into any orbit you want and have the free-flyer meet it there, using its electric propulsion.  That LT now becomes the depot and accumulates prop from other LTs for the specific mission.  When the mission is over, the LT goes home if that's more economical than having the free-flyer drag it to the next set of orbital parameters.  Then the free flyer, with dramatically reduced mass, heads off to its next job.

Note that the free-flyer is now a full-up deployable spacecraft (albeit a very funny-lookin' one), launched on a Starship with big payload doors.  You can start with one, and add them as needed to the constellation.  They should have fairly long lifetimes, but you can dispose of them whenever they're getting worn out and deploy a new one.

Update:  Due to my 3-orders-of-magnitude burn time error (see up-thread), using electric propulsion for this is problematic.  You likely have to use methox to move the free-flyer around, and that makes its economics considerably dicier.  I'll crank it through when I'm done licking my wounds.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/02/2022 08:10 pm

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.041kg/s, and burn time becomes 37.3hours.

Something is wrong with your numbers.
ISP of 1500s is exhaust velocity of 14715m/s.  With 5500kg of prop 1/2mv2 gives  6E11 J.  10kW for 37.3 hours is 1.3E9 J, short by several orders of magnitude.

Quote
Better than I thought it would be.
If it seems too good to be true, it usually is.

Yup, you're right.  I used 600N, rather than 600mN.  Never mind!

So you can't use the free-flyer to move the Starship.  However, per my follow-on post, where I drank all of the free-flyer-as-a-lift-tanker-kit Kool-Aid, with a delta-v of 670m/s to move a 20t-dry free-flyer from one orbit to the next, you'd need 930kg of krypton.  At 0.6N of thrust, (you never go wrong with core SI units!), that's 264 days.

Bummer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 06/02/2022 08:26 pm

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.041kg/s, and burn time becomes 37.3hours.

Something is wrong with your numbers.
ISP of 1500s is exhaust velocity of 14715m/s.  With 5500kg of prop 1/2mv2 gives  6E11 J.  10kW for 37.3 hours is 1.3E9 J, short by several orders of magnitude.

Quote
Better than I thought it would be.
If it seems too good to be true, it usually is.

Yup, you're right.  I used 600N, rather than 600mN.  Never mind!

So you can use the free-flyer to move the Starship.  However, per my follow-on post, where I drank all of the free-flyer-as-a-lift-tanker-kit Kool-Aid, with a delta-v of 670m/s to move a 20t-dry free-flyer from one orbit to the next, you'd need 930kg of krypton.  At 0.6N of thrust, (you never go wrong with core SI units!), that's 264 days.

Bummer.

So even if it doesn't have the kitchen sink on it, a free flying adapter still has merit.
Minimal needs:
1. Grappling arms for connecting ships. Batteries and solar panels to keep charged.
2. QD mechanisms like we see on the ground. Two of them.
3. RCS of some sort to stabilize. Can be recharged when connected to a ship. Probably N2 gas.

It can be passed around and stay in orbit. Ideally never leave in orbit without a ship nearby.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/02/2022 08:34 pm
So even if it doesn't have the kitchen sink on it, a free flying adapter still has merit.
Minimal needs:
1. Grappling arms for connecting ships. Batteries and solar panels to keep charged.
2. QD mechanisms like we see on the ground. Two of them.
3. RCS of some sort to stabilize. Can be recharged when connected to a ship. Probably N2 gas.

It can be passed around and stay in orbit. Ideally never leave in orbit without a ship nearby.

That doesn't solve the orbit-changing problem.  We'd discussed this somewhere way, way up-thread, but if you fix the inclination of a depot, then you have fairly sparse, narrow windows for lunar departures, unless you want the departing LSS to spend a lot of extra delta-v doing plane changes as part of its TLI burn.

I was hoping that the kitchen-sink free-flyer solved most of this problem, because launching a lift tanker to the proper orbit is considerably easier than moving a depot.  But that means that the free-flyer has to... you know, free-fly to meet up with the LT in the new orbit.

I still think there's a pony in here somewhere, if for no other reason than hauling 10-20t of free-flyer through 600-700m/s of delta-v is a lot easier than hauling 100-140t of empty depot through the same change.  But it's not as magical as I thought.

PS:
Quote
3. RCS of some sort to stabilize. Can be recharged when connected to a ship. Probably N2 gas.

You're going to have lots of power if you're providing cryocooling services to the depot.  Pressure-fed methox thrusters are pretty simple, and you can just store a few tonnes of methalox, then use the power to vaporize and pump it into COPVs when you're not doing cooling.

And allowing the Starship RCS impulse bit to get a lot bigger if it doesn't have to do ullage acceleration makes the thruster design a lot simpler.  Not sure if electric ullage acceleration will work, but it's worth looking at--even if you have to provide big delta-v's via methox.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 06/02/2022 08:50 pm
surely just a simple hydraulic ram system (space rated) pushing a QD plate from one ship to the other requires very little displacement capability. A couple of meters of throw means a pipe that extend that distance in to the depot

A couple of metres seems massively excessive for the actual propellant connectors. IMO, you'd want to bring the visiting ship as close to the depot as you can. Almost scraping hulls. But, this is only required after capture, so long as you are separating the free-flight capture/grapple/docking step from the final propellant mating.

The grapples/grabber/softdock/whatever can have a larger stand-off, and those would be used to pull the visiting ship in close after capture. In essence, you'd be pulling the ship into the fuel connector, not pushing the connector out to the ship. The depot-side of the fuel connectors needs to stand proud of the hull, but not by much, and only needs to deploy once and then remain locked (other than for maintenance.)



A male/male adaptor ("gender bender") could be added to provide the stand-off distance needed for the propellant connectors on the depot-side, but you then need a way to attach it, which means robot-arms or the convoluted "free-flyer", both of which seems un-SpaceX-like for early iterations, hence for Artemis.

However, it's possible that a male/male adaptor could fold out over the depot's version of the female QD plate. That way the plumbing length/angles are fixed on the QD plate, the swing-arm that positions the adaptor doesn't need to carry propellant, and the adaptor itself is rigidly connected once latched. Eliminates the need for flexible plumbing, and it should be possible to tuck all the required parts into the skirt for launch with relatively simple deployment.



Aside: I noticed that most renders of the HLS docked to a depot or other Starship place the propellant connectors amidships. We've all assumed any propellant connectors at the aft, either using the QD plates, or near them.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 06/02/2022 09:01 pm
Re: Robot-arms.

Seems like overkill for the early depot, hence for Artemis at all. And, IMO, a berthing process that requires a working robot-arm seems to be higher risk than a more limited type of capture system. Also, IMO, the ship-to-depot docking/capture system has to work without a robot-arm, even if you later intend to add one. The added risk of a robot-arm failure might be unacceptable to NASA without a backup method of refuelling.

So:

Two (or at least two) grapple end-effectors deployed from a linear system contained within the depot, fore and aft. The visiting ships would have two (or more) fold-out grapple fixtures. (The latter have virtually no mass, and hardly any internal volume requirements.) Ship and depot need prox-ops detection and ranging systems (cameras/radar/reflectors/etc) but that is the case for any system.

The end-effector at the nose of the depot would have the greatest flexibility, but nowhere near as much as a robot-arm, just 4 axes with limited range. The z-axis extending linearly from the depot, x/y-axes pivoting 20 degrees or so from perpendicular, and simple rotation around the z-axis. Any final errors are taken up by the flexibility inherent in the soft-capture wires in the end-effector.

The tail end-effector(s) would have much more limited z-axis reach, and no x/y pivot nor z-rotation other than that inherent in the end-effector itself.

The reason for putting the greatest range of motion in the nose grapple is two-fold: 1/ even with extended tanks, hence limited nose volume, you have more internal volume for the hardware needed to give that extra range of motion compared to the skirt. And 2/ it means the visiting ship can approach the depot at an angle, nose-towards-nose, keeping the tails further apart before capture, reducing the risk of collision. After hard-capture, the nose end-effector can pivot the visiting ship into place to "dock" with the tail end-effector(s). Once the tail is also locked, both the fore/aft end-effectors retract on their z-axes, drawing the ship it slowly towards the depot, until their fuel-connectors mate. And the male QD plate has guide-rods and latches for the final precision required.

[edit: having more flexibility in the nose-nose capture system also means you can move the departing ship further away from the depot, and its tail even further from the depot, before releasing it. Reducing the risk of collision after release, as well.]



By the second generation of depots, I would have at least two berthing locations, preferably three, so you can simultaneously dock two/three visiting ships. (Four seems possible without bumping fins, but three adds margin-of-safety.) That offers a bunch of extra functionality, especially after adding a robot-arm. For example, moving cargo/resupply from an Earth-launched Starship into a BEO/Lunar Starship. Moving an unfuelled deep-space probe from the Earth-launched Starship to a second berthing location, where it can also take on propellant for its BEO injection. Or even mate a payload (such as a probe) to a cheap generic expendable BEO-booster stage (based on a single Raptor, for eg), which was previously delivered to the depot separately.

With multiple berthing points, the depot can also evolve into a station, possibly even manned. Any activities that don't mind vibration and potential external contamination can use the station. (Or a number of stations, since you'll likely have a few different depots in different planes, for different destinations/timings.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/02/2022 09:26 pm
Another free-flying depot kit thought:  You could use the kit to cover the dorsal side of a lift tanker in MLI.  Roll the MLI down after attaching to the lift tanker that's acting as the depot, the roll it back up before detaching.

Anybody have an educated guess for what the equilibrium temperature of the TPS system would be when facing Earth's daylit side, and facing cold space in eclipse?  You've got high emissivity, but fairly low temperature (compared to reentry).  And the tiles aren't really reflective; it's just that their absorptance is low.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 10:38 pm
With electric thrusters they can do plane change. None of this repositioning would be fast.

Update:  I dropped a factor of 1000 in my mass flow computation.  The times are unacceptably long.

I hadn't thought about electric propulsion on a depot.  Let's crank that one through.

Figure an average 5º plane change for each lunar mission.  Δv=2*v*sin(Δi/2).  With 500km orbital speed at 7610m/s, that's Δv=664m/s.

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.000041kg/s, and burn time becomes 37.3hours 1553days.

You're probably in eclipse 40% of the time, so that's really 62hours 2588days.  It's not terrible.  Better than I thought it would be.

Worst-case is probably a 20º plane change, which would be Δv=2640m/s, krypton=23.6t, and 268h 6662d burn time.  Eleven days still isn't terrible.  [Yes it is.]
10kW seems reasonable but also arbitrary. If more power helps, do more power.


I know you've updated your figures but here's a question. Whats the dV to raise to an arbitrary 50,000km? Double that to recircularize. While at 50,000 km what would it take to do an arbitrary 5deg plane change? How does it compare to doing it in LEO? For 20deg? Perigee works best for Oberth assist in raising Perigee, and perigee works best for plane changes.


This applies to a full blooded depot or a strap on kit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 11:04 pm
I'm warming to this free-flyer, especially in Old Atlas's incarnation where it's essentially a kit that turns an ordinary EDL-capable lift tanker into a depot.  But let's put everything that's not needed by a lift tanker on the free-flyer.  It would need:

1) A few kW of solar panels.

2) A krypton tank with enough capacity to manage 3000m/s of delta-v for the free-flyer only.  Assuming the free-flyer has a dry mass of 20t, that would be 4.5t of krypton.

3) Enough electric thrusters to move the free-flyer from orbit to orbit in a few days.  Note that these could also be used to provide ullage thrust to the attached Starships much more efficiently than the Starship RCS thrusters.  (This also allows the minimum impulse bit for the Starship RCS thrusters to be substantially larger, which is good for everything except ullage thrust.)

4) Grappling arm(s) to initiate soft capture.

5) Hard-dock points to glue it onto both the depot (semi-permanently) and the target Starship (during prop transfer). Note that this makes the hard points on the Starships passive, and tiny.

6) A male-male pair of QD plates, allowing connection to the two female QD plates on the Starships.

7) Put the cryocoolers on the free flyer, pumping prop through the cryocooler and returning it, subcooled, to the main tanks of the depot.

 8) Radiators for the cryocooler.

9) Sun shade (probably not for LEO, but handy in some cislunar orbits).

10) Extra credit:  You can build a version of this with an IDSS-compliant active-active docking adapter, for use on the Gateway or even between two passive lunar Starship hatches.  This would also allow you to put the docking port somewhere other than the nose, which will be required when you want to have crews launching and returning via EDL, where the nose LOX header tank is needed.

Assuming that you make all your lift tankers with 1500-1600t prop capacity (by moving the domes forward to eat the entire cylindrical payload bay), now any lift tanker can be turned into a depot for as long as you want, for as many missions as you want.  You can launch an LT into any orbit you want and have the free-flyer meet it there, using its electric propulsion.  That LT now becomes the depot and accumulates prop from other LTs for the specific mission.  When the mission is over, the LT goes home if that's more economical than having the free-flyer drag it to the next set of orbital parameters.  Then the free flyer, with dramatically reduced mass, heads off to its next job.

Note that the free-flyer is now a full-up deployable spacecraft (albeit a very funny-lookin' one), launched on a Starship with big payload doors.  You can start with one, and add them as needed to the constellation.  They should have fairly long lifetimes, but you can dispose of them whenever they're getting worn out and deploy a new one.

Update:  Due to my 3-orders-of-magnitude burn time error (see up-thread), using electric propulsion for this is problematic.  You likely have to use methox to move the free-flyer around, and that makes its economics considerably dicier.  I'll crank it through when I'm done licking my wounds.
Hmmm. This is interesting. Technically as doable as a full depot and maybe worthwhile. Got any extra koolaid?


Changing orbits in a few days seems awfully demanding. Less thrust at a higher exhaust velocity gets more dV for any given propellant mass. Or is there some other problem that raises its ugly head? While depots or kits are a rare thing it'll all be choreographed well ahead of time. When there are more on orbit they'll be closer to hand. 


Something to keep in mind is that dV budget is only one of the many moving parts. Accepting inefficiencies here may add other efficiencies elsewhere.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 11:11 pm

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.041kg/s, and burn time becomes 37.3hours.

Something is wrong with your numbers.
ISP of 1500s is exhaust velocity of 14715m/s.  With 5500kg of prop 1/2mv2 gives  6E11 J.  10kW for 37.3 hours is 1.3E9 J, short by several orders of magnitude.

Quote
Better than I thought it would be.
If it seems too good to be true, it usually is.

Yup, you're right.  I used 600N, rather than 600mN.  Never mind!

So you can't use the free-flyer to move the Starship.  However, per my follow-on post, where I drank all of the free-flyer-as-a-lift-tanker-kit Kool-Aid, with a delta-v of 670m/s to move a 20t-dry free-flyer from one orbit to the next, you'd need 930kg of krypton.  At 0.6N of thrust, (you never go wrong with core SI units!), that's 264 days.

Bummer.
Not totally unworkable. Use the kit or depot for lunar launches then eight months later to round out a mars campaign. As more get to orbit and populate the most popular lanes, it gets easier.


Side thought. Except for the QD area most of a strap on kit can be lattice. Maybe carbon fiber.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: ulm_atms on 06/02/2022 11:13 pm
With electric thrusters they can do plane change. None of this repositioning would be fast.

Update:  I dropped a factor of 1000 in my mass flow computation.  The times are unacceptably long.

I hadn't thought about electric propulsion on a depot.  Let's crank that one through.

Figure an average 5º plane change for each lunar mission.  Δv=2*v*sin(Δi/2).  With 500km orbital speed at 7610m/s, that's Δv=664m/s.

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.000041kg/s, and burn time becomes 37.3hours 1553days.

You're probably in eclipse 40% of the time, so that's really 62hours 2588days.  It's not terrible.  Better than I thought it would be.

Worst-case is probably a 20º plane change, which would be Δv=2640m/s, krypton=23.6t, and 268h 6662d burn time.  Eleven days still isn't terrible.  [Yes it is.]
10kW seems reasonable but also arbitrary. If more power helps, do more power.
It's not reasonable...it's way too low.

https://phys.org/news/2018-02-x3-ion-thruster-propel-mars.html

EDIT:Woops...part of my reply got cut off.  The X3 is up to 200kW
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 11:14 pm

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.041kg/s, and burn time becomes 37.3hours.

Something is wrong with your numbers.
ISP of 1500s is exhaust velocity of 14715m/s.  With 5500kg of prop 1/2mv2 gives  6E11 J.  10kW for 37.3 hours is 1.3E9 J, short by several orders of magnitude.

Quote
Better than I thought it would be.
If it seems too good to be true, it usually is.

Yup, you're right.  I used 600N, rather than 600mN.  Never mind!

So you can use the free-flyer to move the Starship.  However, per my follow-on post, where I drank all of the free-flyer-as-a-lift-tanker-kit Kool-Aid, with a delta-v of 670m/s to move a 20t-dry free-flyer from one orbit to the next, you'd need 930kg of krypton.  At 0.6N of thrust, (you never go wrong with core SI units!), that's 264 days.

Bummer.

So even if it doesn't have the kitchen sink on it, a free flying adapter still has merit.
Minimal needs:
1. Grappling arms for connecting ships. Batteries and solar panels to keep charged.
2. QD mechanisms like we see on the ground. Two of them.
3. RCS of some sort to stabilize. Can be recharged when connected to a ship. Probably N2 gas.

It can be passed around and stay in orbit. Ideally never leave in orbit without a ship nearby.
Use reaction wheels or CMG's like the ISS for attitude control. No propellant needed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 11:18 pm
Re: Robot-arms.

Seems like overkill for the early depot, hence for Artemis at all. And, IMO, a berthing process that requires a working robot-arm seems to be higher risk than a more limited type of capture system. Also, IMO, the ship-to-depot docking/capture system has to work without a robot-arm, even if you later intend to add one. The added risk of a robot-arm failure might be unacceptable to NASA without a backup method of refuelling.

So:

Two (or at least two) grapple end-effectors deployed from a linear system contained within the depot, fore and aft. The visiting ships would have two (or more) fold-out grapple fixtures. (The latter have virtually no mass, and hardly any internal volume requirements.) Ship and depot need prox-ops detection and ranging systems (cameras/radar/reflectors/etc) but that is the case for any system.

The end-effector at the nose of the depot would have the greatest flexibility, but nowhere near as much as a robot-arm, just 4 axes with limited range. The z-axis extending linearly from the depot, x/y-axes pivoting 20 degrees or so from perpendicular, and simple rotation around the z-axis. Any final errors are taken up by the flexibility inherent in the soft-capture wires in the end-effector.

The tail end-effector(s) would have much more limited z-axis reach, and no x/y pivot nor z-rotation other than that inherent in the end-effector itself.

The reason for putting the greatest range of motion in the nose grapple is two-fold: 1/ even with extended tanks, hence limited nose volume, you have more internal volume for the hardware needed to give that extra range of motion compared to the skirt. And 2/ it means the visiting ship can approach the depot at an angle, nose-towards-nose, keeping the tails further apart before capture, reducing the risk of collision. After hard-capture, the nose end-effector can pivot the visiting ship into place to "dock" with the tail end-effector(s). Once the tail is also locked, both the fore/aft end-effectors retract on their z-axes, drawing the ship it slowly towards the depot, until their fuel-connectors mate. And the male QD plate has guide-rods and latches for the final precision required.

[edit: having more flexibility in the nose-nose capture system also means you can move the departing ship further away from the depot, and its tail even further from the depot, before releasing it. Reducing the risk of collision after release, as well.]



By the second generation of depots, I would have at least two berthing locations, preferably three, so you can simultaneously dock two/three visiting ships. (Four seems possible without bumping fins, but three adds margin-of-safety.) That offers a bunch of extra functionality, especially after adding a robot-arm. For example, moving cargo/resupply from an Earth-launched Starship into a BEO/Lunar Starship. Moving an unfuelled deep-space probe from the Earth-launched Starship to a second berthing location, where it can also take on propellant for its BEO injection. Or even mate a payload (such as a probe) to a cheap generic expendable BEO-booster stage (based on a single Raptor, for eg), which was previously delivered to the depot separately.

With multiple berthing points, the depot can also evolve into a station, possibly even manned. Any activities that don't mind vibration and potential external contamination can use the station. (Or a number of stations, since you'll likely have a few different depots in different planes, for different destinations/timings.)
Any way you slice it the ships need to hook up. What you described is essentially an arm variant. If it works early on, it might be fine long term.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/02/2022 11:27 pm
With electric thrusters they can do plane change. None of this repositioning would be fast.

Update:  I dropped a factor of 1000 in my mass flow computation.  The times are unacceptably long.

I hadn't thought about electric propulsion on a depot.  Let's crank that one through.

Figure an average 5º plane change for each lunar mission.  Δv=2*v*sin(Δi/2).  With 500km orbital speed at 7610m/s, that's Δv=664m/s.

Let's use dry mass of 120t for the depot, once you add in all the stuff you're talking about (arms, deployable docking and piping, solar panels, insulation, etc.).  With krypton Isp=1500s, that's 5.5t of krypton.

But wait!  There's more!  What are you figuring for available power?  10kW seems excessive, but let's use that.  You get about 60mN/kW with most Hall thrusters.  It's probably less for krypton, but again, let's be optimistic.  That gives thrust = 600mN, so mdot = 0.000041kg/s, and burn time becomes 37.3hours 1553days.

You're probably in eclipse 40% of the time, so that's really 62hours 2588days.  It's not terrible.  Better than I thought it would be.

Worst-case is probably a 20º plane change, which would be Δv=2640m/s, krypton=23.6t, and 268h 6662d burn time.  Eleven days still isn't terrible.  [Yes it is.]
10kW seems reasonable but also arbitrary. If more power helps, do more power.
It's not reasonable...it's way too low.

https://phys.org/news/2018-02-x3-ion-thruster-propel-mars.html (https://phys.org/news/2018-02-x3-ion-thruster-propel-mars.html)

EDIT:Woops...part of my reply got cut off.  The X3 is up to 200kW
That's a lot of juice. The old ISS panels put out around 100kW. Not sure about the new ones.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nomadd on 06/03/2022 12:12 am

That's a lot of juice. The old ISS panels put out around 100kW. Not sure about the new ones.
The old panels put out about 240kw when they were new. The system, including half the time in shadow, averaged about 110kw.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 06/03/2022 02:52 am
[...]
What you described is essentially an arm variant.

Then I described it badly.

What I'm trying to suggest is more like the very last section of the Canadarm. Just one "wrist" joint, except more limited mobility, plus the end-effector without the four power/data latches. The second (aft) grapple would lack even the wrist joint.

Massively simplified compared to Canadarm or the ERA.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 06/03/2022 11:18 am
A couple of things:

Instead of having a free flyer do its orbit change itself. It might be simpler to bring it down to earth on the last ship and launch on the next mission. This free flyer needs to be simple as possible.

When I said
Quote
1. Grappling arms for connecting ships. Batteries and solar panels to keep charged.
2. QD mechanisms like we see on the ground. Two of them.
3. RCS of some sort to stabilize. Can be recharged when connected to a ship. Probably N2 gas.
Grappling could be real simple. Like paul says. The end effectors of the ISS arm.
Electric will need to be used grappling and moving the QD connection.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 06/03/2022 12:13 pm
I'm warming to this free-flyer, especially in Old Atlas's incarnation where it's essentially a kit that turns an ordinary EDL-capable lift tanker into a depot.  But let's put everything that's not needed by a lift tanker on the free-flyer.  It would need:

1) A few kW of solar panels.

2) A krypton tank with enough capacity to manage 3000m/s of delta-v for the free-flyer only.  Assuming the free-flyer has a dry mass of 20t, that would be 4.5t of krypton.

3) Enough electric thrusters to move the free-flyer from orbit to orbit in a few days.  Note that these could also be used to provide ullage thrust to the attached Starships much more efficiently than the Starship RCS thrusters.  (This also allows the minimum impulse bit for the Starship RCS thrusters to be substantially larger, which is good for everything except ullage thrust.)

4) Grappling arm(s) to initiate soft capture.

5) Hard-dock points to glue it onto both the depot (semi-permanently) and the target Starship (during prop transfer). Note that this makes the hard points on the Starships passive, and tiny.

6) A male-male pair of QD plates, allowing connection to the two female QD plates on the Starships.

7) Put the cryocoolers on the free flyer, pumping prop through the cryocooler and returning it, subcooled, to the main tanks of the depot.

8) Radiators for the cryocooler.

9) Sun shade (probably not for LEO, but handy in some cislunar orbits).

10) Extra credit:  You can build a version of this with an IDSS-compliant active-active docking adapter, for use on the Gateway or even between two passive lunar Starship hatches.  This would also allow you to put the docking port somewhere other than the nose, which will be required when you want to have crews launching and returning via EDL, where the nose LOX header tank is needed.

Assuming that you make all your lift tankers with 1500-1600t prop capacity (by moving the domes forward to eat the entire cylindrical payload bay), now any lift tanker can be turned into a depot for as long as you want, for as many missions as you want.  You can launch an LT into any orbit you want and have the free-flyer meet it there, using its electric propulsion.  That LT now becomes the depot and accumulates prop from other LTs for the specific mission.  When the mission is over, the LT goes home if that's more economical than having the free-flyer drag it to the next set of orbital parameters.  Then the free flyer, with dramatically reduced mass, heads off to its next job.

Note that the free-flyer is now a full-up deployable spacecraft (albeit a very funny-lookin' one), launched on a Starship with big payload doors.  You can start with one, and add them as needed to the constellation.  They should have fairly long lifetimes, but you can dispose of them whenever they're getting worn out and deploy a new one.

Update:  Due to my 3-orders-of-magnitude burn time error (see up-thread), using electric propulsion for this is problematic.  You likely have to use methox to move the free-flyer around, and that makes its economics considerably dicier.  I'll crank it through when I'm done licking my wounds.
Lose the electric thrusters. You're already hooked into 6 bar tankage, so tapping off some for cold gas thrusters loses some (lots) ISP, but lets it resupply it's own supply in the course of regular operation.

In the alternative, operationally declare that the strapon kit will never fly unaccompanied in orbit, and have it rely on it's arms and the systems the tanker already has for stability, orbit changes, and so on.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/03/2022 04:36 pm
[...]
What you described is essentially an arm variant.

Then I described it badly.

What I'm trying to suggest is more like the very last section of the Canadarm. Just one "wrist" joint, except more limited mobility, plus the end-effector without the four power/data latches. The second (aft) grapple would lack even the wrist joint.

Massively simplified compared to Canadarm or the ERA.
Your description was clear. Much simplified from a full set of arms and designed for a specific task rather than general purpose. But still a a variant in my book. Terminology again. Maybe we should call them Waldo's.  8)


Words are a two edged sword. So powerful, yet they all too often set unintentional limits to a concept. I now think of Pluto as not-a-planet which means the concept of planet is still somehow bound to it in my mind. Anybody born after the reclassification won't have quite the same connections.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/03/2022 07:13 pm
It's not reasonable...it's way too low.

https://phys.org/news/2018-02-x3-ion-thruster-propel-mars.html

The thrusters aren't the problem, the solar panels are. 200kW worth are gonna be pretty unwieldy, especially for a kit for a lift tanker.  That's like... 550m²?

On the other hand, at 200kW, 60mN/kW, Isp=1500s, and Δv=700m/s, moving a kit weighing 20t of dry mass requires 975kg of Kr, which will take about 23 days to burn with 40% eclipse time.  Not terrible (unless of course I dropped three orders of magnitude again).

Also, I doubt this works for providing the ullage acceleration.  200kW @ 60mN/kW = 12N.  A tanker adding the last load to an otherwise-full depot will have a combined mass of close to 1800t.  12N of force gives you 0.007mm/s² acceleration.  ULA has looked at accelerations as low as 1E-5g, which would be about 0.1mm/s².  That's not the end of the world, but it does mean that you lose the advantages of being able to design Starship thrusters with larger minimum impulse bits.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/03/2022 07:53 pm
What I'm trying to suggest is more like the very last section of the Canadarm. Just one "wrist" joint, except more limited mobility, plus the end-effector without the four power/data latches. The second (aft) grapple would lack even the wrist joint.

Massively simplified compared to Canadarm or the ERA.

Let's break this down into three steps:

1) Getting the end effector to cover the grapple fixture on the target Starship well enough that the snares can grab it.

2) Nulling out residual translation and rotation rates.

3) Translating, rotating, and retracting to align the Starship for hard dock.

#1 isn't very hard, as long as the residuals are modest.  #2 I think can be handled with a "wrist", as you proposed, as long as the arm is capable of dealing with the residual torques.  But I can't see how you can simply retract the arm without an additional "shoulder" and "elbow" to align for hard dock.  You can't just yank on the grappled Starship and expect it to bump against stuff until it slides into place.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/04/2022 03:46 pm
It's not reasonable...it's way too low.

https://phys.org/news/2018-02-x3-ion-thruster-propel-mars.html (https://phys.org/news/2018-02-x3-ion-thruster-propel-mars.html)

The thrusters aren't the problem, the solar panels are. 200kW worth are gonna be pretty unwieldy, especially for a kit for a lift tanker.  That's like... 550m²?

On the other hand, at 200kW, 60mN/kW, Isp=1500s, and Δv=700m/s, moving a kit weighing 20t of dry mass requires 975kg of Kr, which will take about 23 days to burn with 40% eclipse time.  Not terrible (unless of course I dropped three orders of magnitude again).

Also, I doubt this works for providing the ullage acceleration.  200kW @ 60mN/kW = 12N.  A tanker adding the last load to an otherwise-full depot will have a combined mass of close to 1800t.  12N of force gives you 0.007mm/s² acceleration.  ULA has looked at accelerations as low as 1E-5g, which would be about 0.1mm/s².  That's not the end of the world, but it does mean that you lose the advantages of being able to design Starship thrusters with larger minimum impulse bits.
Maybe not as big a problem as it looks. Changing orbit, plane or height, calls for periodic, not continuous burns. More battery, less panel. If electric propulsion fits in well, it just has to be accepted that changing orbit takes the depot/kit out of service for awhile. If that's not acceptable then it's gotta be methalox or launch another one.


For settling props, a very low thrust minimizes slosh. Hypothetically, once the ship has moved one tank length it ready to start transfer. To make it just a smidge more appealing, only props that have settled need be accelerated. Any propellant free floating in the tank stays where it is until the tank catches up with it.


ISTM 12 hours, or multiples of 12 hours between tanker deliveries is to be expected until they have more pads. If they can hook up, settle props, transfer and clear the area in 12 hours, they're doing fine. Later on, when multiple ships need propellant transfer at one time, they'll need to step up the pace.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/04/2022 04:30 pm

For settling props, a very low thrust minimizes slosh. Hypothetically, once the ship has moved one tank length it ready to start transfer. To make it just a smidge more appealing, only props that have settled need be accelerated. Any propellant free floating in the tank stays where it is until the tank catches up with it.

Hypothetically there is an acceleration so low that viscosity and surface tension win and no settling occurs.  1g for six months just about settles small amounts of molasses but does not settle tahini.

Do we have any non-hypothetical data (Either detailed modelling or actual experiments) on just what acceleration is needed for settling?  It's all very well to assume 0.01 m s-2 or 0.0001 m s-2 will work, but that's quite a range.  It seems the first order of business is to run the experiment.

Is there something like a Renolds number that relates time, density, linear scale and other fluid properties to determine settling?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/04/2022 05:15 pm

For settling props, a very low thrust minimizes slosh. Hypothetically, once the ship has moved one tank length it ready to start transfer. To make it just a smidge more appealing, only props that have settled need be accelerated. Any propellant free floating in the tank stays where it is until the tank catches up with it.

Hypothetically there is an acceleration so low that viscosity and surface tension win and no settling occurs.  1g for six months just about settles small amounts of molasses but does not settle tahini.
https://en.wikipedia.org/wiki/Pitch_drop_experiment
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/05/2022 12:53 am
I'm still not convinced developing an entire new vehicle is really the fastest, easiest, or cheapest way to connect a fluid coupler to the existing fluid coupler. Particularly as the existing example of fluid coupling in orbit does not require more than sufficient alignment for docking (Progress), the existing demonstration of 100-tonne-class objects docking (STS Orbiter to ISS) did not require any arms or other apparatus beyond RCS thrust, and SpaceX have demonstrated software control for precision docking. None of this seem to indicate that connecting two Starships requires anything more than the RCS thrust for sufficient precision, nor any coupling methods more complex than the existing QD plate.
Do we have any non-hypothetical data (Either detailed modelling or actual experiments) on just what acceleration is needed for settling?  It's all very well to assume 0.01 m s-2 or 0.0001 m s-2 will work, but that's quite a range.  It seems the first order of business is to run the experiment.
Centaur has tested hyodrolox propellant settling down to microgee levels (so ~0.001 ms-2) in orbit for long duration coasts.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: meekGee on 06/05/2022 02:45 am

For settling props, a very low thrust minimizes slosh. Hypothetically, once the ship has moved one tank length it ready to start transfer. To make it just a smidge more appealing, only props that have settled need be accelerated. Any propellant free floating in the tank stays where it is until the tank catches up with it.

Hypothetically there is an acceleration so low that viscosity and surface tension win and no settling occurs.  1g for six months just about settles small amounts of molasses but does not settle tahini.

Do we have any non-hypothetical data (Either detailed modelling or actual experiments) on just what acceleration is needed for settling?  It's all very well to assume 0.01 m s-2 or 0.0001 m s-2 will work, but that's quite a range.  It seems the first order of business is to run the experiment.

Is there something like a Renolds number that relates time, density, linear scale and other fluid properties to determine settling?
Haha neither Tahini nor molasses are Newtonian...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/05/2022 05:49 pm

Is there something like a Renolds number that relates time, density, linear scale and other fluid properties to determine settling?
Haha neither Tahini nor molasses are Newtonian...

No fluid is Newtonian.  "Newtonian" is just the simplest model that accounts for viscosity in any way.  For many fluids Newtonian is good enough for practical calculations under ordinary conditions*.  The question is do "ordinary conditions" and "good enough" extend across six orders of magnitude of acceleration?  Are we extrapolating or interpolating?

Newtonian fluids also do not account for surface tension and other properties that result in small forces that must become more important at low accelerations.  Fluid dynamics is full of weird effects that can show up as soon as you extend the conditions.

My interpretation of the current state of the art:  Early flights gave unexpected results.  The engineering investigations concentrated on finding specific solutions rather than a general understanding.  The solutions for Saturn, Atlas, Centaur etc. do not address the specific conditions for a refueling.  I expect the first refueling attempts to encounter novel problems which will not be insurmountable but will require surprising solutions.

Another example:
Very naively look at the Bond Number of liquid methane under 1 micro-gee.  You can "expect" droplets of liquid methane on the walls that are 500 times the diameter of the water droplets clinging to your shower curtain.  That's not much of a problem once you restart the main engine, it is a problem if you want to transfer all the fuel using 1 micro-gee ullage.

*Then again ignoring viscosity completely is good enough for some fluids under some conditions.  Also compare Newtonian gravity with general relativity.  Usually Newton is good enough, but sometimes it's not, sometimes gravity is negligible and can be ignored.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/05/2022 10:55 pm
Here's a quote from one of the old ULA depot papers that kinda blew my hair back:

Quote
Similarly, settling allows venting during propellant
transfer to maintain pressure in the receiver tank at
desired levels. With extremely low acceleration,
propellant entering the receiver tank may geyser. To
prevent liquid venting, the propellant transfer process may
need to be accomplished in pulse mode, where propellant
transfer and venting are conducted sequentially.

It's a fun problem, this refueling thing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/06/2022 12:39 am
I'm still not convinced developing an entire new vehicle is really the fastest, easiest, or cheapest way to connect a fluid coupler to the existing fluid coupler. Particularly as the existing example of fluid coupling in orbit does not require more than sufficient alignment for docking (Progress), the existing demonstration of 100-tonne-class objects docking (STS Orbiter to ISS) did not require any arms or other apparatus beyond RCS thrust, and SpaceX have demonstrated software control for precision docking. None of this seem to indicate that connecting two Starships requires anything more than the RCS thrust for sufficient precision, nor any coupling methods more complex than the existing QD plate.
Do we have any non-hypothetical data (Either detailed modelling or actual experiments) on just what acceleration is needed for settling?  It's all very well to assume 0.01 m s-2 or 0.0001 m s-2 will work, but that's quite a range.  It seems the first order of business is to run the experiment.
Centaur has tested hyodrolox propellant settling down to microgee levels (so ~0.001 ms-2) in orbit for long duration coasts.
At its heart a depot is a big tank, just like a tanker. From that starting point one design criteria after another intruded on the simplicity.


- Assuming belly to belly stays the plan, even residual torque will be a lot of torque. Is it easier or less expensive to have additional connection points or reinforce the area around the QD to take the load? I don't know and accept that either could work but don't know where optimal resides.
- Again assuming belly to belly, it looks like the depot will need either a mirror connector or an adapter plate. If a mirror connection, this alone makes the depot a special build. If an adapter plate, it calls for either a one off deployment solution or a general purpose (aka, arm) solution.
- Again assuming belly to belly, there needs to be some sort of standoff to keep the two ships from becoming too intimate. Perhaps there's a solution other than a blister with the mirror QD on it or an adapter of some depth but I don't know what it is. A depot based extendable QD plate is conceivable but ISTM that this 1) rules out the QD also being a structural member, 2) adds a degree of complexity that would introduce too much risk of malfunction in a system that must work multiple times.
- While 'growing the system' active cooling may not be needed. Later, ISTM it's unavoidable, if only for Artemus. For LEO use it makes sense long term. Without active cooling it's all too easy to picture a full depot boiling off on orbit during a weather or technical hold on he ship its waiting for.
- With active cooling comes robust power needs. Panels and batteries.


SX will surely (opinion) run test flights looking for their classic just good enough solutions. If it were only one or two things they could well find them but it's too many things pointing to a the Depot being a variant in its own right. OTOH, they may pull off a stunner and leave me babbling happily into my beer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/06/2022 04:22 am
- Again assuming belly to belly, it looks like the depot will need either a mirror connector or an adapter plate. If a mirror connection, this alone makes the depot a special build. If an adapter plate, it calls for either a one off deployment solution or a general purpose (aka, arm) solution.

The depot doesn't have to be a special build if the QD is not the pathway through which on-orbit refueling occurs.  There can be a second connector on all Starships that's either androgynous or hermaphroditic.  That connector can be part of or associated with the docking/berthing mechanism.

For payload Starships, that adds some parasitic weight to the spacecraft that isn't needed if no refueling is required.  I'd expect that the connectors and docking mechansim would be removable for Starships that don't need it, similar to how the D2 can remove the nose dock when it's not needed.

One other thought on the depot:  If you go all-in on the cryocooler and you have the separate connector mechanism installed (or installable) on all Starships, then most of the stuff you need could simply be a payload that's added to the vestigial payload bay of a lift tanker.  Note that tankers are almost certain to be different from payload Starships, because they really need to hold 1500t-1600t of prop to make the LSS able to do LEO-NRHO-LS-NRHO on one tank of prop.  The LSS itself needs more prop capacity as well.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/06/2022 05:04 pm
- Again assuming belly to belly, it looks like the depot will need either a mirror connector or an adapter plate. If a mirror connection, this alone makes the depot a special build. If an adapter plate, it calls for either a one off deployment solution or a general purpose (aka, arm) solution.

The depot doesn't have to be a special build if the QD is not the pathway through which on-orbit refueling occurs.  There can be a second connector on all Starships that's either androgynous or hermaphroditic.  That connector can be part of or associated with the docking/berthing mechanism.

For payload Starships, that adds some parasitic weight to the spacecraft that isn't needed if no refueling is required.  I'd expect that the connectors and docking mechansim would be removable for Starships that don't need it, similar to how the D2 can remove the nose dock when it's not needed.

One other thought on the depot:  If you go all-in on the cryocooler and you have the separate connector mechanism installed (or installable) on all Starships, then most of the stuff you need could simply be a payload that's added to the vestigial payload bay of a lift tanker.  Note that tankers are almost certain to be different from payload Starships, because they really need to hold 1500t-1600t of prop to make the LSS able to do LEO-NRHO-LS-NRHO on one tank of prop.  The LSS itself needs more prop capacity as well.
Mass is less a constraint for the SS family than any other rocket out there BUT, to duplicate all the plumbing, valving and QD connections on all ships is the mass curse that just keeps giving.


Even if the connections themselves were easily removable back to a valve, there would still be some mass in the plumbing and that valve. Worse yet, unlike the D2 docking collar that fits under the nose cap, a blanking plate would probably be needed and the concept of rapid turnaround takes a ding.


I totally agree that the tankers will be their own variant. Whatever kit the depot or free flyer needs can come up in whatever way makes most sense. Unlike earlier speculation that a launched depot would hit orbit with its own cargo of propellant, a kitted out depot would sacrifice some or all if this.


I'm trying to envision what a free flyer might look like. What I come up with is a boxy structure faced on two sides with QD connectors then a lattice structure extending up the length of the depot/tanker to make a tight connection between ships up high. There might be ullage settling thrusters in the box and maybe something for repositioning. Arms and/or wrists would sprout off the lattice. There might be some lateral lattice to support solar panels and radiators.


The strongest single objection I have to a free flyer is that any tanker in the temporary role of depot would have full heatshield and fins. This throws repositioning out the window. If a free flyer can be broken down and stored for return and reuse in another orbit, or free fly to another orbit this makes sense. If not, and it's in a little used orbit, it becomes a throwaway. Maybe that's a worthwhile trade off.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/06/2022 06:14 pm
I think it is far simpler to eliminate as many variants as possible, whilst making additional mass/hardware an addon rather than being built in or moved to a brand new vehicle:
- Cargo Starship - has a payload bay, carries payload in that bay, may need to receive propellant in LEO for high-mass high-orbit missions.
- Tanker Starship - a Cargo Starship carrying no payload, and a stock of additional ullage gas (or propellant boilers to generate gas on orbit) to force propellant back out of the tanks. No further mods, no extended tanks. Transfers propellants to Depot Starship only, never directly to a Cargo or HLS Starship.
- Depot Starship - a Cargo Starship whose payload bay is occupied with propellant transfer hardware: a copy of the 'squid' portion of the QD arm and QD plate, pressurisation equipment (may be propellant boilers, or just tanks of ullage pressurant) to force fluid from the main tanks across the transfer link, and a cryocooler. May have had its tiles and flaps stripped off (or not installed) and more appropriate insulation applied, or may be otherwise stock and returnable. May dock 'upside down' belly to belly with tankers to simplify fluid draining.
- HLS Starship - a two-off (or more-off if additional contracts are won) extremely modified variant on the Starship platform. methods and engines but otherwise

This means:
- The additional mass borne by all non-Depot Starship variants is the addition of capped-off taps on the fluid plumbing and pressurisation lines between the main and nose-header tanks to allow for propellant tap-off, and possibly additional valving (pinned shut for any ship not operating as a Depot) at the 'top' of fluid line runs out of the main tanks to minimise loss during prop transfer.
- Tankers are 'inefficient' due to the empty nose volume, but can be compensated by launching additional tankers rather than producing a bayless Starship variant. Rocket equation works in their favour, payload is exchanged 1:1 for residual propellant.
- Depot can be recoverable for single-transfer missions where a Cargo Starship needs more propellant than it can carry, but does not require the depot to loiter in LEO for a prolonged period.
- Any given Starship can become a tanker or short-duration Depot without carrying the mass penalty of doing so for its entire life.
- A 'long duration' Depot requires modification for insulation that precludes recovery, so would be desirable only for the few missions where a long coast to accumulate many propellant loads and to wait for the target vehicle is really necessary. If you have the capability to launch fast enough, a long duration depot may not be necessary.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/06/2022 08:17 pm
I think it is far simpler to eliminate as many variants as possible, whilst making additional mass/hardware an addon rather than being built in or moved to a brand new vehicle:
- Cargo Starship - has a payload bay, carries payload in that bay, may need to receive propellant in LEO for high-mass high-orbit missions.

So far so good.

Quote
- Tanker Starship - a Cargo Starship carrying no payload, and a stock of additional ullage gas (or propellant boilers to generate gas on orbit) to force propellant back out of the tanks. No further mods, no extended tanks. Transfers propellants to Depot Starship only, never directly to a Cargo or HLS Starship.

If the only job a tanker had to perform was as a lift tanker, this would work--but it's not.  At the very least, unless you're willing to expend a depot on every Artemis mission after Arty III, you need a tanker to transfer prop from LEO to NRHO, then either transfer it directly to the LSS or to a semi-permanent depot, which then hangs onto it until just before the mission and then transfers it to the LSS.

To do that in one flight also requires a 1500t-capacity tanker.  You can reduce that by 150t if you're willing to fly a BLT with the tanker, but even then you need more than 1200t.

Also, there's very little penalty to increasing the prop capacity on the tanker.  You're not stretching the outer mould line; you're simply moving the common dome and LCH4 top dome forward.  That should be zero GSE change and a very modest structural change:  you just change the stacking order of cylindrical rings and dome assemblies a bit.  There are obviously wiring and plumbing changes as well, but that's about it.

You don't need to launch a 1500t-1600t tanker completely full if it messes up the T/W of the SH/SS system (and it absolutely would).  As long as you only launch with whatever the cargo Starship is rated (1200t right now), you're not changing the structural hydrostatics of the tankage.  You just need the extra tankage to be completely full on-orbit, prior to LEO-NRHO.  Then the tanker can return directly to EDL.

On extra ullage gas:  You're almost certainly better off with an electric pump and a line to equalize ullage pressures between the two tanks.  If you slosh and uncover the outlet with a pump, it's no big deal for a low-power pump.  If you do the same thing with a pressure-fed system, you're toast.  Also, the pump will weigh a lot less than the ullage gas.

Quote
- Depot Starship - a Cargo Starship whose payload bay is occupied with propellant transfer hardware: a copy of the 'squid' portion of the QD arm and QD plate, pressurisation equipment (may be propellant boilers, or just tanks of ullage pressurant) to force fluid from the main tanks across the transfer link, and a cryocooler. May have had its tiles and flaps stripped off (or not installed) and more appropriate insulation applied, or may be otherwise stock and returnable. May dock 'upside down' belly to belly with tankers to simplify fluid draining.

Again, 1200t doesn't do it.  You need at least 1500t.

I think we've beaten the connector issue to death, but I still think an active QD plate with a squid (nice name) is massive overkill.

One other issue that we haven't really discussed as much:  Is it better to spend a lot of time figuring out how to do insulation for a depot (where "insulation" probably means MLI or SOFI, vs. just painting the stupid thing Solar White), or can you just rely on a cryocooler with enough capacity to get you to near-zero boiloff? 

I suspect that the answer to that choice lies with how much power you need to run the cryocooler, and whether that's easily deployable on the depot.  IMO, this would be a major advantage of a free-flyer, because it can deploy however much solar array area it needs, and then route the prop to be subcooled from the tanks, through the cryocooler, and back into the tanks.

Quote
- HLS Starship - a two-off (or more-off if additional contracts are won) extremely modified variant on the Starship platform. methods and engines but otherwise

Certainly more modified, but if you've gone to 1500t tanks and are willing to make do with only a couple of meters of cargo bay (which is plenty for the airlocks and whatever cargo a crewed mission would need to deploy), the LSS looks an awful lot like a tanker with a crew system crammed into the nose.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: shark0302 on 06/06/2022 08:36 pm
So dumb thought. 
Could you have the tanker and depot meet up so that they are off nose to tail. So that instead of the qd plates meet at the same end one essentially is flying tail first, and the other nose first. The you end up doing a  pressurized propellant transfer, maybe unless you have dual pick ups at the top of the tank for the on flying nose first it its the tanker. now you have reasons for two qd plates and both can used for both hard dock and propellant transfer. If you have to have a part that is stupid tough why not make it do more than one thing. Also this qd at the nose would always be male vs the tail end being female.

Sent from my SM-T860 using Tapatalk

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/06/2022 08:58 pm
I'm trying to envision what a free flyer might look like. What I come up with is a boxy structure faced on two sides with QD connectors then a lattice structure extending up the length of the depot/tanker to make a tight connection between ships up high. There might be ullage settling thrusters in the box and maybe something for repositioning. Arms and/or wrists would sprout off the lattice. There might be some lateral lattice to support solar panels and radiators.

Attached is an arm-wave at what I was thinking.

Quote
The strongest single objection I have to a free flyer is that any tanker in the temporary role of depot would have full heatshield and fins. This throws repositioning out the window. If a free flyer can be broken down and stored for return and reuse in another orbit, or free fly to another orbit this makes sense. If not, and it's in a little used orbit, it becomes a throwaway. Maybe that's a worthwhile trade off.

The idea is that you'd launch the tanker-to-be-depot directly to the target orbit, then the free-flyer would change orbit to meet it.  I'd been budgeting a 20t dry mass, but I suspect that's way too much.  The heaviest objects in my picture are the tankage and thrusters; I doubt the whole thing would have a dry mass more than 2-3t.  Assuming pressure-fed methalox (or methox) thrusters with Isp=340s, 700m/s of Δv would require about 700kg of prop, which would be taken from the depot before leaving for its next job.  (Note that I've completely given up on SEP.)

The tanker-that-is-the-depot could either stay in the target orbit if there was more traffic to be handled, but my guess is that it'll almost always be cheaper to de-orbit to EDL, then launch a different tanker-to-be-depot to the new target orbit.

NOTE:  If you can somehow manage to stuff all of the free-flyer hardware into a volume that will fit into the ogive portion of the tanker's remaining payload bay, that's better than a free flyer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/06/2022 08:59 pm
So dumb thought. 
Could you have the tanker and depot meet up so that they are off nose to tail. So that instead of the qd plates meet at the same end one essentially is flying tail first, and the other nose first. The you end up doing a  pressurized propellant transfer, maybe unless you have dual pick ups at the top of the tank for the on flying nose first it its the tanker. now you have reasons for two qd plates and both can used for both hard dock and propellant transfer. If you have to have a part that is stupid tough why not make it do more than one thing. Also this qd at the nose would always be male vs the tail end being female.
Back to basics. Depot must be able to transfer fuel to HLS and to the ships of the Mars fleet. Your design must allow for this. It must also transfer fuel from the tankers and possibly to the tankers. Make sure you account for all of these transfers. Note that the actual Starship tail is occupied by Raptors, so you must figure out exactly where the connection will occur. for all ships with a tail connection. Note that the HLS probably has crew cocking hatch in the nose. Note that it is probably challenging to have a nose port on EDL-capable ships including tankers.

(Minor: nose-to-tail dock has both ships going forward or both ships going backward, not as you stated).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: biosehnsucht on 06/06/2022 09:22 pm
A thought:

If we assume depots never return home, then we can make a lot of changes for convenience:
- Instead of trying to use a free flyer or arm that can reach outside the payload bay around any doors, simply use F9 style fairings that get jettisoned ASAP during launch (probably don't even bother to make them recoverable at first, until you're putting up a lot of depots, and perhaps not even then)
- Since it's not coming back, we get to remove any heat shield tiles and flap gizmos and so forth, plus make any similar optimizations that are possible with engine arrangement (perhaps give it just enough thrust to reach LEO without a refuel, and arrive with empty tanks, if that lets us save some engines)
- With the new lack of restriction on where things stick out of the top of the Depot, rather than trying to turn the QD plate into a QD arm, give it a QD arm apparatus, separate from the QD plate for filling for launch is (which would be normal QD plate). The QD arm, like the one on Stage 0, would have the opposite gender of the QD plates. This means added mass for extra plumbing and valves, but the positional flexibility gained can probably make up for it. Plus if you are clever in how the arms are arranged, you can potentially mate multiple depots together (Depot QD arm to Depot QD plate) to build a single larger facility, without needing any extra hub parts, which might provide simpler ops having multiple starships refuel at one large depot than many starships at many depots.
- Similar to the ARM having a lot of locational flexibility in where it goes and how it sticks out, with jettisoned fairings you now have much easier time deploying solar panels and radiators, etc

With this setup, "empty" cargo starships can serve as "mini-tankers" until larger dedicated tankers are built, and the depot can refuel all types (including other depots, HLS, etc).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/06/2022 09:36 pm
A thought:

If we assume depots never return home...

There's a difference between never returning home (i.e., to Earth EDL), and never changing orbits.  Especially in LEO, orbital inclination changes are very expensive in terms of delta-v, and therefore in terms of propellant.  You also need to make RAAN changes, unless you're willing to wait for the existing RAAN to line up with your departure trajectory.  That's not the end of the world for lunar missions, but it's a bigger deal for martian ones.

It's kinda counterintuitive, but it's usually cheaper to launch directly into the orbit you need rather than maneuvering from one orbit to another.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: biosehnsucht on 06/06/2022 09:52 pm
A thought:

If we assume depots never return home...

There's a difference between never returning home (i.e., to Earth EDL), and never changing orbits.  Especially in LEO, orbital inclination changes are very expensive in terms of delta-v, and therefore in terms of propellant.  You also need to make RAAN changes, unless you're willing to wait for the existing RAAN to line up with your departure trajectory.  That's not the end of the world for lunar missions, but it's a bigger deal for martian ones.

It's kinda counterintuitive, but it's usually cheaper to launch directly into the orbit you need rather than maneuvering from one orbit to another.

With the possible exception of aerobraking into Martian orbit, hopefully no orbital change maneuvers will require a heatshield or fairing :)

Any number of engines sufficient to get it into LEO, should be enough to do on-orbit maneuvering, or maneuvering to most other orbits (i.e. the moon), you just might need to be refueled...

If we need to support aerobraking at Mars, then on those ships you could still have jettisonable fairings, just wait until later, and have a retractable door in the fairing just on those models to let the QD arm come out to refuel the depot so it can reach Mars. But it's also possible that perhaps with no payload, just fuel, you could just use regular propulsive braking to enter Mars orbit, and that might be cheaper / more reliable over all than aero braking, and you could still jettison the fairings and simplify things in Earth orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/06/2022 09:56 pm
I think it is far simpler to eliminate as many variants as possible, whilst making additional mass/hardware an addon rather than being built in or moved to a brand new vehicle:
- Cargo Starship - has a payload bay, carries payload in that bay, may need to receive propellant in LEO for high-mass high-orbit missions.
- Tanker Starship - a Cargo Starship carrying no payload, and a stock of additional ullage gas (or propellant boilers to generate gas on orbit) to force propellant back out of the tanks. No further mods, no extended tanks. Transfers propellants to Depot Starship only, never directly to a Cargo or HLS Starship.
- Depot Starship - a Cargo Starship whose payload bay is occupied with propellant transfer hardware: a copy of the 'squid' portion of the QD arm and QD plate, pressurisation equipment (may be propellant boilers, or just tanks of ullage pressurant) to force fluid from the main tanks across the transfer link, and a cryocooler. May have had its tiles and flaps stripped off (or not installed) and more appropriate insulation applied, or may be otherwise stock and returnable. May dock 'upside down' belly to belly with tankers to simplify fluid draining.
- HLS Starship - a two-off (or more-off if additional contracts are won) extremely modified variant on the Starship platform. methods and engines but otherwise

This means:
- The additional mass borne by all non-Depot Starship variants is the addition of capped-off taps on the fluid plumbing and pressurisation lines between the main and nose-header tanks to allow for propellant tap-off, and possibly additional valving (pinned shut for any ship not operating as a Depot) at the 'top' of fluid line runs out of the main tanks to minimise loss during prop transfer.
- Tankers are 'inefficient' due to the empty nose volume, but can be compensated by launching additional tankers rather than producing a bayless Starship variant. Rocket equation works in their favour, payload is exchanged 1:1 for residual propellant.
- Depot can be recoverable for single-transfer missions where a Cargo Starship needs more propellant than it can carry, but does not require the depot to loiter in LEO for a prolonged period.
- Any given Starship can become a tanker or short-duration Depot without carrying the mass penalty of doing so for its entire life.
- A 'long duration' Depot requires modification for insulation that precludes recovery, so would be desirable only for the few missions where a long coast to accumulate many propellant loads and to wait for the target vehicle is really necessary. If you have the capability to launch fast enough, a long duration depot may not be necessary.
We're potentially seeing five major variants. Crewed, Artemus, orbital cargo delivery, tanker and depot. Crewed will probably be customized per mission. Artemus is it's own special set of problems and very unlike any other variant. Orbital delivery will have sub variants like the pez dispenser and big doors. Tanker, as Rad Mod points out mostly means moving bulkheads with a good payoff for doing so. The tanker will probably be the simplest and most unchanging build. That leaves the depot.


With four other variants and a slew of sub variants, what difference does one more variant make? It's not like it's as inflexible a design as everything that's come before. Design to meet the need. Set commonality as a soft goal. Ignore it when it gets in the way.


This isn't an argument for or against a semi custom designed depot but an argument that commonality is mostly a matter of degree. Methinks Mr. Musk see ease of operations as the highest design goal second only to actually getting SS to work.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/06/2022 10:15 pm
Any number of engines sufficient to get it into LEO, should be enough to do on-orbit maneuvering, or maneuvering to most other orbits (i.e. the moon), you just might need to be refueled...

This is one where I encourage you to do the math¹.  Thrust isn't the problem.  Delta-v is.  The propellant consumed is exponentially proportional to the delta-v required to do the maneuver, and the propellant has to be launched from Earth, at least for the foreseeable future.  It is the principal expense for any mission beyond earth orbit.

__________
¹For inclination changes, Δv=2*v*sin(Δi/2), where:
v is the orbital speed.  For a 500km x 500km orbit, it's about 7610m/s.
Δi is the inclination change.  About 10º will work for almost any lunar mission from one time to the next, but you may need a bit more than that for Mars.

Then, a variation on the rocket equation:

propRequiredForManeuver = (exp(Δv/(Isp * 9.8)) - 1) * inertMass, where:
Isp = specific impulse (about 378s for a Raptor)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/06/2022 10:25 pm
Any number of engines sufficient to get it into LEO, should be enough to do on-orbit maneuvering, or maneuvering to most other orbits (i.e. the moon), you just might need to be refueled...

This is one where I encourage you to do the math¹.  Thrust isn't the problem.  Delta-v is.  The propellant consumed is exponentially proportional to the delta-v required to do the maneuver, and the propellant has to be launched from Earth, at least for the foreseeable future.  It is the principal expense for any mission beyond earth orbit.

__________
¹For inclination changes, Δv=2*v*sin(Δi/2), where:
v is the orbital speed.  For a 500km x 500km orbit, it's about 7610m/s.
Δi is the inclination change.  About 10º will work for almost any lunar mission from one time to the next, but you may need a bit more than that for Mars.

Then, a variation on the rocket equation:

propRequiredForManeuver = (exp(Δv/(Isp * 9.8)) - 1) * inertMass, where:
Isp = specific impulse (about 378s for a Raptor)
Presumably, this math eventually causes you to launch a lot of depots. Eventually, some of them will be near where you need them. With sufficient lead time you could use low-power high efficiency thrust systems to move them where you want them. Perhaps electric thrusters, or if you are very patient you may be able to use solar sails.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/06/2022 10:39 pm
Presumably, this math eventually causes you to launch a lot of depots. Eventually, some of them will be near where you need them. With sufficient lead time you could use low-power high efficiency thrust systems to move them where you want them. Perhaps electric thrusters, or if you are very patient you may be able to use solar sails.

Or you can just launch them to the orbit you need and move the kit that turns them into depots separately.  Even better, you can launch them with the kit that turns them into depots.  Since the depot will weigh at least 95t dry and the kit will weigh a couple of tonnes, this is a lot more efficient than leaving depots in orbits where they'll have very low usage cadence.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/06/2022 11:09 pm
I'm trying to envision what a free flyer might look like. What I come up with is a boxy structure faced on two sides with QD connectors then a lattice structure extending up the length of the depot/tanker to make a tight connection between ships up high. There might be ullage settling thrusters in the box and maybe something for repositioning. Arms and/or wrists would sprout off the lattice. There might be some lateral lattice to support solar panels and radiators.

Attached is an arm-wave at what I was thinking.

Quote
The strongest single objection I have to a free flyer is that any tanker in the temporary role of depot would have full heatshield and fins. This throws repositioning out the window. If a free flyer can be broken down and stored for return and reuse in another orbit, or free fly to another orbit this makes sense. If not, and it's in a little used orbit, it becomes a throwaway. Maybe that's a worthwhile trade off.

The idea is that you'd launch the tanker-to-be-depot directly to the target orbit, then the free-flyer would change orbit to meet it.  I'd been budgeting a 20t dry mass, but I suspect that's way too much.  The heaviest objects in my picture are the tankage and thrusters; I doubt the whole thing would have a dry mass more than 2-3t.  Assuming pressure-fed methalox (or methox) thrusters with Isp=340s, 700m/s of Δv would require about 700kg of prop, which would be taken from the depot before leaving for its next job.  (Note that I've completely given up on SEP.)

The tanker-that-is-the-depot could either stay in the target orbit if there was more traffic to be handled, but my guess is that it'll almost always be cheaper to de-orbit to EDL, then launch a different tanker-to-be-depot to the new target orbit.

NOTE:  If you can somehow manage to stuff all of the free-flyer hardware into a volume that will fit into the ogive portion of the tanker's remaining payload bay, that's better than a free flyer.
Nice. Length up to the top dome?


I was noodling mounting it on the tanker/depot remora style then realized it'd get in the way of GSE. Never thought about tank size. I've a hunch there's not all that much room in the ogive with flap mechanism  (is the O2 header still up there?). Looks like it might need breakdown and be a secondary load on an orbital delivery ship. That's another whole set of problems but it does give a way to get it back for relaunch.


The tanks for a free flyer are a problem. One or two small ones might fit in a tankers cargo space along with the QD box. Stow more on each tanker coming up until there are enough. Accept the ding in propellant load during that first campaign.


Conversely, If the depot is a no return the fins go away and the ogive has more room. Theres most probably room for the QD adapter box up there. An arm (that things gotta be put together) could fit where a front fin nacelle would be. Flexible PV can be wound up like a window shade or a folded fan and stuck into the rear fin nacelles.


That solar white paint will radically (but not moderately) reduce boil off. It will not stop it. At some point a cryo cooler will be needed for at least Artemus. The paint will most likely do while working the kinks out. A campaign might need one last top off tanker to make up the losses. Later on, as the system matures, add the cooler to depots in busy lanes and stop living hand to mouth. Ya got extra props? You'll use it somewhere, sometime. Maybe for repositioning.


ISTM the free flyer creates more problems than it solves. There are complexities it addresses but the only drawback to a full fledged depot is the amount of mass that has to be moved for orbit changes. That seems to be the cost.


As M. Jagger and K. Richards once said "You can't always get what you want..."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 06/06/2022 11:28 pm
A thought:

If we assume depots never return home...

There's a difference between never returning home (i.e., to Earth EDL), and never changing orbits.  Especially in LEO, orbital inclination changes are very expensive in terms of delta-v, and therefore in terms of propellant.  You also need to make RAAN changes, unless you're willing to wait for the existing RAAN to line up with your departure trajectory.  That's not the end of the world for lunar missions, but it's a bigger deal for martian ones.

It's kinda counterintuitive, but it's usually cheaper to launch directly into the orbit you need rather than maneuvering from one orbit to another.

Jon Goff's 3 burn method for depot positioning and departure alignment solves a few of those issues.

There's also the possibility of having a returning depot, or a depot that can do an aeroassist plane change, if it's still a fundamentally landable Starship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/06/2022 11:36 pm
I think it is far simpler to eliminate as many variants as possible, whilst making additional mass/hardware an addon rather than being built in or moved to a brand new vehicle:
- Cargo Starship - has a payload bay, carries payload in that bay, may need to receive propellant in LEO for high-mass high-orbit missions.

So far so good.

Quote
- Tanker Starship - a Cargo Starship carrying no payload, and a stock of additional ullage gas (or propellant boilers to generate gas on orbit) to force propellant back out of the tanks. No further mods, no extended tanks. Transfers propellants to Depot Starship only, never directly to a Cargo or HLS Starship.

If the only job a tanker had to perform was as a lift tanker, this would work--but it's not.  At the very least, unless you're willing to expend a depot on every Artemis mission after Arty III, you need a tanker to transfer prop from LEO to NRHO, then either transfer it directly to the LSS or to a semi-permanent depot, which then hangs onto it until just before the mission and then transfers it to the LSS.

To do that in one flight also requires a 1500t-capacity tanker.  You can reduce that by 150t if you're willing to fly a BLT with the tanker, but even then you need more than 1200t.

Also, there's very little penalty to increasing the prop capacity on the tanker.  You're not stretching the outer mould line; you're simply moving the common dome and LCH4 top dome forward.  That should be zero GSE change and a very modest structural change:  you just change the stacking order of cylindrical rings and dome assemblies a bit.  There are obviously wiring and plumbing changes as well, but that's about it.

You don't need to launch a 1500t-1600t tanker completely full if it messes up the T/W of the SH/SS system (and it absolutely would).  As long as you only launch with whatever the cargo Starship is rated (1200t right now), you're not changing the structural hydrostatics of the tankage.  You just need the extra tankage to be completely full on-orbit, prior to LEO-NRHO.  Then the tanker can return directly to EDL.

On extra ullage gas:  You're almost certainly better off with an electric pump and a line to equalize ullage pressures between the two tanks.  If you slosh and uncover the outlet with a pump, it's no big deal for a low-power pump.  If you do the same thing with a pressure-fed system, you're toast.  Also, the pump will weigh a lot less than the ullage gas.

Quote
- Depot Starship - a Cargo Starship whose payload bay is occupied with propellant transfer hardware: a copy of the 'squid' portion of the QD arm and QD plate, pressurisation equipment (may be propellant boilers, or just tanks of ullage pressurant) to force fluid from the main tanks across the transfer link, and a cryocooler. May have had its tiles and flaps stripped off (or not installed) and more appropriate insulation applied, or may be otherwise stock and returnable. May dock 'upside down' belly to belly with tankers to simplify fluid draining.

Again, 1200t doesn't do it.  You need at least 1500t.

I think we've beaten the connector issue to death, but I still think an active QD plate with a squid (nice name) is massive overkill.

One other issue that we haven't really discussed as much:  Is it better to spend a lot of time figuring out how to do insulation for a depot (where "insulation" probably means MLI or SOFI, vs. just painting the stupid thing Solar White), or can you just rely on a cryocooler with enough capacity to get you to near-zero boiloff? 

I suspect that the answer to that choice lies with how much power you need to run the cryocooler, and whether that's easily deployable on the depot.  IMO, this would be a major advantage of a free-flyer, because it can deploy however much solar array area it needs, and then route the prop to be subcooled from the tanks, through the cryocooler, and back into the tanks.

Quote
- HLS Starship - a two-off (or more-off if additional contracts are won) extremely modified variant on the Starship platform. methods and engines but otherwise

Certainly more modified, but if you've gone to 1500t tanks and are willing to make do with only a couple of meters of cargo bay (which is plenty for the airlocks and whatever cargo a crewed mission would need to deploy), the LSS looks an awful lot like a tanker with a crew system crammed into the nose.
Counterpoint: use two depots. Cheaper to use multiple 'small' returnable depots than a non-returnable jumbo depot.

The core purpose of Starship is to make launches cheap via multiple re-uses. If you are faced with a decision to manufacture a single-use (depot serves one orbital plane) single-purpose, does-it-all-in-one-vehicle system' or launch two generic systems, it should be pretty clear what the cheaper option is. The single-purposes vehicle (HLS Starship) is the option of last resort when you can't in any way use the regular model.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/06/2022 11:40 pm
Any number of engines sufficient to get it into LEO, should be enough to do on-orbit maneuvering, or maneuvering to most other orbits (i.e. the moon), you just might need to be refueled...

This is one where I encourage you to do the math¹.  Thrust isn't the problem.  Delta-v is.  The propellant consumed is exponentially proportional to the delta-v required to do the maneuver, and the propellant has to be launched from Earth, at least for the foreseeable future.  It is the principal expense for any mission beyond earth orbit.

__________
¹For inclination changes, Δv=2*v*sin(Δi/2), where:
v is the orbital speed.  For a 500km x 500km orbit, it's about 7610m/s.
Δi is the inclination change.  About 10º will work for almost any lunar mission from one time to the next, but you may need a bit more than that for Mars.

Then, a variation on the rocket equation:

propRequiredForManeuver = (exp(Δv/(Isp * 9. 8) ) - 1) * inertMass, where:
Isp = specific impulse (about 378s for a Raptor)
I'm truly math challenged but have a seat of the pants understanding of two body orbital mechanics. I think you're doing this the hard way.


LEO is ~8000m/s. Escape velocity is ~11,000m/s. For 3000m/s, apogee can be raised to any arbitrary height. Let's target L2 for an example and pretend we have all the time in the world (cosmos?).


The depot comes to a an asymptotic  velocity of 0 at L2. Point the depot to align with the desired plane and the mildest of puffs sends it on its way. At perigee a 3000m/s burn circularizes. Total expenditure is 6000m/s + 1 puff. This will even allow a worst case turnaround to go back the other way.


If apogee is lower the dV to get there is less and two puffs, larger than that in the example, bring it to a stop and send it on its way in a new plane. Get slick and one puff, 1.4 times the magnitude of the two puff total will do it.


This is where my math skills screw me. I don't know if there is a lower sweet apogee that minimizes total dV.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: meekGee on 06/07/2022 12:10 am

Is there something like a Renolds number that relates time, density, linear scale and other fluid properties to determine settling?
Haha neither Tahini nor molasses are Newtonian...

No fluid is Newtonian.  "Newtonian" is just the simplest model that accounts for viscosity in any way.  For many fluids Newtonian is good enough for practical calculations under ordinary conditions*.  The question is do "ordinary conditions" and "good enough" extend across six orders of magnitude of acceleration?  Are we extrapolating or interpolating?

Newtonian fluids also do not account for surface tension and other properties that result in small forces that must become more important at low accelerations.  Fluid dynamics is full of weird effects that can show up as soon as you extend the conditions.

My interpretation of the current state of the art:  Early flights gave unexpected results.  The engineering investigations concentrated on finding specific solutions rather than a general understanding.  The solutions for Saturn, Atlas, Centaur etc. do not address the specific conditions for a refueling.  I expect the first refueling attempts to encounter novel problems which will not be insurmountable but will require surprising solutions.

Another example:
Very naively look at the Bond Number of liquid methane under 1 micro-gee.  You can "expect" droplets of liquid methane on the walls that are 500 times the diameter of the water droplets clinging to your shower curtain.  That's not much of a problem once you restart the main engine, it is a problem if you want to transfer all the fuel using 1 micro-gee ullage.

*Then again ignoring viscosity completely is good enough for some fluids under some conditions.  Also compare Newtonian gravity with general relativity.  Usually Newton is good enough, but sometimes it's not, sometimes gravity is negligible and can be ignored.
Fair enough.  I just liked the introduction of Tahini into the conversation. It goes with anything!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/07/2022 04:19 am
Any number of engines sufficient to get it into LEO, should be enough to do on-orbit maneuvering, or maneuvering to most other orbits (i.e. the moon), you just might need to be refueled...

This is one where I encourage you to do the math¹.  Thrust isn't the problem.  Delta-v is.  The propellant consumed is exponentially proportional to the delta-v required to do the maneuver, and the propellant has to be launched from Earth, at least for the foreseeable future.  It is the principal expense for any mission beyond earth orbit.

__________
¹For inclination changes, Δv=2*v*sin(Δi/2), where:
v is the orbital speed.  For a 500km x 500km orbit, it's about 7610m/s.
Δi is the inclination change.  About 10º will work for almost any lunar mission from one time to the next, but you may need a bit more than that for Mars.

Then, a variation on the rocket equation:

propRequiredForManeuver = (exp(Δv/(Isp * 9. 8) ) - 1) * inertMass, where:
Isp = specific impulse (about 378s for a Raptor)
I'm truly math challenged but have a seat of the pants understanding of two body orbital mechanics. I think you're doing this the hard way.


LEO is ~8000m/s. Escape velocity is ~11,000m/s. For 3000m/s, apogee can be raised to any arbitrary height. Let's target L2 for an example and pretend we have all the time in the world (cosmos?).


The depot comes to a an asymptotic  velocity of 0 at L2. Point the depot to align with the desired plane and the mildest of puffs sends it on its way. At perigee a 3000m/s burn circularizes. Total expenditure is 6000m/s + 1 puff. This will even allow a worst case turnaround to go back the other way.


If apogee is lower the dV to get there is less and two puffs, larger than that in the example, bring it to a stop and send it on its way in a new plane. Get slick and one puff, 1.4 times the magnitude of the two puff total will do it.


This is where my math skills screw me. I don't know if there is a lower sweet apogee that minimizes total dV.

If you're moving the depot, then you have to expend 1-3km/s to boost up to the HEEO, and then 1-3km/s to drop back to LEO.  That's super-expensive.

However, you can use the 3-burn departure (what Asteroza mentioned just a bit up-thread) for cargo departures:  You refuel in LEO (500x500, let's say).  Then you boost to LEO+2500m/s (roughly GTO).  At apogee, you do your plane change, which only costs 280m/s, instead of 1330m/s for a 10º change from LEO.  Then you do a perigee departure burn.

You can boost up to any apogee you want doing this, but it takes longer and longer before you can do your final departure burn.  For interplanetary stuff, that's probably insignificant.  For lunar stuff, it can increase your trip time by 50% or more.

The increased trip time, plus the second pair of passes through the Van Allen belts, makes this problematic for crewed lunar flights.

And I don't think you need to do much orbit relocation for Mars.  If you're sending out a surge at the synodic window, you refuel a bunch of flights from a depot (or depots) at an inclination that's right for this particular synod.  Even if you have to abandon a depot every 2.2 years, that's not too bad.

However, this is mostly irrelevant for getting prop to NRHO.  Either you send it in a tanker that can do an EDL to get home cheaply, or you send it in an expendable depot.  The only way to recover a non-EDL depot from NRHO is via propulsive return, which costs 10-11 additional tankers flights--and you still wind up needing to return the tanker that brings the extra prop, so you might as well just use it and have done with it.  You can use the 3-burn trick to adjust the inclination of your tanker for NRHO insertion, but you still need the tanker.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/07/2022 04:40 am
Counterpoint: use two depots. Cheaper to use multiple 'small' returnable depots than a non-returnable jumbo depot.

This would make sense if:

1) Depots had TPS and elonerons to allow them to return straight to EDL.  Otherwise, you need still more depots to refuel them to return propulsively to LEO, and more depots to return the ones that refueled the refuelers, and so on--it doesn't work.

2) "Jumbo" depots were actually hard to design or manufacture.  But they're not.  Moving the domes forward is easy, and it almost certainly has to be done for the LSS anyway.

Quote
The core purpose of Starship is to make launches cheap via multiple re-uses. If you are faced with a decision to manufacture a single-use (depot serves one orbital plane) single-purpose, does-it-all-in-one-vehicle system' or launch two generic systems, it should be pretty clear what the cheaper option is. The single-purposes vehicle (HLS Starship) is the option of last resort when you can't in any way use the regular model.

You're presenting a false choice, though.  The third option is simply to swap cargo space that cant'/won't be used for more prop, which doesn't change any of the launch or EDL dynamics (because you never launch with more than 1200t of prop, and you return to EDL the same way as a Starship with 1200t tankage), and has all kinds of uses--including use by the LSS, which needs very little cargo space but needs the extra 300t of prop from LEO quite badly.

This does indeed require that depots are EDL-capable, so that you move them by recovering them and launching them again.  But that doesn't sound that difficult, unless they really need to be so heavily insulated that the TPS system starts working against you.  However, I'm fairly confident that brute-forcing the boil-off problems with a cryocooler will turn out to be considerably cheaper than all the hand-wringing that's necessary to get close to ZBO with only insulation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/07/2022 10:41 am
Counterpoint: use two depots. Cheaper to use multiple 'small' returnable depots than a non-returnable jumbo depot.

This would make sense if:

1) Depots had TPS and elonerons to allow them to return straight to EDL.  Otherwise, you need still more depots to refuel them to return propulsively to LEO, and more depots to return the ones that refueled the refuelers, and so on--it doesn't work.

2) "Jumbo" depots were actually hard to design or manufacture.  But they're not.  Moving the domes forward is easy, and it almost certainly has to be done for the LSS anyway.
Scenario A: you build two Cargo Starships, and fit them as Depots. You fly them for a HLS mission, fill them with propellants, load those propellants onto a HLS Starship, then return them to Earth. You then remove the depot hardware (extra arm and equipment in the cargo bay), and you now have two Cargo Starships to make money with whilst you wait for the next HLS mission. If one of them experiences any issue between now and then, you can pick another Cargo Starship to use.
Scenario B: You design and build a dedicated Jumbo Depot Starship. You fly it for a HLS mission, fill it with propellants, load those propellants onto a HLS Starship. You now have a Jumbo Depot Starship either sitting in orbit doing nothing (and possibly costing you money to actively monitor) or  returned to Earth (and taking up room at your build site) whilst you wait for the next HLS mission. If it experiences any issue between now and then you need to build a new dedicated vehicle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/07/2022 12:13 pm
Scenario A: you build two Cargo Starships, and fit them as Depots. You fly them for a HLS mission, fill them with propellants, load those propellants onto a HLS Starship, then return them to Earth. You then remove the depot hardware (extra arm and equipment in the cargo bay), and you now have two Cargo Starships to make money with whilst you wait for the next HLS mission. If one of them experiences any issue between now and then, you can pick another Cargo Starship to use.
Scenario B: You design and build a dedicated Jumbo Depot Starship. You fly it for a HLS mission, fill it with propellants, load those propellants onto a HLS Starship. You now have a Jumbo Depot Starship either sitting in orbit doing nothing (and possibly costing you money to actively monitor) or  returned to Earth (and taking up room at your build site) whilst you wait for the next HLS mission. If it experiences any issue between now and then you need to build a new dedicated vehicle.

Scenario C:  You build two lift tanker Starships (1500t tankage), one of which you kit out as a depot.  The other one is your lift tanker to fill the depots, 150t of prop at a time (9 launches--the depot bring the first 150t with it when it launches to LEO).  In between HLS missions, you can use the two Starships as lift tankers, although it's much more likely that one of them stays kitted out for depot work.

If you never have a BEO mission other than an LSS mission, then Scenario A beats Scenario C.  But that's no direct-to-GEO missions, no CLPS missions, no Mars test missions, no deep-space scientific missions, no customers that want to put up a cislunar constellation.  What are the odds?

Meanwhile, let's assume that:

a) An LSS dry mass is 115t (95t Starship dry, 20t crew module, and 1500t of tankage)
b) All vanilla cargo Starship and lift tankers are 120t dry and can bring 150t of prop to LEO for transfer.
c) Raptor Isp=378s. 
d) All tankers have tankage for 1500t of prop, but launch with only 1200t.
e) Everything is EDL-capable, and everything eventually returns to EDL. 
f) NRHO tankers return direct to EDL.

Scenario A for Option A (LSS does LEO-NRHO-LS-NRHO):
Cargo-SS-as-depot #1: One launch, brings 150t of prop to LEO.
Cargo-SS-as-depot #2: One launch, brings 150t of prop to LEO.
Cargo-SS-as-lift-tanker #3:  7 launches to fill #1, 1 launch to fill #2.
LSS requires two RPODs to be fully fueled in LEO.
Total tanker launches: 10

Scenario A for Option B (LSS empty in NRHO does NRHO-LS-NRHO.)
Cargo-SS-as-depot #1: One launch, brings 150t of prop.
Cargo-SS-as-depot #2: One launch, brings 150t of prop.
Cargo-SS-as-lift-tanker #3: 7 launches, to fill #1, 4 launches to fill #2.
Both #1 and #2 have to go to NRHO and return to EDL.
LSS requires two RPODs to be fully fueled in NRHO.
Total tanker launches:  13

I'm skipping Scenario B, because it's a strawman.

Scenario C for Option A (LSS does LEO-NRHO-LS-NRHO):
Tanker-SS-as-depot #1:  One launch, brings 150t of prop.
Tanker-SS-as-lift-tanker #2: 9 launches, each bringing 150t of prop.
LSS requires one RPOD to be fully fueled in LEO.
Total tanker launches: 10

Scenario C for Option B (LSS already in NRHO does NRHO-LS-NRHO):
Tanker-SS-as-depot #1: One launch bring 150t of prop.
Tanker-SS-as-lift-tanker #2: 9 launches, each bringing 150t of prop.
Only Depot #1 goes to NRHO.
LSS requires one RPOD to be fully fueled in NRHO.
Total tanker launches: 10
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/07/2022 01:44 pm
Scenario A: you build two Cargo Starships, and fit them as Depots. You fly them for a HLS mission, fill them with propellants, load those propellants onto a HLS Starship, then return them to Earth. You then remove the depot hardware (extra arm and equipment in the cargo bay), and you now have two Cargo Starships to make money with whilst you wait for the next HLS mission. If one of them experiences any issue between now and then, you can pick another Cargo Starship to use.
Scenario B: You design and build a dedicated Jumbo Depot Starship. You fly it for a HLS mission, fill it with propellants, load those propellants onto a HLS Starship. You now have a Jumbo Depot Starship either sitting in orbit doing nothing (and possibly costing you money to actively monitor) or  returned to Earth (and taking up room at your build site) whilst you wait for the next HLS mission. If it experiences any issue between now and then you need to build a new dedicated vehicle.

Scenario C:  You build two lift tanker Starships (1500t tankage), one of which you kit out as a depot.  The other one is your lift tanker to fill the depots, 150t of prop at a time (9 launches--the depot bring the first 150t with it when it launches to LEO).  In between HLS missions, you can use the two Starships as lift tankers, although it's much more likely that one of them stays kitted out for depot work.

If you never have a BEO mission other than an LSS mission, then Scenario A beats Scenario C.  But that's no direct-to-GEO missions, no CLPS missions, no Mars test missions, no deep-space scientific missions, no customers that want to put up a cislunar constellation.  What are the odds?

Meanwhile, let's assume that:

a) An LSS dry mass is 115t (95t Starship dry, 20t crew module, and 1500t of tankage)
b) All vanilla cargo Starship and lift tankers are 120t dry and can bring 150t of prop to LEO for transfer.
c) Raptor Isp=378s. 
d) All tankers have tankage for 1500t of prop, but launch with only 1200t.
e) Everything is EDL-capable, and everything eventually returns to EDL. 
f) NRHO tankers return direct to EDL.

Scenario A for Option A (LSS does LEO-NRHO-LS-NRHO):
Cargo-SS-as-depot #1: One launch, brings 150t of prop to LEO.
Cargo-SS-as-depot #2: One launch, brings 150t of prop to LEO.
Cargo-SS-as-lift-tanker #3:  7 launches to fill #1, 1 launch to fill #2.
LSS requires two RPODs to be fully fueled in LEO.
Total tanker launches: 10

Scenario A for Option B (LSS empty in NRHO does NRHO-LS-NRHO.)
Cargo-SS-as-depot #1: One launch, brings 150t of prop.
Cargo-SS-as-depot #2: One launch, brings 150t of prop.
Cargo-SS-as-lift-tanker #3: 7 launches, to fill #1, 4 launches to fill #2.
Both #1 and #2 have to go to NRHO and return to EDL.
LSS requires two RPODs to be fully fueled in NRHO.
Total tanker launches:  13

I'm skipping Scenario B, because it's a strawman.

Scenario C for Option A (LSS does LEO-NRHO-LS-NRHO):
Tanker-SS-as-depot #1:  One launch, brings 150t of prop.
Tanker-SS-as-lift-tanker #2: 9 launches, each bringing 150t of prop.
LSS requires one RPOD to be fully fueled in LEO.
Total tanker launches: 10

Scenario C for Option B (LSS already in NRHO does NRHO-LS-NRHO):
Tanker-SS-as-depot #1: One launch bring 150t of prop.
Tanker-SS-as-lift-tanker #2: 9 launches, each bringing 150t of prop.
Only Depot #1 goes to NRHO.
LSS requires one RPOD to be fully fueled in NRHO.
Total tanker launches: 10
Your C is just B with the nameplate ground off. You're going to need a lot of launches at full mass to make regular propellant transfer necessary, which unless you have a very high mass unitary payload going to GEO means you may as well launch two Starships with half the payload that can fly themselves to GEO and back (and insert their payloads into the target slots directly) rather than one Starship that needs to meet a tanker (still two launches) and also needs to muck about with phasing to get its payloads into their slots.
Unitary high mass high energy payloads are likely to be an extreme minority of Starship missions. Building dedicated vehicles to service those makes little sense if you can fly two standard vehicles instead for the same outcome. Every new SKU introduced adds more fixed and ongoing costs.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/07/2022 06:06 pm
Here's an orbital question. If a depot is launched on the autumnal equinox, at midnight, with an inclination of 23.5deg it would be orbiting in the plane of the ecliptic. Is there any orbital juju that would economically keep it in the plane of the ecliptic? If it can be kept there it's always lined up to support a mars campaign.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/07/2022 07:03 pm
Your C is just B with the nameplate ground off. You're going to need a lot of launches at full mass to make regular propellant transfer necessary, which unless you have a very high mass unitary payload going to GEO means you may as well launch two Starships with half the payload that can fly themselves to GEO and back (and insert their payloads into the target slots directly) rather than one Starship that needs to meet a tanker (still two launches) and also needs to muck about with phasing to get its payloads into their slots.
Unitary high mass high energy payloads are likely to be an extreme minority of Starship missions. Building dedicated vehicles to service those makes little sense if you can fly two standard vehicles instead for the same outcome. Every new SKU introduced adds more fixed and ongoing costs.

Starship can't do a reusable mission direct to GEO with any payload without 3 tankers' worth of refueling.  However, I agree that these will be rare.  And I doubt that the DoD will tolerate refueling for NatSec payloads--they'll use an FH2R or FHE for quite a while.  But you could get some commercial traffic, and maybe even some cubesat constellations.

Expendable LSSes for CLPS missions require at least 2 tankers' worth of prop, even for a minimal payload.

I'm pretty sure you'll see some cislunar traffic in the Arty 3 timeframe.  DARPA is already sniffing around applications there, and my guess is you'll have a couple of commercial customers having FOMO about commsat applications.  There are also Gateway construction and resupply.  Those will require two tankers' worth apiece if you want to recover the Starships.

I wouldn't rule out a couple of tourist missions to LLO (DearMoon and a Jared Special) in the 2025 timeframe, as long as they're willing to use a D2 for getting to and from LEO, and the LSS does a propulsive LEO insert.  Those need 6-7 tankers.

And then there's Mars.  I'm under no illusions that there are going to be tens of Starships to Mars this decade, but I'd expect there to be a couple of flight test missions each synod, starting in late 2024 (assuming they can get the planetary protection issues ironed out).  Each of those is 4-5 tankers.

Finally, you have refueling flight test.  Those can certainly be done with cargo Starships, but if you had the tankers, you'd use them.

Seems to me that you have a pretty high likelihood of at least one tanker per month within three years.  That's worth dedicating a couple of hulls.

I think the stock keeping and change management expense for hulls is trivial.  We're not talking about hundreds or even tens of Starships in this timeframe--and if we were, dedicated tanker hulls would clearly be worth it.  You're going to have to build the infrastructure for LSS hulls anyway, and those almost certainly need 1500t of tankage, unless NASA wants to incur refueling risk in NRHO for Option A.  Why not use it for tankers at the same time?

One last thing that I wouldn't rule out:  SpaceX could roll the whole fleet over to using larger tankage and stretch the outer mould line.  You get about 56t of prop per meter of height.  A 6m stretch would give you about 1530t total, with the same payload bay size.  Eyeballing mechazilla, it looks like it has enough headroom at the top to deal with that.  It's obviously more dry mass, but if Raptor 2 (or 3) really gets to 250tf, I don't think an extra 10t-20t makes much difference. (Edit:  6m of extra height might be only 5t-10t of dry mass.)  And it gives you a higher mass ratio when refueled.  What's not to like?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/07/2022 10:45 pm
Here's an orbital question. If a depot is launched on the autumnal equinox, at midnight, with an inclination of 23.5deg it would be orbiting in the plane of the ecliptic. Is there any orbital juju that would economically keep it in the plane of the ecliptic? If it can be kept there it's always lined up to support a mars campaign.

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

But finally, Mars is inclined 1.85º to the ecliptic, so this doesn't completely solve your problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/07/2022 11:49 pm
Here's an orbital question. If a depot is launched on the autumnal equinox, at midnight, with an inclination of 23.5deg it would be orbiting in the plane of the ecliptic. Is there any orbital juju that would economically keep it in the plane of the ecliptic? If it can be kept there it's always lined up to support a mars campaign.

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

But finally, Mars is inclined 1.85º to the ecliptic, so this doesn't completely solve your problem.
Thanks.

You said earlier in another context that it's easier (less propellant?) to do a plane change during a launch than to do it on orbit. Still true? What sort of savings might there be? NOTE: 'Launch' is a simple word with a complex set of trades behind it so you don't loose points if you don't ring the bell. :D

For a depot or a tanker reaching orbit massing 150t over dry weight, how much propellant does 670m/s translate to?

Yeah, Mars isn't dead on the ecliptic but if you set a depot there it's never more than 1.85degrees off. It's a handy place for any planetary or asteroid mission. Never perfect, but operationally excellent. Looks like what I've been calling "a busy lane."

There'd be lunar perturbations so anything there would need to do station keeping burns occasionally. Maybe this would be a place for electric thrust to reappear.

Late thought: perturbations might help in fine tuning the orbit to reduce that 1.85 degrees or whatever's needed for a launch elsewhere. I doubt missions will be so fast and furious that this would be undoable in the short term.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/08/2022 12:31 am
I don't know how long it's been since we've seen an update on mars mission conops. Way past the SX 'use by' date.


As mentioned up thread, there will probably be uncrewed launches this decade but probably not the big push described in the past. (7.5 years might go by fast once the FAA log jam breaks, so who's to say?)


Using low energy transfers that take longer than a Hohmann transfer, launches can be spaced out through a synod as is convenient. One depot on the ecliptic, assuming it doesn't break down, could support this. The ships themselves would be relatively inexpensive. If the tanker(s) aren't doing anything else and pad scheduling allows, their use would spread their amortization over more flights and expand operational experience. Not to mention what could be learned about mars.


Propellant would be far and away the largest expense. Like back in the day, yeah I'm up for a pizza run if you buy the gas. Helloooo, NASA.  Want pizza?



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/08/2022 12:53 am

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

What about precession?  By my calculation in a 500kmx500kmx23.5º orbit the RAAN will move completely around the Earth in a bit under 50 days.

Precession is caused by the oblateness of the Earth.  There is no reason for an orbit to remain fixed  w.r.t. the sun and ecliptic.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/08/2022 02:57 am

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

What about precession?  By my calculation in a 500kmx500kmx23.5º orbit the RAAN will move completely around the Earth in a bit under 50 days.

Precession is caused by the oblateness of the Earth.  There is no reason for an orbit to remain fixed  w.r.t. the sun and ecliptic.
Yeah, I made the same mistake myself a few months ago. Then, once I realized the orbit would precess, I assumed it would be trivial for the depot to use ion engines to station keep. But it turns out the cumulative delta-v to keep it from precessing is horrendous.

Upshot: you can't stop it from precessing, and, as Barley says, it's going to go around a bit over once a month.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/08/2022 04:42 am

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

What about precession?  By my calculation in a 500kmx500kmx23.5º orbit the RAAN will move completely around the Earth in a bit under 50 days.

Precession is caused by the oblateness of the Earth.  There is no reason for an orbit to remain fixed  w.r.t. the sun and ecliptic.

You're right (again).  I had a braino and thought the precession was about the ecliptic instead of the equator, which is of course wrong.

This seems like it's kind of a big deal for depots in LEO that are used for interplanetary missions.  On the one hand, you can have an ecliptic inclination that's way out of whack for where you want to go if you expend no delta-v.  One the other hand, by carefully scheduling the increase or reduction of the eccentricity of the orbit, you can move the depot into a good departure alignment for a fairly small amount of delta-v.  But reliable scheduling would be very important.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/08/2022 05:11 am
First, note that Barley poured cold water on my "if you insert into the ecliptic, you stay in the ecliptic" assertion.

Thanks.

You said earlier in another context that it's easier (less propellant?) to do a plane change during a launch than to do it on orbit. Still true? What sort of savings might there be? NOTE: 'Launch' is a simple word with a complex set of trades behind it so you don't loose points if you don't ring the bell.

This is one of those things where you can get varying amounts of free delta-v from the rotation of the Earth.  If you were launching from the equator due north, you'd lose 465m/s that you could have if you were launching due east.

From any other latitude, launching due east gives you 465*cos(latitudeOfLaunch) m/s of free delta-v.  Since most (but not all) parking orbits for lunar or interplanetary departures have fairly low inclinations, you're usually not losing more than 100-150m/s of free delta-v by going straight to the target orbit.¹

Quote
For a depot or a tanker reaching orbit massing 150t over dry weight, how much propellant does 670m/s translate to?

Assuming Isp=378s and dry mass=120t, 45t of prop.

_________
¹This OrbiterWiki page (https://www.orbiterwiki.org/wiki/Launch_Azimuth#:~:text=The%20launch%20azimuth%20is%20the,the%20celestial%20body's%20reference%20plane.) has the full set of azimuth calculations and their delta-v implications.  Trigonometric pain awaits.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 06/08/2022 05:54 am
So dumb thought. 
Could you have the tanker and depot meet up so that they are off nose to tail. So that instead of the qd plates meet at the same end one essentially is flying tail first, and the other nose first. [...]
[...]
Note that the actual Starship tail is occupied by Raptors, so you must figure out exactly where the connection will occur. for all ships with a tail connection. Note that the HLS probably has crew cocking hatch in the nose.
Note that it is probably challenging to have a nose port on EDL-capable ships including tankers.
(Minor: nose-to-tail dock has both ships going forward or both ships going backward, not as you stated).

I believe shark0302 meant flying side-by-side, but facing opposite directions. (Sketched below.)

I suspect they also meant that the fancy stuff for prop-transfer is in the nose of just one ship, the active or male. And is meant to be compatible with the standard (or what becomes the standard) passive/female QD plate at the tail of every ship. No extras are needed in the tail and not every ship needs the fancy in its nose.

Hence the depot would be an active with all the necessary fancy in its nose, along with its other depot-specific long-duration mods. Tankers, HLS, Mars ships can be a passive, without the fancy in the nose.

(Since the actives still have the regular passive QD plate at their tail, they can mate/dock with other actives. If you need that flexibility.)


[I might have interpreted shark0302 wrong, of course.]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/08/2022 07:47 am
I don't know about shark0302, but that's definitely what I was trying to describe. Packing all depot hardware into the payload bay allows for any Starship to be a Depot when needed but when no depots are needed that Starship can still carry payloads like any other (and earn money rather than sitting doing nothing).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/08/2022 02:48 pm
There does not appear to be any reason* for the axis of the ships to be parallel, or anti parallel, or to lie in the same plane.  If you're not using the ground support plumbing the ullage thrust does not need to be along a ship's axis, and even if you are using the ground support plumbing the receiving ship can be off axis and tens of degrees of off axis thrust could be accommodated on the donor.

For example, a back to back X configuration provides clearance for fins and could tolerate a far less rigid dock as long as the plumbing is flexible.

* other than a misplaced sense of order and an inability to get your minds off of a mudball and into space where there is no up or down.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/08/2022 04:05 pm
If you're not using the ground support plumbing the ullage thrust does not need to be along a ship's axis, and even if you are using the ground support plumbing the receiving ship can be off axis and tens of degrees of off axis thrust could be accommodated on the donor
Settling thrust axis must be aligned such that the fluid drain for both tanks is at the lowest point of the tanks along that axis (i.e. you have to fill and drain from the 'bottom'). That's your constraint on vehicle orientation when transferring propellants.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/08/2022 05:07 pm
If you're not using the ground support plumbing the ullage thrust does not need to be along a ship's axis, and even if you are using the ground support plumbing the receiving ship can be off axis and tens of degrees of off axis thrust could be accommodated on the donor
Settling thrust axis must be aligned such that the fluid drain for both tanks is at the lowest point of the tanks along that axis (i.e. you have to fill and drain from the 'bottom'). That's your constraint on vehicle orientation when transferring propellants.
A conical sump tolerates significant misalignment between the axis of symmetry and the axis of thrust.  If you're not using the ground support plumbing, you can put the transfer intake wherever you want.  "Down" is a design choice, not a given.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/08/2022 06:29 pm

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

What about precession?  By my calculation in a 500kmx500kmx23.5º orbit the RAAN will move completely around the Earth in a bit under 50 days.

Precession is caused by the oblateness of the Earth.  There is no reason for an orbit to remain fixed  w.r.t. the sun and ecliptic.
That put the kabosh to that idea - except... if it precesses a full circuit in only 50 days, transfer injection opportunities for any specific ecliptic target come every 25 days. Not too bad.


Two depots, 90deg apart on this busy lane give ecliptic insertion opportunities every 12.5 days. Three with 120 deg spacing give opportunities about every eight days. Assuming a perfectly choreographed depot loading campaign of six tankers (900t) at 12 hour intervals, it would take three days to fill a tanker. Go for a 1200t load from 8 tankers it's four days. Go for a 1600t load on 11 tankers (really 10.6 tankers) and it's 5.5 days.


If saving propellant is an overriding concern the depot loading campaign can be limited to those days where the precession is most advantageous. Truth be said, recycling rockets in an hour isn't going to happen for quite a while. Recycling the launch facility in half a day, including props replenishment, is the immediate goal to work towards. It'll be hard enough.


BTW, Lunar orbit is inclined 1.54deg to the ecliptic. Depots in the discussed orbit can easily support lunar missions. If I'm picturing the precession correctly, when precession is half way between perfect for an ecliptic insertion and it's worst, it's perfect for a GEO insertion. It's looking like an orbit that precesses to parallel to the ecliptic has potential to be a very busy lane.


ISTM that after the initial futzing around this is strong argument for depot(s) that are more a robust facility than a ship on TDY.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/08/2022 07:47 pm
You might want to have a look at "Raan-Agnostic 3-Burn Departure Methodology for Deep Space Missions From Leo Depots (https://www.researchgate.net/profile/John-Carrico/publication/327390433_RAAN-Agnostic_3-Burn_Departure_Methodology_for_Deep_Space_Missions_from_LEO_Depots/links/5b9913be92851c4ba8150f50/RAAN-Agnostic-3-Burn-Departure-Methodology-for-Deep-Space-Missions-from-LEO-Depots.pdf),” which describes a very cool method to efficiently launch interplanetary payloads from a depot in an almost-arbitrary orbit. Not great for crew, since it adds as much as a month to the trajectory (plus two extra trips through the Van Allen belts), but for cargo, it looks pretty good, and it lets you make do with a single depot.

Something I just thought of: if you put the depot at an altitude of 770 km, it should precess in 59 days, which is two (sidereal) lunar orbits, meaning you ought to be able to arrange for once-a-month direct transfers to the moon.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/08/2022 07:57 pm
If you're not using the ground support plumbing the ullage thrust does not need to be along a ship's axis, and even if you are using the ground support plumbing the receiving ship can be off axis and tens of degrees of off axis thrust could be accommodated on the donor
Settling thrust axis must be aligned such that the fluid drain for both tanks is at the lowest point of the tanks along that axis (i.e. you have to fill and drain from the 'bottom'). That's your constraint on vehicle orientation when transferring propellants.
A conical sump tolerates significant misalignment between the axis of symmetry and the axis of thrust.  If you're not using the ground support plumbing, you can put the transfer intake wherever you want.  "Down" is a design choice, not a given.

I think it's fairly safe to say that main tanks are only going to have one outlet, and it's going to do double duty as the outlet that feeds the tanker/depot's own propulsion.  So if you put ullage thrust off-axis, it'll be at the expense of some increased sump losses.  Geometry can only do so much.

However, if it's just a little bit off-axis, that might not matter very much for most tanker situations, because tankers need to hold a prop reserve for return and EDL, where thrust will certainly be on-axis.  It's a bigger deal for a true depot, because missions may be designed around an integral number of tanker launches, and increased sump losses can reduce payload size.  It doesn't matter for the first through n-1th tankers, but that last tanker's transferred (and usable) prop can be non-trivially smaller than the previous ones.

All things considered, on-axis sumps with on-axis ullage thrusts are better.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: shark0302 on 06/08/2022 08:06 pm
So dumb thought. 
Could you have the tanker and depot meet up so that they are off nose to tail. So that instead of the qd plates meet at the same end one essentially is flying tail first, and the other nose first. [...]
[...]
Note that the actual Starship tail is occupied by Raptors, so you must figure out exactly where the connection will occur. for all ships with a tail connection. Note that the HLS probably has crew cocking hatch in the nose.
Note that it is probably challenging to have a nose port on EDL-capable ships including tankers.
(Minor: nose-to-tail dock has both ships going forward or both ships going backward, not as you stated).

I believe shark0302 meant flying side-by-side, but facing opposite directions. (Sketched below.)

I suspect they also meant that the fancy stuff for prop-transfer is in the nose of just one ship, the active or male. And is meant to be compatible with the standard (or what becomes the standard) passive/female QD plate at the tail of every ship. No extras are needed in the tail and not every ship needs the fancy in its nose.

Hence the depot would be an active with all the necessary fancy in its nose, along with its other depot-specific long-duration mods. Tankers, HLS, Mars ships can be a passive, without the fancy in the nose.

(Since the actives still have the regular passive QD plate at their tail, they can mate/dock with other actives. If you need that flexibility.)


[I might have interpreted shark0302 wrong, of course.]
Thanks this is exactly what I meant. Sorry if it wasn't clearer.

Sent from my SM-T860 using Tapatalk

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/08/2022 08:27 pm

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

What about precession?  By my calculation in a 500kmx500kmx23.5º orbit the RAAN will move completely around the Earth in a bit under 50 days.

Precession is caused by the oblateness of the Earth.  There is no reason for an orbit to remain fixed  w.r.t. the sun and ecliptic.
That put the kabosh to that idea - except... if it precesses a full circuit in only 50 days, transfer injection opportunities for any specific ecliptic target come every 25 days. Not too bad.


I might be wrong here, but I believe alignment only occurs once per cycle. After half a cycle the descending node is pointing at the first point of Aeries and that doubles the angle between the orbital plane and the ecliptic.   So at the 25 day mark the angle to the ecliptic is 47º. 

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/08/2022 09:00 pm
You might want to have a look at "Raan-Agnostic 3-Burn Departure Methodology for Deep Space Missions From Leo Depots (https://www.researchgate.net/profile/John-Carrico/publication/327390433_RAAN-Agnostic_3-Burn_Departure_Methodology_for_Deep_Space_Missions_from_LEO_Depots/links/5b9913be92851c4ba8150f50/RAAN-Agnostic-3-Burn-Departure-Methodology-for-Deep-Space-Missions-from-LEO-Depots.pdf),” which describes a very cool method to efficiently launch interplanetary payloads from a depot in an almost-arbitrary orbit. Not great for crew, since it adds as much as a month to the trajectory (plus two extra trips through the Van Allen belts), but for cargo, it looks pretty good, and it lets you make do with a single depot.

Something I just thought of: if you put the depot at an altitude of 770 km, it should precess in 59 days, which is two (sidereal) lunar orbits, meaning you ought to be able to arrange for once-a-month direct transfers to the moon.

Yup, we talked about this method up-thread a bit.  Some of the authors on the paper were co-authors with the paper that Jon Goff wrote.  (Edit: Reading error--he's on this paper too.)

I agree that this will work fine for cargo flights.  It'll probably work OK for crewed Mars flights, where the double-dip through the VA belts is probably the least of the crew's radiation worries.  But it takes too long and the double-dip will be significant for crewed lunar flights.

I do think that it's possible to manage the depot's RAAN pretty cheaply for lunar flights.  However, permanently positioning it and doing refueling ops at 770km seems pretty risky.  Until proven otherwise, depots should be considered major debris hazards.  Therefore, refueling as low as possible is probably a good orbital hygiene measure.

This is yet another reason why I like temporary, mission-oriented depots, instead of permanent or semi-permanent ones.  With a mission-oriented depot, you can do all ops in 300km VLEO, where a tanker collision or a transfer-related explosion won't hurt as much.  This probably comes at the expense of a few tonnes of prop¹ to maintain the orbit while the prop load is aggregated.

A low refueling point also gives you a bit more Oberth effect for whatever insertion burn you're going to do next.  Probably fairly trivial, but it's better'n a sharp stick in the eye.

______________
¹Average drag @ ~300km is 2E-4N/m².  Figure an average 50m² of solar array aspect (a SWAG) and 200m² average Starship aspect (another SWAG, assuming that we're optimizing for drag reduction rather than thermal management, which we brute-force with cryocooling), so we have 0.05N average drag force.  For a lunar mission that takes 10 tanker launches, with the first becoming the depot, with a cadence of 1 launch/2weeks, that's 140days with an average mass of 945t (starts at 270t and ends at 1620t), that's a Δv of 0.6m/s. 

Use a generous fudge factor and move that up to 5m/s and use the average mass of 945t, with a cold-gas thruster with Isp=105s.  That's 4.6t of prop, which is probably not too different than the boil-off, which can be used to drive the thrusters.

Many, many questionable assumptions in the above calculation.  No doubt barley will find another embarrassing dropped decimal or two.  (I haven't exactly been on an arithmetic accuracy tear recently.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/08/2022 10:57 pm
If you're not using the ground support plumbing the ullage thrust does not need to be along a ship's axis, and even if you are using the ground support plumbing the receiving ship can be off axis and tens of degrees of off axis thrust could be accommodated on the donor
Settling thrust axis must be aligned such that the fluid drain for both tanks is at the lowest point of the tanks along that axis (i.e. you have to fill and drain from the 'bottom'). That's your constraint on vehicle orientation when transferring propellants.
A conical sump tolerates significant misalignment between the axis of symmetry and the axis of thrust.  If you're not using the ground support plumbing, you can put the transfer intake wherever you want.  "Down" is a design choice, not a given.

I think it's fairly safe to say that main tanks are only going to have one outlet, and it's going to do double duty as the outlet that feeds the tanker/depot's own propulsion.  So if you put ullage thrust off-axis, it'll be at the expense of some increased sump losses.  Geometry can only do so much.

However, if it's just a little bit off-axis, that might not matter very much for most tanker situations, because tankers need to hold a prop reserve for return and EDL, where thrust will certainly be on-axis.  It's a bigger deal for a true depot, because missions may be designed around an integral number of tanker launches, and increased sump losses can reduce payload size.  It doesn't matter for the first through n-1th tankers, but that last tanker's transferred (and usable) prop can be non-trivially smaller than the previous ones.

All things considered, on-axis sumps with on-axis ullage thrusts are better.
Or settle and do most of the transfer off axis and only that last little bit on axis. Not a 'it has to be this way' but a viable option for some circumstances. Here's my thinking.


We know the options for settling thrust but don't know what will be chosen. Venting for thrust could be done by both ships keeping thrust on axis, or it may be done by special build low power engines on the depot only, giving off axis thrust. If the latter, the second ship could vent for pitch against the depot to get the dregs. Maybe all, maybe only some.


On a related note, all the vids I've seen of F9 RCS is hard blaps of gas. It looks strictly on-off with no long vents or throttling. It doesn't really look like a system with enough nuance for anything like what a depot needs to do. That's a gut reaction, not actual knowledge.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/09/2022 01:12 am

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

What about precession?  By my calculation in a 500kmx500kmx23.5º orbit the RAAN will move completely around the Earth in a bit under 50 days.

Precession is caused by the oblateness of the Earth.  There is no reason for an orbit to remain fixed  w.r.t. the sun and ecliptic.
That put the kabosh to that idea - except... if it precesses a full circuit in only 50 days, transfer injection opportunities for any specific ecliptic target come every 25 days. Not too bad.


I might be wrong here, but I believe alignment only occurs once per cycle. After half a cycle the descending node is pointing at the first point of Aeries and that doubles the angle between the orbital plane and the ecliptic.   So at the 25 day mark the angle to the ecliptic is 47º.
I could be wronger (is that even a word?) than you but ISTM if you're shooting for something on the ecliptic while orbiting at a high inclination from the ecliptic, when a tangent from either ascending or descending node is pointing in the right direction, you can go. Whichever is used, the other node will line up in half a precession. It's not like there's all that much parallax between the extremes except for GEO. I think.


Any small differences between the extremes would be small potato's easily corrected during the usual mid course correction that would be most economical at the slowest point in the transit. I think.


On reread, maybe this is all argument that this isn't all that special an orbit except for GEO.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/09/2022 01:38 am
If you're not using the ground support plumbing the ullage thrust does not need to be along a ship's axis, and even if you are using the ground support plumbing the receiving ship can be off axis and tens of degrees of off axis thrust could be accommodated on the donor
Settling thrust axis must be aligned such that the fluid drain for both tanks is at the lowest point of the tanks along that axis (i.e. you have to fill and drain from the 'bottom'). That's your constraint on vehicle orientation when transferring propellants.
A conical sump tolerates significant misalignment between the axis of symmetry and the axis of thrust.  If you're not using the ground support plumbing, you can put the transfer intake wherever you want.  "Down" is a design choice, not a given.

I think it's fairly safe to say that main tanks are only going to have one outlet, and it's going to do double duty as the outlet that feeds the tanker/depot's own propulsion.  So if you put ullage thrust off-axis, it'll be at the expense of some increased sump losses.  Geometry can only do so much.

However, if it's just a little bit off-axis, that might not matter very much for most tanker situations, because tankers need to hold a prop reserve for return and EDL, where thrust will certainly be on-axis.  It's a bigger deal for a true depot, because missions may be designed around an integral number of tanker launches, and increased sump losses can reduce payload size.  It doesn't matter for the first through n-1th tankers, but that last tanker's transferred (and usable) prop can be non-trivially smaller than the previous ones.

All things considered, on-axis sumps with on-axis ullage thrusts are better.
This just kicked off a thought.


Assume the depot has the thrusters and transfer is off axis, with or without the last minute correction I suggested earlier. Also assume it is bone dry. The two ships will have roughly the same mass during the transfer. Ten transfers later the mass of the tanker will be nothing compared to the mass of the depot. Off axis angle would be very small.


Then a ship shows up and drains the propellant. At the beginning of transfer the depot is the heavy ship and the receiver is light. By the end of the transfer the receiver is heavy and the depot is light. The off axis angle would be high. Unless the depot is expected to keep fairly large residuals this large mass inequality might create an off axis angle that would be difficult to counter.


Possible solutions: either or both ships contribute thrust; live with it. Anything else?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/09/2022 02:51 am
______________
¹Average drag @ ~300km is 2E-4N/m².  Figure an average 50m² of solar array aspect (a SWAG) and 200m² average Starship aspect (another SWAG, assuming that we're optimizing for drag reduction rather than thermal management, which we brute-force with cryocooling), so we have 0.05N average drag force.  For a lunar mission that takes 10 tanker launches, with the first becoming the depot, with a cadence of 1 launch/2weeks, that's 140days with an average mass of 945t (starts at 270t and ends at 1620t), that's a Δv of 0.6m/s. 

Use a generous fudge factor and move that up to 5m/s and use the average mass of 945t, with a cold-gas thruster with Isp=105s.  That's 4.6t of prop, which is probably not too different than the boil-off, which can be used to drive the thrusters.

Many, many questionable assumptions in the above calculation.  No doubt barley will find another embarrassing dropped decimal or two.  (I haven't exactly been on an arithmetic ac5curacy tear recently.)
Looks good once you account for the 5/0.6 = 8.3 fudge factor.  You don't actually need the average mass, since it cancels.  You can also calculate the impulse which amounts to the same thing.  0.05N x 140days x 24hours/day x 3600s/hour / (105s * 9.81m s-2) = 528kg  which is close once you multiply in the fudge factor.


But on a different note you want the depot as low as practicable for efficiency reasons.  300km may well be too high.

Going from 300x300 to 770x770 orbit takes a Δv of 258 m/s.  Plus an extra 120 m/s for reentry from the higher orbit.   So 378 m/s to move the dry mass of 10 tankers that would not need to be boosted if you refuel at 300km.   A raptor vacuums exhaust velocity is 3561m/s so its something like 125 tonnes of propellent just to move the dry mass of the tankers.  That's close to a whole tankers load, and way more than the drag loss.

Refueling at 200kmx200km saves another 2.7 tonnes per tanker.  I'm not sure how low you can go before drag and operational angst out weights the extra fuel.

Even if you need to refuel in a higher energy orbit you want to transfer fuel to as few tankers as possible in low orbit so as little dry mass as possible needs to be moved.   (This is the logic of staging: drop unneeded mass as soon as possible).

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/09/2022 05:38 am
Refueling at 200kmx200km saves another 2.7 tonnes per tanker.  I'm not sure how low you can go before drag and operational angst out weights the extra fuel.

Assuming that you're fairly confident that your depot's thrusters aren't going to fail and it comes launched with 150t of prop, that's a lot of balm to sooth operational angst.

That said, you can over-constrain your VLEO altitude pretty quickly if you want lots of windows to launch tankers to the depot.  I doubt all of these requirements can be met simultaneously:

1) Orbit accessible from both BC and the Cape.  To do this, you'll have to overfly Florida.  The ground track will be moving at a pretty good clip by then, but you have to avoid Tampa on the south and Jacksonville on the north.  Risk to the public gets lower as the ground track speeds up, but higher with higher population density.  I think there might be a hole at about Ocala, but it's not a slam-dunk.  That's an inclination of about 30.4º, if I did the azimuth calculation correctly. 

The other option, which doesn't overfly anything, is 32.7º inclination, which you could launch south from BC through the Yucatan Channel and north or south from the Cape, but you're starting to spend a bit more delta-v with that one, and it might be a bit too inclined for easy lunar access.

2) At least one access per day.  Even with just an access from one or the other of the two sites, you need an orbit with a period of one sidereal day / n - daily RAAN precession.  Finding an orbital period that would allow access from both sites will be very tough.

3) Altitude doesn't exceed about 400km, for the reasons we were discussing.

280km is 16 orbits per sidereal day.  You'd need to be a bit lower than this to offset the precession (which IIRC is negative).  Even with lots of prop, I'm sure that you'd find some people at NASA wearing their brown trousers doing ops that low.  But 15 orbits/day is 550km, which is way too high. 

No doubt there are some orbits with windows that are more or less than exactly 24 hours, but I'm too lazy to figure them out.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 06/09/2022 01:25 pm
There's been handwringing over having RAAN limiting missions, as well as the delta-v cost of inclination changes. However, you can do a plane-change and/or RAAN-change burn while doing the main transfer burn and eliminate almost all the delta-v cost.

For example, if you required an (arbitrarily chosen) 2km/s plane-change and/or RAAN-change burn, combined with a 3.2km/s LEO-to-NRHO burn, it doesn't increase your delta-v requirement to 5.2km/s, but to just 3.77km/s. That is, by combining them into a single manoeuvre, the 2km/s is reduced to ~570m/s. Likewise a 1km/s extra burn is reduced to ~150m/s when combined.

Similarly, if your desired orbital inclination is different from your launch-site, you don't do the plane-change manoeuvre on-orbit, you do it during launch where it is reduced to almost nothing by combining it with the enormous launch delta-v.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/09/2022 05:47 pm
1) Orbit accessible from both BC and the Cape.
If you want two launch pads you lose a lot of constraints if they are both at the same site. This gives up geographic and political diversification.

An extreme version would be to take over missile row and launch all ten tankers at the same time, there would be no more orbital constraints than launching a single rocket.  Plus salvos of SH launches and landings. 8)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/09/2022 06:01 pm
There's been handwringing over having RAAN limiting missions, as well as the delta-v cost of inclination changes. However, you can do a plane-change and/or RAAN-change burn while doing the main transfer burn and eliminate almost all the delta-v cost.

For example, if you required an (arbitrarily chosen) 2km/s plane-change and/or RAAN-change burn, combined with a 3.2km/s LEO-to-NRHO burn, it doesn't increase your delta-v requirement to 5.2km/s, but to just 3.77km/s. That is, by combining them into a single manoeuvre, the 2km/s is reduced to ~570m/s. Likewise a 1km/s extra burn is reduced to ~150m/s when combined.

Yeah, but when it's the first burn of several, even 150m/s isn't trivial.

For a very pointed example, let's look at Option A:

LEO-TLI: 3190
TLI-NRHO: 520¹
NRHO-LS: 2855
LS-NRHO: 2810
Total: 9375m/s

For a 95t dry mass LSS with a 20t crew module and Isp=378s, that's 1327t of prop with no flight performance reserve.  If you add on 150m/s to that TLI, you now need 1387t of prop with no FPR.  That 150m/s costs you 60t of prop--not exactly trivial.

Quote
Similarly, if your desired orbital inclination is different from your launch-site, you don't do the plane-change manoeuvre on-orbit, you do it during launch where it is reduced to almost nothing by combining it with the enormous launch delta-v.

This is the point I've been trying make, though.  With a permanent depot, all RAAN and inclination misalignments have to be taken out with orbital maneuvers, however optimized they are.  With a depot that returns to Earth after each mission and is then launched to the next target orbit, that orbit alignment delta-v is mostly removed.

____________
¹I'm pulling all post-TLI delta-v from the HLS trade study, a copy of which can be found here (https://drive.google.com/file/d/1cv63Mt5S_F9fg5bYz8hOMQS9d0xHLmWu/view?usp=sharing).  See p. 41.  Note that these are for fast transit with two RPODs.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/09/2022 06:13 pm
1) Orbit accessible from both BC and the Cape.
If you want two launch pads you lose a lot of constraints if they are both at the same site. This gives up geographic and political diversification.

An extreme version would be to take over missile row and launch all ten tankers at the same time, there would be no more orbital constraints than launching a single rocket.  Plus salvos of SH launches and landings. 8)

To paraphrase Donald Rumsfeld:  You do Option A with the pads you have, not the pads you wish you had.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/10/2022 04:26 pm
1) Orbit accessible from both BC and the Cape.
If you want two launch pads you lose a lot of constraints if they are both at the same site. This gives up geographic and political diversification.

An extreme version would be to take over missile row and launch all ten tankers at the same time, there would be no more orbital constraints than launching a single rocket.  Plus salvos of SH launches and landings. 8)

To paraphrase Donald Rumsfeld:  You do Option A with the pads you have, not the pads you wish you had.

A completely misplaced quote and a staggering failure of imagination.  Amongst other things Donald Rumsfeld oversaw the development of the B-1 and quite a few ships because he wished he had them.

Also there are no operational SS pads, so clearly we will never have an SS launch.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/10/2022 08:53 pm
Also there are no operational SS pads, so clearly we will never have an SS launch.

True, but we have every expectation that there will be one pad operational this year, and presumably the LC-39A pad will show up sometime in 2Q23, which would be in time for Option A tanker mission planning.  Counting on a second pad at the Cape in that timeframe seems like a bad idea, especially since Florida ops are probably more gated by getting the factory up and running than by pad construction.

I'd kinda guess that they'll just go with one pad and plan for a window every 3-7 days.  It wouldn't be a terrible idea just to use BC for tankers and LC-39A for Starlinks and have done with it.  After Option A, they can reassess.  By then, they may have something cooking at LC-39...C, is it?  In which case, they can look at upping the refueling cadence.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/10/2022 10:09 pm
It'd be nice to get the first one off. Everything they know about RCS behavior is strictly modeling. If the model is dead on, great. I doubt they're planning on that. They will look at attitude behavior with an eye towards props transfer and other docking. They may take RCS in another direction. They may like what they have.

How tight they can get the RCS control will impact almost every aspect of depot operations. It will impact how they actually hook the ships together and how large the standoff needs to be. From those two things alone the decision tree fans out wide enough to make my brain hurt.

Informal poll: how long following a successful first launch before we see two SS's on orbit at the same time? That's a precursor to any attempt to transfer props.  I'd guess six months minimum. Probably more like a year.


Edit: yeah, they used RCS on the hops. Way too different an environment.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Zed_Noir on 06/11/2022 02:07 pm
Informal poll: how long following a successful first launch before we see two SS's on orbit at the same time? That's a precursor to any attempt to transfer props.  I'd guess six months minimum. Probably more like a year.
Really optimistic, the first and second orbital Starships after a successful suborbital flight to Hawaii. Since it is likely both will be doing water landing to verify the powered pitch up maneuver works after reentry from orbit. For them to meet up and video each other's exterior in orbit should provide more data points.

More realistic, maybe the 4th and the 5th orbital Starships.

When there are Starships in close proximity in orbit. SpaceX will likely be recording 4k videos of the exteriors of the Starships. Then will there be movie light arrays on the Starship exterior for illumination?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/11/2022 04:24 pm
Informal poll: how long following a successful first launch before we see two SS's on orbit at the same time? That's a precursor to any attempt to transfer props.  I'd guess six months minimum. Probably more like a year.
Really optimistic, the first and second orbital Starships after a successful suborbital flight to Hawaii. Since it is likely both will be doing water landing to verify the powered pitch up maneuver works after reentry from orbit. For them to meet up and video each other's exterior in orbit should provide more data points.

More realistic, maybe the 4th and the 5th orbital Starships.

When there are Starships in close proximity in orbit. SpaceX will likely be recording 4k videos of the exteriors of the Starships. Then will there be movie light arrays on the Starship exterior for illumination?
Now you have me musing on how long an SS might stay alive on orbit. More batteries. Longer life. More batteries, fewer starlinks. PV will happen at some point. Power Consumption: comms and RCS solenoids. Later on, camera lighting. What else?


I wonder how hard it would be to substitute panels for the last two starlinks and keep them attached.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/11/2022 04:32 pm
Informal poll: how long following a successful first launch before we see two SS's on orbit at the same time? That's a precursor to any attempt to transfer props.  I'd guess six months minimum. Probably more like a year.
Really optimistic, the first and second orbital Starships after a successful suborbital flight to Hawaii. Since it is likely both will be doing water landing to verify the powered pitch up maneuver works after reentry from orbit. For them to meet up and video each other's exterior in orbit should provide more data points.

More realistic, maybe the 4th and the 5th orbital Starships.

When there are Starships in close proximity in orbit. SpaceX will likely be recording 4k videos of the exteriors of the Starships. Then will there be movie light arrays on the Starship exterior for illumination?
Now you have me musing on how long an SS might stay alive on orbit. More batteries. Longer life. More batteries, fewer starlinks. PV will happen at some point. Power Consumption: comms and RCS solenoids. Later on, camera lighting. What else?


I wonder how hard it would be to substitute panels for the last two starlinks and keep them attached.
Starship HLS is contractually obligated to be able to loiter in NHRO for up to 200(?) days to wait if SLS/Orion launch is delayed. HLS will then take up to 14(?) days to complete its mission. I speculate without evidence that the SpaceX will use the same technologies for the other SS variants, with the exception of the solar panels that appear to be specialized for non-EDL starships.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rliebman on 06/11/2022 04:38 pm
can someone provide an illustration/photo/drawing of what the solar panels on the SS will look like / where attached / how deployed??
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 06/11/2022 04:48 pm

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

What about precession?  By my calculation in a 500kmx500kmx23.5º orbit the RAAN will move completely around the Earth in a bit under 50 days.

Precession is caused by the oblateness of the Earth.  There is no reason for an orbit to remain fixed  w.r.t. the sun and ecliptic.
Yeah, I made the same mistake myself a few months ago. Then, once I realized the orbit would precess, I assumed it would be trivial for the depot to use ion engines to station keep. But it turns out the cumulative delta-v to keep it from precessing is horrendous.

Upshot: you can't stop it from precessing, and, as Barley says, it's going to go around a bit over once a month.

...So what you're saying is that each depot in an orbit like this will precess to the correct orientation about twice a mars launch window?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/11/2022 08:53 pm
...So what you're saying is that each depot in an orbit like this will precess to the correct orientation about twice a mars launch window?

Precession rate is dependent on altitude and inclination, so you can engineer it to a certain extent.  However, you need to be at the right inclination (which changes every synod, and even changes a little bit during the window), and you need the altitude to be low enough both to maximize the amount of prop transferred per tanker and to minimize any debris problems if there's an accident.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/11/2022 09:20 pm
Starship HLS is contractually obligated to be able to loiter in NHRO for up to 200(?) days to wait if SLS/Orion launch is delayed. HLS will then take up to 14(?) days to complete its mission. I speculate without evidence that the SpaceX will use the same technologies for the other SS variants, with the exception of the solar panels that appear to be specialized for non-EDL starships.

They need a solar power solution for EDL-capable Starships: 

1) Even if you have depots in cislunar, they still need to be filled by tankers, which have to return to EDL.  (If they're not reusable, then you might as well just expend a depot for every mission, which seems like a waste.)  That's likely at least a week, probably closer to two weeks when you factor in RPODs, both in LEO and in cislunar.

2) GEO missions have an RPOD in LEO, the boost to GTO, the circularization maneuver, the deployment, and the insertion back into EDL.  I can't imagine that all happening in less than 2-3 days.

3) And of course Mars missions need 6-12 months of power, followed by an EDL.

Cargo ships and tankers could deploy solar panels from a payload bay, but that's harder to do with a crewed Mars ship.

The old old old drawings from 2017 and 2019 showed solar arrays deployed from the propulsion section.  My speculation was that the booms with the stowed arrays lived in chines along the outside of the tanks.  On orbit, you'd open doors in the chines, swing the booms out to perpendicular, then expand the arrays.  That still seems like it could work.  But the arrays might interfere with docking for refueling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/12/2022 01:12 am
Starship HLS is contractually obligated to be able to loiter in NHRO for up to 200(?) days to wait if SLS/Orion launch is delayed. HLS will then take up to 14(?) days to complete its mission. I speculate without evidence that the SpaceX will use the same technologies for the other SS variants, with the exception of the solar panels that appear to be specialized for non-EDL starships.

They need a solar power solution for EDL-capable Starships: 

1) Even if you have depots in cislunar, they still need to be filled by tankers, which have to return to EDL.  (If they're not reusable, then you might as well just expend a depot for every mission, which seems like a waste.)  That's likely at least a week, probably closer to two weeks when you factor in RPODs, both in LEO and in cislunar.

2) GEO missions have an RPOD in LEO, the boost to GTO, the circularization maneuver, the deployment, and the insertion back into EDL.  I can't imagine that all happening in less than 2-3 days.

3) And of course Mars missions need 6-12 months of power, followed by an EDL.

Cargo ships and tankers could deploy solar panels from a payload bay, but that's harder to do with a crewed Mars ship.

The old old old drawings from 2017 and 2019 showed solar arrays deployed from the propulsion section.  My speculation was that the booms with the stowed arrays lived in chines along the outside of the tanks.  On orbit, you'd open doors in the chines, swing the booms out to perpendicular, then expand the arrays.  That still seems like it could work.  But the arrays might interfere with docking for refueling.
Apart from Earth-Mars coast, on-board energy requirements could be met with a genset running off of on-board propellants.
A commercial 100kW genset running at 100% duty cycle would consume a little under 25kg LCH4 per hour (warmed to gaseous Methane, an ideal use of exhaust heat) along with somewhere under 50kg of LOX (also warmed, and you want to use less to avoid detonation from stoichiometric combustion. Exhaust gas recirculation may also help). So on the order of 75kg propellant mass per hour in orbit (at 100kW constant energy consumption). IIRC Ultraflex arrays are ~7kg/kW, so ~700kg for a 100kW array - only half duty cycle for many Earth orbits though due to the sun only being on one side, so ~1400kg really. Even if we assume a 1 tonne off-the-shelf genset (rather than something derived more from the aviation world) and ~400kg for propellant, your desired on-orbit lifetime bust be beyond 5 hours before a deployable array becomes the desirable option in terms of mass (and adds all the issues of an array that can be un-deployed, which may rule out ultra low mass options like Ultraflex or iROSA). The lighter your genset is, the better it trades: you probably can't approach the 10kW/kg of a gas turbine at such low power levels, but 100kg for a gas turbine genset is not out of the question, which puts the propellant balance at 1300kg and endurance at 17 hours before the solar array trades better.
Even for multi-day or possibly multi-week coasts, just not having to deal with deploying and recovering a solar array may be worth the mass hit of the extra propellant, particularly if power demand is very peaky around brief operational periods.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/12/2022 04:19 am
Apart from Earth-Mars coast, on-board energy requirements could be met with a genset running off of on-board propellants.
A commercial 100kW genset running at 100% duty cycle would consume a little under 25kg LCH4 per hour (warmed to gaseous Methane, an ideal use of exhaust heat) along with somewhere under 50kg of LOX (also warmed, and you want to use less to avoid detonation from stoichiometric combustion. Exhaust gas recirculation may also help). So on the order of 75kg propellant mass per hour in orbit (at 100kW constant energy consumption). IIRC Ultraflex arrays are ~7kg/kW, so ~700kg for a 100kW array - only half duty cycle for many Earth orbits though due to the sun only being on one side, so ~1400kg really. Even if we assume a 1 tonne off-the-shelf genset (rather than something derived more from the aviation world) and ~400kg for propellant, your desired on-orbit lifetime bust be beyond 5 hours before a deployable array becomes the desirable option in terms of mass (and adds all the issues of an array that can be un-deployed, which may rule out ultra low mass options like Ultraflex or iROSA). The lighter your genset is, the better it trades: you probably can't approach the 10kW/kg of a gas turbine at such low power levels, but 100kg for a gas turbine genset is not out of the question, which puts the propellant balance at 1300kg and endurance at 17 hours before the solar array trades better.
Even for multi-day or possibly multi-week coasts, just not having to deal with deploying and recovering a solar array may be worth the mass hit of the extra propellant, particularly if power demand is very peaky around brief operational periods.

I've thought about methalox APUs for LSS during lunar night, but I'd never really considered it for cislunar or earth-orbit applications in general.  It's a good idea.

It does make sending a full tanker to refuel an Option B LSS in NRHO very, very tight, even if it's stretched to 1500t.  Using fast transit 115t LSS (95t dry + 20t crew module) needs about 450t of prop to do NRHO-LS-NRHO, which is just about what a 1500t tanker can bring from LEO if it wants to return direct to EDL.

Even if you assume that the tanker can make do with an average duty cycle of 20% of what you spec'd up above, and it needs a 3 week lifetime, that's about 7.6t of methalox, which is more than I could allocate for FPR.

However, I think it works if you use a BLT for the tanker.  That requires a lot more APU methalox, because now you're looking at a full lifetime of 60-90 days, which would be up to 32t of methalox.  But I think that still leaves you with enough prop to get back to EDL via fast flyby.  (I'm sure that there's a reverse BLT available too, but I have no clue what it costs.)

It's even kinda close to getting you to Mars.  You can do a minimum energy departure to Mars in late 2028 for C3=3km/s, which is about 3340m/s from a 300x300 parking orbit.  That's a 225 day flight.  At 50kW average, that would be 203t of methalox for the APU.  With only 90t of payload on a 1200t Starship, that's right at the hairy edge of doable.  However, I don't think you could get back to Earth unless you refueled in LMO.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/12/2022 01:51 pm
Informal poll: how long following a successful first launch before we see two SS's on orbit at the same time? That's a precursor to any attempt to transfer props.  I'd guess six months minimum. Probably more like a year.
Really optimistic, the first and second orbital Starships after a successful suborbital flight to Hawaii. Since it is likely both will be doing water landing to verify the powered pitch up maneuver works after reentry from orbit. For them to meet up and video each other's exterior in orbit should provide more data points.

More realistic, maybe the 4th and the 5th orbital Starships.

When there are Starships in close proximity in orbit. SpaceX will likely be recording 4k videos of the exteriors of the Starships. Then will there be movie light arrays on the Starship exterior for illumination?
Now you have me musing on how long an SS might stay alive on orbit. More batteries. Longer life. More batteries, fewer starlinks. PV will happen at some point. Power Consumption: comms and RCS solenoids. Later on, camera lighting. What else?


I wonder how hard it would be to substitute panels for the last two starlinks and keep them attached.
Starship HLS is contractually obligated to be able to loiter in NHRO for up to 200(?) days to wait if SLS/Orion launch is delayed. HLS will then take up to 14(?) days to complete its mission. I speculate without evidence that the SpaceX will use the same technologies for the other SS variants, with the exception of the solar panels that appear to be specialized for non-EDL starships.
I was noodling on quick n dirty for early transfer testing. It'll be awhile before they have the pads or turnaround speed to put a second ship up within the expected lifetime of a first ship. Other than transfer testing, there's no immediate need for PV.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/12/2022 01:57 pm
Informal poll: how long following a successful first launch before we see two SS's on orbit at the same time? That's a precursor to any attempt to transfer props.  I'd guess six months minimum. Probably more like a year.
Really optimistic, the first and second orbital Starships after a successful suborbital flight to Hawaii. Since it is likely both will be doing water landing to verify the powered pitch up maneuver works after reentry from orbit. For them to meet up and video each other's exterior in orbit should provide more data points.

More realistic, maybe the 4th and the 5th orbital Starships.

When there are Starships in close proximity in orbit. SpaceX will likely be recording 4k videos of the exteriors of the Starships. Then will there be movie light arrays on the Starship exterior for illumination?
Now you have me musing on how long an SS might stay alive on orbit. More batteries. Longer life. More batteries, fewer starlinks. PV will happen at some point. Power Consumption: comms and RCS solenoids. Later on, camera lighting. What else?


I wonder how hard it would be to substitute panels for the last two starlinks and keep them attached.
Starship HLS is contractually obligated to be able to loiter in NHRO for up to 200(?) days to wait if SLS/Orion launch is delayed. HLS will then take up to 14(?) days to complete its mission. I speculate without evidence that the SpaceX will use the same technologies for the other SS variants, with the exception of the solar panels that appear to be specialized for non-EDL starships.
I was noodling on quick n dirty for early transfer testing. It'll be awhile before they have the pads or turnaround speed to put a second ship up within the expected lifetime of a first ship. Other than transfer testing, there's no immediate need for PV.
But Depot is non-EDL, so it can use some variant of the HLS solar panels, which do need to be developed "soon". Unless those panels interfere with Depot's thermal management scheme or its docking/transfer scheme.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/12/2022 03:04 pm
Apart from Earth-Mars coast, on-board energy requirements could be met with a genset running off of on-board propellants.
A commercial 100kW genset running at 100% duty cycle would consume a little under 25kg LCH4 per hour (warmed to gaseous Methane, an ideal use of exhaust heat) along with somewhere under 50kg of LOX (also warmed, and you want to use less to avoid detonation from stoichiometric combustion. Exhaust gas recirculation may also help). So on the order of 75kg propellant mass per hour in orbit (at 100kW constant energy consumption). IIRC Ultraflex arrays are ~7kg/kW, so ~700kg for a 100kW array - only half duty cycle for many Earth orbits though due to the sun only being on one side, so ~1400kg really. Even if we assume a 1 tonne off-the-shelf genset (rather than something derived more from the aviation world) and ~400kg for propellant, your desired on-orbit lifetime bust be beyond 5 hours before a deployable array becomes the desirable option in terms of mass (and adds all the issues of an array that can be un-deployed, which may rule out ultra low mass options like Ultraflex or iROSA). The lighter your genset is, the better it trades: you probably can't approach the 10kW/kg of a gas turbine at such low power levels, but 100kg for a gas turbine genset is not out of the question, which puts the propellant balance at 1300kg and endurance at 17 hours before the solar array trades better.
Even for multi-day or possibly multi-week coasts, just not having to deal with deploying and recovering a solar array may be worth the mass hit of the extra propellant, particularly if power demand is very peaky around brief operational periods.

I've thought about methalox APUs for LSS during lunar night, but I'd never really considered it for cislunar or earth-orbit applications in general.  It's a good idea.

It does make sending a full tanker to refuel an Option B LSS in NRHO very, very tight, even if it's stretched to 1500t.  Using fast transit 115t LSS (95t dry + 20t crew module) needs about 450t of prop to do NRHO-LS-NRHO, which is just about what a 1500t tanker can bring from LEO if it wants to return direct to EDL.

Even if you assume that the tanker can make do with an average duty cycle of 20% of what you spec'd up above, and it needs a 3 week lifetime, that's about 7.6t of methalox, which is more than I could allocate for FPR.

However, I think it works if you use a BLT for the tanker.  That requires a lot more APU methalox, because now you're looking at a full lifetime of 60-90 days, which would be up to 32t of methalox.  But I think that still leaves you with enough prop to get back to EDL via fast flyby.  (I'm sure that there's a reverse BLT available too, but I have no clue what it costs.)

It's even kinda close to getting you to Mars.  You can do a minimum energy departure to Mars in late 2028 for C3=3km/s, which is about 3340m/s from a 300x300 parking orbit.  That's a 225 day flight.  At 50kW average, that would be 203t of methalox for the APU.  With only 90t of payload on a 1200t Starship, that's right at the hairy edge of doable.  However, I don't think you could get back to Earth unless you refueled in LMO.
Do we have even a rudimentary energy budget for an SS just hanging out on orbit? 100kW seems excessive. It needs comms, attitude awareness and RCS control, GPS and some level of internal sensing. Most of this can go to low sample rates until some active measure is needed. I'd WAG that 10kW would be more than adequate.


Proximity operations are a different story. Higher comms data rates, inter-ship comms, high rate of sensor sampling, radar, beacons and for at least one ship, increased RCS use.


A depot would have to do this repeatedly. A tanker, once per flight. If the tanker keeps a passive role for hookup & transfer, most of the power load is on the depot. If the depot is intended return, conformal PV is questionable. If doesn't have tiles and fins it looks better for conformal. An active cooler would need more than what conformal PV can produce.


My predictions*:
- early tests will use PV extended from the Pez dispenser to keep itself alive long enough for a second ship to reach it for proof of concept propellant transfer.
- if depot design moves beyond a tanker on TDY, it will shed the tiles and fins and use conformal PV until active cooling becomes a thing. From that point on, it will use non conformal PV. An APU trades well for short duration use. For long duration not so well. As easy as SX makes it look, it's not and all too often things take longer than you'd like.


*I predicted a bull market for 1929.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/12/2022 03:24 pm
Informal poll: how long following a successful first launch before we see two SS's on orbit at the same time? That's a precursor to any attempt to transfer props.  I'd guess six months minimum. Probably more like a year.
Really optimistic, the first and second orbital Starships after a successful suborbital flight to Hawaii. Since it is likely both will be doing water landing to verify the powered pitch up maneuver works after reentry from orbit. For them to meet up and video each other's exterior in orbit should provide more data points.

More realistic, maybe the 4th and the 5th orbital Starships.

When there are Starships in close proximity in orbit. SpaceX will likely be recording 4k videos of the exteriors of the Starships. Then will there be movie light arrays on the Starship exterior for illumination?
Now you have me musing on how long an SS might stay alive on orbit. More batteries. Longer life. More batteries, fewer starlinks. PV will happen at some point. Power Consumption: comms and RCS solenoids. Later on, camera lighting. What else?


I wonder how hard it would be to substitute panels for the last two starlinks and keep them attached.
Starship HLS is contractually obligated to be able to loiter in NHRO for up to 200(?) days to wait if SLS/Orion launch is delayed. HLS will then take up to 14(?) days to complete its mission. I speculate without evidence that the SpaceX will use the same technologies for the other SS variants, with the exception of the solar panels that appear to be specialized for non-EDL starships.
I was noodling on quick n dirty for early transfer testing. It'll be awhile before they have the pads or turnaround speed to put a second ship up within the expected lifetime of a first ship. Other than transfer testing, there's no immediate need for PV.
But Depot is non-EDL, so it can use some variant of the HLS solar panels, which do need to be developed "soon". Unless those panels interfere with Depot's thermal management scheme or its docking/transfer scheme.
True (a shared opinion) that a depot will not be EDL capable, but for those very first tests they'll just use two StarLink delivery ships. Hard to say what the launch interval will be when they first put two ships in proximity but I'd bet against battery power being adequate to keep the first ship alive.


I think the only deployable PV that SX has played with is on StarLink. We should look to that for the model they will first investigate. If two StarLink v2's can fit through the Pez dispenser at least two Starlink panels can too. This may or may not be the seed model for future development but it sure is a convenient approach for an early quick n dirty.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/12/2022 04:26 pm
I think the only deployable PV that SX has played with is on StarLink. We should look to that for the model they will first investigate.
Dragon 1 also had deployable solar arrays. Like with the cold-gas RCS for early Starships (literally some Falcon 9 quads attached to the side) it is not inconceivable that a Dragon 1 array and aerocover (or multiples thereof) could be attached to the side of an early Starship if PV were needed early for some reason.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/12/2022 05:38 pm
True (a shared opinion) that a depot will not be EDL capable, but for those very first tests they'll just use two StarLink delivery ships.
This entire thread is mostly speculation, my posts included. We do not yet know that SpaceX intends to implement buddy transfers between arbitrary Starships. It's possible that Depot must be one of the two Starships for any transfer. Depot is inexpensive except for the docking and transfer hardware, which must be present anyway for the test, so I see no advantage is using any ship other than Depot plus whatever arbitrary ship they have available for the test.

In fact, we don't even know that an arbitrary Starship can transfer propellant to a Depot. The only SS that is required to transfer propellant to a Depot is the tanker. The only SS that is contractually required to be refuelled from a Depot is the HLS, but we know SpaceX wants to refill at least some other types as well.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/12/2022 07:58 pm
True (a shared opinion) that a depot will not be EDL capable, but for those very first tests they'll just use two StarLink delivery ships.
This entire thread is mostly speculation, my posts included. We do not yet know that SpaceX intends to implement buddy transfers between arbitrary Starships. It's possible that Depot must be one of the two Starships for any transfer. Depot is inexpensive except for the docking and transfer hardware, which must be present anyway for the test, so I see no advantage is using any ship other than Depot plus whatever arbitrary ship they have available for the test.

In fact, we don't even know that an arbitrary Starship can transfer propellant to a Depot. The only SS that is required to transfer propellant to a Depot is the tanker. The only SS that is contractually required to be refuelled from a Depot is the HLS, but we know SpaceX wants to refill at least some other types as well.
There's things to test before transferring props. Flying in formation for the ultimate test of min and max RCS impulse needed. Both ships active but under control of one or one active and one fully passive? Or somewhere in between? Tinkering data links between ships. Maybe some active vaguely Canadarm type struts at a much smaller scale for the hookup? Even working out the PV for the ship that has to wait on orbit has to be done. Get all this out of the way and then set up for full body contact to transfer a small amount of props.


No one of the prelim exploration is earth shattering. Most can be modeled but actually doing it counts for a lot. Except for the delayed EDL attempt and the bother of keeping a ship on orbit awaiting a later ship, there's little cost in this prelim work. It could be done with a couple StarLink launchers early in the testing campaign.


I think we agree on Depot Heavy vs a tanker on TDY but Elon might insist on exploring the light route for all the reasons we've heard here, before recognizing the genius to be tapped at NSF and building some flavor of Depot Heavy. 8)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/13/2022 11:36 am
can someone provide an illustration/photo/drawing of what the solar panels on the SS will look like / where attached / how deployed??
Great question. Everybody speculated. Nobody knows. Maybe different for different types of missions.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 06/13/2022 04:20 pm
Refueling at 200kmx200km saves another 2.7 tonnes per tanker.  I'm not sure how low you can go before drag and operational angst out weights the extra fuel.

Assuming that you're fairly confident that your depot's thrusters aren't going to fail and it comes launched with 150t of prop, that's a lot of balm to sooth operational angst.

That said, you can over-constrain your VLEO altitude pretty quickly if you want lots of windows to launch tankers to the depot.  I doubt all of these requirements can be met simultaneously:

1) Orbit accessible from both BC and the Cape.  To do this, you'll have to overfly Florida.  The ground track will be moving at a pretty good clip by then, but you have to avoid Tampa on the south and Jacksonville on the north.  Risk to the public gets lower as the ground track speeds up, but higher with higher population density.  I think there might be a hole at about Ocala, but it's not a slam-dunk.  That's an inclination of about 30.4º, if I did the azimuth calculation correctly. 

The other option, which doesn't overfly anything, is 32.7º inclination, which you could launch south from BC through the Yucatan Channel and north or south from the Cape, but you're starting to spend a bit more delta-v with that one, and it might be a bit too inclined for easy lunar access.

2) At least one access per day.  Even with just an access from one or the other of the two sites, you need an orbit with a period of one sidereal day / n - daily RAAN precession.  Finding an orbital period that would allow access from both sites will be very tough.

3) Altitude doesn't exceed about 400km, for the reasons we were discussing.

280km is 16 orbits per sidereal day.  You'd need to be a bit lower than this to offset the precession (which IIRC is negative).  Even with lots of prop, I'm sure that you'd find some people at NASA wearing their brown trousers doing ops that low.  But 15 orbits/day is 550km, which is way too high. 

No doubt there are some orbits with windows that are more or less than exactly 24 hours, but I'm too lazy to figure them out.


You must not forget that your launch point is moving East at ~0.4km/s so you want more than 16 orbits per sidereal day, you want 16 passes over any fixed meridian. This is 16 orbits only when you're in ~98° (SSO) inclination, the lower the inclination the greater the additional fraction must be. AFAIR for inclinations like 30° you'd have to get as low as 210km or so. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/17/2022 02:13 am
Refueling at 200kmx200km saves another 2.7 tonnes per tanker.  I'm not sure how low you can go before drag and operational angst out weights the extra fuel.

Assuming that you're fairly confident that your depot's thrusters aren't going to fail and it comes launched with 150t of prop, that's a lot of balm to sooth operational angst.

That said, you can over-constrain your VLEO altitude pretty quickly if you want lots of windows to launch tankers to the depot.  I doubt all of these requirements can be met simultaneously:

1) Orbit accessible from both BC and the Cape.  To do this, you'll have to overfly Florida.  The ground track will be moving at a pretty good clip by then, but you have to avoid Tampa on the south and Jacksonville on the north.  Risk to the public gets lower as the ground track speeds up, but higher with higher population density.  I think there might be a hole at about Ocala, but it's not a slam-dunk.  That's an inclination of about 30.4º, if I did the azimuth calculation correctly. 

The other option, which doesn't overfly anything, is 32.7º inclination, which you could launch south from BC through the Yucatan Channel and north or south from the Cape, but you're starting to spend a bit more delta-v with that one, and it might be a bit too inclined for easy lunar access.

2) At least one access per day.  Even with just an access from one or the other of the two sites, you need an orbit with a period of one sidereal day / n - daily RAAN precession.  Finding an orbital period that would allow access from both sites will be very tough.

3) Altitude doesn't exceed about 400km, for the reasons we were discussing.

280km is 16 orbits per sidereal day.  You'd need to be a bit lower than this to offset the precession (which IIRC is negative).  Even with lots of prop, I'm sure that you'd find some people at NASA wearing their brown trousers doing ops that low.  But 15 orbits/day is 550km, which is way too high. 

No doubt there are some orbits with windows that are more or less than exactly 24 hours, but I'm too lazy to figure them out.


You must not forget that your launch point is moving East at ~0.4km/s so you want more than 16 orbits per sidereal day, you want 16 passes over any fixed meridian. This is 16 orbits only when you're in ~98° (SSO) inclination, the lower the inclination the greater the additional fraction must be. AFAIR for inclinations like 30° you'd have to get as low as 210km or so.

When you add in the RAAN precession, this gets surprisingly non-linear and hard to predict.  I think the best you can do is to look for some orbital altitude (assuming circular orbits) and inclination that yields an orbital period:

T = sqrt (μ/r³)

and a precession rate:

p = -(3/2)*(RE²/r²)*(2π/T)*cos(i)

...where RE is Earth's equatorial radius, r if the orbital radius, and i the inclination.

So a valid launch window is when

mod(nT(⍵rot + ⍵p), 2π) ≈ 0  (note¹)

... where ⍵rot is Earth's rotation rate (about 7.29E-5 rad/s).

There's another set of solutions for descending orbital paths, but I'm way too lazy to figure out the trigonometry.  In general, I'd think that unless the inclination was extremely high, the difference between the ascending and descending arcs would be too small to get two launches in on the same orbit.

I don't have a closed form that simply picks an optimal altitude and inclination, but trial-and-error has a fairly nice sweet spot at 362km altitude, 27º inclination, which yields a window every 1 day, 26 mins.  Alignment for this is off by 0.07º, which sounds well within an easily correctable distance.

_________________
¹I'm having brain freeze on whether it's ⍵rot + ⍵p or ⍵rot - ⍵p.  A little help?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/17/2022 04:00 pm
I'm not seeing why this is so complicated. I would think you'd only need to launch your tanker into the same orbital plane as the depot but with a slightly lower perigee. Then let it catch up over the next couple of days.

When you simplify the problem to merely getting into the right orbital plane, assuming an orbital inclination equal to the latitude of KSC, I think you'd get one launch attempt per day from KSC, and that launch would be due east. Boca Chica, being a bit south of that would get two attempts, but they'd be so close together it'd probably be equivalent to one attempt a day--again, with a launch due east or close to it. Allowing for orbital precession just means you get to launch a bit more often than once a day, but not by much.

Am I wrong? Or (worse) am I missing the point somehow?


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/17/2022 07:34 pm
I'm not seeing why this is so complicated. I would think you'd only need to launch your tanker into the same orbital plane as the depot but with a slightly lower perigee. Then let it catch up over the next couple of days.

When you simplify the problem to merely getting into the right orbital plane, assuming an orbital inclination equal to the latitude of KSC, I think you'd get one launch attempt per day from KSC, and that launch would be due east. Boca Chica, being a bit south of that would get two attempts, but they'd be so close together it'd probably be equivalent to one attempt a day--again, with a launch due east or close to it. Allowing for orbital precession just means you get to launch a bit more often than once a day, but not by much.

Am I wrong? Or (worse) am I missing the point somehow?

It's a fair point that you can launch in-plane a lot more often than this and then just phase to get to rendezvous.  But that will require that tankers stay on-orbit longer, because phasing takes time.

Another feature of using windows that are both in-plane and in-phase is that they allow near-direct EDL for the tankers to return.  I don't know if that's a big deal or not.  This wades into yet another swamp concerning what return EDL approaches will be considered safe to the public.  We certainly have lots of evidence that the Shuttle was considered safe for an approach over large chunks of the southwest and southern US, as well as the Florida peninsula itself.  Will that be true for Starship?  I'd kinda think "yes", but the Shuttle had a lot of cross-range while Starship doesn't.

Given that there seems to be a pretty tolerant once-per-day in-plane/in-phase window with an altitude of 360km and inclinations anywhere from 27º to 30º (~30º would get you access to the same orbits from both BC and the Cape without flying over Tampa), I don't see why you wouldn't use that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/17/2022 07:56 pm
I'm not seeing why this is so complicated. I would think you'd only need to launch your tanker into the same orbital plane as the depot but with a slightly lower perigee. Then let it catch up over the next couple of days.

When you simplify the problem to merely getting into the right orbital plane, assuming an orbital inclination equal to the latitude of KSC, I think you'd get one launch attempt per day from KSC, and that launch would be due east. Boca Chica, being a bit south of that would get two attempts, but they'd be so close together it'd probably be equivalent to one attempt a day--again, with a launch due east or close to it. Allowing for orbital precession just means you get to launch a bit more often than once a day, but not by much.

Am I wrong? Or (worse) am I missing the point somehow?

It's a fair point that you can launch in-plane a lot more often than this and then just phase to get to rendezvous.  But that will require that tankers stay on-orbit longer, because phasing takes time.

Another feature of using windows that are both in-plane and in-phase is that they allow near-direct EDL for the tankers to return.  I don't know if that's a big deal or not.  This wades into yet another swamp concerning what return EDL approaches will be considered safe to the public.  We certainly have lots of evidence that the Shuttle was considered safe for an approach over large chunks of the southwest and southern US, as well as the Florida peninsula itself.  Will that be true for Starship?  I'd kinda think "yes", but the Shuttle had a lot of cross-range while Starship doesn't.

Given that there seems to be a pretty tolerant once-per-day in-plane/in-phase window with an altitude of 360km and inclinations anywhere from 27º to 30º (~30º would get you access to the same orbits from both BC and the Cape without flying over Tampa), I don't see why you wouldn't use that.
To optimize the launch cadence of the launch tower, you launch once a day, or twice a day for high inclination planes, and you EDL once a day, or twice a day for high inclination planes. Each tanker takes however long it needs for phasing in each direction. You use as many tankers as needed to maintain this schedule. By comparison to the rest of the mission cost, the amortized capital cost of the extra tankers is inconsequential. If total phasing time is ten days, you need ten extra tankers for once-a-day launches.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/17/2022 08:58 pm
To optimize the launch cadence of the launch tower, you launch once a day, or twice a day for high inclination planes, and you EDL once a day, or twice a day for high inclination planes. Each tanker takes however long it needs for phasing in each direction. You use as many tankers as needed to maintain this schedule. By comparison to the rest of the mission cost, the amortized capital cost of the extra tankers is inconsequential. If total phasing time is ten days, you need ten extra tankers for once-a-day launches.

Pre-rendezvous phasing costs something in terms of boil-off.  Post-transfer phasing for EDL is cheaper, but there are still ops costs associated with stuff loitering on-orbit.

But it seems that this is a moot issue, because a direct-ascent window with high cadence is available.   Why wouldn't you use it?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 06/17/2022 11:10 pm
To optimize the launch cadence of the launch tower, you launch once a day, or twice a day for high inclination planes, and you EDL once a day, or twice a day for high inclination planes. Each tanker takes however long it needs for phasing in each direction. You use as many tankers as needed to maintain this schedule. By comparison to the rest of the mission cost, the amortized capital cost of the extra tankers is inconsequential. If total phasing time is ten days, you need ten extra tankers for once-a-day launches.

Pre-rendezvous phasing costs something in terms of boil-off.  Post-transfer phasing for EDL is cheaper, but there are still ops costs associated with stuff loitering on-orbit.

But it seems that this is a moot issue, because a direct-ascent window with high cadence is available.   Why wouldn't you use it?

Fire up Kerbal and do some in-space refueling.  You will learn a lot.  (direct-ascent window is the best after actually trying it, but trajectory programming direct-from-launch is hard in Kerbal, it'll take some work).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 06/18/2022 03:18 am
Kerbal kind of biases your intuition in weird ways.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/18/2022 03:45 am
Kerbal kind of biases your intuition in weird ways.

My experience is that my intuition is wrong when it comes to any kind of orbital mechanics.  I often get the math wrong, but when it's finally right, I trust it way more than intuition.

(direct-ascent window is the best after actually trying it...).

I think a more likely scenario than a perfect direct ascent is an ascent that's only a little bit out of phase.  It still requires getting a good state vector, computing a rendezvous burn, and then doing the full suite of proximity and docking ops.  But the state vector doesn't have to be as good (and may be computable relative to the target), and the actual transit time for phasing a degree or two is almost negligible.  It's the difference between 15-20 hours of rendezvous ops and a couple of hours.  And with boil-off, time is propellant.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/18/2022 02:03 pm
And with boil-off, time is propellant.
Is that really going to be such a problem, though? Have a look at this paper:
Quote from: Jarred Wilhite, Jason Wendell" link="https://tfaws.nasa.gov/wp-content/uploads/TFAWS2020-CT-103-Wilhite-Paper.pdf
The combination of the thin silver layer with the scattering layer of solar white material results in
a solar reflector that can theoretically reflect 99.9% of the Sun’s energy when used in deep space.
An analytical model in Phase I of this NIAC study showed that a sphere in deep space can reach
cryogenic temperatures (< 50 K) when covered with a 10 mm thick coating of solar white
material, located 1 astronomical unit (1 AU) from the Sun. Thus, application of this solar white
coating has far-reaching benefits, with the most significant being the long-term, deep space
storage of cryogenic liquids for future lunar and planetary missions (i.e. liquid oxygen).

50K is easily lower than the melting point for oxygen and methane.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/18/2022 02:46 pm
And with boil-off, time is propellant.
Is that really going to be such a problem, though? Have a look at this paper:
Quote from: Jarred Wilhite, Jason Wendell" link="https://tfaws.nasa.gov/wp-content/uploads/TFAWS2020-CT-103-Wilhite-Paper.pdf
The combination of the thin silver layer with the scattering layer of solar white material results in
a solar reflector that can theoretically reflect 99.9% of the Sun’s energy when used in deep space.
An analytical model in Phase I of this NIAC study showed that a sphere in deep space can reach
cryogenic temperatures (< 50 K) when covered with a 10 mm thick coating of solar white
material, located 1 astronomical unit (1 AU) from the Sun. Thus, application of this solar white
coating has far-reaching benefits, with the most significant being the long-term, deep space
storage of cryogenic liquids for future lunar and planetary missions (i.e. liquid oxygen).

50K is easily lower than the melting point for oxygen and methane.
Tanker is an EDL-capable starship. It's black (but insulated) on the bottom, ans the top will be subjected to plasma on every re-entry. I don't think we can count on a nice solar white coating.

I also don't think it matters that much. Tanker boil-off per day is a very small percentage of the total propellant budget.

Boil-off is a much bigger issue for Depot and for any long-duration Starship, including HLS and Mars missions. For Depot, a separate sunshade is feasible if it is needed. Something like the JWST shield could be deployed by a separate cargo SS. This works for Depot because Depot will remain in orbit and will be reused for many missions. However, it won't work for Mars transits, and whatever solution is used for Mars transit may be adequate for Depot also.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/18/2022 08:46 pm
I also don't think it matters that much. Tanker boil-off per day is a very small percentage of the total propellant budget.

The Starship to be fueled doesn't care about your percentage; it cares how much prop you can transfer to it for its mission.  Unless you have high-cadence BEO mission ops (not the current plan), or you manage to come up with zero-boiloff tech for the depot right from the git-go, then every drop of prop for the mission is delivered by lift tankers dedicated to the mission campaign.  If your nth tanker comes up a couple of tonnes short on prop, then you'll need an n+1th tanker, to deliver those last couple of tonnes--at exactly the same launch cost.

I agree with you that the only insulation on a lift tanker is the TPS on the ventral side, and that's not really designed to maximize emissivity and reflectance.  I'm willing to believe that you can get the average heat load that's absorbed down to 50W/m².  Starship has roughly 25m-long tanks, which would make the total heat rate into them 35kW.

Enthalpies of vaporization:

LOX: 6.82kJ/mol = 213kJ/kg
LCH4: 8.18kJ/mol = 511kJ/kg

With O:F=3.6, that should average out to 278kJ/kg of prop.  At a 35kW leak rate, that's 126g/s of boil-off = 450kg/hr.

This is, admittedly, a complete arm-wave on the leak rate.  But if you buy the arm-wave, the difference between a 5-hour RPOD and a 15-hour RPOD is 4.5t of prop.  That's nothing to sneeze at.

And again, unless my math is wrong (always a possibility), this just isn't that hard to do.  So if you can engineer the orbit to reduce RPOD time, why wouldn't you?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: warp99 on 06/19/2022 03:08 am
I also don't think it matters that much. Tanker boil-off per day is a very small percentage of the total propellant budget.

The Starship to be fueled doesn't care about your percentage; it cares how much prop you can transfer to it for its mission.  Unless you have high-cadence BEO mission ops (not the current plan), or you manage to come up with zero-boiloff tech for the depot right from the git-go, then every drop of prop for the mission is delivered by lift tankers dedicated to the mission campaign.  If your nth tanker comes up a couple of tonnes short on prop, then you'll need an n+1th tanker, to deliver those last couple of tonnes--at exactly the same launch cost.

I agree with you that the only insulation on a lift tanker is the TPS on the ventral side, and that's not really designed to maximize emissivity and reflectance.  I'm willing to believe that you can get the average heat load that's absorbed down to 50W/m².  Starship has roughly 25m-long tanks, which would make the total heat rate into them 35kW.

Enthalpies of vaporization:

LOX: 6.82kJ/mol = 213kJ/kg
LCH4: 8.18kJ/mol = 511kJ/kg

With O:F=3.6, that should average out to 278kJ/kg of prop.  At a 35kW leak rate, that's 126g/s of boil-off = 450kg/hr.

This is, admittedly, a complete arm-wave on the leak rate.  But if you buy the arm-wave, the difference between a 5-hour RPOD and a 15-hour RPOD is 4.5t of prop.  That's nothing to sneeze at.

And again, unless my math is wrong (always a possibility), this just isn't that hard to do.  So if you can engineer the orbit to reduce RPOD time, why wouldn't you?
You seem to be assuming the 50W/m2 over the full circumference but if you assume it is mainly solar heat gain then the effective surface area decreases to 9 x 25m and the heat gain reduces to 11kW.  Since the depot is likely to be in a low inclination orbit it will spend nearly 50% of the time in shadow which further reduces the heat gain to say 6kW.  Suddenly the mass losses become more reasonable.

The other effect is that the propellant is sub-cooled so there will be no evaporation at all until it has heated up to the boiling point at the working tank ullage pressure which could be up to 6 bar to facilitate transfer to the depot.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/19/2022 04:49 am
You seem to be assuming the 50W/m2 over the full circumference but if you assume it is mainly solar heat gain then the effective surface area decreases to 9 x 25m and the heat gain reduces to 11kW.  Since the depot is likely to be in a low inclination orbit it will spend nearly 50% of the time in shadow which further reduces the heat gain to say 6kW.  Suddenly the mass losses become more reasonable.

The other effect is that the propellant is sub-cooled so there will be no evaporation at all until it has heated up to the boiling point at the working tank ullage pressure which could be up to 6 bar to facilitate transfer to the depot.

You're putting way, way too much analysis into a guess that's way more primitive than figuring out the view factor.  And we can argue boiloff rates forever and it still won't refute my initial argument, which was that there's very little reason not to use an orbit that allows a daily near-direct ascent.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Thrustpuzzle on 06/19/2022 05:11 am
Any boiling due to solar heating doesn't need to vent the gas. I'd expect a simple system that just used passive pressure regulation to use the excess warmer gas to cool the rest. Tank pressure over a threshold of say 2 bar? Open valves to flow GCH4 and GO2 to a solid state fuel cell (or mechanical genset) and vent the H2O and CO2 exhaust. Use the power generated to run a cryocooler.  The higher the solar heating, the more gaseous excess you have to generate more power for more cooling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: schuttle89 on 06/20/2022 04:35 pm
Any boiling due to solar heating doesn't need to vent the gas. I'd expect a simple system that just used passive pressure regulation to use the excess warmer gas to cool the rest. Tank pressure over a threshold of say 2 bar? Open valves to flow GCH4 and GO2 to a solid state fuel cell (or mechanical genset) and vent the H2O and CO2 exhaust. Use the power generated to run a cryocooler.  The higher the solar heating, the more gaseous excess you have to generate more power for more cooling.
Then you need to radiate the heat from the fuel cell and the heat from the cooling loop. Thermodynamics is fun!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BT52 on 06/20/2022 06:26 pm
Good point i agree. Seemingly quite robust system and quite necessary for efficient orbital fuel depo.

But is skin heating  really not possible to manage with huge "shade" of solar arrays? I know that would be quite huge array and moon reflectivity make things even more complex.

Array could serve as JWST  umbrella shade. That array could be used as power source with PV.  I believe main power generator would be electricity anyway and batteries. I know not so power dense but when u are in orbit weight doesn't really matter if u can reach determinated orbit and delta v where is limmit really. Put 10 Tesla pack and call it a day. :) 

Still at end of day you can  still use fuel cell if more power is needed. 

Quite fun project. So many possibilities.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/21/2022 04:25 am
Good point i agree. Seemingly quite robust system and quite necessary for efficient orbital fuel depo.

Just remember that there are two different on-orbit components:  the depots and the lift tankers.  You can put as much effort into the depot to get to zero boiloff as you want, but the lift tanker can't support most of those technologies, because it needs to return to EDL.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/21/2022 04:11 pm
Good point i agree. Seemingly quite robust system and quite necessary for efficient orbital fuel depo.

Just remember that there are two different on-orbit components:  the depots and the lift tankers.  You can put as much effort into the depot to get to zero boiloff as you want, but the lift tanker can't support most of those technologies, because it needs to return to EDL.
Some thoughts on EDL. Most of us recognize the tankers as a dedicated variant. How variant might it get? If the thermal tiles were extended to cover the entire ship that would add what, 80%? We've guesstimated 10t current weight? Add another 8t? If anybody has the numbers handy a fact check would be much appreciated.


What impact would this have on boiloff?  With a semi-WAG of 4.5t/hour unprotected, an 8t solution with perfect insulation equals 1.8 hours boiloff. If half that it equals 3.6 hours boiloff. In an ideal world all tanker launches would minimize loiter. In an ideal world we'd already be on Mars.


Another thought is a tanker will never EDL with payload. ISTM the fore fins are not needed for pitch during EDL. Roll & yaw are a different story. Any nifty mass reduction ideas that could ditch the fore fins? Maybe just smaller?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 06/21/2022 05:33 pm
Another thought is a tanker will never EDL with payload. ISTM the fore fins are not needed for pitch during EDL. Roll & yaw are a different story. Any nifty mass reduction ideas that could ditch the fore fins? Maybe just smaller?
You don't want mass reduction; you want (total) cost reduction.

Lots of ideas in this thread involve special one offs with high engineering costs.  There's a lot to be said for using a standard tanker as a depot, if it can be made to work well enough.  If it can't be made to work, you want to look at the minimum necessary changes.  Only if the minimum is too large should you go for a complete redesign.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/21/2022 06:05 pm
Bingo. The trade to look for is whether the costs of designing, manufacturing, and maintaining a 'low boiloff depot' variant is more or less than flying the additional propellant to top off a near-stock-Starship used as a depot with a 'high' boiloff. Same with a 'high capacity depot': it needs to trade against just flying two near-stock-Starships as depots. The pendulum swings even further if those near-stock-Starships can spend the time they are not needed as depots (most of the time) instead flying payload missions, rather than being idle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 06/21/2022 06:17 pm
Bingo. The trade to look for is whether the costs of designing, manufacturing, and maintaining a 'low boiloff depot' variant is more or less than flying the additional propellant to top off a near-stock-Starship used as a depot with a 'high' boiloff. Same with a 'high capacity depot': it needs to trade against just flying two near-stock-Starships as depots. The pendulum swings even further if those near-stock-Starships can spend the time they are not needed as depots (most of the time) instead flying payload missions, rather than being idle.
Take another look at this Graphic from a NASA paper on IEEE Spectrum:
   https://ntrs.nasa.gov/api/citations/20220003725/downloads/22%203%207%20Kent%20IEEE%20paper.pdf
It's one of the few I've seen that at least tries to get the spacecraft (except Orion) to the same scale.  I know its notional, and possible the designer of the graphic didn't have any more info that we do, but look at the depot, the tankers, and the HLS.
    https://commons.wikimedia.org/wiki/File:Artemis_III_CONOPS.svg
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 06/21/2022 09:13 pm
Bingo. The trade to look for is whether the costs of designing, manufacturing, and maintaining a 'low boiloff depot' variant is more or less than flying the additional propellant to top off a near-stock-Starship used as a depot with a 'high' boiloff. Same with a 'high capacity depot': it needs to trade against just flying two near-stock-Starships as depots. The pendulum swings even further if those near-stock-Starships can spend the time they are not needed as depots (most of the time) instead flying payload missions, rather than being idle.
Any cost model has fixed costs and variable costs. In this case the cost of the depot and tankers would be the fixed cost and the cost of propellant, including boiloff, launch and recycle would be the variable cost. There's a lot of moving parts in this oversimplification.


In the "dedicated depot" model the depot cost would be high, but fixed. This fixed cost would include the R&D. The tanker costs would be low, probably the least expensive variant of all.


If it is possible to make a tanker do double duty as a depot it would need additional systems: some sort of gender bender QD; adequate maneuvering ability to dock/berth/connect; some way to keep the ships connected while settling and transferring propellant; long term power; maybe a sun shield. And all this needs to be an add on kit that does not distorted the OML too much during launch. If it's not an add on kit but part of the structure, it's a dedicated depot. If all tankers are modified to accept the depot gear, even if not installed, they will be more expensive and heavier.


I absolutely agree that SX will do the minimalist approach to get this concept to work. Once. Maybe twice. By the time they do it three times the extra add on kit will be getting more sophisticated and the usual trades will rear their ugly heads. My thinking is that the optimizations will point to just installing all the extras during the build and being done with it. Voila, a dedicated depot.


The first pass could well still be EDL capable. Reuse will drive down the fixed cost per campaign but I expect the SX penchant for tinkering will lead to sequential builds with increasing sophistication. As refueling needs go up and they become comfortable with the overall design and reliability, the trades will point towards keeping at least one on orbit in a busy lane. When one depot staying on orbit is capable of n campaigns, the lowered variable cost due to EDL, recycle and relaunch will make up for the higher cost of the depot with a limited lifetime of n+x campaigns. I have no idea of the value of n or x.


Methinks a depot staying on orbit might actually have a higher campaign life than one that repeatedly goes through launch and EDL. Still, not all launches can be expected to conveniently line up with a permanently on orbit depot. These would need a dedicated depot and would most likely be considered an expensive mission. These depots could be an EDL capable sub-variant or maybe left on orbit 'just in case'.


One last point to ponder: the actual build cost of these ships, all variants, promises to be, by rocket standards, not all that high. Their heritage is a plate stainless water tank with legs and engine, built by a water tank company in open salt air, with expectation of mass production. Operations will be the expensive part.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/22/2022 01:09 pm
Bingo. The trade to look for is whether the costs of designing, manufacturing, and maintaining a 'low boiloff depot' variant is more or less than flying the additional propellant to top off a near-stock-Starship used as a depot with a 'high' boiloff. Same with a 'high capacity depot': it needs to trade against just flying two near-stock-Starships as depots. The pendulum swings even further if those near-stock-Starships can spend the time they are not needed as depots (most of the time) instead flying payload missions, rather than being idle.
Take another look at this Graphic from a NASA paper on IEEE Spectrum:
   https://ntrs.nasa.gov/api/citations/20220003725/downloads/22%203%207%20Kent%20IEEE%20paper.pdf
It's one of the few I've seen that at least tries to get the spacecraft (except Orion) to the same scale.  I know its notional, and possible the designer of the graphic didn't have any more info that we do, but look at the depot, the tankers, and the HLS.
    https://commons.wikimedia.org/wiki/File:Artemis_III_CONOPS.svg
That same document depicts two different HLS Starship designs on adjacent pages. Concept art is concept art, and we know for SpaceX concept art often lags actual hardware let alone future designs (e.g. the images in the printouts stuck to in-production ring segments).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/22/2022 02:00 pm
I had an idea this morning; let me know if it makes any sense. The idea is that it might be simpler to launch tanks of fuel/oxidizer but never transfer propellant from one tank to another; simply move entire tanks.

Obviously Starship isn't designed to let you pop out the old fuel tank and pop in a new one, but if the lunar Starship is to be a variant that doesn't have to EDL through atmosphere, perhaps it makes sense for it to be "open" in that sense. If tanks are pluggable in that sense, you'd just send up (and plug in) as many as you needed for a particular mission--whether it were to Mars or the moon.

I realize that adds a set of (possibly really expensive) changes, but it completely eliminates the challenge of transferring propellant from one tank to another in zero-g.

What got me to thinking about this was the idea that you could make these tanks be zero boiloff for O2 and CH4. Then your "depot" could be little more than a scaffold for holding a few dozen tanks. Normal Starships could bring them up full and (if desired) bring them back empty.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: andrewmcleod on 06/22/2022 02:23 pm
I had an idea this morning; let me know if it makes any sense. The idea is that it might be simpler to launch tanks of fuel/oxidizer but never transfer propellant from one tank to another; simply move entire tanks.

Obviously Starship isn't designed to let you pop out the old fuel tank and pop in a new one, but if the lunar Starship is to be a variant that doesn't have to EDL through atmosphere, perhaps it makes sense for it to be "open" in that sense. If tanks are pluggable in that sense, you'd just send up (and plug in) as many as you needed for a particular mission--whether it were to Mars or the moon.

I realize that adds a set of (possibly really expensive) changes, but it completely eliminates the challenge of transferring propellant from one tank to another in zero-g.

What got me to thinking about this was the idea that you could make these tanks be zero boiloff for O2 and CH4. Then your "depot" could be little more than a scaffold for holding a few dozen tanks. Normal Starships could bring them up full and (if desired) bring them back empty.

The Starship _is_ the tanks. The tanks are not housed within the Starship walls; the Starship walls *are* the walls of the tanks. It's basically two mostly cylindrical tanks with a skirt at the bottom, a cone on the top, and a cylinder joining them together in the middle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 06/22/2022 04:37 pm
I had an idea this morning; let me know if it makes any sense. The idea is that it might be simpler to launch tanks of fuel/oxidizer but never transfer propellant from one tank to another; simply move entire tanks.

Obviously Starship isn't designed to let you pop out the old fuel tank and pop in a new one, but if the lunar Starship is to be a variant that doesn't have to EDL through atmosphere, perhaps it makes sense for it to be "open" in that sense. If tanks are pluggable in that sense, you'd just send up (and plug in) as many as you needed for a particular mission--whether it were to Mars or the moon.

I realize that adds a set of (possibly really expensive) changes, but it completely eliminates the challenge of transferring propellant from one tank to another in zero-g.

What got me to thinking about this was the idea that you could make these tanks be zero boiloff for O2 and CH4. Then your "depot" could be little more than a scaffold for holding a few dozen tanks. Normal Starships could bring them up full and (if desired) bring them back empty.

The Starship _is_ the tanks. The tanks are not housed within the Starship walls; the Starship walls *are* the walls of the tanks. It's basically two mostly cylindrical tanks with a skirt at the bottom, a cone on the top, and a cylinder joining them together in the middle.
Not even that. It's effectively ONE tank, with a single divider in the middle to keep the oxygen on the top and the methane in the bottom from mixing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/22/2022 07:09 pm
I see. And on a launch with no payload, those tanks would be about 20% full once the Starship reached orbit, right?

So to do anything like the approach I had in mind, they'd need to put separate O2 and/or CH4 tanks in the payload bay and carry those to orbit. Then some entirely new design of spacecraft would use them. It might or might not work, but it definitely doesn't seem to be what Elon has in mind.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/22/2022 07:35 pm
I see. And on a launch with no payload, those tanks would be about 20% full once the Starship reached orbit, right?

So to do anything like the approach I had in mind, they'd need to put separate O2 and/or CH4 tanks in the payload bay and carry those to orbit. Then some entirely new design of spacecraft would use them. It might or might not work, but it definitely doesn't seem to be what Elon has in mind.

The other issue is that docking a tank on a connector that has to be structurally sound at at least a couple of gees is a lot harder than docking two Starships on a connector that only has to withstand an ullage acceleration of 1-5mm/s².

However, there's one use case where separate tanks might be really interesting:  the StarKicker, the expendable, deep-space version that Elon has discussed.  Assuming that you jettison the fairing from the front of the vehicle during or after launch, you can dock several drop tanks onto the front of the payload, allowing you to reach even higher C3's than a standalone StarKicker.  The payload obviously has to be engineered with this docking in mind, and the plumbing to get the prop from the drop tanks and into the main downcomers would need to be in place.

Figuring out whether this is worthwhile isn't easy, and the advantage isn't huge.  For example, adding a 150t drop tank with 144t of prop to a 90t dry StarKicker with 1500t of onboard prop and a 20t payload only generates about 316m/s of extra delta-v, and additional drop tanks ahead of that would generate even less.  But there are a few missions (especially extrasolar or Kuiper Belt ones) where having some extra prop in the mains for an Oberth maneuver during a flyby can be pretty handy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: davis_404 on 06/22/2022 07:50 pm
Liquification temp of krypton is -251 F, liquification temp of oxygen is -297.

You see what I'm saying, sure. With a nuclear electric generator, the existing plumbing and cooling system of a deep space Ship can handle the krypton fuel in liquid form for huge array of SpaceX ion drives.

The acceleration would be miniscule, but the nature of ion propulsion is that the speed over time is much, much faster than achievable with a methane propulsion drive in a mighty initial burst.

Elon doesn't want to wait that long to go to Mars, but a outer system science probe mission would have no problem waiting a few years to get up to omigod speed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 06/22/2022 08:30 pm
Liquification temp of krypton is -251 F, liquification temp of oxygen is -297.

You see what I'm saying, sure. With a nuclear electric generator, the existing plumbing and cooling system of a deep space Ship can handle the krypton fuel in liquid form for huge array of SpaceX ion drives.

The acceleration would be miniscule, but the nature of ion propulsion is that the speed over time is much, much faster than achievable with a methane propulsion drive in a mighty initial burst.

Elon doesn't want to wait that long to go to Mars, but a outer system science probe mission would have no problem waiting a few years to get up to omigod speed.

I refer you to here (https://forum.nasaspaceflight.com/index.php?topic=55550.0), where this discussion won't be off-topic.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/22/2022 09:37 pm
The other issue is that docking a tank on a connector that has to be structurally sound at at least a couple of gees is a lot harder than docking two Starships on a connector that only has to withstand an ullage acceleration of 1-5mm/s².
I'm visualizing an open framework of girders that you slot the tanks into--using more tanks if you're going further. Hmm. That starts to sound a lot like my mental picture of the depot itself, except with a Raptor on one end and a payload on the other. I suppose even with just a single Raptor, it won't be possible to keep the g-load low, though, which makes for a much more robust structure than I originally had in mind.

In general, I don't think it's entirely unreasonable to expect that deep-space craft be very different from those that have to EDL through atmosphere and cope with higher gravity. But that doesn't seem to be what Elon wants, and I do see the value of having as few variant vehicles as possible.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 06/23/2022 10:33 am
By the time you've applied all the adaptations to both be capable of using connectible drop-tanks, and optimising for the use of connectible drop-tanks (e.g. structure able to support them, structure large enough to fit more than one, etc), what you have is a new vehicle that happens to use Raptor rather than a modified Starship. Not inherently a bad thing - exoatmospheric-only vehicles assembled in orbit have plenty of advantages of their own - but pretty separate from Starship development.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/23/2022 01:13 pm
By the time you've applied all the adaptations to both be capable of using connectible drop-tanks, and optimising for the use of connectible drop-tanks (e.g. structure able to support them, structure large enough to fit more than one, etc), what you have is a new vehicle that happens to use Raptor rather than a modified Starship. Not inherently a bad thing - exoatmospheric-only vehicles assembled in orbit have plenty of advantages of their own - but pretty separate from Starship development.
Yeah, I was coming to that conclusion too. So I guess it comes down to a question of whether (and how soon) Elon can make this in-orbit refueling work.

And it probably still makes sense for him to produce a version of Starship that never enters atmosphere. Save the weight of the tiles, at least.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 06/29/2022 02:09 pm
It occurs to me that the single most important component of the tanker, from a PR standpoint, is to somehow justify using the same orange foam insulation as the shuttle external tank.

The image of the docked craft paints the entire refueled system as "the next shuttle", with the winged craft attached to an orange tank. That alone will evoke all the unused ideas of the shuttle era for use of external tanks in space. Comparing a refueling starship (with the heat shield away from the "orange tank" depot) to a shuttle on the pad points out how starship avoids what happened in the second shuttle disaster, painting the starship as a SAFER shuttle.

It also gives an easy talking point about how "if we could have refueled a shuttle tank in orbit (like starship is doing), it could have flown to (destination)", which helps sell the refilling concept past the "but 20 dockings is BAD" FUD.

There's probably more PR benifits that come from an easy visual comparison to the shuttle era, but those are the big ones off the top of my head.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 06/29/2022 08:52 pm
It occurs to me that the single most important component of the tanker, from a PR standpoint, is to somehow justify using the same orange foam insulation as the shuttle external tank.
Uh, are you suggesting the depot should be painted orange? Since the depot needs to be zero-boil-off (or as close to it as possible), I think it's pretty likely to be white. And I really don't think the public needs any comparisons with the old Space Shuttle. Once Starship gets a few successful missions under its belt, they'll be happy with it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: whitelancer64 on 06/29/2022 09:35 pm
It occurs to me that the single most important component of the tanker, from a PR standpoint, is to somehow justify using the same orange foam insulation as the shuttle external tank.

*snip*

Nit: the SOFI foam is actually yellow, it turns orange over time from exposure to UV rays.

Also, the foam was one of the problems found in the many studies done on reuse of the Shuttle ET in orbit. The foam can shed from mechanical fracturing, outgassing, and debris strikes, resulting in more orbital debris.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AJW on 06/30/2022 01:11 am
In the fall of 1985, Ford Motor Company marketing decided to link their Ford Aerostar to the Shuttle through a print and television campaign.  This was great from a PR standpoint, until the following January.

https://apnews.com/article/4c8e02f598cf8154c267be124b373a2f



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/04/2022 10:11 pm
Here's a half-baked idea for a capture/docking system.  I'm not sure if it's rigid enough, but what if the active side had four booms stowed in chines on the upper dorsal surface of the Starship?  During docking, they'd deploy from the chines and make a sort of cradle for the other Starship to slide into.  The booms would slide along hard points on the passive Starship until they latched in place.

Note that I'm using the dorsal-to-dorsal, nose-to-tail docking orientation (suggested up-thread by somebody), which allows a depot or tanker "kit" payload to be mounted in the payload bay, with the active quick disconnect and its bellows to extend through a door to attached to the existing QD that will be on all Starships.  I'm pretty sure that I've laid this out to be completely hermaphroditic, although one side will almost certainly be active and the other passive.  The passive side can have hardware depopulated to save weight, but you still have the option of more complex conops where a tanker might need to be active and passive in the same mission.

Pros:

1) It's pretty simple.

2) The booms can be made to an arbitrary stiffness, so some flexibility can soak up positioning errors.  (Note that the latches that the booms use to grab the hard points are an arm-wave in the diagrams below.  I'd think that you'd want them a certain length down the boom, so the ends can deal with positioning errors.  This needs work.)

3) It's probably pretty lightweight.

4) It's pretty much androgynous.  I'd think that you'd typically only outfit a depot with the booms, leaving the lift tankers or payload Starships with only the mass penalty of the hard points, which is almost nothing.  But if you got into a situation where you needed to do tanker-tanker or tanker-payload hookups, instead of just tanker-depot, both vehicles can have booms and one can leave them stowed.  Similarly, you can have two depot kits and everything will still work.

5) I think that this "kit" idea can be extended to add in extra deployable solar panels for cryocooler power and heat rejection, and maybe even fold-out MLI to insulate the dorsal side of a depot.  Note that TPS might be an OK insulator for a depot in LEO, since albedo heat from earth will only build up for half an orbit, and will mostly radiate away on the night side.  However, in deep space, the TPS won't perform nearly as well as MLI.


Cons:

a) It's kinda flimsy.  There really isn't a hard dock involved here, and any imbalance in ullage thrust between the two Starships could overtax the booms.

b) I'm not super sure the docking dynamics work.  One of the booms is going to touch first, and apply torques until the other booms engage.  However, the flexibility ought to minimize the torques.

c) The boom joints at the base, where they attach to the skin, need to rotate in two directions.  There's also the complexity of chine doors to think about.

d) The booms might interfere with other payload doors.  However, if you only populate the booms on depots and lift tankers, this doesn't seem to be too much of a problem.  And the hard points are tiny and can go on any Starship.

As I said, kinda half-baked, but hopefully it adopts the best ideas from the last few pages of posts.  See what you think.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 07/05/2022 03:38 am
Jon Goff was working on StickyBooms with Altius, which might be appropriate for soft/hard capture work?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/05/2022 06:26 am
Jon Goff was working on StickyBooms with Altius, which might be appropriate for soft/hard capture work?

This would be a bit like sticky booms with a latch. But if the geometry works out, the booms would act as guides and shock absorbers as well.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/05/2022 10:52 am
I think the additional booms can be dispensed with as long as sufficiently low settling thrust is used. The 'receiving' Starship helpfully has its nose tip clear of the depot's base, so settling and alignment can be accomplished by the required tank venting: LOX venting from the base to provide settling (with modulation between the base thrust and non-perturbative venting to control acceleration) and CH4 venting from the nose from a tip vent (with modulation between the tip and non-perturbative venting to control angular acceleration) to 'point' the entire stack such that the total thrust vector is always aimed through the CoM. If torsion loads on the depot QD plate are too high even with microgee settling, these can be reduce to near pure shear loads by a single attachment point from the base of the depot to the nose of the receiving Starship, or by supplementary RCS thrust on the depot side.

Controlling QD mechanical loads by active RCS thrust sounds terrifying: formation flying of hundred-ton objects with zero clearance! But that's not actually the problem you need to solve: you don't really care about what position the two vehicles are flying relative to each other (i.e. you can dispense with a relative positioning system if not needed), you only actually care about the mechanical load on the QDs. By using load cells on the QDs (and on the arm the depot's QD is attached to) you can use actual measured loads to directly drive RCS impulses. From there your concern is not formation flying, but watching overall thrust accumulation to ensure the mated pair are not spinning themselves up excessively. To an extent - dependant on actual RCS thruster locations - using differential thrust on each vehicle to counter measured QD loads minimises the total effect of that thrust on the mated pair. As long as the QD plates can unlatch rapidly enough (and ideally internal valves can close rapidly enough, but a small release of flash-vaporised propellant from the QD plate mating volume is survivable and provides a little 'free' separation impulse) if mechanical loads start increasing towards a pre-set limit, the failure mode then becomes a prematurely aborted propellant transfer rather than a mechanical failure.

Note that I'm using the dorsal-to-dorsal, nose-to-tail docking orientation (suggested up-thread by somebody)
Yo.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/05/2022 03:31 pm
I think the additional booms can be dispensed with as long as sufficiently low settling thrust is used. The 'receiving' Starship helpfully has its nose tip clear of the depot's base, so settling and alignment can be accomplished by the required tank venting: LOX venting from the base to provide settling (with modulation between the base thrust and non-perturbative venting to control acceleration) and CH4 venting from the nose from a tip vent (with modulation between the tip and non-perturbative venting to control angular acceleration) to 'point' the entire stack such that the total thrust vector is always aimed through the CoM. If torsion loads on the depot QD plate are too high even with microgee settling, these can be reduce to near pure shear loads by a single attachment point from the base of the depot to the nose of the receiving Starship, or by supplementary RCS thrust on the depot side.

At the very least, you probably need something to hold the two ships together to make the QD connection.

Quote
Controlling QD mechanical loads by active RCS thrust sounds terrifying: formation flying of hundred-ton objects with zero clearance! But that's not actually the problem you need to solve: you don't really care about what position the two vehicles are flying relative to each other (i.e. you can dispense with a relative positioning system if not needed), you only actually care about the mechanical load on the QDs. By using load cells on the QDs (and on the arm the depot's QD is attached to) you can use actual measured loads to directly drive RCS impulses. From there your concern is not formation flying, but watching overall thrust accumulation to ensure the mated pair are not spinning themselves up excessively. To an extent - dependant on actual RCS thruster locations - using differential thrust on each vehicle to counter measured QD loads minimises the total effect of that thrust on the mated pair. As long as the QD plates can unlatch rapidly enough (and ideally internal valves can close rapidly enough, but a small release of flash-vaporised propellant from the QD plate mating volume is survivable and provides a little 'free' separation impulse) if mechanical loads start increasing towards a pre-set limit, the failure mode then becomes a prematurely aborted propellant transfer rather than a mechanical failure.

This sounds fine in nominal ops, but there are a lot of things that can go wrong.  Your load sensors can malfunction.  You can get a stuck RCS (vent gas, high pressure cold gas, hot gas, makes no nevermind).  You can lose a star tracker and wind up with a rotational rate.  So aborts could be... uncommon, but not that uncommon.  Which means you have to worry about failures of the QD system to detach, or to detach destructively if the tensile or shear loads are high enough. 

You also have to worry about disconnect torques that might cause a collision.  The booms might provide enough z-axis force to minimize collisions in a spin-up.

All that said, the booms, even if they're very flexible and forgiving once connected, are a lot more complicated to detach in an emergency than a single QD or QD+one-point connection.

Quote
Note that I'm using the dorsal-to-dorsal, nose-to-tail docking orientation (suggested up-thread by somebody)
Yo.

Yeah, sorry about that.  But I have an excuse!  I'm currently up in the Maine north woods with a 4G internet connection with an adequate downlink, but about 1kbps uplink.  Changing pages is something that one has to stop and consider carefully, especially if somebody else is streaming something.  Dropped ACKs send us into congestion control all the time.  I've got a Starlink coming tomorrow, and then we'll see if I can find a spot on the roof where the trees allow something approaching continuous coverage.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/05/2022 04:01 pm
I think the additional booms can be dispensed with as long as sufficiently low settling thrust is used. The 'receiving' Starship helpfully has its nose tip clear of the depot's base, so settling and alignment can be accomplished by the required tank venting: LOX venting from the base to provide settling (with modulation between the base thrust and non-perturbative venting to control acceleration) and CH4 venting from the nose from a tip vent (with modulation between the tip and non-perturbative venting to control angular acceleration) to 'point' the entire stack such that the total thrust vector is always aimed through the CoM. If torsion loads on the depot QD plate are too high even with microgee settling, these can be reduce to near pure shear loads by a single attachment point from the base of the depot to the nose of the receiving Starship, or by supplementary RCS thrust on the depot side.

At the very least, you probably need something to hold the two ships together to make the QD connection.
I'd just use the QD itself for that. It has alignment pins, it has driveable latches, and on the ship QD side it sits on the end of an 5-axis actuated (possibly 6-axis, not clear if the QD plate itself can 'roll' about its own short axis) arm. It seems rather redundant to add an additional arm with alignment and latching capability just to avoid using your arm with alignment and latching capability.
Quote
Quote
Controlling QD mechanical loads by active RCS thrust sounds terrifying: formation flying of hundred-ton objects with zero clearance! But that's not actually the problem you need to solve: you don't really care about what position the two vehicles are flying relative to each other (i.e. you can dispense with a relative positioning system if not needed), you only actually care about the mechanical load on the QDs. By using load cells on the QDs (and on the arm the depot's QD is attached to) you can use actual measured loads to directly drive RCS impulses. From there your concern is not formation flying, but watching overall thrust accumulation to ensure the mated pair are not spinning themselves up excessively. To an extent - dependant on actual RCS thruster locations - using differential thrust on each vehicle to counter measured QD loads minimises the total effect of that thrust on the mated pair. As long as the QD plates can unlatch rapidly enough (and ideally internal valves can close rapidly enough, but a small release of flash-vaporised propellant from the QD plate mating volume is survivable and provides a little 'free' separation impulse) if mechanical loads start increasing towards a pre-set limit, the failure mode then becomes a prematurely aborted propellant transfer rather than a mechanical failure.

This sounds fine in nominal ops, but there are a lot of things that can go wrong.  Your load sensors can malfunction.  You can get a stuck RCS (vent gas, high pressure cold gas, hot gas, makes no nevermind).  You can lose a star tracker and wind up with a rotational rate.  So aborts could be... uncommon, but not that uncommon.  Which means you have to worry about failures of the QD system to detach, or to detach destructively if the tensile or shear loads are high enough. 

You also have to worry about disconnect torques that might cause a collision.  The booms might provide enough z-axis force to minimize collisions in a spin-up.

All that said, the booms, even if they're very flexible and forgiving once connected, are a lot more complicated to detach in an emergency than a single QD or QD+one-point connection.
If you really need an extra connection point (to limit QD loads to pure shear rather than shear + torsion) a single point could be added for nose-to-tail at the other end. Store a short supplementary arm of very limited capability (only enough articulation to pack away) inside the engine skirt of the depot ship, as in the 'cargo boxes' seen on old renders. Flip this arm out and latch to a hardpoint on the dorsal side of the nose - a lasso around the nose is attractive from mechanical simplicity but risks tile damage.
Less elegant would be to add the arm to the nose section of the receiving Starship (mass added to all vehicles rather than just the depot) and have it align with and latch to the otherwise unused QD plate (with unoccupied alignment pin sockets and latch bars) on the depot ship.
Quote
Quote
Note that I'm using the dorsal-to-dorsal, nose-to-tail docking orientation (suggested up-thread by somebody)
Yo.
Yeah, sorry about that.  But I have an excuse!  I'm currently up in the Maine north woods with a 4G internet connection with an adequate downlink, but about 1kbps uplink.  Changing pages is something that one has to stop and consider carefully, especially if somebody else is streaming something.  Dropped ACKs send us into congestion control all the time.  I've got a Starlink coming tomorrow, and then we'll see if I can find a spot on the roof where the trees allow something approaching continuous coverage.
No worries. If you end up still bandwidth-constrained: right at the very bottom of the footer of the page is the line "RSS Mobile". Click on 'Mobile' and you get the ugly but functional web-based mobile view, which kills all off-site embeds and drops attachment resolution down.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/05/2022 05:20 pm
I'd just use the QD itself for that. It has alignment pins, it has driveable latches, and on the ship QD side it sits on the end of an 5-axis actuated (possibly 6-axis, not clear if the QD plate itself can 'roll' about its own short axis) arm. It seems rather redundant to add an additional arm with alignment and latching capability just to avoid using your arm with alignment and latching capability.

Assuming that the active QD is on the depot side, you can wind up pushing on a fairly light payload Starship (less than 200t).  How much force do you need to make it latch?  And what happens if you've given it a shove not anywhere near the CoM and it doesn't latch?  Seems like a fine way to spin the payload Starship's nose into the depot's tail.

At the very least, you need something to fend off the compressive spinning load in a latch failure.  Better still is something that can handle a tensile load, so you can jab at that sucker over and over if you need to.

Quote
If you really need an extra connection point (to limit QD loads to pure shear rather than shear + torsion) a single point could be added for nose-to-tail at the other end. Store a short supplementary arm of very limited capability (only enough articulation to pack away) inside the engine skirt of the depot ship, as in the 'cargo boxes' seen on old renders. Flip this arm out and latch to a hardpoint on the dorsal side of the nose - a lasso around the nose is attractive from mechanical simplicity but risks tile damage.
Less elegant would be to add the arm to the nose section of the receiving Starship (mass added to all vehicles rather than just the depot) and have it align with and latch to the otherwise unused QD plate (with unoccupied alignment pin sockets and latch bars) on the depot ship.

But then the arm needs some active articulation to find the grapple on the other side, QD or otherwise.  The nice thing about the booms is that you can have relatively sloppy prox ops and still wind up latched, because the cradling effect of four booms will steer the target toward the latches on the booms.  And even if you fail to latch, I don't think you can wind up with a rotational failure that can cause a collision.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/05/2022 06:17 pm
I'd just use the QD itself for that. It has alignment pins, it has driveable latches, and on the ship QD side it sits on the end of an 5-axis actuated (possibly 6-axis, not clear if the QD plate itself can 'roll' about its own short axis) arm. It seems rather redundant to add an additional arm with alignment and latching capability just to avoid using your arm with alignment and latching capability.

Assuming that the active QD is on the depot side, you can wind up pushing on a fairly light payload Starship (less than 200t).  How much force do you need to make it latch?  And what happens if you've given it a shove not anywhere near the CoM and it doesn't latch?  Seems like a fine way to spin the payload Starship's nose into the depot's tail.

At the very least, you need something to fend off the compressive spinning load in a latch failure.  Better still is something that can handle a tensile load, so you can jab at that sucker over and over if you need to.
The current latches appear to be purely rotary (i.e. no force imparted 'against' the plate). If a latching mechanism without active driving is required, the spherical tips of the alignment pins could pull double-duty by adding lateral detents such that fully seating the alignment pins adds a few newtons or tens or newtons of resistance to extraction.
Imparted rotary force does look like an issue in case of a 'bonk' by the QD extension mechanism. Maybe the dorsal side of the depot could inflate a few rows of concave bumpers during prox-ops so that any contact low velocity events are gentle bounces rather than hard impacts. Could even work for the stabilisation during settling thrust if the QD latches are strong enough to 'pull' the two vehicles together against the bumpers.
Quote
But then the arm needs some active articulation to find the grapple on the other side, QD or otherwise.  The nice thing about the booms is that you can have relatively sloppy prox ops and still wind up latched, because the cradling effect of four booms will steer the target toward the latches on the booms.  And even if you fail to latch, I don't think you can wind up with a rotational failure that can cause a collision.
The downside of the booms is that any misalignment rubs the effectors up and down the sides of the Starship, which are covered in tiles.
Some articulation may be needed for alignment, but it only needs to be fairly gross alignment (get the narrow tip of the alignment pin into the wide opening of the socket) which could be accomplished with a lockable complaint mechanism (rather than powered actuation).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/05/2022 07:10 pm
This idea sure has moved ahead. No or minimal separation between ships. An extendable QD plate in the payload bay and the ships hugging in a particularly erotic way. Maybe an arm boom or two between the ships.

Unless both ships are capable of RCS there is zero chance of thrust passing exactly through the CoM except maybe momentarily.   Without some thrust mods somewhere this sounds like a way to move orbital mechanics from deterministic to probabilistic. With both ships pointing in the same direction this becomes a non problem with no further mods needed.

With no standoff between ships there is zero margin for screwup. In my experience screwups sometimes happen no matter how good the planning. The trick is to have enough margin to give a reasonable chance of recovery.

This design moves from a mostly static QD on a standoff blister, admittedly needing a one time gender bender install, to a QD designed to extend from a mostly removed payload bay.

A point making this catawompus setup a little more realistic is that the depot side extendable QD plate and any booms need not be stowed after deploy. They'll be needed again.

The only advantage I can see in catawompus mounting is avoiding the dreaded gender/layout problem when the QD plates face each other and/or the difficulty in deploying a gender bender. The attached drawing shows a simple way around this.

The gender bender would stow for launching on the hull directly above the QD plate. The gender bender could have some streamlining on top but if possible, it should remain open faced to facilitate deploy. The two arms would be aligned such that they would swing the gender bender exactly onto the QD plate and be locked into place using standard QD locking hardware. The gender bender would have to duplicate this hardware in order to ply its trade. The deployment would be powered by a couple springs concentric to pivot points on the arms. The gender bender can be built arbitrarily deep, consistent with launch aerodynamics, to act as a standoff.

The biggest downside to this that I can come up with is the deployment arms would hinder the little bit of QD slop needed for final alignment. Sloppy hinges on the deploy arms might do. Mass should be the same or less than the payload bay QD. Side thought: if the payload bay QD has virtues I've not identified and ends up making sense, just build it into the ships hull like the main QD, and put a pantograph behind it. Deploy would be easier.

Something to keep in mind is that while micro g is most probably adequate to keep propellants settled, something more would most probably be needed to get it initially settled in an acceptable time frame. This seems to imply symmetrical RCS, which in turn implies the ships pointing in the same direction or some new RCS mods.

The idea of a boom or booms makes sense. It seems like a minimalist approach to adding just a little bit more to give a lot of margin. One good RCS belch might make for a bad day otherwise. All SS's have a place to be grabbed by the chopsticks. Two brace booms stowed on either side of the depot, would do the job. Make them telescoping with ullage powered double acting pistons and it has both mobility in that dimension and shock absorber functionality. If the chopstick grab points calls for grossly oversized brace ends, cobble up that suggested universally installed grab point.

ISTM the most reasonable place to stow the braces would be inside the strakes (what are we calling them now?) that the nonexistent fins mount through. A one time latch on a long door lets each brace deploy with the base end already where it needs to be for a wide stance. Motors on the base give two further axis of freedom. The grabber end may or may not need further degrees of freedom. Sure is starting to sound like a low end mini Canadarm.


Edit: went back and saw the notional payload bay QD is mounted in the wall.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/05/2022 07:43 pm
I'd just use the QD itself for that. It has alignment pins, it has driveable latches, and on the ship QD side it sits on the end of an 5-axis actuated (possibly 6-axis, not clear if the QD plate itself can 'roll' about its own short axis) arm. It seems rather redundant to add an additional arm with alignment and latching capability just to avoid using your arm with alignment and latching capability.

Assuming that the active QD is on the depot side, you can wind up pushing on a fairly light payload Starship (less than 200t).  How much force do you need to make it latch?  And what happens if you've given it a shove not anywhere near the CoM and it doesn't latch?  Seems like a fine way to spin the payload Starship's nose into the depot's tail.

At the very least, you need something to fend off the compressive spinning load in a latch failure.  Better still is something that can handle a tensile load, so you can jab at that sucker over and over if you need to.

Quote
If you really need an extra connection point (to limit QD loads to pure shear rather than shear + torsion) a single point could be added for nose-to-tail at the other end. Store a short supplementary arm of very limited capability (only enough articulation to pack away) inside the engine skirt of the depot ship, as in the 'cargo boxes' seen on old renders. Flip this arm out and latch to a hardpoint on the dorsal side of the nose - a lasso around the nose is attractive from mechanical simplicity but risks tile damage.
Less elegant would be to add the arm to the nose section of the receiving Starship (mass added to all vehicles rather than just the depot) and have it align with and latch to the otherwise unused QD plate (with unoccupied alignment pin sockets and latch bars) on the depot ship.

But then the arm needs some active articulation to find the grapple on the other side, QD or otherwise.  The nice thing about the booms is that you can have relatively sloppy prox ops and still wind up latched, because the cradling effect of four booms will steer the target toward the latches on the booms.  And even if you fail to latch, I don't think you can wind up with a rotational failure that can cause a collision.
I don't think it matters which side is active and which side is light. The force is what it is. The relative mass is what it is. Low mass moves a lot. High mass, not so much.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/05/2022 09:09 pm
I'd just use the QD itself for that. It has alignment pins, it has driveable latches, and on the ship QD side it sits on the end of an 5-axis actuated (possibly 6-axis, not clear if the QD plate itself can 'roll' about its own short axis) arm. It seems rather redundant to add an additional arm with alignment and latching capability just to avoid using your arm with alignment and latching capability.

Assuming that the active QD is on the depot side, you can wind up pushing on a fairly light payload Starship (less than 200t).  How much force do you need to make it latch?  And what happens if you've given it a shove not anywhere near the CoM and it doesn't latch?  Seems like a fine way to spin the payload Starship's nose into the depot's tail.

At the very least, you need something to fend off the compressive spinning load in a latch failure.  Better still is something that can handle a tensile load, so you can jab at that sucker over and over if you need to.
The current latches appear to be purely rotary (i.e. no force imparted 'against' the plate). If a latching mechanism without active driving is required, the spherical tips of the alignment pins could pull double-duty by adding lateral detents such that fully seating the alignment pins adds a few newtons or tens or newtons of resistance to extraction.
Imparted rotary force does look like an issue in case of a 'bonk' by the QD extension mechanism. Maybe the dorsal side of the depot could inflate a few rows of concave bumpers during prox-ops so that any contact low velocity events are gentle bounces rather than hard impacts. Could even work for the stabilisation during settling thrust if the QD latches are strong enough to 'pull' the two vehicles together against the bumpers.
Quote
But then the arm needs some active articulation to find the grapple on the other side, QD or otherwise.  The nice thing about the booms is that you can have relatively sloppy prox ops and still wind up latched, because the cradling effect of four booms will steer the target toward the latches on the booms.  And even if you fail to latch, I don't think you can wind up with a rotational failure that can cause a collision.
The downside of the booms is that any misalignment rubs the effectors up and down the sides of the Starship, which are covered in tiles.
Some articulation may be needed for alignment, but it only needs to be fairly gross alignment (get the narrow tip of the alignment pin into the wide opening of the socket) which could be accomplished with a lockable complaint mechanism (rather than powered actuation).
I have to agree that QD latching isn't like sticking a chuck into a service station air line. Most probably rotating and actively powered as you say. A collet, actively clamped onto a simple shaft by a concentric motor would have a lot of tensile strength, and drawing the clamped collets in via an arm waving mechanism would allow an even distribution of force across the QD plate that would be independent of wear. Disconnect might be a problem as lathe collets often need the workpiece to be lightly tapped with a hammer. Maybe the arm waved draw mechanism can do double duty.


ISTM that using the QD plate to do anything more than float a bit to deal with that last bit of misalignment would be asking for trouble. It would be the fulcrum of a very long lever with cryo plumbing on the back side. The shuttle admittedly did this sorta kinda when it docked to the ISS but it didn't cycle the connection every 12-24 hours. How much time did the shuttle spend lining up with the docking mechanism and shedding torque? If it takes a day it won't work for this application.


Unless there is a separate deployment mechanism, any booms would have to be self deploying from their storage location. If there is danger of boom misalignment leading to ANY type of damage there is an argument for Active Boom Control (ABC, my first TLA). ABC doesn't have to have particularly powerful motors on its base or end effector but it will need a bit of power for extension and retraction. With two booms as describe up thread, once the booms have attached to the non-depot ship, and the QD is connected, just reel the booms in with all pivots free floating. Geometry will rule and as the geometry (trig really) hits the intended location, lock the joints and slightly pressurize the telescoping booms to act as final bracing.


Note: all of this is independent of the QD being fore or aft.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/05/2022 10:13 pm
I just realized that all we've been discussing on QD and physical hookup is dependent on the min and max power of the RCS system. Somebody with more skills than me might be able to rough in a max based on 6bar exiting through an aperture of some size then fudging for the vents that exit into the dryer vent covers. Are there other vents for attitude control and non longitudinal translation?


What about minimums? All this impacts everything from matching orbits, maneuvering for physical hookup and ullage settling.


The thrust impact between a butterfly valve at 89deg and 90deg is insignificant. The difference between 1deg and 1.25deg would be huge. I don't doubt that sub degree control at a high level of precision is possible, but at what cost?


One solution might be to do coarse correction at 6bar, allowing the pressure to bleed down for the finer work. That in turn might require pressure be built back up to facilitate propellant transfer. How many kg of methane for each bar? Of course the O2 should be untouched by all this.


Grumble, grumble. Hot gas thrusters for the depot might end up being more props stingy with the benefit tighter control at a lower cost.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/05/2022 11:01 pm
Most of the RCS designs use a binary position valve and not a variable position valve. Thus the impulse control min is how fast can the valve cycle from from closed to open and back to closed. There are current valves used on spacecraft that can operate in pulse mode for pulses as fast as 160ms. Thus for a 100kg thruster max (continuous on) to min (single pulse). The min from such a thruster is 16kgf divided into the mass of the vehicle. The nicety is that the regulation of how much thrust is given is easily controlled. Which is why for monopropellants such valves are commonly used. Even though the min on the example case would be 16kgf. The ability of the fine variants between that and full on is easily much much smaller increments since it is simple a matter of adjusting the on time by a few ms. Which can give a granularity/accuracy of control of >100 to 1 without much difficulty.

In general your primarily haggling over already solved problems for large spacecraft. Note Shuttle was a 100t+ vehicle that had extremely fine RCS control and did not need most of the things being discussed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/06/2022 12:13 am
In general your primarily haggling over already solved problems for large spacecraft. Note Shuttle was a 100t+ vehicle that had extremely fine RCS control and did not need most of the things being discussed.
This is something I've harped on about for a while: moving a large heavy spacecraft into close proximity with another large heavy object and then gently touching the two together is a solved problem, not a new problem to solve.
The APAS ring also had a lot lower mass than the QD plate attached to Starship, and still tolerated reboost thrust (~0.2 milli-gee) with the orbiter joined to the station.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/06/2022 04:42 am
The downside of the booms is that any misalignment rubs the effectors up and down the sides of the Starship, which are covered in tiles.
Some articulation may be needed for alignment, but it only needs to be fairly gross alignment (get the narrow tip of the alignment pin into the wide opening of the socket) which could be accomplished with a lockable complaint mechanism (rather than powered actuation).

I think I had the geometry figured out so that the booms would only rub along the dorsal side, and then probably only within a meter or so of the hard point.  The idea was for the hard point to guide the effector/latch to it as the boom scraped along.  Whether that idea actually works... that's where the "half-baked" part came in.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/06/2022 05:07 am
In general your primarily haggling over already solved problems for large spacecraft. Note Shuttle was a 100t+ vehicle that had extremely fine RCS control and did not need most of the things being discussed.
This is something I've harped on about for a while: moving a large heavy spacecraft into close proximity with another large heavy object and then gently touching the two together is a solved problem, not a new problem to solve.
The APAS ring also had a lot lower mass than the QD plate attached to Starship, and still tolerated reboost thrust (~0.2 milli-gee) with the orbiter joined to the station.

There are some decent-sized caveats to that "solved problem" assertion:

1) Assuming you fill it full on the pad, any Starship is going to have a gross mass of >250t in LEO.  (The division between prop and payload may be different for different missions, but the gross mass will be the same.)  That's more than 2x the gross mass of an Orbiter.

2) You're dealing with some combination of cold gas, hot gas, methox, and/or methalox, not hypergolics.  Minimum impulse bit for anything with an igniter is going to be a lot bigger than the corresponding hypergolic system.  Similarly, minimum impulse bit for a cold gas thruster is going to be a lot lower.

3) There are a lot more roles for Starship's thrusters to play than there were on the Orbiter.  Orbiter used OMS for maneuvers with tens of m/s of delta-v, and it used RCS for attitude control, prox ops, and backup deorbit maneuvers.  Starship has to do all of those things plus ullage burns that go on for minutes to hours, and the LSS has landing thrusters (maybe).

4) You're dealing with a lot more slosh than the Orbiter had.

I'm sure that there's a solution somewhere in the trade space, but I'm less sure that the solution doesn't impact some aspects of docking or refueling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/06/2022 11:20 am
I have to agree that QD latching isn't like sticking a chuck into a service station air line. Most probably rotating and actively powered as you say. A collet, actively clamped onto a simple shaft by a concentric motor would have

It's probably worth reminding people again what the QD passive and active plates actually look like. (See below.)

And I suggest that if you aren't referring to the actual QD system (or a very, very slight mod), you use other terminology for your variant (rather than QD) to avoid talking at cross purposes when someone is specifically referring to the existing mechanism and you're talking about something new.

The active side on the arm (male) contains guide pins, in addition it has rolling latches that lock into lugs on the female plate, clamping the plates together and apparently handling the final, fine alignment.

Images start with a labelled version of the female (ship-side) plate.
Then the male plate.
Then a couple of varying quality images of the male plate closer up, to show the guide-pins as well as the rolling locking latches.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/06/2022 11:37 am
While we're at it, I think everyone should clarify what they mean when they say "QD arm".

The tower arm with the QD assembly is called the "QD arm" (as distinct from the two "chopsticks") but is much, much more than just the QD plate on its specific assembly sub-arm. There's two "claws" that clamp around the SH to stabilise it, there's a bumper system which extends past the SH and I assume is used to bump the aft of the SS against when aligning for stacking. Then finally, on top of the arm holding those features, there's the relatively small QD assembly.

If you say "QD arm" are you talking about just the QD assembly sub-arm, or an entire grabber arm thing in addition to the specific QD part? Because those are very, very different situations.

See images below. Some are from during assembly, not everything's in place. And I've included a bunch of angles because focal foreshortening hides black-on-black trusses against each other, making depth perception difficult.

[edit: Ignore the green circles on the second image, that was already there.]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/06/2022 12:11 pm
If you say "QD arm" are you talking about just the QD assembly sub-arm, or an entire grabber arm thing in addition to the specific QD part? Because those are very, very different situations.
QD arm = the arm the QD plate sits on. It can extend and retract, move up-down-left-right relative to the QD plate, and the plate itself can pitch and yaw on the end oft he arm. It may be able to roll too, but that has not been observed.
Clip of the arm itself in motion: https://twitter.com/CSI_Starbase/status/1530366511784898562

The larger swing-out arm with the claw is the stabiliser arm. The QD arm sits on top of this.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/06/2022 04:40 pm
I have to agree that QD latching isn't like sticking a chuck into a service station air line. Most probably rotating and actively powered as you say. A collet, actively clamped onto a simple shaft by a concentric motor would have

It's probably worth reminding people again what the QD passive and active plates actually look like. (See below.)

And I suggest that if you aren't referring to the actual QD system (or a very, very slight mod), you use other terminology for your variant (rather than QD) to avoid talking at cross purposes when someone is specifically referring to the existing mechanism and you're talking about something new.

The active side on the arm (male) contains guide pins, in addition it has rolling latches that lock into lugs on the female plate, clamping the plates together and apparently handling the final, fine alignment.

Images start with a labelled version of the female (ship-side) plate.
Then the male plate.
Then a couple of varying quality images of the male plate closer up, to show the guide-pins as well as the rolling locking latches.
Can't speak for anybody else but I've been using QD plate much like you do. The gender bender would have an 'active' QD plate on each face to connect to the passive QD plate on the depot on one side and the passive QD plate of the second ship on the other side.

In discussion of latching, I've been speaking of the structural QD latching hooks. Do the structural latches hold the fluid connections together by brute force or is there another mechanism?


Edit: added the underlined words.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/06/2022 05:33 pm
The question we need to focus on isn't about the quick disconnect, it's about the presence or absence of a quick connect

If edzieba is right that the QD can be latched reliably with very little force applied, then that dramatically simplifies the "docking" process.  I'm using scare quotes because, while we all agree that this isn't docking as we usually envision in it, it seems as if there are a wide variety of opinions on how rigid and accurate the docking needs to be.

At one extreme, there's no docking at all, except for the QD.  At the other, we need one or more structurally rigid connections capable of handling worst-case torques produced by the most heavily loaded depot-tanker, depot-payload, or tanker-tanker system.  In between, we've seen suggestions for articulated grapples, fairly dumb booms, with various mechanisms to make them sticky or latched, and plain ol' vanilla IDSS or APAS-like docks that don't happen to need pressurized tunnels.

The tower arm with the QD assembly is called the "QD arm" (as distinct from the two "chopsticks") but is much, much more than just the QD plate on its specific assembly sub-arm. There's two "claws" that clamp around the SH to stabilise it, there's a bumper system which extends past the SH and I assume is used to bump the aft of the SS against when aligning for stacking. Then finally, on top of the arm holding those features, there's the relatively small QD assembly.

If you say "QD arm" are you talking about just the QD assembly sub-arm, or an entire grabber arm thing in addition to the specific QD part? Because those are very, very different situations.

I've been assuming that it's fairly easy to take the cam and the bellows that are part of the GSE arm and adapt that so that the bellows is mostly looped in the remnants of the payload bay (with most of the cylinder consumed by moving the dome and bulkhead forward to contain more prop), with the cam able to extend the active plate through a door.  It wouldn't have to extend more than a meter or two.

But if you need the grabbers and bumpers and the entire stack stuck to the ground by gravity for the connection to be made reliably, then I doubt that you can make a grabber that extends out through a payload door, nor would it be sufficient to stabilize the pair of Starships even if it could.  Hence, the need for "docking".
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/07/2022 01:33 am
The question we need to focus on isn't about the quick disconnect, it's about the presence or absence of a quick connect

If edzieba is right that the QD can be latched reliably with very little force applied, then that dramatically simplifies the "docking" process.  I'm using scare quotes because, while we all agree that this isn't docking as we usually envision in it, it seems as if there are a wide variety of opinions on how rigid and accurate the docking needs to be.

At one extreme, there's no docking at all, except for the QD.  At the other, we need one or more structurally rigid connections capable of handling worst-case torques produced by the most heavily loaded depot-tanker, depot-payload, or tanker-tanker system.  In between, we've seen suggestions for articulated grapples, fairly dumb booms, with various mechanisms to make them sticky or latched, and plain ol' vanilla IDSS or APAS-like docks that don't happen to need pressurized tunnels.

The tower arm with the QD assembly is called the "QD arm" (as distinct from the two "chopsticks") but is much, much more than just the QD plate on its specific assembly sub-arm. There's two "claws" that clamp around the SH to stabilise it, there's a bumper system which extends past the SH and I assume is used to bump the aft of the SS against when aligning for stacking. Then finally, on top of the arm holding those features, there's the relatively small QD assembly.

If you say "QD arm" are you talking about just the QD assembly sub-arm, or an entire grabber arm thing in addition to the specific QD part? Because those are very, very different situations.

I've been assuming that it's fairly easy to take the cam and the bellows that are part of the GSE arm and adapt that so that the bellows is mostly looped in the remnants of the payload bay (with most of the cylinder consumed by moving the dome and bulkhead forward to contain more prop), with the cam able to extend the active plate through a door.  It wouldn't have to extend more than a meter or two.

But if you need the grabbers and bumpers and the entire stack stuck to the ground by gravity for the connection to be made reliably, then I doubt that you can make a grabber that extends out through a payload door, nor would it be sufficient to stabilize the pair of Starships even if it could.  Hence, the need for "docking".
A decent summation. What are the chances that SX engineers have worn the same groves? I'd guess high. The only thing missing is data.


Another guess is that fiddling with StarLink will be high priority on the next flight followed closely by some maneuvering that would explore refueling ops.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: colbourne on 07/07/2022 11:44 am
I think we will end up transferring full fuel tanks (possibly with engines attached) rather than pumping methane and oxygen across. This might require a redesign of the Mars transit craft but should solve many problems.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: chopsticks on 07/07/2022 12:23 pm
I think we will end up transferring full fuel tanks (possibly with engines attached) rather than pumping methane and oxygen across. This might require a redesign of the Mars transit craft but should solve many problems.
Full fuel tanks with engines attached is basically what a rocket is. There's no "separate" tank that can be transfered - the fuel tank is the rocket body.

I don't see the prop transfer as being super hard thing to figure out, there are lots of good ideas of ways it could be done further up in this thread.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/07/2022 04:40 pm
Another guess is that fiddling with StarLink will be high priority on the next flight followed closely by some maneuvering that would explore refueling ops.

It should be noted that, if they really wind up with a dorsal-to-dorsal, front-to-back refueling configuration, getting this to play nice with the Starlink Pez dispenser isn't a slam-dunk.  This might not be essential; after all, Starlink missions don't require refueling.  But if you want to use Starships that have finished deploying their Starlinks  as test targets (or even test tankers) for refueling, then having them compatible might be nice.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/08/2022 03:32 pm
Another guess is that fiddling with StarLink will be high priority on the next flight followed closely by some maneuvering that would explore refueling ops.

It should be noted that, if they really wind up with a dorsal-to-dorsal, front-to-back refueling configuration, getting this to play nice with the Starlink Pez dispenser isn't a slam-dunk.  This might not be essential; after all, Starlink missions don't require refueling.  But if you want to use Starships that have finished deploying their Starlinks  as test targets (or even test tankers) for refueling, then having them compatible might be nice.
I was thinking just the solo (sub)orbital. Work the dispenser and then jinx around to get a feel for the handling. Two ships later on.


Two ships could be a short term problem. The first ship deploys sats then has to loiter until the second ship launches, however long that may be. It needs power and reaction mass. The power load would be computers, comms, RCS control and maybe some heaters. I've no idea of what the total load would be. The longer the loiter, the more batteries needed. PV is an option. At some length of loiter PV may be the only option.


This early in the game, depending on a booster turnaround looks high risk. With only one launch mount ISTM the best shot at a fast turnaround would be two full stacks, both having gone through a full pre launch test suite. Ship A would go back to the production facility after ground tests. Ship B would then test, then launch.


A big unknown is turnaround time for the table and tower. Another bottleneck could be propellant tankers. It's going to take a LOT of semi's. The upshot of this is, we (and SX) have no idea how much loiter time a ship will need while waiting for a second ship to show up.


If somebody has booster propellant capacity I'll work up the number of tankers needed to load a stack.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/08/2022 04:23 pm
Another guess is that fiddling with StarLink will be high priority on the next flight followed closely by some maneuvering that would explore refueling ops.

It should be noted that, if they really wind up with a dorsal-to-dorsal, front-to-back refueling configuration, getting this to play nice with the Starlink Pez dispenser isn't a slam-dunk.  This might not be essential; after all, Starlink missions don't require refueling.  But if you want to use Starships that have finished deploying their Starlinks  as test targets (or even test tankers) for refueling, then having them compatible might be nice.
I'm really not a fan of the catawompus alignment but if it works out to be best, ISTM it can be made to work.


First off, most of the cargo bay on a depot would be filled by O2 tankage and as you (or somebody) pointed out earlier, this would force the QD up into the ogive. Same, same for the Pez dispenser.


Instead of stowing the QD internally, build a dormer with the QD on its face instead of stowing it internally and deploying through a door. If it needs to extend this minimizes the 'tail' it has to drag around.


How all this would interact with aero launch forces could be a problem. If the QD plate gets a bit toasty during maxQ, extend the dormer overhang. Assuming the dormer can be made stable enough to make it through maxQ the problem would be drag. Cut he Pez dispenser down to more closely represent the depot top dome location and sacrifice a few starlinks, then move the QD dormer down a matching amount. This would both drop mass to allow powering through maxQ and reduce the dormers parasitic drag.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/08/2022 05:08 pm
I'm in the camp that rather than extensive modification (enlarging tanks, adding external housings, wrapping in MLI, etc), if you need to store more propellant than a single regular-sized Starship can hold you fly two regular-sized Starships with prop transfer arms in the payload bay, rather than one. At least until flight rates are so obscenely high that you can justify a dedicated variant that will only ever be a depot (i.e. your depot is never idle), you may as well keep the modifications as minimal as possible so that for all the months of the year you do not ned any depots, those ships can return to Earth, have the arm removed, and be used for regular payload launches (paying off their own manufacture) rather than floating around earning nothing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 07/08/2022 05:31 pm
I'm in the camp that rather than extensive modification (enlarging tanks, adding external housings, wrapping in MLI, etc), if you need to store more propellant than a single regular-sized Starship can hold you fly two regular-sized Starships with prop transfer arms in the payload bay, rather than one. At least until flight rates are so obscenely high that you can justify a dedicated variant that will only ever be a depot (i.e. your depot is never idle), you may as well keep the modifications as minimal as possible so that for all the months of the year you do not ned any depots, those ships can return to Earth, have the arm removed, and be used for regular payload launches (paying off their own manufacture) rather than floating around earning nothing.
I have a completely different perspective. Build a depot with all of the transfer hardware and no EDL, launch it into space, and leave it there. Much cheaper than building transfer hardware that can be installed and removed and building all ships to accommodate it. Depot becomes the only variant that has any specialized equipment.  Except for the transfer hardware, Depot is the cheapest possible Starship variant, so the idling cost is minor, and you must build the transfer hardware in some form in any event. Depot can also be taller than a normal starship to accomodate more fuel.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/08/2022 06:27 pm
Much cheaper than building transfer hardware that can be installed and removed and building all ships to accommodate it.
No need to build all ships to accommodate it, just the ones that will pull double-duty as depots*. If you're going to mod some ships anyway, may as well do it as a mod that can keep launching payloads rather than a mod that can't, just in pursuit of an optimum solution rather than a good-enough one. Remember that not all propellant transfers need to be a volume greater than a standard Starship, only Starship HLS has been proposed as needing a larger propellant capacity so only it would ever need an oversized depot. The rest of the time, a regular-Starship-sized depot can refill a regular-sized Starship.
Plus, a jumbo-depot that cannot return is effectively expended (can only ever be re-used if you are going to the same plane again or waste launches and propellants refilling it to perform a plane-change), whereas a returnable depot can be re-used multiple times in the otherwise idle period, either as a depot for alternate planes or for direct payload launches.
And on top of that: for any mission that requires only a small propellant top-up rather than a full re-fill, you can launch a normal-sized returnable depot instead of a single tanker flight, saving you a launch and removing any coupling hardware from the tanker flights (because a tanker would never need to couple directly to any Starship without the depot hardware).

The fewer SKUs you have, and the fewer changes between SKUs, the cheaper overall production can be.
* Unless it works out cheaper to build all ships with tank apex taps and just cap them off when not needed, in order to avoid having two different types of forward dome instead of one.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/08/2022 06:33 pm
First off, most of the cargo bay on a depot would be filled by O2 tankage and as you (or somebody) pointed out earlier, this would force the QD up into the ogive. Same, same for the Pez dispenser.

As long as the depot has enough prop to fill the LSS for its worst-case mission (which is currently LEO-NRHO-LS-NRHO), then it doesn't need any more stretch than that.  For BLT transit, that's about 1535t of prop.  If the LSS launches with a 20t crew module and 10t of cargo, then it arrives in LEO with about 120t of prop, so the depot doesn't need to be bigger than about 1420t, which leaves it with about 3.8m of cylinder.  Goose it up to 1500t for boil-off and ullage limits and you still have more than 2m of cylinder.  That's enough for doors, and you can coil the bellows up into the ogive with no problem.

Quote
Instead of stowing the QD internally, build a dormer with the QD on its face instead of stowing it internally and deploying through a door. If it needs to extend this minimizes the 'tail' it has to drag around.

How all this would interact with aero launch forces could be a problem. If the QD plate gets a bit toasty during maxQ, extend the dormer overhang. Assuming the dormer can be made stable enough to make it through maxQ the problem would be drag. Cut he Pez dispenser down to more closely represent the depot top dome location and sacrifice a few starlinks, then move the QD dormer down a matching amount. This would both drop mass to allow powering through maxQ and reduce the dormers parasitic drag.

You're swatting a fly with a nuclear weapon, and you don't need to.  The bellows loops will fit fine (maybe upside-down, looped up int the ogive), as will everything else you'd need for a tanker or depot.  It's really just another payload, but with some plumbing and an appropriate door/hatch.

But this does require the nose-to-tail configuration to work.  Sure sounds a lot easier than fiddling with the aerodynamics.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/08/2022 06:54 pm
I'm in the camp that rather than extensive modification (enlarging tanks, adding external housings, wrapping in MLI, etc), if you need to store more propellant than a single regular-sized Starship can hold you fly two regular-sized Starships with prop transfer arms in the payload bay, rather than one. At least until flight rates are so obscenely high that you can justify a dedicated variant that will only ever be a depot (i.e. your depot is never idle), you may as well keep the modifications as minimal as possible so that for all the months of the year you do not ned any depots, those ships can return to Earth, have the arm removed, and be used for regular payload launches (paying off their own manufacture) rather than floating around earning nothing.

I think you're going to wind up with a stretched version no matter what.

A lot of this depends on what you need for the LSS.  I've been using Isp=378s, 95t dry mass, 20t crew module, 2t of cargo, and about 1t of returned samples (aka downmass--as long as you're standing on your head).  For that, LEO-BLT-NRHO-LS-NRHO takes 8965m/s of delta-v, and you can indeed make that work for a 1200t LSS.

But that's with 0.07% FPR, which isn't anywhere near what NASA will require for a crewed mission--especially one with some amount of boil-off.  If you figure about 1.7% FPR, you're up to 1430t of prop.  I suspect that 1500t is probably closer to the design center, and that'll give you either bigger payloads with the plan of record or the ability to do LEO-LLO-LS-NRHO, which is key to getting rid of SLS/Orion from the conops.

1500t of tankage leaves you with 2.3m of the cylindrical payload bay left, which is more than enough for HLS purposes.  As long as your crew module fits into the ogive, you're all set.

But if you're doing the tank stretch for the LSS, then you might as well use the same stretch for the tankers and depots.  It's different lengths for the downcomers and wiring harnesses, which is kind of pain, but if you launch with only 1200t of prop, the dynamics should be close to identical.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/08/2022 07:02 pm
Sorry if I missed this somewhere, but, given that an ullage burn is unavoidable, what sort of engines are going to do it? Since you only want a milli-g or less, surely the Raptors can't throttle down that far. And if you're docking the ships nose-to-tail, where is that ullage engine going to be? It seems as though it has to be centered or else it'll impose a torque on the two ships, which makes it seem (to me) as though it has to be a whole separate module that the two starships essentially dock to.

Again, apologies if this has already been thoroughly discussed. I just now started wondering about it because of all the nose-to-tail discussions.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Zed_Noir on 07/08/2022 09:54 pm
Sorry if I missed this somewhere, but, given that an ullage burn is unavoidable, what sort of engines are going to do it? Since you only want a milli-g or less, surely the Raptors can't throttle down that far. And if you're docking the ships nose-to-tail, where is that ullage engine going to be? It seems as though it has to be centered or else it'll impose a torque on the two ships, which makes it seem (to me) as though it has to be a whole separate module that the two starships essentially dock to.

Again, apologies if this has already been thoroughly discussed. I just now started wondering about it because of all the nose-to-tail discussions.
AIUI for ullage, the Starship will just vent off some of the gases from propellants boil off as one option.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/09/2022 12:26 am
AIUI for ullage, the Starship will just vent off some of the gases from propellants boil off as one option.
Okay, but how would that work for two ships that are docked? It can't go out the engines or else they'd spin like a pinwheel.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/09/2022 03:05 am
AIUI for ullage, the Starship will just vent off some of the gases from propellants boil off as one option.
Okay, but how would that work for two ships that are docked? It can't go out the engines or else they'd spin like a pinwheel.

There will be cold gas thrusters, pointing in at least two directions.  You're right that each ship will need its own thrusters.  Note that the thrusts will vary as mass is pumped from one vehicle to another and the center of mass changes.

I'm skeptical that ullage and attitude control thrusters will be the same.  Minimum wet mass to orbit for a Starship will be roughly 270t (assuming the tanks are filled at launch, no matter the payload size).  Maximum wet mass for a full depot could be as much as 1600t.  Figure 1mm/s² acceleration for ullage, and that gives you between 270N and 1600N of ullage thrust.

Now let's look at attitude control.  Doing a really simpleminded estimate of moments of inertia, x-axis would be 5500 to 32,400t-m², and the y- and z-axes would be 169,000 to 1,000,000t-m².  What do you need for decent command authority for attitude control?  Maybe 1º/s² (0.017rad/s²)?  That would require an x-axis thrust of 21 to 126kN, and y- and z-axis thrusts of 118 to 698kN.  Those are big thrusters, probably big enough to be used for the LSS landing thrusters, if they happen to point in the right direction.

This is only half the story, though.  You also need to figure out total impulse for ullage, attitude control, and landing, and how gas or liquid is sequestered and pressure-fed for the various applications.  That will determine whether methox thrusters need to be gas- or liquid-fed, and how big the COPVs for the pressure-fed system need to be.

This is not straightforward.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/09/2022 10:47 am
Sorry if I missed this somewhere, but, given that an ullage burn is unavoidable, what sort of engines are going to do it?

They tested hot-gas methalox thrusters back in 2021. They were intended to replace all RCS thrusters on the Superheavy for the first orbital flight, due to the insane improvement in Isp over cold-gas RCS (from 60s to >300s), but were removed to avoid being a complicating factor. I would assume they'll be kept for ullage burns during fuel transfer, given that the hardware exists and the Isp advantage is ridiculous.

(Presumably a larger version would be used for the cosine landing engines for HLS.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: volker2020 on 07/09/2022 12:26 pm
Currently I guess under the following assumptions:
a) HLS Contract does refer and show a larger depot Starship.
b) Instead of adding additional weight to any standard Starship (be it Tanker, Mars Lander or Lunar Lander) it is probably easier to only have depot Starships left with special tanking hardware.
c) If there is the need for precision control, regarding the docking, it should be enough to have those on the depot variant.
d) I assume that only the tanker has a extra docking port, able to dock to the quick connect of the standard Starship.
e) I assume that there will be more than one Depot Starship on different planes, which will be filled in parallel to avoid plane changes while tanking.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Joffan on 07/09/2022 03:05 pm
Adapting "the best part is no part" to "the best tanker variant is no tanker variant", for the odd case when a a big fill of propellant is required, it seems to me that you could simply use 2 minimally-modified Starships in place of one enlarged tanker. Sure, that would mean 2 filling processes for the mission Starship, but those processes are going to have to be pretty routine and automated, and while space  operations has many challenges, fortunately orbital mechanics is very predictable. Two depot-Starships orbiting together would be straightforward for a company that auto-docks to the ISS (with all of the added restrictions of approaching  and operating near that outpost).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/09/2022 03:34 pm
e) I assume that there will be more than one Depot Starship on different planes, which will be filled in parallel to avoid plane changes while tanking.
There's at least some argument that one depot will suffice. Have a look at this, if you haven't seen it yet: AAS Paper Review: Practical Methodologies For Low Delta-V Penalty, On-Time Departures To Arbitrary Interplanetary Destinations From A Medium-Inclination Low-Earth Orbit Depot (https://selenianboondocks.com/2018/02/aas-paper-review-practical-methodologies-for-low-delta-v-penalty-on-time-departures-to-arbitrary-interplanetary-destinations-from-a-medium-inclination-low-earth-orbit-depot/).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 07/09/2022 04:22 pm
e) I assume that there will be more than one Depot Starship on different planes, which will be filled in parallel to avoid plane changes while tanking.
There's at least some argument that one depot will suffice. Have a look at this, if you haven't seen it yet: AAS Paper Review: Practical Methodologies For Low Delta-V Penalty, On-Time Departures To Arbitrary Interplanetary Destinations From A Medium-Inclination Low-Earth Orbit Depot (https://selenianboondocks.com/2018/02/aas-paper-review-practical-methodologies-for-low-delta-v-penalty-on-time-departures-to-arbitrary-interplanetary-destinations-from-a-medium-inclination-low-earth-orbit-depot/).
Apparently, if you have a smallish number of refuelling operations, you can use a single depot without a big penalty. However, as the cadence increases to more than one tanker a day from a single launch tower, the tanker launches will predominate and you will need multiple depots so the tanker flights don't need to waste fuel on doglegs.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/09/2022 09:59 pm
Sorry if I missed this somewhere, but, given that an ullage burn is unavoidable, what sort of engines are going to do it?

They tested hot-gas methalox thrusters back in 2021. They were intended to replace all RCS thrusters on the Superheavy for the first orbital flight, due to the insane improvement in Isp over cold-gas RCS (from 60s to >300s), but were removed to avoid being a complicating factor. I would assume they'll be kept for ullage burns during fuel transfer, given that the hardware exists and the Isp advantage is ridiculous.

You're talking about an incredibly wide throttling range if you use one thruster for ullage, attitude control, and lunar landing.  It'd be a nice trick, but based on my arm-wave just up thread, you'd have to be able to throttle down to fractions of a percent to use them for settling.

Also, a clarification of terms:  When you say "hot gas methalox", you're talking about a pressure-fed, liquid-injected engine that burns LCH4 and LOX, right?  So "hot gas methox" would be a pressure-fed, gas-injected engine that burns CH4 and O2?  and "cold gas" means a pressure-driven engine that ejects unburnt CH4 (since its molecular weight is lower than O2), and it's presumably fed from some COPV with the proper pressure and temperature characteristics to keep it in a supercritical state?

Using those definitions for terms, note that non-hypergolic hot gas thrusters require an igniter, which makes the minimum impulse bit a lot bigger than a hypergolic thruster, doesn't it?

It's the min impulse bit, along with the wildly different moment of inertia around the x-axis, compared to the y- and z-axes, that makes this whole thing kinda weird.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/09/2022 11:05 pm
I'm in the camp that rather than extensive modification (enlarging tanks, adding external housings, wrapping in MLI, etc), if you need to store more propellant than a single regular-sized Starship can hold you fly two regular-sized Starships with prop transfer arms in the payload bay, rather than one. At least until flight rates are so obscenely high that you can justify a dedicated variant that will only ever be a depot (i.e. your depot is never idle), you may as well keep the modifications as minimal as possible so that for all the months of the year you do not ned any depots, those ships can return to Earth, have the arm removed, and be used for regular payload launches (paying off their own manufacture) rather than floating around earning nothing.
I'm obviously in the 'Depot Heavy' camp but agree that it will start as you suggest. In the end, we just have to wait and see.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/09/2022 11:50 pm
I'm in the camp that rather than extensive modification (enlarging tanks, adding external housings, wrapping in MLI, etc), if you need to store more propellant than a single regular-sized Starship can hold you fly two regular-sized Starships with prop transfer arms in the payload bay, rather than one. At least until flight rates are so obscenely high that you can justify a dedicated variant that will only ever be a depot (i.e. your depot is never idle), you may as well keep the modifications as minimal as possible so that for all the months of the year you do not ned any depots, those ships can return to Earth, have the arm removed, and be used for regular payload launches (paying off their own manufacture) rather than floating around earning nothing.
I have a completely different perspective. Build a depot with all of the transfer hardware and no EDL, launch it into space, and leave it there. Much cheaper than building transfer hardware that can be installed and removed and building all ships to accommodate it. Depot becomes the only variant that has any specialized equipment.  Except for the transfer hardware, Depot is the cheapest possible Starship variant, so the idling cost is minor, and you must build the transfer hardware in some form in any event. Depot can also be taller than a normal starship to accomodate more fuel.
Eventually, this is what I expect but in difference, I expect a fully tricked out depot to be the most expensive short of crewed. Some combination of that solar paint, a sunshield, and/or cryo cooler. A hefty battery pack. PV. Add in any transfer specific plumbing, fittings and maybe a tank stretch, then possibly some deployable booms or arms. Maybe specialized thrusters. Subtract the heat tiles and fins.

Even an expensive idle depot can be cost effective if analyzed at an operational level as it adds a capability that would otherwise be impossible. Even though it's an expensive build, as space hardware goes, it's cheap. If it were in constant use it would be great but realistically, an annual or semiannual Artemus mission, the occasional solar system probe or high mass GEO package, and eventually, Mars. This ignores one depots specific orbit maybe not being ideal for all these missions.

The commercial market for trans LEO hasn't quite figured itself out yet but as it does, we'll see more depots tucked away on the more popular lanes. That's the reason I figure Depot Heavy, designed as a long lived facility, is the logical growth path. It's as much a part of the infrastructure as a launch table/tower.

Edit: added green text.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/10/2022 12:21 am
First off, most of the cargo bay on a depot would be filled by O2 tankage and as you (or somebody) pointed out earlier, this would force the QD up into the ogive. Same, same for the Pez dispenser.

As long as the depot has enough prop to fill the LSS for its worst-case mission (which is currently LEO-NRHO-LS-NRHO), then it doesn't need any more stretch than that.  For BLT transit, that's about 1535t of prop.  If the LSS launches with a 20t crew module and 10t of cargo, then it arrives in LEO with about 120t of prop, so the depot doesn't need to be bigger than about 1420t, which leaves it with about 3.8m of cylinder.  Goose it up to 1500t for boil-off and ullage limits and you still have more than 2m of cylinder.  That's enough for doors, and you can coil the bellows up into the ogive with no problem.

Quote
Instead of stowing the QD internally, build a dormer with the QD on its face instead of stowing it internally and deploying through a door. If it needs to extend this minimizes the 'tail' it has to drag around.

How all this would interact with aero launch forces could be a problem. If the QD plate gets a bit toasty during maxQ, extend the dormer overhang. Assuming the dormer can be made stable enough to make it through maxQ the problem would be drag. Cut he Pez dispenser down to more closely represent the depot top dome location and sacrifice a few starlinks, then move the QD dormer down a matching amount. This would both drop mass to allow powering through maxQ and reduce the dormers parasitic drag.

You're swatting a fly with a nuclear weapon, and you don't need to.  The bellows loops will fit fine (maybe upside-down, looped up int the ogive), as will everything else you'd need for a tanker or depot.  It's really just another payload, but with some plumbing and an appropriate door/hatch.

But this does require the nose-to-tail configuration to work.  Sure sounds a lot easier than fiddling with the aerodynamics.
If this is a Depot Heavy of some flavor and there's room for the QD below the ogive, why bother with the inside QD and door? Mount flush like the GSE QD. If there's aero problems that far up, an aero shield, open faced if possible. Add in whatever mechanism needed, if any, for extending the plate.


I figure this in the field artillery ranking of efforts.  :D



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 07/10/2022 12:31 am
To my mind one thing counting against long term orbiting depots is longevity.

The more time in space, the higher the likelihood of impact damage from orbiting debris or micrometeoroids.
Any additional coatings and or shielding adds to weight, complexity and cost.
A single skin with reflective paint isn't going to do much to prevent punctures...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/10/2022 02:09 am
Sorry if I missed this somewhere, but, given that an ullage burn is unavoidable, what sort of engines are going to do it? Since you only want a milli-g or less, surely the Raptors can't throttle down that far. And if you're docking the ships nose-to-tail, where is that ullage engine going to be? It seems as though it has to be centered or else it'll impose a torque on the two ships, which makes it seem (to me) as though it has to be a whole separate module that the two starships essentially dock to.

Again, apologies if this has already been thoroughly discussed. I just now started wondering about it because of all the nose-to-tail discussions.
You hit on one of the big disadvantages of the nose to tail orientation. They can do things to get around asymmetrical thrust but I don't see the point. It's more complexity for a technique whose only virtue seems to be avoiding a gender bender.


The second big disadvantage is mass. All the propellant plumbing is already designed to feed from the bottom. Why duplicate it to also feed from the top? The design does away with the QD gender bender at the cost of a second QD plate. Admittedly the gender bender would mass more than the second QD plate (it's essentially two active QD plates) but in exchange can be designed to be a spacer for stand-off between ships. There's also a chance that the gender bender can deploy via a very simple mechanism.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Dancing Dog on 07/10/2022 02:26 am
To my mind one thing counting against long term orbiting depots is longevity.

The more time in space, the higher the likelihood of impact damage from orbiting debris or micrometeoroids.
Any additional coatings and or shielding adds to weight, complexity and cost.
A single skin with reflective paint isn't going to do much to prevent punctures...

If the first test tanker gets penetrated too quickly, they could always send the second one up empty, with 300+t of armor instead of the first load of props. Mass is not a limiting factor anymore.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/10/2022 02:43 am
To my mind one thing counting against long term orbiting depots is longevity.

The more time in space, the higher the likelihood of impact damage from orbiting debris or micrometeoroids.
Any additional coatings and or shielding adds to weight, complexity and cost.
A single skin with reflective paint isn't going to do much to prevent punctures...
This is a good point. Here's a paper. It does a nice survey of impacts but without a size/damage distribution.


https://ntrs.nasa.gov/api/citations/20190033989/downloads/20190033989.pdf (https://ntrs.nasa.gov/api/citations/20190033989/downloads/20190033989.pdf)





Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/10/2022 04:44 am
To my mind one thing counting against long term orbiting depots is longevity.

The more time in space, the higher the likelihood of impact damage from orbiting debris or micrometeoroids.
Any additional coatings and or shielding adds to weight, complexity and cost.
A single skin with reflective paint isn't going to do much to prevent punctures...

They're likely to be in VLEO, which is presumably a cleaner environment than higher up in LEO.  It also improves overall fuel efficiency for all missions.  However, there's another reason to stay low: there are punctures that cause a leak, and then there are punctures that cause catastrophic venting or even explosions.  You not only want to minimize your chance of being a debris receiver, you want to minimize the chance of being a debris generator.  Not only is that good stewardship of the orbital environment, it's also nakedly self-interested, because if you blow up one depot and generate a bunch of junk, the next depot, which you'd probably like to put in a similar orbit, has to dodge the mess you made, or go to a less optimal altitude.

I haven't done the arithmetic on how much delta-v you have spend per unit time to maintain a 360ish km orbit for an almost-empty depot with a bunch of solar panels and radiators, but I'll bet it's not trivial.  You should be able to engineer around the possibility that your orbit maintenance prop pushes you over the edge from n to n+1 lift tankers, but why force people to think about it at all?

So there are another couple of reasons why landing your depot when you're not using it makes sense.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/10/2022 04:28 pm
Sorry if I missed this somewhere, but, given that an ullage burn is unavoidable, what sort of engines are going to do it?
They tested hot-gas methalox thrusters back in 2021.
Also, a clarification of terms:  When you say "hot gas methalox", you're talking about a pressure-fed, liquid-injected engine that burns LCH4 and LOX, right?  So "hot gas methox" would be a pressure-fed, gas-injected engine that burns CH4 and O2?

I wasn't differentiating between the two. "It uses ship's native fuel-ox and it burns." Don't know details, they stopped talking about it when they pulled it off of SH for the orbital test.

note that non-hypergolic hot gas thrusters require an igniter, which makes the minimum impulse bit a lot bigger than a hypergolic thruster, doesn't it?

They were talking about it as a replacement for RCS for the landing manoeuvres during Superheavy's chop-stick capture, so I assume they think that fast-start + fine-control is a solved or solvable problem. OTOH, it then went away, so maybe not. But neither ullage nor lunar landing cosines require instant-on. Ullage probably needs to be steerable, to deal with the shifting CoM during prop xfer, but that's someone-elses-problem. And you'd be stupid to waste tens of tonnes of propellant for cold-gas ullage when you know how to build little hot-gas rockets and have plenty of room to stick them in the tail.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: txgho on 07/10/2022 06:51 pm
Fuel Depot I would expect would launch with a stretched body and extra sheath of thin skin painted white.
No fins and no SR. Possibly one or three small engines or miniature raptors for station keeping and the 3 VR.
Additionally arrays of solar panels deployed outside of the approach a tanker or mission ship would connect.
As a depot it would have a mini refinery installed for continual cooling and separation activities.

Altitude for tanker/mission transfers would be smartest at lowest utility altitude for mitigation of any incident or accident.

Initial fueling of a depot on the tower would likely be easiest with a gender reversed adapter to send the depot to orbit with the single interface for connecting to tankers and mission ships.

When idle orbit can be moved outside of LEO drag altitudes.
Every extra piece and component not required for a depot that will never return can be deleted.

Lot of real estate available on the outside of the starship. A full depot would be capable of tanking a LSS and have the fuel to change orbit for loitering.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/11/2022 06:31 am
They were talking about it as a replacement for RCS for the landing manoeuvres during Superheavy's chop-stick capture, so I assume they think that fast-start + fine-control is a solved or solvable problem. OTOH, it then went away, so maybe not. But neither ullage nor lunar landing cosines require instant-on. Ullage probably needs to be steerable, to deal with the shifting CoM during prop xfer, but that's someone-elses-problem. And you'd be stupid to waste tens of tonnes of propellant for cold-gas ullage when you know how to build little hot-gas rockets and have plenty of room to stick them in the tail.

Let's look at the various requirements:

1) Ullage thrust (very low thrust, moderate total impulse, long "burn" time).

2) Fine in-space attitude control (very low thrust, short impulse bit, fairly low total impulse).

3) Coarse in-space attitude control (medium thrust around x, higher thrust around y and z, total impulse long enough to handle an operational phase without having to autogenously replenish your pressure vessels until there's a natural ops pause).

4) Low delta-v maneuvers (10-100m/s delta-v, moderate thrust)

5) Earth landing attitude control (high thrust, fairly coarse min impulse bit, enough total impulse to get to the ground)

6) Lunar landing thrusters (high thrust, moderate total impulse, pretty wide throttling range)

You can handle #3-#6 with a throttleable hot gas thruster with a thrust of about 20kN to 200kN, but #1 and #2 just don't fit in that range.  They need 0.2-2kN.  That's why I think you need two distinct sets of thrusters.  If you can get the hot gas impulse bit low enough, then hot gas is great.  Otherwise, you'll probably need cold gas for the little guys.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 10:44 am
I’ve been pondering Rendezvous Proximity Operations and Docking (RPOD) at an in-space propellant accumulator (depot). In view of the excellent conversations here,  I better post it now.

Operational approaches to propellant refilling in space include navigating into position, grappling between accumulator and ship, and connecting plumbing systems between the two. Some objectives for an RPOD system could be:

(1) An in-space accumulator SpaceX term)would be based on a standard Starship as much as feasible, with as few modifications as possible.

(2) Standard plumbing interfaces should be employed, based on the quick disconnect (QD) interface system employed at Starbase, Texas.

(3) A desirable system would be robust and reliable, even when the operation  was not perfectly nominal.

(4) A simple system is preferred over a more complex approach.

(5) Mechanisms for inter-ship transfer of propellants should reside primarily in the one accumulator rather than in the many visiting Starships. “Strap-on” mechanisms on the accumulator would reinforce this approach.

In the next few posts I will provide sketches to illustrate one way to addresses these objectives.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 10:48 am
This approach employs a straight-forward  docking system and a simple plumbing connection with no moving parts. The in-space depot is a standard SS tanker converted for use as an in-space propellant accumulator. It features three docking posts that serve to lock a visiting vehicle in a rigid grip and to automatically connect plumbing between accumulator and ship at their respective propellant connectors plates (PCPs) built into their dorsal (non-tiled) sides. I am assuming ships would be oriented in the same direction when docked to each other so that catchers on the visiting ship would line up with docking posts on the accumulator. In this position, one of the ships would use ullage pressure, possibly supplemented by a motor in the liquid line, to overcome the small force from thrusters that settle propellants in the tanks.

The sketch below shows three posts mounted on the dorsal side of SS accumulator, labelled A, B, and C. Imaginary lines joining the tops of these posts would form a plane ABC congruent with the plumbing connector (QD) plate shown in the middle of the ship. If a visiting vehicle docks exactly on the tops these posts, its dorsal plumbing would be exactly connected to the complementary connectors of the accumulator, all in the same ABC plane.

UPDATE 7-12-22: Incorporates observations by Paul451. New sketch relocates Propellant Connectors Plate (PCP) previously call QD plate. New underlying rendering was produced by Afael Adamy in 2.26.2020 and published in NASApceflight.com.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 10:51 am
The next sketch shows the three posts A, B, and C with posts B and C protruding from the accumulator surface and post A on the surface . Posts are not perpendicular to the surface, but are perpendicular to the ABC plane, which is the common plane for all  four connectors (including QD plates PCPs)  for two vehicles.

In this approach, the tops of the three posts would feature the male ends of a non-androgynous mating system (not shown). Each visiting ship would carry three corresponding female devices installed on its surface to accurately align with male connectors on the posts.

If a visiting ship is capable of employing its thrusters to align itself accurately with the three posts and to dock upon them, the QD plates will also align accurately and mate automatically.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 10:52 am
To secure each post in its correct position, struts and internal stiffeners would be mounted in two directions. As shown in the sketch below, one strut would occupy a position in a line parallel to the X axis of the ship (as seen from above the post). It would attach near the top of the post and angle “downward” to the accumulator surface. The second strut would attach near the top of the post and angle down across the curved surface of the SS tank section, perpendicular to the X-axis. The sketch below shows C-post mounted to the side of an SS accumulator with these two struts firmly attached.

In addition to struts above the surface, additional stiffeners would be placed below the surface, inside the propellant tank. One stiffener would attach directly underneath the strut attachment point 1 and run to attachment point 2 (base of post); The second stiffener would run from point 2 to  surface attachment point 3 (the  X-direction); and a third stiffener would run  from point 3 back to point 1 to complete a rigid triangulation of attachment points. I assume SpX would install these stiffeners inside the tank on the ground before launching the accumulator ship and that they would not interfere with any tank operations..

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 10:54 am
This type of RPOD might be an simple and reliable system for operating an in-space accumulator. OTOH its drawback could be the original expense of setting up posts on the surface of a ship in space. In the worst case, this would involve sending astronauts in space suits to carry out this task. A better option might be to design a pop-up  apparatus to install  each post into position automatically.

In this approach, an accumulator would be sent into orbit with posts B and C attached to its surface. A male connector device would be installed at the top of each post. The base of each post would be welded to the surface and each post would be attached to its base with a hinge rod. As shown in the sketch below, It would initially lie on the surface parallel to the X-axis with a hinged strut already attached near the top of  post In the sketch below, the first strut for post C would be attached to the accumulator surface at point 3 and would lie folded up next to (or inside) the post during launch. The same configuration would apply to post B (not shown).

Once in proper orbit, unfolding would commence, as suggested in the sketch.. A highly geared-down motor in the base of the post would begin slowly turning the hinge rod until the post was in upright position. Rotary motion of the post would also cause one strut to unfold. When the post reached upright position, the strut hinge would lock in place (permanently).

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 10:56 am
The second strut for post C would unfold in a different manner because it is located above the curvature of the ship . The sketch below suggests that a metal strip with a groove inside could be laid across  the curved surface. To initiate unfolding, the strut would be released from its folded position. From here,  a motorized wheel attached to the bottom of the strut would drive it from the base of the post to point 1, using the groove as a navigation guide, to its final position, where it it would lock in place (permanently).

To operate correctly, the wheel must enter the wheel guide without issue. Thus, when the  post and first strut initially unfold, the wheel must be designed to automatically rotate up to align with the groove/wheel-guide.

Note that the wheel is initially located in a crevice between post C and one side of the accumulator surface. To initiate its travel, the wheel in the groove must ascend almost vertically out of the crevice, which could potentially cause the strut, in its initially vertical position, to jam and halt its unfolding. To ensure that the strut unfolds outward, a spring located on the post would exert a small force on the strut just above its knee joint. In addition, a wire of a specified length would be attached from the spring to the strut so that when the knee passes beyond its final position and the wheel is only part-way to final position, the spring exerts a force in reverse to help the strut  snap into final position and lock.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 11:21 am
Post A on the accumulator is different from posts B and C in that it is located in the center of the accumulator instead of on its side. This means the top of the post would be congruent with plane ABC and the post itself would be emerged immersed inside a fuel tank. The second sketch above showed post A, which is installed inside the tank on the ground before launch, with short internal struts in place. Post A is only long enough to house the male connector device.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/11/2022 02:04 pm
Mentioned this earlier: If you're not talking about the existing QD plate used in ground operations, don't use the term QD.

For example, the QD is at the bottom of the ship, not half-way up the side. If you want a refuelling port half-way up the side, then it's not the QD.

[That said, there's nothing in Ionmars' proposal that requires the refuelling port to be half-way up the side. The tail-end QD would be fine.]
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 02:52 pm
Mentioned this earlier: If you're not talking about the existing QD plate used in ground operations, don't use the term QD.

For example, the QD is at the bottom of the ship, not half-way up the side. If you want a refuelling port half-way up the side, then it's not the QD.

[That said, there's nothing in Ionmars' proposal that requires the refuelling port to be half-way up the side. The tail-end QD would be fine.]
Thank you.
I will make these changes when I can.

How about propellant connectors plate (PCP)?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/11/2022 03:29 pm
And you'd be stupid to waste tens of tonnes of propellant for cold-gas ullage when you know how to build little hot-gas rockets and have plenty of room to stick them in the tail.
Unless of course you have a cold-gas thruster that could do the job and you haven't yet built and tested the little hot-gas rocket.

I'd expect initial refuelings, including the first moon missions, to be rough and ready rather than optimized.

It's very hard to predict rough and ready because it depends on factors with very little visibility.  It might help if some of the suggested solutions include a timeline for adoption.  The only things that can be ruled out a prior for the initial refuelings are things that can't do the job under any circumstances or are not in SpaceX's toolkit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/11/2022 03:33 pm
The second strut for post C would unfold in a different manner because it is located above the curvature of the ship .
Why are you deploying this strut?

Two of the struts can be combined into a fixed triangle*, with one or both struts arced to conform to the hull.

* Hinged at the base with the hull, which is the third side of the triangle.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 04:04 pm
The second strut for post C would unfold in a different manner because it is located above the curvature of the ship .
Why are you deploying this strut?

Two of the struts can be combined into a fixed triangle*, with one or both struts arced to conform to the hull.

* Hinged at the base with the hull, which is the third side of the triangle.
I am reluctant to consider the hull as a strut or stiffener because of its thin 4mm skin. There are many ways of producing rigidity; my particular approach has the post, struts, and stiffeners forming a four-sided (3 sides + base) pyramid, where the edges of each side form a rigid triangle.

Added: In the case of post A, it is so short that the added rigidity of additional stiffeners might not be required and the hull might be strong enough.

Added: For a perfectly nominal docking between vehicles, rigid struts might not even be required. The extra rigidity comes into play if and when a 200 t ship bumps a little too hard agains a post. From what direction and how hard a hit? It is an engineering and cost judgement for what extra strength to build in as a safety margin. There will always be some scenario that only ends badly.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/11/2022 06:04 pm
The main reason for rigid posts and struts is to ensure accuracy and safety during orbital refilling. Accuracy is required to align the PCPs of the two vehicle to prevent leaks at the PCP connections. Precision is required to perform refilling operations repeatedly. Safety is provided by rigid connections between vehicles during refilling; they may move in conjunction like a single unit.

This simple approach also avoids the complexity of a robotic arm or grappler to maintain connections between vehicles during refilling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/11/2022 06:10 pm
The main reason for rigid posts and struts is to ensure accuracy and safety during orbital refilling. Accuracy is required to align the PCPs of the two vehicle to prevent leaks at the PCP connections. Precision is required to perform refilling operations repeatedly. Safety is provided by rigid connections between vehicles during refilling; they may move in conjunction like a single unit.

This simple approach also avoids the complexity of a robotic arm or grappler to maintain connections between vehicles during refilling.
You only need actuation for gross alignment to get the alignment pins in their sockets. Once those are pushed in the latches are driven closed and the QD plates are held in rigid alignment until you drive the latches open again. Actuation of the plates is not a significant burden (it's what's already done for the QD plate on the ship QD arm), less so than adding extra structural hardpoints to a semiballon tank skin.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: VaBlue on 07/11/2022 07:06 pm
This seems like a complicated solution to something airplanes do all the time.  Aerial refueling is pretty commonplace, and has proven to be quite effective and safe (relatively speaking for such a dangerous thing).  Perhaps I'm over-simplifying because I'm not a rocket scientist?  But I can see a refueling arm, complete with flexible hoses and a QD plate, on the depot swinging out to meet the receiver.  Rendezvous is also pretty well known and has proven to be quite stable and reliable.  Match the orbits, mate the QD.  The arm can have plenty of flex available (via joints and extendibility) for sway between vehicles.

As I said, I've probably over-simplified things a great deal, but simple is usually better.  It's certainly more reliable - if it works to begin with.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/12/2022 12:01 am
This seems like a complicated solution to something airplanes do all the time.  Aerial refueling is pretty commonplace, and has proven to be quite effective and safe (relatively speaking for such a dangerous thing).  Perhaps I'm over-simplifying because I'm not a rocket scientist?  But I can see a refueling arm, complete with flexible hoses and a QD plate, on the depot swinging out to meet the receiver.  Rendezvous is also pretty well known and has proven to be quite stable and reliable.  Match the orbits, mate the QD.  The arm can have plenty of flex available (via joints and extendibility) for sway between vehicles.

As I said, I've probably over-simplified things a great deal, but simple is usually better.  It's certainly more reliable - if it works to begin with.
You may be right, but I set out to simplify what I perceived to be a more complicated robotic arm operation, like the QD arm at the Starbase launch tower. Would this arm work safely with two giant structures maneuvering in space with thrusters? Or would it be safer to immobilize their positions before beginning the refill operation?

Docking posts may be more complicated to set up but I believe in-space operations might be reasonably straight-forward  and potentially safer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/12/2022 11:27 am
If you can use RCS to line up and contact a set of rigid posts, you can use RCS to line up and contact the QD plates. Those also have hard-alignment features (ball-ended posts and matching sockets) and driveable latches, so it seems of little benefit to add an entire new docking system just to avoid using them. That you can also translate and rotate one of those plates makes the task easier, not harder.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/12/2022 01:16 pm
If you can use RCS to line up and contact a set of rigid posts, you can use RCS to line up and contact the QD plates. Those also have hard-alignment features (ball-ended posts and matching sockets) and driveable latches, so it seems of little benefit to add an entire new docking system just to avoid using them. That you can also translate and rotate one of those plates makes the task easier, not harder.
I believe you are absolutely correct about the feasibility of joining the two ships at the propellant connections wthout the docking posts.

However, the point is that any small movement of the two 200t ships, especially at the far end, would exert tremendous leveraged force at the connectors that could tear them apart. Safer to have the two ships rigidly attached to one another before refueling begins. Especially desirable to have connection points as far apart as feasible, so they can leverage and counteract any movement.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/12/2022 03:24 pm
Mentioned this earlier: If you're not talking about the existing QD plate used in ground operations, don't use the term QD.

For example, the QD is at the bottom of the ship, not half-way up the side. If you want a refuelling port half-way up the side, then it's not the QD.

[That said, there's nothing in Ionmars' proposal that requires the refuelling port to be half-way up the side. The tail-end QD would be fine.]
Thank you.
I will make these changes when I can.

How about propellant connectors plate (PCP)?
Well thought out and a nice level of detail.


I've some questions and observations.


- When you say a plumbing connection with no moving parts, are you limiting this to extending outward or do you include latching and related?
- Your design depends on the three fixed attachment posts to provide alignment for the PCP. ISTM that between changing sun and earth angle, along with the changing volumes of cryo propellants, it will be a complex thermal environment. Both initial and ongoing dimensions will be variables.
- With the PCP moved to the bottom would it make sense to use it as a hard connection point? Move the two connecting posts up to the top and loose the solo post that is up there. The PCP itself would have a small bit of float and be surrounded by a hard point bezel. This would act as the ultimate index point that the two posts would have to thermally accommodate.
- the two ships will be almost be touching along their length. This leaves no margin for the unexpected.
- Up thread it was suggested that the depot QD/PCP be gender bent to mate with ships and the GSE QD be (I'm floundering here) be gender unbent for depot launches. I like this idea.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/12/2022 03:43 pm
This seems like a complicated solution to something airplanes do all the time.  Aerial refueling is pretty commonplace, and has proven to be quite effective and safe (relatively speaking for such a dangerous thing).  Perhaps I'm over-simplifying because I'm not a rocket scientist?  But I can see a refueling arm, complete with flexible hoses and a QD plate, on the depot swinging out to meet the receiver.  Rendezvous is also pretty well known and has proven to be quite stable and reliable.  Match the orbits, mate the QD.  The arm can have plenty of flex available (via joints and extendibility) for sway between vehicles.

As I said, I've probably over-simplified things a great deal, but simple is usually better.  It's certainly more reliable - if it works to begin with.
Air to air refueling does not involve 100t+ of cryo fuel and oxidizer. It happens within the atmosphere which is both a blessing and a curse. They skip around the curses and embrace the blessings. It happens at a steady 1g.


See: [size=78%]https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/ (https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/)[/size]


Now that's a bit touchy. I expect the first time it was done in jets there were some little green apples in the cockpit.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/12/2022 03:49 pm
If you can use RCS to line up and contact a set of rigid posts, you can use RCS to line up and contact the QD plates.
However, the point is that any small movement of the two 200t ships, especially at the far end, would exert tremendous leveraged force ...
But where do those small movements come from?  If they can be modeled, or measured, you may be able to use the RCS to remove them.

The choice between dynamic and static stability should be data (and experiment) driven, not dogma driven.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/12/2022 06:16 pm

Well thought out and a nice level of detail.

I've some questions and observations.

Thanks. And your questions are spot on; I'll try to answer each one in a short post.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/12/2022 06:43 pm

- When you say a plumbing connection with no moving parts, are you limiting this to extending outward or do you include latching and related?

When I said that I had in mind the visiting ship landing on the posts and latching in place simultaneously with female connectors on the visitor mating with the male projections from the PCP on the accumulator, with no movement of robotic parts to complete the connection (a hard latching).

However, your question reminds me that I don't want to have any "pointy parts" projecting from any SS surface while launching through the atmosphere. So this means to me that it would be necessary for the PCP to launch with male connectors recessed below the surface. Then when in space the PCP would rise up sufficiently (20 cm?)to commence operations.

Still trying to minimize devices on the numerous visiting ships and putting them on the one accumulator.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/12/2022 07:28 pm

- Your design depends on the three fixed attachment posts to provide alignment for the PCP. ISTM that between changing sun and earth angle, along with the changing volumes of cryo propellants, it will be a complex thermal environment. Both initial and ongoing dimensions will be variables.

Correct again.

Varying expansions and contractions of parts would have to be accommodated in the detailed engineering specs and in operating procedures for the accumulator.

One example could be the materials composing the components of the male and female  connectors. The parts should have similar expansion ratios, but to to avoid cold fusion they should not have identical alloys.

Cold fusion is the welding together of identical materials in space. It does not occur on Earth because of microscope debris and oxides that prevent atoms of each component from coming in direct contact.

Another example could be variations in extreme temperatures that could cause small variations in the distances between posts. If one ship-to-post connection was a little off-center, an Earth-bound remote operator might be able to correct small errors. He/she could wait tor the accumulator to pass into Earth's shadow to cool down all components and shrink parts. Then he could release the latches on that post so that the two other post connections would either push or pull on the mis-aligned post connection. When the vehicle passed back into sunlight, re-latch.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/12/2022 07:46 pm

- With the PCP moved to the bottom would it make sense to use it as a hard connection point? Move the two connecting posts up to the top and loose the solo post that is up there. The PCP itself would have a small bit of float and be surrounded by a hard point bezel. This would act as the ultimate index point that the two posts would have to thermally accommodate.

Seems like this could be made to work.    :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/12/2022 11:36 pm

- the two ships will be almost be touching along their length. This leaves no margin for the unexpected.

On the contrary, the docking posts and the rigid connections between two ships ARE the safety margin for the unexpected.   :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/12/2022 11:54 pm
If you can use RCS to line up and contact a set of rigid posts, you can use RCS to line up and contact the QD plates.
However, the point is that any small movement of the two 200t ships, especially at the far end, would exert tremendous leveraged force ...
But where do those small movements come from?  If they can be modeled, or measured, you may be able to use the RCS to remove them.

The choice between dynamic and static stability should be data (and experiment) driven, not dogma driven.

Potential movements derive from unexpected events, e.g.

- A stuck valve causes a thruster to continue thrusting with no ending

- A remote (human) operator on Earth presses the wrong button

- Accumulator is hit by in-space debris

- Software error causes unexpected shift in propellant mass during refueling

In respect to the word "dogma" we are discussing a new out-of-box idea that has no history. So there cannot be "dogma."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/13/2022 05:36 pm

- the two ships will be almost be touching along their length. This leaves no margin for the unexpected.

On the contrary, the docking posts and the rigid connections between two ships ARE the safety margin for the unexpected.   :)
When your methods are to "test lots and don't worry if you break things" most of your margin is in the "test lots".  Testing turns "known unknowns" into "knowns" and flushes out some of the "unknown unknowns".  You don't need to start with hardware margins the way you would if you intend to land on the moon by the 16th launch.

It's not that I think ridged posts etc are bad or won't work and are not worth discussing.  It's just not yet, and maybe never.

I'd expect them to start with the simplest method that might work.  IMHO that would include trying formation flying or using the QD and dynamic stability.   They will learn something.  Maybe it will work, maybe they'll learn that they need a ridged coupling, but I expect them to try unless they know a priori that it can't work, and maybe even then because you can still learn by trying something that can't work.


Potential movements derive from unexpected events, e.g.

- A stuck valve causes a thruster to continue thrusting with no ending

- A remote (human) operator on Earth presses the wrong button

- Software error causes unexpected shift in propellant mass during refueling

These are the sort of events you can essentially eliminate (drive to a very low level) with sufficient testing.
Quote
- Accumulator is hit by in-space debris
And this is something you learn about and can decide whether or not it is worth taking countermeasures.


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/13/2022 07:06 pm

When your methods are to "test lots and don't worry if you break things" most of your margin is in the "test lots".  Testing turns "known unknowns" into "knowns" and flushes out some of the "unknown unknowns".  You don't need to start with hardware margins the way you would if you intend to land on the moon by the 16th launch.

It's not that I think ridged posts etc are bad or won't work and are not worth discussing.  It's just not yet, and maybe never.

I'd expect them to start with the simplest method that might work.  IMHO that would include trying formation flying or using the QD and dynamic stability.   They will learn something.  Maybe it will work, maybe they'll learn that they need a ridged coupling, but I expect them to try unless they know a priori that it can't work, and maybe even then because you can still learn by trying something that can't work.


Potential movements derive from unexpected events, e.g.

- A stuck valve causes a thruster to continue thrusting with no ending

- A remote (human) operator on Earth presses the wrong button

- Software error causes unexpected shift in propellant mass during refueling

These are the sort of events you can essentially eliminate (drive to a very low level) with sufficient testing.
Quote
- Accumulator is hit by in-space debris
And this is something you learn about and can decide whether or not it is worth taking countermeasures.
I agree with your development principles and I also agree that we should begin with the easy solution first.

It's just that when two ships are floating in space with the intent of transferring fluids, you will require something like a robotic arm to make connections between their PCPs before refueling can proceed. I don't see that approach being simpler than the docking posts, just more familiar.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/14/2022 07:21 am
It's just that when two ships are floating in space with the intent of transferring fluids, you will require something like a robotic arm to make connections between their PCPs before refueling can proceed. I don't see that approach being simpler than the docking posts, just more familiar.
A robot arm is not a necessity. Progress has been transferring propellants for over 40 years without one. If you can align and mate two surfaces, you can make plumbing connections and transfer fluids.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: frith01 on 07/14/2022 07:56 am
It's just that when two ships are floating in space with the intent of transferring fluids, you will require something like a robotic arm to make connections between their PCPs before refueling can proceed. I don't see that approach being simpler than the docking posts, just more familiar.
A robot arm is not a necessity. Progress has been transferring propellants for over 40 years without one. If you can align and mate two surfaces, you can make plumbing connections and transfer fluids.

Best part is no part, BUT there is a very high likelihood of Starship needing an arm (or two) for multiple purposes anyway.  Might as well use it if its going to be needed.  (as a crane for off-loading , releasing cargo, exchanging cargo, inspecting heat shield , solar panel setup, etc.)

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/14/2022 02:17 pm
It's just that when two ships are floating in space with the intent of transferring fluids, you will require something like a robotic arm to make connections between their PCPs before refueling can proceed. I don't see that approach being simpler than the docking posts, just more familiar.
A robot arm is not a necessity. Progress has been transferring propellants for over 40 years without one. If you can align and mate two surfaces, you can make plumbing connections and transfer fluids.

Yeah, but Progress flows propellants through a docking interface.  If anything, that's more evidence that a free-flying interface without something to make it more rigid won't work.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/14/2022 05:04 pm
It's just that when two ships are floating in space with the intent of transferring fluids, you will require something like a robotic arm to make connections between their PCPs before refueling can proceed. I don't see that approach being simpler than the docking posts, just more familiar.
A robot arm is not a necessity. Progress has been transferring propellants for over 40 years without one. If you can align and mate two surfaces, you can make plumbing connections and transfer fluids.

Best part is no part, BUT there is a very high likelihood of Starship needing an arm (or two) for multiple purposes anyway.  Might as well use it if its going to be needed.  (as a crane for off-loading , releasing cargo, exchanging cargo, inspecting heat shield , solar panel setup, etc.)
You are correct about SpaceX needing robotic arm(s) for multiple functions.

Each arm/device would operate from a location on Starship that is different, depending on the function, e.g. a crane operating from the payload bay to manage cargo or a mobile robot to inspect tiles and repair solar panels. I don't think you want one do-it-all robot for all these functions because it could become larger and more complicated than Canadarm operating across the framework of ISS. For transferring propellants between ships you would want a simple robot that operates near or on the PCP, just to join propellant connectors.

Unless you had some other alternative.  :)

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Hog on 07/14/2022 05:16 pm
This seems like a complicated solution to something airplanes do all the time.  Aerial refueling is pretty commonplace, and has proven to be quite effective and safe (relatively speaking for such a dangerous thing).  Perhaps I'm over-simplifying because I'm not a rocket scientist?  But I can see a refueling arm, complete with flexible hoses and a QD plate, on the depot swinging out to meet the receiver.  Rendezvous is also pretty well known and has proven to be quite stable and reliable.  Match the orbits, mate the QD.  The arm can have plenty of flex available (via joints and extendibility) for sway between vehicles.

As I said, I've probably over-simplified things a great deal, but simple is usually better.  It's certainly more reliable - if it works to begin with.
Air to air refueling does not involve 100t+ of cryo fuel and oxidizer. It happens within the atmosphere which is both a blessing and a curse. They skip around the curses and embrace the blessings. It happens at a steady 1g.


See: [size=78%]https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/ (https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/)[/size]


Now that's a bit touchy. I expect the first time it was done in jets there were some little green apples in the cockpit.
Bold and Bold emphasis mine.
a) While typical AAR doesn't involve cryo fluids, it can involve 100+ tonne fuel weights.

b) Not all AAR missions occur at 1g.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: SDSmith on 07/14/2022 05:31 pm
The original concept was docking tail to tail and refueling with lines in the engine bay that would connect. I believe it was killed due to the amount of plumbing required (memory can be fuzzy).

Why not return to docking tail to tail but have a robotic arm that can carry a plate to connect to the existing plate? Since the tail area is currently designed to dock with SH it might be easier. The tanker would be responsible for having the robotic arm and make the connection with SS.


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/14/2022 05:37 pm

Yeah, but Progress flows propellants through a docking interface.  If anything, that's more evidence that a free-flying interface without something to make it more rigid won't work.

If anything it's nothing.

Evidence that "it is possible to transfer fuel while docked" is not evidence that "it is not possible to transfer fuel while not docked".  Formally this logical fallacy is called "denying the antecedent", informally it's a dope slap.

Progress was designed to transfer pressurized cargo such as food as well as fuel.  Transferring food more or less requires docking.  It is no surprise that fuel is transferred after docking since Progress is already docked.  Evidence that the Soviets tried, but fail, to transfer fuel without a hard dock would be relevant, but the observation that Progress docks does not provide that evidence.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/14/2022 05:41 pm
For SpaceX using an approach where certain elements would be common: end effectors, joints, structural light weight tubing that can be made to various lengths, and as well as that iconic Canadarm "dumb" target grapple point. These arms "Robots" can be double ended or have a permanent fixed end. In significant quantities SpaceX could make a cheaper copy of a Canadarm3 arm. Such that with 2 providers of arms and a broader use of them the prices will drop. You could even have a version strong enough to operate in a 1/6 G field. A 2 ton of lifting force means that an object on the Moon of up to 6t mass could be lifted and placed on top of a Lunar surface habitat without having to have much significant other support than a mounting target grapple on the habitat itself that the new object is being lifted to mate on top. The basic here is that a pair of arms can be moved around to wherever the work is at easily.

The concepts are common so go with a generic and stay away from custom special unique designs all across the arm for each application. NASA has successfully gone away from the special to the generic mover/grabber.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/14/2022 06:04 pm
Does anyone have a notion of when refueling tests might start?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/14/2022 06:18 pm
Does anyone have a notion of when refueling tests might start?
Very soon after Starlink V2 deployments start as in the next flight. Starlink operational deployments will likely start on the first launch out of 39A in very late 2022 or early 2023. SpaceX is building enough R2 engines fast enough to build both a new SH and a SS every month. So even if SH reuse has yet to be achieved. Expect a launch rate of ~1 launch a month when adding launches at both KSC and BC.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/14/2022 06:22 pm
For SpaceX using an approach where certain elements would be common: end effectors, joints, structural light weight tubing that can be made to various lengths, and as well as that iconic Canadarm "dumb" target grapple point. These arms "Robots" can be double ended or have a permanent fixed end. In significant quantities SpaceX could make a cheaper copy of a Canadarm3 arm. Such that with 2 providers of arms and a broader use of them the prices will drop. You could even have a version strong enough to operate in a 1/6 G field. A 2 ton of lifting force means that an object on the Moon of up to 6t mass could be lifted and placed on top of a Lunar surface habitat without having to have much significant other support than a mounting target grapple on the habitat itself that the new object is being lifted to mate on top. The basic here is that a pair of arms can be moved around to wherever the work is at easily.

The concepts are common so go with a generic and stay away from custom special unique designs all across the arm for each application. NASA has successfully gone away from the special to the generic mover/grabber.
Might even be better with no robotic arm.

When SS lands on the three landing posts with their respective PCPs perfectly aligned with each other, they would be almost touching. The male connectors on the accumulator PCP would sink into the female receptors on the PCP of the visiting ship and become latched.. No robot needed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/14/2022 07:22 pm
Does anyone have a notion of when refueling tests might start?

Notion?  No.  Opinion?  Of course!

1) Every Starship launch is a test for both SuperHeavy and Starship recovery.

2) Until Starlink deployment is working, no work on refueling at all.

3) When Starship has on-orbit life of at least a week, you can test refueling with each launch, simply by leaving the previous launch in the Starlink deployment orbit as a target for the next launch to practice.  Then the previous launch can test recovery, and the new launch waits for the next mission to act as a target.¹

4) I don't think there's a lot of reason to work on a depot until both SuperHeavy and Starship are not only recoverable but also reusable.  There's probably a worst case where Starship isn't recoverable and SpaceX still does the Option A mission, but it's expensive for them.  I doubt that the system is viable if SuperHeavy can't be recovered and reused.

5) Depot has to be tested before Option A testing (either uncrewed or crewed) can proceed.

____________________
¹Weasel words:  If you don't yet have Starship recovery sorta-kinda working, and you think that there's a good chance that failed refueling ops are going to render the Starship unrecoverable (a collision, an explosion, a hung docking or QD failure, pranging the TPS, etc.), then it probably makes sense to forgo refueling tests until you get comfy with Starship recovery--it's more important.  But I suspect that eliminating most of the catastrophic refueling accidents doesn't take very long.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 07/14/2022 07:38 pm
A fully fueled 1200 ton depot, at 1 mili-g acceleration, is only 1.2 force-tons on the connector. Most tankers will only be around 400-ish tons when they transfer propellant, so 400 kilos of force.

Just throwing that out there, in the name of some hard numbers.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/14/2022 09:20 pm
The second strut for post C would unfold in a different manner because it is located above the curvature of the ship .
Why are you deploying this strut?

Two of the struts can be combined into a fixed triangle*, with one or both struts arced to conform to the hull.

* Hinged at the base with the hull, which is the third side of the triangle.
I am reluctant to consider the hull as a strut or stiffener because of its thin 4mm skin. There are many ways of producing rigidity; my particular approach has the post, struts, and stiffeners forming a four-sided (3 sides + base) pyramid, where the edges of each side form a rigid triangle.

Added: In the case of post A, it is so short that the added rigidity of additional stiffeners might not be required and the hull might be strong enough.

Added: For a perfectly nominal docking between vehicles, rigid struts might not even be required. The extra rigidity comes into play if and when a 200 t ship bumps a little too hard agains a post. From what direction and how hard a hit? It is an engineering and cost judgement for what extra strength to build in as a safety margin. There will always be some scenario that only ends badly.
The engine mounts on a Huey, H model and earlier, were a tripod, a bipod, and a monopod. The apex of the tripod and bipod were on the engine. Similar mounting problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/14/2022 09:21 pm
A fully fueled 1200 ton depot, at 1 mili-g acceleration, is only 1.2 force-tons on the connector. Most tankers will only be around 400-ish tons when they transfer propellant, so 400 kilos of force.

Just throwing that out there, in the name of some hard numbers.
But not during the transfer from the Depot to the LSS in which both could be at 50% prop loaded or ~800t each. A NOTE here is that it is highly likely that the LSS and Depot would hold 1500t of prop. So the amount of force is 800kg. But that is static force. With both SS having active RCS the max force could be several tons but over very short duration. The worst kind of situation for any QD. The intent would be to have some level of additional force relief to lesson QD failures/leaks.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Keldor on 07/14/2022 09:35 pm
A fully fueled 1200 ton depot, at 1 mili-g acceleration, is only 1.2 force-tons on the connector. Most tankers will only be around 400-ish tons when they transfer propellant, so 400 kilos of force.

Just throwing that out there, in the name of some hard numbers.
But not during the transfer from the Depot to the LSS in which both could be at 50% prop loaded or ~800t each. A NOTE here is that it is highly likely that the LSS and Depot would hold 1500t of prop. So the amount of force is 800kg. But that is static force. With both SS having active RCS the max force could be several tons but over very short duration. The worst kind of situation for any QD. The intent would be to have some level of additional force relief to lesson QD failures/leaks.

Alright, but both sides have RCS systems, and it really isn't hard to "fly in formation".  If both sides can match their accelleration within a 1% error margin, then the remaining load through the QD is now only 8kg.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/14/2022 09:57 pm
A fully fueled 1200 ton depot, at 1 mili-g acceleration, is only 1.2 force-tons on the connector. Most tankers will only be around 400-ish tons when they transfer propellant, so 400 kilos of force.

Just throwing that out there, in the name of some hard numbers.
But not during the transfer from the Depot to the LSS in which both could be at 50% prop loaded or ~800t each. A NOTE here is that it is highly likely that the LSS and Depot would hold 1500t of prop. So the amount of force is 800kg. But that is static force. With both SS having active RCS the max force could be several tons but over very short duration. The worst kind of situation for any QD. The intent would be to have some level of additional force relief to lesson QD failures/leaks.

Alright, but both sides have RCS systems, and it really isn't hard to "fly in formation".  If both sides can match their accelleration within a 1% error margin, then the remaining load through the QD is now only 8kg.
Prop transfer is not a static case. It is dynamic and no assumptions can be made as to how little the forces would be but only how high they could. Also the system performing the prop settling acceleration can either provide enough thrust for when the Depot is almost full, in which case the acceleration when it is empty and docked to a tanker could be 4 times as much. The alternate is a much more complex system that adjusts thrust to deal with the total lower mass to 25% that of when full.

When dealing with feedback control GNC RCS systems they act more like as a controlled state of chaos. Expecting a systems to act exactly a certain way is sure way of creating a failure.

NOTE have experience with large system LV GNC software and hardware.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/14/2022 10:09 pm
It's just that when two ships are floating in space with the intent of transferring fluids, you will require something like a robotic arm to make connections between their PCPs before refueling can proceed. I don't see that approach being simpler than the docking posts, just more familiar.
A robot arm is not a necessity. Progress has been transferring propellants for over 40 years without one. If you can align and mate two surfaces, you can make plumbing connections and transfer fluids.

Best part is no part, BUT there is a very high likelihood of Starship needing an arm (or two) for multiple purposes anyway.  Might as well use it if its going to be needed.  (as a crane for off-loading , releasing cargo, exchanging cargo, inspecting heat shield , solar panel setup, etc.)
You are correct about SpaceX needing robotic arm(s) for multiple functions.

Each arm/device would operate from a location on Starship that is different, depending on the function, e.g. a crane operating from the payload bay to manage cargo or a mobile robot to inspect tiles and repair solar panels. I don't think you want one do-it-all robot for all these functions because it could become larger and more complicated than Canadarm operating across the framework of ISS. For transferring propellants between ships you would want a simple robot that operates near or on the PCP, just to join propellant connectors.

Unless you had some other alternative.  :)
The current arm has attachment points scattered about the ISS. It also has an end effector at each end that can grab the attachment points and inchworm itself end-over-end from one point to the next. It goes where it's needed.

An arm is a general purpose tool. In my experience, a GP tool can do many things but a special purpose tool will fulfill its purpose better at the cost of only being able to do that one thing. If a specific job must be done repeatedly a special tool may be justified.

The struts would be deployed once. If an arm is available, use it. Same for PV, radiators, cargo deployment or transfers. Maybe to draw the ships together. Maybe even to aid in removing engines so repositioning becomes more practical.

This is for a fully matured system. Start simple, maybe a design lifetime of only one refueling campaign, then ditch it. Next one will be better. Eventually (10 years?) it'll be Depot Heavy, a complex bit of kit with its place in the burgeoning space infrastructure.

Edit: Ninja'd by OldAtlas.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/14/2022 10:24 pm
This seems like a complicated solution to something airplanes do all the time.  Aerial refueling is pretty commonplace, and has proven to be quite effective and safe (relatively speaking for such a dangerous thing).  Perhaps I'm over-simplifying because I'm not a rocket scientist?  But I can see a refueling arm, complete with flexible hoses and a QD plate, on the depot swinging out to meet the receiver.  Rendezvous is also pretty well known and has proven to be quite stable and reliable.  Match the orbits, mate the QD.  The arm can have plenty of flex available (via joints and extendibility) for sway between vehicles.

As I said, I've probably over-simplified things a great deal, but simple is usually better.  It's certainly more reliable - if it works to begin with.
Air to air refueling does not involve 100t+ of cryo fuel and oxidizer. It happens within the atmosphere which is both a blessing and a curse. They skip around the curses and embrace the blessings. It happens at a steady 1g.


See: [size=78%]https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/ (https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/)[/size]


Now that's a bit touchy. I expect the first time it was done in jets there were some little green apples in the cockpit.
Bold and Bold emphasis mine.
a) While typical AAR doesn't involve cryo fluids, it can involve 100+ tonne fuel weights.

b) Not all AAR missions occur at 1g.
I stand corrected. KC-10 at 166 tons. Can you explain missions not at 1g?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/14/2022 10:56 pm
A fully fueled 1200 ton depot, at 1 mili-g acceleration, is only 1.2 force-tons on the connector. Most tankers will only be around 400-ish tons when they transfer propellant, so 400 kilos of force.

Just throwing that out there, in the name of some hard numbers.
But not during the transfer from the Depot to the LSS in which both could be at 50% prop loaded or ~800t each. A NOTE here is that it is highly likely that the LSS and Depot would hold 1500t of prop. So the amount of force is 800kg. But that is static force. With both SS having active RCS the max force could be several tons but over very short duration. The worst kind of situation for any QD. The intent would be to have some level of additional force relief to lesson QD failures/leaks.

Alright, but both sides have RCS systems, and it really isn't hard to "fly in formation".  If both sides can match their accelleration within a 1% error margin, then the remaining load through the QD is now only 8kg.
Upthread it was pointed out that cold gas would most probably be on/off with the length of 'on' defining impulse. With a mass differential between ships it might be hard to keep the thrust within 1%.

I'd hate to see all ships needing special thrusters for propellant transfer. Best if only the depot has special hardware but then it needs to either vector control or have proportional thrust through two or more thrusters to keep thrust through the CG.

It looks like either all ships get special refueling thrusters or they're rigidly connected somewhere other than the QD.

A point to keep in mind is that every fluid movement will be adding or cancelling impulse somewhere. It'll all balance out at the system level in the long run but till that happens ISTM that the ships will have some independent jiggling.


Edit: Ninja'd by OldAtlas AGAIN! Great minds... 8)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/14/2022 11:29 pm
At the moment the only way I can see to solve the controlled state of the two bodies after docking under acceleration. That is to have a well modeled on orbit structural, prop slosh/settling, and RCS models. For both the Tanker and the Depot as well as for the LSS. Then make the Tanker a slave to the Depot's GNC. Where the Depot's GNC completely controls the RCS on the Tanker or the LSS. Thus the acceleration of both vehicles are managed to be the same within some level of error that over some time averages to very close to 0. This basically means that at least one of the QD plates need to move independently of the vehicle which is their owner. It might actually be possible to do this without any hard docks at all and only a soft dock. So long as there are no GNC or RCS anomalies that create issues that have errors greater than what the flexible plate can handle. If that occurs then the QD separates immediately after the valves close and the lighter if its GNC and RCS is operating OK to move away. Else the one with a fully functional GNC RCS moves away.

The other method is a hard dock with enough strength to handle the excess forces from the GNC RCS errors by the two vehicles.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Hog on 07/15/2022 10:09 am
This seems like a complicated solution to something airplanes do all the time.  Aerial refueling is pretty commonplace, and has proven to be quite effective and safe (relatively speaking for such a dangerous thing).  Perhaps I'm over-simplifying because I'm not a rocket scientist?  But I can see a refueling arm, complete with flexible hoses and a QD plate, on the depot swinging out to meet the receiver.  Rendezvous is also pretty well known and has proven to be quite stable and reliable.  Match the orbits, mate the QD.  The arm can have plenty of flex available (via joints and extendibility) for sway between vehicles.

As I said, I've probably over-simplified things a great deal, but simple is usually better.  It's certainly more reliable - if it works to begin with.
Air to air refueling does not involve 100t+ of cryo fuel and oxidizer. It happens within the atmosphere which is both a blessing and a curse. They skip around the curses and embrace the blessings. It happens at a steady 1g.


See: [size=78%]https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/ (https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/)[/size]


Now that's a bit touchy. I expect the first time it was done in jets there were some little green apples in the cockpit.
Bold and Bold emphasis mine.
a) While typical AAR doesn't involve cryo fluids, it can involve 100+ tonne fuel weights.

b) Not all AAR missions occur at 1g.
I stand corrected. KC-10 at 166 tons. Can you explain missions not at 1g?
Not sure if it's done much anymore with other types, but during A-12/SR-71/M-21/YF-12 "Blackbird" AAR missions, the refueller/refuellee combo would "connect" at higher altitudes as both a/c were in a slight dive. Resulting in small amounts of less than 1g.
If the operation was performed in straight/level flight, the refuellee would have to "stroke a single engine into min A/B" whilst controlling its speed with the opposite throttle.(pulling the non-A/B engine down from mil power.)
These 2 maneuvers were required as the refueller would begin at near max speed and as the refuellee got heavy with fuel, it couldn't maintain speed with both engines at Mil power, therefore a descending maneuver or the single engine min a/b maneuver were required.  IIRC  The single min a/b maneuver became the norm.

Conversely, if the combo had to perform a Wifferdill turn, the combo could see g loading surpassing 1 g.
AIUI

xxxxxx
Apparently Gemini-10 had to perform a Wifferdil "the need for the maneuver was a result of misalignment of the inertial guidance platform, causing axial drift of the approach trajectory, prior to closing to docking distance." Wiki

attachment
KC-135 and B-52 engaged in a Wifferdill/dale turn during AAR ops
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/15/2022 11:28 pm
This seems like a complicated solution to something airplanes do all the time.  Aerial refueling is pretty commonplace, and has proven to be quite effective and safe (relatively speaking for such a dangerous thing).  Perhaps I'm over-simplifying because I'm not a rocket scientist?  But I can see a refueling arm, complete with flexible hoses and a QD plate, on the depot swinging out to meet the receiver.  Rendezvous is also pretty well known and has proven to be quite stable and reliable.  Match the orbits, mate the QD.  The arm can have plenty of flex available (via joints and extendibility) for sway between vehicles.

As I said, I've probably over-simplified things a great deal, but simple is usually better.  It's certainly more reliable - if it works to begin with.
Air to air refueling does not involve 100t+ of cryo fuel and oxidizer. It happens within the atmosphere which is both a blessing and a curse. They skip around the curses and embrace the blessings. It happens at a steady 1g.


See: [size=78%]https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/ (https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/197385/first-air-to-air-refueling/)[/size]


Now that's a bit touchy. I expect the first time it was done in jets there were some little green apples in the cockpit.
Bold and Bold emphasis mine.
a) While typical AAR doesn't involve cryo fluids, it can involve 100+ tonne fuel weights.

b) Not all AAR missions occur at 1g.
I stand corrected. KC-10 at 166 tons. Can you explain missions not at 1g?
Not sure if it's done much anymore with other types, but during A-12/SR-71/M-21/YF-12 "Blackbird" AAR missions, the refueller/refuellee combo would "connect" at higher altitudes as both a/c were in a slight dive. Resulting in small amounts of less than 1g.
If the operation was performed in straight/level flight, the refuellee would have to "stroke a single engine into min A/B" whilst controlling its speed with the opposite throttle.(pulling the non-A/B engine down from mil power.)
These 2 maneuvers were required as the refueller would begin at near max speed and as the refuellee got heavy with fuel, it couldn't maintain speed with both engines at Mil power, therefore a descending maneuver or the single engine min a/b maneuver were required.  IIRC  The single min a/b maneuver became the norm.

Conversely, if the combo had to perform a Wifferdill turn, the combo could see g loading surpassing 1 g.
AIUI

xxxxxx
Apparently Gemini-10 had to perform a Wifferdil "the need for the maneuver was a result of misalignment of the inertial guidance platform, causing axial drift of the approach trajectory, prior to closing to docking distance." Wiki

attachment
KC-135 and B-52 engaged in a Wifferdill/dale turn during AAR ops
Never knew about that. Still, what I was getting at was gravity settled fluids. No ullage settling. 'Crept that wacky turn.


That semi-botched Russian docking a couple months back, the problem was a leaking tank bellows. Gas pressure outside the bellows normally compresses the bellows filled with propellant. Don't remember the exact details of the problem. Anyways, that shows a bellows as a known Russian engineering approach to handling ullage. I'd venture a guess this is how they handle propellant transfer to the ISS. In microgravity a bellows or thrust settling is what you got. Until they start spinning things up.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 07/16/2022 08:45 am
A fully fueled 1200 ton depot, at 1 mili-g acceleration, is only 1.2 force-tons on the connector. Most tankers will only be around 400-ish tons when they transfer propellant, so 400 kilos of force.

Just throwing that out there, in the name of some hard numbers.
But not during the transfer from the Depot to the LSS in which both could be at 50% prop loaded or ~800t each. A NOTE here is that it is highly likely that the LSS and Depot would hold 1500t of prop. So the amount of force is 800kg. But that is static force. With both SS having active RCS the max force could be several tons but over very short duration. The worst kind of situation for any QD. The intent would be to have some level of additional force relief to lesson QD failures/leaks.

Alright, but both sides have RCS systems, and it really isn't hard to "fly in formation".  If both sides can match their accelleration within a 1% error margin, then the remaining load through the QD is now only 8kg.
Upthread it was pointed out that cold gas would most probably be on/off with the length of 'on' defining impulse. With a mass differential between ships it might be hard to keep the thrust within 1%.

I'd hate to see all ships needing special thrusters for propellant transfer. Best if only the depot has special hardware but then it needs to either vector control or have proportional thrust through two or more thrusters to keep thrust through the CG.

It looks like either all ships get special refueling thrusters or they're rigidly connected somewhere other than the QD.

A point to keep in mind is that every fluid movement will be adding or cancelling impulse somewhere. It'll all balance out at the system level in the long run but till that happens ISTM that the ships will have some independent jiggling.


Edit: Ninja'd by OldAtlas AGAIN! Great minds... 8)
It's worth mentioning that you don't need to count both ships mass, as the worst case is one tankee "dangling" from the connector while the connected ship provides all the thrust. It doesn't matter if the ship providing the thrust is partially or fully fueled, only how much mass is being held "at arms length" by the connector... And at the expected settling acceleration, that's under 2 force tons at the end of the lever-arm.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/16/2022 10:35 am
ISTM that (1) flying in formation with an arm to transfer fluid and (2) using landing posts are both feasible and practical approaches to in-space refilling.

One metric to compare the two could be total propellant usage [p(t)] during refilling operations.

(1) p(t) = p(to establish connections between vehicles + p(to maintain formation during refilling)

(2) p(t) = p(to establish connections between vehicles)


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/16/2022 11:13 am
Propellant consumption for landing on three posts is probably greater than usage of propellants when connecting at one location while in formation. But the difference may be small.

The sketch below is from Reply #1087 above.
If a visiting ship has its B-mating device correctly aligned directly across from post B, then its A-mating device should be almost aligned at the far end. With a few extra pushes from thrusters, A and B mating devices would be aligned with their respective posts and C would be aligned automatically. Final docking could commence.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/16/2022 06:34 pm
Does anyone have a notion of when refueling tests might start?

Notion?  No.  Opinion?  Of course!

1) Every Starship launch is a test for both SuperHeavy and Starship recovery.

2) Until Starlink deployment is working, no work on refueling at all.

This depends in part on how you define "work on refueling" and "refuelling tests".

There is some ship handling and (cryogenic liquid) cargo handling that can be tested early and perhaps on every launch.  Some of this is general purpose but I would expect the refueling people to be very interested and maybe request particular tests.  Some of this could be almost invisible unless SpaceX makes an announcement.

I'm thinking about things like testing precision control, flying a fixed path, or in formation with a starlink satellite.  Putting a camera and sensors in the fuel tank and watching what happens to residual fuel under ullage and maneuvering thrust.  Pumping fuel between main and header tanks.  Perhaps pumping residual fuel overboard.

At the least this would help verify modelling and actual capabilities, which is a big help if you're designing refueling procedures based on a model and presumed capabilities.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/16/2022 08:52 pm
Does anyone have a notion of when refueling tests might start?

Notion?  No.  Opinion?  Of course!

1) Every Starship launch is a test for both SuperHeavy and Starship recovery.

2) Until Starlink deployment is working, no work on refueling at all.

This depends in part on how you define "work on refueling" and "refuelling tests".

There is some ship handling and (cryogenic liquid) cargo handling that can be tested early and perhaps on every launch.  Some of this is general purpose but I would expect the refueling people to be very interested and maybe request particular tests.  Some of this could be almost invisible unless SpaceX makes an announcement.

I'm thinking about things like testing precision control, flying a fixed path, or in formation with a starlink satellite.  Putting a camera and sensors in the fuel tank and watching what happens to residual fuel under ullage and maneuvering thrust.  Pumping fuel between main and header tanks.  Perhaps pumping residual fuel overboard.

At the least this would help verify modelling and actual capabilities, which is a big help if you're designing refueling procedures based on a model and presumed capabilities.
A campaign of subsystem tests mostly on just needing a single SS to test them. Only when "docking", QD seal, flight control, prop settling with 2 vehicles, and transfer test phase would 2 on orbit SS are needed. Hopefully all of this will be over a short list of flights but could take longer if significant difficulties and significant redesigns of prop transfer subsystems is needed. Likely most of 2023 with hopefully a successful transfer design and demonstration before the end of 2023. Unfortunately testing and proofing and redesign consumes time even for SpaceX. Almost all of these can be concurrent to the improvements being made on the SS to be able to recover and then refly an SS. Also at the same time the SS are in a semi operation mission mode of deploying Starlink V2 sats. 2023 is going to be a busy and revealing year for Starship's capabilities on a practical standpoint.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/17/2022 05:32 pm
Does anyone have a notion of when refueling tests might start?

Notion?  No.  Opinion?  Of course!

1) Every Starship launch is a test for both SuperHeavy and Starship recovery.

2) Until Starlink deployment is working, no work on refueling at all.

This depends in part on how you define "work on refueling" and "refuelling tests".

There is some ship handling and (cryogenic liquid) cargo handling that can be tested early and perhaps on every launch.  Some of this is general purpose but I would expect the refueling people to be very interested and maybe request particular tests.  Some of this could be almost invisible unless SpaceX makes an announcement.

I'm thinking about things like testing precision control, flying a fixed path, or in formation with a starlink satellite.  Putting a camera and sensors in the fuel tank and watching what happens to residual fuel under ullage and maneuvering thrust.  Pumping fuel between main and header tanks.  Perhaps pumping residual fuel overboard.

At the least this would help verify modelling and actual capabilities, which is a big help if you're designing refueling procedures based on a model and presumed capabilities.
Another thing to test is long term capabilities. It'll be a while before they'll be able to launch more often than once a month (guesstimate), so if they plan to do any proximity tests they need to keep a ship alive for a month while it waits for the second ship. They'll need power and propellant and attitude control.

They'll need to know what sort of boiloff to expect (more testing) and my gut sez LEO boiloff will be high enough that it'll be a problem. Maybe a shade, maybe some sacrificial thermal white on the dorsal side. Either of those should be visible.

For power they could add more batteries but without knowing the expected power draw, it's hard to figure. AIUI, batteries don't last too long with the thermal cycling at LEO. Is it normal practice to insulate sat batteries? Maybe we'll see some combination of battery insulation and sacrificial conformal PV.


Edit to add: that last got me thinking. Insulation and heaters would do for battery conditioning in the cold. Would long term use benefit from cooling during the hot cycle? If so, could this be a twofer for maintaining ullage pressure/RCS gasses on the methane side?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/17/2022 07:23 pm
Does anyone have a notion of when refueling tests might start?

Notion?  No.  Opinion?  Of course!

1) Every Starship launch is a test for both SuperHeavy and Starship recovery.

2) Until Starlink deployment is working, no work on refueling at all.

This depends in part on how you define "work on refueling" and "refuelling tests".

There is some ship handling and (cryogenic liquid) cargo handling that can be tested early and perhaps on every launch.  Some of this is general purpose but I would expect the refueling people to be very interested and maybe request particular tests.  Some of this could be almost invisible unless SpaceX makes an announcement.

I'm thinking about things like testing precision control, flying a fixed path, or in formation with a starlink satellite.  Putting a camera and sensors in the fuel tank and watching what happens to residual fuel under ullage and maneuvering thrust.  Pumping fuel between main and header tanks.  Perhaps pumping residual fuel overboard.

At the least this would help verify modelling and actual capabilities, which is a big help if you're designing refueling procedures based on a model and presumed capabilities.
A campaign of subsystem tests mostly on just needing a single SS to test them. Only when "docking", QD seal, flight control, prop settling with 2 vehicles, and transfer test phase would 2 on orbit SS are needed. Hopefully all of this will be over a short list of flights but could take longer if significant difficulties and significant redesigns of prop transfer subsystems is needed. Likely most of 2023 with hopefully a successful transfer design and demonstration before the end of 2023. Unfortunately testing and proofing and redesign consumes time even for SpaceX. Almost all of these can be concurrent to the improvements being made on the SS to be able to recover and then refly an SS. Also at the same time the SS are in a semi operation mission mode of deploying Starlink V2 sats. 2023 is going to be a busy and revealing year for Starship's capabilities on a practical standpoint.

I would add two-ship prox ops to the list of things you probably don't want to risk as long as collecting EDL data is more important, but yeah, I agree with most of that list.

If Starship runs into some difficult or intractable problems with EDL, then SpaceX has some decisions to make with respect to Option A HLS.  The moment that refueling becomes the critical path item, then the riskier refueling tests need to preempt the most robust versions of EDL testing.  That's an expensive decision for SpaceX, but IMO it pays back more in the long run than a pure "first EDL, then refueling" approach.

Ideally, every Starship launched can be used for both a refueling RPOD and an EDL test.  Realistically, anything even slightly risky with refueling probably gets deferred until the EDL folks have achieved recoverability--or proven that this version of Starship is barking up the wrong tree.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/18/2022 03:03 pm
Does anyone have a notion of when refueling tests might start?

Notion?  No.  Opinion?  Of course!

1) Every Starship launch is a test for both SuperHeavy and Starship recovery.

2) Until Starlink deployment is working, no work on refueling at all.

This depends in part on how you define "work on refueling" and "refuelling tests".

There is some ship handling and (cryogenic liquid) cargo handling that can be tested early and perhaps on every launch.  Some of this is general purpose but I would expect the refueling people to be very interested and maybe request particular tests.  Some of this could be almost invisible unless SpaceX makes an announcement.

I'm thinking about things like testing precision control, flying a fixed path, or in formation with a starlink satellite.  Putting a camera and sensors in the fuel tank and watching what happens to residual fuel under ullage and maneuvering thrust.  Pumping fuel between main and header tanks.  Perhaps pumping residual fuel overboard.

At the least this would help verify modelling and actual capabilities, which is a big help if you're designing refueling procedures based on a model and presumed capabilities.
A campaign of subsystem tests mostly on just needing a single SS to test them. Only when "docking", QD seal, flight control, prop settling with 2 vehicles, and transfer test phase would 2 on orbit SS are needed. Hopefully all of this will be over a short list of flights but could take longer if significant difficulties and significant redesigns of prop transfer subsystems is needed. Likely most of 2023 with hopefully a successful transfer design and demonstration before the end of 2023. Unfortunately testing and proofing and redesign consumes time even for SpaceX. Almost all of these can be concurrent to the improvements being made on the SS to be able to recover and then refly an SS. Also at the same time the SS are in a semi operation mission mode of deploying Starlink V2 sats. 2023 is going to be a busy and revealing year for Starship's capabilities on a practical standpoint.

I would add two-ship prox ops to the list of things you probably don't want to risk as long as collecting EDL data is more important, but yeah, I agree with most of that list.

If Starship runs into some difficult or intractable problems with EDL, then SpaceX has some decisions to make with respect to Option A HLS.  The moment that refueling becomes the critical path item, then the riskier refueling tests need to preempt the most robust versions of EDL testing.  That's an expensive decision for SpaceX, but IMO it pays back more in the long run than a pure "first EDL, then refueling" approach.

Ideally, every Starship launched can be used for both a refueling RPOD and an EDL test.  Realistically, anything even slightly risky with refueling probably gets deferred until the EDL folks have achieved recoverability--or proven that this version of Starship is barking up the wrong tree.
Agree. EDL is the tough nut. The first big risk in transfer testing is when two ships make contact. All precursor testing is fairly risk free.


AIUI, Boca will be allowed six launches a year. The Cape may come on line but let's ignore it for now. They can launch every two months. How deep is the assembly pipeline? By assembly I mean "putting it all together".  Let's call it six weeks for the sake of argument. This is important because it sets some limits on iteration.


Here's a worst case scenario. SX decides the tile retaining pins need a redesign. The ship that is stacking already has pins and the barrel sections waiting for the next ship has them too. SX is impossible to predict but let me put on my faux CTO hat. I'd say "it'll never make it through EDL but we have starlinks to launch. Launch that sucker and let's do some risky propellant transfer tests. Scrap the barrels and start over on that one."


The pipeline may be deeper, the Cape may come on line and most changes won't reach that far back into the pipeline. Still, this shows how two conflicting priorities MAY be resolved.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/18/2022 06:08 pm
By indication of the R2 engine production rates and even if no reuse has yet to be achieved on SS or an SH. SpaceX would be able to launch in total 1 Starship a month. 4 or 5 at BC and 7 or 8 at KSC 39A in 2023.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/19/2022 12:16 am
"[...] and let's do some risky propellant transfer tests. [...]"

IMO, once you understand your RCS at a reasonable level, prox.ops isn't that risky. You know where your limitations and issues are, you know where to put extra stops and check-points. Hence the chance of a mishap that causes loss-of-vehicle and therefore prevents an EDL test is low. (Other failures that don't destroy the ship, obviously, will be much higher.) Hence RPOD tests are likely before EDL is even vaguely dialled in.

Indeed, EDL feels higher risk than RPOD for a company that built Dragon successfully, and that may result in them pushing to do RPOD early because you're losing so many ships during reentry, necessitating building more, which is a wasted opportunity if you're not also iterating new RPOD & prop-xfer hardware in each new ship as well.

That last paragraph might not have been very clear... So...

Say there's 6 generations of hardware needed in order to iterate a reasonably final design for the docking/refuelling, and there's 10 generations to get EDL right.

Doing EDL first, then RPOD, takes 16 generations. Doing RPOD then EDL takes 10, which you had to do anyway because of the EDL failures.

You would have to lose more than 6 ships to EDL-preventing orbital collisions/ruptures/etc in order to be worse off testing EDL+RPOD simultaneously. And I find that unlikely. One or two? Sure. But six? No. If they can't do prox.ops after three catastrophic failures (and subsequent redesigns), they'll need to fundamentally change strategy anyway.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/19/2022 03:46 am
IMO, once you understand your RCS at a reasonable level, prox.ops isn't that risky. You know where your limitations and issues are, you know where to put extra stops and check-points.

In the hope that somebody will explain to me why this isn't a big deal, I'm going to pound the table yet again on the RCS issue, with two major objections:

1) The thrust range required to accomplish all of the different requirements of the RCS system (attitude control across axes with wildly different moments of inertia, ullage, prox ops, small delta-v maneuvers, and the swoop-n-slam maneuver for landing) is huge.

2) Nobody's explained how a hot (ignited) gas thruster has a sufficiently small minimum impulse bit for adequate control.

This is a lot different than the Shuttle, both in terms of the mass and moments, because the Shuttle was basically unfueled on-orbit, and because the Shuttle was able to leverage a wealth of prior art with hypergolics.  Starship has neither advantage.

I still think that the most likely outcome is a combination of fairly hefty hot-gas and fairly feeble cold-gas thrusters, integrated into a single guidance and control system.  That's doable--hey, it's only software--but I don't think you get to retire oodles of risk right at the git-go.

Quote
Indeed, EDL feels higher risk than RPOD for a company that built Dragon successfully, and that may result in them pushing to do RPOD early because you're losing so many ships during reentry, necessitating building more, which is a wasted opportunity if you're not also iterating new RPOD & prop-xfer hardware in each new ship as well.

If you're right and I'm wrong about RCS, then I agree with you here.  But you have to retire whatever RCS risk there is before you can start using the "every launch has an RPOD and an EDL test" strategy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/19/2022 09:42 am
1) The thrust range required to accomplish all of the different requirements of the RCS system (attitude control across axes with wildly different moments of inertia, ullage, prox ops, small delta-v maneuvers, and the swoop-n-slam maneuver for landing) is huge.
Your minimum thrust bit can be arbitrarily low, and your maximum RCS thrust level only places a limit on how long a given manoeuvre takes. High-thrust RCS has been eliminated as a requirement for EDL, so RCS thrust can be very low indeed if desired. The sole RPOD event that may require high thrust (rapid back-away) can be accomplished with the existing Starship dorsal tank dump vents.
Quote
2) Nobody's explained how a hot (ignited) gas thruster has a sufficiently small minimum impulse bit for adequate control.
Hot-gas thrusters may not even be present, let alone the sole RCS mechanism. RCS thrust of record is ullage gas venting (cold has thrusters) which can be of incredibly low thrust and highly variable in thrust - depending only on upstream valving to modulate final valve inlet pressure and on final valve actuation speed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/19/2022 11:50 am
IMO, once you understand your RCS at a reasonable level, prox.ops isn't that risky. You know where your limitations and issues are, you know where to put extra stops and check-points.
I'm going to pound the table yet again on the RCS issue, with two major objections:
1) The thrust range required to accomplish all of the different requirements of the RCS system [...] is huge. [...]

Your objections are design issues, not testing. It's not like they are going to launch a ship without knowing the thrust of its RCS system.

What matters for testing is whether it's reliable under flight conditions, whether it performs as spec, whether it has weird fail modes, etc.

Remember, the RCS is a component of EDL. If the RCS isn't reliable, you can't reliably control vehicle orientation, so testing the reliability of RCS isn't something you can do "after" you've perfected EDL, even if you wanted to. The RCS has to perform as expected in order to merely orient the ship, so you are "testing" that system by just operating in space, long before you do any prox.ops, let alone anything more demanding. In doing so, you'll characterise the RCS performance in orbit in detail, that lets you design your test envelope for incremental RPOD testing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/19/2022 04:36 pm
RCS is not just thrusters. It includes the IMU sensors, computers, software (very important), batteries, IMU calibration systems (likely GPS for LEO ops), and communications command and control (C3) (which includes telemetry).

Such that SpaceX is not assuming that RCS will work for the first flight correctly on orbit. Evidence is by the trajectory where reentry will occur without RCS working correctly on orbit. An absolute necessity is that RCS works to orient the vehicle for a reentry burn.

If data is gathered on the on-orbit behavior of the RCS on this first mission, then most of the info will be there to develop the detailed models needed for most of the eventual 2 ship maneuvers for a prop transfer.

All of this is not that SpaceX doesn't think that the RCS on orbit will work. But that until shown it does work to be conservative and reduce risks from uncontrolled reentry points.

As more understanding on the ability to control the SS under all of its operating modes (RCS) for launch, on-orbit, and EDL the remaining mode of operation of docking 2 SS on-orbit should not be difficult beyond final validation by actually accomplishing it on-orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/19/2022 05:06 pm
As a comparison, SpaceX flew a single Dragon 1 flight before engaging in ISS prox-ops, and Dragon 2 went straight to docking. Since Starship would not be rendezvousing with the ISS and potentially risking human life in case of a failure, and instead merely risking a second unmanned test vehicle - which would likely be obsoleted and scrapped on return anyway, at least for the first few flights - I suspect the only reason they would not test prox-ops on the first flight would be the need to then delay it and sit around and wait for a second vehicle to be ready.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/19/2022 07:26 pm
Your minimum thrust bit can be arbitrarily low...

Not if it's hot gas with an igniter, it can't.

Quote
...and your maximum RCS thrust level only places a limit on how long a given manoeuvre takes.  High-thrust RCS has been eliminated as a requirement for EDL, so RCS thrust can be very low indeed if desired.

The maximum must be high enough to produce a controlled system.  You have to be able to null out a rotation within some period of time to prevent collisions. 

Relying on Dragon as a good example of the kind of rate requirements you need, if you create a simpleminded model using a 12t D2 with an average moment of inertia of of 48,000 kg-m², that gives you an x-axis angular acceleration of 0.14º/s.  If you scale that to a 270t Starship with a CoM that's 20m from the bottom, to get a similar y- or z-axis acceleration requires a thruster (or thrusters) with 264kN of thrust.

Lots of caveats with that calculation, given that I guessed on the CoM, and maybe you can go somewhat lower on your angular acceleration.  But I doubt you can go much lower.  The r² term for the moment is not your friend.

Quote
The sole RPOD event that may require high thrust (rapid back-away) can be accomplished with the existing Starship dorsal tank dump vents.

Don't forget the lunar landing thrusters.

Quote
Quote
2) Nobody's explained how a hot (ignited) gas thruster has a sufficiently small minimum impulse bit for adequate control.
Hot-gas thrusters may not even be present, let alone the sole RCS mechanism. RCS thrust of record is ullage gas venting (cold has thrusters) which can be of incredibly low thrust and highly variable in thrust - depending only on upstream valving to modulate final valve inlet pressure and on final valve actuation speed.

My point is that cold gas simply isn't going to do it. 

Let's assume that we have 6bar methane at 250K with an Isp of 100s.  Assume a tank with 33t of LCH4 (i.e., 150t of methalox), leaving a volume of about 540m³ of ullage.  PV=nRT, so that's 2.5t of GCH4.  That's 9.1m/s of total delta-v.  Not exactly a generous maneuvering budget.

Now, you can heat the tank to produce more gas, but that seems silly.  Why spend mass on battery storage when you can produce energy by setting methalox on fire?

A much better solution is to use cold gas for extremely fine maneuvering, but use hot (ignited) gas for coarser maneuvers.

Bringing this back around to something close to on-topic, the point is that the RCS system for Starship is likely to be complicated.  Will it be complicated enough to require real prox ops tests with real spacecraft before the bugs are worked out, or can all the data be acquire in free-space tests on the way to EDL?  I don't know the answer to that.  My guess is that they'll probably need a couple of prox ops tests before they're willing to proceed to a real docking/refueling test.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/19/2022 07:32 pm
As a comparison, SpaceX flew a single Dragon 1 flight before engaging in ISS prox-ops, and Dragon 2 went straight to docking. Since Starship would not be rendezvousing with the ISS and potentially risking human life in case of a failure, and instead merely risking a second unmanned test vehicle - which would likely be obsoleted and scrapped on return anyway, at least for the first few flights - I suspect the only reason they would not test prox-ops on the first flight would be the need to then delay it and sit around and wait for a second vehicle to be ready.

But both D1 and D2 are pretty normal-sized and -shaped capsules, using hypergolics with decades of flight history on them.  Starship is not only huge but also long and skinny, and it can't use hypergolics.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/19/2022 11:00 pm
Your minimum thrust bit can be arbitrarily low...

Not if it's hot gas with an igniter, it can't.

Quote
...and your maximum RCS thrust level only places a limit on how long a given manoeuvre takes.  High-thrust RCS has been eliminated as a requirement for EDL, so RCS thrust can be very low indeed if desired.

The maximum must be high enough to produce a controlled system.  You have to be able to null out a rotation within some period of time to prevent collisions. 

Relying on Dragon as a good example of the kind of rate requirements you need, if you create a simpleminded model using a 12t D2 with an average moment of inertia of of 48,000 kg-m², that gives you an x-axis angular acceleration of 0.14º/s.  If you scale that to a 270t Starship with a CoM that's 20m from the bottom, to get a similar y- or z-axis acceleration requires a thruster (or thrusters) with 264kN of thrust.

Lots of caveats with that calculation, given that I guessed on the CoM, and maybe you can go somewhat lower on your angular acceleration.  But I doubt you can go much lower.  The r² term for the moment is not your friend.

Quote
The sole RPOD event that may require high thrust (rapid back-away) can be accomplished with the existing Starship dorsal tank dump vents.

Don't forget the lunar landing thrusters.

Quote
Quote
2) Nobody's explained how a hot (ignited) gas thruster has a sufficiently small minimum impulse bit for adequate control.
Hot-gas thrusters may not even be present, let alone the sole RCS mechanism. RCS thrust of record is ullage gas venting (cold has thrusters) which can be of incredibly low thrust and highly variable in thrust - depending only on upstream valving to modulate final valve inlet pressure and on final valve actuation speed.

My point is that cold gas simply isn't going to do it. 

Let's assume that we have 6bar methane at 250K with an Isp of 100s.  Assume a tank with 33t of LCH4 (i.e., 150t of methalox), leaving a volume of about 540m³ of ullage.  PV=nRT, so that's 2.5t of GCH4.  That's 9.1m/s of total delta-v.  Not exactly a generous maneuvering budget.

Now, you can heat the tank to produce more gas, but that seems silly.  Why spend mass on battery storage when you can produce energy by setting methalox on fire?

A much better solution is to use cold gas for extremely fine maneuvering, but use hot (ignited) gas for coarser maneuvers.

Bringing this back around to something close to on-topic, the point is that the RCS system for Starship is likely to be complicated.  Will it be complicated enough to require real prox ops tests with real spacecraft before the bugs are worked out, or can all the data be acquire in free-space tests on the way to EDL?  I don't know the answer to that.  My guess is that they'll probably need a couple of prox ops tests before they're willing to proceed to a real docking/refueling test.
I 99%+ in agreement. Some thoughts:

If we assume all pre contact maneuvering is done by the depot the tanks will start off at 6bar. This will go down with maneuvering unless there is a replenishment system. This could be done by vaporizing propellants in the COPV's prior to proximity ops. It still needs electrical power, just not all at once. Its reasonable to assume that it will be awhile before they can launch back to back so one ship will most likely have some conformal PV.

BTW, in a real world depot, there will be multiple proximity ops. After each one the ullage volume of the depot will be less, and the mass greater. Despite the glimmering of hope in the previous paragraph, the long term RCS problem will be much worse than than your estimates.

The only solution that I come up with is hot gas RCS, same as you. FWIW, I see no reason that igniter torches can't be left on during proximity ops for 'instant on' thrust. Just have to make sure they work in pairs to cancel torch thrust.

This is another system that forces the depot to be a very different and complex build. Yeah, do some strap on kludges for development but at some point it will have to refine and be its own unique thing.


Edit to add: on rereading the above, the increasing mass of the depot is a good argument for the other ship taking on some positioning duties. Not a slam dunk but worth thinking about.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/20/2022 04:19 am
The only solution that I come up with is hot gas RCS, same as you. FWIW, I see no reason that igniter torches can't be left on during proximity ops for 'instant on' thrust. Just have to make sure they work in pairs to cancel torch thrust.

I've been assuming that the amount of propellant required to get reasonably complete combustion is much larger for methox (or methalox) than it is for hypergols.  That in turn makes the minimum impulse bit larger for a methox thruster than for the hypergols.

Is that a good assumption?

Note also that an igniter torch is producing thrust.  Probably not very much, but we are talking about fine control here.

Quote
This is another system that forces the depot to be a very different and complex build. Yeah, do some strap on kludges for development but at some point it will have to refine and be its own unique thing.

Edit to add: on rereading the above, the increasing mass of the depot is a good argument for the other ship taking on some positioning duties. Not a slam dunk but worth thinking about.

I don't think the complexity of this system is high enough to warrant a completely different design for the depot vs. a tanker or payload Starship.¹  Ultimately, this is simply a spacecraft that has two different kinds of thrusters:  one (hot) for coarse maneuvers, sub-100m/s delta-v burns, and lunar landings, and a second kind for ullage and fine maneuvers.²  The plumbing is the plumbing, but I doubt that the complexity substantially increases risk, or adds much to the manufacturing cost.

But it does have to be figured out, tested, and debugged.  Like about a hundred other systems on Starship.

_____________
¹FWIW, I still think that almost everything you need to turn a vanilla Starship into a depot or a lift tanker can be provided by a payload kit and an appropriate door.  I still expect all Starship versions other than the LSS to have TPS and to return to EDL at the end of the mission.  I do expect the LCH4 dome and LOX/LCH4 bulkhead to be moved forward for depots and LSSes to handle 1500-1600t of prop.  One could also handle that prop extension via auxiliary tanks as part of the kit, but it seems easier just to juggle Starship cylinder segments and rejigger the plumbing and electrical runs.

²This is hardly novel.  The Shuttle had vernier RCS, primary RCS, and OMS.  D2 has Draco and SuperDraco.  Starliner has abort engines, OMACs, and RCS thrusters.  Starship would be unique in using hot gas methox or methalox for one system and cold gas for the other, but cold gas is considerably simpler.  Overall, the Starship thruster system would likely be of similar complexity to other spacecraft.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/20/2022 04:43 am
The only solution that I come up with is hot gas RCS, same as you. FWIW, I see no reason that igniter torches can't be left on during proximity ops for 'instant on' thrust. Just have to make sure they work in pairs to cancel torch thrust.
I've been assuming that the amount of propellant required to get reasonably complete combustion is much larger for methox (or methalox) than it is for hypergols.  That in turn makes the minimum impulse bit larger for a methox thruster than for the hypergols.
Is that a good assumption?

I don't think it is. Rockets scale down easier than they scale up. If you need less thrust, it's just a smaller engine. The smallest viable gas burning rocket engine is going to be well below anything useful for a vehicle the size of Starship. It's hard to imagine that "not small enough" is a limiting factor.

Note also that an igniter torch is producing thrust.  Probably not very much, but we are talking about fine control here.

Just have to make sure they work in pairs to cancel torch thrust.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 07/20/2022 09:40 am
Accordring to the latest info they are not doing bipropellant trhusters, they are going to use warm/hot ullage gas. This is even simpler than hypergolic thrusters.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/20/2022 10:31 am
Accordring to the latest info they are not doing bipropellant trhusters, they are going to use warm/hot ullage gas. This is even simpler than hypergolic thrusters.

The last time I looked into it, that claim seems to be the result of a single comment Musk made in an Tim Dodd interview, where he called something a "methane thruster" instead of methalox or (most likely) methanox. I've assumed it was just Elon-imprecision, not a strict description.

Has there been some other, more specific confirmation?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/20/2022 10:35 am
Your minimum thrust bit can be arbitrarily low...

Not if it's hot gas with an igniter, it can't.
The current number of hot-gas thrusters on Starship is zero.
Quote
You have to be able to null out a rotation within some period of time to prevent collisions. 
Rotation from what origin? Both mating bodies are cooperative, both are manoeuvrable, and both are running the same flight software. Both can agree to null all rotation prior to approach and to not approach until rotation rates are within acceptable limits. A worst-case stuck-on- thruster demands no more rotation acceleration requirements as can be provided by the same thruster.
This is unlike the ISS, where the ISS is somewhat of an uncooperative target (it does not manoeuvre under command of dragon, or indeed at all during final approach), and both vehicle have dissimilar propulsion capabilities (e.g. Dragon's back-away capability is sized conservatively for extreme events. In theory if Nauka had decided to start its dance during a vehicle approach that vehicle should have been capable of backing away to a safe distance).
Quote
Don't forget the lunar landing thrusters.
In addition to maybe not being installed at all, they are not used during docking (no good reason to).

Quote
My point is that cold gas simply isn't going to do it. 

Let's assume that we have 6bar methane at 250K with an Isp of 100s.  Assume a tank with 33t of LCH4 (i.e., 150t of methalox), leaving a volume of about 540m³ of ullage.  PV=nRT, so that's 2.5t of GCH4.  That's 9.1m/s of total delta-v.  Not exactly a generous maneuvering budget.

Now, you can heat the tank to produce more gas, but that seems silly.  Why spend mass on battery storage when you can produce energy by setting methalox on fire?
Why add an entire new propulsion system when you can use your post-coast ullage gas production system to produce more ullage gas? Sure hot-gas thrusters offer more thrust and ISP than cold-gas. But Starship is optimised for cheap and simple, not eking out every drop of performance from a limited mass budget. "Use more propellant, operate the thrusters for longer" is the simpler and cheaper option. There is no need for propellant transfer RPOD to be particularly fast.

Accordring to the latest info they are not doing bipropellant trhusters, they are going to use warm/hot ullage gas. This is even simpler than hypergolic thrusters.

The last time I looked into it, that claim seems to be the result of a single comment Musk made in an Tim Dodd interview, where he called something a "methane thruster" instead of methalox or (most likely) methanox. I've assumed it was just Elon-imprecision, not a strict description.

Has there been some other, more specific confirmation?
Yes, in the same Tim Dodd interview series Elon clarified he did indeed mean that hot-gas thrusters were being eliminated entirely from both vehicles, that RCS was going to be ullage-vent only, and that they also wanted to remove the landing thrusters from Lunar Starship if at all possible.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/20/2022 12:19 pm
https://twitter.com/davidnagysfgang/status/1548738746815840259

Quote
With Masten Space Systems unfortunately folding, MM1 will almost certainly not be moving ahead

It's been removed from the Artemis Launch Schedule. and become the first victim of what I call the "CLPS Bubble"
Artemis launch schedule lists tanker launch for Starship refuelling demo as the third flight. There has previously been a NASA/SpaceX contract for an internal propellant transfer demo (cryogenic prop movement between inside a vehicle) so it is unclear if this is that mission, or a full rendezvous and transfer demo.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/20/2022 07:52 pm

Note also that an igniter torch is producing thrust.  Probably not very much, but we are talking about fine control here.

With pairs of igniters facing in opposite directions you could use differential gas flow for vernier control.

Also swiveling a raptor is another control system.  How precisely they can control position and attitude at SECO will affect the maneuvering propellent budget.  If you can light a raptor in LEO you can probably target the south Pacific, if not the landing area, possibly a useful contingency if other attitude control fails.


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/20/2022 08:15 pm
Rotation from what origin? Both mating bodies are cooperative, both are manoeuvrable, and both are running the same flight software. Both can agree to null all rotation prior to approach and to not approach until rotation rates are within acceptable limits. A worst-case stuck-on- thruster demands no more rotation acceleration requirements as can be provided by the same thruster.
This is unlike the ISS, where the ISS is somewhat of an uncooperative target (it does not manoeuvre under command of dragon, or indeed at all during final approach), and both vehicle have dissimilar propulsion capabilities (e.g. Dragon's back-away capability is sized conservatively for extreme events. In theory if Nauka had decided to start its dance during a vehicle approach that vehicle should have been capable of backing away to a safe distance).

Are you saying that you could move from prox ops to docking with inadequate command authority to null out unexpected rotations?  That sounds crazy to me.  If you have sufficient back-away authority, maybe that's OK, but an unanticipated rotation of either Starship requires the other one to get out the way, since they almost never share an axis of rotation.
 
Quote
Quote
Don't forget the lunar landing thrusters.
In addition to maybe not being installed at all, they are not used during docking (no good reason to).

If you rummage up-thread, you'll find that the thrust requirements for the landing thrusters are very close to those for producing the same angular acceleration on Starship as on the D2.  They may choose not to install them in the waist positions with the cosine losses, but one hot gas thruster size will work for coarse attitude control, modest delta-v maneuvers, and landing.

Quote
Why add an entire new propulsion system when you can use your post-coast ullage gas production system to produce more ullage gas? Sure hot-gas thrusters offer more thrust and ISP than cold-gas. But Starship is optimised for cheap and simple, not eking out every drop of performance from a limited mass budget. "Use more propellant, operate the thrusters for longer" is the simpler and cheaper option. There is no need for propellant transfer RPOD to be particularly fast.

You really need to do the math on this.  It's always possible (maybe even likely) that I've messed something up, but there simply isn't enough impulse available from ullage gas--especially if you're feeding a nozzle at considerably less than 6bar.

I agree that Elon said to Tim Dodd what Elon said.  I still don't understand it.  I can't make the math work out.  The only thing I can think of to reconcile the statement and the math is that he meant that the first launch would use only cold gas thrusters, which probably would work, since there will be no RPOD to deal with.  There's some evidence here (https://twitter.com/elonmusk/status/1407969457411067905?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1407969457411067905%7Ctwgr%5E%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Fwww.teslarati.com%2Fspacex-starship-hot-gas-thruster-photos%2F) to back this up.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/20/2022 11:20 pm
The only solution that I come up with is hot gas RCS, same as you. FWIW, I see no reason that igniter torches can't be left on during proximity ops for 'instant on' thrust. Just have to make sure they work in pairs to cancel torch thrust.

I've been assuming that the amount of propellant required to get reasonably complete combustion is much larger for methox (or methalox) than it is for hypergols.  That in turn makes the minimum impulse bit larger for a methox thruster than for the hypergols.

Is that a good assumption?

Note also that an igniter torch is producing thrust.  Probably not very much, but we are talking about fine control here.

Quote
This is another system that forces the depot to be a very different and complex build. Yeah, do some strap on kludges for development but at some point it will have to refine and be its own unique thing.

Edit to add: on rereading the above, the increasing mass of the depot is a good argument for the other ship taking on some positioning duties. Not a slam dunk but worth thinking about.

I don't think the complexity of this system is high enough to warrant a completely different design for the depot vs. a tanker or payload Starship.¹  Ultimately, this is simply a spacecraft that has two different kinds of thrusters:  one (hot) for coarse maneuvers, sub-100m/s delta-v burns, and lunar landings, and a second kind for ullage and fine maneuvers.²  The plumbing is the plumbing, but I doubt that the complexity substantially increases risk, or adds much to the manufacturing cost.

But it does have to be figured out, tested, and debugged.  Like about a hundred other systems on Starship.

_____________
¹FWIW, I still think that almost everything you need to turn a vanilla Starship into a depot or a lift tanker can be provided by a payload kit and an appropriate door.  I still expect all Starship versions other than the LSS to have TPS and to return to EDL at the end of the mission.  I do expect the LCH4 dome and LOX/LCH4 bulkhead to be moved forward for depots and LSSes to handle 1500-1600t of prop.  One could also handle that prop extension via auxiliary tanks as part of the kit, but it seems easier just to juggle Starship cylinder segments and rejigger the plumbing and electrical runs.

²This is hardly novel.  The Shuttle had vernier RCS, primary RCS, and OMS.  D2 has Draco and SuperDraco.  Starliner has abort engines, OMACs, and RCS thrusters.  Starship would be unique in using hot gas methox or methalox for one system and cold gas for the other, but cold gas is considerably simpler.  Overall, the Starship thruster system would likely be of similar complexity to other spacecraft.
Depot return is where we differ. Yes, the basic structure for the depot and the tanker will most probably be the same. Sort of like a 707 and a KC-135. That analogy shouldn't be taken too far.


A depot needs either a QD gender bender or there needs to be a second QD. Either way, it's a heavy assembly. Your idea of a catawompus alignment with a gender bent QD extending out from the truncated cargo bay is innovative but adds complexity and mass. Also, If the fins stay on, the space in the ogive may be less than you expect. Those fins need a robust support structure. If anything else needs to deploy it's gonna get tight up there and deployment mechanisms are may end up way too complex short of a full fledged Canadarm.


Another issue is that ullage settling has to orient to the visiting tanker. It has to accelerate in the usual direction to keep propellant down around the QD. The depot would be opposite. Everything coming into it will have to flow 'up hill'. The flow will either be all the way up to where GSE normally enters the respective tanks, or shorter and entering the 'top' of the respective tanks. If the first, the propellant will fall the length of the respective tanks to splash into what would normally be the top. That would give the RCS an 'interesting' workout. Add in the mass of the extra plumbing to reach to the true GSE plumbing, and expect to loose any storage for deployable bits that might be stowed in the side nacelles. If the second, another hole must pierce the tanks near their tops, thus making at least one significant change from a generic tanker.


Catawompus can be made to work but ISTM every bit of it either needs a workaround or makes another piece of the puzzle harder. Deploying s gender bender to the original QD plate has the one difficulty of deployment, which IMO, would not be all that difficult.


A bare naked depot will have boil off. Slap some Solar white paint on it and life gets easier but don't expect the paint to survive EDL. If that proves inadequate (unproven assumption) a sun screen or active cooling will be needed. You'll most likely get a shade to deploy. You'll most likely get PV and a radiator deployed to support active cooling. I seriously doubt any of that can undeploy. It would have to jettison for EDL. So much for full reusability.


It would be easier for me to accept a depot to be a one use throwaway than to be landed and reused.


The cape is ~28.5 deg. A depot at this inclination is not ideal for any specific mission but it can be made to do for many of them. A family of depots at this inclination in parallel orbits will work for more. Most likely very few missions needing a depot will max out on mass. People and institutions just don't think this way yet. A payload mass limit to allow refueling in a sub optimal plane sounds like a solid SX 'good enough'. Yeah, sorry. We can only do 40 tons for your Jupiter probe. :o  LOL.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: robot_enthusiast on 07/21/2022 01:09 am
Accordring to the latest info they are not doing bipropellant trhusters, they are going to use warm/hot ullage gas. This is even simpler than hypergolic thrusters.

The last time I looked into it, that claim seems to be the result of a single comment Musk made in an Tim Dodd interview, where he called something a "methane thruster" instead of methalox or (most likely) methanox. I've assumed it was just Elon-imprecision, not a strict description.

Has there been some other, more specific confirmation?
Musk talked at length about it in this interview with Tim, starting at around 3 minutes. There is no misinterpretation.
https://youtu.be/3Ux6B3bvO0w
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/21/2022 05:41 am

Are you saying that you could move from prox ops to docking with inadequate command authority to null out unexpected rotations?  That sounds crazy to me.  If you have sufficient back-away authority, maybe that's OK, but an unanticipated rotation of either Starship requires the other one to get out the way, since they almost never share an axis of rotation.
 
Or you test a lot, make unanticipated rotations rare, and accept that once in a great while things may go boom.

Because even if you don't accept that once in a great while things will go boom, they will still go boom.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/21/2022 12:31 pm
Rotation from what origin? Both mating bodies are cooperative, both are manoeuvrable, and both are running the same flight software. Both can agree to null all rotation prior to approach and to not approach until rotation rates are within acceptable limits. A worst-case stuck-on- thruster demands no more rotation acceleration requirements as can be provided by the same thruster.
This is unlike the ISS, where the ISS is somewhat of an uncooperative target (it does not manoeuvre under command of dragon, or indeed at all during final approach), and both vehicle have dissimilar propulsion capabilities (e.g. Dragon's back-away capability is sized conservatively for extreme events. In theory if Nauka had decided to start its dance during a vehicle approach that vehicle should have been capable of backing away to a safe distance).

Are you saying that you could move from prox ops to docking with inadequate command authority to null out unexpected rotations?  That sounds crazy to me.  If you have sufficient back-away authority, maybe that's OK, but an unanticipated rotation of either Starship requires the other one to get out the way, since they almost never share an axis of rotation.
Unanticipated rotations from what source? Unless spontaneous violations of the conservation of momentum occur, that leaves the thrusters themselves (which by definition have the thrust as the thrusters you would be countering with), or catastrophic tank bursts. It's possible there could be a requirement for a rapid backaway in that event, but "don't burst the tanks" seems a better option.
Quote
You really need to do the math on this.  It's always possible (maybe even likely) that I've messed something up, but there simply isn't enough impulse available from ullage gas--especially if you're feeding a nozzle at considerably less than 6bar.
With ullage gas production from propellant heating (already necessary for repressurisation) your are limited by raw propellant volume (lots), energy available within the batteries for reheat (from initial charge level to depleted), or energy/time for battery recharge (after depletion of starting charge). Total delta-V for RPOD will be dominated by velocity mismatch between vehicles rather than actual approach and proxops (e.g. if you start from matched velocity 1km apart, even a mere 1m/s approach rate for 2m/s delta-V only has  penalty of a ~15m wait) leaving 7m/s delta-V from ullage gas even if you never use the batteries at all.
For gross velocity nulling prior to approach, the Raptors are pretty capable: Starship has a dry mas on the order of 100 tonnes, and a payload on the order of 100 tonnes, so a worst-case (a depot with more full tanks will mass even more) mass is around 200 tonnes. Raptor 2 has ~230 tonnes-force (~2.3MN) thrust, so at 40% throttle (~1MN) a single raptor on a Tanker is imparting ~5m/s per second of operation.
We've seen Raptors start up and abort very rapidly, so its not unreasonable to assume the minimum impulse bit is even lower than that, but even at 5m/s that's within the capacity of the ullage alone to null velocity difference and perform RPOD.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/22/2022 01:18 am
Got a question. When the GSE QD plate for the SS withdraws, does it pull back linear or does it follow an arc? Anybody catch this?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/22/2022 02:15 am
Got a question. When the GSE QD plate for the SS withdraws, does it pull back linear or does it follow an arc? Anybody catch this?
No arc. Straight back.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/22/2022 11:15 am
Main extension ram is angled slightly downwards (so single actuator motion would be 'out and down'), but the arm has 5DoF so composite retraction motion could be in any direction other than roll of the plate about its short axis. Straight out would be the obvious option: no need to wait for t0 for retract, as the plate is designed to reconnect without issue so can be disconnected well before t0.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Hog on 07/22/2022 11:21 am
Main extension ram is angled slightly downwards (so single actuator motion would be 'out and down'), but the arm has 5DoF so composite retraction motion could be in any direction other than roll of the plate about its short axis. Straight out would be the obvious option: no need to wait for t0 for retract, as the plate is designed to reconnect without issue so can be disconnected well before t0.
5DoF=?  I'm going to guess, 5 axis of freedom?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/22/2022 11:25 am
Main extension ram is angled slightly downwards (so single actuator motion would be 'out and down'), but the arm has 5DoF so composite retraction motion could be in any direction other than roll of the plate about its short axis. Straight out would be the obvious option: no need to wait for t0 for retract, as the plate is designed to reconnect without issue so can be disconnected well before t0.
5DoF=?  I'm going to guess, 5 axis of freedom?
Yes.
- Extend-retract (single long actuator visible)
- Translate up/down (two 'A frame' actuators supporting the joint at the base of the plate)
- Translate left/right (same two 'A frame' actuators supporting the joint at the base of the plate)
- Pitch/tilt (actuators on rear of QD plate bearing against joint)
- Yaw/swing (actuators on rear of QD plate bearing against joint)
Roll is the absent axis from the actuators and joints visible.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/22/2022 04:42 pm
Got a question. When the GSE QD plate for the SS withdraws, does it pull back linear or does it follow an arc? Anybody catch this?
No arc. Straight back.
Reason I was asking is that somebody (wish I could remember who) suggested upthread that a depot could be built with a gender bent QD plate and a gender unbender adapter added to the GSM QD plate for Depot launches. If it were withdrawing through an arc this would be hard. Since it's withdrawing linear, the only other modification needed would be to adapt the linkages that bring the cover down.


ISTM that this would be the most trouble free and low mass approach to gender bending. Has a certain elegance too.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 07/22/2022 11:03 pm
Got a question. When the GSE QD plate for the SS withdraws, does it pull back linear or does it follow an arc? Anybody catch this?
No arc. Straight back.
Reason I was asking is that somebody (wish I could remember who) suggested upthread that a depot could be built with a gender bent QD plate and a gender unbender adapter added to the GSM QD plate for Depot launches. If it were withdrawing through an arc this would be hard. Since it's withdrawing linear, the only other modification needed would be to adapt the linkages that bring the cover down.


ISTM that this would be the most trouble free and low mass approach to gender bending. Has a certain elegance too.

There's no cover for the SS quick disconnect.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/23/2022 06:20 pm
Got a question. When the GSE QD plate for the SS withdraws, does it pull back linear or does it follow an arc? Anybody catch this?
No arc. Straight back.
Reason I was asking is that somebody (wish I could remember who) suggested upthread that a depot could be built with a gender bent QD plate and a gender unbender adapter added to the GSM QD plate for Depot launches. If it were withdrawing through an arc this would be hard. Since it's withdrawing linear, the only other modification needed would be to adapt the linkages that bring the cover down.


ISTM that this would be the most trouble free and low mass approach to gender bending. Has a certain elegance too.

There's no cover for the SS quick disconnect.
I meant the GSE QD on the tower. It's got a clamshell that covers it as it retracts.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: daavery on 07/23/2022 11:36 pm
the "GSE" (Booster QD) is just for the booster. the SQD ( starship QD ) on the mid tower QD arm is for Starship and has no covers. it is protected by the entire arm swinging 90 degrees away before launch
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: russianhalo117 on 07/24/2022 12:07 am
Got a question. When the GSE QD plate for the SS withdraws, does it pull back linear or does it follow an arc? Anybody catch this?
No arc. Straight back.
Reason I was asking is that somebody (wish I could remember who) suggested upthread that a depot could be built with a gender bent QD plate and a gender unbender adapter added to the GSM QD plate for Depot launches. If it were withdrawing through an arc this would be hard. Since it's withdrawing linear, the only other modification needed would be to adapt the linkages that bring the cover down.


ISTM that this would be the most trouble free and low mass approach to gender bending. Has a certain elegance too.
Not presently possible without OLM Booster QD modifications as the QD is fully mechanically connected to the QD blast protection housing cover by arms and rods which dictate the precise alignment and connection with the booster FSE plate when extended. There is not room not the buolt in adjustment capability to add in such an adapter plate. It is not designed at all for that. The ship QD is also not meant for this and the extra drag and bump out of such an adapter. I would expect in a future Starship block the addition of androgynous or custom docking and fuel transfer hardware. I do not expect the current interface to be used onorbit in its current iteration.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/24/2022 12:59 am
the "GSE" (Booster QD) is just for the booster. the SQD ( starship QD ) on the mid tower QD arm is for Starship and has no covers. it is protected by the entire arm swinging 90 degrees away before launch
Hmmmm. Your right. I had the booster QD in my minds eye.


Does it retract a bit before swinging away? Seem like it should if only to completely disengage from the fluid connections and latches. Even if it doesn't retract it seems a relatively minor mod to give it some travel to make room for an adapter to a gender bent depot QD.


I doubt the QD design is fixed in stone yet. Its never even been used and it'll be awhile before they need to do even a PoC fluid transfer between ships.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: russianhalo117 on 07/24/2022 03:56 am
the "GSE" (Booster QD) is just for the booster. the SQD ( starship QD ) on the mid tower QD arm is for Starship and has no covers. it is protected by the entire arm swinging 90 degrees away before launch
Hmmmm. Your right. I had the booster QD in my minds eye.


Does it retract a bit before swinging away? Seem like it should if only to completely disengage from the fluid connections and latches. Even if it doesn't retract it seems a relatively minor mod to give it some travel to make room for an adapter to a gender bent depot QD.


I doubt the QD design is fixed in stone yet. Its never even been used and it'll be awhile before they need to do even a PoC fluid transfer between ships.
Yes the three hydraulic pistons retract from their fully extended positions to pull the GSE plate side of the Ship QD in and down tight against the top of the arm before/as the arm swings (they have performed a test of each method).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/24/2022 08:18 pm
Instead of using ullage burns to settle the propellants, why not just rotate the depot on its long axis? I'm sure this has been discussed somewhere in this long thread, but I hunted for it and didn't find it.

Just for fun, I played with some of the parameters, assuming the depot rotated at 1 rpm, and using SpinCalc (http://www.artificial-gravity.com/sw/SpinCalc/).

If the tanks in the depot are 4.5 m in radius, then the gravity at the rim will be 5 milli-g, and it'll be above 1 milli-g from 0.9 m from the axis. That means 96% of the volume of the tank will have gravity of 1 milli-g or more.

Yes, you'll need to dock with a spinning object, but it's not spinning very fast, so maybe that's easier than trying to accelerate two objects of different (and changing) mass. It may be more of a problem keeping it stable, since things don't like to spin on their long axis, but I think there are various ways around that. And, of course, you'd need to fill it and drain it from ports on the sides, but that doesn't sound all that hard.

I'm sure someone, somewhere has analyzed this in detail, but I haven't found a paper on it. I gather this has already been discussed and discarded, but I'd be interested to know why.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: frith01 on 07/24/2022 10:55 pm
If you spin the rocket, instead of propulsion in one direction, you have several issues.
1.  1/3 of your propellant will be heading in the wrong direction (top of tank),  1/3 of your propellant wont be going anywhere (middle)
2.  difficult to ensure the spin planes are co-planar. 
3.  Most importantly, rotation will act as a gyroscope, so will require more energy to re-direct the vehicle, and avoid other known space debris or continue on toward destination.  (remember, re-fueling is on the way to somewhere.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/25/2022 01:38 am
If you spin the rocket, instead of propulsion in one direction, you have several issues.
1.  1/3 of your propellant will be heading in the wrong direction (top of tank),  1/3 of your propellant wont be going anywhere (middle)
2.  difficult to ensure the spin planes are co-planar. 
3.  Most importantly, rotation will act as a gyroscope, so will require more energy to re-direct the vehicle, and avoid other known space debris or continue on toward destination.  (remember, re-fueling is on the way to somewhere.)
I don't understand your point #1 at all, I'm afraid. I think all of the propellant will be on the edges of the tank. That is, on the "bottom" as defined by the rotation.

I think your point #2 means that you think docking would be hard, despite how easy it looked in 2001, the movie. I'm not sure that's actually true, but I'd be interested to hear why.

As for point #3, because the rotation rate is relatively slow, the angular momentum shouldn't be very large. Also, distinguishing the spaceship from the depot, I would assume that the spaceship would initiate a roll before docking, dock and refuel, undock and spin down, and then head off for wherever its destination is.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: daavery on 07/25/2022 01:51 am
spinning the vehicle on it's own axis does little to the contents of the tanks and even if it did forcing fluid to the wall of the tank is the least efficient path to being able to collect and transfer it. whereas a slight acceleration collects the liquid at the end of the tank where it is east to collect and move
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/25/2022 05:43 am
Also:

Spin around the long axis (low moment of inertia axis) is secularly unstable and will in time become chaotic, eventually ending in a flat spin.  This doesn't say how fast the transition will occur and it is possible to stabilize almost anything if you add enough control authority (i.e. thrusters) so it's not hopeless.  But large masses of fluid that do not move as a rigid body tend to increase the instability and speed of transition.  Sloshing, eddies etc. amplify asymmetries and dissipate energy both of which increase instability about the low inertia axis.

At the very least this mode would require some detailed and difficult modelling including the fluid dynamics.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 07/25/2022 06:11 am
Why did SpaceX do away with the tail to tail fueling?  A little thrust going toward the tanker would make the tanker empty into A Starship.  This might still be done but on the side, Starship facing one way tanker the other but docked and emptying tanker.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Llian Rhydderch on 07/25/2022 11:17 am
Why did SpaceX do away with the tail to tail fueling?  A little thrust going toward the tanker would make the tanker empty into A Starship.  This might still be done but on the side, Starship facing one way tanker the other but docked and emptying tanker.

SpaceX never had tail to tail fueling. That was merely a concept for how it might be done that was whipped into some pretty graphics Musk showed at the IAC as he was announcing the overall conceptual design for this very large launch vehicle 5 or 6 years ago.

SpaceX is strongly committed to a process of iterative and incremental development, relentless work on improving the production process (so they build a lot of test articles), the overall meta objective of building a Mars-capable super heavy lift rocket that could start the process of making humans multiplanetary, and economically staying afloat during the capital-intensive multibillion dollar development-only phase.

Thus, they are concentrating on the big objectives of landing the vehicles, making reuse possible, and getting to orbit with the largest rocket ever built.  When they can do that, we'll see the details of the vehicle-to-vehicle propellant transfer begin detailed design around one of the several plausible concepts for how it might be done.  NSFers are debating various concepts for what might be most plausible here in this thread.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/25/2022 12:31 pm
Why did SpaceX do away with the tail to tail fueling?  A little thrust going toward the tanker would make the tanker empty into A Starship.  This might still be done but on the side, Starship facing one way tanker the other but docked and emptying tanker.

SpaceX never had tail to tail fueling.
All post-hopper test flights (SN5, SN6, SN8, SN9, SN10, SN11, SN15) flew from launch mounts with tail propellant loading.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/25/2022 01:33 pm
Also:

Spin around the long axis (low moment of inertia axis) is secularly unstable and will in time become chaotic, eventually ending in a flat spin.  This doesn't say how fast the transition will occur and it is possible to stabilize almost anything if you add enough control authority (i.e. thrusters) so it's not hopeless.  But large masses of fluid that do not move as a rigid body tend to increase the instability and speed of transition.  Sloshing, eddies etc. amplify asymmetries and dissipate energy both of which increase instability about the low inertia axis.

At the very least this mode would require some detailed and difficult modelling including the fluid dynamics.
Yes, I knew about this issue, but there seem to be various ways around it, such as pumping fluid around on purpose, and adding things to change the moment of inertia (e.g. weights on long booms attached to the middle). I agree there'd be some difficult modelling, but I think that's going to be true of any solution. The devil is always in the details. The question is which devil is the easiest to deal with.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/25/2022 02:15 pm
Also:

Spin around the long axis (low moment of inertia axis) is secularly unstable and will in time become chaotic, eventually ending in a flat spin.  This doesn't say how fast the transition will occur and it is possible to stabilize almost anything if you add enough control authority (i.e. thrusters) so it's not hopeless.  But large masses of fluid that do not move as a rigid body tend to increase the instability and speed of transition.  Sloshing, eddies etc. amplify asymmetries and dissipate energy both of which increase instability about the low inertia axis.

At the very least this mode would require some detailed and difficult modelling including the fluid dynamics.
Yes, I knew about this issue, but there seem to be various ways around it, such as pumping fluid around on purpose, and adding things to change the moment of inertia (e.g. weights on long booms attached to the middle). I agree there'd be some difficult modelling, but I think that's going to be true of any solution. The devil is always in the details. The question is which devil is the easiest to deal with.
Pumping propellants from the base of the tank means draining from a single point where there already happens to be a fill-drain line.
Pumping from the tank walls (where the propellant will be flung if strongly coupled to the tank wall rotation) means adding a new series of annular drain ports around the entire tank circumference, and across the entire length of the tank (because propellant swill not pool to a single location). More plumbing, additional plumbing, and awkward plumbing. And if the fluid is not well coupled to the tank walls (e.g. floating in a blob in the middle of the tank) then you can spin all you want and it will not move its way to the tank walls.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/25/2022 03:11 pm
Pumping from the tank walls (where the propellant will be flung if strongly coupled to the tank wall rotation) means adding a new series of annular drain ports around the entire tank circumference, and across the entire length of the tank (because propellant swill not pool to a single location). More plumbing, additional plumbing, and awkward plumbing. And if the fluid is not well coupled to the tank walls (e.g. floating in a blob in the middle of the tank) then you can spin all you want and it will not move its way to the tank walls.
I hadn't thought about the problem that propellant might just sit in the middle and not rotate. It would seem that that wouldn't be an insuperable problem, though. Naively, I'd think something as simple as a few wires across the inside of the tank would be enough to break that up, but I'm sure there are better ways to tackle it.

You might be able to get away with drain ports at the edges of the end of the cylinder, since those are on the bottom both during rotation and during acceleration. Not sure how much trouble that would be in practice, though. As far as worrying that it won't all pool to a single location, I think that only matters if you are trying to run the thing dry. For the fuel depot, at least, I think it's probably okay if the last 1% to 5% is hard to get out.

On the question of plumbing, I seem to remember that it's already necessary to "stir" the tanks from time to time. If, hypothetically speaking, the instability problem could be dealt with by something as simple as pumping fluid from the ends of the tank to the middle, that might simultaneously handle the need for stirring. After all, the instability is caused by the fact that there's lower kinetic energy for the same angular momentum if the fluid flows to the ends of the tank, so pumping some of it back towards the middle seems like the first thing to try.

Anyway, the hope would be that you'd end up with different plumbing, but not necessarily a whole lot of extra plumbing.

Again, it would seem that someone has surely already tried to model this, but I've not been able to find anything online.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 07/25/2022 03:19 pm
Well as long as we are spinning in this topic...

Tether 2 starships from the nose and spin'er up.
Tether has 2 fuel lines to transfer in and out.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 07/25/2022 03:44 pm
Pumping from the tank walls (where the propellant will be flung if strongly coupled to the tank wall rotation) means adding a new series of annular drain ports around the entire tank circumference, and across the entire length of the tank (because propellant swill not pool to a single location). More plumbing, additional plumbing, and awkward plumbing. And if the fluid is not well coupled to the tank walls (e.g. floating in a blob in the middle of the tank) then you can spin all you want and it will not move its way to the tank walls.
I hadn't thought about the problem that propellant might just sit in the middle and not rotate. It would seem that that wouldn't be an insuperable problem, though. Naively, I'd think something as simple as a few wires across the inside of the tank would be enough to break that up, but I'm sure there are better ways to tackle it.
:) The solution to the problem is baffling :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 07/25/2022 03:55 pm
Pumping from the tank walls (where the propellant will be flung if strongly coupled to the tank wall rotation) means adding a new series of annular drain ports around the entire tank circumference, and across the entire length of the tank (because propellant swill not pool to a single location). More plumbing, additional plumbing, and awkward plumbing. And if the fluid is not well coupled to the tank walls (e.g. floating in a blob in the middle of the tank) then you can spin all you want and it will not move its way to the tank walls.
I hadn't thought about the problem that propellant might just sit in the middle and not rotate. It would seem that that wouldn't be an insuperable problem, though. Naively, I'd think something as simple as a few wires across the inside of the tank would be enough to break that up, but I'm sure there are better ways to tackle it.

You might be able to get away with drain ports at the edges of the end of the cylinder, since those are on the bottom both during rotation and during acceleration. Not sure how much trouble that would be in practice, though. As far as worrying that it won't all pool to a single location, I think that only matters if you are trying to run the thing dry. For the fuel depot, at least, I think it's probably okay if the last 1% to 5% is hard to get out.

On the question of plumbing, I seem to remember that it's already necessary to "stir" the tanks from time to time. If, hypothetically speaking, the instability problem could be dealt with by something as simple as pumping fluid from the ends of the tank to the middle, that might simultaneously handle the need for stirring. After all, the instability is caused by the fact that there's lower kinetic energy for the same angular momentum if the fluid flows to the ends of the tank, so pumping some of it back towards the middle seems like the first thing to try.

Anyway, the hope would be that you'd end up with different plumbing, but not necessarily a whole lot of extra plumbing.

Again, it would seem that someone has surely already tried to model this, but I've not been able to find anything online.
There’s ullage gas which would transfer rotation forces to the center, and propellant sitting exactly in the center (and how would propellant get precisely in the center with exactly zero relative velocity anyway?) would be an unstable equilibrium, and rotation being surface tension would quickly fling propellant to the sides. So wires are not required.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 07/25/2022 04:08 pm
If you want to use spin I'd look at end over end.  This is stable.  If you're lucky (or good) and can get the spin axis between the two tanks (possibly by manipulating the load in the header tanks) the walls of the tanks will sweep up all the fluids which will collect at opposite ends in inverted dome shaped sumps.  At most you'd need to add an additional drain at the top of the upper tank.

Once the fluid is gathered in the donor ship the center of gravity (and rotation) can move somewhat.  As long as it stays between the liquid surfaces in the two tanks the liquids may remain under control.
 

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/25/2022 05:57 pm
If you spin the rocket, instead of propulsion in one direction, you have several issues.
1.  1/3 of your propellant will be heading in the wrong direction (top of tank),  1/3 of your propellant wont be going anywhere (middle)
2.  difficult to ensure the spin planes are co-planar. 
3.  Most importantly, rotation will act as a gyroscope, so will require more energy to re-direct the vehicle, and avoid other known space debris or continue on toward destination.  (remember, re-fueling is on the way to somewhere.)
I don't understand your point #1 at all, I'm afraid. I think all of the propellant will be on the edges of the tank. That is, on the "bottom" as defined by the rotation.

I think your point #2 means that you think docking would be hard, despite how easy it looked in 2001, the movie. I'm not sure that's actually true, but I'd be interested to hear why.

As for point #3, because the rotation rate is relatively slow, the angular momentum shouldn't be very large. Also, distinguishing the spaceship from the depot, I would assume that the spaceship would initiate a roll before docking, dock and refuel, undock and spin down, and then head off for wherever its destination is.
Frith01 misunderstood the spin axis in your concept.


If rotation around a common center of gravity is initiated after docking, the rotating target problem goes away. It's not impossible that letting the two ships 'fling' apart after disconnect could be put to a useful orbital input. As for 2001, Kubric and Clark had the docking ship aligned with and sharing the stations axis of rotation. A much easier problem, not to mention great visuals.


The plumbing problem is tough. It would add mass and complexity. If the ships are dorsal to dorsal the tapoff would have to penetrate the heat shield of the tanker and the depot too if it has one. If ventral to ventral, the heat shield is ok. If side to side, the heat shield problem is back.


All three approaches require plumbing, most likely external, to reach from the side opposite the rotational axis of the delivering ship and around its bulk to the receiving ship. It would probably also have to span around the receiving ship if a goal is to have only one connection point no matter what the ships role.


Most of the plumbing issues brought up elsewhere are surmountable at some unknown and possibly acceptable tradeoff cost. My gut sez that the problem outlined above would make it unworkable.


I've not yet read all the comments on this idea so some of this may already be Ninja'd.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/25/2022 06:03 pm
Well as long as we are spinning in this topic...

Tether 2 starships from the nose and spin'er up.
Tether has 2 fuel lines to transfer in and out.
Tethers are hard and IIUC, experiments attempting this are mixed. It'd be another new technology to learn.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/25/2022 11:48 pm
As for 2001, Kubric and Clark had the docking ship aligned with and sharing the stations axis of rotation. A much easier problem, not to mention great visuals.
That's pretty much what I'm visualizing, although (obviously) the depot is a lot smaller than the space station in the movie. You dock the nose of the starship with some sort of adaptor that facilitates leaving or receiving fuel. This probably works a lot better with a custom-designed depot rather than just repurposing a Starship, but you've still got to somehow make it work for the tanker vehicles and the ultimate "customer" vehicles.

On the ground, I assume fueling is simply done through the tops of the tanks--just like filling up a car. I can see why people would be unenthusiastic about trying to do that from the side or the bottom. Although you certainly won't try to get the fuel out the same way, so it seems that any refueling plan is going to involve extra plumbing--for the tanker vehicles and the depot, anyway.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 07/26/2022 12:41 am
For cryo fluids you do not fill from above but from below to minimize agitation and sometimes even more dangerous actions by the fluid caused by stimulating it beyond safe levels. One of the items that is always dreaded is when cryo fluid contacts warm surfaces and explosively from the standpoint of a pressure event boils violently. This risk of this type of event mitigation is done by filling from the bottom. Such an uncontrolled event can cause the tank to split and a complete loss of vehicle. When dealing with cryo it is safest to go slow and as least amount of agitation as possible.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: frith01 on 07/26/2022 03:57 am
When talking about spinning starship in orbit, please review this video about "Bizarre Behavior of Rotating Bodies"  which leads to the conclusion that the only stable orientation with spinning is the heavy ends out (engine to nose, end over end)

Jump to 11:20 to see the effect of liquid filled cylinder spinning in space station.

https://www.youtube.com/watch?v=1VPfZ_XzisU (https://www.youtube.com/watch?v=1VPfZ_XzisU)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 07/26/2022 12:16 pm
So this got me wondering about future refueling stations.

A hub with starships nose into the hub. I visualize six starships docking to the hub by the nose. Then fuel lines extending(robotically) to the fill/drain ports near the engine section. Could we also have manned starships dock this way? Robotically extended tunnels for people to move to the hub and the lounge. 2001 images float in my head.

Question:
If the hub is slowly rotating to provide settling g could a starship maneuver with RCS to dock? 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/26/2022 03:42 pm
As for 2001, Kubric and Clark had the docking ship aligned with and sharing the stations axis of rotation. A much easier problem, not to mention great visuals.
That's pretty much what I'm visualizing, although (obviously) the depot is a lot smaller than the space station in the movie. You dock the nose of the starship with some sort of adaptor that facilitates leaving or receiving fuel. This probably works a lot better with a custom-designed depot rather than just repurposing a Starship, but you've still got to somehow make it work for the tanker vehicles and the ultimate "customer" vehicles.

On the ground, I assume fueling is simply done through the tops of the tanks--just like filling up a car. I can see why people would be unenthusiastic about trying to do that from the side or the bottom. Although you certainly won't try to get the fuel out the same way, so it seems that any refueling plan is going to involve extra plumbing--for the tanker vehicles and the depot, anyway.
Ahhh. I stand corrected. Was picturing side by side.


I could be wrong but I think GSE fills the tanks from the bottom. No splash and the pressure head is the same if you're stuffing fluid into the bottom of a nearly full tank or stuffing it up a pipe to dump it in at the top.


One reason I favor QD to QD transfer is that it adds exactly zero additional plumbing except for a gender bender. Give the depot a mirror image of a standard QD, and use a removable adapter on the GSE that services it and even this minor complexity stays on the ground.


There was a time that 36 and 48 inch pipe wrenches were my companions. Plumbing is so conceptually simple and so very deceptive. It has soooo many failure modes, all wanting to jump out at you. Worse yet, sometimes they want to remain hidden and bedevil you. Anything that minimizes plumbing has my vote.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/26/2022 04:51 pm

Ahhh. I stand corrected. Was picturing side by side.


I could be wrong but I think GSE fills the tanks from the bottom. No splash and the pressure head is the same if you're stuffing fluid into the bottom of a nearly full tank or stuffing it up a pipe to dump it in at the top.


One reason I favor QD to QD transfer is that it adds exactly zero additional plumbing except for a gender bender. Give the depot a mirror image of a standard QD, and use a removable adapter on the GSE that services it and even this minor complexity stays on the ground.


There was a time that 36 and 48 inch pipe wrenches were my companions. Plumbing is so conceptually simple and so very deceptive. It has soooo many failure modes, all wanting to jump out at you. Worse yet, sometimes they want to remain hidden and bedevil you. Anything that minimizes plumbing has my vote.
Given those constraints, I see two possibilities:
1) Ullage burn with the two vehicles side-by-side and a short connecter near the bottom of each one. This is far-and-away the easiest, assuming you can come up with some way to automatically make that connection. (This is not just plumbing--it's plumbing that moves!) This all seems to be discussed in Settled Cryogenic Propellant Transfer (https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf), which I just now discovered and read.

2) Nose-to-nose docking with a temporary spin and a really long connector from bottom to bottom. For this I'm visualizing a docking adaptor with really long connector pipes that are external to both starships. This is clearly more plumbing than option #1, but I think it's the minimum given that you wanted to a) rotate and b) refuel/unload from the bottom of both vehicles. Note that in this scenario, if it's true that you only need about 1 milli-g at the bottom of the tanks, the rotation rate is only about 8 revolutions per hour. If the 10-micro-g number from Settled Cryogenic Propellant Transfer is correct, then 19 revolutions per day would suffice.

Option 2 would only be worth considering if the fueling in option 1 took so long that the ullage burn wasn't feasible and the burn required to spin and de-spin the two starships was a lot smaller. If the ullage burns are so small they can be done as part of the orbit raising and/or venting you have to do anyway in LEO, then I can see why you wouldn't consider anything else.

Both scenarios have to cope with the fact that the center of mass is going to move during this operation. Assuming both vehicles participate in the ullage burn, that's probably pretty easy to handle, since you'll just increase the thrust on the one that's filling and decrease it on the one that's emptying. In case #2, you can probably ignore it and just live with the fact that the rotation won't be centered on the docking unit.

So does anyone have any idea how long the ullage burns are likely to take and how much fuel they'll expend? From previous posts, I gather that even fully throttled down, this can't be done with the Raptor engines. I've seen some discussion, but I don't think there was any conclusion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 07/27/2022 09:41 am
Why did SpaceX do away with the tail to tail fueling?  A little thrust going toward the tanker would make the tanker empty into A Starship.  This might still be done but on the side, Starship facing one way tanker the other but docked and emptying tanker.

Little thrust by itself wouldn't be enough. The flow rate at milligee acceleration wound be extremely slow. You need pressure differential (the easiest way is to have ullage pressure difference, but you could also use a pump).

You need the minimal gravity to ensure the tank outlet in covered by liquid, not for pumping.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/27/2022 04:05 pm
You need the minimal gravity to ensure the tank outlet in covered by liquid, not for pumping.
Any idea how long it sloshes around before it settles? And I guess the microgravity accelerations means that fueling the target vehicle "upside down" isn't a problem, as compared to the issues with fueling a vehicle on Earth from the top of the tank.

In the tail-to-tail configuration, I was wondering where the ullage engines would go. Or are they likely so small that it doesn't much matter?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 07/27/2022 05:09 pm
You need the minimal gravity to ensure the tank outlet in covered by liquid, not for pumping.
Any idea how long it sloshes around before it settles? And I guess the microgravity accelerations means that fueling the target vehicle "upside down" isn't a problem, as compared to the issues with fueling a vehicle on Earth from the top of the tank.

In the tail-to-tail configuration, I was wondering where the ullage engines would go. Or are they likely so small that it doesn't much matter?

It will take a few minutes. Generally check out how long restartable upper stages do pre burn venting/thrusting. And scale that up by a square root of height difference between typical upper stages ans Starship tankage. Or by 6th degree root of volume difference.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/27/2022 06:02 pm
You need the minimal gravity to ensure the tank outlet in covered by liquid, not for pumping.
Any idea how long it sloshes around before it settles? And I guess the microgravity accelerations means that fueling the target vehicle "upside down" isn't a problem, as compared to the issues with fueling a vehicle on Earth from the top of the tank.

In the tail-to-tail configuration, I was wondering where the ullage engines would go. Or are they likely so small that it doesn't much matter?

It will take a few minutes. Generally check out how long restartable upper stages do pre burn venting/thrusting. And scale that up by a square root of height difference between typical upper stages ans Starship tankage. Or by 6th degree root of volume difference.
Isn't there's a big difference between that and refueling, though? For the restart case, as you say, there just needs to be enough propellant over the intakes to get the engines started. Once the engines have fired for even a second, the thrust will settle the contents of the tanks. But with refueling, I would think the ullage burn would need to settle the contents more thoroughly. Of course, when it comes to refueling, I guess it's probably okay if the pump sucks vapor from time to time, whereas that generally wrecks engines, so maybe it's not an issue.

Separate question: Won't there also be a need to pump the pressurization gas from the recipient vehicle back into the donor? I haven't heard that talked about, but (in the spirit of the recent discussion), that's also extra plumbing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: daavery on 07/27/2022 08:53 pm

Separate question: Won't there also be a need to pump the pressurization gas from the recipient vehicle back into the donor? I haven't heard that talked about, but (in the spirit of the recent discussion), that's also extra plumbing.

just vent the receiving vehicle tank ( vent downward to add ullage thrust)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/27/2022 11:05 pm

Separate question: Won't there also be a need to pump the pressurization gas from the recipient vehicle back into the donor? I haven't heard that talked about, but (in the spirit of the recent discussion), that's also extra plumbing.

just vent the receiving vehicle tank ( vent downward to add ullage thrust)

You'd also have to pressurize the sending tank.  It's much easier just to tie the two ullage spaces together with a simple line.  It only has to be wide enough to move enough gas to offset the mass flow of the liquid.  Maybe 1cm?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: daavery on 07/27/2022 11:25 pm

Separate question: Won't there also be a need to pump the pressurization gas from the recipient vehicle back into the donor? I haven't heard that talked about, but (in the spirit of the recent discussion), that's also extra plumbing.

just vent the receiving vehicle tank ( vent downward to add ullage thrust)

You'd also have to pressurize the sending tank.  It's much easier just to tie the two ullage spaces together with a simple line.  It only has to be wide enough to move enough gas to offset the mass flow of the liquid.  Maybe 1cm?

depending on your config all that will move the fluids is differential pressure - you want different ullage pressures in the 2 tanks
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/27/2022 11:42 pm

Separate question: Won't there also be a need to pump the pressurization gas from the recipient vehicle back into the donor? I haven't heard that talked about, but (in the spirit of the recent discussion), that's also extra plumbing.

just vent the receiving vehicle tank ( vent downward to add ullage thrust)

You'd also have to pressurize the sending tank.  It's much easier just to tie the two ullage spaces together with a simple line.  It only has to be wide enough to move enough gas to offset the mass flow of the liquid.  Maybe 1cm?

depending on your config all that will move the fluids is differential pressure - you want different ullage pressures in the 2 tanks

I did this computation somewhere up thread, but IIRC it requires something like a 100W electric pump to do a 150t transfer in under an hour, as long as the ullage spaces are connected. 

And there's another little problem with differential pressure:  If for some reason you accidentally uncover the inlet from the source tank, all of that differential pressure equalizes almost instantly, and you have a big mess.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BT52 on 07/27/2022 11:56 pm
100W? little low i think. Sir could u share your initial conditions. Or at least provide .txt or even spreadsheet Thanks.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/28/2022 01:55 am

Separate question: Won't there also be a need to pump the pressurization gas from the recipient vehicle back into the donor? I haven't heard that talked about, but (in the spirit of the recent discussion), that's also extra plumbing.

just vent the receiving vehicle tank ( vent downward to add ullage thrust)

You'd also have to pressurize the sending tank.  It's much easier just to tie the two ullage spaces together with a simple line.  It only has to be wide enough to move enough gas to offset the mass flow of the liquid.  Maybe 1cm?
There's already autog press plumbing in the engine bay not far from the QD plate. Does anybody have that labeled QD pic handy? It might already be there.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/28/2022 02:19 am
Are numbers available for how long it takes GSE to fill the SS? This sets an upper limit on transfer rates - not that I think they'll go that fast.


On the ground the GSE has to overcome a pressure head caused by the weight of fuel above it plus viscosity/friction losses. On orbit the weight caused by settling thrust will be minuscule in comparison and I'd expect the viscosity/friction to dominate and to be roughly the same as experienced by the GSE. From this a high end BOE pump power number should be available without getting too esoteric.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: daavery on 07/28/2022 03:38 am
friction/viscosity is probably orders of magnitude greater on the ground - tank farm to rocket is on the order of 500 ft of pipe and needs probably 10 elbows which adds another 400-500 ft worth of friction. for reference the OTF has over 500hp of turbine pumps for CH4  and over 1000hp  for LOX
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 07/28/2022 07:26 am
There's already autog press plumbing in the engine bay not far from the QD plate. Does anybody have that labeled QD pic handy? It might already be there.

Female (ship-side) QD plate below. Labelled. And male (GSE-side) belower.

"Prepress" are the two you want.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 07/28/2022 10:37 am
100W? little low i think. Sir could u share your initial conditions. Or at least provide .txt or even spreadsheet Thanks.

Yup. Something is off, IMO.

100W would be more likely good for for a 1t per hour rather than 150t per hour.

Likely the calculation has some error or it assumed no friction/no viscosity case -- while especially at low flow rates it would actually dominate.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/28/2022 03:28 pm
So would it make the most sense to build a special docking station with something equivalent to two quick disconnect arms? It would sit between two starships, connected to both, and transfer fuel from one to the other exactly as the system does on the ground. Each starship would be responsible for the ullage burn (and keeping it balanced as the center of mass moved during fueling), but the pumps would be in the docking station. The docking station would need its own power, and big enough solar panels might be a problem, but perhaps big batteries would suffice with solar panels large enough to recharge them in 24 hours or so.

That certainly minimizes the plumbing (no change to either starship), and if you can use the same QD structure that's used on the ground, then even that's not a new part.

Does that all make sense? Or is that more or less what people have been talking about already and I'm just slow figuring it out? :-)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/28/2022 03:35 pm
100W? little low i think. Sir could u share your initial conditions. Or at least provide .txt or even spreadsheet Thanks.

Yup. Something is off, IMO.

100W would be more likely good for for a 1t per hour rather than 150t per hour.

Likely the calculation has some error or it assumed no friction/no viscosity case -- while especially at low flow rates it would actually dominate.

My initial back-of-napkin was here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2368599#msg2368599).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/28/2022 08:33 pm
There's already autog press plumbing in the engine bay not far from the QD plate. Does anybody have that labeled QD pic handy? It might already be there.

Female (ship-side) QD plate below. Labelled. And male (GSE-side) belower.

"Prepress" are the two you want.
Thankee. Prepress it is. The only reason I can come up with to not interconnect the ullage space in two ships mating for propellant transfer is the possibility of using a pressure differential of up to 6bar to move the the fluids.


Assuming the receiving ship has zero ullage pressure and the delivering ship is at six bar, the delivering ship must have enough stored ullage gas to equal the volume of the moved fluid at six bar. The mass of this gas vs pump mass + batteries/PV is probably the trade to look at.


Ullage pressure might or might not move the fluids fast enough for a production system. A pump and related can scale as necessary. Pressure only should be good enough for proof of concept and early testing at minimum.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/28/2022 08:46 pm
So would it make the most sense to build a special docking station with something equivalent to two quick disconnect arms? It would sit between two starships, connected to both, and transfer fuel from one to the other exactly as the system does on the ground. Each starship would be responsible for the ullage burn (and keeping it balanced as the center of mass moved during fueling), but the pumps would be in the docking station. The docking station would need its own power, and big enough solar panels might be a problem, but perhaps big batteries would suffice with solar panels large enough to recharge them in 24 hours or so.

That certainly minimizes the plumbing (no change to either starship), and if you can use the same QD structure that's used on the ground, then even that's not a new part.

Does that all make sense? Or is that more or less what people have been talking about already and I'm just slow figuring it out? :-)
LoL. Yup. I think this one's been discussed. It's prime advantage is both the accumulated and the tanker would be exactly the same build.


Downside is it has no tanks, so there is zero chance of moving it if that becomes desirable, short of hitching a ride.


Which way is best? I doubt anybody at SpaceX really knows yet. I sure don't. That sucker's got to fly so it's working in a vacuum instead of us working in a vacuum.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: robot_enthusiast on 07/29/2022 03:00 am
So would it make the most sense to build a special docking station with something equivalent to two quick disconnect arms? It would sit between two starships, connected to both, and transfer fuel from one to the other exactly as the system does on the ground. Each starship would be responsible for the ullage burn (and keeping it balanced as the center of mass moved during fueling), but the pumps would be in the docking station. The docking station would need its own power, and big enough solar panels might be a problem, but perhaps big batteries would suffice with solar panels large enough to recharge them in 24 hours or so.

That certainly minimizes the plumbing (no change to either starship), and if you can use the same QD structure that's used on the ground, then even that's not a new part.

Does that all make sense? Or is that more or less what people have been talking about already and I'm just slow figuring it out? :-)
The problem with that is that the system lacks the medium term storage needed to enable flexibility in the schedule for the accumulation flights. This desire for flexibility with the scheduling of tanker flights seems to have driven SpaceX down the path they are currently pursuing, with dedicated orbital-only depots. From all appearances, the depot ships will be based off of the tankers, with as few changes as possible to suit their needs. Deleted entry hardware, additional thermal insulation, possible extra tank capacity to account for boiloff, and hardware for propellant transfer with docked ships (compatible QD plate and likely dedicated ullage thrusters) seem to be their medium-long term solution to the problem. In the long term, once they've mastered active cooling solutions to allow for minimal boiloff over longer time periods, the potential limitations from being locked into a specific inclination will be minimized by putting tankers in every useful orbit. In the shorter term (ie. HLS contracts), the relatively low-cost nature of the depot design should allow missions to be profitable even with single-use depots.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/29/2022 03:22 pm
So would it make the most sense to build a special docking station with something equivalent to two quick disconnect arms? It would sit between two starships, connected to both, and transfer fuel from one to the other exactly as the system does on the ground. Each starship would be responsible for the ullage burn (and keeping it balanced as the center of mass moved during fueling), but the pumps would be in the docking station. The docking station would need its own power, and big enough solar panels might be a problem, but perhaps big batteries would suffice with solar panels large enough to recharge them in 24 hours or so.

That certainly minimizes the plumbing (no change to either starship), and if you can use the same QD structure that's used on the ground, then even that's not a new part.

Does that all make sense? Or is that more or less what people have been talking about already and I'm just slow figuring it out? :-)
The problem with that is that the system lacks the medium term storage needed to enable flexibility in the schedule for the accumulation flights. This desire for flexibility with the scheduling of tanker flights seems to have driven SpaceX down the path they are currently pursuing, with dedicated orbital-only depots. From all appearances, the depot ships will be based off of the tankers, with as few changes as possible to suit their needs. Deleted entry hardware, additional thermal insulation, possible extra tank capacity to account for boiloff, and hardware for propellant transfer with docked ships (compatible QD plate and likely dedicated ullage thrusters) seem to be their medium-long term solution to the problem. In the long term, once they've mastered active cooling solutions to allow for minimal boiloff over longer time periods, the potential limitations from being locked into a specific inclination will be minimized by putting tankers in every useful orbit. In the shorter term (ie. HLS contracts), the relatively low-cost nature of the depot design should allow missions to be profitable even with single-use depots.
Completely agree but for one point. We have logic'd through all the identifiable issue, often with divergent conclusions, but we have not seen a shred of hard evidence as to specifics of what SX really intends because right now their intentions are only general. 


That amazing ship hasn't even made it to orbit yet. No responsible engineering organization that specializes in rapid iterative development would expend much energy optimizing a complex, not yet needed capability for a system that has only demonstrated a few of the many capabilities needed for that complex operation.


They've done what I call noodling. That's what we do here. The difference is they have more information and they can read our memos and we can't read theirs.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/29/2022 07:51 pm
Given all this, it seems to me that there's a lot of functionality that's needed for refueling but probably isn't needed in any given Starship. Does it make sense for them to build a specialized "docking station" to handle all of that and avoid complicating the Starship design?

The rough idea is a sort of cylinder that two Starships dock to which has quick disconnect arms that reach down to attach to each one. Lacking a clear idea of a) how they dock and b) where the QD ports are going to be, I can't be more specific. The dock would handle all the ullage burns and would need solar panels to provide power to pump the propellants around and keep them cool.

That's a whole new structure that has to fit into a Starship and unfold itself in space, but maybe that's not so bad. It seems better than adding a whole lot of hardware to every Starship that would only be used during refueling. Does that all seem reasonable?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: groknull on 07/29/2022 08:32 pm
Given all this, it seems to me that there's a lot of functionality that's needed for refueling but probably isn't needed in any given Starship. Does it make sense for them to build a specialized "docking station" to handle all of that and avoid complicating the Starship design?

The rough idea is a sort of cylinder that two Starships dock to which has quick disconnect arms that reach down to attach to each one. Lacking a clear idea of a) how they dock and b) where the QD ports are going to be, I can't be more specific. The dock would handle all the ullage burns and would need solar panels to provide power to pump the propellants around and keep them cool.

That's a whole new structure that has to fit into a Starship and unfold itself in space, but maybe that's not so bad. It seems better than adding a whole lot of hardware to every Starship that would only be used during refueling. Does that all seem reasonable?

Deployable free-flyers and tethered / robot arm deployed adapters have been discussed earlier in this thread.

IMO further discussion of those concepts is not unreasonable.  Formalization of each concept group may help discussion.  That would probably clarify assumptions, questions, benefits, and drawbacks.

e.g.
- <concept name> (to help identify which concept is being discussed)
- <general concept description> (deployable free flyer; arm deployed adapter; adapter attached to depot tanker; adapter attached to ferry tanker; gender bent depot tanker with GSE adapter; etc.)
- <free flyer. (yes | no)
- <orbit lifetime> (goes up with, comes home with depot tanker; stays in orbit for a refilling campaign; stays in orbit permanently; stays in orbit until obsolete, then retrieved)
- <includes thrusters> (yes | no)
- <thruster type(s)> (ion; cold gas; MethaLOX; MethOX; etc.)
- <plane change capable?> (yes | no; time required for the plane change)
- <includes batteries and solar arrays> (yes | no)
- <includes depot methane and LOX propellant tanks?> (yes | no)
- <extensible?> (fixed size, single purpose; modular - configuration on ground; modular - extensible in orbit)
- <evolution of, or adaptation of existing tech?> (Starlink bus, Dragon components, Dragon XL components, Starship components)
- <adaptable / intended for adaptation> (free flyer adapter intended to be used in multiple Earth orbits, plus around Mars; free flyer adapter is just an initial implementation of future general purpose ferry or work platform; etc.)
- <alignment with SpaceX goals> (If bespoke hardware, how does this help with getting to Mars?; Does this accelerate or delay Starship development or deployment?>
- <applicable timeframe> (When is this concept or version of concept valid?; When will evolving tech or a large number of Starships built make this concept obsolete?)

As OTV Booster and others have pointed out, SpaceX isn't showing all of their cards, and will change their hand to adapt to changing situations, so we don't have enough information to accurately guess what they are going to do.  We can, however, have a lot of fun playing with concepts and looking at the trade offs.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/29/2022 09:01 pm
Yet another half-baked geometrical thought:

If you put the depot's hardware (active side of the QD, cryocooler, pumps, etc.) in the ogive portion of the nose, it's straightforward to dock the depot and the other ship (either a lift tanker or a payload Starship to be refueled), dorsal-to-dorsal but nose-to-tail.  However, as has been pointed out, then you have to deal with ullage management in two different directions.

So what about this?  Imagine sort of a backwards chomper, i.e. a fairing door whose bottom is at about the level where the ogive portion of the nose begins, whose top is a bit below the level of the LOX header tank, and whose sides are roughly at the midline of the fairing.  Now put the whole fairing on an arm that can lift it clear of the depot and flip it over.  On the inside of the removed door are solar cells and docking hardware.  When a Starship wishes to use the depot, it slides into the (now concave) door segment and latches.  The active QD then extends from the open ogive space and connects to the Starship.

You now have both Starships pointing in the same direction, which removes the ullage issues.  What you've sacrificed is some amount of torsional stability between the two Starships.  Note that this is still a depot (or other kind of tanker) that can return to EDL, because everything will close up and stow, leaving the TPS alone.

I'm not quite sure I have the geometry right in the attached very bad picture.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/29/2022 09:34 pm
Given all this, it seems to me that there's a lot of functionality that's needed for refueling but probably isn't needed in any given Starship. Does it make sense for them to build a specialized "docking station" to handle all of that and avoid complicating the Starship design?

The rough idea is a sort of cylinder that two Starships dock to which has quick disconnect arms that reach down to attach to each one. Lacking a clear idea of a) how they dock and b) where the QD ports are going to be, I can't be more specific. The dock would handle all the ullage burns and would need solar panels to provide power to pump the propellants around and keep them cool.

That's a whole new structure that has to fit into a Starship and unfold itself in space, but maybe that's not so bad. It seems better than adding a whole lot of hardware to every Starship that would only be used during refueling. Does that all seem reasonable?
If you take that docking station and give it full sized tanks, it is a depot heavy. Tanks are the only thing missing from it.


The virtue I do see in a deployable docking station over a stay on orbit depot is that it has no raptors to waste. If it needs to change orbits it will need only one, saving that mass and the mass of the tanks.


On the down side, it's something new to design. Maybe that's a good trade.


A point in defense of the depot heavy which you might have missed is that no other ships need any changes. The depot would supply all the services of your docking station and all other ships would dress out plain jane. Well, maybe some attachment points which might also be needed for the docking station.


Still, saving the mass of the engines and tankage for orbit change does sound like a good trade if it can be designed to fit in and deploy from an SS. Mass shouldn't  be a problem.


Late thought. Somebody correct me if I've got this wrong. I think we're going to need a depot on high earth or lunar orbit. Under this plan that would mean a loaded tanker and a docking station. Is there anything outrageous about a docking station hitching a ride on a full tanker? It's has to attach anyway. It'll put the CoM a bit off the centerline but hey, that's why the engines gymbal.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 07/29/2022 10:27 pm
Yet another half-baked geometrical thought:

If you put the depot's hardware (active side of the QD, cryocooler, pumps, etc.) in the ogive portion of the nose, it's straightforward to dock the depot and the other ship (either a lift tanker or a payload Starship to be refueled), dorsal-to-dorsal but nose-to-tail.  However, as has been pointed out, then you have to deal with ullage management in two different directions.

So what about this?  Imagine sort of a backwards chomper, i.e. a fairing door whose bottom is at about the level where the ogive portion of the nose begins, whose top is a bit below the level of the LOX header tank, and whose sides are roughly at the midline of the fairing.  Now put the whole fairing on an arm that can lift it clear of the depot and flip it over.  On the inside of the removed door are solar cells and docking hardware.  When a Starship wishes to use the depot, it slides into the (now concave) door segment and latches.  The active QD then extends from the open ogive space and connects to the Starship.

You now have both Starships pointing in the same direction, which removes the ullage issues.  What you've sacrificed is some amount of torsional stability between the two Starships.  Note that this is still a depot (or other kind of tanker) that can return to EDL, because everything will close up and stow, leaving the TPS alone.

I'm not quite sure I have the geometry right in the attached very bad picture.
This is what I've been proposing. The folding multi-use load-bearing fairing can probably be omitted for a simpler non load bearing panel, and the QD arm itself used for local grappling and mating. No need for two pieces of added hardware that can extend, actuate, and latch, when you can use one.
Depending on the final 'chomper' geometry (would need to be able to fold far enough to clear, but this may be needed for payload deployment anyway) this could even mean the QD assembly is entirely self-contained within the payload bay, so the depot/tanker can have that unit removed (and the extra plumbing blanked off) and the vehicle be used for payload launches when no depots or tankers are needed in order to pay for itself, rather than sitting around doing nothing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/29/2022 11:54 pm
Yet another half-baked geometrical thought:

If you put the depot's hardware (active side of the QD, cryocooler, pumps, etc.) in the ogive portion of the nose, it's straightforward to dock the depot and the other ship (either a lift tanker or a payload Starship to be refueled), dorsal-to-dorsal but nose-to-tail.  However, as has been pointed out, then you have to deal with ullage management in two different directions.

So what about this?  Imagine sort of a backwards chomper, i.e. a fairing door whose bottom is at about the level where the ogive portion of the nose begins, whose top is a bit below the level of the LOX header tank, and whose sides are roughly at the midline of the fairing.  Now put the whole fairing on an arm that can lift it clear of the depot and flip it over.  On the inside of the removed door are solar cells and docking hardware.  When a Starship wishes to use the depot, it slides into the (now concave) door segment and latches.  The active QD then extends from the open ogive space and connects to the Starship.

You now have both Starships pointing in the same direction, which removes the ullage issues.  What you've sacrificed is some amount of torsional stability between the two Starships.  Note that this is still a depot (or other kind of tanker) that can return to EDL, because everything will close up and stow, leaving the TPS alone.

I'm not quite sure I have the geometry right in the attached very bad picture.
Yes. Much better than catawompus. My objections are some quibbles, and some not so quibbly.


That door. You're trying to make the inside diameter, probably with reinforcement, mate up to the outside diameter. Won't work, although if there were some mating surfaces that lined up with the edges it might. On the other hand, do you really want a hinged hatch as part of a latching arrangement? Maybe edzieba is right and nothing more than the QD is needed to attach. I don't agree but...


I think you came up with some numbers on the amount of power needed to run a cooler. Don't remember what they were except they were big. I don't think what you're showing would be enough. If EDL is intended, Solar white paint is out. I'm not seeing an easy sun shade deploy. Same for PV. Maybe cheap flexible throwaway PV? Put it conformal over the heat tiles and let it burn away? Where does the radiator go?


They changed the methane header again. It's up in the nose these days. Also, do you remember the boxy reinforcing rings we saw on the early builds? That may have changed. If it hasn't, there isn't very much room up there. I've no idea of the size of a cryo cooler. Maybe it'll fit. If not, the tanks may not stretch to 1600t without a full ship stretch.


The big deploy stuff, PV, radiators and sun shade, are a big reason I see the depot as something different than a plain jane tanker. Or maybe a docking station and an accumulator tanker. If it needs a cryo cooler it will need both PV and radiator. Maybe it can get away with just a sun shade. That needs to deploy. It's gotta have something or that propellant will boil off.


Edit to add: with the methane header moved up to the nose there's an easy tapoff point. I think this system would demand active pumping. Gotta think on this some more.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/30/2022 01:06 am
Yet another half-baked geometrical thought:

If you put the depot's hardware (active side of the QD, cryocooler, pumps, etc.) in the ogive portion of the nose, it's straightforward to dock the depot and the other ship (either a lift tanker or a payload Starship to be refueled), dorsal-to-dorsal but nose-to-tail.  However, as has been pointed out, then you have to deal with ullage management in two different directions.

So what about this?  Imagine sort of a backwards chomper, i.e. a fairing door whose bottom is at about the level where the ogive portion of the nose begins, whose top is a bit below the level of the LOX header tank, and whose sides are roughly at the midline of the fairing.  Now put the whole fairing on an arm that can lift it clear of the depot and flip it over.  On the inside of the removed door are solar cells and docking hardware.  When a Starship wishes to use the depot, it slides into the (now concave) door segment and latches.  The active QD then extends from the open ogive space and connects to the Starship.

You now have both Starships pointing in the same direction, which removes the ullage issues.  What you've sacrificed is some amount of torsional stability between the two Starships.  Note that this is still a depot (or other kind of tanker) that can return to EDL, because everything will close up and stow, leaving the TPS alone.

I'm not quite sure I have the geometry right in the attached very bad picture.
This is what I've been proposing. The folding multi-use load-bearing fairing can probably be omitted for a simpler non load bearing panel, and the QD arm itself used for local grappling and mating. No need for two pieces of added hardware that can extend, actuate, and latch, when you can use one.
Depending on the final 'chomper' geometry (would need to be able to fold far enough to clear, but this may be needed for payload deployment anyway) this could even mean the QD assembly is entirely self-contained within the payload bay, so the depot/tanker can have that unit removed (and the extra plumbing blanked off) and the vehicle be used for payload launches when no depots or tankers are needed in order to pay for itself, rather than sitting around doing nothing.
I think the advantages of stretch tanks have been well established. That would make the tanker a different variant than a dry cargo SS. Indeed, having a special cargo hatch for a refueling QD plate, either for a depot variant or all tankers for commonality, would demand it be a variant. The current StarLink deployment slot shows that SX is not shy about optimizing the design to fit the mission. The upshot of this is that if they left an accumulator on orbit it would only impact the tanker fleet, not the dry cargo fleet. The tanker is probably the simplest and least expensive variant they will ever have.


It looks like they will need two accumulators for Artemus but it's not clear that they will need two tanker fleets. Until the first Mars push there's no clear reason that they'd need more than one tanker fleet. Indeed, unless SS is a total reuse flop, there's every reason to believe their production capacity will be greater than the demand. A few accumulators that stay on orbit will not put any launch strain in SX.


If the accumulator is going to stay on orbit it needs no fins or heatshield. At this point it is yet another variant and can be called a depot. Why do half measures. Ditch the second QD plate in the nose and permanently deploy any PV, radiators, sun shade, thrusters and anything else it needs and be done with it.


It won't start out this way. They'll play around with two ships with overlapping  missions. Then move to a dry cargo ship and a tanker, moving further ahead on each trial, adding only what they need. Eventually they'll have it figured out.


Then they'll either build a full tilt depot or a docking station. I'm warming to the docking station.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/30/2022 04:07 am
If you take that docking station and give it full sized tanks, it is a depot heavy. Tanks are the only thing missing from it.
Actually, I was thinking of the docking station as also being a way for crew to move from one Starship to another as well. My thought was that although you want to bring up the crew after you've already got your mission vehicle fully fueled, you really don't want to transfer all that fuel from one vehicle to another if you can help it. But if the docking station is a tube that two Starships dock to, why not make it a passageway for the crew and small amounts of last-minute cargo?

Not sure if that's really necessary, but it makes the station rather different from yet another Starship variant.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 07/30/2022 04:12 am
- <thruster type(s)> (ion; cold gas; MethaLOX; MethOX; etc.)
- <plane change capable?> (yes | no; time required for the plane change)
Two questions: First, do ion engines provide anything like the thrust required for ullage? I know it doesn't take much, but even so.
Second, I thought the cost of plane changes in LEO was prohibitive. Is that really on the table?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nomadd on 07/30/2022 04:57 am
- <thruster type(s)> (ion; cold gas; MethaLOX; MethOX; etc.)
- <plane change capable?> (yes | no; time required for the plane change)
Two questions: First, do ion engines provide anything like the thrust required for ullage? I know it doesn't take much, but even so.
Second, I thought the cost of plane changes in LEO was prohibitive. Is that really on the table?
That would be like a butterfly trying to move an elephant.
 Inclination changes are hard. Plane changes just take time. Most Starlinks go to different planes than they start out in.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 07/30/2022 10:25 am
Following the suggestion of groknull I outlined this RPOD concept for comparative discussion.

Concept: Landing Docking Posts

Where posted:  Short series beginning at p. 55 Reply #1086, with sketches.

Description: Accumulator/depot in orbit is visited by SS resupply tankers and SS outgoing ships. They  land dock on three posts jutting from the accumulator, whose projecting ends are in the same plane as a plumbing connector plate on the ventral surface of the accumulator. Each visiting ship has a corresponding connector plate that automatically joins to the accumulator plate upon landing docking, requiring no other robotic arms or gadgets.

Free flyer: no

Orbital lifetime: Until obsolete

Includes thrusters?  Use thrusters of standard Starships

Plane change capable: Not addressed, but assumes appropriate orbital position around a planetary body that serves multiple missions.

Batteries and solar arrays: Not addressed, but beefed up arrays and batteries would allow accumulator to recharge batteries of a visiting ship while receiving or providing propellants.

Both LOX and CH4 tanks? Yes. Accumulator is a standard SS tanker with strap-on posts that automatically erect in space.

Modular: No extensions in space are planned other than erecting the posts.

Adaption of other existing tech: No. Minimal changes to Starship, assuming SS is “existing” tech.

Adaptable to multiple orbits: Only to the extent that any standard SS could theoretically change to a different orbit.

Alignment with SpaceX goals: Refueling in space is required. This is one approach.

Edited: landing changed to docking
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 07/30/2022 12:06 pm
If you take that docking station and give it full sized tanks, it is a depot heavy. Tanks are the only thing missing from it.
Actually, I was thinking of the docking station as also being a way for crew to move from one Starship to another as well. My thought was that although you want to bring up the crew after you've already got your mission vehicle fully fueled, you really don't want to transfer all that fuel from one vehicle to another if you can help it. But if the docking station is a tube that two Starships dock to, why not make it a passageway for the crew and small amounts of last-minute cargo?

Not sure if that's really necessary, but it makes the station rather different from yet another Starship variant.

Eventually, like in the future.

The fuel transfer and cargo moving will be the same thing.
Sort of like the 7/11 of space.
I like the idea of nose docking to a slowing rotating hub. That way ullage is free.
Now just have connectors on the hub for fuel and people/cargo.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 07/30/2022 06:57 pm
This is what I've been proposing. The folding multi-use load-bearing fairing can probably be omitted for a simpler non load bearing panel, and the QD arm itself used for local grappling and mating. No need for two pieces of added hardware that can extend, actuate, and latch, when you can use one.

I still don't think you can get sufficient rigidity (torsional or otherwise) from a one-point latch, or even a multi-point latch with the attachments within a short distance of one another.  The nice thing about using the inside of the door as the second point of contact is that it's simple and far enough away from the active QD to be rigid.

Quote
Depending on the final 'chomper' geometry (would need to be able to fold far enough to clear, but this may be needed for payload deployment anyway) this could even mean the QD assembly is entirely self-contained within the payload bay, so the depot/tanker can have that unit removed (and the extra plumbing blanked off) and the vehicle be used for payload launches when no depots or tankers are needed in order to pay for itself, rather than sitting around doing nothing.

Largely agree here, although I still think that we're going to see an LSS with 1500t-1600t tanks, and you might as well use that layout for lift tankers and depots.  But if everything remains at 1200t, then yes, you could conceivably have a single propulsion section layout that was good for everything.

We know that there are going to be several different noses.  In order that they'll appear:

1) The Pez Dispenser is only good for Starlinks or third-party birds that conform to Starlink form factor and specs.

2) There's going to be a tanker/depot nose, unless the free-flyer is viable.  (I've come to the conclusion that it would only be a solution of last resort, with a payload kit being far, far preferable.)

3) There's going to be an LSS hatch + crew module version.

4) There's going to be some chomper-like version for deploying bulky payloads into microgravity.

5) I expect to see a jettisonable fairing for expendable deep space missions, but that one will probably wait a while.

It's possible that the chomper and the tanker nose will be the same, but I'd put my bet on it being unique, and designed to work with a 1500t (of propellant) propulsion section. 

I think you came up with some numbers on the amount of power needed to run a cooler. Don't remember what they were except they were big. I don't think what you're showing would be enough. If EDL is intended, Solar white paint is out. I'm not seeing an easy sun shade deploy. Same for PV. Maybe cheap flexible throwaway PV? Put it conformal over the heat tiles and let it burn away? Where does the radiator go?

I think the cooling numbers were associated with a self-contained Sabaitier/electrolysis/RWGS plant for Mars.  That's considerably more challenging.  I haven't learned how to model heating in VLEO.

That said, there's an awful lot to be said for brute force.  If you can deploy enough PV and heat rejection, then you don't need anything other than bare-metal tanks.  And if you're going to swing most of the top of the nose out of the way, you're looking at at least a 6m x 6m trapezoidal hole.  You can deploy an awful lot of stuff out of a hole like that--even if it it contains active QD and docking hardware.  And it's not like PV and radiators with large surface areas are low-TRL technologies.  As for cryocoolers:  they're sized for the job, usually as multi-stage systems to keep efficiency up.  Again, you're talking about hundreds of m³ of space in the nose, even with stretched tanks.

It would be, admittedly, a bit a of Rube Goldberg, especially if you want to fold everything back up for recovery.  But if you can't do that, you can probably jettison the PV and radiators for recovery, or you can simply leave the depot on orbit.  Eventually, there will be a recoverable design.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/31/2022 05:05 am
A problem we're having is that we don't have a clue what boiloff rates to expect and what the various mitigation might give us. That impacts the BFS (Big Floppy Stuff) a depot or accumulator might need. PV, radiators and sunshield. There's also Solar white paint and spray on foam. That last isn't all that popular but it's there to look at.

I found this 2020 thesis on modeling boiloff mitigation. https://repository.tudelft.nl/islandora/object/uuid:f1e1c77e-f5be-4e6b-a813-e7f955760603/datastream/OBJ/download (https://repository.tudelft.nl/islandora/object/uuid:f1e1c77e-f5be-4e6b-a813-e7f955760603/datastream/OBJ/download)

It's 143 pages long and in the abstract it has an interesting comment.
Quote
Due to a lack of experimental and flight data of existing spacecraft, the Boil-off Monte Carlo program is not validated.

Even the real rocket scientists have only a hypothetical understanding of the problem. My bets are that we might see a StarLink launch or three that explore boiloff mitigation. Might see one with solar white and one with spray-on foam. Rotisserie spins. Anything that's quick, easy and on point.

Edit: cleanup and add: maybe just settin there for a day keeping the shiny side to the heat and a day with the tiles to the heat. Start and end each test with mild acceleration to get a quick n dirty on boiloff.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 07/31/2022 05:41 pm
Found an on point paper from 2006. Only 8 pages. Maybe somebody already linked it already. It's worth the read. https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf (https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf)


Broadly, it covers most everything we've been talking about except the SpaceX specific QD hookup. The most interesting points are:
- a list of the TRL's of the sub tasks
- settling thrust of down to 2x10^-5g already demonstrated.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 08/01/2022 12:59 pm
Largely agree here, although I still think that we're going to see an LSS with 1500t-1600t tanks, and you might as well use that layout for lift tankers and depots.  But if everything remains at 1200t, then yes, you could conceivably have a single propulsion section layout that was good for everything.
The Lunar Starship is the only vehicle that would need more than one 'standard' Starship's worth of propellant to be transferred, and those will be launching maybe once a year, if that. The rest of the time, a standard capacity Starship is sufficient as a tanker or depot for all regular launches (albeit most launches will likely require zero tankers, let alone a depot).
Rather than using a custom stretch depot Starship that may-or-may-not be able to ever deorbit (and even if it can, that's a lot of extra R&D work to make sure it actually works) and require s abunch of unique handling equipment and procedures to accommodate its greater size; for that once-a-year occasion that actually requires that capacity you can launch two regular sized depots instead.

In terms of cost and time, it's always preferable to make more copies of the same item than to make more one-off custom item variants whenever you can possibly help it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/01/2022 06:33 pm
Largely agree here, although I still think that we're going to see an LSS with 1500t-1600t tanks, and you might as well use that layout for lift tankers and depots.  But if everything remains at 1200t, then yes, you could conceivably have a single propulsion section layout that was good for everything.
The Lunar Starship is the only vehicle that would need more than one 'standard' Starship's worth of propellant to be transferred, and those will be launching maybe once a year, if that. The rest of the time, a standard capacity Starship is sufficient as a tanker or depot for all regular launches (albeit most launches will likely require zero tankers, let alone a depot).
Rather than using a custom stretch depot Starship that may-or-may-not be able to ever deorbit (and even if it can, that's a lot of extra R&D work to make sure it actually works) and require s abunch of unique handling equipment and procedures to accommodate its greater size; for that once-a-year occasion that actually requires that capacity you can launch two regular sized depots instead.

In terms of cost and time, it's always preferable to make more copies of the same item than to make more one-off custom item variants whenever you can possibly help it.

I think this is less work (and less manufacturing cost) than you think.  The overall length of the Starship doesn't change, just where you put the LOX/LCH4 bulkhead and the LCH4 dome.  Remember that, to a first-order approximation, you're just stacking ring segments, domes/bulkheads, and noses.  Changing the order of how these segments are stacked is trivial.  (Changing their plumbing and wiring is less than trivial, but it's still pretty easy.)

If you divide the manufacturing into a propulsion section, a nose, and various accoutrements, you can build everything you need:

1) Propulsion Section:
a) Short 1200t
b) Long 1550t.

2) Nose Section (long goes on short propulsion, short on long propulsion):
a) Starlink Pez Dispenser (long nose)
b) Tanker and/or Depot (short nose)
c) LSS nose (short nose)
d) Large micro-gravity payload nose "chomper" (long nose)
e) Jettisonable fairing for deep space expendable mission (probably short nose)
f) A crew launch nose (long nose, and possibly exotic escape system).

3) Various accoutrements:
a) TPS and bare metal surface.
b) Solar white surface.
c) Elonerons.
d) Header tanks.
e) Landing legs.
f) Lunar thruster package.

From these you make:

A) Vanilla Starship
Short propulsion
Pez dispenser or chomper long noses.
TPS/bare metal surface.
Elonerons, header tanks.
Cadence (first 5 years of operations):  20/yr Pez, 10/yr chomper.

B) Tanker (EDL-capable)
Long propulsion
Tanker short nose (with or without depot kit, depending on application)
TPS/bare metal surface
Elonerons, header tanks.
Cadence: 10/yr

Note that you can make a non-EDL depot with no elonerons, no headers, and solar white surface.

C) Lunar Starship
Long propulsion
LSS short nose with crew module, cargo deck, and crew access
Solar white surface.
Thrusters, landing legs.
Cadence: 1/yr

D) Deep-space "StarKicker"
Long propulsion
Short jettisonable fairing nose
No accoutrements at all
Cadence: 1/yr

E) Crew-Launch Starship
Short Propulsion
Long nose (possibility with exotic escape capabilities)
TPS/bare metal surface.
Header tanks, elonerons.
Possibly landing legs for aborts.
Cadence (first 5 years): 0/yr

All the real work is in the nose variants, and even then everything fits into either a long or short version.  Bear in mind that they must do the long propulsion version for LSS.  Once you have the tooling built for that, why wouldn't you use it when it made your life easier?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 08/01/2022 07:32 pm
The HSS Starship doe snot move the tanks up into the nose. the nose there is where the habitation and cargo areas are. The HSS Starship significantly lengthens the overall stack to achieve the greater tank capacity.

The reason that's a problem for a depot variant is that the HSS Starship also omits the flaps and the ability to re-enter and land. That means once you launch it, you don't get it back and can't use it for anything else. If you want to make a similarly stretched vehilce but add the flaps and TPS back on, you now have a bunch of unique structural work (longer vehicle 'spine'), EDL behaviour modelling, and testing, for one very limited use variant, since this stretched depot is only ever actually  needed[/i[ for HSS.
On the other hand, two regular-sized Starships used as depots with prop transfer hardware in the cargo bay only need to play depot for a week or two per year to handle HSS Starship, and the rest of the year can spend their time launching payloads just like any other Starship. They are also the right size for any regular Starship that needs in-orbit prop transfer (e.g. for direct GTO injection).

Or using your numbers: instead of spreading making 20x, 10x and 10x of 3 different variants, you make 40x of one variant and keep and modifications within the payload bay.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/01/2022 07:46 pm
Largely agree here, although I still think that we're going to see an LSS with 1500t-1600t tanks, and you might as well use that layout for lift tankers and depots.  But if everything remains at 1200t, then yes, you could conceivably have a single propulsion section layout that was good for everything.
The Lunar Starship is the only vehicle that would need more than one 'standard' Starship's worth of propellant to be transferred, and those will be launching maybe once a year, if that. The rest of the time, a standard capacity Starship is sufficient as a tanker or depot for all regular launches (albeit most launches will likely require zero tankers, let alone a depot).
Rather than using a custom stretch depot Starship that may-or-may-not be able to ever deorbit (and even if it can, that's a lot of extra R&D work to make sure it actually works) and require s abunch of unique handling equipment and procedures to accommodate its greater size; for that once-a-year occasion that actually requires that capacity you can launch two regular sized depots instead.

In terms of cost and time, it's always preferable to make more copies of the same item than to make more one-off custom item variants whenever you can possibly help it.
The 1600t tanker can work without a change to the OML. The common dome moves up and the upper dome moves up into what would otherwise be the cargo bay. All associated plumbing and wiring does need to stretch to match. Total tank stretch would be 6.4m for boiling props. A bit less for sub chilled. Landing characteristics should be very little different than an empty SS of standard internal dimensions.


There is one issue that might make a 1600t tanker unworkable. The static head pressure at the bottom of the tanks will be higher than normal. Ullage pressure and acceleration loading will need to be rethought. It's possible that this has already been baked into the design. Maybe it can be handled by ullage pressure and the acceleration profile. If it calls for a major redesign of the tanks and domes it probably won't happen. At least not until SSv2.


On the issue of special builds, it seems more of a non issue. If your building 10k widgets a day customizing even one is a big deal. The automotive industry routinely works with 'build sheets' that show every option available. The build sheet for a semi truck goes on for pages. It's low volume, almost kit like, assembly.


If SS were built the same way as SLS I'd agree with you on this. If the Pez dispenser works out we'll see that as a cargo variant with a very specific build sheet and IMO we'll see any other variant that's useful and can build on the core design. An unexpected benefit of rapid iteration at this level of sophistication is an unprecedented ability to be easily modify. What we have today is SN1 modified inch by inch and step by step.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/01/2022 08:02 pm
The HSS Starship doe snot move the tanks up into the nose. the nose there is where the habitation and cargo areas are. The HSS Starship significantly lengthens the overall stack to achieve the greater tank capacity.

Do you have a source for this?

Crew area doesn't need to be in the cylindrical section.  Indeed, the bottom of the cylindrical section needs to be in vacuum, because that's where the cargo hatch, airlocks, and elevator have to be.

You only need about 2m of cylindrical height to do this.  And the ogive portion of the nose provides more than enough volume to support the pressurized crew module, especially since there are no header tanks.

That gives you at least 6m of room for the LCH4 dome to be moved forward.  At 3.6:1 O:F and using boiling densities, you get an average density of 885kg/m³.  So 6m of extra tankage gives you about 340t more prop than the vanilla tankage, i.e., 1540t total.

Quote
The reason that's a problem for a depot variant is that the HSS Starship also omits the flaps and the ability to re-enter and land. That means once you launch it, you don't get it back and can't use it for anything else. If you want to make a similarly stretched vehilce but add the flaps and TPS back on, you now have a bunch of unique structural work (longer vehicle 'spine'), EDL behaviour modelling, and testing, for one very limited use variant, since this stretched depot is only ever actually  needed for HSS.

But this argument is based on the assumption that the outer mould line is really stretched.  I don't think it is.  You can put header tanks into a long propulsion section with no problem, and you'll wind up with exactly the same balance and dynamics as the short propulsion version during EDL, when the mains are mostly empty.  The only time the CoM will be substantially different is when the thing is loaded to the gills with propellant--but that only happens in vacuum, where you don't have aerodynamic issues.

Quote
On the other hand, two regular-sized Starships used as depots with prop transfer hardware in the cargo bay only need to play depot for a week or two per year to handle HSS Starship, and the rest of the year can spend their time launching payloads just like any other Starship. They are also the right size for any regular Starship that needs in-orbit prop transfer (e.g. for direct GTO injection).

Or using your numbers: instead of spreading making 20x, 10x and 10x of 3 different variants, you make 40x of one variant and keep and modifications within the payload bay.

Again, this is all based on the eloneron dynamics of the short and long propulsion sections being very different.  If the dynamics are the same, then, since you've already conceded that the LSS version needs the long propulsion module, why wouldn't you use it and save yourself the cost of an extra depot (even if the depot turns out to be an ordinary lift tanker, which I think we both think will be more likely than not)?

The other thing a long propulsion tanker allows is that you can easily play with how much you fill it for launch, optimizing usable prop delivered to orbit.  There's no fooling with auxiliary tankage in the payload bay, which would need to be vented and managed at launch.  Just fill the tanks a little more or less until you find the optimum.  Then every time the Raptor people manage to eke out a few extra kN from each engine, you can just fill the lift tanker a bit more.

You can't do that with a 1200t tank.  If the thrust increases, you can still only put 1200t into it, not 1220t or 1280t or 1300t.  But if the tank will take up to 1500t, then all of those attempts to optimize things are easy.

PS:  There's also nothing that prevents you from supporting a short nose for third-party payloads.  Even with only 2m of cylindrical height, the ogive is so huge that any satellite that was designed to be dual-sourced for launch would be well inside the static envelope.

So if you're really worried about some horrible imbalance between long and short propulsion modules, use the long ones for payload when they're not needed for tanking or LSS duties.  Or expend them for StarKicker missions.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bryan Hayward on 08/01/2022 08:18 pm

2) Nose Section (long goes on short propulsion, short on long propulsion):
...
e) Jettisonable fairing for deep space expendable mission (probably short nose)

From these you make:

D) Deep-space "StarKicker"
...
Short jettisonable fairing nose
...

 (... is snippage)
One nit I am going to ask about.  I realize that the fairing is likely to be extra mass you don't want to push around. However, it is another layer between the crew and space hazards. Do we really want to lose it?

Otherwise, I agree it seems reasonable that making versions of SS is relatively simple, once the base version is proven. The one exception may be the docking structure/mechanism on the depot as that may not be restricted to the nose, per other posts in this thread.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/01/2022 11:32 pm
e) Jettisonable fairing for deep space expendable mission (probably short nose)
I realize that the fairing is likely to be extra mass you don't want to push around. However, it is another layer between the crew and space hazards. Do we really want to lose it?

Unless crew count as "expendable", I'm assuming he's talking about BEO robotic missions.



doe snot

Oh dear.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 08/02/2022 03:18 am

There is one issue that might make a 1600t tanker unworkable. The static head pressure at the bottom of the tanks will be higher than normal.

Pressure at the bottom of the tank depends mostly on the thrust and not very much on the amount of fuel.  Unless you add engines when you lengthen the tank this should be manageable.

The easy way to think about this is that pressure is force per unit area, and all the force comes from the engines.

The hard way us that the extra mass reduces the acceleration, which reduces the head.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/02/2022 05:36 am
doe snot

Oh dear.

Shouldn't that be "Oh deer"?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/02/2022 03:51 pm

There is one issue that might make a 1600t tanker unworkable. The static head pressure at the bottom of the tanks will be higher than normal.

Pressure at the bottom of the tank depends mostly on the thrust and not very much on the amount of fuel.  Unless you add engines when you lengthen the tank this should be manageable.

The easy way to think about this is that pressure is force per unit area, and all the force comes from the engines.

The hard way us that the extra mass reduces the acceleration, which reduces the head.
Yes, the greater load, the lower the T/W and the lower the acceleration which lowers the head. But, increasing prop load by 33% doesn't decrease acceleration by 33% unless you have a magical zero dry mass.


F=MA. Engines supply force. Propellant supplies mass. Head at the bottom very much depends on the mass of the propellant. The only reason we have static head pressure is because earth supplies a constant 1g acceleration.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/02/2022 06:23 pm

There is one issue that might make a 1600t tanker unworkable. The static head pressure at the bottom of the tanks will be higher than normal.

Pressure at the bottom of the tank depends mostly on the thrust and not very much on the amount of fuel.  Unless you add engines when you lengthen the tank this should be manageable.

The easy way to think about this is that pressure is force per unit area, and all the force comes from the engines.

The hard way us that the extra mass reduces the acceleration, which reduces the head.
Yes, the greater load, the lower the T/W and the lower the acceleration which lowers the head. But, increasing prop load by 33% doesn't decrease acceleration by 33% unless you have a magical zero dry mass.


F=MA. Engines supply force. Propellant supplies mass. Head at the bottom very much depends on the mass of the propellant. The only reason we have static head pressure is because earth supplies a constant 1g acceleration.

Just do the math.  p = ρah.

You don't have to worry about loads at launch, because the stretched tanks won't be full.  In microgravity, I get a worst case 5.6kPa difference.  It's nothing like what the tanks would be subjected to, even with only 1200t of prop, at launch.  Everything's linear, and you know that the hydrostatic pressure drops to near zero at burnout, so the worst case is always going to be at startup with full tanks.

Here's an example:
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bryan Hayward on 08/02/2022 08:58 pm
e) Jettisonable fairing for deep space expendable mission (probably short nose)
I realize that the fairing is likely to be extra mass you don't want to push around. However, it is another layer between the crew and space hazards. Do we really want to lose it?

Unless crew count as "expendable", I'm assuming he's talking about BEO robotic missions.
I was thinking "expendable" in terms of SS means no re-use of the ship, not necessarily a statement about the lack of return of the whole mission or all its components. But I do hope that we'll be able to figure out crew protection on long haul trips.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/02/2022 09:48 pm
I was thinking "expendable" in terms of SS means no re-use of the ship, not necessarily a statement about the lack of return of the whole mission or all its components. But I do hope that we'll be able to figure out crew protection on long haul trips.

I'd expect a crew nose that can handle launch and EDL would evolve to handle at least trips to Mars and back.  Anything beyond that is probably science fiction for a while.

The "StarKicker" idea has been broached by Elon a few times, as a way to put heavy interplanetary (or even extrasolar) probes into the fastest transfers possible.  It doesn't necessarily have to have a jettisonable fairing, but it reduces the dry mass a bit and may have a couple of other advantages that are even more off-topic.  I'd expect SpaceX to do this eventually, but it's probably pretty low on the list--unless NASA drops by and says, "Hey, could you help us get 20t to Neptune right now?"
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/03/2022 12:38 am
Had an afternoon of profitable thinking. It brought on a prediction.


In 10 years, third party retanking will be a measurable portion of SX's SS 'for hire' launch income.


Here's my thinking. Given the current lamentable shape the world is in, the earth imaging market is bull. One major failing of sat imaging is that it is predictable. Activities can be timed around the sat's schedule and avoid exposure.


The biggest of these sats carry a good load of propellants and do have maneuverability but dV is limited and is a carefully managed resource. Refueling changes the game. Opinion: Once SX demonstrates on orbit transfer some agencies will want to have some serious discussions with them.


This says nothing directly about the technical issues we've been discussing. "nothing directly." If I hit the mark up above it changes assumptions and opens up the trade space. Not right away. But maybe in 5-8 years as a new generation of sats come on line with retanking baked in.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/03/2022 01:50 am

There is one issue that might make a 1600t tanker unworkable. The static head pressure at the bottom of the tanks will be higher than normal.

Pressure at the bottom of the tank depends mostly on the thrust and not very much on the amount of fuel.  Unless you add engines when you lengthen the tank this should be manageable.

The easy way to think about this is that pressure is force per unit area, and all the force comes from the engines.

The hard way us that the extra mass reduces the acceleration, which reduces the head.
Yes, the greater load, the lower the T/W and the lower the acceleration which lowers the head. But, increasing prop load by 33% doesn't decrease acceleration by 33% unless you have a magical zero dry mass.


F=MA. Engines supply force. Propellant supplies mass. Head at the bottom very much depends on the mass of the propellant. The only reason we have static head pressure is because earth supplies a constant 1g acceleration.

Just do the math.  p = ρah.

You don't have to worry about loads at launch, because the stretched tanks won't be full.  In microgravity, I get a worst case 5.6kPa difference.  It's nothing like what the tanks would be subjected to, even with only 1200t of prop, at launch.  Everything's linear, and you know that the hydrostatic pressure drops to near zero at burnout, so the worst case is always going to be at startup with full tanks.

Here's an example:
Hmm did a sanity check. Took you're numbers for the weight for a cubic meter of methane, and the height of the tank. Multiplied by 1.5, assuming that is the T/W. Then consulted my oracle for 1bar in kg/m^2 (10197.16) and divided by that. Got  a .6bar head looking at the methane. A higher number but within BOE spitting distance. I was assuming it would me much higher. This is low enough that it retires my worries about pressure management. Nowhere near the margins.


Hmmm, Wait. Don't we need ~4bar for the pump inlets? Well, that margin just shrank.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/03/2022 05:08 am
Hmm did a sanity check. Took you're numbers for the weight for a cubic meter of methane, and the height of the tank. Multiplied by 1.5, assuming that is the T/W. Then consulted my oracle for 1bar in kg/m^2 (10197.16) and divided by that. Got  a .6bar head looking at the methane. A higher number but within BOE spitting distance. I was assuming it would me much higher. This is low enough that it retires my worries about pressure management. Nowhere near the margins.

Hmmm, Wait. Don't we need ~4bar for the pump inlets? Well, that margin just shrank.

You don't care about T/GLOW when looking at Starship (not SuperHeavy) hydrostatics.  You care about burnout T/W for the SuperHeavy.  That's something that's almost certainly acceleration-limited.  3g?  5g?  Whatever it is, it's way, way more acceleration than a full Starship will generate at startup.

If you need 4bar for the pumps, then you can reduce ullage pressure by however much the stretch in the tanks increases the hydrostatic head.  Bottom line:  The tanks are already strong enough to accommodate the stretch, as long as they're not filled all the way on the pad--which they can't be, if you want a reasonable T/GLOW.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TomH on 08/03/2022 06:13 am
One major failing of sat imaging is that it is predictable. Activities can be timed around the sat's schedule and avoid exposure.
......propellants.....dV is limited..... Refueling changes the game. Opinion: Once SX demonstrates on orbit transfer some agencies will want to have some serious discussions with them...... maybe in 5-8 years as a new generation of sats come on line with retanking baked in.

Well, an alternative to that might be that SS makes spy birds become as plentiful as Starlinks: there's always a dozen or so over every part of Earth, moving in multiple directions. No maneuvering and retasking then required. Of course, eventually LEO will become like a pinball machine, the debris field so dense that nothing can penetrate in or out without severe damage.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/03/2022 04:29 pm
One major failing of sat imaging is that it is predictable. Activities can be timed around the sat's schedule and avoid exposure.
......propellants.....dV is limited..... Refueling changes the game. Opinion: Once SX demonstrates on orbit transfer some agencies will want to have some serious discussions with them...... maybe in 5-8 years as a new generation of sats come on line with retanking baked in.

Well, an alternative to that might be that SS makes spy birds become as plentiful as Starlinks: there's always a dozen or so over every part of Earth, moving in multiple directions. No maneuvering and retasking then required. Of course, eventually LEO will become like a pinball machine, the debris field so dense that nothing can penetrate in or out without severe damage.
There's a lot that can be done with a lot of small sats. What they can't do is the resolution that comes with a 3m optical system. That's expensive. SX can drop the cost of launch but the glass is what it is.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bryan Hayward on 08/03/2022 05:43 pm
(snip)
The "StarKicker" idea (snip)  I'd expect SpaceX to do this eventually, but it's probably pretty low on the list--unless NASA drops by and says, "Hey, could you help us get 20t to Neptune right now?"
I'd say the most likely and important need for 20t in a hurry would be to "Asteroid X orbit" (where X is a heretofore undetected extinction level event sized body) for a re-direct mission. If I were planetary defense authority, that's what I would be asking someone deeply familiar with SpX intentions to consider. Paying for the difference between a Starkicker and their Mars concept would be considerably cheaper than making a bespoke system from the ground up (a la SLS).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/03/2022 06:38 pm
To deflect an extinction level asteroid with Starship would require a launch capacity comparable to what the full fledged Martian city would need, like at LEAST 1 million tons to orbit per year, if not 100 million or 1 billion. Even if you’re using nukes, it’d require quite a lot of lift.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Zed_Noir on 08/03/2022 09:41 pm
To deflect an extinction level asteroid with Starship would require a launch capacity comparable to what the full fledged Martian city would need, like at LEAST 1 million tons to orbit per year, if not 100 million or 1 billion. Even if you’re using nukes, it’d require quite a lot of lift.
With sufficient lead time. Someone can build missiles using Starkicker frames armed with MT nuke boosted with about 20 tonnes of Tritium each. Drastically reduces the number of nukes and thus spacecrafts (lift) required especially if the missiles are use in salvos with simultaneous initiation of the payloads for a focused cone blast.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bryan Hayward on 08/03/2022 09:45 pm
To deflect an extinction level asteroid with Starship would require a launch capacity comparable to what the full fledged Martian city would need, like at LEAST 1 million tons to orbit per year, if not 100 million or 1 billion. Even if you’re using nukes, it’d require quite a lot of lift.

There are a lot of imponderables here, such as asteroid (or comet) mass/composition/orbit, but I'm assuming for the sake of in-space refueling discussions that getting out to "X" in a hurry is the point.  That means refueling as fast and as much as possible to get there in time for an infinitesimal nudge to matter. Would liquid refueling even be the point? Is Starkicker a robust solution for "get there ASAP" (neglecting the chances of mission success, admittedly small)?

That got me started on a different idea, which I've sent to Advanced Concepts so as not to send us off-topic.
https://forum.nasaspaceflight.com/index.php?topic=56871.0 (https://forum.nasaspaceflight.com/index.php?topic=56871.0)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Redclaws on 08/04/2022 01:21 am
This is really, really, really off topic.  Spectacularly so.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/07/2022 10:10 pm
Does anybody have an estimate for how much solar wing and radiator area it would take just to have a depot with bare metal tanks and a brute-force cryocooler that could keep up with VLEO levels of boiloff?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/10/2022 02:57 am
Does anybody have an estimate for how much solar wing and radiator area it would take just to have a depot with bare metal tanks and a brute-force cryocooler that could keep up with VLEO levels of boiloff?
Somebody (I thought it was you) said the power to run a cryo cooler would be 2-3 times the power needed to evaporate the propellant. Figure the sun is putting out 1400 W/m^2 at a 55-60% duty cycle. Worst case it's broiling the stainless half the ship. Are we talking about a ship with a heat shield? More options but still nothing worse than stainless flat on to the sun. Stainless has sucko reflectance (I'm not even sure of the term so I can't even look it up) and sucko thermal transmission for a metal.


To add only a minor inconvenience to this indeterminant mess the earth will throw back heat that varies from day to night and between landmass and water, plus's a few other variables like snow and cloud cover, rock vs forest etc. Maybe NOAA or NASA has numbers somewhere. If I remember, I'll look for it tomorrow. Bedtime now.


Oh, one other thing. ZBO or something less?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/10/2022 04:54 am
Does anybody have an estimate for how much solar wing and radiator area it would take just to have a depot with bare metal tanks and a brute-force cryocooler that could keep up with VLEO levels of boiloff?
Somebody (I thought it was you) said the power to run a cryo cooler would be 2-3 times the power needed to evaporate the propellant. Figure the sun is putting out 1400 W/m^2 at a 55-60% duty cycle. Worst case it's broiling the stainless half the ship. Are we talking about a ship with a heat shield? More options but still nothing worse than stainless flat on to the sun. Stainless has sucko reflectance (I'm not even sure of the term so I can't even look it up) and sucko thermal transmission for a metal.


To add only a minor inconvenience to this indeterminant mess the earth will throw back heat that varies from day to night and between landmass and water, plus's a few other variables like snow and cloud cover, rock vs forest etc. Maybe NOAA or NASA has numbers somewhere. If I remember, I'll look for it tomorrow. Bedtime now.


Oh, one other thing. ZBO or something less?

Let's assume ZBO.

First, even in earth orbit, you can point the nose of the tanker/depot at the sun, which effectively puts the main tanks out of reach of direct heating.  There will obviously be some re-radiation from the nose onto the top of the LCH4 tank, but it's going to be dramatically reduced.  This will require some small amount of propellant, because tidal forces want the nose to point toward the center of the Earth, but I'd guess that the amount needed over a refueling campaign would be tiny.

That leaves you with albedo heating from the Earth, which is non-trivial but nothing like direct insolation.

I'm not sure what you're driving at comparing the cryocooler power to the evaporation.  If you're saying that evaporative cooling is more efficient, that may be true--but since we're assuming near-ZBO, it's probably irrelevant.

TPS tiles should have better-than-average emissivity and of course ridiculously low thermal conductivity, but I doubt that helps very much in steady state.  It'll help a bit if you can arrange to have the tiles pointed at Earth on the daylit side and pointed at space on the night side, allowing the heat that piled up on the daylit side (and didn't get conducted towards the bare metal very much, raising the tile surface temperature) to partially radiate away before the next orbit.  But I'd think that this effect would be fairly modest.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/10/2022 09:56 am
but still nothing worse than stainless flat on to the sun. Stainless has sucko reflectance (I'm not even sure of the term so I can't even look it up) and sucko thermal transmission for a metal.

Reflectance is the term. Stainless is typically around 50%, not great, but there's plenty of worse materials. However, not-very-special white paint might be between 75-85%, compared to polished silver at around 90%. There are special whites that exceed 90%. And various "whitest whites" that the internet loves vying for the momentary record, but they are notoriously fragile. Nonetheless, you should be able to find a suitable white coating that gets 85% reflectance.

Of course, highly reflective materials tend to have awful emissivity, so hold onto their heat once they have it. But there'll be an optimal paint with the right balance of high reflectance and high emissivity.

I assume the trade tilts wildly towards preferring reflectance over emissivity for any sun-facing side.

So, if a depot doesn't need to re-enter, you should expect at least a white coating. And ideally, IMO, a good white coating over a material with low thermal conductivity. Painting the regular heat tiles might be a quick'n'dirty way to get both. But since you don't need a material that can survive the temperatures of re-entry, there's probably vastly better options for thermal tiles that are robust enough to survive launch. Low conductivity lets you keep the heat on the surface material and let it radiate away on the Earth's night-side before it penetrates to the tank.

To add only a minor inconvenience to this indeterminant mess the earth will throw back heat that varies from day to night and between landmass and water, plus's a few other variables like snow and cloud cover, rock vs forest etc. Maybe NOAA or NASA has numbers somewhere. If I remember, I'll look for it tomorrow. Bedtime now.

Earth's albedo is around 0.3 (30%) and the depot will be in an orbit where it's roughly 50% of the sky. There's no reason to try to get any more precise than that, we lack the necessary details about Starship's thermal properties to calculate anything useful, anyway.

Oh, one other thing. ZBO or something less?

That's a decision you make after you've worked out the thermal properties of Starship.

Similarly, you need to know how much heat you are getting rid of in order to work out radiator area, to know whether you can rely on flat-panel radiator on (or built into) the body, or will need to deploy "wings". And hence where you would put the radiators.




First, even in earth orbit, you can point the nose of the tanker/depot at the sun, which effectively puts the main tanks out of reach of direct heating.

Reduces exposed area from roughly 450m² (50x9m, ignoring the nose taper) down to roughly 65m² (ignoring the sun's half-degree "width".) A 7-fold reduction while also keeping that direct heat away from the tank walls.

It also means you can tune each of the three surfaces (nose, dorsal, ventral) for three separate tasks. Make the nose highly reflective, pointed to the sun. Angle the ventral, heat-shield side to Earth and give it a reflective coating, while the low thermal conductivity of the tiles (or a substitute) provide insulation. Point the dorsal side away from both the sun and Earth, with a highly transmissive and highly emissive coating, to help naturally radiate away heat from the tanks to the deep-space background.

And you get all of that before you need to worry about any multi-layered thermal shielding, cryo-cooling, etc. Reducing the amount of energy they have to deal with.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/10/2022 10:16 am
Does anybody have an estimate for how much solar wing and radiator area it would take just to have a depot with bare metal tanks and a brute-force cryocooler that could keep up with VLEO levels of boiloff?
Somebody (I thought it was you) said the power to run a cryo cooler would be 2-3 times the power needed to evaporate the propellant. Figure the sun is putting out 1400 W/m^2 at a 55-60% duty cycle. Worst case it's broiling the stainless half the ship. Are we talking about a ship with a heat shield? More options but still nothing worse than stainless flat on to the sun. Stainless has sucko reflectance (I'm not even sure of the term so I can't even look it up) and sucko thermal transmission for a metal.


To add only a minor inconvenience to this indeterminant mess the earth will throw back heat that varies from day to night and between landmass and water, plus's a few other variables like snow and cloud cover, rock vs forest etc. Maybe NOAA or NASA has numbers somewhere. If I remember, I'll look for it tomorrow. Bedtime now.


Oh, one other thing. ZBO or something less?

Let's assume ZBO.

First, even in earth orbit, you can point the nose of the tanker/depot at the sun, which effectively puts the main tanks out of reach of direct heating.  There will obviously be some re-radiation from the nose onto the top of the LCH4 tank, but it's going to be dramatically reduced.  This will require some small amount of propellant, because tidal forces want the nose to point toward the center of the Earth, but I'd guess that the amount needed over a refueling campaign would be tiny.

That leaves you with albedo heating from the Earth, which is non-trivial but nothing like direct insolation.

I'm not sure what you're driving at comparing the cryocooler power to the evaporation.  If you're saying that evaporative cooling is more efficient, that may be true--but since we're assuming near-ZBO, it's probably irrelevant.

TPS tiles should have better-than-average emissivity and of course ridiculously low thermal conductivity, but I doubt that helps very much in steady state.  It'll help a bit if you can arrange to have the tiles pointed at Earth on the daylit side and pointed at space on the night side, allowing the heat that piled up on the daylit side (and didn't get conducted towards the bare metal very much, raising the tile surface temperature) to partially radiate away before the next orbit.  But I'd think that this effect would be fairly modest.
I was looking for upper bounds, worst case, which would be tank wall to the sun. Nose on would be the way to go in the real world. I wonder if some multilayer insulation on the tank top would be of benefit. Much of the nose has tiles so there's some help there. More on this below.


That 2-3x number is something I read in these forums. AIUI, if the propellant gets 1kW of heating it will take 2-3kW of power to remove that heat and IIRC, it was presented as a rule of thumb and included real world inefficiencies. I lack the personal tools needed to verify this but would very much like to know it's validity.


The orientation of the tiles in relation to the earth is not straightforward. Picture earth at the origin,  the ship at the 12 o'clock position and the sun also at 12 but much further out. The nose is pointed towards the sun, the tail towards earth. The tiles can point anywhere because they're not in play. At the 3 and 9 o'clock position the nose is still pointed at the sun, the tail is pointed off into space and the ship orients on its roll axis to point the tiles towards earth. At the six o'clock the nose is still pointing towards the sun but the earth is in the way, the tail is pointing into space and again, the tiles are free to radiate in whatever direction they want. I fantasize a tile glaze with the properties of thermal white paint.


Earth's thermal input is nowhere near that of the sun but it is not insignificant. There's that specular reflection from the oceans. Then there is good reflection from snow and clouds. Emissivity varies widely. The bare red soil of Arizona reemits much more than the Amazon. Throw in general greenhouse and local greenhouse caused by humidity and it's a very complex model. Especially for VLEO where the ground view is highly localized.


A complete refueling campaign for one client ship could conceivably be done in a few days. Two launch sites each launching every 12 hours. Hookup, transfer, unhooking then clearing the area every six hours. Boiloff probably wouldn't be much of a problem. Realistically, two weeks sounds optimistic for first efforts.


With the tankers broiling in the sun during approach and severe attitude restrictions during transfer, I can't imagine boiloff not being a problem.


IIUC correctly, tidal forces will not only draw one end of the ship towards the earth but because of gravitational irregularities, the ship will end up tumbling. If a depot is a one campaign accumulator intended to return to earth for reuse, using thrust to counter this would be the way to go. If it's a long term facility, CMG's or reaction wheels would make more sense.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/10/2022 01:28 pm
but still nothing worse than stainless flat on to the sun. Stainless has sucko reflectance (I'm not even sure of the term so I can't even look it up) and sucko thermal transmission for a metal.

Reflectance is the term. Stainless is typically around 50%, not great, but there's plenty of worse materials. However, not-very-special white paint might be between 75-85%, compared to polished silver at around 90%. There are special whites that exceed 90%. And various "whitest whites" that the internet loves vying for the momentary record, but they are notoriously fragile. Nonetheless, you should be able to find a suitable white coating that gets 85% reflectance.

Of course, highly reflective materials tend to have awful emissivity, so hold onto their heat once they have it. But there'll be an optimal paint with the right balance of high reflectance and high emissivity.

I assume the trade tilts wildly towards preferring reflectance over emissivity for any sun-facing side.

So, if a depot doesn't need to re-enter, you should expect at least a white coating. And ideally, IMO, a good white coating over a material with low thermal conductivity. Painting the regular heat tiles might be a quick'n'dirty way to get both. But since you don't need a material that can survive the temperatures of re-entry, there's probably vastly better options for thermal tiles that are robust enough to survive launch. Low conductivity lets you keep the heat on the surface material and let it radiate away on the Earth's night-side before it penetrates to the tank.

To add only a minor inconvenience to this indeterminant mess the earth will throw back heat that varies from day to night and between landmass and water, plus's a few other variables like snow and cloud cover, rock vs forest etc. Maybe NOAA or NASA has numbers somewhere. If I remember, I'll look for it tomorrow. Bedtime now.

Earth's albedo is around 0.3 (30%) and the depot will be in an orbit where it's roughly 50% of the sky. There's no reason to try to get any more precise than that, we lack the necessary details about Starship's thermal properties to calculate anything useful, anyway.

Oh, one other thing. ZBO or something less?

That's a decision you make after you've worked out the thermal properties of Starship.

Similarly, you need to know how much heat you are getting rid of in order to work out radiator area, to know whether you can rely on flat-panel radiator on (or built into) the body, or will need to deploy "wings". And hence where you would put the radiators.




First, even in earth orbit, you can point the nose of the tanker/depot at the sun, which effectively puts the main tanks out of reach of direct heating.

Reduces exposed area from roughly 450m² (50x9m, ignoring the nose taper) down to roughly 65m² (ignoring the sun's half-degree "width".) A 7-fold reduction while also keeping that direct heat away from the tank walls.

It also means you can tune each of the three surfaces (nose, dorsal, ventral) for three separate tasks. Make the nose highly reflective, pointed to the sun. Angle the ventral, heat-shield side to Earth and give it a reflective coating, while the low thermal conductivity of the tiles (or a substitute) provide insulation. Point the dorsal side away from both the sun and Earth, with a highly transmissive and highly emissive coating, to help naturally radiate away heat from the tanks to the deep-space background.

And you get all of that before you need to worry about any multi-layered thermal shielding, cryo-cooling, etc. Reducing the amount of energy they have to deal with.



Found an interesting paper from 2020. [size=78%]https://tfaws.nasa.gov/wp-content/uploads/TFAWS2020-CT-103-Wilhite-Paper.pdf (https://tfaws.nasa.gov/wp-content/uploads/TFAWS2020-CT-103-Wilhite-Paper.pdf)[/size]


It's about a newer high thermal rejection coating. Among the interesting points is a theoretical 99.9% rejection and a treated sphere at 1 AU could theoretically reach equilibrium at <50K. It postulates a 10mm thickness which I doubt would survive EDL, which would rule out return and relaunch. This paper may have been quoted earlier in this thread or another on NSF.


You make an excellent point in that the ships thermal properties need to be known before focusing on how good is good enough. We sometimes get ahead of ourselves here.


The average albedo OTOH, is not so helpful. VLEO gives such a restricted view of the surface that worst case needs to be explored. It's not impossible that some areas will be so far from the average that mitigation aimed at the average will still allow significant boil off. This may not be, but it should be looked at.


The question of a tanker/accumulator vs depot heavy is one I've sunk my teeth into. The accumulator would work well at VLEO. Depot Heavy would need to be higher. The higher up, the less the earth fills the sky and the more the average of the albedo dominates local extremes. Soooo many moving parts.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 08/10/2022 02:14 pm
Remember that as well as active cryocoolers and/or added passive heat rejection mechanisms (deployable sunshields, non-reenterable coatings and MLI, etc), both also need to trade against the no-part option of launch more propellant and let any boil-off boil off. Depending on loiter time and final prop load required (which will not always be a full load), that will not necessarily even require an additional launch, and instead mean greater prop loads for any tanker flight(s) topping up the accumulation tanker/depot.

At some point down the road you may end up with enough propellant in orbit over a long enough period of time that the cost of boiloff trades poorly against the cost of developing and operating a low boiloff system. But I suspect that trade point is far further off than the first few years of operation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/10/2022 03:49 pm
Remember that as well as active cryocoolers and/or added passive heat rejection mechanisms (deployable sunshields, non-reenterable coatings and MLI, etc), both also need to trade against the no-part option of launch more propellant and let any boil-off boil off. Depending on loiter time and final prop load required (which will not always be a full load), that will not necessarily even require an additional launch, and instead mean greater prop loads for any tanker flight(s) topping up the accumulation tanker/depot.

At some point down the road, you may end up with enough propellant in orbit over a long enough period of time that the cost of boiloff trades poorly against the cost of developing and operating a low boiloff system. But I suspect that trade point is far further off than the first few years of operation.
I agree that it's worth at least knowing how bad the boiloff problem is if you do nothing at all, but, as a practical matter, it would seem reasonable to at least do the cheap things to minimize boiloff in the vehicle that accumulates the propellants. That is, keep the nose pointed at the sun, rotate it so the heat shield faces the Earth, cover the nose with Solar White (the paint version (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf), not the tile version), etc. Then see how far it falls short.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 08/10/2022 03:56 pm
Remember that as well as active cryocoolers and/or added passive heat rejection mechanisms (deployable sunshields, non-reenterable coatings and MLI, etc), both also need to trade against the no-part option of launch more propellant and let any boil-off boil off. Depending on loiter time and final prop load required (which will not always be a full load), that will not necessarily even require an additional launch, and instead mean greater prop loads for any tanker flight(s) topping up the accumulation tanker/depot.

At some point down the road, you may end up with enough propellant in orbit over a long enough period of time that the cost of boiloff trades poorly against the cost of developing and operating a low boiloff system. But I suspect that trade point is far further off than the first few years of operation.
I agree that it's worth at least knowing how bad the boiloff problem is if you do nothing at all, but, as a practical matter, it would seem reasonable to at least do the cheap things to minimize boiloff in the vehicle that accumulates the propellants. That is, keep the nose pointed at the sun, rotate it so the heat shield faces the Earth, cover the nose with Solar White (the paint version (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf), not the tile version), etc. Then see how far it falls short.
Why would Depot have a TPS tiles at all rather than a specialized shield? It will never EDL. IF you are using a Tanker as a Depot, then you don't have a long-term storage problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 08/10/2022 04:03 pm
Remember that as well as active cryocoolers and/or added passive heat rejection mechanisms (deployable sunshields, non-reenterable coatings and MLI, etc), both also need to trade against the no-part option of launch more propellant and let any boil-off boil off. Depending on loiter time and final prop load required (which will not always be a full load), that will not necessarily even require an additional launch, and instead mean greater prop loads for any tanker flight(s) topping up the accumulation tanker/depot.

At some point down the road, you may end up with enough propellant in orbit over a long enough period of time that the cost of boiloff trades poorly against the cost of developing and operating a low boiloff system. But I suspect that trade point is far further off than the first few years of operation.
I agree that it's worth at least knowing how bad the boiloff problem is if you do nothing at all, but, as a practical matter, it would seem reasonable to at least do the cheap things to minimize boiloff in the vehicle that accumulates the propellants. That is, keep the nose pointed at the sun, rotate it so the heat shield faces the Earth, cover the nose with Solar White (the paint version (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf), not the tile version), etc. Then see how far it falls short.
Why would Depot have a TPS tiles at all rather than a specialized shield? It will never EDL. IF you are using a Tanker as a Depot, then you don't have a long-term storage problem.
A non-returning depot is a premature optimisation, and an unnecessary expenditure of a Starship. There is currently only one mission profile (HLS) that might require more propellant than could be carried by a standard Starship, it would be flying at most once per year if that, and could also be serviced with two regular sized Starships. Returning the Starship used as a depot means you get to use it again for something else rather than it floating in orbit doing nothing for a year or more (assuming it even has sufficient in-orbit endurance to do so).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 08/10/2022 05:31 pm

The average albedo OTOH, is not so helpful. VLEO gives such a restricted view of the surface that worst case needs to be explored. It's not impossible that some areas will be so far from the average that mitigation aimed at the average will still allow significant boil off. This may not be, but it should be looked at.

The thermal time constants will matter here. 

If they are several days, or hours, an average should be reasonably good.  The correct average to use might not be the arithmetic average, and it might not be over the entire Earth, but there should be a single number rather than having to model the ground in detail.

OTOH if the time constants are seconds then a pass over the worst case spot could be significant.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 08/10/2022 05:31 pm
IIUC correctly, tidal forces will not only draw one end of the ship towards the earth but because of gravitational irregularities, the ship will end up tumbling. If a depot is a one campaign accumulator intended to return to earth for reuse, using thrust to counter this would be the way to go. If it's a long term facility, CMG's or reaction wheels would make more sense.
When working the dynamics remember that a tank full of fluid is not a rigid body.  A tank half full of liquid is even less rigid.  For a tanker some effects will be orders of magnitude greater than for a rigid body.

I'm not sure exactly which effects will dominate, but I have a difficult time envisioning the nose pointing to the sun while it slowly rotates around the long axis to keep one side towards Earth.  It really wants to enter an end over end tumble. It's going to take work to stop that.

(Also all the attitude control work ends up heating the fluid -- probably not significant but somebody should probably calculate that to confirm it).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/10/2022 05:58 pm
Remember that as well as active cryocoolers and/or added passive heat rejection mechanisms (deployable sunshields, non-reenterable coatings and MLI, etc), both also need to trade against the no-part option of launch more propellant and let any boil-off boil off. Depending on loiter time and final prop load required (which will not always be a full load), that will not necessarily even require an additional launch, and instead mean greater prop loads for any tanker flight(s) topping up the accumulation tanker/depot.

At some point down the road you may end up with enough propellant in orbit over a long enough period of time that the cost of boiloff trades poorly against the cost of developing and operating a low boiloff system. But I suspect that trade point is far further off than the first few years of operation.
I suspect you're right on the timeline, but it will happen.


Once refueling is shown to work I expect it to take off, so to speak. Give it 3-5 years and infrastructure, not all of it SX, will start popping up. I'd expect it to be commercial sat servicing for the most part. Probably hypergolics to start with. 3-5 years is long enough to get refueling built into new sat launches.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/10/2022 06:10 pm
Remember that as well as active cryocoolers and/or added passive heat rejection mechanisms (deployable sunshields, non-reenterable coatings and MLI, etc), both also need to trade against the no-part option of launch more propellant and let any boil-off boil off. Depending on loiter time and final prop load required (which will not always be a full load), that will not necessarily even require an additional launch, and instead mean greater prop loads for any tanker flight(s) topping up the accumulation tanker/depot.

At some point down the road you may end up with enough propellant in orbit over a long enough period of time that the cost of boiloff trades poorly against the cost of developing and operating a low boiloff system. But I suspect that trade point is far further off than the first few years of operation.

If you're only interested in a depot in LEO, the slop with the last lift tanker launch may easily cover the boiloff.  However, you also need to deal with cislunar refueling, where the way the delta-v budget works out gives you extremely slim margins, even with a full 1500t tanker taking prop to NRHO for an LSS.

Given that you want to recover that tanker (otherwise, you expend one for every lunar mission), you're probably restricted to bare metal tanks and TPS.  But then that makes it essential minimize transit boiloff without solar white coatings, sunshields, etc.  If you only have passive cooling, then you're in a race as to whether you lose more prop doing a 40-90 day BLT and saving 350m/s of delta-v but allowing a longer time for boiloff to occur, or doing the fast transit to NRHO and spending that delta-v.  Even a fairly modest amount of cryocooling solves a lot of problems here.

Also, don't forget that the LSS itself has to store prop for 60-90 days in NRHO, as well as enduring the daylight surface temperatures near the poles.  I'm sure that passive cooling will play some role in the LSS propellant management, but will it be enough to hold boiloff to a reasonable number to meet the requirements?  I wouldn't be surprised to see some kind of cryocooling built into even the Option A LSS.  And if that tech's available for the LSS, then it's likely available for tankers/depots.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/10/2022 06:21 pm
IIUC correctly, tidal forces will not only draw one end of the ship towards the earth but because of gravitational irregularities, the ship will end up tumbling. If a depot is a one campaign accumulator intended to return to earth for reuse, using thrust to counter this would be the way to go. If it's a long term facility, CMG's or reaction wheels would make more sense.
When working the dynamics remember that a tank full of fluid is not a rigid body.  A tank half full of liquid is even less rigid.  For a tanker some effects will be orders of magnitude greater than for a rigid body.

I'm not sure exactly which effects will dominate, but I have a difficult time envisioning the nose pointing to the sun while it slowly rotates around the long axis to keep one side towards Earth.  It really wants to enter an end over end tumble. It's going to take work to stop that.

(Also all the attitude control work ends up heating the fluid -- probably not significant but somebody should probably calculate that to confirm it).

A fair point.  But I think the x-axis rotation (to keep TPS pointed at daylit Earth and space on the night side) could be omitted if the dynamics get too hairy.  The first-order requirement is to keep direct sun off of the tanks as much as possible, by pointing the nose at the sun.

It'd be interesting to look at the tidal effects on the propellant if the nose is pointed in some direction other than radially wrt the Earth.  Are they strong enough to force prop into two blobs, or are they small enough that they can't overcome surface tension in half an orbit?  What does that do to the dynamics of the tanker/depot itself?  How does that change based on the amount of prop in the depot?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/10/2022 06:30 pm
The question of a tanker/accumulator vs depot heavy is one I've sunk my teeth into. The accumulator would work well at VLEO. Depot Heavy would need to be higher. The higher up, the less the earth fills the sky and the more the average of the albedo dominates local extremes. Soooo many moving parts.

Two big reasons to stay in VLEO:

1) It's more prop-efficient as an overall refueling architecture.

2) The chance of a depot/tanker accident has to be at least an order of magnitude higher than other spacecraft, and it can generate an awful lot of debris.  Even if it just failed passively and couldn't deorbit, it's a pretty big hazard.  Getting it to passively deorbit in a few months is probably pretty high on the requirements list.

This is yet another reason why I like plain ol' tankers with a depot kit, as opposed to non-EDL depots, and why I think it's better to bring down the accumulator when its mission is done.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/10/2022 06:35 pm
The thermal time constants will matter here. 

If they are several days, or hours, an average should be reasonably good.  The correct average to use might not be the arithmetic average, and it might not be over the entire Earth, but there should be a single number rather than having to model the ground in detail.

OTOH if the time constants are seconds then a pass over the worst case spot could be significant.

Since you're trying to minimize the flux into hundreds or even thousands of tonnes of propellant, I'd guess that an average will be just fine.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/10/2022 06:41 pm
Given that you want to recover that tanker (otherwise, you expend one for every lunar mission), you're probably restricted to bare metal tanks and TPS.  But then that makes it essential minimize transit boiloff without solar white coatings, sunshields, etc. 
Table 1 in the article I linked to above (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf) said that applying 5 mils (about 1/8 mm) of Solar White paint on top of stainless steel reduced absorption of solar energy from 53% to 8.5%. That seems like a very fine improvement for very little cost. Especially if you only need to paint the nose.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/10/2022 07:26 pm
Given that you want to recover that tanker (otherwise, you expend one for every lunar mission), you're probably restricted to bare metal tanks and TPS.  But then that makes it essential minimize transit boiloff without solar white coatings, sunshields, etc. 
Table 1 in the article I linked to above (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf) said that applying 5 mils (about 1/8 mm) of Solar White paint on top of stainless steel reduced absorption of solar energy from 53% to 8.5%. That seems like a very fine improvement for very little cost. Especially if you only need to paint the nose.

Well, more than half the nose of any EDL-capable Starship is TPS, which you almost certainly can't paint.  The question would be whether to clean off the solar white after EDL would be worth it, vs. just soaking up the rays.  Remember, the nose for a tanker is kinda like the outside of a dewar.  Only a fraction of the heat that's radiated inward, into the payload bay, will warm the top of the LCH4 tank. 

I'll bet a hunk of MLI in the payload bay to cover the dome would be vastly easier to engineer than figuring out how to paint the nose.  The paint won't survive entry.  I have no clue what scorching it would do to the heat distribution during EDL.  Probably nothing good.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/10/2022 08:34 pm
Well, more than half the nose of any EDL-capable Starship is TPS, which you almost certainly can't paint. 
There's another problem: Whatever you paint, it needs to be reflective, so painting the black TPS isn't going to work in the first place. I think that eliminates this idea--at least for a regular Starship doing depot duty. If the lunar Starship isn't going to have tiles in the first place, though, then maybe it does make sense to paint the stainless steel with Solar White so the propellant will last longer. Assuming it's determined that just pointing the nose at the sun isn't good enough for the time it takes to refuel it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/11/2022 12:47 am
Given that you want to recover that tanker (otherwise, you expend one for every lunar mission), you're probably restricted to bare metal tanks and TPS.  But then that makes it essential minimize transit boiloff without solar white coatings, sunshields, etc. 
Table 1 in the article I linked to above (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf) said that applying 5 mils (about 1/8 mm) of Solar White paint on top of stainless steel reduced absorption of solar energy from 53% to 8.5%. That seems like a very fine improvement for very little cost. Especially if you only need to paint the nose.

Well, more than half the nose of any EDL-capable Starship is TPS, which you almost certainly can't paint.  The question would be whether to clean off the solar white after EDL would be worth it, vs. just soaking up the rays.  Remember, the nose for a tanker is kinda like the outside of a dewar.  Only a fraction of the heat that's radiated inward, into the payload bay, will warm the top of the LCH4 tank. 

I'll bet a hunk of MLI in the payload bay to cover the dome would be vastly easier to engineer than figuring out how to paint the nose.  The paint won't survive entry.  I have no clue what scorching it would do to the heat distribution during EDL.  Probably nothing good.
In the solar white article I linked a page or two back, the active ingredient in the paint is yttrium oxide. The outer surface of the tiles is black vitrification. In other words, a glaze, and glazes are routinely pigmented. Is there a chance the black is choice, not chance? It would be very Elon like.


Yttrium oxide is considered to be rare earth like if not actually a rare earth. Rare earths are used as glaze pigments. If yttrium oxide gives solar white properties to the tiles as a glaze (a total unknown - at least to me) and it doesn't do anything wonky, it sounds as if it would be good for the tiles in general, not just for cryo insulation.


I hate when there are that many if's. I know next to nothing about pottery.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/11/2022 01:41 am

Just found a paywall Ed paper on Y2O3 that hits on its chromatic behavior when added to Zirconia in varying concentrations. Most of the meat is hidden and what is public strains my level of understanding. The references shown may point to something on point for heat tile glaze for those with the skill and interest.


One takeaway: Y2O3 exhibits infrared photoluminescence when laser pumped which explains its high emissivity in the IR. 


https://www.sciencedirect.com/science/article/abs/pii/S0749603608002760 (https://www.sciencedirect.com/science/article/abs/pii/S0749603608002760)


My search also also hit some Chinese patients using Y2O3 as a low concentration element in powdered glass glazes but nothing specifically as a pigment. The text appeared to be google translated and was a trial to understand.


Edit: paywalled, not paywall Ed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/11/2022 06:25 am
The average albedo OTOH, is not so helpful. VLEO gives such a restricted view of the surface that worst case needs to be explored. It's not impossible that some areas will be so far from the average that mitigation aimed at the average will still allow significant boil off. This may not be, but it should be looked at.

My point was that we're so far away from understanding the thermal properties of Starship, let alone a depot, that the difference between assuming 0.3 global albedo and the specific time-variant radiance from a specific (but unknown) path over the Earth are not even within our rounding-off-errors.

It's like we're booking an appointment 50 years away, somewhere within a random 24-month window, and you're asking whether "Morning or afternoon?" will be better for weather and traffic.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 08/11/2022 12:56 pm
The outer surface of the tiles is black vitrification. In other words, a glaze, and glazes are routinely pigmented. Is there a chance the black is choice, not chance? It would be very Elon like.
The outer borosilciate glass surface is very functional: it's what rejects the majority of re-entry energy via radiation, and what protects the vehicle from chemical attack from the plasma. The sintered silica tile body behind it helps prevent heat conducted from the extremely hot coating from reaching the vehicle skin (and acts as a substrate for the coating) but that coating is the vital component for surviving re-entry.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 08/11/2022 03:17 pm
The outer surface of the tiles is black vitrification. In other words, a glaze, and glazes are routinely pigmented. Is there a chance the black is choice, not chance? It would be very Elon like.

There's a reason they call it "Black Body Radiation"

The black maximizes the Stefan-Boltzmann coefficient, which is the primary method of removing heat from a surface that can reach 1300degC.  The equation has a term where the energy moved is proportional to temperature change to the 4th power, times that coefficient.   Stainless steel is 0.35, the black borosilicate about 0.95 or better.   The black paint they added behind the forward flaps by itself improves the coefficient to above 0.9 (but without all other benefits of the tile)

Maybe this will help you if you your depot spends half time in the shade and half time in the sun in LEO.  The half time in the shade allows the black surface to maximize the emission of the energy gained while being in the sun, while the time in the sun heats up the tiles (which by themselves have about 10MJ / square meter heat capacity, if I'm remembering the math correctly), with the tiles and the mineral fiber mat underneath acting as insulation.

You'd have to do actual math to figure out if this is better or worse than reflecting the energy from the Sun before it is absorbed.  There's a lot of that kind of math on the heat shield thread.

The tiles themselves were designed to be bathed in intense heat for about 30 minutes and are allowed to pass a lot of that heat through to the interior as long as the skin doesn't exceed 800degC or so.  Not exactly what you are looking for, but your scenario has vastly different things happening.

If your depot needs to polar orbit the moon, there is no shade, so I doubt the black surface would be better than a sun shade or white surface.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/11/2022 04:46 pm
The outer surface of the tiles is black vitrification. In other words, a glaze, and glazes are routinely pigmented. Is there a chance the black is choice, not chance? It would be very Elon like.

There's a reason they call it "Black Body Radiation"

The black maximizes the Stefan-Boltzmann coefficient, which is the primary method of removing heat from a surface that can reach 1300degC.  The equation has a term where the energy moved is proportional to temperature change to the 4th power, times that coefficient.   Stainless steel is 0.35, the black borosilicate about 0.95 or better.   The black paint they added behind the forward flaps by itself improves the coefficient to above 0.9 (but without all other benefits of the tile)

Maybe this will help you if you your depot spends half time in the shade and half time in the sun in LEO.  The half time in the shade allows the black surface to maximize the emission of the energy gained while being in the sun, while the time in the sun heats up the tiles (which by themselves have about 10MJ / square meter heat capacity, if I'm remembering the math correctly), with the tiles and the mineral fiber mat underneath acting as insulation.

You'd have to do actual math to figure out if this is better or worse than reflecting the energy from the Sun before it is absorbed.  There's a lot of that kind of math on the heat shield thread.

The tiles themselves were designed to be bathed in intense heat for about 30 minutes and are allowed to pass a lot of that heat through to the interior as long as the skin doesn't exceed 800degC or so.  Not exactly what you are looking for, but your scenario has vastly different things happening.

If your depot needs to polar orbit the moon, there is no shade, so I doubt the black surface would be better than a sun shade or white surface.
The whole point of the solar white paint is that it's highly reflective at visible wavelengths and highly emissive in the infrared. Being white in the visible, it rejects much of that input. What it doesn't reject becomes heat. This in turn is emitted in the infrared where, if we could see in the infrared, the paint would normally appear black.


The active component of solar white paint in the 2020 paper is Y2O3. Another paper I linked mentioned Y2O3 as fluorescing in the infrared when pumped by a laser. This makes me suspect that at least part of the infrared emission isn't linked directly to the usual black body considerations.


The energy input/output that follows somewhat more conventional black body considerations would, if examined within a restricted IR range, appear to be much hotter than what a perfect black body would normally exhibit. Within that IR emission range Mr Stefan and Mr Boltzmann become even more your friends than usual.


These are the properties I was noodling for the tile glaze. It would appear white in the visible. Elon likes white.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/11/2022 06:53 pm
I did a plot of black-body radiation distribution from 0 to 8 microns for the Starship tiles (1650 K) and the surface of the sun (5778 K). I picked 8μ because Solar White reflects light for shorter wavelengths (down to 0.25μ, below which neither the sun nor the tiles emit much of anything). It's pretty amazing because it reflects all but 0.2% of the sun's light.

Unfortunately, this range is so broad that it's only going to have 5% emissivity for the tiles. That pretty much breaks them completely.

You can't compromise and use something that's a bit less reflective because then it won't keep cryogenics cold. So I think this completely rules out the idea of painting the heat-resistant tiles with something like Solar White. It just won't work.

Note: numbers on the x-axis are meters.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/11/2022 08:12 pm
One takeaway: Y2O3 exhibits infrared photoluminescence when laser pumped which explains its high emissivity in the IR. 

I wouldn't get too hung up on the details, because I doubt that this is exactly what SpaceX will use on the LSS.

My dumb explanation of solar white:

1) It's really reflective in UV, visible light, and mid-infrared.
2) It's really emissive in low infrared.

It's the combination of these two properties that makes it effective.

Edit:  Hopelessly ninja'd.  That's what happens when you don't pay attention to the "you might want to recheck the thread" warning after leaving something sitting overnight.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 08/11/2022 11:07 pm
I did a plot of black-body radiation distribution from 0 to 8 microns for the Starship tiles (1650 K) and the surface of the sun (5778 K). I picked 8μ because Solar White reflects light for shorter wavelengths (down to 0.25μ, below which neither the sun nor the tiles emit much of anything). It's pretty amazing because it reflects all but 0.2% of the sun's light.

Unfortunately, this range is so broad that it's only going to have 5% emissivity for the tiles. That pretty much breaks them completely.

You can't compromise and use something that's a bit less reflective because then it won't keep cryogenics cold. So I think this completely rules out the idea of painting the heat-resistant tiles with something like Solar White. It just won't work.

Note: numbers on the x-axis are meters.

Nice plot,  Can you include an estimate for the solar white paint?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/12/2022 12:50 am
One takeaway: Y2O3 exhibits infrared photoluminescence when laser pumped which explains its high emissivity in the IR. 

I wouldn't get too hung up on the details, because I doubt that this is exactly what SpaceX will use on the LSS.

My dumb explanation of solar white:

1) It's really reflective in UV, visible light, and mid-infrared.
2) It's really emissive in low infrared.

It's the combination of these two properties that makes it effective.

Edit:  Hopelessly ninja'd.  That's what happens when you don't pay attention to the "you might want to recheck the thread" warning after leaving something sitting overnight.


Ahhh, sometimes getting down in the weeds is fun. You posted this 9 page paper: [size=78%]https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf)[/size]


Just finished it. Excellent read. The point that really jumped out at me was that Y2O3 is hydrophobic.


Don't get me wrong. I'm not doing a 'Y2O3 will fix all the problems' type of thing but there are hints that it might be nearly as good as sliced bread. Many unresolved issues too.


On the plus side:
- They are exploring both paint and ceramic tiles with the understanding that the investigation is looking at the tiles for cryo purposes, not EDL.
- Y2O3 has promising thermal properties
- by itself, it's expected to stand up well to atomic Oxygen. Testing is planned.
- it's hydrophobic


On the down side:
- the thermal properties are sensitive to the metallic substrate and stainless (specifics unknown) is a poor performing substrate.
- the KBr binder used in the paint is easily dissolved in water.
- the work is preliminary and many questions specific to use in the real world of rockets are yet to be answered. OTOH, there has been at least one sample mounted outside the ISS and parallel work is planned.


There's more, pro and con. It is intriguing but not a slam dunk. Other binders need to be explored. Investigating the ceramic form as a component of an EDL heat shield isn't mentioned. It's worth a deeper dive to see if there's other research going on. Loves them footnotes.





Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/12/2022 02:40 am
Nice plot,  Can you include an estimate for the solar white paint?
At what temperature? Over what wavelengths? Can you better articulate what you're looking for?

The plots above are for black-body radiation. Solar White isn't black-body from 0.25μ to 8μ, but it pretty much is outside that range. Or, at least, that's what the papers claim.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 08/12/2022 01:22 pm
Does anybody have an estimate for how much solar wing and radiator area it would take just to have a depot with bare metal tanks and a brute-force cryocooler that could keep up with VLEO levels of boiloff?

The flux from the Earth day side to ~250m^2 of Starship's stainless tanks side (let's ignore the light falling on the nose cone, skirt or flaps) would be ~35kW. Multiply by 1 - reflectance (0.6) and you have 14kW flux while flying over the day side.

If you point heat shield side towards the Earth, you get 35 * 0.95 ie ~33kW. But this one could be averaged over day-night cycle because of low thermal conductivity of the shield. The day nigt cycle is not 50:50, because Earth shadow intersceting quite a bit less than 50% of the orbit. Also summer/winter time insolation grow with inclination. So the duty cycle is about 70% for low inclinations and get's much worse (>90% for ISS like inclinations). So about 23-30kW if you face the Earth with  heatshield.

So, stainless towards the earth it is. You will have some radiation hitting some heathsield because the earth takes like 1/3 of the sky so it will illuminate the sides, but at a slant angle.

So BOTE result is 15kW to get rid of if you can't average and you have to deal with the momentary peak flux on the day side. If you could average things by for example allowing the pressure to rise on the day side you could get the power to get rid of down to 11-14kW depending on inclination.

I assumed the ship pointing its nose to the Sun and MLI placed in the main bay just above methane tank. I ignored radiative cooling because at 70K the black body power is around 1.25W/m^2 so is negligible. I also ignored night side heating -- the flux would be about 13W per m^2 of Starship side (40W/m^2 from black body at 220K times 1/3 of the sky occupied by it) --  it's be less than 10% of day side light so while not negligible, for BOTE calculations could be ignored.


So it's 11-14kW to get rid of. What's the expected efficiency of practical cryo heat pumps? I suspect it's awful, AFAIR 10-20%. So 55 to 140kW range for the electric power.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/12/2022 03:51 pm
In The Spacecraft Thermal Control Handbook, Chapter 2: Spacecraft Thermal Environments (Clawson, J.F. & Gilmore, David G., 2002) (http://matthewwturner.com/uah/IPT2008_summer/baselines/LOW%20Files/Thermal/Spacecraft%20Thermal%20Control%20Handbook/02.pdf), I think Table 2.2 (p. 27) gives us what we really need to compute the thermal loading caused by the Earth. It gives the "hot case" albedo and IR heating for a hypothetical satellite orbiting just 30 km above the surface. Yeah, that's an impossible altitude, but you can adjust the numbers based on the real orbit of the depot.

Is it safe to assume that we really only care about the average thermal loading over a 24-hour period--given the thermal inertia of the tanks? We don't really care about temporary spikes or anything like that.

In general, this document is rich in information. I think it's well worth a look from anyone who's really interested in this topic.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/12/2022 05:49 pm
So it's 11-14kW to get rid of. What's the expected efficiency of practical cryo heat pumps? I suspect it's awful, AFAIR 10-20%. So 55 to 140kW range for the electric power.

So 150kW would be about 370m² of panels.  That's 15% of an ISS, all to deploy and stow robotically--reliably--if you want to get the Starship back.  Plus, you need the same amount of heat rejection.  Presumably, you deploy that as the back of your solar array.

That's not an easy engineering task.  Not impossible, but not the simplest thing that SpaceX could do (which is what they'll do).  That makes a non-recoverable depot a lot more likely.

PS:

You will have some radiation hitting some heathsield because the earth takes like 1/3 of the sky so it will illuminate the sides, but at a slant angle.

Radiative flux from a sphere to a cylinder has a computable view factor.  It's horrid, as you can see below, but you can use the pretty graph and see that it's going to be about 45% in VLEO.  So that helps a little.

The other computation that we need is how much heat power a particular boiloff rate gobbles up and expels.  This obviously strays away from full ZBO, but if you have some max amount of electrical power in your budget, it should let you compute the boiloff rate.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/13/2022 03:31 am
So BOTE result is 15kW to get rid of if you can't average and you have to deal with the momentary peak flux on the day side. If you could average things by for example allowing the pressure to rise on the day side you could get the power to get rid of down to 11-14kW depending on inclination.

Using 15kW as the worst case, but at 24hr/day. (Ie, 15kJ/second.) And a heat-of-vaporisation figure for methane of 511 kJ/kg, that means boil-off-only cooling would cost 2.5 tonnes of methane per day.

If your refuelling mission cycle takes 30 days from empty, through accumulation, to final target-vehicle transfer and empty again, you lose 75 tonnes of methane. I don't think it's reasonable to assume a faster cycle for a first generation depot.

(Oxygen has worse heat-of-vap (213kJ/kg), so you're only hurting yourself if you let it evaporate. Likewise, proper heat capacity of methane is around 2 kJ/kg per K so ignored at this level of chicken scratchings.)

So, stainless towards the earth it is.

There's no reason to have bare stainless steel. Even if you intend to re-enter the depot-ships, you can paint the dorsal side with something reflective. Just going from your 60% best-case reflectance to a hand-waving 90% for a white coating, and assuming 2/3rds of your 15kW is coming from the bare SS and 1/3rd through the heat-tiles on the sides, that coating alone reduces heat-input by half. And much better if you can also give at least the side-tiles a non-black coating. Between them, coatings alone could get the heat-input down below 4kW. Boil-off-only cooling is thus reduced to below 2/3 tonnes of methane per day, maybe down to half a tonne/day. 30 day mission ops for fuel accumulation costs you 15 tonnes of methane.

And 15 tonnes feels viable for a first generation depot.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/13/2022 04:16 am
Re: Above.

Of course this was only the Earth-side heating. It assumes pointing the nose at the sun effectively eliminates heat-xfer to the tanks. IMO, that's not a given, since the nose isn't reflective and isn't designed for the job of being a sun-shield.




You will have some radiation hitting some heathsield because the earth takes like 1/3 of the sky so it will illuminate the sides, but at a slant angle.
Radiative flux from a sphere to a cylinder has a computable view factor.  It's horrid, as you can see below, but you can use the pretty graph and see that it's going to be about 45% in VLEO.  So that helps a little.

It's going to be complicated because the angle of the ship to Earth isn't constant, but rotates 360 degrees each orbit if the nose points to the sun.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/13/2022 12:16 pm
Re: Above.

Of course this was only the Earth-side heating. It assumes pointing the nose at the sun effectively eliminates heat-xfer to the tanks. IMO, that's not a given, since the nose isn't reflective and isn't designed for the job of being a sun-shield.




You will have some radiation hitting some heathsield because the earth takes like 1/3 of the sky so it will illuminate the sides, but at a slant angle.
Radiative flux from a sphere to a cylinder has a computable view factor.  It's horrid, as you can see below, but you can use the pretty graph and see that it's going to be about 45% in VLEO.  So that helps a little.

It's going to be complicated because the angle of the ship to Earth isn't constant, but rotates 360 degrees each orbit if the nose points to the sun.
Mild confusion on your diagram. It shows the lighter (skin) side towards earth. This works if it's covered with thermal white paint. If not, wouldn't it work better with the tile side earthward?


No, the nose isn't designed as a sun shield but that doesn't mean it can't do less than perfect double duty. Orbit as you show. Much of the nose is already covered in tiles. Top the tank dome in MLI. Not the greatest. An SX first hack to get things working. Improve later.


Edit: added "doesn't"
Edit 2: just reread sbec's  post from 8/7 on stainless vs tiles facing earth. Still scratching my head.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/13/2022 12:31 pm
So BOTE result is 15kW to get rid of if you can't average and you have to deal with the momentary peak flux on the day side. If you could average things by for example allowing the pressure to rise on the day side you could get the power to get rid of down to 11-14kW depending on inclination.

Using 15kW as the worst case, but at 24hr/day. (Ie, 15kJ/second.) And a heat-of-vaporisation figure for methane of 511 kJ/kg, that means boil-off-only cooling would cost 2.5 tonnes of methane per day.

If your refuelling mission cycle takes 30 days from empty, through accumulation, to final target-vehicle transfer and empty again, you lose 75 tonnes of methane. I don't think it's reasonable to assume a faster cycle for a first generation depot.

(Oxygen has worse heat-of-vap (213kJ/kg), so you're only hurting yourself if you let it evaporate. Likewise, proper heat capacity of methane is around 2 kJ/kg per K so ignored at this level of chicken scratchings.)

So, stainless towards the earth it is.

There's no reason to have bare stainless steel. Even if you intend to re-enter the depot-ships, you can paint the dorsal side with something reflective. Just going from your 60% best-case reflectance to a hand-waving 90% for a white coating, and assuming 2/3rds of your 15kW is coming from the bare SS and 1/3rd through the heat-tiles on the sides, that coating alone reduces heat-input by half. And much better if you can also give at least the side-tiles a non-black coating. Between them, coatings alone could get the heat-input down below 4kW. Boil-off-only cooling is thus reduced to below 2/3 tonnes of methane per day, maybe down to half a tonne/day. 30 day mission ops for fuel accumulation costs you 15 tonnes of methane.

And 15 tonnes feels viable for a first generation depot.
It's all reasonable except the paint gets toasted during EDL. Painting's easy. Prep is the killer. That ship will never relaunch as a depot without a lot of prep. As an alternative, use only virgin tankers w/paint as one time accumulators, then recycle to a tanker only role.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/13/2022 01:13 pm
So it's 11-14kW to get rid of. What's the expected efficiency of practical cryo heat pumps? I suspect it's awful, AFAIR 10-20%. So 55 to 140kW range for the electric power.

So 150kW would be about 370m² of panels.  That's 15% of an ISS, all to deploy and stow robotically--reliably--if you want to get the Starship back.  Plus, you need the same amount of heat rejection.  Presumably, you deploy that as the back of your solar array.

That's not an easy engineering task.  Not impossible, but not the simplest thing that SpaceX could do (which is what they'll do).  That makes a non-recoverable depot a lot more likely.

PS:

You will have some radiation hitting some heathsield because the earth takes like 1/3 of the sky so it will illuminate the sides, but at a slant angle.

Radiative flux from a sphere to a cylinder has a computable view factor.  It's horrid, as you can see below, but you can use the pretty graph and see that it's going to be about 45% in VLEO.  So that helps a little.

The other computation that we need is how much heat power a particular boiloff rate gobbles up and expels.  This obviously strays away from full ZBO, but if you have some max amount of electrical power in your budget, it should let you compute the boiloff rate.


A thought.

Quote
That's not an easy engineering task.  Not impossible, but not the simplest thing that SpaceX could do (which is what they'll do).  That makes a non-recoverable depot a lot more likely.


That and radiators and sun shades are why I keep coming back to depot heavy. Deploying stuff is hard and expensive. Tucking it away for redeploy is really, really, hard and expensive.


Here's an alternative for PV. Your numbers assumed state of the art panels with 40% conversion efficiently. What if they use twice as many dirt cheap 20% panels and cut them loose after braking thrust and before encountering significant atmosphere? It rubs against reusability but...


The radiator is quite a bit smaller than the panels and probably easier to refold.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Jim on 08/13/2022 01:46 pm
the boil off would be used for fuel cells or turbo generator.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 08/13/2022 02:29 pm
Using 15kW as the worst case, but at 24hr/day. (Ie, 15kJ/second.) And a heat-of-vaporisation figure for methane of 511 kJ/kg, that means boil-off-only cooling would cost 2.5 tonnes of methane per day.

If your refuelling mission cycle takes 30 days from empty, through accumulation, to final target-vehicle transfer and empty again, you lose 75 tonnes of methane. I don't think it's reasonable to assume a faster cycle for a first generation depot.

Heat of vaporization:  511kJ/kg
Heat of combustion:  56MJ/kg

Even at 10% efficiency cooling by burning methane is 10x more efficient than evaporating it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nevyn72 on 08/13/2022 03:11 pm
Using 15kW as the worst case, but at 24hr/day. (Ie, 15kJ/second.) And a heat-of-vaporisation figure for methane of 511 kJ/kg, that means boil-off-only cooling would cost 2.5 tonnes of methane per day.

If your refuelling mission cycle takes 30 days from empty, through accumulation, to final target-vehicle transfer and empty again, you lose 75 tonnes of methane. I don't think it's reasonable to assume a faster cycle for a first generation depot.

Heat of vaporization:  511kJ/kg
Heat of combustion:  56MJ/kg

Even at 10% efficiency cooling by burning methane is 10x more efficient than evaporating it.

Excellent, then as Jim says, burn the boil-off and get both!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Nomadd on 08/13/2022 03:55 pm
 If they go with a depot and sunshade, has anybody here figured how much an issue earthshine would be?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/13/2022 04:59 pm
It's going to be complicated because the angle of the ship to Earth isn't constant, but rotates 360 degrees each orbit if the nose points to the sun.

According to the view factor plot I included above (https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=50157.0;attach=2120012;image), when h = H/R (H is the orbital radius, R is the Earth's radius) approaches 1, the angle becomes less relevant and everything approaches 0.45.  In a 250km VLEO, h=1.04.  This seems non-intuitive to me, but I trust whoever did that plot (http://webserver.dmt.upm.es/~isidoro/tc3/Radiation%20View%20factors.pdf) a lot more than I trust my intuition on this subject.

That would reduce the problem to figuring out a reasonable average albedo and specular flux from an entire Earth orbit, then using that as the basis for an approximation of the heating.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/13/2022 05:30 pm
Using 15kW as the worst case, but at 24hr/day. (Ie, 15kJ/second.) And a heat-of-vaporisation figure for methane of 511 kJ/kg, that means boil-off-only cooling would cost 2.5 tonnes of methane per day.

If your refuelling mission cycle takes 30 days from empty, through accumulation, to final target-vehicle transfer and empty again, you lose 75 tonnes of methane. I don't think it's reasonable to assume a faster cycle for a first generation depot.

Heat of vaporization:  511kJ/kg
Heat of combustion:  56MJ/kg

Even at 10% efficiency cooling by burning methane is 10x more efficient than evaporating it.

I'd think that you'd let it evaporate, then burn it.  That way you get both. 

But if you're working off of evaporated stuff, then you probably need to factor in battery charging efficiency as well.  I'd think that you need to run a cryocooler for a while to get it chilled down and operating at peak (albeit lousy) efficiency.  You can't do that if you're only running your APU every time the tanks need to vent, so you'd have to dump the energy into the battery until you have enough for an efficient cryocooler run.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/13/2022 09:14 pm
I looked online for into on methane-burning fuel cells, and everything I see suggests that they're a) experimental and b) run at 600° C or more. And I think you're still going to need a sizeable radiator.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 08/14/2022 04:18 am
I looked online for into on methane-burning fuel cells, and everything I see suggests that they're a) experimental and b) run at 600° C or more. And I think you're still going to need a sizeable radiator.

Methane turbines are around us and generate a large part of our power, power backup for homes, etc.  Far better than 10% efficiency.

I'm curious if one could find one that would work in space and use oxygen instead of air.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: BT52 on 08/14/2022 07:57 am
You mean using Raptor powerhead as generator? :D Doable. I m sure unnecessarily tho. I would like to keep propellant passive as much as possible. Well at least how long this prop will be in orbit? 1-2days max? So any advanced sys is unnecessarily complication. I would even argue they would make slightly thicker SS skin. If was necessarily to raise operational pressure and skin stresses.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: sebk on 08/14/2022 09:52 am
Using 15kW as the worst case, but at 24hr/day. (Ie, 15kJ/second.) And a heat-of-vaporisation figure for methane of 511 kJ/kg, that means boil-off-only cooling would cost 2.5 tonnes of methane per day.

If your refuelling mission cycle takes 30 days from empty, through accumulation, to final target-vehicle transfer and empty again, you lose 75 tonnes of methane. I don't think it's reasonable to assume a faster cycle for a first generation depot.

Heat of vaporization:  511kJ/kg
Heat of combustion:  56MJ/kg

Even at 10% efficiency cooling by burning methane is 10x more efficient than evaporating it.

You must include the mass of oxygen which is ~4x bigger. Also you should include oxygen heat of vaporization..
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/14/2022 01:29 pm
So, stainless towards the earth it is.
There's no reason to have bare stainless steel. Even if you intend to re-enter the depot-ships, you can paint the dorsal side with something reflective.
It's all reasonable except the paint gets toasted during EDL. Painting's easy. Prep is the killer. That ship will never relaunch as a depot without a lot of prep.

Cleaning stainless steel and reapplying some paint doesn't seem like it could possibly cost more than building a new depot. Given the difference it makes in thermal input, whether that energy is removed via boiloff or cryo-cooling, it's worth the extra hassle even on recovered/reused depot-ships.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/14/2022 03:44 pm
Using 15kW as the worst case, but at 24hr/day. (Ie, 15kJ/second.) And a heat-of-vaporisation figure for methane of 511 kJ/kg, that means boil-off-only cooling would cost 2.5 tonnes of methane per day.

If your refuelling mission cycle takes 30 days from empty, through accumulation, to final target-vehicle transfer and empty again, you lose 75 tonnes of methane. I don't think it's reasonable to assume a faster cycle for a first generation depot.

Heat of vaporization:  511kJ/kg
Heat of combustion:  56MJ/kg

Even at 10% efficiency cooling by burning methane is 10x more efficient than evaporating it.

I'd think that you'd let it evaporate, then burn it.  That way you get both. 

But if you're working off of evaporated stuff, then you probably need to factor in battery charging efficiency as well.  I'd think that you need to run a cryocooler for a while to get it chilled down and operating at peak (albeit lousy) efficiency.  You can't do that if you're only running your APU every time the tanks need to vent, so you'd have to dump the energy into the battery until you have enough for an efficient cryocooler run.
Or, run a smaller cooler at a higher duty cycle. Unknowns: do cryo coolers gain efficiency with size? Is it acceptable to allow random subcooling throughout the duty cycle?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/14/2022 04:40 pm
I looked online for into on methane-burning fuel cells, and everything I see suggests that they're a) experimental and b) run at 600° C or more. And I think you're still going to need a sizeable radiator.

Methane turbines are around us and generate a large part of our power, power backup for homes, etc.  Far better than 10% efficiency.

I'm curious if one could find one that would work in space and use oxygen instead of air.
Turbine engines do not scale down well. Bigger = more efficient.


An idea I've put out is for a constant rpm constant load two stroke. Liquid propellant used for cooling and combustion.  Inexpensive, stone simple to maintain. When run off of liquid propellant the compression stroke has, in effect, already been done. Should be crazy efficient at small scale.


For those with minimal experience with piston ICE's, with constant rpm and load the complexity of 'breathing' and mixing are simplified and efficiency can approach the theoretical Carnot limits. Get rid of the compression stroke, add in the perfectly characterized cooling requirements, and it's better yet.


The major trade to look at is tapping unvaporized propellant vs increased efficiency. I noodled feeding it with boiloff but the cooling gets tricky and the compression stroke shows up again. Maybe gas fed can work, maybe not.


Being a simple easily maintained system, it's good for Mars use too.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/14/2022 04:55 pm
So, stainless towards the earth it is.
There's no reason to have bare stainless steel. Even if you intend to re-enter the depot-ships, you can paint the dorsal side with something reflective.
It's all reasonable except the paint gets toasted during EDL. Painting's easy. Prep is the killer. That ship will never relaunch as a depot without a lot of prep.

Cleaning stainless steel and reapplying some paint doesn't seem like it could possibly cost more than building a new depot. Given the difference it makes in thermal input, whether that energy is removed via boiloff or cryo-cooling, it's worth the extra hassle even on recovered/reused depot-ships.
I'm in the middle of deck rail and trim painting. If I let you put a sharp stick in my eye would you do it for me?


More seriously, if we're talking about a tanker with some add-ons, why bother? Let it show its heritage and let us wonks be happy. Swap in a new, or as yet unpainted tanker for the next accumulator. If it's a Depot Heavy, do a refurb.


Edit to add: what I've been reading about solar white paint mentions thickness from 3-10mm. It's more of a coating than paint. Might be a pain to remove and re-prep.


Edit 2: The paint substrate material and surface texture are an inherent part of the thermal performance. A wire wheel will most likely remove the coating and not hurt the stainless but it would alter the surface texture. Maybe to the good, maybe not.


In small scale tests investigating Y2O3, an aluminum substrate was found to add good rejection properties. Stainless had the poorest rejection of the tested substrates. My knowledge of aluminum coating is limited to vapor deposition onto glass in a good vacuum (3x10^-5mm of mercury). I'm not aware of any electroplating or chemical deposit techniques but I am widely renowned for being ignorant.  :D  I wonder how chrome works. A little work on the fins and 1960-70's car vibes would rule.


Repreping the surfaces looks a bit more complicated than preping my deck trim.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/14/2022 05:17 pm
Edit to add: what I've been reading about solar white paint mentions thickness from 3-10mm. It's more of a coating than paint. Might be a pain to remove and re-prep.

Some of the solar white work uses a silvered backing to reflect as much IR as possible.  That would be a huge pain to reapply on a reusable tanker or depot.  I'm not sure how much that contributes to the the overall reflectance.  Probably doesn't have much to do with the emissivity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/14/2022 07:10 pm
Edit to add: what I've been reading about solar white paint mentions thickness from 3-10mm. It's more of a coating than paint. Might be a pain to remove and re-prep.

Some of the solar white work uses a silvered backing to reflect as much IR as possible.  That would be a huge pain to reapply on a reusable tanker or depot.  I'm not sure how much that contributes to the the overall reflectance.  Probably doesn't have much to do with the emissivity.


Fron the papers cited earlier, it looks like you have your choice of paint or ceramic tiles.

The tiles have a silver backing and reflect 99.8% of solar radiation while still being over 90% emissive from 8 microns on down. Perfect for a dedicated depot (or maybe a sun shade), but too fragile for reentry.

The paint can go on any sort of reflective backing, but it's not as reflective. Applied to unpolished stainless steel, they measured it at 90% reflective with ten coats (5 mm total thickness). Maybe worth it just to reduce the size/cost of a cooler. Or to eliminate the need for one entirely, depending on the timeframe. Probably dirt cheap, so no big issue if it doesn't survive reentry.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/14/2022 07:30 pm
Edit to add: what I've been reading about solar white paint mentions thickness from 3-10mm. It's more of a coating than paint. Might be a pain to remove and re-prep.

Some of the solar white work uses a silvered backing to reflect as much IR as possible.  That would be a huge pain to reapply on a reusable tanker or depot.  I'm not sure how much that contributes to the the overall reflectance.  Probably doesn't have much to do with the emissivity.


Fron the papers cited earlier, it looks like you have your choice of paint or ceramic tiles.

The tiles have a silver backing and reflect 99.8% of solar radiation while still being over 90% emissive from 8 microns on down. Perfect for a dedicated depot (or maybe a sun shade), but too fragile for reentry.

The paint can go on any sort of reflective backing, but it's not as reflective. Applied to unpolished stainless steel, they measured it at 90% reflective with ten coats (5 mm total thickness). Maybe worth it just to reduce the size/cost of a cooler. Or to eliminate the need for one entirely, depending on the timeframe. Probably dirt cheap, so no big issue if it doesn't survive reentry.
The tiles mentioned were not really intended or explored for EDL. I wonder if anybody's been looking at different fillers. If the heat shield tiles had a Y2O3 glaze they might be able to thin them for lower mass.


Edit to add: Y2O3 is hydrophobic. That's a big thing for the tiles. Not a total solution but a step in the right direction.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/14/2022 08:35 pm
The tiles mentioned were not really intended or explored for EDL. I wonder if anybody's been looking at different fillers. If the heat shield tiles had a Y2O3 glaze they might be able to thin them for lower mass.


Edit to add: Y2O3 is hydrophobic. That's a big thing for the tiles. Not a total solution but a step in the right direction.
I think the trouble is that the heat-shield tiles need to be emissive in the range from about 1 to 8 microns. That rules out coating them with anything like Y2O3. Or anything else that reflects most solar energy, for that matter.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/14/2022 08:42 pm
The problems with refolding and stowing the deployables, and the impact of EDL on solar white leads me (again) to the conclusion that while testing may use near plain vanilla tankers as accumulators, an operational depot will most likely never do EDL.

As pointed out by edzeba, the only specific near future use is Artemus. One in LEO or a bit higher, and one in HEO or lunar orbit. The high one is easily a throwaway. The low one could conceivably return if brute force 'launch more tankers' can successfully counter boil off. This last depends on real world understanding of boil off and how long a transfer campaign would take.

If the not yet available numbers reflecting real world experience show that more tankers can't keep up or drags out the campaign too long, boil off mitigation will be done. Everything we've looked at makes successful EDL look chancy. The low ship would then be a throwaway.

This is not necessarily a bad thing. Why cry over a wasted SS? How many have we seen scrapped before a fire was lit under them? How many more will we see? The first Artemus accumulator/depot will be crude compared to what follows.

The follow-on's will explore ways to either design for EDL or for further use in later missions. My gut says the latter. I also expect that a lot of the ones earmarked for further use will quickly become obsolescent and be splashed in - probably before ever being reused. It's what SX does, and it seems to work.

One tanker depot on LEO will align for lunar insertion every 14 days. A second, in a plane 90 degrees to the first opens up an opportunity every 7 days. Correcting the earth exit to insert to any lunar orbital plane that crosses the earth-moon line is relatively trivial if done around L1. An Earth or sun synchronous family is harder. There's a lot of operational flexibility here.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/14/2022 09:36 pm
The tiles mentioned were not really intended or explored for EDL. I wonder if anybody's been looking at different fillers. If the heat shield tiles had a Y2O3 glaze they might be able to thin them for lower mass.


Edit to add: Y2O3 is hydrophobic. That's a big thing for the tiles. Not a total solution but a step in the right direction.
I think the trouble is that the heat-shield tiles need to be emissive in the range from about 1 to 8 microns. That rules out coating them with anything like Y2O3. Or anything else that reflects most solar energy, for that matter.
Maybe I'm misunderstanding. See pg3, bottom chart showing a piece of the Y2O3 transmission spectrum. I'd cut n paste but it's not cooperating. [size=78%]https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf)[/size]


The chart shows low transmission from about 2 to 10 microns. I've assumed this to mean it readily adsorbs these wavelengths which should also mean it readily emits at these wavelengths. Wrong?


I've tried looking up an emission spectra and hit paywalls and yttria mixed with other things. Also, why 2-10 microns? Isn't this a bit past the far UV? I thought the ideal emission range was in the near IR.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/14/2022 11:09 pm
The tiles mentioned were not really intended or explored for EDL. I wonder if anybody's been looking at different fillers. If the heat shield tiles had a Y2O3 glaze they might be able to thin them for lower mass.


Edit to add: Y2O3 is hydrophobic. That's a big thing for the tiles. Not a total solution but a step in the right direction.
I think the trouble is that the heat-shield tiles need to be emissive in the range from about 1 to 8 microns. That rules out coating them with anything like Y2O3. Or anything else that reflects most solar energy, for that matter.
Maybe I'm misunderstanding. See pg3, bottom chart showing a piece of the Y2O3 transmission spectrum. I'd cut n paste but it's not cooperating. [size=78%]https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf)[/size]


The chart shows low transmission from about 2 to 10 microns. I've assumed this to mean it readily adsorbs these wavelengths which should also mean it readily emits at these wavelengths. Wrong?


I've tried looking up an emission spectra and hit paywalls and yttria mixed with other things. Also, why 2-10 microns? Isn't this a bit past the far UV? I thought the ideal emission range was in the near IR.
Ah. In this case, low transmission means it reflects those wavelengths, which means it is almost completely non-emissive in that range. Otherwise it wouldn't look white in the visible.

The problem is that the heat tiles get up to 1650 Kelvin, which means their emission curve (see the chart I posted two pages back) has a lot of overlap with the solar one. That chart shows that over 90% of the IR spectrum at 1650 is between 1 and 8 microns. That has way too much overlap with the solar spectrum. Solar white, in particular, is super reflective from 1/4 micron to 8 microns.

That means you can't make a single kind of tile that serves both purposes.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/15/2022 04:12 pm
The tiles mentioned were not really intended or explored for EDL. I wonder if anybody's been looking at different fillers. If the heat shield tiles had a Y2O3 glaze they might be able to thin them for lower mass.


Edit to add: Y2O3 is hydrophobic. That's a big thing for the tiles. Not a total solution but a step in the right direction.
I think the trouble is that the heat-shield tiles need to be emissive in the range from about 1 to 8 microns. That rules out coating them with anything like Y2O3. Or anything else that reflects most solar energy, for that matter.
Maybe I'm misunderstanding. See pg3, bottom chart showing a piece of the Y2O3 transmission spectrum. I'd cut n paste but it's not cooperating. [size=78%]https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf)[/size]


The chart shows low transmission from about 2 to 10 microns. I've assumed this to mean it readily adsorbs these wavelengths which should also mean it readily emits at these wavelengths. Wrong?


I've tried looking up an emission spectra and hit paywalls and yttria mixed with other things. Also, why 2-10 microns? Isn't this a bit past the far UV? I thought the ideal emission range was in the near IR.
Ah. In this case, low transmission means it reflects those wavelengths, which means it is almost completely non-emissive in that range. Otherwise it wouldn't look white in the visible.

The problem is that the heat tiles get up to 1650 Kelvin, which means their emission curve (see the chart I posted two pages back) has a lot of overlap with the solar one. That chart shows that over 90% of the IR spectrum at 1650 is between 1 and 8 microns. That has way too much overlap with the solar spectrum. Solar white, in particular, is super reflective from 1/4 micron to 8 microns.

That means you can't make a single kind of tile that serves both purposes.
Thank you. Between some slippery decimals, antique eyes and a bad assumption, I was off track.


I was going to leave my response at that but started thinking and came to a conclusion that will make me sound like a whiney 'it's my idea and I'm in love with it' type of guy. I'll risk it.


There's one assumption. From 10-100 microns, except for a couple of notches, the chart shows high transmission. The assumption is that, as in the visible range, this equates to emissivity.


Here's my thinking. Tiles with solar white characteristics in pre-EDL sunlight should run cooler than black tiles. In the early stages of EDL the thermal input is first friction, then compression heating. The solar white would most probably have about the same rejection characteristics as the black tiles. Starting from a lower temp would probably not have a significant impact but it might be worth a BOE for somebody that knew how to do it.


Where there would be a difference is when the plasma starts to build up and radiative input becomes dominant. The solar white will reject more of this than the black, dropping the peak temperature and shifting the black body emissions away from the visible and deeper into the IR. How much? I have not a clue.


If the black body peak moves far enough it's a strong Planck in my argument for solar white tiles.  :D
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Jim on 08/15/2022 04:52 pm

An idea I've put out is for a constant rpm constant load two stroke. Liquid propellant used for cooling and combustion.  Inexpensive, stone simple to maintain. When run off of liquid propellant the compression stroke has, in effect, already been done. Should be crazy efficient at small scale.


For those with minimal experience with piston ICE's, with constant rpm and load the complexity of 'breathing' and mixing are simplified and efficiency can approach the theoretical Carnot limits. Get rid of the compression stroke, add in the perfectly characterized cooling requirements, and it's better yet.


The major trade to look at is tapping unvaporized propellant vs increased efficiency. I noodled feeding it with boiloff but the cooling gets tricky and the compression stroke shows up again. Maybe gas fed can work, maybe not.

Being a simple easily maintained system, it's good for Mars use too.


some related info

https://www.ulalaunch.com/docs/default-source/extended-duration/development-status-of-an-integrated-propulsion-and-power-system-for-long-duration-cryogenic-spaceflight-2012.pdf
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/15/2022 04:52 pm
Where there would be a difference is when the plasma starts to build up and radiative input becomes dominant. The solar white will reject more of this than the black, dropping the peak temperature and shifting the black body emissions away from the visible and deeper into the IR. How much? I have not a clue.
I really liked that idea too, but what I could find wasn't encouraging.

"Convection is the primary means of heat transfer to a vehicle entering Earth’s atmosphere at speeds under about 15,000 m/s." (Advanced Aerospace Medicine On-line--Tutorial, Section III Space Operations, Chapter 4 Basic Concepts of Manned Spacecraft Design, 4.1.7 Returning from Space (https://www.faa.gov/about/office_org/headquarters_offices/avs/offices/aam/cami/library/online_libraries/aerospace_medicine/tutorial/media/iii.4.1.7_returning_from_space.pdf), p 322, Federal Aviation Administration, Retrieved 15 Aug 2022).

15 kps is well above Earth escape velocity (about 11 kps), and very-low-earth-orbit velocity is under 8 kps. So the heat transfer will mostly be convective, not radiative, and, hence, I don't think Solar White is going to help.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/15/2022 05:47 pm

An idea I've put out is for a constant rpm constant load two stroke. Liquid propellant used for cooling and combustion.  Inexpensive, stone simple to maintain. When run off of liquid propellant the compression stroke has, in effect, already been done. Should be crazy efficient at small scale.

For those with minimal experience with piston ICE's, with constant rpm and load the complexity of 'breathing' and mixing are simplified and efficiency can approach the theoretical Carnot limits. Get rid of the compression stroke, add in the perfectly characterized cooling requirements, and it's better yet.


The major trade to look at is tapping unvaporized propellant vs increased efficiency. I noodled feeding it with boiloff but the cooling gets tricky and the compression stroke shows up again. Maybe gas fed can work, maybe not.

Being a simple easily maintained system, it's good for Mars use too.


some related info

https://www.ulalaunch.com/docs/default-source/extended-duration/development-status-of-an-integrated-propulsion-and-power-system-for-long-duration-cryogenic-spaceflight-2012.pdf
Very interesting. I like the idea of eliminating hydrazine, helium, most of the batteries, etc. That document seems to be from about 2011, so I wondered whatever happened to the idea. Looking further, I also found "ACES Stage Concept: Higher Performance, New Capabilities, at a Lower Recurring Cost (https://web.archive.org/web/20160313195421/http://www.ulalaunch.com/uploads/docs/Published_Papers/Upper_Stages/ACES-Stage_Concept-AIAASpace_2015.pdf)," Section IV: Integrated Vehicle Fluids (IVF), Jonathan Barr, ULA, 2015, which was a little easier read, albeit less detailed.

ULA abandoned ACES in 2020 (https://spacenews.com/ula-studying-long-term-upgrades-to-vulcan/) in favor of the Centaur V, but I can't find any indication whether Centaur V uses the internal-combustion-engine approach of IVF or not. It certainly offers long-duration missions with multiple relights (https://en.wikipedia.org/wiki/Vulcan_Centaur#Long-endurance_upper_stages) though.

If you put something like this on Starship, I'm wondering where the exhaust would go. Some of the documentation for ACES makes it sound like it used the main engine bell. Does that make sense?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: JayWee on 08/15/2022 06:08 pm
https://www.ulalaunch.com/docs/default-source/extended-duration/development-status-of-an-integrated-propulsion-and-power-system-for-long-duration-cryogenic-spaceflight-2012.pdf
Very interesting. I like the idea of eliminating hydrazine, helium, most of the batteries, etc. That document seems to be from about 2011, so I wondered whatever happened to the idea.
ULA's parents (and Shelby) put a stop to it. Would jeopardize SLS.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/15/2022 06:16 pm
Where there would be a difference is when the plasma starts to build up and radiative input becomes dominant. The solar white will reject more of this than the black, dropping the peak temperature and shifting the black body emissions away from the visible and deeper into the IR. How much? I have not a clue.
I really liked that idea too, but what I could find wasn't encouraging.

"Convection is the primary means of heat transfer to a vehicle entering Earth’s atmosphere at speeds under about 15,000 m/s." (Advanced Aerospace Medicine On-line--Tutorial, Section III Space Operations, Chapter 4 Basic Concepts of Manned Spacecraft Design, 4.1.7 Returning from Space (https://www.faa.gov/about/office_org/headquarters_offices/avs/offices/aam/cami/library/online_libraries/aerospace_medicine/tutorial/media/iii.4.1.7_returning_from_space.pdf), p 322, Federal Aviation Administration, Retrieved 15 Aug 2022).

15 kps is well above Earth escape velocity (about 11 kps), and very-low-earth-orbit velocity is under 8 kps. So the heat transfer will mostly be convective, not radiative, and, hence, I don't think Solar White is going to help.
Grumble, grumble. Thought radiative dominated at 8km/s reentry. Sniff, sniff. I guess I did love that idea just a little.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/15/2022 06:19 pm
https://www.ulalaunch.com/docs/default-source/extended-duration/development-status-of-an-integrated-propulsion-and-power-system-for-long-duration-cryogenic-spaceflight-2012.pdf (https://www.ulalaunch.com/docs/default-source/extended-duration/development-status-of-an-integrated-propulsion-and-power-system-for-long-duration-cryogenic-spaceflight-2012.pdf)
Very interesting. I like the idea of eliminating hydrazine, helium, most of the batteries, etc. That document seems to be from about 2011, so I wondered whatever happened to the idea.
ULA's parents (and Shelby) put a stop to it. Would jeopardize SLS.
Shelby's day is past. Can't patent an idea. It's too good an idea for somebody to not go there.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Jim on 08/15/2022 06:28 pm


If you put something like this on Starship, I'm wondering where the exhaust would go. Some of the documentation for ACES makes it sound like it used the main engine bell. Does that make sense?

In separate nozzles for settling.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/15/2022 06:35 pm

An idea I've put out is for a constant rpm constant load two stroke. Liquid propellant used for cooling and combustion.  Inexpensive, stone simple to maintain. When run off of liquid propellant the compression stroke has, in effect, already been done. Should be crazy efficient at small scale.

For those with minimal experience with piston ICE's, with constant rpm and load the complexity of 'breathing' and mixing are simplified and efficiency can approach the theoretical Carnot limits. Get rid of the compression stroke, add in the perfectly characterized cooling requirements, and it's better yet.


The major trade to look at is tapping unvaporized propellant vs increased efficiency. I noodled feeding it with boiloff but the cooling gets tricky and the compression stroke shows up again. Maybe gas fed can work, maybe not.

Being a simple easily maintained system, it's good for Mars use too.


some related info

https://www.ulalaunch.com/docs/default-source/extended-duration/development-status-of-an-integrated-propulsion-and-power-system-for-long-duration-cryogenic-spaceflight-2012.pdf (https://www.ulalaunch.com/docs/default-source/extended-duration/development-status-of-an-integrated-propulsion-and-power-system-for-long-duration-cryogenic-spaceflight-2012.pdf)
Very interesting. I like the idea of eliminating hydrazine, helium, most of the batteries, etc. That document seems to be from about 2011, so I wondered whatever happened to the idea. Looking further, I also found "ACES Stage Concept: Higher Performance, New Capabilities, at a Lower Recurring Cost (https://web.archive.org/web/20160313195421/http://www.ulalaunch.com/uploads/docs/Published_Papers/Upper_Stages/ACES-Stage_Concept-AIAASpace_2015.pdf)," Section IV: Integrated Vehicle Fluids (IVF), Jonathan Barr, ULA, 2015, which was a little easier read, albeit less detailed.

ULA abandoned ACES in 2020 (https://spacenews.com/ula-studying-long-term-upgrades-to-vulcan/) in favor of the Centaur V, but I can't find any indication whether Centaur V uses the internal-combustion-engine approach of IVF or not. It certainly offers long-duration missions with multiple relights (https://en.wikipedia.org/wiki/Vulcan_Centaur#Long-endurance_upper_stages) though.

If you put something like this on Starship, I'm wondering where the exhaust would go. Some of the documentation for ACES makes it sound like it used the main engine bell. Does that make sense?
The two red thingies on opposite sides are the units. There are little nozzles extending from them. KISS. If they do another three vacuum engines on SS it'll get crowded but still workable.


The engineering team for this system could put three support gearheads in hog heaven.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/17/2022 06:58 pm
Hmm. Looking at the pic flashed an idea. Use three units evenly spaced. Three rotating cranks. Three sources of mild torque. The nozzles are intended for settling thrust but can be used to add to the crank torque. Between the two there would be some fine low propellant cost attitude control.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/17/2022 08:40 pm
 
The tiles mentioned were not really intended or explored for EDL. I wonder if anybody's been looking at different fillers. If the heat shield tiles had a Y2O3 glaze they might be able to thin them for lower mass.


Edit to add: Y2O3 is hydrophobic. That's a big thing for the tiles. Not a total solution but a step in the right direction.
I think the trouble is that the heat-shield tiles need to be emissive in the range from about 1 to 8 microns. That rules out coating them with anything like Y2O3. Or anything else that reflects most solar energy, for that matter.
Maybe I'm misunderstanding. See pg3, bottom chart showing a piece of the Y2O3 transmission spectrum. I'd cut n paste but it's not cooperating. [size=78%]https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf)[/size]


The chart shows low transmission from about 2 to 10 microns. I've assumed this to mean it readily adsorbs these wavelengths which should also mean it readily emits at these wavelengths. Wrong?


I've tried looking up an emission spectra and hit paywalls and yttria mixed with other things. Also, why 2-10 microns? Isn't this a bit past the far UV? I thought the ideal emission range was in the near IR.
Ah. In this case, low transmission means it reflects those wavelengths, which means it is almost completely non-emissive in that range. Otherwise it wouldn't look white in the visible.

The problem is that the heat tiles get up to 1650 Kelvin, which means their emission curve (see the chart I posted two pages back) has a lot of overlap with the solar one. That chart shows that over 90% of the IR spectrum at 1650 is between 1 and 8 microns. That has way too much overlap with the solar spectrum. Solar white, in particular, is super reflective from 1/4 micron to 8 microns.

That means you can't make a single kind of tile that serves both purposes.

One obvious solution here is...

don't use a solar white paint with a cutoff at 8 microns. Use something with a different cutoff.

Based on your earlier graphs, the ideal cutoff wavelength for a "designer" selective surface material would be somewhere around 1-1.5 microns.

Edit: it seems that Z-93 paint (which SpaceX has used before (https://www.nasa.gov/mission_pages/station/research/news/dragon_coating.html)) has a cutoff around 3 microns (https://www.spectroscopyonline.com/view/optical-measurement-of-the-reflectance-behavior-of-z93-the-thermal-coating-on-the-international). Getting closer!

Tiles can't simply be painted (or can they? ???), but that's not my point. Rather, this is intended to show that all solar selective surfaces don't have the same cutoff at 8 microns.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 08/17/2022 11:27 pm
The tiles mentioned were not really intended or explored for EDL. I wonder if anybody's been looking at different fillers. If the heat shield tiles had a Y2O3 glaze they might be able to thin them for lower mass.

Why not paint the top of Starship white, leave the tiles black, and point the white side at the sun or other radiating entity?

The white will reflect the heat.

The black tiles will radiate some heat out, which by design they are quite good at.

So will this work?

At solar incidence of 1360 W/m^2 and reflectivity of 98% on the white top, 27 watts/m^2 emission would be needed of the tiles, giving an equilibrium temperature of ~150K, above 112k boiling point of methane.  Close, but no cigar, though it would reduce the evaporation rate of the methane by about 8/27 = 30% as compared to the entire Starship being white.  (8W is tile emission at boiling point)

In LEO while on the sunward side the tiles getting earth radiation, at a rate of 400W/m^2.  An hour soak would impart 1.44MJ on each square meter of tile, and the tiles can store 9MJ of heat per square meter before heating up the back side of the tiles (i.e. the mat and stainless steel and thus the fuel) significantly.

Alas, I lack the knowledge to figure out how hot the outside of the tiles get when there's 1.44MJ of energy stored per square meter of tile, so I can't figure out whether that will radiate off while in the shade of the earth.  Anyone got a curve or equation for shuttle tile soak, heat conduction, and radiation?

The LEO scenario can't be better than deep space, and all white would eliminate 392 W/m^2 from earth radiance while on the sunward side of the earth.  Maybe not worth breaking out the shuttle performance curve references and solving a dynamic heat equation.

So for no modification saving painting the top white, one gains a 30% reduction in boil off in deep space as compared to an all-white Starship and at least 7 free orbits of the Earth before having to boil off methalox. 

So definitely helps the deep space scenario, probably worse for LEO.

But it is almost free of major design changes.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/18/2022 12:38 am
One obvious solution here is...

don't use a solar white paint with a cutoff at 8 microns. Use something with a different cutoff.

Based on your earlier graphs, the ideal cutoff wavelength for a "designer" selective surface material would be somewhere around 1-1.5 microns.
Yes, but then you give up on the ability of Solar White paint to keep the inside of a tank at 50 K. I'm not sure how much warmer it would be with a 1.5 micron cutoff, but my guess is you're talking dry ice, not LOX.

A lot depends on what your goals are. If you want a system to passively keep cryogens liquid even in full sunlight, then you need that 8 micron cutoff. But if you're just trying to reduce the costs of cooling, then, sure, any old white paint is probably fine.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/20/2022 08:14 am
No no, by "optimum" I mean that a cutoff at 1-1.5 microns will result in the coldest equilibrium temperature. Colder than 8 microns.

The optimum spectral strategy AIUI is to be 100% reflective in parts of the electromagnetic spectrum where incoming flux exceeds black body emission, and 100% emissive otherwise.

If the vertical axis is spectral power (ie the area under the curve is power), the ideal wavelength cutoff is simply where the curves cross.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/20/2022 05:59 pm
No no, by "optimum" I mean that a cutoff at 1-1.5 microns will result in the coldest equilibrium temperature. Colder than 8 microns.

The optimum spectral strategy AIUI is to be 100% reflective in parts of the electromagnetic spectrum where incoming flux exceeds black body emission, and 100% emissive otherwise.

If the vertical axis is spectral power (ie the area under the curve is power), the ideal wavelength cutoff is simply where the curves cross.
Ah. No, the curves I showed above are normalized so the area under each one is 1. If you look at the unnormalized curves, the radiance from the sun is overwhelming and the curves don't cross at all.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/22/2022 10:28 am
No no, by "optimum" I mean that a cutoff at 1-1.5 microns will result in the coldest equilibrium temperature. Colder than 8 microns.

The optimum spectral strategy AIUI is to be 100% reflective in parts of the electromagnetic spectrum where incoming flux exceeds black body emission, and 100% emissive otherwise.

If the vertical axis is spectral power (ie the area under the curve is power), the ideal wavelength cutoff is simply where the curves cross.
Ah. No, the curves I showed above are normalized so the area under each one is 1. If you look at the unnormalized curves, the radiance from the sun is overwhelming and the curves don't cross at all.

Did you account for the view factor?

You're not comparing 1 square meter of the Sun's photosphere with 1 square meter of Starship's exterior. You're comparing the hemispherical thermal emissions from Starship (view solid angle = 2 pi steradians) with the insolation from the Sun (view solid angle = 6.794×10−5 steradians).

That gets you the radiant balance for a surface under direct perpendicular sunlight.

Then derate the insolation to account for orbital night (~1/2), the projected geometry of a cylinder (1/pi), and vehicle pointing (for the tank wall this is sin(θ), where θ is the angle between the nose pointing direction and the Sun).

Plus add the emitted heat and reflected light from the Earth, averaged over one orbit. The emitted heat is simply a ~250 K blackbody emitting 235 watts per square meter (https://arxiv.org/pdf/physics/0503119.pdf), derated 50% for view factor. The reflected light can be approximated as Lambertian solar reflector (black body temperature = 6000 K, times the Sun view factor above) with a reflectivity of 31% (Earth's mean albedo), plus 50% derate for view factor, and 1/pi for Earth's projected geometry.

Then see where the curves cross. ;)


Edit: if the surface is at radiative equilibrium, then normalizing the area under the curves to 1 should work, actually. The catch is, you have to make sure the chosen surface temperature for Starship is indeed the correct equilibrium temperature.

Hopefully this helps!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/23/2022 09:23 pm
Then derate the insolation to account for orbital night (~1/2), the projected geometry of a cylinder (1/pi), and vehicle pointing (for the tank wall this is sin(θ), where θ is the angle between the nose pointing direction and the Sun).

Plus add the emitted heat and reflected light from the Earth, averaged over one orbit. The emitted heat is simply a ~250 K blackbody emitting 235 watts per square meter (https://arxiv.org/pdf/physics/0503119.pdf), derated 50% for view factor. The reflected light can be approximated as Lambertian solar reflector (black body temperature = 6000 K, times the Sun view factor above) with a reflectivity of 31% (Earth's mean albedo), plus 50% derate for view factor, and 1/pi for Earth's projected geometry.

Then see where the curves cross. ;)

The view factor for Earth-to-Starship in LEO is roughly 45%, irrespective of orientation.  I was kinda surprised by this.  See here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2395765#msg2395765).

If you point the nose (or the tail, I guess, although that's were the LOX tank dome is), at the sun, the flux should be close to a point-source at infinity projected onto the nose cross-section, which is SolarConstant * π4.5².  No view factor hand-wringing required.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/24/2022 08:07 am
Then derate the insolation to account for orbital night (~1/2), the projected geometry of a cylinder (1/pi), and vehicle pointing (for the tank wall this is sin(θ), where θ is the angle between the nose pointing direction and the Sun).

Plus add the emitted heat and reflected light from the Earth, averaged over one orbit. The emitted heat is simply a ~250 K blackbody emitting 235 watts per square meter (https://arxiv.org/pdf/physics/0503119.pdf), derated 50% for view factor. The reflected light can be approximated as Lambertian solar reflector (black body temperature = 6000 K, times the Sun view factor above) with a reflectivity of 31% (Earth's mean albedo), plus 50% derate for view factor, and 1/pi for Earth's projected geometry.

Then see where the curves cross. ;)

The view factor for Earth-to-Starship in LEO is roughly 45%, irrespective of orientation.  I was kinda surprised by this.

Yep. I came to the same conclusion.


Doing (roughly) the above math, I don't know where this 8,000 nanometer spectral cutoff comes from. By my curves, to minimize tank heating the surface should ideally be maximally reflective through the entire infrared band.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/24/2022 10:21 pm
Doing (roughly) the above math, I don't know where this 8,000 nanometer spectral cutoff comes from. By my curves, to minimize tank heating the surface should ideally be maximally reflective through the entire infrared band.
Well, there's this quote from the guys who came up with Solar White:

"First, assume that a perfect coating was developed with a cutoff at about 4 µm, so that only 1% of the Sun’s energy was absorbed, i.e. a ratio alpha/epsilon = 0.01. For a sphere the temperature now drops to 88 K, below the 90 K needed to maintain LOX, but still too high to run a superconductor. But if we can find a material with a transition wavelength at about 8 µm, where only 0.1% of the Sun’s energy is absorbed, then the sphere temperature will drop to about 50 K. This would not only allow superconductors to operate, but would allow LOX storage to occur at higher density and at lower pressure. Figure 7 shows the temperatures that could be achieved if a perfect material were available to coat a sphere at 1 AU from the Sun."
(Cryogenic Selective Surfaces: Final Report on a Phase I NIAC Study (https://www.nasa.gov/sites/default/files/atoms/files/cryogenic_selective_surfaces_final_report_niac_phase_i.pdf), p. 10, February 2016, Robert C. Youngquist and Mark A. Nurge)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/28/2022 01:52 am
Doing (roughly) the above math, I don't know where this 8,000 nanometer spectral cutoff comes from. By my curves, to minimize tank heating the surface should ideally be maximally reflective through the entire infrared band.
Well, there's this quote from the guys who came up with Solar White:

"First, assume that a perfect coating was developed with a cutoff at about 4 µm, so that only 1% of the Sun’s energy was absorbed, i.e. a ratio alpha/epsilon = 0.01. For a sphere the temperature now drops to 88 K, below the 90 K needed to maintain LOX, but still too high to run a superconductor. But if we can find a material with a transition wavelength at about 8 µm, where only 0.1% of the Sun’s energy is absorbed, then the sphere temperature will drop to about 50 K. This would not only allow superconductors to operate, but would allow LOX storage to occur at higher density and at lower pressure. Figure 7 shows the temperatures that could be achieved if a perfect material were available to coat a sphere at 1 AU from the Sun."
(Cryogenic Selective Surfaces: Final Report on a Phase I NIAC Study (https://www.nasa.gov/sites/default/files/atoms/files/cryogenic_selective_surfaces_final_report_niac_phase_i.pdf), p. 10, February 2016, Robert C. Youngquist and Mark A. Nurge)

That explains it.

Their 8 micron cutoff can only reach LOX temperatures in deep space. The authors don't say it's the "ideal" cutoff either (Figure 7).

We're doing calculations for LEO (significant IR heating), so 8 microns is not ideal here.

In LEO Solar White is also incapable of reaching LOX temperatures, which the authors themselves acknowledge (Section 2.3).

Phew! Mystery solved.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/28/2022 06:06 pm
We're doing calculations for LEO (significant IR heating), so 8 microns is not ideal here.

In LEO Solar White is also incapable of reaching LOX temperatures, which the authors themselves acknowledge (Section 2.3).
What I would say is that, for a Starship in LEO, there is no completely passive ZBO solution. Not at the moment, anyway. There must be some cryocooler, with all the extra hardware that implies. The best that Solar White can do for you is reduce the mass of the cryocooler.

Are we in broad agreement?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/29/2022 04:57 am
What I would say is that, for a Starship in LEO, there is no completely passive ZBO solution. Not at the moment, anyway. There must be some cryocooler, with all the extra hardware that implies. The best that Solar White can do for you is reduce the mass of the cryocooler.

Or use boil-off.

I did a back-of-the-envelope calc earlier in the thread (using someone else's thermal figures), and depending on how much heat you can reject, selectively boiling-off methane (boiling off LOx is less effective) can remove the remaining heat for sub-20tonnes of loss for a modestly slow refuelling cycle. Given the "propellant is cheap" model of operations, that's doable for a first generation depot.

But to make the decision, you have to know what kind of heat input you are dealing with.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/29/2022 08:19 am
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.

Boil-off can be made arbitrarily low by adding more MLI layers.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 08/29/2022 02:02 pm
What I would say is that, for a Starship in LEO, there is no completely passive ZBO solution. Not at the moment, anyway. There must be some cryocooler, with all the extra hardware that implies. The best that Solar White can do for you is reduce the mass of the cryocooler.

Or use boil-off.

I did a back-of-the-envelope calc earlier in the thread (using someone else's thermal figures), and depending on how much heat you can reject, selectively boiling-off methane (boiling off LOx is less effective) can remove the remaining heat for sub-20tonnes of loss for a modestly slow refuelling cycle. Given the "propellant is cheap" model of operations, that's doable for a first generation depot.

But to make the decision, you have to know what kind of heat input you are dealing with.

I may be having a thermodynamic brain fart, but when you do this just from an energy balance standpoint, you've got a certain amount of heat seeping in, and in equilibrium you have to remove that heat by boiling prop and venting it, right?  So the heat rate that seeps in, divided by the enthalpy of vaporization, gives you the boiloff rate.

No cooling effects need to be considered.  Yeah, they're there, but the heat just warms stuff up until it vaporizes more prop.  If you know the heat balance, you know the boiloff rate.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/29/2022 04:34 pm
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.
I wouldn't say no advantage. At a minimum, you could use the paint (not the tiles) on the parts of the ship that don't have to be jet black and that would reject 90% of the heat from sun exposure at very little cost. But it definitely doesn't buy you ZBO. Not with a standard Starship, anyway.

At this point, I'd like to talk a little bit about exactly how the fuel is going to be transported. (Apologies if this has already been discussed. Extra apologies if it's already been discarded with good reason!) :-)

I know Elon doesn't like depos, but I'm starting to think it makes more sense to have a special "fuel depot" (for want of a better word) that a standard-issue Starship would carry as cargo rather than have a special tanker version of Starship. That would mean that two Starships would never dock to each other directly--not for purposes of fueling, anyway.

My reasoning is as follows:

First, that means you don't have an extra version of Starship with extra-large tanks. Second, the fuel depot would have the hardware required to mate to the QD ports on a Starship and do the refueling, so you wouldn't need a version of Starship with that equipment on it somewhere. (Or, worse, add that hardware to every Starship.)

The basic fuel depot would just be fuel tanks, the docking attachment, and pumps. It wouldn't ever leave the cargo bay of the Starship that brought it up to orbit. It's the sort of thing you'd use to refuel the Lunar Starship, assuming that only needs one extra load of fuel. You'd launch the two Starships at about the same time so boiloff wouldn't be a big concern.

A more complex fuel depot would be larger and have a cryocooler, radiator, extra insulation, and QD ports identical to those on a Starship. It would be put in orbit by itself and get filled up on multiple visits by Starships carrying basic depots full of fuel and LOX. (Note that you wouldn't need to develop this until after the lunar missions.)

That sort of assembly probably would benefit from extensive use of Solar White tiles. Yes, it would still need a cryocooler, but the tiles ought to reduce the mass of the cooler, the radiator, and the solar panels needed to power it.

You would cope with the orbital precession problem by timing the initial orbit and the refueling orbits so the depot is in the right place at the right time to refuel the Starship dedicated to the mission. After the mission Starship departed, you could send up another Starship to retrieve the empty depot.

If you wanted to use the same depot for multiple missions, you could use something like the RAAN-Agnostic 3-Burn Departure Methadology for Deep Space Missions from LEO Depots (https://selenianboondocks.com/2018/09/aas-paper-review-raan-agnostic-3-burn-departure-methodology-for-deep-space-missions-from-leo-depots-part-1-of-2/) that has been talked about earlier.

Again, apologies if this has already been debated/discarded.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/29/2022 04:38 pm
What I would say is that, for a Starship in LEO, there is no completely passive ZBO solution. Not at the moment, anyway. There must be some cryocooler, with all the extra hardware that implies. The best that Solar White can do for you is reduce the mass of the cryocooler.

Or use boil-off.

I did a back-of-the-envelope calc earlier in the thread (using someone else's thermal figures), and depending on how much heat you can reject, selectively boiling-off methane (boiling off LOx is less effective) can remove the remaining heat for sub-20tonnes of loss for a modestly slow refuelling cycle. Given the "propellant is cheap" model of operations, that's doable for a first generation depot.

But to make the decision, you have to know what kind of heat input you are dealing with.
Maybe I'm missing something, but how is this different from just doing nothing? (Nothing except venting gas as needed based on the pressure, that is.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/29/2022 05:23 pm
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.
I wouldn't say no advantage. [...]

I mean there's no advantage to the selectivity, ie the perfect paint would completely reject all wavelengths. There is no intentional cutoff.

Same principle, but in reverse: https://en.wikipedia.org/wiki/Selective_surface

The closest thing that exists in real life is MLI.


First, that means you don't have an extra version of Starship with extra-large tanks. Second, the fuel depot would have the hardware required to mate to the QD ports on a Starship and do the refueling, so you wouldn't need a version of Starship

Developing an entirely new vehicle from scratch is worse than reusing non-recurring engineering.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/29/2022 05:36 pm
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.
I wouldn't say no advantage. [...]

I mean there's no advantage to the selectivity, ie the perfect paint would completely reject all wavelengths. There is no intentional cutoff.

Same principle, but in reverse: https://en.wikipedia.org/wiki/Selective_surface

The closest thing that exists in real life is MLI.
There is some advantage: a selective coating can reject the wavelengths of sunlight but still efficiently radiate heat to space in so-called thermal IR wavelengths.

There are thermodynamic limits to this, but it is possible.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/29/2022 05:45 pm
Developing an entirely new vehicle from scratch is worse than reusing non-recurring engineering.
Sure, but I don't think what I'm describing is a vehicle. It's little more than two tanks, two pumps, and two hoses. In the basic form, it doesn't even leave the cargo bay of a Starship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 08/29/2022 05:47 pm
In thinking about the case of each tanker shipping to LEO ~200t of prop. And the HLS and all the tankers have same size tanks of 1500t. So that they can lift the 200t to LEO and the HLS can have the extra DV to not have any mass restrictions problems to met the mission requirements of leaving LEO going to NRHO, leaving HRHO to Surface, return from surface back to NRHO after having been on the surface for 30+ days and all of this time having boil off occur.

It is much more simple to manage the amount of prop than many other items. A side note is that since the collector has only 1500t tanks but the tankers bring up 1600t total there is 100t of prop that can boil away while waiting for all those tankers to dock to the collector so it can dock to the HLS to transfer prop.

Thus special measures to reduce boil off may not be applicable for the prop collector at LEO. But for the depot function at HRHO there is likely some reasoning for some level of additional boil off mitigation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/29/2022 05:48 pm
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.
I wouldn't say no advantage. [...]

I mean there's no advantage to the selectivity, ie the perfect paint would completely reject all wavelengths. There is no intentional cutoff.

Same principle, but in reverse: https://en.wikipedia.org/wiki/Selective_surface

The closest thing that exists in real life is MLI.
There is some advantage: a selective coating can reject the wavelengths of sunlight but still efficiently radiate heat to space in so-called thermal IR wavelengths.

There are thermodynamic limits to this, but it is possible.

Nope. That's exactly what I'm disclaiming.

The radiated heat into space is miniscule. It is totally swamped by the extra thermal IR absorbed from the Earth.

Note that we've all been using a "single particle" model here. If you have a surface that only ever points toward space and never has a view of the Earth, then the story changes, obviously.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/29/2022 05:54 pm
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.
I wouldn't say no advantage. [...]

I mean there's no advantage to the selectivity, ie the perfect paint would completely reject all wavelengths. There is no intentional cutoff.

Same principle, but in reverse: https://en.wikipedia.org/wiki/Selective_surface

The closest thing that exists in real life is MLI.
There is some advantage: a selective coating can reject the wavelengths of sunlight but still efficiently radiate heat to space in so-called thermal IR wavelengths.

There are thermodynamic limits to this, but it is possible.

Nope. That's exactly what I'm disclaiming.

The radiated heat into space is miniscule. It is totally swamped by the extra thermal IR absorbed from the Earth.
Sure, it depends on how good the solar rejection is, and the orbit you’re in. If
You body is at human body temperature, you’re emitting 500W/m^2 and receiving 1350W/m^2 from the sun. I think it’s perfectly doable to make a coating that isn’t just insulating but actually allows heat to escape. At 6 bar, methane’s boiling point is 138K, so you emit a factor of 25 times less. But STILL feasible to cool if your emissivity is 100 times higher at thermal IR than sunshine wavelengths.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/29/2022 05:56 pm
Developing an entirely new vehicle from scratch is worse than reusing non-recurring engineering.
Sure, but I don't think what I'm describing is a vehicle. It's little more than two tanks, two pumps, and two hoses. In the basic form, it doesn't even leave the cargo bay of a Starship.

So... a box within a box?

Delete the tanks, use Starship's. Now you just two pumps and two hoses. That's a win.

(Now obviously it's not that simple since you need insulation too, but you didn't list that in your list.  ;))
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/29/2022 06:03 pm
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.
I wouldn't say no advantage. [...]

I mean there's no advantage to the selectivity, ie the perfect paint would completely reject all wavelengths. There is no intentional cutoff.

Same principle, but in reverse: https://en.wikipedia.org/wiki/Selective_surface

The closest thing that exists in real life is MLI.
There is some advantage: a selective coating can reject the wavelengths of sunlight but still efficiently radiate heat to space in so-called thermal IR wavelengths.

There are thermodynamic limits to this, but it is possible.

Nope. That's exactly what I'm disclaiming.

The radiated heat into space is miniscule. It is totally swamped by the extra thermal IR absorbed from the Earth.
Sure, it depends on how good the solar rejection is, and the orbit you’re in. If
You body is at human body temperature, you’re emitting 500W/m^2 and receiving 1350W/m^2 from the sun. I think it’s perfectly doable to make a coating that isn’t just insulating but actually allows heat to escape. At 6 bar, methane’s boiling point is 138K, so you emit a factor of 25 times less.

I ran numbers at 50 K, as requested. I can re-run them at this temperature. Give me a bit.

Edit: the boiling point of LOX at 6 bar is only 111.2 K. I'll run that too.

But STILL feasible to cool if your emissivity is 100 times higher at thermal IR than sunshine wavelengths.

That's the problem, it's not just the sunshine wavelengths. That's what was throwing us off before.

You also have to add in the incoming thermal IR from the Earth (if your surface has any non-zero view factor).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/29/2022 06:21 pm
Delete the tanks, use Starship's. Now you just two pumps and two hoses. That's a win.
But is that a win? Now you've got another kind of Starship--one that can't be used for any other purpose. Is that desirable? Also, what kind of plumbing do you need to draw propellant from the bottoms of the LOX and CH4 tanks? That is, where do the attachments go and how does the docking work? Ideally, the attachments would be at the bottom of the tanks, since that's where the pipes will need to attach. But can you add a "male QD" at the bottom of a Starship without major changes? And if you try to put it anywhere else, don't you get unacceptable amounts of extra plumbing?

Putting it all in the cargo bay just seems a lot cleaner to me. (Not that I'm an expert, of course.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 08/29/2022 06:24 pm
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.
I wouldn't say no advantage. [...]

I mean there's no advantage to the selectivity, ie the perfect paint would completely reject all wavelengths. There is no intentional cutoff.

Same principle, but in reverse: https://en.wikipedia.org/wiki/Selective_surface

The closest thing that exists in real life is MLI.
There is some advantage: a selective coating can reject the wavelengths of sunlight but still efficiently radiate heat to space in so-called thermal IR wavelengths.

There are thermodynamic limits to this, but it is possible.

Nope. That's exactly what I'm disclaiming.

The radiated heat into space is miniscule. It is totally swamped by the extra thermal IR absorbed from the Earth.
Sure, it depends on how good the solar rejection is, and the orbit you’re in. If
You body is at human body temperature, you’re emitting 500W/m^2 and receiving 1350W/m^2 from the sun. I think it’s perfectly doable to make a coating that isn’t just insulating but actually allows heat to escape. At 6 bar, methane’s boiling point is 138K, so you emit a factor of 25 times less.

I ran numbers at 50 K, as requested. I can re-run them at this temperature. Give me a bit.

Edit: the boiling point of LOX at 6 bar is only 111.2 K. I'll run that too.

But STILL feasible to cool if your emissivity is 100 times higher at thermal IR than sunshine wavelengths.

That's the problem, it's not just the sunshine wavelengths. That's what was throwing us off before.

You also have to add in the incoming thermal IR from the Earth (if your surface has any non-zero view factor).
I already ran the numbers. That’s in my post. It’s just proportional to absolute temperature to the fourth power. 5.67e-8W/(m^2*Kelvin^4) is the Proportionality constant, the Stefan-Boltzmann constant.

Might be in an elliptical orbit or something so Earthshine isn’t bad, averaged overall the whole orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 08/29/2022 06:31 pm
...for a Starship in LEO, there is no completely passive ZBO solution. Not at the moment, anyway. There must be some cryocooler, with all the extra hardware that implies.

Well, for a dedicated depot Starship, you might use a passive V-Groove radiator, as on SPHEREx.  In fact, SPHEREx cools passively to LOX freezing point, past depot requirement.

--

Conceivably, three Mylar sheets might be stored in a tiny depot payload "cap".  In LEO, they're released down the hull, under slow rotation with small propulsive acceleration.

The deployed form requires little structure; e.g., a few magnets at the depot base, plus cables.  Cables could be wound on spools on the inner sheet edge.  As the sheet descends past magnets under propulsive acceleration, each cable's iron tip attaches to a magnet.  Then propulsion ends, rotation rate increases, and cables unwind, to deploy the inner sheet.  The other sheets deploy on further cables.  No cryocooler, coatings, or insulation required.

QD is exposed at depot base.  To initiate filling, the depot halts rotation for QD access.  (Inverted positioning for QD attachment?)

A simple formula to calculate V-Groove passive cryogenic cooling is given in Bhandari et al. 2020.

Refs.

Bhandari, P., Moore, B., Bolton, D. and Aboobaker, A., 2020, July. Design and Analysis of V-Groove Passive Cryogenic Radiators for Spaceborne Telescopes & Instruments. (https://ttu-ir.tdl.org/bitstream/handle/2346/86248/ICES-2020-24.pdf?sequence=1) 2020 International Conference on Environmental Systems.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/29/2022 06:51 pm
Delete the tanks, use Starship's. Now you just two pumps and two hoses. That's a win.
But is that a win? Now you've got another kind of Starship--one that can't be used for any other purpose. Is that desirable?

Versus some completely new hardware of equal size that is also not usable for any other purpose? Yeah.

Also, what kind of plumbing do you need to draw propellant from the bottoms of the LOX and CH4 tanks? That is, where do the attachments go and how does the docking work? Ideally, the attachments would be at the bottom of the tanks, since that's where the pipes will need to attach. But can you add a "male QD" at the bottom of a Starship without major changes? And if you try to put it anywhere else, don't you get unacceptable amounts of extra plumbing?

Don't worry. It's just two tubes and two pumps!  :)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/29/2022 07:00 pm
I already ran the numbers. That’s in my post. It’s just proportional to absolute temperature to the fourth power. 5.67e-8W/(m^2*Kelvin^4) is the Proportionality constant, the Stefan-Boltzmann constant.

I'm doing something different, actually. See upthread.

Might be in an elliptical orbit or something so Earthshine isn’t bad, averaged overall the whole orbit.

Might be. Many won't be.

It would be a huge limitation if there was no good solution for LEO.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/29/2022 09:39 pm
If there's no advantage to using a selective solar coating (which seems to be true), then we're back to MLI.
I wouldn't say no advantage. At a minimum, you could use the paint (not the tiles) on the parts of the ship that don't have to be jet black and that would reject 90% of the heat from sun exposure at very little cost. But it definitely doesn't buy you ZBO. Not with a standard Starship, anyway.

At this point, I'd like to talk a little bit about exactly how the fuel is going to be transported. (Apologies if this has already been discussed. Extra apologies if it's already been discarded with good reason!) :-)

I know Elon doesn't like depos, but I'm starting to think it makes more sense to have a special "fuel depot" (for want of a better word) that a standard-issue Starship would carry as cargo rather than have a special tanker version of Starship. That would mean that two Starships would never dock to each other directly--not for purposes of fueling, anyway.

My reasoning is as follows:

First, that means you don't have an extra version of Starship with extra-large tanks. Second, the fuel depot would have the hardware required to mate to the QD ports on a Starship and do the refueling, so you wouldn't need a version of Starship with that equipment on it somewhere. (Or, worse, add that hardware to every Starship.)

The basic fuel depot would just be fuel tanks, the docking attachment, and pumps. It wouldn't ever leave the cargo bay of the Starship that brought it up to orbit. It's the sort of thing you'd use to refuel the Lunar Starship, assuming that only needs one extra load of fuel. You'd launch the two Starships at about the same time so boiloff wouldn't be a big concern.

A more complex fuel depot would be larger and have a cryocooler, radiator, extra insulation, and QD ports identical to those on a Starship. It would be put in orbit by itself and get filled up on multiple visits by Starships carrying basic depots full of fuel and LOX. (Note that you wouldn't need to develop this until after the lunar missions.)

That sort of assembly probably would benefit from extensive use of Solar White tiles. Yes, it would still need a cryocooler, but the tiles ought to reduce the mass of the cooler, the radiator, and the solar panels needed to power it.

You would cope with the orbital precession problem by timing the initial orbit and the refueling orbits so the depot is in the right place at the right time to refuel the Starship dedicated to the mission. After the mission Starship departed, you could send up another Starship to retrieve the empty depot.

If you wanted to use the same depot for multiple missions, you could use something like the RAAN-Agnostic 3-Burn Departure Methadology for Deep Space Missions from LEO Depots (https://selenianboondocks.com/2018/09/aas-paper-review-raan-agnostic-3-burn-departure-methodology-for-deep-space-missions-from-leo-depots-part-1-of-2/) that has been talked about earlier.

Again, apologies if this has already been debated/discarded.
There has been discussion on several of the ideas you've offered up.

Keeping all the transfer hardware kit as a cargo has good points and bad points. It all boils down to the classic "it depends."

First off, a disclaimer: I am opposed to something that stays inside a cargo bay. It could work if the ship carrying it is nosed up between the tanker and the receiving ship and just acts as a fuel pump but that has safety issues for a crewed receiver. If it has its own tankage so it can act as an accumulator and that tankage is separate from the ship it is mounted in, it's not enough. If it uses the ships tanks, it needs some one off plumbing, which moves it in the direction of a special build. Then at transfer time it will need to pump up hill just to get the props to the point of transfer.

An important point you addressed is GSE plate gender mating. The four solutions so far discussed were: 1) the tanker deploying a gender bender on top of its GSE plate; 2) a tanker having a second, gender bent GSE plate; 3) the tanker being already gender bent and a temporary adapter mounted on the ground GSE for launch; 4) A free flyer with all transfer kit mounted and two gender bent GSE fittings.

If memory serves I suggested #1, and I now formally renounce it as 3 & 4 seem better ideas. Number 2 calls for extra mass and a small bit of extra plumbing and valving. IMO, 3 is a better approach. Number 3 requires an accumulator to have specialized hardware which makes it a depot. IMO this is the best approach if a formal depot is decided upon. Number 4 is much like what you suggested and IMO is the way to go if a specialized depot is to be avoided.

The idea of snaking a couple of hoses is not going to work. At minimum there also needs to be ullage transfer to minimize pump size and power, with the benefit of minimizing wastage. Past that, looking long term, it's not unreasonable to expect a transfer of any other fluids needed on later builds, and juicing up batteries has a certain appeal if a transfer campaign get drawn out. That and to my knowledge, fluid transfer has never been done between two bodies not firmly connected. Might as well take advantage of the GSE hardware that's already there and not get too fancy.

A minimal free flyer would be a boxy structure with two interconnected gender bent GSE plates, pumps and a small amount of propellant tankage and thrusters to allow it to maneuver onto the GSE plate of a tanker acting as an accumulator. The tanker would then take over maneuvering to mate the free flyer with another tanker for transfer. We'd see two SS's dorsal to dorsal with the free flyer connecting them at the GSE plates.

The debate on how tightly the two ships would need to be bound was ongoing with no consensus that I remember. This issue applies to all transfer schemes, not just the free flyer.

A possible strong point for the free flyer is it's central position offers a good location for ullage settling thrust. That might imply a larger gimbaled engine than that needed for its own maneuvering. Maybe not. A classic "it depends".

Moving up from a minimal free flyer adds PV, a radiator and cryocooler. Maybe MLI. This would call for a structure of some sort, maybe a telescoping truss, that would extend parallel to the ship(s) it's attached to. This has the advantage of offering more attachment points if they're needed.

A big disadvantage here is figuring out what to do with all the foldy stuff when two ships are involved. PV and radiators could probably be swing out of the way but MLI would be a challenge, but maybe not needed if the cryocooler is sized big enough, which leads us back to the current discussion on expected boiloff.

A free flyer could stay attached to one tanker for use over many refueling campaigns or be bought back into the cargo bay for return after one campaign. They might have to jettison the floppy stuff for a return. James Web was awesome but can you imagine the teams reaction if they got orders to design for refolding and redeploy?

I've had a dog in this fight but I've tried to keep this summary as dispassionate and neutral (with some identified opinion interjected) as possible. Summaries are useful in pulling together and clarifying discussions scattered through many pages. If I've missed, I apologize. Corrections welcomed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/29/2022 10:10 pm
Thanks! That was hugely helpful. For the first time, I think I actually understand a few of the terms I saw in discussions further upthread.

I am opposed to something that stays inside a cargo bay. It could work if the ship carrying it is nosed up between the tanker and the receiving ship and just acts as a fuel pump but that has safety issues for a crewed receiver.
What are the safety issues? Is there a real concern that a vehicle being refilled might explode?

An important point you addressed is GSE plate gender mating. The four solutions so far discussed were: 1) the tanker deploying a gender bender on top of its GSE plate; 2) a tanker having a second, gender bent GSE plate; 3) the tanker being already gender bent and a temporary adapter mounted on the ground GSE for launch; 4) A free flyer with all transfer kit mounted and two gender bent GSE fittings.

Hmm. I actually visualized a "free flyer" depot with two sets of fittings: one male and one female. My thought was that you'd fill it up through the female fittings and it would fill a Starship via the male fittings. A basic depot would get filled on the ground before loading into the cargo bay. An advanced depot would get repeated loads of fuel while in orbit and then (when full) would unload into a Starship. There would never be a time when two Starships were connected directly to each other. (Although the case where the depot is secured inside the cargo bay is, admittedly, not very different from that.) You could even chain multiple depots together, if there were any advantage to be had from that. (E.g. greater capacity but sharing the same power source/coolant or something.)

The idea of snaking a couple of hoses is not going to work. At minimum there also needs to be ullage transfer to minimize pump size and power, with the benefit of minimizing wastage. Past that, looking long term, it's not unreasonable to expect a transfer of any other fluids needed on later builds, and juicing up batteries has a certain appeal if a transfer campaign get drawn out. That and to my knowledge, fluid transfer has never been done between two bodies not firmly connected. Might as well take advantage of the GSE hardware that's already there and not get too fancy.

I guess I'm just concerned that the GSE hardware either won't be long enough to reach or else it'll protrude too much at launch time. Perhaps that's a silly thing to worry about, but I've had trouble visualizing how it would actually work. Likewise, the free flyer with two male GSE adaptors that bridges two Starships troubles me because I don't see how it ever gets into place without someone in a space suit moving it around. Of course, given the name, I guess it could have its own propulsion and guidance system, but, man, that seems really complicated. At that point, it really is a vehicle. (It's a beautifully symmetrical solution, though.)

Thanks again for your clear and helpful responses.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/29/2022 11:07 pm
Delete the tanks, use Starship's. Now you just two pumps and two hoses. That's a win.
But is that a win? Now you've got another kind of Starship--one that can't be used for any other purpose. Is that desirable?

Orbital refuelling seems to be a core part of Starship's standard operating procedure. Tankers are therefore going to get plenty of use. There's no savings in having a special payload-bay tank that goes in a regular cargo Starship. Even if there was an advantage to having separate tanks (simplified plumbing, say) then it would still make sense to have dedicated tankers that have that plumbing built in.

Personally, I like box-in-a-box. There's a reason shipping containers revolutionised transport. But it's against SpaceX's design philosophy, and dedicated tankers aren't going to be sitting around wasted, so there's no reason for it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/29/2022 11:07 pm
I am opposed to something that stays inside a cargo bay. It could work if the ship carrying it is nosed up between the tanker and the receiving ship and just acts as a fuel pump but that has safety issues for a crewed receiver.
What are the safety issues? Is there a real concern that a vehicle being refilled might explode?

A loss of pressure during refuelling could cause a BLEVE.

BLEVE's are bad.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/29/2022 11:15 pm
What I would say is that, for a Starship in LEO, there is no completely passive ZBO solution. Not at the moment, anyway. There must be some cryocooler, with all the extra hardware that implies. The best that Solar White can do for you is reduce the mass of the cryocooler.
Or use boil-off. [...]
Maybe I'm missing something, but how is this different from just doing nothing?

I was reacting to "There must be some cryocooler". Not necessarily. Just because you can't passively cool to ZBO, doesn't mean you can't passively reflect enough sunlight/Earthlight to reach acceptable levels of boil-off.

The choice isn't "reflect/emit until ZBO" or else "you must use a cryocooler". There's a third choice, "propellant is cheap, reflect enough until you can tolerate the boil-off losses". There's also a continuum with the cryocooler, where you tolerate a certain amount of boil-off to reduce the demand on the cryocooler, or as part of the cooling loop. (For example, using methalox boil-off to dump heat from a LOx-only cryocooler. Effectively an open-cycle radiator.)

Hence you are trading "reflective coatings", "MLI", "cryocooler", "boil-off". The mass optimum would be some combination. The development-cost optimum for first-gen depots would favour non-MLI reflective coatings plus boil-off.

(Aside: In the boil-off scenario, venting wouldn't be passive, however. You want to actively pump boil-off gas into a low pressure expansion chamber that lets it boil-off at lower temperatures, to avoid having the entire propellant load hovering around the boiling point. This also allows you to cool the LOx via just methane boil-off. (Methane's latent heat content is greater than oxygen's, so it's more mass-efficient.) Thankfully, the energy calcs are the same.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/30/2022 06:53 am
Thanks! That was hugely helpful. For the first time, I think I actually understand a few of the terms I saw in discussions further upthread.

I am opposed to something that stays inside a cargo bay. It could work if the ship carrying it is nosed up between the tanker and the receiving ship and just acts as a fuel pump but that has safety issues for a crewed receiver.
What are the safety issues? Is there a real concern that a vehicle being refilled might explode?

An important point you addressed is GSE plate gender mating. The four solutions so far discussed were: 1) the tanker deploying a gender bender on top of its GSE plate; 2) a tanker having a second, gender bent GSE plate; 3) the tanker being already gender bent and a temporary adapter mounted on the ground GSE for launch; 4) A free flyer with all transfer kit mounted and two gender bent GSE fittings.

Hmm. I actually visualized a "free flyer" depot with two sets of fittings: one male and one female. My thought was that you'd fill it up through the female fittings and it would fill a Starship via the male fittings. A basic depot would get filled on the ground before loading into the cargo bay. An advanced depot would get repeated loads of fuel while in orbit and then (when full) would unload into a Starship. There would never be a time when two Starships were connected directly to each other. (Although the case where the depot is secured inside the cargo bay is, admittedly, not very different from that.) You could even chain multiple depots together, if there were any advantage to be had from that. (E.g. greater capacity but sharing the same power source/coolant or something.)

The idea of snaking a couple of hoses is not going to work. At minimum there also needs to be ullage transfer to minimize pump size and power, with the benefit of minimizing wastage. Past that, looking long term, it's not unreasonable to expect a transfer of any other fluids needed on later builds, and juicing up batteries has a certain appeal if a transfer campaign get drawn out. That and to my knowledge, fluid transfer has never been done between two bodies not firmly connected. Might as well take advantage of the GSE hardware that's already there and not get too fancy.

I guess I'm just concerned that the GSE hardware either won't be long enough to reach or else it'll protrude too much at launch time. Perhaps that's a silly thing to worry about, but I've had trouble visualizing how it would actually work. Likewise, the free flyer with two male GSE adaptors that bridges two Starships troubles me because I don't see how it ever gets into place without someone in a space suit moving it around. Of course, given the name, I guess it could have its own propulsion and guidance system, but, man, that seems really complicated. At that point, it really is a vehicle. (It's a beautifully symmetrical solution, though.)

Thanks again for your clear and helpful responses.
You're right about the safety issue with kit in nose design. I was having cranial flatulence. Still, it adds another GSE hookup and has to pump up hill. The plumbing has to reach down to the main tanks. I just don't see any advantage over the dedicated depot or the free flyer except for deploying, and enough downside to give it the hairy eyeball.


I wouldn't worry too much about a free flyer having guidance and control issues. It needn't move fast, so not much dV is needed. Aligning itself and attaching doesn't call for anything a Dragon doesn't already do routinely. The computers they use for engine controllers would probably be overkill. The optics could be GoPro.


Interesting factoid: one ton of propellant takes up 0.016m of tank height. A nominal 150t of propellant would need a 2.4m tank stretch. Just a matter of moving domes, wiring, plumbing and vents. This would make the tankers their own variant and save the weight and operational inconvenience of cargo bay tanks. Some see that as a problem. I don't. Opinion: Hardware rich won't end when operations start. Tanker use will be slow at first but my money says use will pick up as the utility proves itself. A 40t probe to the Oort cloud? No problemo.


The free flyer would need two of one type of GSE plate. It would exactly mimic the plate on the ground GSE, which would be exactly what the ship GSE plate expects to mate up with. The gender of a quick disconnect doesn't dictate flow direction. Gas station air line QD's line up the way they do for the convenience of having a barb on the air chuck or impact driver instead of a coupler.


There was discussion about how much of a stand-off would be needed between two ships. Opinions on the high end were a couple of meters. On the low end, zip. I'd say the minimum with the free flyer would be however much it would take to route all the internal crossovers, L to R and R to L, and pumps. Maybe around 1.5m? For a depot, maybe mount the gender bent QD on a pantograph with flex hose behind it for some minimal clearance. This would be too mushy for a firm connection between the ships so either another mechanism or some struts or something.


But then there were those who felt that a firm connection wasn't necessary...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/30/2022 06:55 am
I already ran the numbers. That’s in my post. It’s just proportional to absolute temperature to the fourth power. 5.67e-8W/(m^2*Kelvin^4) is the Proportionality constant, the Stefan-Boltzmann constant.

I'm doing something different, actually. See upthread.

Ok, so I ran the numbers and it's confirmed. At both LOX and CH4 boiling points, there's no advantage to using a selective surface assuming you're in LEO and your surface has the average Earth view factor.

Curves for LOX (111.2 K) (https://www.wolframalpha.com/input?i=plot%3A+y%3D5.9552e28%2F%28x%5E5*%28exp%2848938%2Fx%29+-+1%29%29+%2B+4.0459e23%2F%28x%5E5+*+%28exp%282740.5%2Fx%29+-+1%29%29%2C+y+%3D+1.19104e29%2F%28x%5E5+*+%28exp%28129386%2Fx%29+-+1%29%29++from+x%3D0+to+50000)

Curves for CH4 (138.4 K) (https://www.wolframalpha.com/input?i=plot%3A+y%3D5.9552e28%2F%28x%5E5*%28exp%2848938%2Fx%29+-+1%29%29+%2B+4.0459e23%2F%28x%5E5+*+%28exp%282740.5%2Fx%29+-+1%29%29%2C+y+%3D+1.19104e29%2F%28x%5E5+*+%28exp%28129386%2Fx%29+-+1%29%29++from+x%3D0+to+50000)

The upper curve is the combined solar + Earth IR incoming spectrum, modeled as blackbodies. The lower curve is the blackbody emission spectrum. Horizontal axis is wavelength in nanometers, vertical axis is arbitrarily scaled but area under the curves is proportional to power per square meter.

As you can see, the curves never cross. That means there's no upper wavelength cutoff beyond which a totally emissive blackbody would "beat" a totally reflective surface.



Btw SPHEREx is precisely the special case I described earlier: the cone always points up, so it never has a view of the Earth. It also needs to be located in a dusk/dawn polar orbit, such that it never has a view of the Sun. There are downsides to using a disk/dawn orbit for a depot: it's less efficient than launching eastward, and it has less flexibility for targeting a specific right ascension for departure.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/30/2022 01:35 pm
As you can see, the curves never cross. That means there's no upper wavelength cutoff beyond which a totally emissive blackbody would "beat" a totally reflective surface.
Does such a surface exist, though?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/30/2022 03:17 pm
You're right about the safety issue with kit in nose design. I was having cranial flatulence. Still, it adds another GSE hookup and has to pump up hill. The plumbing has to reach down to the main tanks. I just don't see any advantage over the dedicated depot or the free flyer except for deploying, and enough downside to give it the hairy eyeball.
I still don't think I'm clearly conveying what I'm visualizing. Let me give it another try. First, remember that I'm assuming no modifications to the cargo Starship whatsoever. It has a female GSE hookup down at the bottom of the tanks and that's all. There is no extra plumbing running up to the nose of it.

The basic depot has a male GSE hookup. It fits in the cargo hold of a standard starship and is firmly braced in there. It can move out a little bit (relative to the cargo hold), probably on rails or something, so it's always firmly attached. It does not have any external plumbing.

When cargo starship A refills starship B, A opens the cargo doors, lines up the depot GSE with B's GSE and connects them. Once the cargo doors are open, the depot rolls out, and the two GSE's can mate without any other contact between the two vehicles.

Both vehicles do a gentle ullage burn while the propellants are transferred. Because the docking is asymmetrical, the ullage burns will have to be calculated to generate net zero torque. Note, though, that even a symmetrical docking will have this problem because the masses of the two starships won't be the same and the center of mass will change dynamically during the refilling operation.

After the depot is depleted, they disconnect and the depot retracts. A goes home, and B goes on its mission. (Assuming it only needed one load.)

As a result, there's no plumbing that reaches down to the main tanks; they are refilled from the bottom, just as they are on the launch pad.

Is that clearer?
Interesting factoid: one ton of propellant takes up 0.016m of tank height. A nominal 150t of propellant would need a 2.4m tank stretch. Just a matter of moving domes, wiring, plumbing and vents. This would make the tankers their own variant and save the weight and operational inconvenience of cargo bay tanks. Some see that as a problem. I don't. Opinion: Hardware rich won't end when operations start. Tanker use will be slow at first but my money says use will pick up as the utility proves itself. A 40t probe to the Oort cloud? No problemo.
I just can't help thinking that if you're trying to save money by mass-producing these things, it will help a lot to have fewer versions. Right now I'm seeing a cargo version, a crew version, a tanker version, a lunar version, etc. and that just seems like a problem to me. Obviously it's easier if the lunar version is just the crew version minus the last steps (that attach tiles), and maybe the cargo version and the crew version share all the work up to the point where the upper part of the vehicle is configured, but the tanker version seems like a very different work flow. Also, there probably won't be very many lunar and crew versions built, so maybe it's okay that they're expensive to build; almost everything will be cargo starships, after all. But you'd likely need a lot of tankers.

Hence my thinking that it's a win if the basic depot is as simple as possible and just fits into the cargo hold. Again, I'm no expert here--just trying to think the whole thing through.
The free flyer would need two of one type of GSE plate. It would exactly mimic the plate on the ground GSE, which would be exactly what the ship GSE plate expects to mate up with. The gender of a quick disconnect doesn't dictate flow direction.
Yeah, I realize that. I had been thinking you'd want a female GSE on any depot just so you could fill it on the ground using the same hardware that fills a starship, but perhaps that's unnecessary. Also, it would let you chain multiple depots together--if that had any value. It also occurred to me that it might be easier to make pumps that only work in one direction, but that's sheer speculation on my part.

In that case, though, why does a depot need two GSE's? Why not just one male one that it uses to fill up from one starship (or on the ground) and then later pump the propellant into another one? It only seems you'd need two attachments if you wanted to be connected to two starships at the same time, but with a depot, I don't think you'd ever want to do that.

If a single refill (of 150 tons) is sufficient for the lunar missions, then a basic depot is all that would be required for a while. But I'm still fascinated by the idea of a "jumbo depot" that could hold much more propellant indefinitely.

Using your 0.016 m/mT figure, it would seem that an empty jumbo depot that filled the 18m cargo bay could hold 1125 tons of propellant. It would take 7.5 flights to fill such a thing, and three such depots could completely refill a starship (if there were a reason to do that--maybe if you wanted to take months hauling propellant up so you could quickly fill up a fleet of starships all heading to Mars in the same launch window).

If we were going to construct such a "depot complex," I could imagine dividing the functionality between "storage modules" (just tanks and plumbing) and a single "command module." The command module would have solar panels, cryocooler, radiator, communication links, ullage engine etc. (but minimal fuel tanks) that would attach to one or more storage modules.

But that's starting to be pretty pie-in-the-sky. :-)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Hog on 08/30/2022 04:14 pm
BLEVE=Boiling Liquid Expanding Vapour Explosion
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: jimvela on 08/30/2022 04:54 pm
As you can see, the curves never cross. That means there's no upper wavelength cutoff beyond which a totally emissive blackbody would "beat" a totally reflective surface.
Does such a surface exist, though?

No, there's no such thing as a totally emissive blackbody nor a totally reflective surface.

I don't think the analysis in the thread has touched on all the available coatings. 
Just from my own experience, for example, there are things like BIRB which is used on cryogenic telescope missions.

I'm sure whomever does the work on the Starship in-space cryogenic systems will be aware of these kinds of options.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Comga on 08/30/2022 07:15 pm
(snip)
Btw SPHEREx is precisely the special case I described earlier: the cone always points up, so it never has a view of the Earth. It also needs to be located in a dusk/dawn polar orbit, such that it never has a view of the Sun. There are downsides to using a disk/dawn orbit for a depot: it's less efficient than launching eastward, and it has less flexibility for targeting a specific right ascension for departure.

Jumping in late to this discussion, so apologies if this has been said before.
The dusk-dawn SSO orbit is a limiting case, one that is particularly well suited to all-sky surveys like SPHEREX and WISE (https://www.nasa.gov/mission_pages/WISE/mission/index.html).
(And particularly unsuited to a depot from which to start interplanetary trajectories, as you said.)
There, the angle from the local vertical to the Sun is always near 90 degrees.  (probably >90 minus 23.5 degree inclination of the Earth's rotation with respect to its orbit minus ~7 degrees from the inclination to the pole.)
However, the same goal of never pointing the radiator at the Sun can be achieved at lower inclinations for which the precession does not match the orbital rate of the Earth, i.e. with a variable beta angle.

Take the other extreme, an equatorial orbit.
A depot in equatorial orbit oriented to the Local Velocity, Local Horizon (LVLH) would need an Earth shield and a sunshield.
This could be cylindrical, with axis pointed at the orbit poles, perpendicular to velocity and to the local zenith.
A cryoradiator could point out either end.
However, it's Field Of View would be limited to 90 - 23.5 =~66 degrees from the axis, to prevent the Sun from shining into it regardless of season.
(The beta angle is zero at the equinoxes, when the pane of the equator passes through the Sun, and grows to ~22 deg at the solstices.)
Such radiators are commonly made with "Etendue (https://www.eckop.com/resources/optics/illumination/etendue/) converters", non-imaging optics, sometimes implemented with Winston cones (https://scienceworld.wolfram.com/physics/WinstonCone.html).  (See some one dimensional implementations here (https://www.winstonconeoptics.com/).)

For higher inclination orbits, the angle is reduced, and by conservation of Etendue, the far end of the "cone" gets proportionally larger.
So it's a trade between making the inclination high enough to go through the departure points and restricting the size.

So it's not just the spectral properties that influence the solar and Earth heating and the thermal radiation. 
The geometry is also important.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/30/2022 07:20 pm
You're right about the safety issue with kit in nose design. I was having cranial flatulence. Still, it adds another GSE hookup and has to pump up hill. The plumbing has to reach down to the main tanks. I just don't see any advantage over the dedicated depot or the free flyer except for deploying, and enough downside to give it the hairy eyeball.
I still don't think I'm clearly conveying what I'm visualizing. Let me give it another try. First, remember that I'm assuming no modifications to the cargo Starship whatsoever. It has a female GSE hookup down at the bottom of the tanks and that's all. There is no extra plumbing running up to the nose of it.

The basic depot has a male GSE hookup. It fits in the cargo hold of a standard starship and is firmly braced in there. It can move out a little bit (relative to the cargo hold), probably on rails or something, so it's always firmly attached. It does not have any external plumbing.

When cargo starship A refills starship B, A opens the cargo doors, lines up the depot GSE with B's GSE and connects them. Once the cargo doors are open, the depot rolls out, and the two GSE's can mate without any other contact between the two vehicles.

Both vehicles do a gentle ullage burn while the propellants are transferred. Because the docking is asymmetrical, the ullage burns will have to be calculated to generate net zero torque. Note, though, that even a symmetrical docking will have this problem because the masses of the two starships won't be the same and the center of mass will change dynamically during the refilling operation.

After the depot is depleted, they disconnect and the depot retracts. A goes home, and B goes on its mission. (Assuming it only needed one load.)

As a result, there's no plumbing that reaches down to the main tanks; they are refilled from the bottom, just as they are on the launch pad.

Is that clearer?
Interesting factoid: one ton of propellant takes up 0.016m of tank height. A nominal 150t of propellant would need a 2.4m tank stretch. Just a matter of moving domes, wiring, plumbing and vents. This would make the tankers their own variant and save the weight and operational inconvenience of cargo bay tanks. Some see that as a problem. I don't. Opinion: Hardware rich won't end when operations start. Tanker use will be slow at first but my money says use will pick up as the utility proves itself. A 40t probe to the Oort cloud? No problemo.
I just can't help thinking that if you're trying to save money by mass-producing these things, it will help a lot to have fewer versions. Right now I'm seeing a cargo version, a crew version, a tanker version, a lunar version, etc. and that just seems like a problem to me. Obviously it's easier if the lunar version is just the crew version minus the last steps (that attach tiles), and maybe the cargo version and the crew version share all the work up to the point where the upper part of the vehicle is configured, but the tanker version seems like a very different work flow. Also, there probably won't be very many lunar and crew versions built, so maybe it's okay that they're expensive to build; almost everything will be cargo starships, after all. But you'd likely need a lot of tankers.

Hence my thinking that it's a win if the basic depot is as simple as possible and just fits into the cargo hold. Again, I'm no expert here--just trying to think the whole thing through.
The free flyer would need two of one type of GSE plate. It would exactly mimic the plate on the ground GSE, which would be exactly what the ship GSE plate expects to mate up with. The gender of a quick disconnect doesn't dictate flow direction.
Yeah, I realize that. I had been thinking you'd want a female GSE on any depot just so you could fill it on the ground using the same hardware that fills a starship, but perhaps that's unnecessary. Also, it would let you chain multiple depots together--if that had any value. It also occurred to me that it might be easier to make pumps that only work in one direction, but that's sheer speculation on my part.

In that case, though, why does a depot need two GSE's? Why not just one male one that it uses to fill up from one starship (or on the ground) and then later pump the propellant into another one? It only seems you'd need two attachments if you wanted to be connected to two starships at the same time, but with a depot, I don't think you'd ever want to do that.

If a single refill (of 150 tons) is sufficient for the lunar missions, then a basic depot is all that would be required for a while. But I'm still fascinated by the idea of a "jumbo depot" that could hold much more propellant indefinitely.

Using your 0.016 m/mT figure, it would seem that an empty jumbo depot that filled the 18m cargo bay could hold 1125 tons of propellant. It would take 7.5 flights to fill such a thing, and three such depots could completely refill a starship (if there were a reason to do that--maybe if you wanted to take months hauling propellant up so you could quickly fill up a fleet of starships all heading to Mars in the same launch window).

If we were going to construct such a "depot complex," I could imagine dividing the functionality between "storage modules" (just tanks and plumbing) and a single "command module." The command module would have solar panels, cryocooler, radiator, communication links, ullage engine etc. (but minimal fuel tanks) that would attach to one or more storage modules.

But that's starting to be pretty pie-in-the-sky. :-)
I'd like to do the embedded quotes but between fat fingers and a small phone... Bear with me.

Sorry to be so dense. IIUC, the cargo/tanker has GSE up in the nose along with propellant tanks. It noses up to the QD plate on the receiving ship, its QD plate extends, and it pumps over the propellant from the cargo bay tanks. Is this right?

The nominal 150t is more a placeholder based on last stated cargo capacity to LEO (100t?) than a firm number. If I've got your conops right, total propellant load would be the nominal 150t minus the mass of the internal tankage, pumps, QD plate and plate extension mechanism - for every cargo/tanker load. A bespoke tanker would have the same size and design as a cargo ship. The only difference would be the top and common dome slid up taking up some of the cargo space. That and longer wiring harnesses and plumbing.


The actual numbers will change as the ships capabilities evolve. Twark or Rad Mod or maybe Robo ran some numbers awhile back claiming that a tanker with a small stretch could load a total of 1600 or 1650t of propellant and deliver quite a bit more than 150t. A dedicated depot would most probably be a tanker variant with extra kit so this plays into supporting early Artemus missions where the extra propellant can be put to good use. Don't ask. It was a very detailed discussion.

As spaceship modifications go, moving the domes is no big thing. The number of rings per barrel would change but the assembly techniques stay the same and in the same order. Less difficulty than a pickup assembly line accommodating the occasional crew cab or dually rear end, short bed/long bed. The extra cost is offset to some degree by not needing a chomper or cargo hatch. The upside is less parasitic mass. The downside is dedicated use. Long term, this strikes me as a good tradeoff. An actual stretch would be a bit more difficult but not outrageously so.

The multi use cargo/tanker looks like a good proof of concept and maybe even an early operational architecture but not so good long term. Which points up an ongoing problem here at NSF. We often do not differentiate between short term and long term vision. Talking about SX makes it worse because they have a bad habit of changing plans in the blink of an eye and that pie in the sky comes on fast.


There is one point in your cargo/tanker idea where safety would be a concern. This may be where my brain fart happened. If a crewed ship needs refueling it should only happen once. Multiple small transfers each carry the same safety risks as one big gulp. This can be got around by waiting for the transfers to be finished before the crew launches and transfers over to the refueled ship. Another complexity, but maybe the best way.


You're correct in expecting the depot to have what you call female and what I call gender bent GSE. My terminology comes from electronics adapters (called gender benders) used to reverse a cables gender. This 'off gender' GSE connection complicates ground handling. Hey, that suckers does have to be fueled and launched. The solution that IMO, solves the problem best is to put an adapter (a gender bender) on the ground side GSE for depot launches only. Put it on. Launch. Take it off. The extra mass stays on the ground and most likely only a couple hours labor at each end. Depot launches will be rare so this wouldn't be much overall system impact.

A depot would have only the one QD plate. It's the free flyer that would have two of the same gender as the ground side GSE.  One side would mount to a tanker with standard QD plate that becomes a temporary accumulator. The other would mate to the tankers (also standard QD plate) arriving to fill the accumulator, and ultimately to the receiving ship (also with standard QD plate). Keep in mind, the free flyer and depot are not complimentary. The are competing concepts with the same goal.


What you call a command module is pretty much what the free flyer would be. It packs in whatever functionality a depot needs but allows plain Jane tankers to become storage tanks (accumulators). If you build the command module into a tanker you have - a depot. The free flyer concept that's been thrown around would be a new design with all that implies.


One free flyer variant gives it enough tankage for substantial dV so it can shift to where the next traffic is expected. Another has it staying on the tanker to hitch a ride to the next orbit. A third packs it back into a cargo bay for return and relaunch. They all have strengths and weaknesses. IMO it's way too early for hard decisions in depot vs free flyer, let alone the variants. Just gonna set back and enjoy the show.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/30/2022 07:40 pm
Personally, I like box-in-a-box. There's a reason shipping containers revolutionised transport. But it's against SpaceX's design philosophy, and dedicated tankers aren't going to be sitting around wasted, so there's no reason for it.
Another reason to go with "box-in-a-box" is that it would let them refill vehicles that used different kinds of fuel. If they draw straight from the tanks of a special tanker starship, then the only thing they can ever refuel is something that uses CH4 and O2. But if the propellants are carried in separate tanks in the cargo hold, then they could fill up (in theory) a vehicle that used H2 and O2, or a nuclear-thermal vehicle that just used H2 or an ion drive that wanted xenon or krypton or whatever. Obviously the depots would be different, but the cargo starships would be the same.

Of course, that assumes there will ever be very many vehicles like that needing to be refueled.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/30/2022 08:22 pm
IIUC, the cargo/tanker has GSE up in the nose along with propellant tanks. It noses up to the QD plate on the receiving ship, its QD plate extends, and it pumps over the propellant from the cargo bay tanks. Is this right?
Maybe I've been messing the terms up. I think QD is the "quick disconnect," which is the female socket that's at the bottom of a Starship. GSE is the "ground support equipment." This is the male plug that goes into the QD when a starship is filled on the ground. A "gender bender" is an adaptor that's male on one side and female on the other, with plumbing crossed over as needed to make it work. (Otherwise it'll be the mirror image of what you want on one side or the other.) Are we in agreement here? (Sorry if I've been mixing them up before.)

Given that, I've been visualizing the tanker starship as having GSE instead of a QD at the bottom. They attach a gender bender on the ground to fill it up, but they pull that off and the GSE needs to retract into the starship before it launches. In space, it extends the GSE and mates with the QD on a standard starship.

Are we still together?

If you want one tanker to refill another, you'll have to somehow attach a gender bender to one of them. This is what a "free flyer" is good for. (Among other things)

For what I was envisioning, the "depot" in the cargo hold would consist of two small tanks plus GSE. You'd fill these tanks the same way you would a tanker starship. In space, the cargo door would open, the GSE would extend, and it would mate with the QD on the target vehicle.

The actual numbers will change as the ships capabilities evolve. Twark or Rad Mod or maybe Robo ran some numbers awhile back claiming that a tanker with a small stretch could load a total of 1600 or 1650t of propellant and deliver quite a bit more than 150t. A dedicated depot would most probably be a tanker variant with extra kit so this plays into supporting early Artemus missions where the extra propellant can be put to good use. Don't ask. It was a very detailed discussion.
I guess the real question is "how much more propellant?" If it's twice as much, then, yeah, I can see how it might be worth it. But if it's just 20% more, maybe not.

Perhaps I should go back through the earlier posts on the thread. I suspect they'd make a lot more sense to me now.

There is one point in your cargo/tanker idea where safety would be a concern. This may be where my brain fart happened. If a crewed ship needs refueling it should only happen once. Multiple small transfers each carry the same safety risks as one big gulp. This can be got around by waiting for the transfers to be finished before the crew launches and transfers over to the refueled ship. Another complexity, but maybe the best way.
Yeah, I thought about this too. That's where a bespoke depot seems to make the most sense. You fill it up--no matter how long it takes--and when it's got enough propellant, then you send up the crewed ship and fill it just once. That's the case where special reflectors and/or cryocoolers make sense as well--given how long the propellant has to survive in LEO--and all that extra hardware really does seem like it should be on a depot--not a starship.

You're correct in expecting the depot to have what you call female and what I call gender bent GSE. My terminology comes from electronics adapters (called gender benders) used to reverse a cables gender.
Yep. I've soldered RS232 cables for this purpose myself, although not since Jimmy Carter was president. :-)

A depot would have only the one QD plate. It's the free flyer that would have two of the same gender as the ground side GSE.  One side would mount to a tanker with standard QD plate that becomes a temporary accumulator. The other would mate to the tankers (also standard QD plate) arriving to fill the accumulator, and ultimately to the receiving ship (also with standard QD plate). Keep in mind, the free flyer and depot are not complimentary. The are competing concepts with the same goal.

What you call a command module is pretty much what the free flyer would be. It packs in whatever functionality a depot needs but allows plain Jane tankers to become storage tanks (accumulators). If you build the command module into a tanker you have - a depot. The free flyer concept that's been thrown around would be a new design with all that implies.
Ah, that makes sense, although the gender combinations are a little challenging. E.g. if you want one tanker to fill another, you need a twisted female-female free flyer. But if you want a tanker to fill a regular starship, you need a straight-through male-female one. And if you know you'll always have a free flyer, then you want to make all starships female and all free flyers twisted male-male.

IMO it's way too early for hard decisions in depot vs free flyer, let alone the variants. Just gonna set back and enjoy the show.
Sure, but it's a lot easier to enjoy the show if you start off knowing what language they're speaking! :-)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 08/31/2022 02:39 am
The basic depot [...] fits in the cargo hold of a standard starship and is firmly braced in there.
[...]
When cargo starship A refills starship B

Just to clarify, the terminology of the thread (and the NASA HLS contract) is:

Tanker: Starship that launches fuel, either to an orbiting depot, or directly to another Starship to be refuelled.

Depot: An accumulator in orbit that is refuelled by a number of tanker-flights until it is full (or full enough). Then-and-only-then the mission vehicle launches and docks with the depot to take on a full (or full enough) propellant load for its mission.

What you've described as a "depot" is what we are calling a "tanker". And your model of operations lacks a depot and requires the mission vehicle to act as the accumulator on orbit. The model of operations described by the HLS contract has an accumulator (depot) in orbit being filled by a series of tankers.

This doesn't change your argument for a modular tanks and plumbing in a cargo-vehicle to replace dedicated tankers. But it isn't the "depot", it's a "tanker".



I just can't help thinking that if you're trying to save money by mass-producing these things, it will help a lot to have fewer versions. Right now I'm seeing a cargo version, a crew version, a tanker version, a lunar version, etc. and that just seems like a problem to me.

Meh, it's pretty standard with aircraft. The hard-part is the basic design. Variations are trivial by comparison.

But you'd likely need a lot of tankers.

That's why it makes sense to specialise.

You only want a box-in-a-box if the particular box rarely flies, because you don't want a vehicle that spends most of its time not earning money, unless it's earning NASA-money. That isn't the case for tankers, because refuelling is the standard-operating-procedure for Starship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/31/2022 04:01 am
Personally, I like box-in-a-box. There's a reason shipping containers revolutionised transport. But it's against SpaceX's design philosophy, and dedicated tankers aren't going to be sitting around wasted, so there's no reason for it.
Another reason to go with "box-in-a-box" is that it would let them refill vehicles that used different kinds of fuel. If they draw straight from the tanks of a special tanker starship, then the only thing they can ever refuel is something that uses CH4 and O2. But if the propellants are carried in separate tanks in the cargo hold, then they could fill up (in theory) a vehicle that used H2 and O2, or a nuclear-thermal vehicle that just used H2 or an ion drive that wanted xenon or krypton or whatever. Obviously the depots would be different, but the cargo starships would be the same.

Of course, that assumes there will ever be very many vehicles like that needing to be refueled.


Quote
Of course, that assumes there will ever be very many vehicles like that needing to be refueled.
LOL. Get in your way back machine to around 1915 and try to explain a megaplex truck stop with 800 parking spots, a movie theater, an 8 bay repair facility, a hotel, a full service restaurant, half a dozen fast food joints and three C stores. You'll be locked up as a lunatic.


Back to spaceships. Right now SX is concerned with refueling their own ship. We're talking 1200+ tons of propellant. This is the cake. Customers would be the icing.


Customers in the foreseeable future will need only a few tons. This is where tanks and a transfer rig in the cargo bay make sense. SS to orbit. Open the pod bay Hal. A self propelled mini depot emerges and lights off to service a customer. Think spy sats.


Refueling works best if the target is designed for it. If someone starts designing for it today it won't show up for three years minimum. In the real world, probably 3-4 years after SX shows it to work.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: AC in NC on 08/31/2022 04:45 am
Of course, that assumes there will ever be very many vehicles like that needing to be refueled.
LOL. Get in your way back machine to around 1915 and try to explain a megaplex truck stop with 800 parking spots, a movie theater, an 8 bay repair facility, a hotel, a full service restaurant, half a dozen fast food joints and three C stores. You'll be locked up as a lunatic.

And of course:

Quote
I think there is a world market for maybe five computers.
-- Thomas Watson, IBM Founder, 1943
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/31/2022 07:15 am
As you can see, the curves never cross. That means there's no upper wavelength cutoff beyond which a totally emissive blackbody would "beat" a totally reflective surface.
Does such a surface exist, though?

Obviously not, but you're missing the point entirely.

First you find the optimum. Then you try to design a material that gets as close as possible to that optimum. But if you don't know the "ideal" first, you don't know what spectral properties you're aiming for.


Additionally, if even that "ideal" material isn't able to achieve your primary goal (for instance, passive ZBO), then you know it's impossible for any real material. You can therefore avoid investing millions of dollars in material science research looking for something that provably can't exist.

"Success being one of the possible outcomes is very important."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/31/2022 07:55 am
I don't think the analysis in the thread has touched on all the available coatings.

That's the beautiful part about it. You don't have to exhaustively search through all available coatings.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/31/2022 08:16 am
If you don't care about all that math and you just want to skip straight to "the answer" (as if there ever was such a thing!  ::))...


If your depot doesn't reenter, Nomex blankets over MLI.


This also contributes some MMOD protection, which is a nice risk buy-down.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 08/31/2022 02:26 pm
(snip)
Btw SPHEREx is precisely the special case I described earlier: the cone always points up, so it never has a view of the Earth. It also needs to be located in a dusk/dawn polar orbit, such that it never has a view of the Sun. There are downsides to using a disk/dawn orbit for a depot: it's less efficient than launching eastward, and it has less flexibility for targeting a specific right ascension for departure.

Jumping in late to this discussion, so apologies if this has been said before.
The dusk-dawn SSO orbit is a limiting case, one that is particularly well suited to all-sky surveys like SPHEREX and WISE (https://www.nasa.gov/mission_pages/WISE/mission/index.html).
(And particularly unsuited to a depot from which to start interplanetary trajectories, as you said.)
There, the angle from the local vertical to the Sun is always near 90 degrees.  (probably >90 minus 23.5 degree inclination of the Earth's rotation with respect to its orbit minus ~7 degrees from the inclination to the pole.)
However, the same goal of never pointing the radiator at the Sun can be achieved at lower inclinations for which the precession does not match the orbital rate of the Earth, i.e. with a variable beta angle.

Take the other extreme, an equatorial orbit.
A depot in equatorial orbit oriented to the Local Velocity, Local Horizon (LVLH) would need an Earth shield and a sunshield.
This could be cylindrical, with axis pointed at the orbit poles, perpendicular to velocity and to the local zenith.
A cryoradiator could point out either end.
However, it's Field Of View would be limited to 90 - 23.5 =~66 degrees from the axis, to prevent the Sun from shining into it regardless of season.
(The beta angle is zero at the equinoxes, when the pane of the equator passes through the Sun, and grows to ~22 deg at the solstices.)
Such radiators are commonly made with "Etendue (https://www.eckop.com/resources/optics/illumination/etendue/) converters", non-imaging optics, sometimes implemented with Winston cones (https://scienceworld.wolfram.com/physics/WinstonCone.html).  (See some one dimensional implementations here (https://www.winstonconeoptics.com/).)

For higher inclination orbits, the angle is reduced, and by conservation of Etendue, the far end of the "cone" gets proportionally larger.
So it's a trade between making the inclination high enough to go through the departure points and restricting the size.

So it's not just the spectral properties that influence the solar and Earth heating and the thermal radiation. 
The geometry is also important.

More to the point: 

- A sun-sync orbit could freeze a passive V-Groove depot (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2401672#msg2401672) solid.

- A low-inclination orbit doesn't just maximize depot payload.  It also melts ice.

Q:  Which liquid would you load in the exposed tank?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/31/2022 05:09 pm
Just to clarify, the terminology of the thread (and the NASA HLS contract) is:

Tanker: Starship that launches fuel, either to an orbiting depot, or directly to another Starship to be refuelled.

Depot: An accumulator in orbit that is refuelled by a number of tanker-flights until it is full (or full enough). Then-and-only-then the mission vehicle launches and docks with the depot to take on a full (or full enough) propellant load for its mission.
So consider a specialized starship with extended tanks (and no cargo hold)--what I've been calling a "tanker." Is that still a tanker by your definition? (I would expect it to be.) If you launch it empty with a plan to fill it in orbit, does it turn into a depot? Even though it's the same vehicle?

Now consider a cargo starship with a couple of tanks of propellant in the cargo bay. Surely that's not also a tanker? It turns into a tanker when you put the tanks in the cargo hold, and it quits being one when you take them out?

Imagine that you launch that same cargo starship with empty tanks in the hold. Does that mean it magically turns into a depot as soon as it's in orbit--assuming the plan were to fill and then empty the tanks?

Surely it makes no sense to reclassify the launch vehicle merely based on the cargo in it!

Anyway, it seems to me like you're missing a term to distinguish bespoke hardware vs. multipurpose hardware. Unless I'm missing some other consideration . . .

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Luc on 08/31/2022 05:24 pm
Rather than a specialized depot with coatings, what about an orbiting tubular sunshade/solar wing that starships “park in” - basically fly in formation with while exchanging/storing fuel? I’m not talking about anything elaborate, just a big, hollow milar structure with basic station keeping from one of their satellites.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rakaydos on 08/31/2022 05:52 pm
If you don't care about all that math and you just want to skip straight to "the answer" (as if there ever was such a thing!  ::))...


If your depot doesn't reenter, Nomex blankets over MLI.


This also contributes some MMOD protection, which is a nice risk buy-down.
Can you paint it shuttle-tank orange? Gotta have some nice clarity in the visuals when explaining it to 80 year old politicians.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/31/2022 06:59 pm
- A sun-sync orbit could freeze a V-Groove depot solid.

- A low-inclination orbit doesn't just maximize depot payload.  It also melts ice.

It's almost like... it's a trade-off.

SSO + V-Groove is simply a bad trade compared to LEO 28° + MLI/Nomex. More complex sunshade design / deployment and a less efficient orbit.



Folks here love it when you and I get into it ( ::) ), so that's the last I'll say in this exchange. You get the last word, buddy!! I'll just pretend like you said I'm wrong by hyperlinking to your own post, and call it a day. ;)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 08/31/2022 11:24 pm
Rather than a specialized depot with coatings, what about an orbiting tubular sunshade/solar wing that starships “park in” - basically fly in formation with while exchanging/storing fuel? I’m not talking about anything elaborate, just a big, hollow milar structure with basic station keeping from one of their satellites.

An inflatable beam tensegrity structure with stretched mylar films would be fairly simple, and there are designs for UV cured inflatable booms so it stays stiff after the deployment gas leaks out. Either self deploy with an inchworm robot arm slipping it over the nose, or something that can be swung around from the freeflyer mating device that's been proposed here. However, the PR optics on what would be a shiny condom for Staship would, at the least, be a dangerous meme resource for our favorite memelord...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 08/31/2022 11:25 pm
IIUC, the cargo/tanker has GSE up in the nose along with propellant tanks. It noses up to the QD plate on the receiving ship, its QD plate extends, and it pumps over the propellant from the cargo bay tanks. Is this right?
Maybe I've been messing the terms up. I think QD is the "quick disconnect," which is the female socket that's at the bottom of a Starship. GSE is the "ground support equipment." This is the male plug that goes into the QD when a starship is filled on the ground. A "gender bender" is an adaptor that's male on one side and female on the other, with plumbing crossed over as needed to make it work. (Otherwise it'll be the mirror image of what you want on one side or the other.) Are we in agreement here? (Sorry if I've been mixing them up before.)

Given that, I've been visualizing the tanker starship as having GSE instead of a QD at the bottom. They attach a gender bender on the ground to fill it up, but they pull that off and the GSE needs to retract into the starship before it launches. In space, it extends the GSE and mates with the QD on a standard starship.

Are we still together?

If you want one tanker to refill another, you'll have to somehow attach a gender bender to one of them. This is what a "free flyer" is good for. (Among other things)

For what I was envisioning, the "depot" in the cargo hold would consist of two small tanks plus GSE. You'd fill these tanks the same way you would a tanker starship. In space, the cargo door would open, the GSE would extend, and it would mate with the QD on the target vehicle.

The actual numbers will change as the ships capabilities evolve. Twark or Rad Mod or maybe Robo ran some numbers awhile back claiming that a tanker with a small stretch could load a total of 1600 or 1650t of propellant and deliver quite a bit more than 150t. A dedicated depot would most probably be a tanker variant with extra kit so this plays into supporting early Artemus missions where the extra propellant can be put to good use. Don't ask. It was a very detailed discussion.
I guess the real question is "how much more propellant?" If it's twice as much, then, yeah, I can see how it might be worth it. But if it's just 20% more, maybe not.

Perhaps I should go back through the earlier posts on the thread. I suspect they'd make a lot more sense to me now.

There is one point in your cargo/tanker idea where safety would be a concern. This may be where my brain fart happened. If a crewed ship needs refueling it should only happen once. Multiple small transfers each carry the same safety risks as one big gulp. This can be got around by waiting for the transfers to be finished before the crew launches and transfers over to the refueled ship. Another complexity, but maybe the best way.
Yeah, I thought about this too. That's where a bespoke depot seems to make the most sense. You fill it up--no matter how long it takes--and when it's got enough propellant, then you send up the crewed ship and fill it just once. That's the case where special reflectors and/or cryocoolers make sense as well--given how long the propellant has to survive in LEO--and all that extra hardware really does seem like it should be on a depot--not a starship.

You're correct in expecting the depot to have what you call female and what I call gender bent GSE. My terminology comes from electronics adapters (called gender benders) used to reverse a cables gender.
Yep. I've soldered RS232 cables for this purpose myself, although not since Jimmy Carter was president. :-)

A depot would have only the one QD plate. It's the free flyer that would have two of the same gender as the ground side GSE.  One side would mount to a tanker with standard QD plate that becomes a temporary accumulator. The other would mate to the tankers (also standard QD plate) arriving to fill the accumulator, and ultimately to the receiving ship (also with standard QD plate). Keep in mind, the free flyer and depot are not complimentary. The are competing concepts with the same goal.

What you call a command module is pretty much what the free flyer would be. It packs in whatever functionality a depot needs but allows plain Jane tankers to become storage tanks (accumulators). If you build the command module into a tanker you have - a depot. The free flyer concept that's been thrown around would be a new design with all that implies.
Ah, that makes sense, although the gender combinations are a little challenging. E.g. if you want one tanker to fill another, you need a twisted female-female free flyer. But if you want a tanker to fill a regular starship, you need a straight-through male-female one. And if you know you'll always have a free flyer, then you want to make all starships female and all free flyers twisted male-male.

IMO it's way too early for hard decisions in depot vs free flyer, let alone the variants. Just gonna set back and enjoy the show.
Sure, but it's a lot easier to enjoy the show if you start off knowing what language they're speaking! :-)
Let's not sweat terminology unless it hinders communications rather than help it. As long as we end up on the same page, the specific label is unimportant. We're working on it.

I've been a bit loose on QD plate and GSE. The QD plate that is part of the ship might arguably be considered GSE. It depends where you want to split that hair. Part of the launch tower GSE is a QD plate. Following your convention it would be male. The one on the ship, female.

In the spirit of keeping hardware as common as possible, all ships would have a female QD plate so that all ships can properly mate to the male GSE QD plate on the tower. The problem is that a depot needs a QD plate that will mate to the tower GSE so it can be filled and launched. It must be female. Then, to mate to any other ship, it's QD plate must be male, because other ships QD plates are female. From this, arise all discussion of gender benders and related.

Ideally, the depot would be the only ship with major adaptions specific to refueling. Something has to be done and whatever it is, it logically should be done on the tanker depot. Or better yet, on the ground equipment. That's where the idea of a gender bender on the towers GSE QD plate comes in.

The ground GSE is male. Ships are ordinarily female. The depot needs to be male to do its job on orbit.  So an adapter (gender bender), female on both faces, is mounted between the ground side QD plate and the depot side QD plate. It stays on the ground.

I envision the depot QD plate to be in a bit of a dog house sticking out of the ship. This would allow some stand-off between the ships. If necessary, it could be designed to extend. This would preclude using the QD plate connection as a firm connection between ships so if they need to do a tight lashup, the lashup would need be elsewhere.

The most reasonable alternative IMO, is the free flyer. Simple version is a cube with male QD plates on either side. It needs pumps, plumbing, smarts, thrusters to maneuver and fuel. If small enough it might launch in the truncated cargo bay of the first tanker to act as an accumulator. Ity bitty bay door opens and it exits and works it's way down the hull and one of its male QD plates does unmentionable intimate things to the normal female QD plate on the tanker accumulator. And they lived happily ever-after.

If too large, it has to find another ride. The heavy version would have PV, radiator, cryocooler, and/or some MLI. Might have substantial tankage to allow orbit changes. It would need another ride.

These are IMO, the direction SX will be looking. In usual fashion they'll start simple, maybe something like the rig and tankage in the cargo bay. But these two designs or something like them, will be what they'll be looking at as the final product.

Refueling is central to what SX wants to do, not to mention that it is (I think) a formal milestone in the Artemus contract. This implies they will start bending metal as early as possible. Maybe two or three launches following the first reasonably successful (sub)orbital flight will see the first attempt.

This will still be early enough in the overall starship development process that even if they can recover a ship successfully, the changes will be coming too fast to bother except as a test of reflight. Well, maybe once they get a StarLink system unkinked they'll reuse them out of practicality as soon as possible.

StarLink launch, as it stands now, looks like it will be a candidate for a dedicated variant. If a big tank tanker adds only 20% propellant it gives six tankers worth for the launch of five tankers. Actually, better than that because there is zero cargo handling & fueling other than normal propellant loading. This is another variant candidate.

If starship were isogrid construction and Boeing were doing the work, variants would be out of the question without five years lead time and a ton of money. Not sure if that's a metric ton or not. This is SX. This whole shebang started with a water tower construction company doing contract work to build Hoppy in the open, within spitting distance of the beach. The whole point is to optimize operations, not keep the fabrication costs as low as possible. Fabrication costs are important but it's only a means to an end and they seem to have it well in hand.

There may be some unlabeled opinions above.  :D


Edit, changed word.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/01/2022 12:04 am
Well, maybe once they get a StarLink system unkinked they'll reuse them out of practicality as soon as possible.
Of course! I see it now! The free flyer will be a modified StarLink satellite! (Inspired by the folks above who assured me that it's always easier to modify something you've already built.) :-)

Seriously, I have to say the Free Flyer is certainly the cleanest option, if it can be made to work. Then you don't make any changes to your base starships, and you can always put up specialized depots later, if you want to. Everything except a free flyer would have female QD plates--even the depots.

Then the question is: what is the simplest possible free-flyer design?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/01/2022 01:45 am
SSO + V-Groove is simply a bad trade compared to LEO 28° + MLI/Nomex.

No, the sun-sync orbit was never a requirement of V-Groove passive radiators, just your misreading.  Posters seem unfamiliar with V-Groove, despite long heritage.  Mass-efficiency is hard to beat.

And it's methane (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2402494#msg2402494) in the exposed tank, because...?

Rather than a specialized depot with coatings, what about an orbiting tubular sunshade/solar wing that starships “park in” - basically fly in formation with while exchanging/storing fuel? I’m not talking about anything elaborate, just a big, hollow milar structure with basic station keeping from one of their satellites.

Well, a single-layer sunshade or parallel-layering sunshade wouldn't give cryogenic temperature in LEO; you'd need more complex structural units, like V-Groove.  Example:  Compare the V-Groove middle-shield temperatures to inner-shield temperatures in Bhandari et al. 2020 (https://ttu-ir.tdl.org/bitstream/handle/2346/86248/ICES-2020-24.pdf?sequence=1) figures.  By inference, single-shield temperature would be even higher.

...the Free Flyer is certainly the cleanest option, if it can be made to work. Then you don't make any changes to your base starships, and you can always put up specialized depots later, if you want to.

If you built for Martian fleet support, you could arrange a "free-flying structure" also for passive liquid water storage, docks, cargo warehouses, stevedore services, and other protected fleet facilities, giving a more valuable, multi-purpose depot. 

E.g., the VLEO "Terrestation" (https://forum.nasaspaceflight.com/index.php?topic=45299.msg1818436#msg1818436) concept.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/01/2022 04:26 am
Just to clarify, the terminology of the thread (and the NASA HLS contract) is:
Tanker: Starship that launches fuel, either to an orbiting depot, or directly to another Starship to be refuelled.
Depot: An accumulator in orbit that is refuelled by a number of tanker-flights until it is full (or full enough). Then-and-only-then the mission vehicle launches and docks with the depot to take on a full (or full enough) propellant load for its mission.
So consider a specialized starship with extended tanks (and no cargo hold)--what I've been calling a "tanker." Is that still a tanker by your definition? (I would expect it to be.) If you launch it empty with a plan to fill it in orbit, does it turn into a depot? Even though it's the same vehicle?
Now consider a cargo starship with a couple of tanks of propellant in the cargo bay. Surely that's not also a tanker? It turns into a tanker when you put the tanks in the cargo hold, and it quits being one when you take them out?
Imagine that you launch that same cargo starship with empty tanks in the hold. Does that mean it magically turns into a depot as soon as it's in orbit--assuming the plan were to fill and then empty the tanks?
Surely it makes no sense to reclassify the launch vehicle merely based on the cargo in it!
Anyway, it seems to me like you're missing a term to distinguish bespoke hardware vs. multipurpose hardware. Unless I'm missing some other consideration . . .

Dude, I wasn't trying to get into some debate on Plutonic [Platonic] ideals. I was just letting you know that you are using terminology that has been used differently previous in the thread (and in the HLS contract); and judging by some replies, it's likely to cause confusion.

[edit: Words are hard]

For example, your concept only replaces the tankers, it doesn't replace the depot. You are instead using the mission HLS as the accumulator. That just transfers the cost of thermal management/boil-off losses to the HLS, it doesn't solve it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/01/2022 01:43 pm
Dude, I wasn't trying to get into some debate on Plutonic ideals.
Only if we're planning to refill the Starship with magma. :-)

I was just letting you know that you are using terminology that has been used differently previous in the thread (and in the HLS contract); and judging by some replies, it's likely to cause confusion.
Do you have a link to the part of the HLS contract you're talking about?

I did see a post (way back) trying to impose the definition you're talking about, and I think I saw one or two people try to use it, but I'm not convinced that it's a) widely used or b) particularly useful.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/01/2022 02:14 pm
Well, maybe once they get a StarLink system unkinked they'll reuse them out of practicality as soon as possible.
Of course! I see it now! The free flyer will be a modified StarLink satellite! (Inspired by the folks above who assured me that it's always easier to modify something you've already built.) :-)

Seriously, I have to say the Free Flyer is certainly the cleanest option, if it can be made to work. Then you don't make any changes to your base starships, and you can always put up specialized depots later, if you want to. Everything except a free flyer would have female QD plates--even the depots.

Then the question is: what is the simplest possible free-flyer design?
Wellll, it depends on the needs. That's what the discussion about heat rejection and solar white paint is all about. If they can cycle a tanker to the accumulator every 15 minutes (yeah, right) then get the receiving ship in there right after the final tanker, the basic cube free flyer would be fine. If it's two weeks between tankers it's a whole different problem. How long does it take to transfer a tanker load? How often can they launch a tanker? Lots of moving parts. One SpaceX sized step at a time.


You were joking about modifying a StarLink for the free flyer. There is some truth there. The first Ford Mustang was a Falcon with a few inches added to the frame, a bigger engine and sexier bling. Any overlap in needs between StarLink and free flyer can be carried over in hardware and software. Just don't expect gross physical structure to carry over.


My gut says they'll shoot for either a free flyer or a full tilt depot but not cross over from one to the other unless they hit a wall or decide to add capabilities beyond propellant transfer. Which is better? Too many unknown at this time. Thermal and its mitigation is at the top of the list.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: docmordrid on 09/01/2022 02:59 pm
As regards Starlink customization, SpaceX is already doing it.

https://twitter.com/Sandra_I_Erwin/status/1313225103392546816

Sandra Erwin @Sandra_I_Erwin
Breaking news: SpaceX, L3Harris win Space Development Agency contracts to build missile-warning satellites. SpaceX is developing a new satellite for DoD based on the Starlink design. SpaceNews
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/01/2022 04:36 pm
Some insight on research on atmospheric thermal behavior that seems pertinent.

https://www.nasaspaceflight.com/2022/08/jpss-2-testing-complete/ (https://www.nasaspaceflight.com/2022/08/jpss-2-testing-complete/)

Quote
Lastly, the Clouds and Earth’s Radiant Energy System (CERES) instrument will measure the energy being absorbed and emitted by Earth’s atmosphere. CERES is currently flying on Suomi NPP and NOAA-20 and previously helped scientists understand the links between absorption and energy emissions that affect energy balances.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 09/01/2022 05:11 pm
A 'free flyer' is a needless complexity. Since:
- an adapter will only ever do anything when right next to (and in contact with) a Starship
- is only ever needed when a Starship is transferring propellants to another Starship
- that situation (prop transfer) only ever needs to occur between a tanker and a depot, and a depot and a 'target' Starship
- any adapter (free flyer or not) needs to be launchable by a Starship or smaller vehicle
- therefor any adapter by definition can be launched by a Starship
You can eliminate all redundant free-flight hardware (power, propulsion, avionics, control, position sensing, etc) by attaching the adapter to a Starship, ideally the depot. That way the adapter is launched by the Depot Starship rather than by an entirely separate launch of its own, so the propellant 'cost' for the combined depot launch is lower overall than with a separate launch. If the 'free flight' adapter is intended to launch with the depot anyway, then it is a direct mass and complexity (and cost and development time) saving to eliminate the free flight capability and leave just the adapter attached.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/01/2022 06:02 pm
A 'free flyer' is a needless complexity. Since:
- an adapter will only ever do anything when right next to (and in contact with) a Starship
- is only ever needed when a Starship is transferring propellants to another Starship
- that situation (prop transfer) only ever needs to occur between a tanker and a depot, and a depot and a 'target' Starship
- any adapter (free flyer or not) needs to be launchable by a Starship or smaller vehicle
- therefor any adapter by definition can be launched by a Starship
You can eliminate all redundant free-flight hardware (power, propulsion, avionics, control, position sensing, etc) by attaching the adapter to a Starship, ideally the depot. That way the adapter is launched by the Depot Starship rather than by an entirely separate launch of its own, so the propellant 'cost' for the combined depot launch is lower overall than with a separate launch. If the 'free flight' adapter is intended to launch with the depot anyway, then it is a direct mass and complexity (and cost and development time) saving to eliminate the free flight capability and leave just the adapter attached.
IIRC, you expect that the two ships need little gap between them and the QD connection is the only connection needed. No other supports necessary. If I have this right, and your conjectures are correct, this can be a very simple system. While I very much respect the technical skill you bring to the table, I have to disagree. I'd be gobsmacked if it ends up that easy.


Gobsmacked, but not unhappy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/01/2022 11:25 pm
A 'free flyer' is a needless complexity. Since:
- an adapter will only ever do anything when right next to (and in contact with) a Starship
- is only ever needed when a Starship is transferring propellants to another Starship
- that situation (prop transfer) only ever needs to occur between a tanker and a depot, and a depot and a 'target' Starship
- any adapter (free flyer or not) needs to be launchable by a Starship or smaller vehicle
- therefor any adapter by definition can be launched by a Starship
You can eliminate all redundant free-flight hardware (power, propulsion, avionics, control, position sensing, etc) by attaching the adapter to a Starship, ideally the depot. That way the adapter is launched by the Depot Starship rather than by an entirely separate launch of its own, so the propellant 'cost' for the combined depot launch is lower overall than with a separate launch. If the 'free flight' adapter is intended to launch with the depot anyway, then it is a direct mass and complexity (and cost and development time) saving to eliminate the free flight capability and leave just the adapter attached.
IIRC, you expect that the two ships need little gap between them and the QD connection is the only connection needed. No other supports necessary. If I have this right, and your conjectures are correct, this can be a very simple system. While I very much respect the technical skill you bring to the table, I have to disagree. I'd be gobsmacked if it ends up that easy.


Gobsmacked, but not unhappy.

We're sorta re-litigating some stuff, but this is ultimately about torsional load.  I doubt that a single set of bellows hoses will support the torques that an almost-full depot can exert against a completely full lift tanker (which should be worst case), but you can build some kind of reinforcing truss around the plumbing that takes up the loads.

As that truss gets wider, the individual loads on any of its connecting pins obviously go down.  So the question is whether they go down enough that a single subunit (i.e., plumbing, docking guides, and connecting pins) can support the worst-case torsion.  If not, then you need two subsystems, separated by as much distance as you can manage, and things are complicated.

However, in either case, I don't see a free-flyer helping out much, unless it's two connecting systems mounted on either end of a long truss.  And... that's just a kludge.

(Note:  If it's truly flying free, then it has propulsion.  Even if you want some fairly robust structure, you can deploy an awful lot of stuff from a payload bay.  Some kind of robot that walks end-over-end would probably do fine, and it's still vastly simpler than a true free flyer.  But even that is overkill.)

A quick summary:  We've talked about four different geometries (cheesy diagrams attached):

1) The old-timey tail-to-tail docking (apparently abandoned).

2) Dorsal-to-dorsal, nose-to-nose, aligned docking (what the most recent aspirational artwork showed).  No connectors shown, and this would require something that deployed. (Free flyer, walker, extender, something in a chine--who knows?  It's not my fave.)

3) Dorsal-to-dorsal nose-to-tail, overlapped docking.  This could be androgynous, but it can also have full plumbing coming out of one payload bay and a stub that just stabilizes things in the other.  (I think I've given up on this one.)

4) Dorsal-to-dorsal, noses in the same direction, but overlapped docking, with the payload bay of one close to the tail QD of the other.  In this case, the single grapple and the plumbing are all that's taking the torsional loads.

OTV Booster seems to be on Team #2.  I think edzieba and I are mostly on Team #4 (although we disagree on whether the torsional loads are a problem).  I used to like #3, but I think #4 does everything it does and is simpler, moments of inertia permitting.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/01/2022 11:33 pm
RRM3 (https://nexis.gsfc.nasa.gov/RRM3.html) (Robotic Refueling Mission-3) Flex Hose Robotic Demonstration on ISS

https://www.youtube.com/watch?v=cSV2H_secbQ
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/02/2022 02:03 am
A 'free flyer' is a needless complexity. Since:
- an adapter will only ever do anything when right next to (and in contact with) a Starship
- is only ever needed when a Starship is transferring propellants to another Starship
- that situation (prop transfer) only ever needs to occur between a tanker and a depot, and a depot and a 'target' Starship
- any adapter (free flyer or not) needs to be launchable by a Starship or smaller vehicle
- therefor any adapter by definition can be launched by a Starship
You can eliminate all redundant free-flight hardware (power, propulsion, avionics, control, position sensing, etc) by attaching the adapter to a Starship, ideally the depot. That way the adapter is launched by the Depot Starship rather than by an entirely separate launch of its own, so the propellant 'cost' for the combined depot launch is lower overall than with a separate launch. If the 'free flight' adapter is intended to launch with the depot anyway, then it is a direct mass and complexity (and cost and development time) saving to eliminate the free flight capability and leave just the adapter attached.
IIRC, you expect that the two ships need little gap between them and the QD connection is the only connection needed. No other supports necessary. If I have this right, and your conjectures are correct, this can be a very simple system. While I very much respect the technical skill you bring to the table, I have to disagree. I'd be gobsmacked if it ends up that easy.


Gobsmacked, but not unhappy.

We're sorta re-litigating some stuff, but this is ultimately about torsional load.  I doubt that a single set of bellows hoses will support the torques that an almost-full depot can exert against a completely full lift tanker (which should be worst case), but you can build some kind of reinforcing truss around the plumbing that takes up the loads.

As that truss gets wider, the individual loads on any of its connecting pins obviously go down.  So the question is whether they go down enough that a single subunit (i.e., plumbing, docking guides, and connecting pins) can support the worst-case torsion.  If not, then you need two subsystems, separated by as much distance as you can manage, and things are complicated.

However, in either case, I don't see a free-flyer helping out much, unless it's two connecting systems mounted on either end of a long truss.  And... that's just a kludge.

(Note:  If it's truly flying free, then it has propulsion.  Even if you want some fairly robust structure, you can deploy an awful lot of stuff from a payload bay.  Some kind of robot that walks end-over-end would probably do fine, and it's still vastly simpler than a true free flyer.  But even that is overkill.)

A quick summary:  We've talked about four different geometries (cheesy diagrams attached):

1) The old-timey tail-to-tail docking (apparently abandoned).

2) Dorsal-to-dorsal, nose-to-nose, aligned docking (what the most recent aspirational artwork showed).  No connectors shown, and this would require something that deployed. (Free flyer, walker, extender, something in a chine--who knows?  It's not my fave.)

3) Dorsal-to-dorsal nose-to-tail, overlapped docking.  This could be androgynous, but it can also have full plumbing coming out of one payload bay and a stub that just stabilizes things in the other.  (I think I've given up on this one.)

4) Dorsal-to-dorsal, noses in the same direction, but overlapped docking, with the payload bay of one close to the tail QD of the other.  In this case, the single grapple and the plumbing are all that's taking the torsional loads.

OTV Booster seems to be on Team #2.  I think edzieba and I are mostly on Team #4 (although we disagree on whether the torsional loads are a problem).  I used to like #3, but I think #4 does everything it does and is simpler, moments of inertia permitting.
Yeah we've been down this road before but I now see something that confuses me more than usual.

Quote
I doubt that a single set of bellows hoses will support the torques that an almost-full depot can exert against a completely full lift tanker...
I don't think anybody is suggesting bellows hoses would be structural. They just carry fluid. Am I misunderstanding?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/02/2022 03:18 am
I doubt that a single set of bellows hoses will support the torques that an almost-full depot can exert against a completely full lift tanker...
I don't think anybody is suggesting bellows hoses would be structural. They just carry fluid. Am I misunderstanding?

Seems to me that there are only three possibilities:

1) The hoses are so flexible that formation-flying works and you just have to figure out how to get the hoses connected.  This has been discussed--I'm skeptical that the station-keeping is good enough to make this reliable.

2) The hoses are rigid enough to bear structural loads that hold the two ships together.  I, like you, think this is unlikely.

3) Some other structure bears the loads and the hoses don't have to worry about bearing any loads other than pressure.  This then decomposes into two sub-cases:
a) The load-bearing members are very close to the hoses.
b) They're farther away from the hoses.

If you had, say, just the QD plate and nothing else, that would be sub-case a), and I'm skeptical that it work work.  If there are structural members within, say, a meter of the hoses, would that work?  I don't know.

On the other hand, if you have one structural member latched onto the QD and one latched several meters away, I'm pretty confident the pair would bear the loads pretty easily.

In my little cartoons above, it's pretty clear the case #4, for all of its nice features, is the worst case in term of torsion.  (Well, nose-to-nose would be worse, but it's also dumb.)  But you can probably construct some set of trusses that will bear loads a meter or so from the hoses.  Is that good enough?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 09/02/2022 04:46 am
Dumb question, but does use of flexible hoses add the additional problem of both hose pressure stiffening the hose and whipping around both vehicles, and such movement also buckling the hose, altering flowrate?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/02/2022 05:12 am
I doubt that a single set of bellows hoses will support the torques that an almost-full depot can exert against a completely full lift tanker...
I don't think anybody is suggesting bellows hoses would be structural. They just carry fluid. Am I misunderstanding?

Seems to me that there are only three possibilities:

1) The hoses are so flexible that formation-flying works and you just have to figure out how to get the hoses connected.  This has been discussed--I'm skeptical that the station-keeping is good enough to make this reliable.

2) The hoses are rigid enough to bear structural loads that hold the two ships together.  I, like you, think this is unlikely.

3) Some other structure bears the loads and the hoses don't have to worry about bearing any loads other than pressure.  This then decomposes into two sub-cases:
a) The load-bearing members are very close to the hoses.
b) They're farther away from the hoses.

If you had, say, just the QD plate and nothing else, that would be sub-case a), and I'm skeptical that it work work.  If there are structural members within, say, a meter of the hoses, would that work?  I don't know.

On the other hand, if you have one structural member latched onto the QD and one latched several meters away, I'm pretty confident the pair would bear the loads pretty easily.

In my little cartoons above, it's pretty clear the case #4, for all of its nice features, is the worst case in term of torsion.  (Well, nose-to-nose would be worse, but it's also dumb.)  But you can probably construct some set of trusses that will bear loads a meter or so from the hoses.  Is that good enough?
Well, the only thing I can think of along these lines is the depot QD mounted on a pantograph or possibly some jack screws so it can extend a bit to the tanker QD. I can't see this by itself as structurally sufficient for a safe lashup. I'm addressing two ships dorsal to dorsal facing in the same direction.


A close alternative would be the depot QD mounted on the face of a dog house blistering out a bit from he hull. Might need a retractable cover to protect the connections during ascent and EDL if it's intended to return.


The QD plates have locking lugs to pull them firmly together. The fluid connections themselves might each need a little float to allow self centering. The thermal environment will be all over the place and the dimensions will change.


Speaking of thermal, the whole ship(s) will most likely change dimensions throughout orbit. This will be a challenge for any struts holding the two ships in place. Maybe some intentional slop where they connect. I wonder how much an empty starship would warp with full sun on one side for 50 minutes. Propellant would modify this but it gives an upper limit.


However the two ships get hitched, I don't think they will spontaneously start wild gyrations. More likely little things from transferring propellants or small misalignment from ullage thrust. If something weird does start, a stuck thruster for example, the software will need a fast acting abort mode.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/02/2022 06:21 am
However the two ships get hitched, I don't think they will spontaneously start wild gyrations. More likely little things from transferring propellants or small misalignment from ullage thrust. If something weird does start, a stuck thruster for example, the software will need a fast acting abort mode.

The fast acting abort only works if there's a solution where the ships don't collide post-disconnect.  It's not immediately apparent to me that such a solution always exists--especially if one of the ships is having a thruster malfunction.

The way to avoid problems is with a rigid connection.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/02/2022 06:36 am
A question:  Where is the LOX downcomer for the nose header tank routed?  Is there an external chine with a line that connects it to the LOX main tank?  Or is it routed through the methane tank?

All of the top of methane tank-to-QD configurations (i.e., #3 and #4 in my diagrams) rely on a LOX line being available to grab in the payload bay.  Without that line, the QD-to-QD configuration (#2) starts to make more sense.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: daavery on 09/02/2022 06:44 am
the LOX downcomer runs along the center of the ventral(windward) side.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: john smith 19 on 09/02/2022 06:52 am
I'm very late to this as I was offline for so long.

But wasn't at least some of these problems sorted out when FH was talking about cross-filling during flight?

Once SX decided on specialst "core" stages for FH  that would mean abandoning the idea of gender-neutral connectors

Likewise from Armadillio Aerospace's experience with differential tank pressures we know it can take as little as 2psi to push flow between different propellant tanks of the same type (which played havoc with ending feed).

And of course orbital loading will have a lot more time to play with than the few minutes of an FH flight.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 09/02/2022 10:08 am
However the two ships get hitched, I don't think they will spontaneously start wild gyrations. More likely little things from transferring propellants or small misalignment from ullage thrust. If something weird does start, a stuck thruster for example, the software will need a fast acting abort mode.

The fast acting abort only works if there's a solution where the ships don't collide post-disconnect.  It's not immediately apparent to me that such a solution always exists--especially if one of the ships is having a thruster malfunction.

The way to avoid problems is with a rigid connection.

The "Nauka's gone wild" problem. Do you hang on for dear life and fight it, or run.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 09/02/2022 12:57 pm
The bellows are an complete nonissue: the 'problem' of how to have flexible cryogenic couplings on a rigid but moveable linkage without those bellow being load-bearing is not only a solved problem, but there is a solution currently sat on top of the arm halfway up the orbital launch tower, with said solution having already demonstrated the capability to move around, connect to a ship, and transfer cryogenic propellants.
3) Dorsal-to-dorsal nose-to-tail, overlapped docking.  This could be androgynous, but it can also have full plumbing coming out of one payload bay and a stub that just stabilizes things in the other.  (I think I've given up on this one.)

4) Dorsal-to-dorsal, noses in the same direction, but overlapped docking, with the payload bay of one close to the tail QD of the other.  In this case, the single grapple and the plumbing are all that's taking the torsional loads.

OTV Booster seems to be on Team #2.  I think edzieba and I are mostly on Team #4 (although we disagree on whether the torsional loads are a problem).  I used to like #3, but I think #4 does everything it does and is simpler, moments of inertia permitting.
I'm in camp 3, because it means all your drain-propellant-in-orbit-for-transfer plumbing is at the apex of the tanks and out of the way during normal operations, rather than at the nadir where all the drain-propellant-for-propulsion plumbing also needs to be. It also minimises torque about the connection, and minimises the movement of the centre of mass of the system (think of the setup as a heavy glob of propellants that some rigid lightweight shells happen to move around the outside of). It also allows for a 'bumper' to be deployed at the 'unused' (or inactive) nose-to-tail meeting point to aid bracing, if needed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/02/2022 02:54 pm
Dumb question, but does use of flexible hoses add the additional problem of both hose pressure stiffening the hose and whipping around both vehicles, and such movement also buckling the hose, altering flowrate?

However the two ships get hitched, I don't think they will spontaneously start wild gyrations. More likely little things from transferring propellants or small misalignment from ullage thrust. If something weird does start, a stuck thruster for example, the software will need a fast acting abort mode.

The fast acting abort only works if there's a solution where the ships don't collide post-disconnect.  It's not immediately apparent to me that such a solution always exists--especially if one of the ships is having a thruster malfunction.

The way to avoid problems is with a rigid connection.

The "Nauka's gone wild" problem. Do you hang on for dear life and fight it, or run.

Remember, single-tanker refill would likely be a brief experimental phase.  It wouldn't make sense to dedicate a fleet of Starship tankers to depot duty, long-term.  At fleet scale, inflatable passive cryogenic tanks would be far more mass-efficient and cost-effective. 

If you deployed inflatables on a rigid truss, that truss could also give rigid mounting points for docks, with robust cradling connectors for QD plates and hulls.  A station-keeping engine is needed, but little more.  Such a structure could retire tanker-dock concerns such as crimped hoses and gyration.

https://www.youtube.com/watch?v=vcbVD7AwEC4
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/02/2022 02:56 pm
However the two ships get hitched, I don't think they will spontaneously start wild gyrations. More likely little things from transferring propellants or small misalignment from ullage thrust. If something weird does start, a stuck thruster for example, the software will need a fast acting abort mode.

The fast acting abort only works if there's a solution where the ships don't collide post-disconnect.  It's not immediately apparent to me that such a solution always exists--especially if one of the ships is having a thruster malfunction.

The way to avoid problems is with a rigid connection.
'Tween the devil and the deep blue sea. A tough call. If things go sideways and a tanker and depot are at risk it's one thing. If a crewed ship is involved it's very different. How to deal with it would call for a detailed analysis of potential failure modes and a lot of modeling. Decisions on possible control inputs, yada, yada. If there comes a point where there's nothing to loose, ya gotta go for it. Rigid connection or not.


I agreed with a rigid connection until thermal thoughts crept in. Ever notice that everything on the ISS is linear attachment? I often thought that running two or more rows of modules with cross connect access would give more rigidity. It would be a great help in giving the truss a broader footprint, during boost burns, in reducing moment when changing attitude and even in reducing MMOD armor,  but it's not done. My conjecture is differential heating would cause unmanageable stress.


Assume one standoff up towards the nose about 30m off. One solution might be a pivot at either end of the stand-off that would allow some swing fore and aft. If the ship expands 5mm (example only) in length between the QD and the standoff, and the separation is a nominal 1m, the change in angle between ships would too small to worry about, except for the control software. The latched QD's should be the unmoving origin of the coordinate system. Plug in other numbers if desired.


It's not exactly a rigid connection but maybe good enough?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/02/2022 03:23 pm
Dumb question, but does use of flexible hoses add the additional problem of both hose pressure stiffening the hose and whipping around both vehicles, and such movement also buckling the hose, altering flowrate?

However the two ships get hitched, I don't think they will spontaneously start wild gyrations. More likely little things from transferring propellants or small misalignment from ullage thrust. If something weird does start, a stuck thruster for example, the software will need a fast acting abort mode.

The fast acting abort only works if there's a solution where the ships don't collide post-disconnect.  It's not immediately apparent to me that such a solution always exists--especially if one of the ships is having a thruster malfunction.

The way to avoid problems is with a rigid connection.

The "Nauka's gone wild" problem. Do you hang on for dear life and fight it, or run.

Remember, single-tanker refill would likely be a brief experimental phase.  It wouldn't make sense to dedicate a fleet of Starship tankers to depot duty, long-term.  At fleet scale, inflatable passive cryogenic tanks would be far more mass-efficient and cost-effective. 

If you deployed inflatables on a rigid truss, that truss could also give rigid mounting points for docks, with robust cradling connectors for QD plates and hulls.  A station-keeping engine is needed, but little more.  Such a structure could retire tanker-dock concerns such as crimped hoses and gyration.

https://www.youtube.com/watch?v=vcbVD7AwEC4 (https://www.youtube.com/watch?v=vcbVD7AwEC4)
I'm unclear. Do you propose the inflatables be delivered as cargo with propellant already loaded or empty to be filled by tankers?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/02/2022 04:49 pm
Thinking it through, neither way makes sense in the short to medium term. Long term - maybe.


The truss with QD would have the same function as the free flyer tricked out with all the goodies. And it would have a lot more structure to hold the inflatables. The free flyer OTOH, would only need enough structure to hold the goodies and two QD's.


The free flyer is intended to mate up with an unmodified tanker carrying the first load of propellant, which would take on the role of accumulator. After other tankers offload into it and it transfers its load to the receiving ship, it's free to return earthwards and continue life as a plain ol tanker. The free flyer might stay on orbit for another refueling campaign or return. Too many hypothetical moving parts to know for sure.


The tanker/accumulator serves as both tankage and structure, removing the need for the inflatables and the structure to hold them. The tanker/accumulator would be out of tanker service for only one campaign and it would start off the campaign delivering the first load of propellant. If the whole system is honed and oiled to where a tanker can be launched every 12 hours AND a returning tanker can be turned around fast enough to participate, one tanker staying on orbit for the entire campaign makes little difference.


The only advantage obvious in a truss with inflatables is the ability to hold more than one tanker/accumulator worth of propellant. Even this advantage is questionable because the ship design is expected (not a done deal) to evolve. Once a tanker design has been superseded it becomes a candidate for one last flight and hulk status as tankage.


The truss itself is not a bad idea. The core of a free flyer is a box with two QD's. It might or might not have tankage beyond that needed to deploy onto the accumulator. It would be sandwiched between the accumulator and another ship and it would have some combination of PV/radiator/cryocooler and MLI. It needs some structure for all of this and a (probably) telescoping truss seems a natural fit. Again, a lot of hypothetical moving parts.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/02/2022 06:24 pm
the LOX downcomer runs along the center of the ventral(windward) side.

Where does it go when it gets to the LCH4 dome?  Through it?  It can't get routed to the outside of the vehicle if it's on the ventral side, because it would interfere with the TPS.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 09/02/2022 06:29 pm
the LOX downcomer runs along the center of the ventral(windward) side.

Where does it go when it gets to the LCH4 dome?  Through it?  It can't get routed to the outside of the vehicle if it's on the ventral side, because it would interfere with the TPS.
Through it, inside. Flyover images of previous ships in the process of being scrapped show where it penetrates the forward and common domes, in the middle of the ventral side.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/02/2022 07:25 pm
The tanker/accumulator serves as both tankage and structure, removing the need for the inflatables and the structure to hold them.. 

The only advantage obvious in a truss with inflatables is the ability to hold more than one tanker/accumulator worth of propellant. Even this advantage is questionable because the ship design is expected (not a done deal) to evolve. Once a tanker design has been superseded it becomes a candidate for one last flight and hulk status as tankage.

No, the $ advantage of inflatables is a depot needing very few tankers. 

If a tanker had 12-hour turnaround, and it delivered 1/5 of a load, it could fill an inflatable depot for ~ 300 crews each synod, by itself. 

Otherwise, a fleet of tankers would be needed.  E.g., given a 14-day window and preloading of depot tankers at synod start, at least 46 tankers would be needed. 

Scale up ~ 10x (https://aeon.co/essays/elon-musk-puts-his-case-for-a-multi-planet-civilisation) for settlement cargo flights.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/02/2022 08:15 pm
The bellows are an complete nonissue: the 'problem' of how to have flexible cryogenic couplings on a rigid but moveable linkage without those bellow being load-bearing is not only a solved problem, but there is a solution currently sat on top of the arm halfway up the orbital launch tower, with said solution having already demonstrated the capability to move around, connect to a ship, and transfer cryogenic propellants.

The arm can handle a small amount of translation, from wind and thermal expansion.  But the amount of torsion on orbit in an emergency is much, much larger.  There's no way that the arm hardware could withstand that.
 
Quote
3) Dorsal-to-dorsal nose-to-tail, overlapped docking.  This could be androgynous, but it can also have full plumbing coming out of one payload bay and a stub that just stabilizes things in the other.  (I think I've given up on this one.)

4) Dorsal-to-dorsal, noses in the same direction, but overlapped docking, with the payload bay of one close to the tail QD of the other.  In this case, the single grapple and the plumbing are all that's taking the torsional loads.

OTV Booster seems to be on Team #2.  I think edzieba and I are mostly on Team #4 (although we disagree on whether the torsional loads are a problem).  I used to like #3, but I think #4 does everything it does and is simpler, moments of inertia permitting.
I'm in camp 3, because it means all your drain-propellant-in-orbit-for-transfer plumbing is at the apex of the tanks and out of the way during normal operations, rather than at the nadir where all the drain-propellant-for-propulsion plumbing also needs to be. It also minimises torque about the connection, and minimises the movement of the centre of mass of the system (think of the setup as a heavy glob of propellants that some rigid lightweight shells happen to move around the outside of). It also allows for a 'bumper' to be deployed at the 'unused' (or inactive) nose-to-tail meeting point to aid bracing, if needed.

When I went and actually drew this out, I got horribly confused, but Option #2 starts to look a lot better.  Take a look and see what you think.  Option #3 is really complicated, although it does indeed separate the transfer sources from the engine sumps, and it's the most mechanically stable.  Note that the pros and cons haven't been thought through completely, but are a nice basis for discussion (or the hurling of rotten fruit).

 ¯\_(ツ)_/¯
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/02/2022 08:15 pm
The tanker/accumulator serves as both tankage and structure, removing the need for the inflatables and the structure to hold them.. 

The only advantage obvious in a truss with inflatables is the ability to hold more than one tanker/accumulator worth of propellant. Even this advantage is questionable because the ship design is expected (not a done deal) to evolve. Once a tanker design has been superseded it becomes a candidate for one last flight and hulk status as tankage.

No, the $ advantage of inflatables is a depot needing very few tankers. 

If a tanker had 12-hour turnaround, and it delivered 1/5 of a load, it could fill an inflatable depot for ~ 300 ships each synod, by itself. 

Otherwise, a fleet of tankers would be needed.  E.g., given a 14-day window and preloading of depot tankers at synod start, at least 46 tankers would be needed. 

Scale up ~ 10x (https://aeon.co/essays/elon-musk-puts-his-case-for-a-multi-planet-civilisation) for settlement cargo flights.
What size inflatables are we talking about?


What you're talking about would come under what I call long term. Definitely more than 5 years. Probably under 10 but not at the scale you envision. Still, hulk tankers would be a lot less R&D, and the hardware would already be in hand. Has anybody ever played with cryo temp inflatables?


On the number of tankers, we might have a terminology problem here. What I call a tanker is the ship that has no other job than lifting propellant onto orbit, transferring it to an accumulator or a depot, then going home. An accumulator would structurally be a tanker but with a more specialized role. Somewhat the same for depots.


Yes accumulators and depots would be stacking up  and at the scale described, a lot of them. It's an interesting question. How many hulk candidates would be available? Maybe not enough.


Something to consider. There are low energy transits to mars and they are not terribly sensitive to a specific launch window. The downside is that they take longer. This isn't a problem with cargo. I think a year long cargo launch campaign would work, but IANARS. Each launch would be more direct and higher energy than the last.


As an ideal, everything would land before the crew mission lights off but probably not realistic. Anyway, spreading the launches does away with the need for massive amounts of propellant on orbit at one time.


When multiple settler ships are launching the equation changes and this idea starts having traction. We might be talking 12-15 years. By that time we might have space hardened bacteria that can link pseudopods and move in a line to poop out a bio printed tank. Arthur C. Clark! Where are you when I need a story written?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 09/02/2022 08:28 pm
Why did SpaceX abandon end to end refueling?  To me that would be the simplest and no flexible connections needed. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/02/2022 09:23 pm
No, the $ advantage of inflatables is a depot needing very few tankers. 

If a tanker had 12-hour turnaround, and it delivered 1/5 of a load, it could fill an inflatable depot for ~ 300 ships each synod, by itself. 

Otherwise, a fleet of tankers would be needed.  E.g., given a 14-day window and preloading of depot tankers at synod start, at least 46 tankers would be needed. 

Scale up ~ 10x (https://aeon.co/essays/elon-musk-puts-his-case-for-a-multi-planet-civilisation) for settlement cargo flights.

What you're talking about would come under what I call long term. Definitely more than 5 years. Probably under 10 but not at the scale you envision...  Has anybody ever played with cryo temp inflatables?

Crew-2029 is "definitely more than 5 years"...

Example:  A basic inflatable depot design:  "Antaios", patented in 2017 (https://patents.google.com/patent/US9676499B2/en).

https://www.youtube.com/watch?v=-7zTfwkjezM
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/02/2022 09:51 pm
The bellows are an complete nonissue: the 'problem' of how to have flexible cryogenic couplings on a rigid but moveable linkage without those bellow being load-bearing is not only a solved problem, but there is a solution currently sat on top of the arm halfway up the orbital launch tower, with said solution having already demonstrated the capability to move around, connect to a ship, and transfer cryogenic propellants.

The arm can handle a small amount of translation, from wind and thermal expansion.  But the amount of torsion on orbit in an emergency is much, much larger.  There's no way that the arm hardware could withstand that.
 
Quote
3) Dorsal-to-dorsal nose-to-tail, overlapped docking.  This could be androgynous, but it can also have full plumbing coming out of one payload bay and a stub that just stabilizes things in the other.  (I think I've given up on this one.)

4) Dorsal-to-dorsal, noses in the same direction, but overlapped docking, with the payload bay of one close to the tail QD of the other.  In this case, the single grapple and the plumbing are all that's taking the torsional loads.

OTV Booster seems to be on Team #2.  I think edzieba and I are mostly on Team #4 (although we disagree on whether the torsional loads are a problem).  I used to like #3, but I think #4 does everything it does and is simpler, moments of inertia permitting.
I'm in camp 3, because it means all your drain-propellant-in-orbit-for-transfer plumbing is at the apex of the tanks and out of the way during normal operations, rather than at the nadir where all the drain-propellant-for-propulsion plumbing also needs to be. It also minimises torque about the connection, and minimises the movement of the centre of mass of the system (think of the setup as a heavy glob of propellants that some rigid lightweight shells happen to move around the outside of). It also allows for a 'bumper' to be deployed at the 'unused' (or inactive) nose-to-tail meeting point to aid bracing, if needed.

When I went and actually drew this out, I got horribly confused, but Option #2 starts to look a lot better.  Take a look and see what you think.  Option #3 is really complicated, although it does indeed separate the transfer sources from the engine sumps, and it's the most mechanically stable.  Note that the pros and cons haven't been thought through completely, but are a nice basis for discussion (or the hurling of rotten fruit).

 ¯\_(ツ)_/¯
Nice rendering. I'm curious. Why is commingling the feed and the engine sumps a bad thing? This is how it loads in the ground.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: john smith 19 on 09/03/2022 07:26 am

When I went and actually drew this out, I got horribly confused, but Option #2 starts to look a lot better.  Take a look and see what you think.  Option #3 is really complicated, although it does indeed separate the transfer sources from the engine sumps, and it's the most mechanically stable.  Note that the pros and cons haven't been thought through completely, but are a nice basis for discussion (or the hurling of rotten fruit).

 ¯\_(ツ)_/¯
The options are complex.  A key question would be how much specific work needs to be done to get from a standard Starship to a tanker (or depot) variant?

SX likes to avoid special versions of stuff as long as possible (F9 and cargo Dragon were flying looong before FH and crew Dragon started flying). Also once SS starts making orbit they will want it to start being productive. That means starlink launches.

This suggests they would prefer options that either a)Can be installed (or removed) as a package or b) Can be permanently installed on all SS because the payload hit is minor or zero.

Both make which specific SS is doing what role an operational decision, rather than a build choice, but b) means that decision can be deferred to (more or less) the day of launch, giving greater flexibility.

How does that change the optics?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/03/2022 08:29 pm

When I went and actually drew this out, I got horribly confused, but Option #2 starts to look a lot better.  Take a look and see what you think.  Option #3 is really complicated, although it does indeed separate the transfer sources from the engine sumps, and it's the most mechanically stable.  Note that the pros and cons haven't been thought through completely, but are a nice basis for discussion (or the hurling of rotten fruit).

 ¯\_(ツ)_/¯
The options are complex.  A key question would be how much specific work needs to be done to get from a standard Starship to a tanker (or depot) variant?

SX likes to avoid special versions of stuff as long as possible (F9 and cargo Dragon were flying looong before FH and crew Dragon started flying). Also once SS starts making orbit they will want it to start being productive. That means starlink launches.

This suggests they would prefer options that either a)Can be installed (or removed) as a package or b) Can be permanently installed on all SS because the payload hit is minor or zero.

Sounds right to me.  It also makes me lean more toward Option #2, where the existing plumbing is used, and there's an adapter to allow two female QDs to talk to a male-male adapter.

I think you could make something like the attached deploy out of the payload bay.  At an arm-wave, you have payload bay doors that open and slide out a stowed version of everything you need.  Then the stowed version needs to unfold itself as it extends back along the dorsal surface.  All mechanical engineering that I attempt always turns out to be a kludge, but I can draw a block diagram.

If you do this right, this is a complete "depot kit" that can turn any lift tanker into a depot or, more accurately, a tanker that's enabled to manage the transfer of prop in either direction.  The sole modification you need on vanilla Starships or lunar Starships to use this is a pair of passive posts for the grapples to grab.  (Note that "active" and "passive" refer to Starships with and without depot kits.  They don't denote the direction of fueling.  Depots have to be able to both receive and send prop.)

Ideally, a deployed depot kit could re-stow itself, allowing the tanker to which it was attached to return to EDL.  If that could happen, then there's no need for depots in cislunar orbits, which are likely more varied than the LEO orbit(s) needed for refueling.  A tanker with a stowed depot kit can refuel at the LEO depot, boost to whatever cislunar (or HEEO) orbit is required, deploy its depot kit, refuel whatever LSS or vanilla Starship needs it, stow the kit, and return directly to EDL, where it can be reused as an ordinary tanker.

If the kit can't re-stow, then you have permanent depots in cislunar that can't return.  In this case, you have a passive tanker (no depot kit) that refuels at the LEO depot, boosts to the cislunar orbit (or HEEO) in which a second depot lives, refuels it, and then returns to EDL.  Not quite as good or as flexible, but almost.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/04/2022 03:20 pm

The options are complex.  A key question would be how much specific work needs to be done to get from a standard Starship to a tanker (or depot) variant?

SX likes to avoid special versions of stuff as long as possible (F9 and cargo Dragon were flying looong before FH and crew Dragon started flying). Also once SS starts making orbit they will want it to start being productive. That means starlink launches.

This suggests they would prefer options that either a)Can be installed (or removed) as a package or b) Can be permanently installed on all SS because the payload hit is minor or zero.

Sounds right to me.  It also makes me lean more toward Option #2, where the existing plumbing is used, and there's an adapter to allow two female QDs to talk to a male-male adapter.

I think you could make something like the attached deploy out of the payload bay.  At an arm-wave, you have payload bay doors that open and slide out a stowed version of everything you need.  Then the stowed version needs to unfold itself as it extends back along the dorsal surface.  All mechanical engineering that I attempt always turns out to be a kludge, but I can draw a block diagram.

If you do this right, this is a complete "depot kit" that can turn any lift tanker into a depot or, more accurately, a tanker that's enabled to manage the transfer of prop in either direction.  The sole modification you need on vanilla Starships or lunar Starships to use this is a pair of passive posts for the grapples to grab.  (Note that "active" and "passive" refer to Starships with and without depot kits.  They don't denote the direction of fueling.  Depots have to be able to both receive and send prop.)

Ideally, a deployed depot kit could re-stow itself, allowing the tanker to which it was attached to return to EDL.  If that could happen, then there's no need for depots in cislunar orbits, which are likely more varied than the LEO orbit(s) needed for refueling.  A tanker with a stowed depot kit can refuel at the LEO depot, boost to whatever cislunar (or HEEO) orbit is required, deploy its depot kit, refuel whatever LSS or vanilla Starship needs it, stow the kit, and return directly to EDL, where it can be reused as an ordinary tanker.

If the kit can't re-stow, then you have permanent depots in cislunar that can't return.  In this case, you have a passive tanker (no depot kit) that refuels at the LEO depot, boosts to the cislunar orbit (or HEEO) in which a second depot lives, refuels it, and then returns to EDL.  Not quite as good or as flexible, but almost.
Yes! Using the existing plumbing in the way it's designed and oriented to work simplifies a lot.


The adapter you described is essentially the free flyer without free flying. A semi-rhetorical question: is it easier/lighter/cheaper to design/build a mechanism to move the adapter down the hull or to put a small cold gas RCS+translation system on the adapter? It's hard to judge with so many final details TBD, but the question will always be there.


On one hand a manipulating system might have welcome capabilities for other purposes. OTOH, an adapter that can detach and hang around for another campaign, or even move to another orbit, might offer good utility.


Stuffing all the floppy bits back into the cargo bay for return sounds iffy. For that matter, doing stretch tanks and shrinking the cargo bay looks too small for the adapter and everything else, although that really does need tight examination. Your earlier arguments for stretch tanks was on point.


If this rig needs another ride it's a strong argument for a multiple campaign lifetime, which implies free flying capabilities. Besides, free flyer sounds cool. Calling the rig an adapter misses a lot of functionality and is boring. Maybe something like Propellant Transfer and Service Device? Um. PTSD's already taken.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/04/2022 03:31 pm
A "gender bender" is an adaptor that's male on one side and female on the other, with plumbing crossed over as needed to make it work. (Otherwise it'll be the mirror image of what you want on one side or the other.) Are we in agreement here?

Nope. It'a trivial to modify the GSE connector to be mirror-symmetrical, with male and female connectors swapped across the plane of symmetry.

This trick works either way you dock (69 or db). It makes no difference: you don't need a "gender bender" component either way.

Obviously we don't see this on Starship today, but equally obvious is that Starship is currently a prototype and not the final version. If we're gonna baseline future changes (eg new parts), let's at least baseline smart changes.

[diagram has a part labeled "gender bender"]

Fortunately there's no need for this.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/04/2022 07:31 pm
[diagram has a part labeled "gender bender"]

Fortunately there's no need for this.

But there's probably a need for something, if only to create a ship-to-ship separation that's manageable.  Also, whatever you wind up using still has to function as a genuine QD.  I imagine that an androgynous topology (which I agree is easy) is a little bit dicier to engineer for a clean, reliable separation at liftoff.  Not impossible, but if you need something anyway, why fool with best practice?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wes_wilson on 09/04/2022 08:04 pm
[diagram has a part labeled "gender bender"]

Fortunately there's no need for this.

But there's probably a need for something, if only to create a ship-to-ship separation that's manageable.  Also, whatever you wind up using still has to function as a genuine QD.  I imagine that an androgynous topology (which I agree is easy) is a little bit dicier to engineer for a clean, reliable separation at liftoff.  Not impossible, but if you need something anyway, why fool with best practice?

This has probably been shot down somewhere in the prior 28 pages, but I couldn't find it skimming through.

If going the depot approach, would the Super Heavy booster not be a good fit to serve as the depot tanker?  Other threads have indicated it's ssto capable.  It and starship already have ports used for fueling starship while they're both on the OLM.  It and starship already have some clamping mechanisms that hold them together through stage separation. 

Put a sh in orbit, use it as the depot/accumulator.  Filled, it would hold nearly 3 starships worth of fuel/oxidizer. 

Lastly, although it has no heatshield, it has enough dv to reach orbit.  There, it can be refueled by starships and then it would have the same dv as when sitting on the ground.  Which is enough dv to be brought back without a heatshield for occasional and infrequent on-ground servicing.

Has there been an exploration of using SH as the depot?  Seems a closer starting point than using Starship.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/04/2022 08:20 pm
[diagram has a part labeled "gender bender"]

Fortunately there's no need for this.

But there's probably a need for something, if only to create a ship-to-ship separation that's manageable.  Also, whatever you wind up using still has to function as a genuine QD.  I imagine that an androgynous topology (which I agree is easy) is a little bit dicier to engineer for a clean, reliable separation at liftoff.  Not impossible, but if you need something anyway, why fool with best practice?
We are back full circle. Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated. Image from Toledospring.com.


Unless there's a cryo variant already available it would be another materials R&D project. Not impossible but not trivial. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/04/2022 08:25 pm
If going the depot approach, would the Super Heavy booster not be a good fit to serve as the tanker?

It's going to be a while before they need to accumulate more prop than will fit in one Starship.  Hence, a Starship will make a perfectly acceptable depot.

When they finally do get to the point where they need more prop than a single Starship will hold, a SuperHeavy would be a pretty terrible waste of perfectly good engines.  Even if they need four or five Starship-sized depots for some massive mission, that's still fewer engines than a single SuperHeavy.

Last but hardly least, turning a SuperHeavy into a depot is a lot of work.  Getting Starship ready to be refueled is probably about 75% of the work you'd need to get a Starship-based depot up and running, but roughly 0% of the work you'd need to get an SH-based depot going.

PS:  Just noticed that you said SH as the tanker, not the depot.  That won't work.  If you get an SH into orbit, it's not coming down.  The reason it can do RTLS is because it's not going that fast at separation.  At reentry from orbital speed, it'll burn up.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: wes_wilson on 09/04/2022 08:34 pm
If going the depot approach, would the Super Heavy booster not be a good fit to serve as the tanker?

It's going to be a while before they need to accumulate more prop than will fit in one Starship.  Hence, a Starship will make a perfectly acceptable depot.

When they finally do get to the point where they need more prop than a single Starship will hold, a SuperHeavy would be a pretty terrible waste of perfectly good engines.  Even if they need four or five Starship-sized depots for some massive mission, that's still fewer engines than a single SuperHeavy.

Last but hardly least, turning a SuperHeavy into a depot is a lot of work.  Getting Starship ready to be refueled is probably about 75% of the work you'd need to get a Starship-based depot up and running, but roughly 0% of the work you'd need to get an SH-based depot going.

PS:  Just noticed that you said SH as the tanker, not the depot.  That won't work.  If you get an SH into orbit, it's not coming down.  The reason it can do RTLS is because it's not going that fast at separation.  At reentry from orbital speed, it'll burn up.

Thanks, I meant depot, not tanker. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/04/2022 08:37 pm
Im pretty sure they’re using a stretched Starship as a depot. It’s possible that the depot will use the launch tanks, but it’s also possible they’ll use separate tanks which will have better thermal insulation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/04/2022 09:16 pm
[diagram has a part labeled "gender bender"]

Fortunately there's no need for this.

But there's probably a need for something, if only to create a ship-to-ship separation that's manageable.  Also, whatever you wind up using still has to function as a genuine QD.  I imagine that an androgynous topology (which I agree is easy) is a little bit dicier to engineer for a clean, reliable separation at liftoff.  Not impossible, but if you need something anyway, why fool with best practice?

This has probably been shot down somewhere in the prior 28 pages, but I couldn't find it skimming through.

If going the depot approach, would the Super Heavy booster not be a good fit to serve as the depot tanker?  Other threads have indicated it's ssto capable.  It and starship already have ports used for fueling starship while they're both on the OLM.  It and starship already have some clamping mechanisms that hold them together through stage separation. 

Put a sh in orbit, use it as the depot/accumulator.  Filled, it would hold nearly 3 starships worth of fuel/oxidizer. 

Lastly, although it has no heatshield, it has enough dv to reach orbit.  There, it can be refueled by starships and then it would have the same dv as when sitting on the ground.  Which is enough dv to be brought back without a heatshield for occasional and infrequent on-ground servicing.

Has there been an exploration of using SH as the depot?  Seems a closer starting point than using Starship.
The big upside would be the volume. All the difficulties involved with bringing it back and extending the software for a use never intended is the start of the downsides. It's probably technically possible. Maybe worth playing with for mid to long term planning but not for immediate use.


As the design evolves and EDL shows success, non state of the art ships will become available for tanker and depot/accumulator R&D. The first known need for tankers and depot/accumulators is Artemus. They'll only need 'good enough' for that but I think the system will evolve fast from then on. By the time they start mars ops the SH might make sense. Probably use an early mark and let it burn up on reentry.


When they get some history under their belt they'll have a clearer sense of direction.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Okie_Steve on 09/04/2022 09:34 pm
At what point does fineness become a problem with all the proposed stretching of 9m booster/ship/depot/flying tanks/etc?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/04/2022 10:16 pm
I imagine that an androgynous topology... is a little bit dicier to engineer for a clean, reliable separation at liftoff.

Why?



Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated.

You're confusing androgynous connectors with an androgynous interface plate. You can make an androgynous interface plate (when mirrored across some plane) while still using entirely male-female connectors.

You simply put a female connector in the "mirror image" location of every male connector, and vice-versa.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/04/2022 10:26 pm
Eta Space (https://etaspace.com) "Cryo-Dock" design, successor to 2024 LOXSAT-1.  (Sound off.)

Quote
"Eta Space announces Cryo-Dock™, a full scale commercial propellant depot capable of refueling multiple spacecraft with cryogenic propellants. Cryo-Dock™ design is already underway and is aimed to launch after the LOXSAT-1 mission funded by NASA. Eta Space is focused on cryogenic fluid management (CFM) technologies for the New Space Age and Clean Energy Economy."

https://www.youtube.com/watch?v=Xcm7Q797jPI&t=85s
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/04/2022 11:14 pm
From previous posts, I gather the following are key design constraints:

1) Minimize new plumbing.

2) Don't create anything you don't have to. Modifying existing hardware is better than all-new hardware. In particular, shrinking/stretching tanks is relatively easy.

3) There has to be an ullage burn. Anything else (e.g. rotation) would involve way too much entirely new hardware.

I'll add a couple more that seem obvious to me:

4) There can't be any big protrusions from the sides or front of the vehicle. (It has to be aerodynamic during launch.) Things might get extruded later, but, at launch, the rocket needs to be pretty smooth.

5) It has to be completely automatic; no human beings can be involved in the process except on the ground.

Not being an expert, it's hard to weigh these against each other, but applying them to the excellent drawings The Radical Moderate provided raises at least a few questions. I'd love to hear responses from people who really are experts.
(https://forum.nasaspaceflight.com/assets/50157.0/2124398.jpg)
This option, #2, seems far and away the most sensible except for the question of how the gender bender gets put in place. It can't be there at launch time or else it'd violate the no-protrusion rule, and if it's extruded after launch, I think that amounts to making a big change to the vehicle. (But maybe not--like I said; I'm no expert.) If it has to be moved into place, it either needs to fly there (which seems a lot of new hardware!) or else someone needs to put it there (which violates the all-automatic rule).
(https://forum.nasaspaceflight.com/assets/50157.0/2124400.jpg)
I don't see how this option (#3) can work at all. Not only does it have lots of new plumbing, I don't see how it can do the ullage burn without spinning like a pinwheel or (maybe) having the acceleration at a crazy angle. (Maybe I just lack imagination.)
(https://forum.nasaspaceflight.com/assets/50157.0/2124402.jpg)
This one (option #4) also seems to have way too much new plumbing. This is why I found myself (several posts back) thinking that it might work better to create a new device consisting of just propellant tanks plus a GSE plug and stick that in the cargo bay of a regular cargo starship. (The "source ship" in the picture.") Then you wouldn't have any new plumbing, and you could think of the depot as a really stripped down starship. (One with no engines and much smaller tanks.) So it makes no changes to the base starship at all, the only new construction is a modification of what already exists, and nothing protrudes until the cargo bay doors open. It won't carry as much propellant as a tanker-only starship, but maybe that's not a deal breaker.

You've still got the problem of holding the two vehicles together during fueling. I don't have any ideas for how to make that work, short of hoping that synchronized flying isn't infeasible, given the very low accelerations involved.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/04/2022 11:39 pm
We are back full circle. Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated.

I don't think that's a problem, because you're not going face-to-face.  Just use whatever male parts are on the tower-side QD for all the connections on the other side of the center line.  But you'll still need to deal with all the aerodynamics and deployment stuff, to say nothing of the actual disconnect dynamics.

Quote
The adapter you described is essentially the free flyer without free flying. A semi-rhetorical question: is it easier/lighter/cheaper to design/build a mechanism to move the adapter down the hull or to put a small cold gas RCS+translation system on the adapter? It's hard to judge with so many final details TBD, but the question will always be there.

If you're using cold gas, then you'll only be flying freely once, to get deployed.  If that's what it takes... meh.  Sure.  But not my first choice.

I think you can do this with extendable masts, if I measured correctly.  See the attached, highly arm-wavy cartoons.

PS:  I realized that I've conflated the terms "boom" and "mast", and also that "mast" implies something that sticks up.  All these booms/masts are are just extendable trusses to push the two hunks of the depot kit to the right spots.

There's also no reason why there have to be two masts.  You could put everything on the back and probably be fine, except it's a long way to extend something without tacking anything down as you go.  Also, putting the solar array where I did has the nice property of not interfering with the elonerons.  You might want to rotate them 90º if you adopt a nose-to-sun orientation to keep the tanks as cool as possible.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/04/2022 11:53 pm
(https://forum.nasaspaceflight.com/assets/50157.0/2124398.jpg)
This option, #2, seems far and away the most sensible except for the question of how the gender bender gets put in place. It can't be there at launch time or else it'd violate the no-protrusion rule, and if it's extruded after launch, I think that amounts to making a big change to the vehicle. (But maybe not--like I said; I'm no expert.) If it has to be moved into place, it either needs to fly there (which seems a lot of new hardware!) or else someone needs to put it there (which violates the all-automatic rule).

This, but

   • Delete the gender-bender and implement androgynous mirror-symmetrical interface plate

    • Tilt the interface plates at a slight angle, so the Starships have clearance between them

    • For docking the Starships both rotate and translate into position. This is done so the (off-axis) docking kick will exactly null out the rotation+translation. You "just" time-reverse the kinematics to find the exact docking approach maneuver.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/05/2022 03:33 am
This option, #2, seems far and away the most sensible except for the question of how the gender bender gets put in place. It can't be there at launch time or else it'd violate the no-protrusion rule, and if it's extruded after launch, I think that amounts to making a big change to the vehicle.

A couple of passive grappling posts is no big deal.  You can put them in shrouds and have them swing up and lock after you reach orbit.  I suspect that these are table stakes for all of the options.

Quote
I don't see how this option (#3) can work at all. Not only does it have lots of new plumbing, I don't see how it can do the ullage burn without spinning like a pinwheel or (maybe) having the acceleration at a crazy angle. (Maybe I just lack imagination.)

If you have both ships burning cooperatively, this isn't a huge deal.  But yeah, it has a lot of new plumbing.

Quote
You've still got the problem of holding the two vehicles together during fueling. I don't have any ideas for how to make that work, short of hoping that synchronized flying isn't infeasible, given the very low accelerations involved.

Grapple fixtures.  They're well-understood tech.  See here (https://en.wikipedia.org/wiki/Grapple_fixture).  But this does require good enough proximity ops that a simple active arm can reach out and grab them.  But you need two, spaced far enough apart that they can handle worst-case torsion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/05/2022 03:49 am
This, but

   • Delete the gender-bender and implement androgynous mirror-symmetrical interface plate

    • Tilt the interface plates at a slight angle, so the Starships have clearance between them

    • For docking the Starships both rotate and translate into position. This is done so the (off-axis) docking kick will exactly null out the rotation+translation. You "just" time-reverse the kinematics to find the exact docking approach maneuver.

I assume that when you say "tilt," you're talking about a rotation in the x-y plane, not the x-z plane?  In the x-z plane, the tails will collide.

If you only have one grappling point, or even two spaced too closely together, the torsion is going to be a problem.  Keeping the x-axes aligned lets you grab on in two widely spaced spots.

I don't worry too much about whether you need an adapter or whether your can make direct androgynous contact.  Unless a purely passively cooled depot works, you're going to need some cryocooling.  And unless I'm wrong about the torsion being a problem, you're going to need some active grappling at least a few meters away.  Once that stuff's there, you need to deploy something.  Making "something" include an adapter is no big deal, and provides a nice way for the cryocooling to get ahold of boiloff gases.

I imagine that an androgynous topology... is a little bit dicier to engineer for a clean, reliable separation at liftoff.

Why?

It's certainly doable.  I just don't think you need to sink the engineering time into something with so many moving parts to get it sufficiently reliable--especially given my objections just above.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/05/2022 08:14 am
[...]

I note that your boom-deployment is very similar to how the lunar elevator works. The module would need motors to "crawl" down the guide-rails instead of being lowered by a winch, due to the lack of gravity, but the actual way it deploys from the cargo-area is very similar. And it seems like HLS will launch with a permanent guide-rail attached to the body, so there's no reason why the depot couldn't.

(Not sure if the guide-rails can tolerate re-entry, however.)

If the adaptor can be pushed out of the depot cargo-area, another alternative to a free-flyer would be for the first visiting tanker to dock with it at the depot-nose, then transfer it to the depot's QD plate for the actual prop. xfer; but leave it attached to the depot when undocking. Free-flight without a free-flyer.




If going the depot approach, would the Super Heavy booster not be a good fit to serve as the [depot]?
It's going to be a while before they need to accumulate more prop than will fit in one Starship.  Hence, a Starship will make a perfectly acceptable depot.

There's debate over the best architecture for the reusable HLS. Whether it stays in lunar orbit and is refuelled by a tanker ferrying propellant from LEO and back, but then you also need a cargo-ship/logistics-module to ferry mission equipment to HLS; or whether the HLS comes back to LEO to be refuelled/reequipped, in which case it still needs a tanker to NRHO to refuel it for the return-to-LEO.

Either way, the implication is that two ships have to travel from LEO to lunar orbit (and back) prior to every crew mission. Either HLS and a tanker, or the tanker and a cargo-ship.

A "Nearly three times" Super-depot has enough capacity to refuel both ships, with extra for a large boil-off reserve.

Of course, a simpler solution might be just using two or more regular depots, instead of one Super-depot.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/05/2022 08:30 am
At what point does fineness become a problem with all the proposed stretching of 9m booster/ship/depot/flying tanks/etc?
You don't need to change the outer mould line.  You just move the LCH4 dome and inter-tank bulkhead forward to consume some of the cylindrical space in the payload bay.  It gets you to about 1500t of prop, which is enough to make the margins more comfy for the lunar Starship LEO-NRHO-LS-NRHO leg.

The LSS probably also has to do this same trick, which will make the unpressurized cargo hold of the LSS considerably smaller--probably not much more than 2-3m high.  But that's enough room for airlocks, the elevator hardware, and a modest amount of cargo.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/05/2022 08:40 am
I note that your boom-deployment is very similar to how the lunar elevator works. The module would need motors to "crawl" down the guide-rails instead of being lowered by a winch, due to the lack of gravity, but the actual way it deploys from the cargo-area is very similar. And it seems like HLS will launch with a permanent guide-rail attached to the body, so there's no reason why the depot couldn't.

The rail would help quite a bit.  I have no idea how long a lightweight boom/mast/truss can get with a mass on the end before it stops being adequately rigid.  The rail would eliminate having to worry about that.

Quote
(Not sure if the guide-rails can tolerate re-entry, however.)

Worst case, you jettison the whole shebang before EDL.  That's obviously suboptimal, but the cost of the depot kit is considerably lower than the cost of the Starship itself.

Quote
If the adaptor can be pushed out of the depot cargo-area, another alternative to a free-flyer would be for the first visiting tanker to dock with it at the depot-nose, then transfer it to the depot's QD plate for the actual prop. xfer; but leave it attached to the depot when undocking. Free-flight without a free-flyer.

Probably acceptable, but it eliminates the ability for depots to back up their kit and return to EDL with it.  As I said a bit up-thread, that's not the end of the world, but when refueling in high orbits is required, putting the depot into a customized orbit for the mission, then returning to be put into the next customized orbit, would be handy.

For NRHO, I'd think you'd put a permanent depot in both LEO and NRHO, with a plain ol' tanker shuttling between the two, transferring prop.

Quote
There's debate over the best architecture for the reusable HLS. Whether it stays in lunar orbit and is refueled by a tanker ferrying propellant from LEO and back, but then you also need a cargo-ship/logistics-module to ferry mission equipment to HLS; or whether the HLS comes back to LEO to be refueled/reequipped, in which case it still needs a tanker to NRHO to refuel it for the return-to-LEO.

...Of course, a simpler solution might be just using two or more regular depots, instead of one Super-depot.

Yes, even an LSS doing LEO-LS-NRHO-LEOpropulsive still needs a refueling in NRHO.  But that prop has to be there before you'll get a green light to launch the crew, so you can stage both the tanker filling the cislunar depot and the LSS itself from the same LEO depot (with more tankers filling up in between, obviously).

I'm sure if this stuff gets to high cadence, bigger depots--or gangs of depots--will make sense.  But I doubt it's in the Option B timeframe.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/05/2022 11:19 am
This, but

   • Delete the gender-bender and implement androgynous mirror-symmetrical interface plate

    • Tilt the interface plates at a slight angle, so the Starships have clearance between them

    • For docking the Starships both rotate and translate into position. This is done so the (off-axis) docking kick will exactly null out the rotation+translation. You "just" time-reverse the kinematics to find the exact docking approach maneuver.

I assume that when you say "tilt," you're talking about a rotation in the x-y plane, not the x-z plane?  In the x-z plane, the tails will collide.

So don't do that. :P

Move it down an inch.

You can also have it stick out a bit. The No Protrusions "rule" is fictional anyway.

If you only have one grappling point, or even two spaced too closely together, the torsion is going to be a problem.  Keeping the x-axes aligned lets you grab on in two widely spaced spots.

This is one of those FUD fan theories that sounds superficially convincing until SpaceX blows right past it.

I don't worry too much about whether you need an adapter or whether your can make direct androgynous contact.

You should. Elon Musk would.

I imagine that an androgynous topology... is a little bit dicier to engineer for a clean, reliable separation at liftoff.

Why?

It's certainly doable.  I just don't think you need to sink the engineering time into something with so many moving parts to get it sufficiently reliable

It's the same number of moving parts, just rearranged.

Your "adapter" is adding moving parts, however.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/05/2022 04:19 pm
It'a trivial to modify the GSE connector to be mirror-symmetrical, with male and female connectors swapped across the plane of symmetry.
Okay, I think I finally see what you're trying to say here. Tell me if this is what you've got in mind. To simplify, imagine we're designing a connector to fill/drain a single tank. Visualize a rectangle with a socket on the left and a spigot on the right. A hose runs from the socket to the bottom of the tank and another hose runs from the spigot to the top of the tank. The idea is that you'd pump liquid into the socket and extract air (ullage gas) from the spigot.

Now imagine that you have two of these contraptions. They'll just plug into each other. You don't need a version that's a mirror image, and you don't even need to flip one of them over. You've got a single type of connector with no adaptors needed.

However, you've now connected the top of one tank with the bottom of the other one, but you need them to be connected bottom-to-bottom and top-to-top (unless you somehow made option #3 work). How do you switch the pipes around? It seems like you still do need an adaptor that's essentially two pipes that cross over and have these universal connectors on each end.

There's also the question of where the pumps are. I don't think a normal starship has pumps either to suck fuel in or push it out (except into the engines). This problem isn't specific to your universal connector design, of course, but the adaptor (given that you have to have one) seems like a nice place to put the pumps. Otherwise they're extra weight that starships don't need once they're fully fueled.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/05/2022 04:30 pm
This option, #2, seems far and away the most sensible except for the question of how the gender bender gets put in place. It can't be there at launch time or else it'd violate the no-protrusion rule, and if it's extruded after launch, I think that amounts to making a big change to the vehicle.

A couple of passive grappling posts is no big deal.  You can put them in shrouds and have them swing up and lock after you reach orbit.  I suspect that these are table stakes for all of the options.
So they'd essentially be bumps on the outside of the ship? Or would they extend inside and force a change to the shape of the propellant tanks? (It's that last that I've been assuming [possibly wrongly] would be a really big deal.)

Quote
I don't see how this option (#3) can work at all. Not only does it have lots of new plumbing, I don't see how it can do the ullage burn without spinning like a pinwheel or (maybe) having the acceleration at a crazy angle. (Maybe I just lack imagination.)
If you have both ships burning cooperatively, this isn't a huge deal.
Even if they're burning cooperatively, they're pointing in opposite directions. I don't see how you could do this without spinning around the center of mass. Unless you put extra engines in the nose or something.

Quote
You've still got the problem of holding the two vehicles together during fueling. I don't have any ideas for how to make that work, short of hoping that synchronized flying isn't infeasible, given the very low accelerations involved.
Grapple fixtures.  They're well-understood tech.  See here (https://en.wikipedia.org/wiki/Grapple_fixture).  But this does require good enough proximity ops that a simple active arm can reach out and grab them.  But you need two, spaced far enough apart that they can handle worst-case torsion.
This makes sense, but I'm wondering how much mass that would add.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/05/2022 06:07 pm
So they'd essentially be bumps on the outside of the ship? Or would they extend inside and force a change to the shape of the propellant tanks? (It's that last that I've been assuming [possibly wrongly] would be a really big deal.)

Yes, just bumps.  Possibly even just naked FRGF's sticking up into the airflow on the dorsal side.  They're not very big.

Quote
Even if they're burning cooperatively, they're pointing in opposite directions. I don't see how you could do this without spinning around the center of mass. Unless you put extra engines in the nose or something.

If it's just a regular RCS pod (likely), then each pod has thrusters that point in multiple directions.

Quote
Grapple fixtures.  They're well-understood tech.  See here (https://en.wikipedia.org/wiki/Grapple_fixture).  But this does require good enough proximity ops that a simple active arm can reach out and grab them.  But you need two, spaced far enough apart that they can handle worst-case torsion.
This makes sense, but I'm wondering how much mass that would add.

A few kg at most.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/05/2022 11:44 pm
Some cryocooler histories, experiments, and capabilities.

Duband, L., 2015. Space cryocooler developments. (https://www.sciencedirect.com/science/article/pii/S1875389215003843/pdf?md5=8bae0bd27d1eac2f38504c39bb32d476&pid=1-s2.0-S1875389215003843-main.pdf) Physics Procedia, 67, pp.1-10.

Ross, R.G., 2007. Aerospace coolers: a 50-year quest for long-life cryogenic cooling in space. (https://www2.jpl.nasa.gov/adv_tech/coolers/Cool_ppr/CEC2005%2050yr%20History%20of%20Cryo%20in%20Space.pdf) In Cryogenic Engineering (pp. 225-284). Springer, New York, NY.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 12:32 am
Tell me if this is what you've got in mind. To simplify, imagine we're designing a connector to fill/drain a single tank. Visualize a rectangle with a socket on the left and a spigot on the right. A hose runs from the socket to the bottom of the tank and another hose runs from the spigot to the top of the tank. The idea is that you'd pump liquid into the socket and extract air (ullage gas) from the spigot.

Now imagine that you have two of these contraptions. They'll just plug into each other. You don't need a version that's a mirror image, and you don't even need to flip one of them over. You've got a single type of connector with no adaptors needed.

Yep you're getting it.

You do have to flip one of them over, either vertically or horizontally. In theory you could even do it diagonally (if you're a sadomasochist).

Why? Because geometry. You can't have two Starships in the exact same orientation and plug them in. The... you know, Starship part will collide. :)

However, you've now connected the top of one tank with the bottom of the other one, but you need them to be connected bottom-to-bottom and top-to-top (unless you somehow made option #3 work). How do you switch the pipes around?

You switch the connectors around, of course.

If you want the bottom to connect to the top

What do you need that for, dude?

It's easy, yes. But I'm also not sure why you'd do it.

you just make those connectors their respective "mirror images" across the symmetry plane.

It seems like you still do need an adaptor that's essentially two pipes that cross over and have these universal connectors on each end.

It's easy. No adapter needed:


        |
  Top   |  Bottom
   X    |    O
        |
        |<-- plane of symmetry



Rotate my ASCII art picture 90° if you assume docking in the "69" configuration.

There's also the question of where the pumps are. I don't think a normal starship has pumps either to suck fuel in or push it out (except into the engines). This problem isn't specific to your universal connector design, of course, but the adaptor (given that you have to have one) seems like a nice place to put the pumps. Otherwise they're extra weight that starships don't need once they're fully fueled.

It seems like an overcomplicated place.

You're launching the mass of the pumps anyway (most logically on the depot side). No need to launch extra mass on top of that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/06/2022 12:46 am
I imagine that an androgynous topology... is a little bit dicier to engineer for a clean, reliable separation at liftoff.

Why?



Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated.

You're confusing androgynous connectors with an androgynous interface plate. You can make an androgynous interface plate (when mirrored across some plane) while still using entirely male-female connectors.

You simply put a female connector in the "mirror image" location of every male connector, and vice-versa.
Yes. That is obvious. Unfortunately if a ship is given a 'mirror image' QD plate so it can interface to tankers and receiving ships it can not interface to the ground side GSE QD plate. The discussion has been looking at:
1) a ship dedicated to depot service with a mirror QD plate and an adapter (gender bender) that is temporarily installed on the ground side GSE, or
2) a standard tanker or general purpose ship that takes on temporary duty as an accumulator after an adapter is installed on orbit. This can deployed via several mechanisms and be anything from a simple adapter to an adapter with MLI and/or PV, radiator and cryocooler.


See the pics. The first is unmodified and does not work. The second shows a mirror plate. It will work and would be implemented as described above. Exactly how the M/F are arranged is open to discussion but is irrelevant to the core issue of getting things lined out to work.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 01:20 am
androgynous topology [...] engineer for a clean, reliable separation at liftoff.

Just realized that this is an upside, not a downside.

If the QD and the on-orbit refilling connector are common, it gives them experience (including the ability to troubleshoot) on the ground which is directly applicable go on-orbit refilling.

Increased commonality between Stage 0 and Stage 2 should increase reliability, lower complexity, and lower cost.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/06/2022 01:21 am
We are back full circle. Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated.

I don't think that's a problem, because you're not going face-to-face.  Just use whatever male parts are on the tower-side QD for all the connections on the other side of the center line.  But you'll still need to deal with all the aerodynamics and deployment stuff, to say nothing of the actual disconnect dynamics.

Quote
The adapter you described is essentially the free flyer without free flying. A semi-rhetorical question: is it easier/lighter/cheaper to design/build a mechanism to move the adapter down the hull or to put a small cold gas RCS+translation system on the adapter? It's hard to judge with so many final details TBD, but the question will always be there.

If you're using cold gas, then you'll only be flying freely once, to get deployed.  If that's what it takes... meh.  Sure.  But not my first choice.

I think you can do this with extendable masts, if I measured correctly.  See the attached, highly arm-wavy cartoons.

PS:  I realized that I've conflated the terms "boom" and "mast", and also that "mast" implies something that sticks up.  All these booms/masts are are just extendable trusses to push the two hunks of the depot kit to the right spots.

There's also no reason why there have to be two masts.  You could put everything on the back and probably be fine, except it's a long way to extend something without tacking anything down as you go.  Also, putting the solar array where I did has the nice property of not interfering with the elonerons.  You might want to rotate them 90º if you adopt a nose-to-sun orientation to keep the tanks as cool as possible.
Yes, virtually as I envisioned it. And yes, cold gas for one deploy and return to the bay. It can take an hour or two to get into position with very little gas, and not really too complex. For the truss, like the emergency cell tower in the pic, but much lighter. Credit to: [size=78%]https://solaristechservices.com/portable-cell-tower-use-it-lose-it-budget/ (https://solaristechservices.com/portable-cell-tower-use-it-lose-it-budget/)[/size]

Where the panels get tricky is when a second ship is attached. That and I think we still don't have a consensus on how much heat to expect in LEO, so we don't have a size for the PV.

Hmmm. Just had a thought. Always a bad beginning. Once the gizmo is outside the bay, extend the truss towards the rear of the ship. Rear part does a grapple. Front part releases and the truss contracts bring the front part down to the rear part. Repeat as necessary. It is its own deployment arm. It's my idea and I'm in love with it. Still think gas would be easier.


Edit: just looked at your pics again. It's missing the second QD plate for the second ship. Also, why not put the forward grapple on the forward part?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 01:35 am
Trusses and QDs and masts, oh my!!



How is any of this easier than a mild redesign of the interface plate + soft capture on the depot side?

Starship will need solar panels and radiators anyway. I doubt they'll do the truss thing for that. Why not design it once instead of twice?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 03:43 am

If a tanker had 12-hour turnaround, and it delivered 1/5 of a load, it could fill an inflatable depot for ~ 300 crews each synod, by itself. 

Otherwise, a fleet of tankers would be needed.  E.g., given a 14-day window and preloading of depot tankers at synod start, at least 46 tankers would be needed. 

Scale up ~ 10x (https://aeon.co/essays/elon-musk-puts-his-case-for-a-multi-planet-civilisation) for settlement cargo flights.

The obvious idea is to use the Starships themselves as the "depot." This eliminates the entire depot fleet, which is a massive savings.

Instead you just transfer the passengers at the last minute. You can use 500 passenger P2P-derived vehicles to do that. This further increases the efficiency of launching within that 14-day window, because each launch serves five Starships instead of one.

Delete depots, get taxis.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: john smith 19 on 09/06/2022 07:19 am
We are back full circle. Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated. Image from Toledospring.com.


Unless there's a cryo variant already available it would be another materials R&D project. Not impossible but not trivial.
And something I'd assuemd had been thrashed out by SX when they were talking about FH having cross-flow from the boosters to the core. The generic term for this is a "Flange coupling."

However AIUI there's a lot more choice for OTS systems if you designate one side male and one female. Fortunately you don't need the 5000psi that some NASA designs tested in the late 60's (and which were complex and heavy).

One thing I will note from the various CP's of NASA's on-orbit servicing workshops was that "floppy" stuff, like cables and hoses is much harder to automate than connectors in rigid locations relative to each other. The former needs either an astronaught to plug them together or something with equivalent dexterity and eye/hand coordination. The latter likely needs more precise construction or a bit of "Selective compliance" built into it (like a SCARA robot arm).

Again it's the idea of not funding and running a specific R&D effort soley for this application.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/06/2022 10:20 am
Quote
If the adaptor can be pushed out of the depot cargo-area, another alternative to a free-flyer would be for the first visiting tanker to dock with it at the depot-nose, then transfer it to the depot's QD plate for the actual prop. xfer; but leave it attached to the depot when undocking. Free-flight without a free-flyer.
Probably acceptable, but it eliminates the ability for depots to back up their kit and return to EDL with it.

If you want to return the non-free-flying adaptor, you just reverse the deployment method. Last visiting ship moves it back to the cargo-area.

["Just"]

But because it's not a free-flyer (a spacecraft with full RPOD capability), the adaptor is vastly simplified, lowering its cost, making it closer to disposable. Especially early on.

(However, if an adaptor module is required, I prefer the rail deployment. Hmmm, call it "the crawler". I like the symmetry with the lander using the same system for the elevator. But I'm just pointing out that there's a third option, not just free-flyer vs crawler.)




Twark Main,

I get the impression that you've missed a couple of points about the reasoning behind people suggesting an adaptor:

It's not just to put a male/male adaptor between the female-QD ships. It is also meant to carry all the depot-necessary equipment. Cryo-coolers, pumps, perhaps solar-arrays, radiators, etc, and all the docking hardware, in a single package. Anything too large to reasonably fit in the engine-skirt, or be carried externally during launch (and certainly EDL.) The alternative is to house those in the nose of a dedicated depot-variant of Starship, which requires extra plumbing up/down the depot-ship and eliminates the possibility of refuelling regular Starships directly from tankers, or from each other.




Aside: It might be worth reposting the QD-plate architecture again. Remember, it has to be compatible with the GSE QD-arm, also shown below.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 11:05 am
We are back full circle. Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated. Image from Toledospring.com.


Unless there's a cryo variant already available it would be another materials R&D project. Not impossible but not trivial.
And something I'd assuemd had been thrashed out by SX when they were talking about FH having cross-flow from the boosters to the core. The generic term for this is a "Flange coupling."

However AIUI there's a lot more choice for OTS systems if you designate one side male and one female.

As repeatedly indicated, you can do this on the connector level without doing this on the interface level.

So the "sides" can be identical, while still using only non-androgynous male/female connectors.


We are back full circle. Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated. Image from Toledospring.com.


Unless there's a cryo variant already available it would be another materials R&D project. Not impossible but not trivial.
And something I'd assuemd had been thrashed out by SX when they were talking about FH having cross-flow from the boosters to the core. The generic term for this is a "Flange coupling."

However AIUI there's a lot more choice for OTS systems if you designate one side male and one female.

As repeatedly indicated, you can do this on the connector level without doing this on the interface level.

So both "sides" can be identical, while still having the interface plate use only non-androgynous male/female connectors.

There is no tradeoff here. We can have our cake and eat it too.

Twark Main,

I get the impression that you've missed

Didn't miss them. Just disagree.


It is also meant to carry all the depot-necessary equipment. Cryo-coolers, pumps,

Put it in the depot.

perhaps solar-arrays, radiators

Definitely put it in the depot, instead of designing it for Starship twice.

, etc, and all the docking hardware

You know what I'm gonna say, right?

Anything too large to reasonably fit in the engine-skirt, or be carried externally during launch (and certainly EDL.)

This hand-wringing about having nothing stick out even an inch on the backside during EDL is truly becoming a sickness.

The alternative is to house those in the nose of a dedicated depot-variant of Starship, which requires extra plumbing up/down the depot-ship

No need to mount it on the top. Side mount is fine. This eliminates your "extra plumbing."

and eliminates the possibility of refuelling regular Starships directly from tankers, or from each other.

This is the only real downside, as far as I can see. The latter is unnecessary, but the former could be useful.

Aside: It might be worth reposting the QD-plate architecture again. Remember, it has to be compatible with the GSE QD-arm, also shown below.

Two things:

   • it doesn't "have to" be compatible with that interface. You can also change the interface. It's a prototype, after all!

   • that QD interface looks suspiciously like it's already being designed with my proposed design in mind, even in this early prototyping phase. It already has the matching pairs across a mirror plane, they "just" need to re-plumb a bit...  ???
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/06/2022 11:16 am
A couple of passive grappling posts is no big deal.  You can put them in shrouds and have them swing up and lock after you reach orbit.  I suspect that these are table stakes for all of the options.
So they'd essentially be bumps on the outside of the ship? Or would they extend inside and force a change to the shape of the propellant tanks? (It's that last that I've been assuming [possibly wrongly] would be a really big deal.)

Grapple fixtures are used on Dragon, they fold out. (On Starship, they can probably be left exposed, but if necessary they can be folded away.) They're stupidly simple and mass 13 kg, but that's only rated for 30 tonnes, so you'd need a heavier/stronger custom variant.

Note, the grapple fixtures are only on the tanker & HLS, you need an "end effector" on the depot side. But you don't need a whole robot-arm, and it's a fairly simple system compared to a conventional docking system and highly tolerant of off-axis contact. (See the schematics, below.)

You should be able to make a modular end-effector that mounts on or in the depot's engine-skirt, next to the actual refuelling plumbing.

But, as I've mentioned in the thread before, I'd want to add more flexibility for prox.ops and a second attachment point at the nose to stabilise the ship during ullage thrust:
 
So I'd have a larger end-effector (or rather, a longer one) in the nose of the depot/accumulator for the initial contact, which extends through a hatch to create a larger stand-off distance during prox.ops. This would also have the necessary additional prox.ops hardware (such as radar/reflectors/cameras/etc.) It should also be able to tilt slightly, allowing the ships to be angled during approach to add standoff distance between the tails, to reduce the risk of collisions.

Once that initial single solid attachment is made, the nose-effector pulls the visiting ship inwards until it reaches the shorter end-effector on the depot at the engine skirt, next to a male QD-plate. This aligns the QD-plates, and the tail-side end-effector pulls the QD-plates together for final attachment.

The two effectors could be standard modules that can be added to any off-the-assembly-line Starship, to turn it into a depot/accumulator. (Assuming you don't need other depot hardware (extra-power/radiators/cryocooling) for a fast-turn-around refuelling mission.)

I've attached some pictures and NASA schematics of the grapple-fixtures. Bottom image is animated, you might have to click-through to run it.) I don't think I ever made a sketch of my proposal, if my explanation isn't not clear, just ask.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 11:39 am
it's a fairly simple system compared to a conventional docking system

Not really. On the contrary, a soft-capture system seems to be the "minimal implementation" of the necessary robot manipulator (especially after you later add a second one).

and highly tolerant of off-axis contact. (See the schematics, below.)

Is this anything but a non-theoretical advantage? I don't recall Dragon having any problems with off-axis contact.


But, as I've mentioned in the thread before, I'd want to add... a second attachment point at the nose to stabilise the ship during ullage thrust

Ahh yes, the other "sickness" in this thread.  ::)
 

the tail-side end-effector pulls the QD-plates together for final attachment.

So rather than applying smooth uniform pressure around the interface plate (like a soft capture mechanism retracting), you want to apply force in a highly cantilevered way?

That sounds like a recipe for binding and jamming. It also sounds very heavy, since you'll need to add more metal to control flexing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/06/2022 11:42 am
As repeatedly indicated, you can do this on the connector level without doing this on the interface level.
So the "sides" can be identical, while still using only non-androgynous male/female connectors.

If words fail to get your idea across, show pictures.

Put it in the depot.

My preference is for a dedicated depot-variant Starship, but the sub-topic you are replying to (the free-flying "gender bender") was explicitly about having a minimally altered Starship-tanker as the accumulator/depot that can be recovered/relaunched into different orbits for specific missions/windows.

Aside: It might be worth reposting the QD-plate architecture again. Remember, it has to be compatible with the GSE QD-arm, also shown below.
Two things:
• it doesn't "have to" be compatible with that interface. You can also change the interface. It's a prototype, after all!

It doesn't have to be compatible with the current design, but it has to be compatible with whatever GSE fuelling infrastructure is used for all launches, as a depot has to self-launch but also has to be compatible with visiting ships.

It also has to be able to dock with those visiting ships. Essentially, it has to duplicate the functionality of the GSE QD-arm during on-orbit prop. xfer, while still being compatible with that same GSE arm during launch. Ideally without having to build extra hardware mass into every Starship.

Personally I think the solution is, again, a dedicated depot that has a conventional (female) QD-plate for launch and a second (male) QD-system for prop. xfer on-orbit. The extra internal plumbing is trivial, and it means the male QD-system can have extra flexibility & functions (such as docking systems, the ability to extend out from the depot for required docking stand-off distance.)

But, again, not the sub-topic you were replying to.

It already has the matching pairs across a mirror plane, they "just" need to re-plumb a bit...  ???

The pairs are for different propellants, you can't cross-them when mating ship-to-ship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 11:48 am
As repeatedly indicated, you can do this on the connector level without doing this on the interface level.
So the "sides" can be identical, while still using only non-androgynous male/female connectors.

If words fail to get your idea across, show pictures.

I did. ASCII art counts, right?

Put it in the depot.

My preference is for a dedicated depot-variant Starship, but the sub-topic you are replying to (the free-flying "gender bender") was explicitly about having a minimally altered Starship-tanker as the accumulator/depot that can be recovered/relaunched into different orbits for specific missions/windows.

Fair enough.

Aside: It might be worth reposting the QD-plate architecture again. Remember, it has to be compatible with the GSE QD-arm, also shown below.
Two things:
• it doesn't "have to" be compatible with that interface. You can also change the interface. It's a prototype, after all!

It doesn't have to be compatible with the current design, but it has to be compatible with whatever GSE fuelling infrastructure is used for all launches, as a depot has to self-launch, but also has to be compatible with other ships.

It also has to be able to dock with those visiting ships. Essentially, it has to duplicate the functionality of the GSE QD-arm during on-orbit prop. xfer, while still being compatible with that same GSE arm during launch. Ideally without having to build extra hardware mass into every Starship.


Yes, that's exactly what using a mirror-symmetrical interface plate does. All connectors are identical, including the connector on the depot. So the depot can connect (unmodified!) to the standard QD arm.

It already has the matching pairs across a mirror plane, they "just" need to re-plumb a bit...  ???

The pairs are for different propellants, you can't cross-them when mating ship-to-ship.

Yes I noticed. Hence "re-plumb."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 12:04 pm
Maybe it will help to show a worked example of a full QD interface plate?

(I was hesitant to do this because people will inevitably get bogged down in trivial bike-shedding details)

X and O represent male and female fluid connectors.

x and o represent male and female electrical/data connectors.


        |
Data/   | Data/
Power A | Power B
   x    |    o
  CH4   |  CH4
 Drain  |  Vent
    X   |   O
  LOX   |  LOX
 Drain  |  Vent
    X   |   O
        |
        |<-- plane of symmetry



This is an androgynous interface made from 100% non-androgynous connectors.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/06/2022 12:39 pm
I think the solution is, again, a dedicated depot that has a conventional (female) QD-plate for launch and a second (male) QD-system for prop. xfer on-orbit. The extra internal plumbing is trivial...

It's good to examine real-world design decisions, puzzling out engineers' reasoning where unobvious. 

E.g., Cryo-Dock (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404365#msg2404365).  What decisions can be seen there?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2022 02:41 pm
What puzzles me about Cryo-Dock™ is that it doesn't seem to have a provision to refuel it, nor does it have an obvious provision for it to return to Earth. So it seems like a one-use thing. Given that, why does it have that big engine on the end? It seems wasteful for it to use fuel to change orbit.

It's cool the way the arm manages to move the other vehicle without it twisting around. How does it avoid a torque that rips the connector off? Just go very slowly?

The connector is male-male on the depot side, female-female on the ship side. This is why I find myself thinking the depot cannot be refueled. Not by another depot, anyway. That implies that it launches full--it's not an "accumulator." But is that reasonable, given the size of the F9 fairing? Can it lift a full depot to LEO?

Also, why did they show SLS apparently launching the exact same depot? Shouldn't it at least be a bigger one?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2022 02:47 pm
Maybe it will help to show a worked example of a full QD interface plate?
Yep. When you turn one of them around, they plug into each other just fine. But you end up with fill/drain connected to vent. You need fill/drain connected to fill/drain and vent connected to vent. (You'll have a similar problem with your power/data connection.) How do you fix that without an adaptor?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/06/2022 03:08 pm
...why does it have that big engine on the end? It seems wasteful for it to use fuel to change orbit.

...Can [F9] lift a full depot to LEO?

A full depot far exceeds F9 LEO payload capacity, clearly.  Without that engine, it doesn't reach orbit.

Also, why did they show SLS apparently launching the exact same depot? Shouldn't it at least be a bigger one?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/06/2022 03:10 pm
Maybe it will help to show a worked example of a full QD interface plate?
(I was hesitant to do this because people will inevitably get bogged down in trivial bike-shedding details)
[linear layout]

As Greg noted, F/D and gas vent are different lines that can't be crossed. But that's just a matter of extending your layout another layer. [Power/Data, helium/etc for GSE, GM prepress/vent, GOx prepress/vent, LM F/D, LOx F/D. All in a vertical line.] No bikes need to be shed.

This kind of linear layout was discussed earlier in the threat. I can't remember why everyone moved past it. [Shrug] Might have just got buried under other discussions.

My only issue is that anyone proposing a different layout make that clear and not just say "QD". Since most people will assume that anyone just saying "QD" is referring to the existing layout or something very similar. It causes unnecessary confusion, as it did here. Make it clear when you are heavily varying something that people are already talking about. (In this case, "linear QD variant" or some such wording.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: meekGee on 09/06/2022 03:46 pm


I doubt that a single set of bellows hoses will support the torques that an almost-full depot can exert against a completely full lift tanker...
I don't think anybody is suggesting bellows hoses would be structural. They just carry fluid. Am I misunderstanding?

Seems to me that there are only three possibilities:

1) The hoses are so flexible that formation-flying works and you just have to figure out how to get the hoses connected.  This has been discussed--I'm skeptical that the station-keeping is good enough to make this reliable.

2) The hoses are rigid enough to bear structural loads that hold the two ships together.  I, like you, think this is unlikely.

3) Some other structure bears the loads and the hoses don't have to worry about bearing any loads other than pressure.  This then decomposes into two sub-cases:
a) The load-bearing members are very close to the hoses.
b) They're farther away from the hoses.

If you had, say, just the QD plate and nothing else, that would be sub-case a), and I'm skeptical that it work work.  If there are structural members within, say, a meter of the hoses, would that work?  I don't know.

On the other hand, if you have one structural member latched onto the QD and one latched several meters away, I'm pretty confident the pair would bear the loads pretty easily.

In my little cartoons above, it's pretty clear the case #4, for all of its nice features, is the worst case in term of torsion.  (Well, nose-to-nose would be worse, but it's also dumb.)  But you can probably construct some set of trusses that will bear loads a meter or so from the hoses.  Is that good enough?

Case 3b, since sooner or later there will be a need to dock SS to SS for other things (e.g. crew transfer, orbit boost, etc) and why build two mechanisms?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/06/2022 04:16 pm
We are back full circle. Androgynous means face to face seals along the lines of a Westinghouse air brake glad hand but cryo rated. Image from Toledospring.com.


Unless there's a cryo variant already available it would be another materials R&D project. Not impossible but not trivial.
And something I'd assuemd had been thrashed out by SX when they were talking about FH having cross-flow from the boosters to the core. The generic term for this is a "Flange coupling."

However AIUI there's a lot more choice for OTS systems if you designate one side male and one female. Fortunately you don't need the 5000psi that some NASA designs tested in the late 60's (and which were complex and heavy).

One thing I will note from the various CP's of NASA's on-orbit servicing workshops was that "floppy" stuff, like cables and hoses is much harder to automate than connectors in rigid locations relative to each other. The former needs either an astronaught to plug them together or something with equivalent dexterity and eye/hand coordination. The latter likely needs more precise construction or a bit of "Selective compliance" built into it (like a SCARA robot arm).

Again it's the idea of not funding and running a specific R&D effort soley for this application.
I was less than clear. Floppy stuff=PV, radiators and MLI. Deploying this stuff is always a point of concern. Folding it back up for later reuse would be... Well, interesting.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/06/2022 05:15 pm


I doubt that a single set of bellows hoses will support the torques that an almost-full depot can exert against a completely full lift tanker...
I don't think anybody is suggesting bellows hoses would be structural. They just carry fluid. Am I misunderstanding?

Seems to me that there are only three possibilities:

1) The hoses are so flexible that formation-flying works and you just have to figure out how to get the hoses connected.  This has been discussed--I'm skeptical that the station-keeping is good enough to make this reliable.

2) The hoses are rigid enough to bear structural loads that hold the two ships together.  I, like you, think this is unlikely.

3) Some other structure bears the loads and the hoses don't have to worry about bearing any loads other than pressure.  This then decomposes into two sub-cases:
a) The load-bearing members are very close to the hoses.
b) They're farther away from the hoses.

If you had, say, just the QD plate and nothing else, that would be sub-case a), and I'm skeptical that it work work.  If there are structural members within, say, a meter of the hoses, would that work?  I don't know.

On the other hand, if you have one structural member latched onto the QD and one latched several meters away, I'm pretty confident the pair would bear the loads pretty easily.

In my little cartoons above, it's pretty clear the case #4, for all of its nice features, is the worst case in term of torsion.  (Well, nose-to-nose would be worse, but it's also dumb.)  But you can probably construct some set of trusses that will bear loads a meter or so from the hoses.  Is that good enough?

Case 3b, since sooner or later there will be a need to dock SS to SS for other things (e.g. crew transfer, orbit boost, etc) and why build two mechanisms?
IIUC you're suggesting using the QD plate and any associated grapples for hooking up two ships?


The QD connectors will have a currently unknown but finite cycle limit. Each lashup will preempt a refueling. Not unreasonable for an emergency but questionable for sop. Boosting orbit seems out of the question. Way too much loading in an area not designed for it.


If refueling needs grapple points, we have a seed for a lashup not involving the QD itself. Use the installed grapple points and maybe one or two others (depends on the number needed for refueling) and keep the ships just far enough apart that the QD's don't touch. This also works around any QD/mirror QD issues.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/06/2022 05:48 pm
Fail Hot vs. Fail Cold

Fail Hot

RRM3 was unable to transfer LCH4 on-orbit because the cryocooler suffered electrical failure (Breon et al. 2020). CH4 vented 72 hours after failure, despite use of advanced Quest MLI (Dye et al. 2010).

Fail Cold

Passive radiators can freeze both LCH4 and LOX in LEO (Korngut et al. 2018).  Here electrical failure freezes propellant, for max long-term passive uncontrolled storage.  Recovery of propellant remains possible with repair.

Refs.

Breon, S.R., Boyle, R.F., Francom, M.B., DeLee, C.H., Francis, J.J., Mustafi, S., Barfknecht, P.W., McGuire, J.M., Krenn, A.G., Zimmerli, G.A. and Hauser, D.M., 2020, March. Robotic Refueling Mission-3—an overview. (https://iopscience.iop.org/article/10.1088/1757-899X/755/1/012002/pdf) In IOP Conference Series: Materials Science and Engineering (Vol. 755, No. 1, p. 012002). IOP Publishing.

Dye, S., Kopelove, A. and Mills, G.L., 2010, April. Integrated and load responsive multilayer insulation. (https://aip.scitation.org/doi/pdf/10.1063/1.3422464) In AIP Conference Proceedings (Vol. 1218, No. 1, pp. 946-953). American Institute of Physics.

Korngut, P.M., Bock, J.J., Akeson, R., Ashby, M., Bleem, L., Boland, J., Bolton, D., Bradford, S., Braun, D., Bryan, S. and Capak, P., 2018, July. SPHEREx: an all-sky NIR spectral survey. (https://authors.library.caltech.edu/88799/1/106981U.pdf) In Space Telescopes and Instrumentation 2018: Optical, Infrared, and Millimeter Wave (Vol. 10698, pp. 576-586). SPIE.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2022 08:06 pm
As Greg noted, F/D and gas vent are different lines that can't be crossed. But that's just a matter of extending your layout another layer. [Power/Data, helium/etc for GSE, GM prepress/vent, GOx prepress/vent, LM F/D, LOx F/D. All in a vertical line.] No bikes need to be shed.
I'm afraid I don't see how that would work. Are these connectors male or female? What does "another layer" mean? How does putting them in a vertical line help anything?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 08:14 pm
Maybe it will help to show a worked example of a full QD interface plate?
Yep. When you turn one of them around, they plug into each other just fine. But you end up with fill/drain connected to vent. You need fill/drain connected to fill/drain and vent connected to vent. (...) How do you fix that without an adaptor?

There seems to be a bit of disagreement on whether that's a "need," but...

That's easy too. If you want to connect like-to-like, you can just have two connectors (one male and one female) on either side. No adapters or androgynous ports needed.

This arguably increases reliability, because now you have redundancy. If one connector fails you can isolate it and refill (more slowly) on the remaining good connection, and return the hardware for troubleshooting. Without any such redundancy, the entire mission would be a loss.

Edit: Looks like Paul451 beat me to it! (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404880#msg2404880)

You'll have a similar problem with your power/data connection.

I'm afraid I don't follow. What problem are you referring to here?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 08:23 pm
If there are structural members within, say, a meter of the hoses, would that work?  I don't know.

On the other hand, if you have one structural member latched onto the QD and one latched several meters away, I'm pretty confident the pair would bear the loads pretty easily.

The rare (or maybe not so rare) "Argument from I Didn't Feel Like Doing The Math."  :-\

NASA seems to have no problem with structural connections close to fluid connections (see: all their docking standards), so why should TheRadicalModerate?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2022 08:51 pm
That's easy too. If you want to connect like-to-like, you can just have two connectors (one male and one female) on either side. No adapters or androgynous ports needed.
Yes, but if you do that you've simply built an adaptor into the QD plate. Just to be sure we're talking about the same thing, on the methane row (for example), we'd have something like this:

G   F/D     F/D G

X    O   |   X    O

But now you've added plumbing, and it's twisty plumbing with lots of Y connectors. Maybe I'm overreacting, but I was impressed by someone's post a month or so ago talking about how difficult plumbing for cryogens is.

Anyway, it shows it can be done--in theory. The question is whether having a universal connector is worth making each connector twice as complicated.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/06/2022 09:04 pm
That's easy too. If you want to connect like-to-like, you can just have two connectors (one male and one female) on either side. No adapters or androgynous ports needed.
Yes, but if you do that you've simply built an adaptor into the QD plate.

Not really, no.

Just to be sure we're talking about the same thing, on the methane row (for example), we'd have something like this:

G   F/D     F/D G

X    O   |   X    O

That's another one that will work!

Happy to see people getting the hang of it.

But now you've added plumbing,

Still less than the adapter. Remember that in this scenario each connection only needs to support half the flow rate, so the size and mass of the connectors is reduced.

it's twisty plumbing with lots of Y connectors.

Welcome to redundant fluid system design! :)

The adapter is a brittle design. If the connector on either "side" fails, the whole refilling operation fails.

By comparison, using an androgynous interface plate adds reliability/redundancy.

Removing single points of failure is better than adding single points of failure.

Anyway, it shows it can be done--in theory. The question is whether having a universal connector is worth making each connector twice as complicated.

Not "twice as complicated."

Certainly not twice as heavy either, which is a relevant metric. Having an "adapter" is more than twice as heavy, by contrast.

We all know there ain't no free lunch, but (of the two) my lunch does seem to be the better bargain.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/06/2022 09:38 pm
That's easy too. If you want to connect like-to-like, you can just have two connectors (one male and one female) on either side. No adapters or androgynous ports needed.
Yes, but if you do that you've simply built an adaptor into the QD plate. Just to be sure we're talking about the same thing, on the methane row (for example), we'd have something like this:

G   F/D     F/D G

X    O   |   X    O

But now you've added plumbing, and it's twisty plumbing with lots of Y connectors. Maybe I'm overreacting, but I was impressed by someone's post a month or so ago talking about how difficult plumbing for cryogens is.

Anyway, it shows it can be done--in theory. The question is whether having a universal connector is worth making each connector twice as complicated.

Eta Space (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404365#msg2404365) made their connector decision, shunning such complexity.  The engineers had reason, yes? 

What puzzles me about Cryo-Dock™ is that it doesn't seem to have a provision to refuel it...

And you noticed their propulsive transfer?  What prevents propulsive refill of the depot there?  Any future reusable Eta Space tanker could present a QD plate; depots and tankers can fly with complementary connectors, after all.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 09/07/2022 08:48 am
A couple of passive grappling posts is no big deal.  You can put them in shrouds and have them swing up and lock after you reach orbit.  I suspect that these are table stakes for all of the options.
So they'd essentially be bumps on the outside of the ship? Or would they extend inside and force a change to the shape of the propellant tanks? (It's that last that I've been assuming [possibly wrongly] would be a really big deal.)

Grapple fixtures are used on Dragon, they fold out. (On Starship, they can probably be left exposed, but if necessary they can be folded away.) They're stupidly simple and mass 13 kg, but that's only rated for 30 tonnes, so you'd need a heavier/stronger custom variant.

Note, the grapple fixtures are only on the tanker & HLS, you need an "end effector" on the depot side. But you don't need a whole robot-arm, and it's a fairly simple system compared to a conventional docking system and highly tolerant of off-axis contact. (See the schematics, below.)

You should be able to make a modular end-effector that mounts on or in the depot's engine-skirt, next to the actual refuelling plumbing.

But, as I've mentioned in the thread before, I'd want to add more flexibility for prox.ops and a second attachment point at the nose to stabilise the ship during ullage thrust:
 
So I'd have a larger end-effector (or rather, a longer one) in the nose of the depot/accumulator for the initial contact, which extends through a hatch to create a larger stand-off distance during prox.ops. This would also have the necessary additional prox.ops hardware (such as radar/reflectors/cameras/etc.) It should also be able to tilt slightly, allowing the ships to be angled during approach to add standoff distance between the tails, to reduce the risk of collisions.

Once that initial single solid attachment is made, the nose-effector pulls the visiting ship inwards until it reaches the shorter end-effector on the depot at the engine skirt, next to a male QD-plate. This aligns the QD-plates, and the tail-side end-effector pulls the QD-plates together for final attachment.

The two effectors could be standard modules that can be added to any off-the-assembly-line Starship, to turn it into a depot/accumulator. (Assuming you don't need other depot hardware (extra-power/radiators/cryocooling) for a fast-turn-around refuelling mission.)

I've attached some pictures and NASA schematics of the grapple-fixtures. Bottom image is animated, you might have to click-through to run it.) I don't think I ever made a sketch of my proposal, if my explanation isn't not clear, just ask.


Hrm, in the soft capture realm, you also have JonGoff's stickyboom sticky plates as well, plus the electropermanent magnets proposed for some small scale docking adapters like DogTags.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 09/07/2022 10:34 am
For the "long vertically mirrored QD plate" concept it work, it actually needs to fit onto Starship. That's currently constrained by aft skirt height being finite, and required pot area being nonzero.
You also end up with every connector becoming at least a pair of connectors, doubling or more dry the mass (even if each individual connector is half the size, you duplicate non-scaling items like latches, and add extra Y plumbing pieces). By adding an androgynous QD plate to every Starship in order to avoid needing to fly an 'adapter' plate on one Starship (whichever you've designated to pull depot duty), you've added the same mass as the adapter to every Starship rather than one Starship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Overtone on 09/07/2022 01:01 pm

I haven't followed the whole thread. Please forgive me if this repeats earlier discussion.

The two starships aren't rigid objects.  They are lightweight shells surrounding many tons of fuel.  In the normal case, the starships have partially filled tanks - e.g. a tanker reaching orbit has 150 tons of fuel in tanks sized for many times that amount. The tanks don't have extensive baffle systems.  Maybe a few baffles associated with managing slosh during the landing flip.

What this means is that any change in the sign of acceleration in any of the 3 translation or 3 roll axes will be followed some 10s of seconds later by a hard-to-predict series of wobbles or bumps in the orientation and acceleration of the starship - as tons of fuel splashes down onto the other side of the tank or turbulently changes its flow direction. Needless to say, those uncontrolled wobbles and bumps are hugely dangerous for prox-ops.

It seems to me the choice of refueling geometry will be affected as much by this challenge as by any of the mechanical engineering challenges that have been discussed so thoroughly above.  Those of you advocating for each of the refueling geometries: how would you get into and out of that configuration without any change in the sign of acceleration in any of the 6 axes when the two starships are in proximity but not grappled?  This is basically a slow, ponderous hoverslam.

You need both starships to be accelerating continuously in the same direction, throughout the prox-ops, in any translation or roll axis that isn't continuously zero.  You can move, roll or pitch them relative to the other by adjusting their relative acceleration. However, watch out: since you are in a spiraling reference frame in orbit, relative motion along the orbital path is coupled to relative motion in orbital altitude.

The nose-to-tail geometry, both pointing in the same direction, appears to allow a safe prox-ops approach.  The trailing starship would approach from below, rising up as it translates along the orbital path into position, while pitching its nose towards the tail of the leading ship (continuously decelerating in pitch rate until grapples activate).  I don't see a safe prox-ops for the other geometries.  Do you?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/07/2022 02:59 pm
What this means is that any change in the sign of acceleration in any of the 3 translation or 3 roll axes will be followed some 10s of seconds later by a hard-to-predict series of wobbles or bumps in the orientation and acceleration of the starship - as tons of fuel splashes down onto the other side of the tank or turbulently changes its flow direction. Needless to say, those uncontrolled wobbles and bumps are hugely dangerous for prox-ops.

.
.
.

The nose-to-tail geometry, both pointing in the same direction, appears to allow a safe prox-ops approach.  The trailing starship would approach from below, rising up as it translates along the orbital path into position, while pitching its nose towards the tail of the leading ship (continuously decelerating in pitch rate until grapples activate).  I don't see a safe prox-ops for the other geometries.  Do you?

Since the accelerations in question will be on the order of 20 micro-g, this may not be a bad as you think. Check this document (https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf) for "Low Acceleration Settling."
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/07/2022 08:26 pm
If there are structural members within, say, a meter of the hoses, would that work?  I don't know.

On the other hand, if you have one structural member latched onto the QD and one latched several meters away, I'm pretty confident the pair would bear the loads pretty easily.

The rare (or maybe not so rare) "Argument from I Didn't Feel Like Doing The Math."

Challenged accepted--sorta.  See below.

Quote
NASA seems to have no problem with structural connections close to fluid connections (see: all their docking standards), so why should TheRadicalModerate?

It's not a question of structural connections close to fluid connections.  It's a question of whether you can build a structural connection that couples reliably and can still support the worst-case torsion of the system.

You want to put the connection waaaaaaaay at the back of the vehicle, far from the center of gravity, and you want a soft-dock instead of a grapple/berthing.  (You can't do a grapple without an arm, and that really will mess up the aerodynamics, even if your androgynous QD plate doesn't.)  That subjects the vehicle to two separate torsional risks:

1) An RCS error (stuck thruster, software problem, etc.).  This is dependent on the RCS thruster, for which we don't know the thrust.  But to have decent control authority over a 270t vehicle that's 50m long, it's fair to say that the thrust isn't going to be as small as on most other vehicles.

2) Much more serious IMO, you have a big soft-capture problem, because all of the translation velocity errors are going to generate big torques.

Let's continue with your Orbiter comparison, because we know that the Orbiter docked away from the center of mass:

Orbiter Length:  37m
Docking position (from nose): 11.6m (derived from measuring pixels)
Nominal Center of Mass: 67% of length = 24.8m (I got this from here (https://archive.org/stream/nasa_techdoc_19790013835/19790013835_djvu.txt))
Dry Mass: 78t
Max Payload for ISS docking:  Couldn't find this, but I'd be surprised if it was >8t.

So, assuming that the up-payload was placed on the CoM, that would give you a moment of inertia of (86t)*(24.8-11.6)² = 1.5E7kg-m².

(Note that I'm assuming that the ISS has infinite mass for this little exercise.  I'll also be assuming that the depot Starship has the same infinite mass.  So the Orbiter and the tanker Starship are the sole active components in the system.  Obviously not true, but close enough for government work.)

We can't really compute torques without knowing a whole bunch about stiffness of the vehicles and the torsional springs in the docking system, but we can compute an angular momentum in Orbiter-normalized units, where ⍵orb = (24.8-11.6) * (maxTranslate).  This would be the rotational speed after contact if the Orbiter and ISS were completely rigid.

Let's assume that maxTranslate is the same for the Orbiter and a tanker, so if we know the CoM of the tanker and the docking point, we can get an apples-to-apples comparison.

I have a model that shows that a Starship tanker with 150t of prop and 120t dry mass has a CoM that's about 30m from the nose.  If we assume that the QD is 49m from the nose, then Itanker = (270t)*(49-30)² = 9.7E7kg-m², and ⍵tanker is 1.4x higher than the Orbiter.

All told, angular momentum for the Starship tanker is 9.1x larger than that on the Orbiter, given similar translational velocity limits.

I never got an answer from you (other than something cryptic and non-responsive) about whether your "tilt the QD plate" scheme tilted it in the x-y plane or x-z plane.  Either way, though, a docking miss is going to result in a collision:  tilted in the x-y plane, you're going to have elonerons colliding, and in the x-z plane, the fuselages will collide.

The more I think about it, I'm pretty sure that any kind of traditional soft-dock is fraught with peril.  I think they'll use grappling--kinda like berthing but with just enough stability to connect the plumbing.  Given that you can't put active grapples on the Starship during launch or landing, they'll have to be deployed.  And if you're going to do that, deploying the QD-to-QD adapter seems like the least difficult thing to do.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Overtone on 09/08/2022 12:13 am
What this means is that any change in the sign of acceleration in any of the 3 translation or 3 roll axes will be followed some 10s of seconds later by a hard-to-predict series of wobbles or bumps in the orientation and acceleration of the starship - as tons of fuel splashes down onto the other side of the tank or turbulently changes its flow direction. Needless to say, those uncontrolled wobbles and bumps are hugely dangerous for prox-ops.

.
.
.

The nose-to-tail geometry, both pointing in the same direction, appears to allow a safe prox-ops approach.  The trailing starship would approach from below, rising up as it translates along the orbital path into position, while pitching its nose towards the tail of the leading ship (continuously decelerating in pitch rate until grapples activate).  I don't see a safe prox-ops for the other geometries.  Do you?

Since the accelerations in question will be on the order of 20 micro-g, this may not be a bad as you think. Check this document (https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf) for "Low Acceleration Settling."

Thanks for the very interesting reference!   However, I don't think it directly addresses my concern.  Those low accelerations are for settling propellant.   If starships relied on those low accelerations during docking ops, wouldn't the time from <at relative rest at standoff distance> to <grappled and ready to connect to the QD> be excessively long?

Do you happen to have a reference that shows typical accelerations used during precise navigation in the last 100 meters of current docking operations?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 09/08/2022 09:26 am
What this means is that any change in the sign of acceleration in any of the 3 translation or 3 roll axes will be followed some 10s of seconds later by a hard-to-predict series of wobbles or bumps in the orientation and acceleration of the starship - as tons of fuel splashes down onto the other side of the tank or turbulently changes its flow direction. Needless to say, those uncontrolled wobbles and bumps are hugely dangerous for prox-ops.

.
.
.

The nose-to-tail geometry, both pointing in the same direction, appears to allow a safe prox-ops approach.  The trailing starship would approach from below, rising up as it translates along the orbital path into position, while pitching its nose towards the tail of the leading ship (continuously decelerating in pitch rate until grapples activate).  I don't see a safe prox-ops for the other geometries.  Do you?

Since the accelerations in question will be on the order of 20 micro-g, this may not be a bad as you think. Check this document (https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf) for "Low Acceleration Settling."

Thanks for the very interesting reference!   However, I don't think it directly addresses my concern.  Those low accelerations are for settling propellant.   If starships relied on those low accelerations during docking ops, wouldn't the time from <at relative rest at standoff distance> to <grappled and ready to connect to the QD> be excessively long?

Do you happen to have a reference that shows typical accelerations used during precise navigation in the last 100 meters of current docking operations?
Closing from 100 metres with an acceleration cap of 20 microgee = ~1400s for a brachistochrone approach: accelerate halfway to about 0.15m/s, then decelerate for the other half to a stop. Or less than 25 minutes.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/08/2022 04:01 pm
Closing from 100 metres with an acceleration cap of 20 microgee = ~1400s for a brachistochrone approach: accelerate halfway to about 0.15m/s, then decelerate for the other half to a stop. Or less than 25 minutes.
I get the same answer. (I don't know about you, but I always like it when someone double-checks my math--at least if they get the same answer.) :-)

Fuel use seems to be really small too. If I'm doing this right, (and assuming equal specific impulse), the fuel used to accelerate for one second at one full g should last almost 14 hours at 20 µg.

This must be why they're not really worried about settling burns wasting a whole lot of fuel. (Lots of people have said this, but I never ran the numbers before.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/08/2022 06:58 pm
Closing from 100 metres with an acceleration cap of 20 microgee = ~1400s for a brachistochrone approach: accelerate halfway to about 0.15m/s, then decelerate for the other half to a stop. Or less than 25 minutes.

I'd expect to spend a lot more propellant on attitude control during prox ops than on actually closing and braking.

Overtone's note about propellant slosh isn't as serious as it would be at higher acceleration, but it still puts limits on your attitude and translation errors, especially just before docking/grappling.  I wonder if a cryogenic propellant management device could tolerate a large enough acceleration to be useful?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/08/2022 10:37 pm
I had a thought about making an adaptor work without having to fly it down from the cargo bay. Suppose it just lay on its side right next to the QD, attached to the body of the rocket by a hinge. I visualize it lying along the body of the rocket from the QD forward.

Once in orbit, it would rotate out 90 degrees, plugging itself into the QD in the process. The adaptor could then be as long as necessary--or even be a robot arm--with a suitable connector at the end for the other Starship.

This presupposes a few things that I'm not really sure of. First, is it possible to design something like that that rotates into place and yet still makes a tight-enough seal? (And doesn't interfere with the GSE that fuels it on the ground.) Second, is this something small enough to be covered with a "shroud" or something like that and still keep the rocket aerodynamic enough?

It still has to be powered and controlled somehow (and, presumably, the pumps that control the actual refueling operation would go inside the adaptor somehow), but it at least seems reasonable to me. (Apologies if someone has already proposed this.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/08/2022 10:37 pm
Closing from 100 metres with an acceleration cap of 20 microgee = ~1400s for a brachistochrone approach: accelerate halfway to about 0.15m/s, then decelerate for the other half to a stop. Or less than 25 minutes.

I'd expect to spend a lot more propellant on attitude control during prox ops than on actually closing and braking.

Overtone's note about propellant slosh isn't as serious as it would be at higher acceleration, but it still puts limits on your attitude and translation errors, especially just before docking/grappling.  I wonder if a cryogenic propellant management device could tolerate a large enough acceleration to be useful?
During a low acceleration maneuver surface tension would probably be the dominating force influencing propellant moving as a large glob vs splashing.


If it's moving in a glob might real time modeling predict the resulting thrusts needed to keep it as a single body and still achieve the objective of the maneuver? Am I overthinking this?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/09/2022 11:10 am
I had a thought about making an adaptor work without having to fly it down from the cargo bay. Suppose it just lay on its side right next to the QD, attached to the body of the rocket by a hinge. I visualize it lying along the body of the rocket from the QD forward.

You could certainly do this with a simple adaptor plate. (Or just have a second (male) QD plate on the depot.) But my understanding was that the "adaptor" was proposed to carry all the depot-specific hardware, including the cryocooler/docking-systems/etc, allowing any tanker to serve as the depot/accumulator. Something that size, probably can't be kept in in the engine-skirt, but would have to be carried into orbit in the nose. (This is the case, whether the depot hardware is carried as a free-flyer/crawler, or built in permanently.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/09/2022 02:30 pm
I had a thought about making an adaptor work without having to fly it down from the cargo bay. Suppose it just lay on its side right next to the QD, attached to the body of the rocket by a hinge. I visualize it lying along the body of the rocket from the QD forward.
You could certainly do this with a simple adaptor plate. (Or just have a second (male) QD plate on the depot.) But my understanding was that the "adaptor" was proposed to carry all the depot-specific hardware, including the cryocooler/docking-systems/etc, allowing any tanker to serve as the depot/accumulator. Something that size, probably can't be kept in in the engine-skirt, but would have to be carried into orbit in the nose. (This is the case, whether the depot hardware is carried as a free-flyer/crawler, or built in permanently.)
I was thinking of the adaptor handling the docking system, but not the cryocooling. I could see the cryocooler sitting at the top of the Starship, assuming it can cool the tanks from there--via conduction, maybe.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/09/2022 05:01 pm
I had a thought about making an adaptor work without having to fly it down from the cargo bay. Suppose it just lay on its side right next to the QD, attached to the body of the rocket by a hinge. I visualize it lying along the body of the rocket from the QD forward.
You could certainly do this with a simple adaptor plate. (Or just have a second (male) QD plate on the depot.) But my understanding was that the "adaptor" was proposed to carry all the depot-specific hardware, including the cryocooler/docking-systems/etc, allowing any tanker to serve as the depot/accumulator. Something that size, probably can't be kept in in the engine-skirt, but would have to be carried into orbit in the nose. (This is the case, whether the depot hardware is carried as a free-flyer/crawler, or built in permanently.)
I was thinking of the adaptor handling the docking system, but not the cryocooling. I could see the cryocooler sitting at the top of the Starship, assuming it can cool the tanks from there--via conduction, maybe.
That would require new plumbing. If you take boil off from the QD repress lines, you can reliquify and feed the liquid back through the fill/drain lines.

If you’re going to add plumbing, the nose-to-tail configuration might make more sense. It’s a lot easier to deploy.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: john smith 19 on 09/09/2022 06:28 pm

I was less than clear. Floppy stuff=PV, radiators and MLI. Deploying this stuff is always a point of concern. Folding it back up for later reuse would be... Well, interesting.
Absolutely.

Hence my belief they would be the key issues for a Mars SS flight, and my belief they'd have to take a "minimum viable vehicle" IE no radiators, no PV, if they wanted to even try for a launch this year.

The "Space mechanisms" conference proceedings make very interesting reading for people who like seeing (to coin a phrase) "tricky mechanical gadgets." One of my favourites was the system that deployed the "aero-disk" on the front of the Trident missile to lower its drag. An amazing mechno-explosive subsystem. Stunning.  8)

One theme that repeatedly comes up in these volumes are the issues around  friction. :(
 Incompatible materials, faulty lubrication etc stopping something deploying (or less frequently stowing). Thermal cycling also seems to have caused an inordinate amount of trouble over the years. Parts distorted so they can't  move. Parts (specifically long booms for example) unevenly heated so they bend, and the mass at the end starts to cause the boom to rotate (in a very peculiar motion) going into and out of Earth eclipse.

While I expect there are several options that could be made to work eventually keeping it as simple as possible sounds like a good idea. Also keeping things fairly close together seems like a good way to keep things stiff, and hence less prone to distortion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 09/09/2022 06:41 pm
"NASA’s Initial Artemis Human Landing System" (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf)

w/depot.  h/t Chris Bergin.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Hog on 09/09/2022 06:45 pm
h/t= hat tip = tip of the hat

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/09/2022 08:29 pm
"NASA’s Initial Artemis Human Landing System" (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf)

w/depot.  h/t Chris Bergin.
Well, that was educational! Still lots of issues to talk about, but that seems to remove a lot of degrees of freedom.

So if I read this correctly, the plan is for a depot with no active cryocooler--just "exterior optical properties" and lots of spare capacity to allow for tolerance of some amount of boiloff. That simplifies things a lot.

Presumably something like Solar White would suffice for the HLS, since it only needs to "loiter" in the NRHO.

And, as others have said, the tanker is just a regular starship with larger tanks and little or no cargo space.

Instead of (or maybe in addition to) a QD plate, I'd expect the depo to have something like the existing GSE that's used to fill/drain the starships on the launchpad. Exactly how they'd squeeze that in there is a good question, but it seems to make the most sense, given everything else.

At least, that's my read on it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 09/09/2022 08:37 pm
I still think the tail to tail docking for fuel transfer was the most optimum safety and good solid docking of all connections. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Stan-1967 on 09/10/2022 02:38 am
Closing from 100 metres with an acceleration cap of 20 microgee = ~1400s for a brachistochrone approach: accelerate halfway to about 0.15m/s, then decelerate for the other half to a stop. Or less than 25 minutes.

I'd expect to spend a lot more propellant on attitude control during prox ops than on actually closing and braking.

Overtone's note about propellant slosh isn't as serious as it would be at higher acceleration, but it still puts limits on your attitude and translation errors, especially just before docking/grappling.  I wonder if a cryogenic propellant management device could tolerate a large enough acceleration to be useful?

I appreciate all the thought going into this discussion on docking & prop slosh. 

Would it make sense to dock using primarily V-Bar approaches between the tanker & depot? If you align the longitudinal axis of the ships with the velocity vector, you would be settling the prop  towards the aft end of each respective vehicle during your approach & hard dock.  You would get some minimal sloshing from counter thrusting in the radial vector.  It seems like prop baffles already in the bottom of the LOX/CH4 tanks should dampen the transient forces when hard dock occurs.

The last 10 meters or so may be better done on an R-Bar, as I think the slosh from the lighter vessel that undergoes a greater differential acceleration at the instant of hard dock would see the prop translating such that the tank walls would absorb the forces in compression or tension along the outer radius of the tank walls, vs. having prop slosh against the domes?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/10/2022 02:50 pm
I was thinking of the adaptor handling the docking system, but not the cryocooling. I could see the cryocooler sitting at the top of the Starship, assuming it can cool the tanks from there--via conduction, maybe.
That would require new plumbing. If you take boil off from the QD repress lines, you can reliquify and feed the liquid back through the fill/drain lines.
Well, that's what you do if you're trying to be perfect, but I was thinking of simply cooling the stainless steel top of the upper tank as a way of sucking out some of the excess heat. Without adding any plumbing below that point.

Looking at a cut-away view of Starship (https://www.humanmars.net/2021/01/cutaway-schematic-of-spacex-starship.html), I see the methane tank is above the oxygen tank, which messes this idea up, but imagine that, in the depot, the oxygen tank were at the top, not the bottom. (Is there a reason the methane tank is on top?) Then you could cool the top of the LOX tank to just above the freezing temperature of oxygen. There would be a temperature gradient, with the tank getting warmer the further down you went, but steel conducts heat pretty well, so there might not be that much difference from the top to the bottom. Even if it didn't work perfectly, that'd be okay if the goal were merely to reduce boiloff--not eliminate it entirely.

You wouldn't want the methane tank to freeze, of course, but there's already insulation separating them, and adding heat (at the very bottom, of course) is a much easier problem. Also, you could do an ullage burn from time to time, which would pull the colder tank contents to the warmer bottoms and lift the warmer contents (and all the gas) to the cold upper surfaces.

No extra plumbing required. However, I suspect swapping the oxygen and methane tanks would count as a major change. Pity. Otherwise, this would offer a way to add cryocooling with (what seems to me to be) a simple incremental change. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 09/10/2022 05:12 pm
I was thinking of the adaptor handling the docking system, but not the cryocooling. I could see the cryocooler sitting at the top of the Starship, assuming it can cool the tanks from there--via conduction, maybe.
The cryocooler is cooling the cryogenic liquid inside a large steel tank. Cooling of the tank walls will be by convection whether you want it to or not.
(Edited to fix the quoting.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/10/2022 05:30 pm

I was less than clear. Floppy stuff=PV, radiators and MLI. Deploying this stuff is always a point of concern. Folding it back up for later reuse would be... Well, interesting.
Absolutely.

Hence my belief they would be the key issues for a Mars SS flight, and my belief they'd have to take a "minimum viable vehicle" IE no radiators, no PV, if they wanted to even try for a launch this year.

The "Space mechanisms" conference proceedings make very interesting reading for people who like seeing (to coin a phrase) "tricky mechanical gadgets." One of my favourites was the system that deployed the "aero-disk" on the front of the Trident missile to lower its drag. An amazing mechno-explosive subsystem. Stunning.  8)

One theme that repeatedly comes up in these volumes are the issues around  friction. :(
 Incompatible materials, faulty lubrication etc stopping something deploying (or less frequently stowing). Thermal cycling also seems to have caused an inordinate amount of trouble over the years. Parts distorted so they can't  move. Parts (specifically long booms for example) unevenly heated so they bend, and the mass at the end starts to cause the boom to rotate (in a very peculiar motion) going into and out of Earth eclipse.

While I expect there are several options that could be made to work eventually keeping it as simple as possible sounds like a good idea. Also keeping things fairly close together seems like a good way to keep things stiff, and hence less prone to distortion.
A tangent, not to be taken too far. Most folding structures appear to be light, high precision, close tolerance designs. Would there be any gains in a heavier, low tolerance and sloppy structure? It's not like an extra 20 kilos is going to hurt much.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/10/2022 05:44 pm
"NASA’s Initial Artemis Human Landing System" (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf)

w/depot.  h/t Chris Bergin.
Interesting. It shows a dedicated stretched depot. The tanker is based on a standard SS. 'Based on' does not preclude stretched tanks within the standard OML (outer mold line). Have these guys been reading our mail? 8)


A question for real rocket scientists. Could a dedicated depot of this design make orbit with minimum propellant and only three engines? A lot of 'ifs' there. Just sorta, kinda, maybe?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/10/2022 07:07 pm
I was thinking of the adaptor handling the docking system, but not the cryocooling. I could see the cryocooler sitting at the top of the Starship, assuming it can cool the tanks from there--via conduction, maybe.
That would require new plumbing. If you take boil off from the QD repress lines, you can reliquify and feed the liquid back through the fill/drain lines.
Well, that's what you do if you're trying to be perfect, but I was thinking of simply cooling the stainless steel top of the upper tank as a way of sucking out some of the excess heat. Without adding any plumbing below that point.

Looking at a cut-away view of Starship (https://www.humanmars.net/2021/01/cutaway-schematic-of-spacex-starship.html), I see the methane tank is above the oxygen tank, which messes this idea up, but imagine that, in the depot, the oxygen tank were at the top, not the bottom. (Is there a reason the methane tank is on top?) Then you could cool the top of the LOX tank to just above the freezing temperature of oxygen. There would be a temperature gradient, with the tank getting warmer the further down you went, but steel conducts heat pretty well, so there might not be that much difference from the top to the bottom. Even if it didn't work perfectly, that'd be okay if the goal were merely to reduce boiloff--not eliminate it entirely.

You wouldn't want the methane tank to freeze, of course, but there's already insulation separating them, and adding heat (at the very bottom, of course) is a much easier problem. Also, you could do an ullage burn from time to time, which would pull the colder tank contents to the warmer bottoms and lift the warmer contents (and all the gas) to the cold upper surfaces.

No extra plumbing required. However, I suspect swapping the oxygen and methane tanks would count as a major change. Pity. Otherwise, this would offer a way to add cryocooling with (what seems to me to be) a simple incremental change.
A couple of misconceptions. Stainless steel has lousy thermal conductivity for a metal. And there is no insulation between LOX and LCH4. Just that poor thermally conductive stainless steel common dome.


If there are separate tank fill and drain connections, little extra plumbing is needed for a cooler. If the cryo cooler is part of a free flyer there would be taps in the adapter and external lines to the cooler. If they go with the dedicated tanker, as it looks like they're doing, it gets more complex, which is one reason for my question about the tanker making orbit with only three engines.


Three engines lowers dry mass making orbit changes and the depot itself, less expensive.  If they decide to do orbit changes. Three engines also open up a lot of space in the engine bay. They'll most likely (reading the paper LMT linked is next on my list) start without a cooler and depend on solar white paint but if a cooler is needed an engine bay location makes any plumbing runs manageably short. The cooler lines would most likely be much smaller than the main feeds. An engine bay cooler is also convenient for deploying a radiator.


If there is no separate drain line it gets more complex but not impossible.


Side note: If they do an engine bay cooler it's hard to say where the solar panels would be. Deploying from the engine bay puts them near the primary draw, the cooler. It also puts them far from the batteries, up above the methane tank. That would call for substantial cabling. OTOH, putting the PV in the nose, nearer the batteries, reduces that cable run but the run from the batteries to the cooler would still be substantial. Much hinges on what the power draw and operating cycle of the cooler might be. IF they can launch with three engines and IF the cooler goes into the engine bay it might be a reason for relocating the batteries and solar panels to the engine bay for this low production rate variant.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/10/2022 08:01 pm
"NASA’s Initial Artemis Human Landing System" (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf)

w/depot.  h/t Chris Bergin.

There are a few weird things in here:

1) Every single author is from MSFC.  There are no SpaceX authors.  It gives the paper a, "See, MSFC isn't irrelevant after all!" tone.

2)  There are multiple references to the SuperHeavy booster "delivering x to Earth orbit."  That's incorrect, and a very "SLS Block 1 Core" kind of thing to say.  It might just be a verbal tic, or it might indicate that the person in charge of drafting the text was a PR person who didn't understand the system that well.

3) "Aft docking mechanism designs—key to the SpaceX propellant transfer architecture - have continued to mature."  (p. 3, sec. 2.2.) Does this mean that tail-to-tail refueling is still on the table, that there's a QD-to-QD dock, or was the paper drafted before SpaceX threw in the towel on tail-to-tail?

4) Last paragraph of p.3:

Quote
Additional focus has also been placed on the development of the single spacecraft-to-spacecraft interface that exists in this architecture: the docking of HLS Starship with Orion. The unique architecture proposed by SpaceX for landing on the Moon uses one single-stage landing spacecraft. This means that, while several rendezvous and docking maneuvers will be required to load propellant into the HLS Starship before the crewed mission begins, there are no major docking or undocking events required to execute the lander’s mission, apart from the rendezvous in NRHO with Orion.

It's not the only spacecraft-to-spacecraft interface; refueling is such an interface.  And several RPODs are not required to load propellant into the HLS Starship; only one is, with the full depot.  All of the other RPODs are between tankers and the depot.

A lot of this could be chalked up to sloppy wording, but it also seems like this might have been a paper that sat on the shelf for a while, waiting for IAC.  As a result, it might be a bit stale.

Update:

5) Stretching the outer mould line of the depot makes no sense whatsoever.  Moving the domes and bulkheads forward makes sense, but the depot will never be filling a vehicle larger than the capacity of the existing tanks.  Unless there's a really, really serious boiloff problem, everything can be done with one OML.  And even then, topping the depot up with a tanker launch just before the LSS launch solves the problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/10/2022 08:22 pm
A couple of misconceptions. Stainless steel has lousy thermal conductivity for a metal. And there is no insulation between LOX and LCH4. Just that poor thermally conductive stainless steel common dome.

Not only that, but the bulk of the cooling effects would be lost to solar or albedo radiation impinging on the LCH4 tank sides.  The tank walls would be back in equilibrium before  the cooling effects made it to the LOX tank.

Quote
If there are separate tank fill and drain connections, little extra plumbing is needed for a cooler. If the cryo cooler is part of a free flyer there would be taps in the adapter and external lines to the cooler. If they go with the dedicated tanker, as it looks like they're doing, it gets more complex, which is one reason for my question about the tanker making orbit with only three engines.

I've been assuming that the process of tying the two ullage gas pre-press systems together is a matter of opening valves that already exist.  If that's the case, then tapping off the pre-press lines in the adapter is simple.  Then you have the following cycle:

1) As boiloff pressure builds up, rather than venting to space, you open the pre-press valves and gas flows into the crycooler.

2) Cryocoolers (one for each of GOX and GCH4) re-liquify the gas and store the liquid in a dewar.  I'd think the temperature drop across the system is small enough that a single-stage Joule-Thomson cooler would work.  All the isothermal heat coming out of the compressor would have to be rejected via radiators.  And you'd need a microgravity-tolerant liquid-gas separator downstream of the J-T valves for both systems.

3) When the dewar is full, open the F/D valves to allow the liquid to be pumped back into the mains.

4) Lather, rinse, repeat, for as long as you need to.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/10/2022 08:22 pm
A lot of this could be chalked up to sloppy wording, but it also seems like this might have been a paper that sat on the shelf for a while, waiting for IAC.  As a result, it might be a bit stale.
I wondered how old it was, since it has a 2020 copyright date. However, it makes reference to events in 2022, so maybe you're right. Someone might have written it two years ago and then massively edited it just recently to try to bring it up to date.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/10/2022 08:26 pm
If they do an engine bay cooler it's hard to say where the solar panels would be. Deploying from the engine bay puts them near the primary draw, the cooler. It also puts them far from the batteries, up above the methane tank. That would call for substantial cabling. OTOH, putting the PV in the nose, nearer the batteries, reduces that cable run but the run from the batteries to the cooler would still be substantial. Much hinges on what the power draw and operating cycle of the cooler might be. IF they can launch with three engines and IF the cooler goes into the engine bay it might be a reason for relocating the batteries and solar panels to the engine bay for this low production rate variant.
I like the idea of removing the vacuum engines and using that space for the cryocooler. (If that's really feasible.) In the depo, why do the batteries need to be in the nose? What's up there for them to power? Why not put the solar panels, cryocooler, and batteries all down at the bottom?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/11/2022 05:49 pm
If they do an engine bay cooler it's hard to say where the solar panels would be. Deploying from the engine bay puts them near the primary draw, the cooler. It also puts them far from the batteries, up above the methane tank. That would call for substantial cabling. OTOH, putting the PV in the nose, nearer the batteries, reduces that cable run but the run from the batteries to the cooler would still be substantial. Much hinges on what the power draw and operating cycle of the cooler might be. IF they can launch with three engines and IF the cooler goes into the engine bay it might be a reason for relocating the batteries and solar panels to the engine bay for this low production rate variant.
I like the idea of removing the vacuum engines and using that space for the cryocooler. (If that's really feasible.) In the depo, why do the batteries need to be in the nose? What's up there for them to power? Why not put the solar panels, cryocooler, and batteries all down at the bottom?
My memories of the battery location discussion are that they're up top for bellyflop balance, thermal and convenience. In a general purpose design the top dome battery location is also convenient to cargo, cargo handling and crew modules.


With no landing the bellyflop balance is irrelevant. A depot will have no crew or cargo except what's in the tanks. With no Vacs there's a lot of room up high near where the thrust dome meets the rings. This might be enough to keep them from the heat. Insulation is always an option if necessary.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/11/2022 06:27 pm
A couple of misconceptions. Stainless steel has lousy thermal conductivity for a metal. And there is no insulation between LOX and LCH4. Just that poor thermally conductive stainless steel common dome.

Not only that, but the bulk of the cooling effects would be lost to solar or albedo radiation impinging on the LCH4 tank sides.  The tank walls would be back in equilibrium before  the cooling effects made it to the LOX tank.

Quote
If there are separate tank fill and drain connections, little extra plumbing is needed for a cooler. If the cryo cooler is part of a free flyer there would be taps in the adapter and external lines to the cooler. If they go with the dedicated tanker, as it looks like they're doing, it gets more complex, which is one reason for my question about the tanker making orbit with only three engines.

I've been assuming that the process of tying the two ullage gas pre-press systems together is a matter of opening valves that already exist.  If that's the case, then tapping off the pre-press lines in the adapter is simple.  Then you have the following cycle:

1) As boiloff pressure builds up, rather than venting to space, you open the pre-press valves and gas flows into the crycooler.

2) Cryocoolers (one for each of GOX and GCH4) re-liquify the gas and store the liquid in a dewar.  I'd think the temperature drop across the system is small enough that a single-stage Joule-Thomson cooler would work.  All the isothermal heat coming out of the compressor would have to be rejected via radiators.  And you'd need a microgravity-tolerant liquid-gas separator downstream of the J-T valves for both systems.

3) When the dewar is full, open the F/D valves to allow the liquid to be pumped back into the mains.

4) Lather, rinse, repeat, for as long as you need to.
Hmmm. I've been assuming a constant liquid circulation but keep running into niggling problems that cooling the gas gets around. I think I like it.


Problem 1). Micro thrust is needed to feed propellant to the cooler.
Problem 2). Both fill and drain orifices would be at the bottom of the tank. Thinking it through there is probably only one orifice and GSE valving appropriate to the use. So a new stand pipe in each tank to return freshly cooled propellant to the tank. Hopefully this drizzle onto the warmer propellant would mix things up enough that the cooler propellant would not just sink to the bottom of the tank only to cycle through the cooler prematurely.


Can we do away with the dewar and immediately reinject through the fill line? Naw, extra plumbing. Batch pump it back up the ullage pressure line. Wonder how big a dewar would be needed.


A possible way to simplify. Run a LOX cooler only. Run methane boiloff through a heat exchanger and keep cooling the same LOX until the methane ullage pressure is down. Then work on the LOX ullage pressure. It could be sized to do both at the same time.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/11/2022 06:44 pm
Problem 1). Micro thrust is needed to feed propellant to the cooler.

Obviously not a problem if you're cooling the gas.  And J-T expansion valves are cheap and easy.  But, as I said, you need some kind of microgravity vapor-liquid separator downstream of the valve.

Quote
Problem 2). Both fill and drain orifices would be at the bottom of the tank. Thinking it through there is probably only one orifice and GSE valving appropriate to the use. So a new stand pipe in each tank to return freshly cooled propellant to the tank. Hopefully this drizzle onto the warmer propellant would mix things up enough that the cooler propellant would not just sink to the bottom of the tank only to cycle through the cooler prematurely.

Can we do away with the dewar and immediately reinject through the fill line? Naw, extra plumbing. Batch pump it back up the ullage pressure line. Wonder how big a dewar would be needed.

If you keep the dewar, then you can do an ullage burn and use all of the existing F/D architecture to drain it when it gets full.

Quote
A possible way to simplify. Run a LOX cooler only. Run methane boiloff through a heat exchanger and keep cooling the same LOX until the methane ullage pressure is down. Then work on the LOX ullage pressure. It could be sized to do both at the same time.

That's not bad. However, you could probably run both the GOX and GCH4 through compressors geared off a common motor, and through a common heat exchanger to remove the adiabatic heat, upstream of their respective J-T valves.  But it would save you the mass of the extra dewar and plumbing.  I doubt that this is a case where mass is at a huge premium--but volume might be, depending on how it's deployed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/11/2022 07:04 pm
It seems to me that the simplest thing would be something like

1) Wait for pressure to reach a certain point in either tank
2) Start an ullage burn.
3) Wait long enough to expect the gas to separate from the liquid
4) Suck in the GOX and GCH4  and liquify them (as TheRadicalModerate describes above)
5) Pump the liquids through the corresponding fill lines
6) When the pressure gets low enough, stop.

Obviously you'd need a bit more than this (e.g. stop on either tank if you start sucking liquid) but it seems simple enough.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/11/2022 08:22 pm
It seems to me that the simplest thing would be something like

1) Wait for pressure to reach a certain point in either tank
2) Start an ullage burn.
3) Wait long enough to expect the gas to separate from the liquid
4) Suck in the GOX and GCH4  and liquify them (as TheRadicalModerate describes above)
5) Pump the liquids through the corresponding fill lines
6) When the pressure gets low enough, stop.

Obviously you'd need a bit more than this (e.g. stop on either tank if you start sucking liquid) but it seems simple enough.

You need enough liquid that the losses from running it through warm pipes don't overwhelm the system.  To that extent, biggish storage dewars are your friends.

PS:  J-T coolers recirculate the cooled gas that doesn't liquify to cool the incoming gas from the compressors.  As such, it takes a while to liquify a decent amount of gas.  You probably don't want an ullage burn that long.

I think something similar to a cryo propellant management device could probably separate the liquid from the gas, though.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/13/2022 08:32 pm
It seems to me that the simplest thing would be something like

1) Wait for pressure to reach a certain point in either tank
2) Start an ullage burn.
3) Wait long enough to expect the gas to separate from the liquid
4) Suck in the GOX and GCH4  and liquify them (as TheRadicalModerate describes above)
5) Pump the liquids through the corresponding fill lines
6) When the pressure gets low enough, stop.

Obviously you'd need a bit more than this (e.g. stop on either tank if you start sucking liquid) but it seems simple enough.

You need enough liquid that the losses from running it through warm pipes don't overwhelm the system.  To that extent, biggish storage dewars are your friends.

PS:  J-T coolers recirculate the cooled gas that doesn't liquify to cool the incoming gas from the compressors.  As such, it takes a while to liquify a decent amount of gas.  You probably don't want an ullage burn that long.

I think something similar to a cryo propellant management device could probably separate the liquid from the gas, though.
Not really sure what a cryo propellant management device is. Maybe a centrifugal separator? Liquid would be flung out to be directed to the dewar, gas would pass through to be directed to the cooler. Perhaps put the separator at the top, tank end, of the ullage pipe and liquid returned directly to the tank. When actually using the pipe for ullage control the centrifugal separator would be locked and present an impedance load, but not a blockage.


I think this could work but I'm not sure it's the best answer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/13/2022 08:55 pm
Not really sure what a cryo propellant management device is. Maybe a centrifugal separator? Liquid would be flung out to be directed to the dewar, gas would pass through to be directed to the cooler. Perhaps put the separator at the top, tank end, of the ullage pipe and liquid returned directly to the tank. When actually using the pipe for ullage control the centrifugal separator would be locked and present an impedance load, but not a blockage.

I think this could work but I'm not sure it's the best answer.

Cryo PMDs are kinda low-TRL, but the storable PMDs mostly work using tricks with surface tension.

It's not quite the same application, though.  Most PMDs exist to make sure that a blob of prop is always covering the inlet to your pressure-fed thrusters, so you don't blow ullage gas through them.  Here, you'd be trying to aggregate tiny droplets into blobs.

You don't want to keep your transfer plumbing constantly chilled-down.  I think (but can't prove) that doing batch transfers is more efficient, because a single chill-down, which admittedly wastes some of the liquids you just created, wastes less than some continuously-chilled dedicated plumbing.  Remember, all the existing plumbing is designed to transfer tonnes of prop, and flashing a few kg to gas at the start is no big deal.  But if you're periodically trickling liquid from the cryocooler, a large percentage of what you just created will be lost.  By transferring a few hundred kg at a time, you should minimize those losses.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/13/2022 09:38 pm
You don't want to keep your transfer plumbing constantly chilled-down.  I think (but can't prove) that doing batch transfers is more efficient, because a single chill-down, which admittedly wastes some of the liquids you just created, wastes less than some continuously-chilled dedicated plumbing.
But what temperature is the plumbing going to be at? With a reflective coating but no cooling, the vehicle exterior will be about 200K (I think), mostly from heating by the Earth itself. The LOX and LCH4 are going to be around 90K (where methane freezes and oxygen boils). Since the plumbing is all connected to the tanks with fairly short pipes, it seems reasonable that it'll be very close to the same temperature as the tanks.

Or is my intuition off-base here?



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/13/2022 09:54 pm
You don't want to keep your transfer plumbing constantly chilled-down.  I think (but can't prove) that doing batch transfers is more efficient, because a single chill-down, which admittedly wastes some of the liquids you just created, wastes less than some continuously-chilled dedicated plumbing.
But what temperature is the plumbing going to be at? With a reflective coating but no cooling, the vehicle exterior will be about 200K (I think), mostly from heating by the Earth itself. The LOX and LCH4 are going to be around 90K (where methane freezes and oxygen boils). Since the plumbing is all connected to the tanks with fairly short pipes, it seems reasonable that it'll be very close to the same temperature as the tanks.

Or is my intuition off-base here?

The QD's are going to be subjected to Earth albedo and reflection, and my guess is that the plumbing inside the skirt will have the same problem.  Wrapping the skirt pipes in MLI will help, be there will still be some losses.  And if the spacer between the QDs has any profile, it may be in direct sun.

There's a thresholding problem here.  Send 100kg through the pipes, and maybe 5kg boil off, leaving 95kg.  But send 1mL through the pipes, and it all boils off--unless the pipes are permanently refrigerated.  So it's a question of how big a batch, and how often.  It'll be hard to model, unless you assume big batches.

Note that this is an argument for a permanently refrigerated holding dewar, which should be much less expensive to refrigerate than the plumbing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/14/2022 03:04 am
The thing about being in orbit is everything is essentially vacuum insulated, or it’s nearly trivial to make it so. That means whether you let it warm up or not, it won’t be a huge problem.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/14/2022 05:42 am
The thing about being in orbit is everything is essentially vacuum insulated, or it’s nearly trivial to make it so. That means whether you let it warm up or not, it won’t be a huge problem.

Eventually I'm going to frame the problem properly.  One more try:

Suppose you transfer 100kg of cryogens once every hour.  During that hour, let's say the plumbing warms up to equilibrium in 10 minutes.  (All numbers derived via rectal extraction.)  Let's further say that 5kg of the cryogens boil off cooling it to the point where boiloff essentially drops to zero during the transfer.  Total loss:  5%.

Suppose you transfer 200kg every 2 hours.  Same 5kg boiloff, over twice as much prop.  Total loss: 2.5%.

Now let's transfer the same 100kg/hr, but do it in six 16.67kg batches, every 10 minutes.  Presumably, 5kg of each batch would still boil off before the plumbing cooled down.  Total loss:  30%.

There's obviously some point where cryogens just kinda ooze into the plumbing, the pumps can't pump it, and it all boils off.  The percentage of boiloff gets worse and worse as you make the batches smaller.  Things get weirder when the plumbing doesn't reach equilibrium between batches, but it's still going to be suboptimal.

Also remember that we're trying to reuse the F/D lines to return liquid prop to the mains.  Those aren't small lines.  They have a lot of surface area.  If you want things to be relatively efficient, you need to move fairly big batches, fairly infrequently.

An alternative is to plumb new lines for the cryocooler to use, make them very thin, very short, and highly insulated.  But that's new plumbing.  We've been trying to avoid that, right?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: volker2020 on 09/14/2022 05:48 am
(Is there a reason the methane tank is on top?)
One good reason is the mass. Methane is much lighter than oxygen.
CH4 main tank: 263.8t
LOX main tank: 981.3t
Putting it up, means that the rockets get heavier, since the center of gravity moves up, away from the engines and you need extra reinforcements, so that the rocket methane thank, now having to carry the weight and inertia of the 980t lox tank does not collapse.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/14/2022 05:51 am
(Is there a reason the methane tank is on top?)
One good reason is the mass. Methane is much lighter than oxygen.
CH4 main tank: 263.8t
LOX main tank: 981.3t
Putting it up, means that the rockets get heavier, since the center of gravity moves up, away from the engines and you need extra reinforcements, so that the rocket methane thank, now having to carry the weight and inertia of the 980t lox tank does not collapse.
It’s also safer in the case that you have a common dome. If the dome fails, the methane will float (and probably freeze) on top of the liquid oxygen instead of mixing like it would if methane was below oxygen.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/14/2022 05:55 am
The thing about being in orbit is everything is essentially vacuum insulated, or it’s nearly trivial to make it so. That means whether you let it warm up or not, it won’t be a huge problem.

Eventually I'm going to frame the problem properly.  One more try:

Suppose you transfer 100kg of cryogens once every hour.  During that hour, let's say the plumbing warms up to equilibrium in 10 minutes.  (All numbers derived via rectal extraction.)  Let's further say that 5kg of the cryogens boil off cooling it to the point where boiloff essentially drops to zero during the transfer.  Total loss:  5%.

Suppose you transfer 200kg every 2 hours.  Same 5kg boiloff, over twice as much prop.  Total loss: 2.5%.

Now let's transfer the same 100kg/hr, but do it in six 16.67kg batches, every 10 minutes.  Presumably, 5kg of each batch would still boil off before the plumbing cooled down.  Total loss:  30%.

There's obviously some point where cryogens just kinda ooze into the plumbing, the pumps can't pump it, and it all boils off.  The percentage of boiloff gets worse and worse as you make the batches smaller.  Things get weirder when the plumbing doesn't reach equilibrium between batches, but it's still going to be suboptimal.

Also remember that we're trying to reuse the F/D lines to return liquid prop to the mains.  Those aren't small lines.  They have a lot of surface area.  If you want things to be relatively efficient, you need to move fairly big batches, fairly infrequently.

An alternative is to plumb new lines for the cryocooler to use, make them very thin, very short, and highly insulated.  But that's new plumbing.  We've been trying to avoid that, right?
Oh I understood your point and don’t disagree with it, although probably the pipes will stay cold a lot longer than 10 minutes. I was just making an aside.

EDIT:
MLI insulates to like 1.45W/m^2. So 1mm thick stainless steel only warms up like 100 degrees after 3 days.

The whole Starship, if 50m long and 8m wide exposed to Earthshine on one side… should only boil off about 87t of liquid oxygen *per year*.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 09/14/2022 07:10 am
(Is there a reason the methane tank is on top?)
One good reason is the mass. Methane is much lighter than oxygen.
CH4 main tank: 263.8t
LOX main tank: 981.3t
Putting it up, means that the rockets get heavier, since the center of gravity moves up, away from the engines and you need extra reinforcements, so that the rocket methane thank, now having to carry the weight and inertia of the 980t lox tank does not collapse.
It’s also safer in the case that you have a common dome. If the dome fails, the methane will float (and probably freeze) on top of the liquid oxygen instead of mixing like it would if methane was below oxygen.
LOX and LCH4 are miscible so it will float on top like alcohol does on water  ;)

Since the mixture is a shock sensitive primary explosive at mass ratios from 16 down past stoichiometric I think the best case scenario is that the initial failure leads to ignition.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/14/2022 12:28 pm
(Is there a reason the methane tank is on top?)
One good reason is the mass. Methane is much lighter than oxygen.
CH4 main tank: 263.8t
LOX main tank: 981.3t
Putting it up, means that the rockets get heavier, since the center of gravity moves up, away from the engines and you need extra reinforcements, so that the rocket methane thank, now having to carry the weight and inertia of the 980t lox tank does not collapse.
It’s also safer in the case that you have a common dome. If the dome fails, the methane will float (and probably freeze) on top of the liquid oxygen instead of mixing like it would if methane was below oxygen.
LOX and LCH4 are miscible so it will float on top like alcohol does on water  ;)

Since the mixture is a shock sensitive primary explosive at mass ratios from 16 down past stoichiometric I think the best case scenario is that the initial failure leads to ignition.
Nope. Subcooled methane and LOx are not miscible because the methane will freeze.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/14/2022 02:16 pm
Nope. Subcooled methane and LOx are not miscible because the methane will freeze.
Is the "because" an argument or an assertion about the chemistry of LOx and methane at the temperature in question?

There are plenty of solids that are soluble in various different liquids, so the conclusion does not follow without additional information.

Or are you simply arguing that the technical term is soluble rather than miscible?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/14/2022 02:24 pm
Nope. Subcooled methane and LOx are not miscible because the methane will freeze.
Is the "because" an argument or an assertion about the chemistry of LOx and methane at the temperature in question?

There are plenty of solids that are soluble in various different liquids, so the conclusion does not follow without additional information.

Or are you simply arguing that the technical term is soluble rather than miscible?
I’m arguing that the rate of mixing or solvating will be dramatically slowed by the fact that methane and LOx have vastly different densities to begin with, especially if the methane is already starting off on top and both are subcooled, leading to methane freezing on the interface, much like kerosene would.

Things can only mix on an interface.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Proponent on 09/14/2022 03:18 pm

First, you'll need a launch site with a lower latitude, or you have to be willing to make the plane change on-orbit.  (A 5º change from 500km x 500km x 28.5º would be 670m/s, which isn't so bad if you only have to do it once.)

Next, if you set up an LEO whose plane is in the ecliptic, it ought to stay in the ecliptic.

What about precession?  By my calculation in a 500kmx500kmx23.5º orbit the RAAN will move completely around the Earth in a bit under 50 days.

Precession is caused by the oblateness of the Earth.  There is no reason for an orbit to remain fixed  w.r.t. the sun and ecliptic.

Don't feel bad, TheRadicalModerate: no less than von Braun himself made the same mistake in his book Project Mars (https://archive.org/details/ProjectMars).  He describes a military space station in an Earth-centric orbit perpendicular to the ecliptic to provide maximum solar energy over the course of a year.  For assembling a Mars fleet, he has another station in the ecliptic plane.

The funny thing is, that von Braun was well aware of precession: he mentions the precession of the moon's orbit.  Apparently it had not occurred to him that his space stations' orbits would precess as well.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: eriblo on 09/14/2022 05:31 pm
Nope. Subcooled methane and LOx are not miscible because the methane will freeze.
Is the "because" an argument or an assertion about the chemistry of LOx and methane at the temperature in question?

There are plenty of solids that are soluble in various different liquids, so the conclusion does not follow without additional information.

Or are you simply arguing that the technical term is soluble rather than miscible?
I’m arguing that the rate of mixing or solvating will be dramatically slowed by the fact that methane and LOx have vastly different densities to begin with, especially if the methane is already starting off on top and both are subcooled, leading to methane freezing on the interface, much like kerosene would.

Things can only mix on an interface.
My scepticism regarding frozen methane came from looking at a methane-oxygen phase diagram and confirming that the (partial) melting point is lower than the weighted average of the melting points (extrapolated below 77 K but most likely always true). This suggests that mixing liquid methane and liquid oxygen will always result in a liquid (as the mixing itself is unlikely to be highly endothermic).

I expect that you can float liquid methane on top of liquid oxygen (just like you can alcohol on water) but I have a hard time imagining the kinematics of the methane freezing. I will concede that the layering might in theory be done gently enough for heat conduction to outpace the combination of diffusion and heat of mixing + fusion, especially under low accelerations, but I imagine the conditions required to be slightly different from a rocket launch + dome failure.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: llanitedave on 09/14/2022 06:45 pm
Nope. Subcooled methane and LOx are not miscible because the methane will freeze.
Is the "because" an argument or an assertion about the chemistry of LOx and methane at the temperature in question?

There are plenty of solids that are soluble in various different liquids, so the conclusion does not follow without additional information.

Or are you simply arguing that the technical term is soluble rather than miscible?
I’m arguing that the rate of mixing or solvating will be dramatically slowed by the fact that methane and LOx have vastly different densities to begin with, especially if the methane is already starting off on top and both are subcooled, leading to methane freezing on the interface, much like kerosene would.

Things can only mix on an interface.
My scepticism regarding frozen methane came from looking at a methane-oxygen phase diagram and confirming that the (partial) melting point is lower than the weighted average of the melting points (extrapolated below 77 K but most likely always true). This suggests that mixing liquid methane and liquid oxygen will always result in a liquid (as the mixing itself is unlikely to be highly endothermic).

I expect that you can float liquid methane on top of liquid oxygen (just like you can alcohol on water) but I have a hard time imagining the kinematics of the methane freezing. I will concede that the layering might in theory be done gently enough for heat conduction to outpace the combination of diffusion and heat of mixing + fusion, especially under low accelerations, but I imagine the conditions required to be slightly different from a rocket launch + dome failure.
Even if the methane did freeze, that would not be an instantaneous event.  There would always be some minor volume of mixed methane and LOX, enough to support ignition/detonation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/14/2022 07:19 pm
Seems like we might be off-topic with the whole "will methane freeze if the inter-tank bulkhead fails" discussion.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/14/2022 07:56 pm
Nope. Subcooled methane and LOx are not miscible because the methane will freeze.
Is the "because" an argument or an assertion about the chemistry of LOx and methane at the temperature in question?

There are plenty of solids that are soluble in various different liquids, so the conclusion does not follow without additional information.

Or are you simply arguing that the technical term is soluble rather than miscible?
I’m arguing that the rate of mixing or solvating will be dramatically slowed by the fact that methane and LOx have vastly different densities to begin with, especially if the methane is already starting off on top and both are subcooled, leading to methane freezing on the interface, much like kerosene would.

Things can only mix on an interface.
My scepticism regarding frozen methane came from looking at a methane-oxygen phase diagram and confirming that the (partial) melting point is lower than the weighted average of the melting points (extrapolated below 77 K but most likely always true). This suggests that mixing liquid methane and liquid oxygen will always result in a liquid (as the mixing itself is unlikely to be highly endothermic).

I expect that you can float liquid methane on top of liquid oxygen (just like you can alcohol on water) but I have a hard time imagining the kinematics of the methane freezing. I will concede that the layering might in theory be done gently enough for heat conduction to outpace the combination of diffusion and heat of mixing + fusion, especially under low accelerations, but I imagine the conditions required to be slightly different from a rocket launch + dome failure.
Even if the methane did freeze, that would not be an instantaneous event.  There would always be some minor volume of mixed methane and LOX, enough to support ignition/detonation.
The concern(-troll?) has been that all the methane and oxygen would fully mix into a highly explosive mixture, so you’d get an absolutely enormous detonation if it lit off, not just a fireball.

However that’s extremely unlikely for the reasons I mentioned. Both solid an liquid methane are far less dense than liquid oxygen (by more than a 2:1 ratio, much greater than oil and water), so the two would not mix well if you removed a barrier separating them. The buoyancy of methane would mean you’d only get some small amount of mixing at the interface. And where there WAS an interface, the cold liquid oxygen would freeze the methane to a solid, which would sink to the bottom of the methane and float above the oxygen, forming an additional barrier to mixing of the two liquids.

It wouldn’t be a *safe* by any stretch, but it also isn’t going to be a Hiroshima-level detonation (which is what the naive calculation of them being fully mixed would say). Probably the amount fully mixed would be on the order of 1% or so, at least for a while. Especially in the case of subcooled propellants (which ensure the fuel would be frozen when it touches the LOx), this isn’t much different than kerolox.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/14/2022 07:57 pm
The thing about being in orbit is everything is essentially vacuum insulated, or it’s nearly trivial to make it so. That means whether you let it warm up or not, it won’t be a huge problem.

Eventually I'm going to frame the problem properly.  One more try:

Suppose you transfer 100kg of cryogens once every hour.  During that hour, let's say the plumbing warms up to equilibrium in 10 minutes.  (All numbers derived via rectal extraction.)  Let's further say that 5kg of the cryogens boil off cooling it to the point where boiloff essentially drops to zero during the transfer.  Total loss:  5%.

Suppose you transfer 200kg every 2 hours.  Same 5kg boiloff, over twice as much prop.  Total loss: 2.5%.

Now let's transfer the same 100kg/hr, but do it in six 16.67kg batches, every 10 minutes.  Presumably, 5kg of each batch would still boil off before the plumbing cooled down.  Total loss:  30%.

There's obviously some point where cryogens just kinda ooze into the plumbing, the pumps can't pump it, and it all boils off.  The percentage of boiloff gets worse and worse as you make the batches smaller.  Things get weirder when the plumbing doesn't reach equilibrium between batches, but it's still going to be suboptimal.

Also remember that we're trying to reuse the F/D lines to return liquid prop to the mains.  Those aren't small lines.  They have a lot of surface area.  If you want things to be relatively efficient, you need to move fairly big batches, fairly infrequently.

An alternative is to plumb new lines for the cryocooler to use, make them very thin, very short, and highly insulated.  But that's new plumbing.  We've been trying to avoid that, right?
No new plumbing is a goal, not the 11th commandment. Still it would be good to avoid it if possible.


A thought. When the depot is not actively engaged in transfer ops, how about the old standby of a rotisserie spin? It would call for conformal PV and radiators, which in turn places demands on orientation.


Thermal white works well against solar radiation but not so well for earth input, so maybe keep the nose earthward and the walls, PV and radiators sunward as much as possible. Tricky for a spinning object in orbit. The radiator would need to broken up into zones with only the most advantageous zone being active. No cooling would be available when the sun transits, so cooling would have to shut down then run off of battery and/or PV when the alignment allows. Tricky ops.


I haven't thought this through all the way but it might allow the cooler to work directly with liquid which (I think) would be more thermally efficient and would simplify pumping. Separating gas and liquid also becomes a non problem.


What appears as a downside actually becomes an upside. The propellant would be directly against the walls where the heat is coming from, BUT convection would keep the propellant well mixed and avoid thermal stratification. No worries about the cooler repeatedly cooling the same bit of propellant if working directly with liquid.


A real downside is losses for spinup and down to allow transfer ops. Another is some simple plumbing additions. There's probably more. As I said, I haven't thought this through all the way.


IIRC, one thing we don't have is a realistic understanding of how much thermal input to expect from earth (or did I miss it?). PV, battery and radiator capacity all hinge on this - for all the ideas we've been fielding.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/14/2022 08:18 pm
Nope. Subcooled methane and LOx are not miscible because the methane will freeze.
Is the "because" an argument or an assertion about the chemistry of LOx and methane at the temperature in question?

There are plenty of solids that are soluble in various different liquids, so the conclusion does not follow without additional information.

Or are you simply arguing that the technical term is soluble rather than miscible?
I’m arguing that the rate of mixing or solvating will be dramatically slowed by the fact that methane and LOx have vastly different densities to begin with, especially if the methane is already starting off on top and both are subcooled, leading to methane freezing on the interface, much like kerosene would.

Things can only mix on an interface.
My scepticism regarding frozen methane came from looking at a methane-oxygen phase diagram and confirming that the (partial) melting point is lower than the weighted average of the melting points (extrapolated below 77 K but most likely always true). This suggests that mixing liquid methane and liquid oxygen will always result in a liquid (as the mixing itself is unlikely to be highly endothermic).

I expect that you can float liquid methane on top of liquid oxygen (just like you can alcohol on water) but I have a hard time imagining the kinematics of the methane freezing. I will concede that the layering might in theory be done gently enough for heat conduction to outpace the combination of diffusion and heat of mixing + fusion, especially under low accelerations, but I imagine the conditions required to be slightly different from a rocket launch + dome failure.
Even if the methane did freeze, that would not be an instantaneous event.  There would always be some minor volume of mixed methane and LOX, enough to support ignition/detonation.
The concern(-troll?) has been that all the methane and oxygen would fully mix into a highly explosive mixture, so you’d get an absolutely enormous detonation if it lit off, not just a fireball.

However that’s extremely unlikely for the reasons I mentioned. Both solid an liquid methane are far less dense than liquid oxygen (by more than a 2:1 ratio, much greater than oil and water), so the two would not mix well if you removed a barrier separating them. The buoyancy of methane would mean you’d only get some small amount of mixing at the interface. And where there WAS an interface, the cold liquid oxygen would freeze the methane to a solid, which would sink to the bottom of the methane and float above the oxygen, forming an additional barrier to mixing of the two liquids.

It wouldn’t be a *safe* by any stretch, but it also isn’t going to be a Hiroshima-level detonation (which is what the naive calculation of them being fully mixed would say). Probably the amount fully mixed would be on the order of 1% or so, at least for a while. Especially in the case of subcooled propellants (which ensure the fuel would be frozen when it touches the LOx), this isn’t much different than kerolox.
Don't want the concern troll label but somehow I see nothing good from a common dome collapse. In the real world I would expect lot of splashing leading to a chaotic mix of thermal interaction. Probably no Hiroshima, but a starship sized somewhat fizzled fuel-air munition would really screw up your day.


There has to be an easier way to transfer props. (That gets it back on topic).  ::)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/14/2022 08:23 pm
The thing about being in orbit is everything is essentially vacuum insulated, or it’s nearly trivial to make it so. That means whether you let it warm up or not, it won’t be a huge problem.

Eventually I'm going to frame the problem properly.  One more try:

Suppose you transfer 100kg of cryogens once every hour.  During that hour, let's say the plumbing warms up to equilibrium in 10 minutes.  (All numbers derived via rectal extraction.)  Let's further say that 5kg of the cryogens boil off cooling it to the point where boiloff essentially drops to zero during the transfer.  Total loss:  5%.

Suppose you transfer 200kg every 2 hours.  Same 5kg boiloff, over twice as much prop.  Total loss: 2.5%.

Now let's transfer the same 100kg/hr, but do it in six 16.67kg batches, every 10 minutes.  Presumably, 5kg of each batch would still boil off before the plumbing cooled down.  Total loss:  30%.

There's obviously some point where cryogens just kinda ooze into the plumbing, the pumps can't pump it, and it all boils off.  The percentage of boiloff gets worse and worse as you make the batches smaller.  Things get weirder when the plumbing doesn't reach equilibrium between batches, but it's still going to be suboptimal.

Also remember that we're trying to reuse the F/D lines to return liquid prop to the mains.  Those aren't small lines.  They have a lot of surface area.  If you want things to be relatively efficient, you need to move fairly big batches, fairly infrequently.

An alternative is to plumb new lines for the cryocooler to use, make them very thin, very short, and highly insulated.  But that's new plumbing.  We've been trying to avoid that, right?
Oh I understood your point and don’t disagree with it, although probably the pipes will stay cold a lot longer than 10 minutes. I was just making an aside.

EDIT:
MLI insulates to like 1.45W/m^2. So 1mm thick stainless steel only warms up like 100 degrees after 3 days.

The whole Starship, if 50m long and 8m wide exposed to Earthshine on one side… should only boil off about 87t of liquid oxygen *per year*.
I'm unclear. Is this assuming the ship is wrapped in MLI, naked stainless or Solar white? That's much lower than I would gut guess.


BTW, 8m diameter? A typo?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/14/2022 09:48 pm
The whole Starship, if 50m long and 8m wide exposed to Earthshine on one side… should only boil off about 87t of liquid oxygen *per year*.
I'm unclear. Is this assuming the ship is wrapped in MLI, naked stainless or Solar white? That's much lower than I would gut guess.

I would bet heavily on either all-passive cooling or all-active cooling, but not a combination of both.

A depot wrapped completely in MLI seems... hard...  But you could certainly insulate any plumbing that got a substantial thermal load.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/14/2022 10:04 pm
I've been wondering whether, over time, the methane would freeze and/or the oxygen would boil as it all comes to equilibrium. I almost wonder whether it would make sense to have a thermoelectric (i.e. sold-state) cooler between them, keeping the oxygen side maybe 40 K cooler than the methane side. That's essentially cooling the oxygen and heating the methane (so not actually shedding any of the absorbed heat from outside), but maybe that's okay, given that there's a more traditional cooler down at the bottom.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/14/2022 11:57 pm
I've been wondering whether, over time, the methane would freeze and/or the oxygen would boil as it all comes to equilibrium. I almost wonder whether it would make sense to have a thermoelectric (i.e. sold-state) cooler between them, keeping the oxygen side maybe 40 K cooler than the methane side. That's essentially cooling the oxygen and heating the methane (so not actually shedding any of the absorbed heat from outside), but maybe that's okay, given that there's a more traditional cooler down at the bottom.

In microgravity the liquids aren’t necessarily in conductive contact.   Probably not very efficient.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/15/2022 10:05 pm
The whole Starship, if 50m long and 8m wide exposed to Earthshine on one side… should only boil off about 87t of liquid oxygen *per year*.
I'm unclear. Is this assuming the ship is wrapped in MLI, naked stainless or Solar white? That's much lower than I would gut guess.

I would bet heavily on either all-passive cooling or all-active cooling, but not a combination of both.

A depot wrapped completely in MLI seems... hard...  But you could certainly insulate any plumbing that got a substantial thermal load.
Why no combo?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: spacenut on 09/15/2022 11:16 pm
If the depot were insulated with some type of foam like the shuttle tank or SLS tank, a smaller active cooling system would be needed.  Thus, lowering launch mass.  In the place of fins, solar panels could be installed for active cooling power.  So you can have a combo. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/15/2022 11:31 pm
You can do MLI underneath metal skin. The reason Shuttle used foam is because MLI at sea level pressure doesn’t work as well as foam does at sea level pressure.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: F3ar13ss on 09/16/2022 12:12 am
Musk is right; the depot idea does suck a bit. Unless you either; 1) give it hardware to change its orbit when needed (or attach a Booster to it from time to time), or 2) have a simply incredible number of those things stationed in all sorts of orbits around Earth and Moon.

The thing is, launching anything is so cheap now, that any combination of those solutions isn't really a big deal except for the time it takes to fully stage and maintain.

My idea is to make a tank only vehicle that stores something like 3-5 ships of fuel. And launch it into orbit like a chunky center stage of a falcon heavy. You could use two, three, four, whatever number of Falcon 9 boosters needed to get it up there. You could scatter these things everywhere up there.

You could even put them at Lagrange points. You put them wherever.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/16/2022 12:32 am
When did Elon say the depot sucks?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 09/16/2022 01:34 am
Musk is right; the depot idea does suck a bit. Unless you either; 1) give it hardware to change its orbit when needed (or attach a Booster to it from time to time), or 2) have a simply incredible number of those things stationed in all sorts of orbits around Earth and Moon.

The thing is, launching anything is so cheap now, that any combination of those solutions isn't really a big deal except for the time it takes to fully stage and maintain.

My idea is to make a tank only vehicle that stores something like 3-5 ships of fuel. And launch it into orbit like a chunky center stage of a falcon heavy. You could use two, three, four, whatever number of Falcon 9 boosters needed to get it up there. You could scatter these things everywhere up there.

You could even put them at Lagrange points. You put them wherever.
Raptor engines are (will be) cheap. Just launch your big tank atop an SH. That's what a Depot is, maybe. If Raptiors were more expensive, you could go the trouble to recover them and return them to Earth, but at 6x $250,000 it's just not worth it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/16/2022 02:13 am
Musk is right; the depot idea does suck a bit. Unless you either; 1) give it hardware to change its orbit when needed (or attach a Booster to it from time to time), or 2) have a simply incredible number of those things stationed in all sorts of orbits around Earth and Moon.

The thing is, launching anything is so cheap now, that any combination of those solutions isn't really a big deal except for the time it takes to fully stage and maintain.

My idea is to make a tank only vehicle that stores something like 3-5 ships of fuel. And launch it into orbit like a chunky center stage of a falcon heavy. You could use two, three, four, whatever number of Falcon 9 boosters needed to get it up there. You could scatter these things everywhere up there.

You could even put them at Lagrange points. You put them wherever.
Raptor engines are (will be) cheap. Just launch your big tank atop an SH. That's what a Depot is, maybe. If Raptiors were more expensive, you could go the trouble to recover them and return them to Earth, but at 6x $250,000 it's just not worth it.
Plus it gives you the option to use the depot as a HUGE kick stage.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/16/2022 03:40 am
You can do MLI underneath metal skin. The reason Shuttle used foam is because MLI at sea level pressure doesn’t work as well as foam does at sea level pressure.

Are you thinking of adding a second, outer skin to the Starship?  If so, how does this survive mechanically?  How much dry mass does it add?  And how are you going to pump out the air efficiently?  What happens to the outer mould line?

Or are you thinking of putting MLI inside the tank?  If so, how do you keep the prop from oozing in between the layers and ruining the insulating effects?  How do you keep slosh from tearing the insulation to pieces?  And does kapton survive mechanically at cryogenic temperatures?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/16/2022 04:17 am
You can do MLI underneath metal skin. The reason Shuttle used foam is because MLI at sea level pressure doesn’t work as well as foam does at sea level pressure.

Are you thinking of adding a second, outer skin to the Starship?  If so, how does this survive mechanically?  How much dry mass does it add?  And how are you going to pump out the air efficiently?  What happens to the outer mould line?

Or are you thinking of putting MLI inside the tank?  If so, how do you keep the prop from oozing in between the layers and ruining the insulating effects?  How do you keep slosh from tearing the insulation to pieces?  And does kapton survive mechanically at cryogenic temperatures?

Just think of, like, any dewar or vacuum-insulated lines for cryogenic filling. Same thing, except no need to pump it all the way to vacuum (just purge with nitrogen) until it's in orbit, so it will be much lower dry mass than a comparable dewar on Earth. And yeah, stuff like kapton is fine.

MLI is pretty standard in dewars for liquid nitrogen.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/16/2022 05:14 am
Are you thinking of adding a second, outer skin to the Starship? ... how does this survive mechanically?

A standard Nomex blanket covering seems to do fine on ascent. See: shuttle.

how are you going to pump out the air efficiently?

Perforated MLI is a standard product for precisely this reason. Nothing special or unusual about it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/16/2022 06:51 am
You can do MLI underneath metal skin. The reason Shuttle used foam is because MLI at sea level pressure doesn’t work as well as foam does at sea level pressure.

Are you thinking of adding a second, outer skin to the Starship?  If so, how does this survive mechanically?  How much dry mass does it add?  And how are you going to pump out the air efficiently?  What happens to the outer mould line?

Or are you thinking of putting MLI inside the tank?  If so, how do you keep the prop from oozing in between the layers and ruining the insulating effects?  How do you keep slosh from tearing the insulation to pieces?  And does kapton survive mechanically at cryogenic temperatures?

Just think of, like, any dewar or vacuum-insulated lines for cryogenic filling. Same thing, except no need to pump it all the way to vacuum (just purge with nitrogen) until it's in orbit, so it will be much lower dry mass than a comparable dewar on Earth. And yeah, stuff like kapton is fine.

MLI is pretty standard in dewars for liquid nitrogen.

Sounds heavy--and mechanically problematic--to me.  You're gonna launch a Starship with effectively double-walled tanks?

how are you going to pump out the air efficiently?

Perforated MLI is a standard product for precisely this reason. Nothing special or unusual about it.

That won't work very well sandwiched between two pieces of sheet metal.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/16/2022 07:11 am
You can do MLI underneath metal skin. The reason Shuttle used foam is because MLI at sea level pressure doesn’t work as well as foam does at sea level pressure.

Are you thinking of adding a second, outer skin to the Starship?  If so, how does this survive mechanically?  How much dry mass does it add?  And how are you going to pump out the air efficiently?  What happens to the outer mould line?

Or are you thinking of putting MLI inside the tank?  If so, how do you keep the prop from oozing in between the layers and ruining the insulating effects?  How do you keep slosh from tearing the insulation to pieces?  And does kapton survive mechanically at cryogenic temperatures?

Just think of, like, any dewar or vacuum-insulated lines for cryogenic filling. Same thing, except no need to pump it all the way to vacuum (just purge with nitrogen) until it's in orbit, so it will be much lower dry mass than a comparable dewar on Earth. And yeah, stuff like kapton is fine.

MLI is pretty standard in dewars for liquid nitrogen.

Sounds heavy--and mechanically problematic--to me.  You're gonna launch a Starship with effectively double-walled tanks?


Yes? I mean, the outer one is essentially just a fairing. If you want it to actually store well, that’s a good idea.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/16/2022 07:13 am
how are you going to pump out the air efficiently?

Perforated MLI is a standard product for precisely this reason. Nothing special or unusual about it.

That won't work very well sandwiched between two pieces of sheet metal.

Hence my remark about Nomex blankets, which you weirdly omitted from your quotation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/16/2022 07:14 am
You can do MLI underneath metal skin. The reason Shuttle used foam is because MLI at sea level pressure doesn’t work as well as foam does at sea level pressure.

Are you thinking of adding a second, outer skin to the Starship?  If so, how does this survive mechanically?  How much dry mass does it add?  And how are you going to pump out the air efficiently?  What happens to the outer mould line?

Or are you thinking of putting MLI inside the tank?  If so, how do you keep the prop from oozing in between the layers and ruining the insulating effects?  How do you keep slosh from tearing the insulation to pieces?  And does kapton survive mechanically at cryogenic temperatures?

Just think of, like, any dewar or vacuum-insulated lines for cryogenic filling. Same thing, except no need to pump it all the way to vacuum (just purge with nitrogen) until it's in orbit, so it will be much lower dry mass than a comparable dewar on Earth. And yeah, stuff like kapton is fine.

MLI is pretty standard in dewars for liquid nitrogen.

Sounds heavy--and mechanically problematic--to me.  You're gonna launch a Starship with effectively double-walled tanks?


Yes? I mean, the outer one is essentially just a fairing. If you want it to actually store well and not have foam that popcorns off, that’s a good idea.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: robot_enthusiast on 09/16/2022 08:57 am
Musk is right; the depot idea does suck a bit. Unless you either; 1) give it hardware to change its orbit when needed
...
The depot needs to have engines anyway to get into orbit, there would be nothing stopping it from using stored propellants to adjust its orbit as necessary. Launching a few tankers worth of prop to move it will probably be cheaper than building a new depot.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/16/2022 03:55 pm
You can do MLI underneath metal skin. The reason Shuttle used foam is because MLI at sea level pressure doesn’t work as well as foam does at sea level pressure.

Are you thinking of adding a second, outer skin to the Starship?  If so, how does this survive mechanically?  How much dry mass does it add?  And how are you going to pump out the air efficiently?  What happens to the outer mould line?

Or are you thinking of putting MLI inside the tank?  If so, how do you keep the prop from oozing in between the layers and ruining the insulating effects?  How do you keep slosh from tearing the insulation to pieces?  And does kapton survive mechanically at cryogenic temperatures?

Just think of, like, any dewar or vacuum-insulated lines for cryogenic filling. Same thing, except no need to pump it all the way to vacuum (just purge with nitrogen) until it's in orbit, so it will be much lower dry mass than a comparable dewar on Earth. And yeah, stuff like kapton is fine.

MLI is pretty standard in dewars for liquid nitrogen.

Sounds heavy--and mechanically problematic--to me.  You're gonna launch a Starship with effectively double-walled tanks?


Yes? I mean, the outer one is essentially just a fairing. If you want it to actually store well, that’s a good idea.
Sounds good in theory but the execution sounds expensive. Real expensive.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/16/2022 04:02 pm
Musk is right; the depot idea does suck a bit. Unless you either; 1) give it hardware to change its orbit when needed (or attach a Booster to it from time to time), or 2) have a simply incredible number of those things stationed in all sorts of orbits around Earth and Moon.
There is another option (https://selenianboondocks.com/2018/02/aas-paper-review-practical-methodologies-for-low-delta-v-penalty-on-time-departures-to-arbitrary-interplanetary-destinations-from-a-medium-inclination-low-earth-orbit-depot/). (Start reading at "The Show Stopper.") The three-burn methodology they propose here looks really promising. It does add about ten days to any mission, plus two trips through the Van Allen belts, but the extra fuel cost is tiny, and it lets you launch from a depot regardless of RAAN. But it'll only matter when someone wants to use a depot to go somewhere beyond the moon. Still, I think it fixes the biggest problem with depots, eliminating the need to make big orbit changes or to have multiple depots.

My idea is to make a tank only vehicle that stores something like 3-5 ships of fuel. And launch it into orbit like a chunky center stage of a falcon heavy. You could use two, three, four, whatever number of Falcon 9 boosters needed to get it up there. You could scatter these things everywhere up there.

You could even put them at Lagrange points. You put them wherever.
Are you thinking of adding Falcon 9 boosters to the side of a Starship booster to let the Starship reach orbit with more fuel in it? Or are you thinking of a totally new design? If the former, these things aren't Legos; you can't just plug them together any old way and expect them to work. If the latter, that would take years and years to develop. Regardless, assuming the three-burn system really works, there's no need for it that I can see.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/16/2022 05:49 pm
The three-burn methodology they propose here looks really promising. It does add about ten days to any mission, plus two trips through the Van Allen belts, but the extra fuel cost is tiny, and it lets you launch from a depot regardless of RAAN. But it'll only matter when someone wants to use a depot to go somewhere beyond the moon. Still, I think it fixes the biggest problem with depots, eliminating the need to make big orbit changes or to have multiple depots.

As you say, this works great for interplanetary but more-or-less doubles transit time to the Moon, while significantly increasing VA belt radiation exposure.  And I'd say that the tolerable radiation exposure for a lunar passenger is probably commensurate with the risk that passenger is willing to assume:  Somebody going to Mars is assuming a bigger risk, and the extra risk from radiation can be increased, while a mere lunar traveller probably has somewhat lower risk appetite.  Astronauts going to the Moon will likely expect to perform several lunar missions over the course of their career, but it will be extremely rare for somebody to make more than one Mars trip.  Minimizing radiation exposure to stay under lifetime limits is therefore considerably more important.

One thing I'm not clear on is how sensitive lunar departures are to inclination, which is hard to change with a depot at a fixed inclination.

The obvious way to eliminate all targeting problems is to bring the depot down between missions, and re-launch it to the proper target orbit for the next mission.  This is why I'm quite enamored of the idea of a depot being just another lift tanker, outfitted with enough deployable payload to do cryocooling, berthing/docking, and whatever QD adaptation is required.  But that obviously requires a viable design for all the extra stuff.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/16/2022 06:38 pm
Or if the depot has any aerodynamic lifting capacity, the depot can do a lifting aerodynamic plane change, which uses about half the delta-v as a purely propulsive delta-v.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bryan Hayward on 09/16/2022 06:43 pm
One thing I'm not clear on is how sensitive lunar departures are to inclination, which is hard to change with a depot at a fixed inclination.

According to the AAS link, lunar missions don't have as much of a plane change problem, allowing a lot of launch opportunities with a stable depot orbit.

"[the author] should probably also note that this problem is not a problem for LEO depot missions to the Moon. Since the moon orbits around the earth approximately every 28ish days, and orbits in a direction opposite to the direction of LEO orbit nodal precession, it means you get coplanar lunar departure opportunities for most LEO orbits approximately every 7-9 days. And if we ever get to the point where we’re flying to the Moon so often that that isn’t frequent enough, we’ve basically already won, haven’t we?"
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 09/16/2022 08:12 pm
So have the depot parts launch for every mission.

Fuel depot assembled to design.
1. quick and dirty just a tanker
2. Insulated and spin stabilized so fuel stays in correct place.
3. We could actually get spin stabilized waiting room assembled for each mission.

If the parts can be assembled quickly then launch them when you know the desired RAAN.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/16/2022 08:16 pm
One thing I'm not clear on is how sensitive lunar departures are to inclination, which is hard to change with a depot at a fixed inclination.

According to the AAS link, lunar missions don't have as much of a plane change problem, allowing a lot of launch opportunities with a stable depot orbit.

"[the author] should probably also note that this problem is not a problem for LEO depot missions to the Moon. Since the moon orbits around the earth approximately every 28ish days, and orbits in a direction opposite to the direction of LEO orbit nodal precession, it means you get coplanar lunar departure opportunities for most LEO orbits approximately every 7-9 days. And if we ever get to the point where we’re flying to the Moon so often that that isn’t frequent enough, we’ve basically already won, haven’t we?"

A related question:  Does this change when targeting the lunar poles?  I don't even have a good feel for how delta-v changes when TLI is set up for:

1) TLI for equatorial LLO
2) TLI for polar LLO
3) TLI for fast NRHO insertion
4) TLI for BLT

Presumably, inclination plays a factor in all of these, in addition to raw C3.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/16/2022 08:25 pm
A related question:  Does this change when targeting the lunar poles?  I don't even have a good feel for how delta-v changes when TLI is set up for:

1) TLI for equatorial LLO
2) TLI for polar LLO
3) TLI for fast NRHO insertion
4) TLI for BLT

Presumably, inclination plays a factor in all of these, in addition to raw C3.
Why would the impact of the inclination of the parking orbit around the Earth differ for those different orbits? It's essentially about whether you waste any delta-v from the orbit when you transfer to the lunar trajectory. But once you're on that trajectory (no matter how much or little delta-v it cost you), I can't see how the original inclination would factor in.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/16/2022 08:26 pm
So have the depot parts launch for every mission.

Fuel depot assembled to design.
1. quick and dirty just a tanker
2. Insulated and spin stabilized so fuel stays in correct place.
3. We could actually get spin stabilized waiting room assembled for each mission.

If the parts can be assembled quickly then launch them when you know the desired RAAN.

I've been assuming that the depot's payload is the parts.  See here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404058#msg2404058) and here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404380#msg2404380).

I think spin-stabiized is off the table for a long time.  Whatever the depot turns out to be, it's almost certainly a Starship variant.  You can't dock with it if it's spinning.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 09/16/2022 08:48 pm
So have the depot parts launch for every mission.

Fuel depot assembled to design.
1. quick and dirty just a tanker
2. Insulated and spin stabilized so fuel stays in correct place.
3. We could actually get spin stabilized waiting room assembled for each mission.

If the parts can be assembled quickly then launch them when you know the desired RAAN.

I've been assuming that the depot's payload is the parts.  See here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404058#msg2404058) and here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404380#msg2404380).

I think spin-stabiized is off the table for a long time.  Whatever the depot turns out to be, it's almost certainly a Starship variant.  You can't dock with it if it's spinning.

Really slow spinning depot(s) to provide the micro/milli g for settling would not be hard to dock to.
Spinning for people is a different matter and a different vehicle (or section of one).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/16/2022 08:51 pm
A related question:  Does this change when targeting the lunar poles?  I don't even have a good feel for how delta-v changes when TLI is set up for:

1) TLI for equatorial LLO
2) TLI for polar LLO
3) TLI for fast NRHO insertion
4) TLI for BLT

Presumably, inclination plays a factor in all of these, in addition to raw C3.
Why would the impact of the inclination of the parking orbit around the Earth differ for those different orbits? It's essentially about whether you waste any delta-v from the orbit when you transfer to the lunar trajectory. But once you're on that trajectory (no matter how much or little delta-v it cost you), I can't see how the original inclination would factor in.

That sounds right as far as it goes, but it's basically assuming that the delta-v for a plane change at TLI is trivial.  I'd be surprised if it's more than a few hundred m/s for the entire gamut of possible transit orbits to various types of lunar orbit injection, but a few hundred m/s is non-trivial.

In all cases, if you're doing a plane change at TLI, then the total delta-v will have a plane-change component baked into it (in the usual, hard-to-visualize vector sum way that these things go).  So the question is:  How big is that component?

Note that injection into a retrograde LLO is partially a gravity assist maneuver, and the angle of the "asymptote" (to the extent that you have such a clean parameter in a 3-body system) is crucial.

I would assume that there are no free-return requirements for the LSS (HLS Starship), since it's uncrewed during transit.  But that might not be true later in the Artemis campaign, if a crewed LSS or EDL-capable Starship is staged from LEO.  That would also have inclination implications--especially at high polar latitudes.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/16/2022 08:58 pm
Really slow spinning depot(s) to provide the micro/milli g for settling would not be hard to dock to.

Describe how that wouldn't be hard.  Note that "The Blue Danube" can't be the background soundtrack to your description.

Note also that the depot Starship will not have the same axis of rotation at various fill levels unless tail-to-tail docking is back on the table.  I don't know why SpaceX ruled that one out, but I suspect that the risk of pranging multiple engine bells on both ships might have had something to do with it.  And that's considerably less terrifying than even minuscule rotations.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/16/2022 09:03 pm
Don’t need to stay settled at all times, just during transfer.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/16/2022 10:16 pm
Really slow spinning depot(s) to provide the micro/milli g for settling would not be hard to dock to.
Spinning for people is a different matter and a different vehicle (or section of one).
How do you want to spin it? If it's around the long axis, that's not stable. Spinning it slowly just delays things--absent some sort of active stabilization, it's going to end up spinning across the short axis (or tumbling). Also, how are you going to add/remove fuel? All the plumbing is in the wrong place--the worst place. Or are you talking about some entirely new vehicle?

If you're planning to spin it on the short axis, you've still got stability issues, and now the plumbing for the methane tank is upside down--plus I think the center of gravity will be inside the LOX tank, which will mean you'll get liquid out of the gas line.

I really do love the idea of rotating structures in space, but I don't think anything SpaceX has done or is likely to do in the next 10 years is going to use them.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/17/2022 12:22 am
Really slow spinning depot(s) to provide the micro/milli g for settling would not be hard to dock to.
Spinning for people is a different matter and a different vehicle (or section of one).
How do you want to spin it? If it's around the long axis, that's not stable. Spinning it slowly just delays things--absent some sort of active stabilization, it's going to end up spinning across the short axis (or tumbling). Also, how are you going to add/remove fuel? All the plumbing is in the wrong place--the worst place. Or are you talking about some entirely new vehicle?

If you're planning to spin it on the short axis, you've still got stability issues, and now the plumbing for the methane tank is upside down--plus I think the center of gravity will be inside the LOX tank, which will mean you'll get liquid out of the gas line.

I really do love the idea of rotating structures in space, but I don't think anything SpaceX has done or is likely to do in the next 10 years is going to use them.
Doesn't matter if it's unstable if you have attitude control.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/17/2022 12:34 am
Doesn't matter if it's unstable if you have attitude control.
Well, maybe, but (to me) it looks like yet another thing to have to work out, yet another thing that could go wrong, and something that might have weird interactions with other needs for attitude control. But I don't claim to be an expert.

That still leaves you with the plumbing issues though.

I should add that I was in love with the "spin it around its long axis" idea just a few months ago, so I'm mostly just repeating the arguments that convinced me that it just wasn't going to happen. They still seem like pretty good arguments to me.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: meekGee on 09/17/2022 02:24 am
Nope. Subcooled methane and LOx are not miscible because the methane will freeze.
Is the "because" an argument or an assertion about the chemistry of LOx and methane at the temperature in question?

There are plenty of solids that are soluble in various different liquids, so the conclusion does not follow without additional information.

Or are you simply arguing that the technical term is soluble rather than miscible?
I’m arguing that the rate of mixing or solvating will be dramatically slowed by the fact that methane and LOx have vastly different densities to begin with, especially if the methane is already starting off on top and both are subcooled, leading to methane freezing on the interface, much like kerosene would.

Things can only mix on an interface.
My scepticism regarding frozen methane came from looking at a methane-oxygen phase diagram and confirming that the (partial) melting point is lower than the weighted average of the melting points (extrapolated below 77 K but most likely always true). This suggests that mixing liquid methane and liquid oxygen will always result in a liquid (as the mixing itself is unlikely to be highly endothermic).

I expect that you can float liquid methane on top of liquid oxygen (just like you can alcohol on water) but I have a hard time imagining the kinematics of the methane freezing. I will concede that the layering might in theory be done gently enough for heat conduction to outpace the combination of diffusion and heat of mixing + fusion, especially under low accelerations, but I imagine the conditions required to be slightly different from a rocket launch + dome failure.
Even if the methane did freeze, that would not be an instantaneous event.  There would always be some minor volume of mixed methane and LOX, enough to support ignition/detonation.
The concern(-troll?) has been that all the methane and oxygen would fully mix into a highly explosive mixture, so you’d get an absolutely enormous detonation if it lit off, not just a fireball.

However that’s extremely unlikely for the reasons I mentioned. Both solid an liquid methane are far less dense than liquid oxygen (by more than a 2:1 ratio, much greater than oil and water), so the two would not mix well if you removed a barrier separating them. The buoyancy of methane would mean you’d only get some small amount of mixing at the interface. And where there WAS an interface, the cold liquid oxygen would freeze the methane to a solid, which would sink to the bottom of the methane and float above the oxygen, forming an additional barrier to mixing of the two liquids.

It wouldn’t be a *safe* by any stretch, but it also isn’t going to be a Hiroshima-level detonation (which is what the naive calculation of them being fully mixed would say). Probably the amount fully mixed would be on the order of 1% or so, at least for a while. Especially in the case of subcooled propellants (which ensure the fuel would be frozen when it touches the LOx), this isn’t much different than kerolox.
Buoyancy needs gravity or acceleration...  So it depends on what the rocket is doing...  It may still be thrusting, or it may be decelerating...

I agree the fluids won't be comprehensively pre-mixed, but there may be enough local mixing to start something, which would then simultaneously cause more structural damage but also disperse the contents of the tanks, inhibiting further mixing, but helping atmospheric oxygen play a role.

So ...   Depends?

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/17/2022 02:59 am
In a nominal case, if the booster is no longer thrusting, then it's coasting in near zero gee and near vacuum, so there's effectively no buoyancy (and in any case, it's nearly empty). It starts feeling acceleration again as drag, in the same direction as it reenters engines-first like F9.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/17/2022 02:40 pm
A related question:  Does this change when targeting the lunar poles?  I don't even have a good feel for how delta-v changes when TLI is set up for:

1) TLI for equatorial LLO
2) TLI for polar LLO
3) TLI for fast NRHO insertion
4) TLI for BLT

Presumably, inclination plays a factor in all of these, in addition to raw C3.
Why would the impact of the inclination of the parking orbit around the Earth differ for those different orbits? It's essentially about whether you waste any delta-v from the orbit when you transfer to the lunar trajectory. But once you're on that trajectory (no matter how much or little delta-v it cost you), I can't see how the original inclination would factor in.

That sounds right as far as it goes, but it's basically assuming that the delta-v for a plane change at TLI is trivial.  I'd be surprised if it's more than a few hundred m/s for the entire gamut of possible transit orbits to various types of lunar orbit injection, but a few hundred m/s is non-trivial.

In all cases, if you're doing a plane change at TLI, then the total delta-v will have a plane-change component baked into it (in the usual, hard-to-visualize vector sum way that these things go).  So the question is:  How big is that component?

Note that injection into a retrograde LLO is partially a gravity assist maneuver, and the angle of the "asymptote" (to the extent that you have such a clean parameter in a 3-body system) is crucial.

I would assume that there are no free-return requirements for the LSS (HLS Starship), since it's uncrewed during transit.  But that might not be true later in the Artemis campaign, if a crewed LSS or EDL-capable Starship is staged from LEO.  That would also have inclination implications--especially at high polar latitudes.
A close to worst possible case would be an equatorial LEO to lunar polar plane change. At L1 earth and lunar gravity balance out. Aim TLI at L1. Velocity at L1 can be arbitrarily low. It could conceivably be zero. THIS is where the plane change is most economical. It's a variation on the 3 burn RAAN except it goes on to a different body.


The downsides are 1) the more economical the plane change, the longer the transit time and 2) this only works for lunar orbits that pass through the earth-moon line. It would not work for an earth synchronous orbit equivalent to a LEO sun synchronous orbit, for example.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/17/2022 03:20 pm
Really slow spinning depot(s) to provide the micro/milli g for settling would not be hard to dock to.

Describe how that wouldn't be hard.  Note that "The Blue Danube" can't be the background soundtrack to your description.

Note also that the depot Starship will not have the same axis of rotation at various fill levels unless tail-to-tail docking is back on the table.  I don't know why SpaceX ruled that one out, but I suspect that the risk of pranging multiple engine bells on both ships might have had something to do with it.  And that's considerably less terrifying than even minuscule rotations.
Yeah, no spin during transfer ops. Depending on he trades, low rpm spin may help control boiloff while waiting between transfer ops. Weather, safety and mechanical delays are a fact of life. More so for a crewed launch.


The last render we saw, maybe past its 'use by' date, implies that tankers are not intended to return and may be used for multiple campaigns. There may be propellant dregs after a campaign. Good boiloff control allows for opportunistic tanker flights instead of a concerted effort. If it costs less to retain rather than replace propellant, why not? That's where the trade offs come in.


Among the tradeoff are: how much propellant it takes so spin up/down, how much this saves per unit time and how long a delay is expected. Alas, it is beyond my capabilities to calculate any of this. RSI (Rocket Scientist Intervention) is welcome.


If we can't have Blue Danube how about some Lead Zep?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/17/2022 09:55 pm
A close to worst possible case would be an equatorial LEO to lunar polar plane change.

At lunar distance, a polar orbit (at NRHO-distance from the moon) is around half a degree difference in TLI angle vs lunar equator. A half degree plane change during TLI is around 1/20,000th of original orbital velocity, or around 35cm/s additional delta-v.

Plane changes needed to inject into equator/polar orbit at distant targets are never an issue. Only changing the local RAAN to match launch windows, and in this case, it's been mentioned that precession gives a window every week.

Hence any depot in any Earth orbit can serve the moon. It's just not an issue.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TaurusLittrow on 09/18/2022 11:34 am
What's the latest thinking on plans for on-orbit propellant transfer (Tanker-to-Depot-to-HLS).

Aft docking? Side-by-side?

Ullage motors to settle the prop?

Internal tank bladders, baffles?

Prop transfer is an esstential part of the Starship system and Artemis architecture yet not a lot news.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/18/2022 03:42 pm
This NASA document, NASA’s Initial Artemis Human Landing System (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf), seems to answer a number of those questions.

There will be three new Starship variants:

1) A reusable tanker, which has no cargo space but the tanks are extended so it can carry as much fuel as possible but still fly back to the ground to get more.

2) A depot, which is like the tanker, but with no tiles and no landing gear, so it can be a bit longer and hold a bit more fuel. The depot is designed to passively reduce boiloff in low Earth orbit. From the picture (figure 2) it looks like it has a very reflective surface. Repeated tanker flights will fill up the depot. The depot can't return to Earth, but you could use it for multiple flights of the HLS.

3) The HLS (human landing system) for Artemis. In addition to whatever modifications it needs to land on the moon, it's designed to passively reduce boiloff in NRHO. It looks to be covered in Solar White or something similar, which could let it "loiter" indefinitely near Gateway without the LOX and LCH4 boiling off. The HLS will rendezvous with the depot, fill up with fuel, and fly off to Gateway in NRHO to wait for the astronauts to arrive. No clue what happens with HLS after a flight.

Based on that, I'd say that, for sure, there will be ullage motors, and I'd expect the docking will be side-by-side. Not sure about baffles or bladders.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 09/18/2022 03:54 pm
What's the latest thinking on plans for on-orbit propellant transfer (Tanker-to-Depot-to-HLS).

Aft docking? Side-by-side?

Ullage motors to settle the prop?

Internal tank bladders, baffles?

Prop transfer is an esstential part of the Starship system and Artemis architecture yet not a lot news.
This thread is now 79 pages long, and everything you asked about is under discussion. It does not appear to be converging. There is no agreement on even some fundamentals. For instance, the HLS contract requires fuel transfer from tanker to depot and from depot to HLS, but most of the posts on the thread seem to assume that other transfers (tanker-to-tanker, tanker-to-HLS) will also be implemented.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: billh on 09/18/2022 05:47 pm
What's the latest thinking on plans for on-orbit propellant transfer (Tanker-to-Depot-to-HLS).

Aft docking? Side-by-side?

Ullage motors to settle the prop?

Internal tank bladders, baffles?

Prop transfer is an esstential part of the Starship system and Artemis architecture yet not a lot news.
This thread is now 79 pages long, and everything you asked about is under discussion. It does not appear to be converging. There is no agreement on even some fundamentals. For instance, the HLS contract requires fuel transfer from tanker to depot and from depot to HLS, but most of the posts on the thread seem to assume that other transfers (tanker-to-tanker, tanker-to-HLS) will also be implemented.
The facts-to-speculation ratio is too low. I would suggest waiting for some new bone to chew on before resuming some of these discussions.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TaurusLittrow on 09/18/2022 06:37 pm
What's the latest thinking on plans for on-orbit propellant transfer (Tanker-to-Depot-to-HLS).

Aft docking? Side-by-side?

Ullage motors to settle the prop?

Internal tank bladders, baffles?

Prop transfer is an esstential part of the Starship system and Artemis architecture yet not a lot news.
This thread is now 79 pages long, and everything you asked about is under discussion. It does not appear to be converging. There is no agreement on even some fundamentals. For instance, the HLS contract requires fuel transfer from tanker to depot and from depot to HLS, but most of the posts on the thread seem to assume that other transfers (tanker-to-tanker, tanker-to-HLS) will also be implemented.
The facts-to-speculation ratio is too low. I would suggest waiting for some new bone to chew on before resuming some of these discussions.

Well, I asked in case there were some factoids out there. Seems odd that so little info is available even about fundamental parameter and ConOps.

A paper by MSFC engineers presented at IAC says "Aft docking mechanism designs...continued to mature." So looks like aft docking is it.

But the rest is a mystery for now.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/18/2022 11:52 pm
A paper by MSFC engineers presented at IAC says "Aft docking mechanism designs...continued to mature." So looks like aft docking is it.
Depending on what "aft docking" means. Looking online a bit, it seems that that merely means that the docking mechanism is located at the back end of the vehicle. If that's all it means, then the docking could be both "aft" and side-by-side.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 09/19/2022 01:05 am
A paper by MSFC engineers presented at IAC says "Aft docking mechanism designs...continued to mature." So looks like aft docking is it.
Depending on what "aft docking" means. Looking online a bit, it seems that that merely means that the docking mechanism is located at the back end of the vehicle. If that's all it means, then the docking could be both "aft" and side-by-side.

aft+SxS implies a very stiff docking mechanism though if that also means no forward stabilization.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/19/2022 01:18 am
aft+SxS implies a very stiff docking mechanism though if that also means no forward stabilization.
How so? At this point, I'm not sure we can rule out "docking" that's little more than a "prehensile hose" plus carefully coordinated flying. The best argument for aft+SxS is that it adds almost no new plumbing. Is there a good argument in favor of any other configuration?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: greybeardengineer on 09/19/2022 03:37 am
aft+SxS implies a very stiff docking mechanism though if that also means no forward stabilization.
How so? At this point, I'm not sure we can rule out "docking" that's little more than a "prehensile hose" plus carefully coordinated flying. The best argument for aft+SxS is that it adds almost no new plumbing. Is there a good argument in favor of any other configuration?

For crewed Starships it will be desirable to be able directly transfer personnel and small cargo items between two docked ships. The most reasonable configuration is +Z long edge to +Z long edge, noses in same direction. This suggests two mating regions - one in the tail section for propellant transfer and one in the cylindrical portion of the forward section for docking adaptor/transfer tunnel. The same basic pattern can be followed for uncrewed variants but with just the mechanical support function for the forward mating region.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 09/19/2022 10:54 am
aft+SxS implies a very stiff docking mechanism though if that also means no forward stabilization.
How so? At this point, I'm not sure we can rule out "docking" that's little more than a "prehensile hose" plus carefully coordinated flying. The best argument for aft+SxS is that it adds almost no new plumbing. Is there a good argument in favor of any other configuration?

For crewed Starships it will be desirable to be able directly transfer personnel and small cargo items between two docked ships. The most reasonable configuration is +Z long edge to +Z long edge, noses in same direction. This suggests two mating regions - one in the tail section for propellant transfer and one in the cylindrical portion of the forward section for docking adaptor/transfer tunnel. The same basic pattern can be followed for uncrewed variants but with just the mechanical support function for the forward mating region.
There are currently no CONOPS that require crewed Starships to dock with each other (let alone transfer cargo). Nor is there any that also need to transfer propellant at the same time.

A lot of people are making the assumption of premature optimisations: e.g. that tankers will have extended tanks, that the depot will have a boiloff minimisation system, that tankers need to transfer propellant to anything other than a depot, etc. None of those are actually necessary for the HLS CONOPS to work. 'stock' tankers and a depot that just allows boiloff to happen may require an additional tanker flight or two (and maybe an occasional replenishment flight if the in-orbit loiter time requirement is applied to the depot as well as the HLS Starship itself) but does not prevent the architecture from working. A tanker never needs to transfer to anything other than the depot, so any sort of coupler androgyny is unnecessary. I'm even of the opinion that the 'long' depot is a red herring and two standard-sized depots would be an adequate substitute.

Think of the absolute minimum viable option, and only add to it when necessary. Optimisations can come after functionality.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/19/2022 12:35 pm
I do actually think there would be benefit to boiloff mitigation, whether depot or HLS. That ultimately limits endurance.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/19/2022 01:30 pm
For crewed Starships it will be desirable to be able directly transfer personnel and small cargo items between two docked ships. The most reasonable configuration is +Z long edge to +Z long edge, noses in same direction. This suggests two mating regions - one in the tail section for propellant transfer and one in the cylindrical portion of the forward section for docking adaptor/transfer tunnel. The same basic pattern can be followed for uncrewed variants but with just the mechanical support function for the forward mating region.
I think this is an excellent suggestion, but isn't it just an elaboration of side-by-side aft docking?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: rsdavis9 on 09/19/2022 03:52 pm
Anyways my slow rotation was to me visualized as:
1. 2+ starships docked nose to nose with a hub connecting the 2+ noses together.
2. Slow rotation for 1milli g is 448s /rev
          1/sqrt(.001g/50m)*2pi = 448s
          a=w^2*r

So basically we have 2 starships with acceleration the same direction as on the ground. This is good.

Piping to connect the QD's together. Probably up the length of the ship to the hub where they are docked to.

If you are worried about uneven weight between them, then just dock more starships nose to nose. It is going to take multiple tankers for refueling.
 
The receiving ship (the starship being refueled) doesn't necessarily require settling. Introducing fuel at bottom should fill without sloshing? So you could have just the tankers spinning and a pipe coming out of the hub at the nose (small rotating joint on pipe) going to the receiving ship.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/19/2022 04:24 pm
Piping to connect the QD's together. Probably up the length of the ship to the hub where they are docked to.
Not a bad idea, but it violates a few principles that (some of us, at least) have been trying to follow:

1) Minimize new plumbing. This involves really long pipes--about as long as they can be.
2) Minimize changes to the basic Starship. These new, long pipes are going to be part of every Starship--or at least, every one that needs to be refueled.
3) Minimize new technology. At the moment, we've really got almost zero experience with rotating structures in space. Stabilizing two (or more) rotating Starships while the center of mass changes radically seems like quite of lot of new tech to master.

Add to that the fact that the fuel required for these very gentle ullage burns is very small, and it really seems like ullage burns--even ones that are hours long--are clearly the way to go for some time to come.

It's a pity. I really have a soft spot for rotating space habitats, but I don't see anything happening with them for the next decade at least.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/19/2022 06:40 pm
A lot of people are making the assumption of premature optimisations: e.g. that tankers will have extended tanks, that the depot will have a boiloff minimisation system, that tankers need to transfer propellant to anything other than a depot, etc. None of those are actually necessary for the HLS CONOPS to work. 'stock' tankers and a depot that just allows boiloff to happen may require an additional tanker flight or two (and maybe an occasional replenishment flight if the in-orbit loiter time requirement is applied to the depot as well as the HLS Starship itself) but does not prevent the architecture from working. A tanker never needs to transfer to anything other than the depot, so any sort of coupler androgyny is unnecessary. I'm even of the opinion that the 'long' depot is a red herring and two standard-sized depots would be an adequate substitute.

Think of the absolute minimum viable option, and only add to it when necessary. Optimisations can come after functionality.

Given the delta-v budget that it needs to do LEO-NRHO-LS-NRHO on one tank of prop, including both FPR and a boiloff budget, it seems likely that the LSS (i.e., the HLS version of Starship) will need stretched tanks, though.  So the minimum viable suite of components might have a depot that was based on LSS, rather than on a vanilla Starship.  That would also allow the depot to inherit the white paint and any other boiloff-mitigating features from the LSS.

That would imply that the depot was indeed non-EDL-capable.  But a lift tanker could still just be a vanilla Starship with no payload.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 09/19/2022 07:04 pm
A lot of people are making the assumption of premature optimisations: e.g. that tankers will have extended tanks, that the depot will have a boiloff minimisation system, that tankers need to transfer propellant to anything other than a depot, etc. None of those are actually necessary for the HLS CONOPS to work. 'stock' tankers and a depot that just allows boiloff to happen may require an additional tanker flight or two (and maybe an occasional replenishment flight if the in-orbit loiter time requirement is applied to the depot as well as the HLS Starship itself) but does not prevent the architecture from working. A tanker never needs to transfer to anything other than the depot, so any sort of coupler androgyny is unnecessary. I'm even of the opinion that the 'long' depot is a red herring and two standard-sized depots would be an adequate substitute.

Think of the absolute minimum viable option, and only add to it when necessary. Optimisations can come after functionality.

Given the delta-v budget that it needs to do LEO-NRHO-LS-NRHO on one tank of prop, including both FPR and a boiloff budget, it seems likely that the LSS (i.e., the HLS version of Starship) will need stretched tanks, though.  So the minimum viable suite of components might have a depot that was based on LSS, rather than on a vanilla Starship.  That would also allow the depot to inherit the white paint and any other boiloff-mitigating features from the LSS.

That would imply that the depot was indeed non-EDL-capable.  But a lift tanker could still just be a vanilla Starship with no payload.
But there will not yet be a "vanilla Starship" before they need tanker. The only earlier EDL-capable Starship we actually have evidence for is the Pez dispenser. I suppose you can use Pez dispenser as a taanker, but would think that the modification needed to expand the tanks into the payload area are quite small: much smaller than just about any other SS modification.

Depot is a bigger deal. It's likely to be stretched, and it may need to bear the entire burden of the fuel transfer process (connections, pumps, RCS, etc.) to minimize refueling HW on all the other SS.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/19/2022 07:23 pm
aft+SxS implies a very stiff docking mechanism though if that also means no forward stabilization.
How so? At this point, I'm not sure we can rule out "docking" that's little more than a "prehensile hose" plus carefully coordinated flying. The best argument for aft+SxS is that it adds almost no new plumbing. Is there a good argument in favor of any other configuration?
I've yet to figure out what +SxS means. Be careful if the docking label. It's carried a couplea three different meanings in this thread.


Passing a hose and pumping is a tough one. If flying side by side the orbits will intersect in 1/4 orbit. If over and under, one will be a bit faster, one a bit slower. Corrections needed either way. It can work but for transferring fluids, it gives me the creeps.


A solid physical connection that makes the two ships one orbital body gives me a warm passive system fuzzy. QD/QD and the old concept of butt to butt both do this. Butt to Butt also puts both ships in exactly the same orbit for whatever that's worth. Nobody outside SX knows why this was abandoned. Every other option we've discussed seems to add complexity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/19/2022 08:39 pm
aft+SxS implies a very stiff docking mechanism though if that also means no forward stabilization.
How so? At this point, I'm not sure we can rule out "docking" that's little more than a "prehensile hose" plus carefully coordinated flying. The best argument for aft+SxS is that it adds almost no new plumbing. Is there a good argument in favor of any other configuration?

For crewed Starships it will be desirable to be able directly transfer personnel and small cargo items between two docked ships. The most reasonable configuration is +Z long edge to +Z long edge, noses in same direction. This suggests two mating regions - one in the tail section for propellant transfer and one in the cylindrical portion of the forward section for docking adaptor/transfer tunnel. The same basic pattern can be followed for uncrewed variants but with just the mechanical support function for the forward mating region.
There are currently no CONOPS that require crewed Starships to dock with each other (let alone transfer cargo). Nor is there any that also need to transfer propellant at the same time.

A lot of people are making the assumption of premature optimisations: e.g. that tankers will have extended tanks, that the depot will have a boiloff minimisation system, that tankers need to transfer propellant to anything other than a depot, etc. None of those are actually necessary for the HLS CONOPS to work. 'stock' tankers and a depot that just allows boiloff to happen may require an additional tanker flight or two (and maybe an occasional replenishment flight if the in-orbit loiter time requirement is applied to the depot as well as the HLS Starship itself) but does not prevent the architecture from working. A tanker never needs to transfer to anything other than the depot, so any sort of coupler androgyny is unnecessary. I'm even of the opinion that the 'long' depot is a red herring and two standard-sized depots would be an adequate substitute.

Think of the absolute minimum viable option, and only add to it when necessary. Optimisations can come after functionality.
Premature optimization avoidance is a good thing. Educated ideas of where optimization will be needed are good things too. It's always good to look one or two moves ahead with the option of changing the game plan if things don't work out as expected.


The tanker tank stretch is a good example. Nobody is claiming this need be done before a proof of concept is done. It need not even follow close on the heels of PoC or even for the first operational depot fill campaign. The complexity needed for a tank stretch is less than what we've been seeing on the periodic design upgrades. To most of us the utility and ease of the idea makes it a no brainer - definitely in the long view, probably in the mid view, and maybe in the short view.


IF QD/QD transfer becomes real and IF additional bracing further up the hull becomes real, THEN extending the idea only a little bit to allow ships to hook up for personnel/cargo transfer becomes a design option. There may be no current conops needing this but that doesn't stop it from being an attractive option for future conops.


It would be foolish for SX to force fit the QD and bracing scheme for propellant transfer if it doesn't want to work out. OTOH, if it shows promise of easy expansion for additional capabilities it would warrant some extra effort to overcome minor to middling development gotchas. Never pass up a twofer.


Musk is a dreamer and at the same time a hardheaded business man (let's ignore Twitter) with good engineering instincts. He successfully embraces a minimalist development process but once a system's proven itself it's fair game for optimization. We've seen one part optimizing while other parts are still waiting for a PoC.


BTW, I do agree that tankers will only transfer to the depot. Nobody expects to drive up to a highway gasoline tanker for a fillup.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/19/2022 09:53 pm
But there will not yet be a "vanilla Starship" before they need tanker. The only earlier EDL-capable Starship we actually have evidence for is the Pez dispenser. I suppose you can use Pez dispenser as a taanker, but would think that the modification needed to expand the tanks into the payload area are quite small: much smaller than just about any other SS modification.

Depot is a bigger deal. It's likely to be stretched, and it may need to bear the entire burden of the fuel transfer process (connections, pumps, RCS, etc.) to minimize refueling HW on all the other SS.

You're over-parsing "vanilla Starship."  The early flights will still use an EDL-capable vehicle with 1200t of tankage. 

LSS, I'm pretty sure, needs more than 1200t of tankage, and it can't be EDL-capable if its dry mass is to be reasonable.  Ignoring any payload left on the lunar surface (which will likely be trivial for Option A) LEO-NRHO(BLT)-LS-NRHO requires 8965m/s of delta-v.  Using Isp=378s, that makes the maximum inert mass (dry mass + crew module + payload + let's say 20t of prop for FPR and boiloff allowance) 117t.  Take out the 20t prop reserve and you're at 97t for dry mass and crew module.  That might get the job done, but margins would be imprudently tight.  It's going to make sense to trade away some cylindrical payload height for more prop.  (NB: It's not really a stretch, just a rearrangement of ring, bulkhead, and dome segments.)

It would of course be possible simply to make all Starships have 1300t-1400t of tankage, but that's a bad deal for the Pez dispenser, which relies on the full 8m of cylindrical payload height.  So, given that you'll likely have two different tankage configurations, and the lift tankers really don't need to be stretched, then making the depot and LSS rely on a common stretch makes sense, with the lift tanker being vanilla.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/19/2022 09:58 pm
There are two ways to increase delta-v capability. Increase propellant capacity or reduce dry mass. Don’t underestimate the latter.

Additionally, it’s possible to pick an elliptical (although maybe still LEO?… this buys you ~400-500m/s if you stay at LEO and up to 3km/s if you go higher) orbit for refueling or even use the depot itself as a sort of departure stage. I wouldn’t bet on a tank stretch of HLS being the way they solve this.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/19/2022 11:29 pm
I've posted to the F9 US reusability discussion. https://forum.nasaspaceflight.com/index.php?topic=42637.new#new (https://forum.nasaspaceflight.com/index.php?topic=42637.new#new)


It's about reassessing a miniSS. Let's move the off topic discussions there. It's gotta be bad if I'm complaining about off topic.


That new post gives reasons for a reassessment. Send a raspberry, reply or ignore it.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/20/2022 05:43 pm
There are two ways to increase delta-v capability. Increase propellant capacity or reduce dry mass. Don’t underestimate the latter.

The computation I did was to pick the maximum inert mass (including a guess of 20t for FPR/boiloff) that would accommodate the delta-v budget available.  So dry + crew module mass is already factored in.

Quote
Additionally, it’s possible to pick an elliptical (although maybe still LEO?… this buys you ~400-500m/s if you stay at LEO and up to 3km/s if you go higher) orbit for refueling or even use the depot itself as a sort of departure stage. I wouldn’t bet on a tank stretch of HLS being the way they solve this.

300 x 2000km will indeed give you an extra 425m/s if you're shooting for C3=-1.5, which is about what a BLT needs.  That takes you down to a total budget of 8540m/s, which leaves you with an inert mass of 133t.  Take out the same 20t for FPR and boiloff, and you're at 113t for dry mass and crew module.

That's more margin.  Still not great, but possible.

To figure out whether that's a viable depot orbit, you'd have to look at your launch window opportunities.  There are two different constraints:

1) How often does the RAAN and argument of perigee line up with your TLI parameters?

2) How often does the RAAN and argument of perigee line up with your ability to launch tankers?

Intuitively, it seems like 300x2000 would give you one window per month for your TLI.  BLTs are a little bit weird in this respect, in that they require the Moon to be at some narrower range of angles to the ecliptic than a fast TLI, but one a month is a decent arm-wave.

If the depot is in a non-circular orbit, then phasing for post-launch rendezvous is more expensive.  If you restrict tanker launches to windows where the depot is close to perigee the launch site's RAAN is close to the depot's, there are a lot fewer cheap windows for tankers.  It's beyond my competence to figure this out explicitly, but it's a non-trivial constraint.

I suspect that a true HEEO (>LEO+2500m/s) for a depot will turn out to be unacceptably restrictive in terms of the two different kinds of windows.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/20/2022 07:47 pm

If the depot is in a non-circular orbit, then phasing for post-launch rendezvous is more expensive.  If you restrict tanker launches to windows where the depot is close to perigee the launch site's RAAN is close to the depot's, there are a lot fewer cheap windows for tankers.  It's beyond my competence to figure this out explicitly, but it's a non-trivial constraint.

Even if departure is from a non-circular orbit, you'd fill the depot in a low circular orbit.  It's easier to rendezvous and more fuel efficient, because you are not boosting 9/10 empty tankers to a higher orbit,

It uses less fuel to transfer the dry weight of the depot to and from the intermediate orbit than to transfer the dry weight of half a dozen or more tankers to the intermediate orbit.  (Assuming the dry weight of the tanker is less than 3 times the dry weight of a tanker).



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/20/2022 08:38 pm
Additionally, it’s possible to pick an elliptical (although maybe still LEO?… this buys you ~400-500m/s if you stay at LEO and up to 3km/s if you go higher) orbit for refueling or even use the depot itself as a sort of departure stage.
I'm trying to think how this offers a benefit, but I'm not seeing it. As Barley says, you really don't want the tankers to all have to do this, but I'm not seeing the advantage in having the mission Starship do it either. It seems to me that you want all refueling operations in the lowest feasible orbit. (i.e., consistent with costs from drag and boiloff.) Otherwise, you're boosting mass you didn't have to boost.

An obvious exception would be if you could use something like an ion drive to lift the filled depot to a higher circular orbit with very little fuel cost. Then it would make sense to push it up high enough that a Starship coming to refuel would be just about empty when it arrived. Otherwise, I don't see it. What am I missing?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 09/20/2022 11:32 pm
I'll just drop a link for Jon Goff's blog post on roving depots associated with HEEO for interplanetary departure, in the context of Jon Goff's 3 burn departure paper, as reference info...

https://selenianboondocks.com/2021/02/an-updated-propellant-depot-taxonomy-part-vi-roving-depots/ (https://selenianboondocks.com/2021/02/an-updated-propellant-depot-taxonomy-part-vi-roving-depots/)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/21/2022 03:58 pm
Additionally, it’s possible to pick an elliptical (although maybe still LEO?… this buys you ~400-500m/s if you stay at LEO and up to 3km/s if you go higher) orbit for refueling or even use the depot itself as a sort of departure stage.
I'm trying to think how this offers a benefit, but I'm not seeing it. As Barley says, you really don't want the tankers to all have to do this, but I'm not seeing the advantage in having the mission Starship do it either. It seems to me that you want all refueling operations in the lowest feasible orbit. (i.e., consistent with costs from drag and boiloff.) Otherwise, you're boosting mass you didn't have to boost.

An obvious exception would be if you could use something like an ion drive to lift the filled depot to a higher circular orbit with very little fuel cost. Then it would make sense to push it up high enough that a Starship coming to refuel would be just about empty when it arrived. Otherwise, I don't see it. What am I missing?

You're correct that lower energy refueling is always more efficient, but it's not the issue at hand.  The question is whether an LSS with only 1200t of tank capacity can do LEO-NRHO-LS-NRHO on one tank of prop.  Increasing the energy of the refueling orbit helps with that.  I'm still not sure it closes for levels of eccentricity that don't have huge additional complications, but it certainly improves the situation.

If it still doesn't close, then the LSS almost certainly needs a bigger tank, which likely means that the depot needs a bigger tank as well.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/21/2022 04:03 pm
Or you lower the dry mass.

Or you use the depot’s engines for a boost.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/21/2022 04:12 pm
Additionally, it’s possible to pick an elliptical (although maybe still LEO?… this buys you ~400-500m/s if you stay at LEO and up to 3km/s if you go higher) orbit for refueling or even use the depot itself as a sort of departure stage.
I'm trying to think how this offers a benefit, but I'm not seeing it. As Barley says, you really don't want the tankers to all have to do this, but I'm not seeing the advantage in having the mission Starship do it either. It seems to me that you want all refueling operations in the lowest feasible orbit. (i.e., consistent with costs from drag and boiloff.) Otherwise, you're boosting mass you didn't have to boost.

An obvious exception would be if you could use something like an ion drive to lift the filled depot to a higher circular orbit with very little fuel cost. Then it would make sense to push it up high enough that a Starship coming to refuel would be just about empty when it arrived. Otherwise, I don't see it. What am I missing?

You're correct that lower energy refueling is always more efficient, but it's not the issue at hand.  The question is whether an LSS with only 1200t of tank capacity can do LEO-NRHO-LS-NRHO on one tank of prop.  Increasing the energy of the refueling orbit helps with that.  I'm still not sure it closes for levels of eccentricity that don't have huge additional complications, but it certainly improves the situation.

If it still doesn't close, then the LSS almost certainly needs a bigger tank, which likely means that the depot needs a bigger tank as well.
Circularizing to the high perigee does not take nearly as much dV as getting to that perigee (I think). It would make rendezvous   easier.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/21/2022 05:46 pm
Circularizing to the high perigee does not take nearly as much dV as getting to that perigee (I think). It would make rendezvous   easier.
Delta V from a moderately high circular orbit to escape is higher than from an eccentric orbit with the same apogee.  From the high circular orbit you first want to burn retrograde so you can do a low perigee Oberth burn. .
Delta V from a high circular orbit is higher than from a eccentric orbit with the same energy.  Because you get a better Oberth effect from the lower perigee of the eccentric orbit

Also you have to recover tankers in a circular orbit propulsively.

One way to make rendezvous easier is for the lander and one or more moveable depots to assemble in loose formation in LEO then remain in formation as they boost the apogee then transfer fuel.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/21/2022 06:07 pm
Or you lower the dry mass.

You can calculate the dry mass you'd need to make it work.  I did that up-thread.  It needs to be improbably low.

Quote
Or you use the depot’s engines for a boost.

Not with a tenuous docking between the depot and the LSS, you can't.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/21/2022 06:43 pm
Or you lower the dry mass.

You can calculate the dry mass you'd need to make it work.  I did that up-thread.  It needs to be improbably low.

Nah. If you can get good mass fraction by stretching the tank for more fuel with high confidence, then I don’t see how you can rule out getting the same mass fraction by just taking less dry mass with you. 80 tons dry mass is not actually out of the question.

And who said anything about the docking needing to be tenuous?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/21/2022 06:53 pm
Depending on how long Starship HLS can last with low boiloff, you can trade time for significantly lower delta-v. Maybe another 400-600m/s, on top of the ~400m/s you get from filling up at high LEO.

That’s up to 1km/s in saved delta-v, and you can just do a high apogee elliptical orbit depot staging point, maybe all the way to GTO-like orbits or even NHRO. That saves 3-4m/s or so.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/21/2022 07:52 pm
One way to make rendezvous easier is for the lander and one or more moveable depots to assemble in loose formation in LEO then remain in formation as they boost the apogee and transfer fuel.
I'm thinking this isn't really going to work.

Going from a circular orbit at 200 km altitude to a highly elliptical one with perigee at 200 km and apogee at 400,000 km, I figure (using vis viva) that you need delta-v of 3.136 kps. The vacuum Raptor has specific impulse of 3.56 kps, so the mass fraction will be exp(-3.136/3.56) = 41.4%.

That means that if you have a ship in LEO with dry mass plus payload of D and carrying propellants with mass F and you boost it to HEEO, the total mass that arrives there will be (D+F)*0.414 and since D can't change, that means the remaining fuel will be (D+F)*0.414 - D = 0.414F - 0.586D. As long as D is small compared to F, you'll keep about 41% of your fuel.

So consider a depot with dry mass (say) of 100 tons and carrying (say) 2000 tons of propellant in LEO. If we boost that into HEEO, it's still have 2100*.0414 - 100*.0586 = 770 tons to HEEO.

Now consider a tanker with dry mass of 100 tons carrying 150 tons of propellant in LEO. This can only deliver 150*.41 - 100*0.586 = 3.5 tons of propellant--it essentially arrives empty.

If you actually want a depot in HEEO, I think the most sensible way to refuel it is by filling another depot in LEO and sending that second depot up to refill the first one. Tankers should never leave LEO.

(Someone should check my math, of course.)

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/21/2022 08:11 pm
Circularizing to the high perigee does not take nearly as much dV as getting to that perigee (I think). It would make rendezvous   easier.

But then you don't get the Oberth advantages of the low perigee, and you don't reduce the post-refueling delta-v requirements enough to fit the mission into 1200t of prop.  (I don't think it fits even then, but it's at least close.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/21/2022 08:18 pm
Or you lower the dry mass.

You can calculate the dry mass you'd need to make it work.  I did that up-thread.  It needs to be improbably low.

Nah. If you can get good mass fraction by stretching the tank for more fuel with high confidence, then I don’t see how you can rule out getting the same mass fraction by just taking less dry mass with you. 80 tons dry mass is not actually out of the question.

And who said anything about the docking needing to be tenuous?

If you're arguing that stretching the tanks by trading payload bay height for more prop doesn't change the dry mass, then we're in violent agreement--that's just a rearrangement of ring segments, domes, and bulkheads.  But I thought the argument was you can stick with 1200t simply by doing dry mass reduction magic.  I don't think that's going to happen.

If you have dorsal-dorsal, aft-to-aft docking, then I think it's pretty tenuous--especially with 2MN engines involved.  The only way I can think of using depot thrust is if you have on-axis nose-to-tail docking.  That ain't happening any time soon.  (IIRC, we analyzed that quite a while back--it has some interesting properties.  But it's sufficiently different that the depot required isn't really a Starship variant any more.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/21/2022 09:08 pm
Why is anybody suggesting a depot boost, pushing the lander, rather than simply boosting the depot and lander in parallel then transferring fuel?

About the only scenario where I find a depot boost to be compelling is a two stage Oberth burn with the staging velocity above escape velocity.  For anything else you can find time for a refueling, taking an extra orbit if necessary.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/21/2022 09:15 pm
Why is anybody suggesting a depot boost, pushing the lander, rather than simply boosting the depot and lander in parallel then transferring fuel?
Does that mean that you'd attach the nose of the depot to the engines of the lander and then just push it? Is the idea that both the lander and the depot would be fully fueled so that at the end of it you'd have a fully fueled lander in some higher orbit with no need to transfer fuel?

Or do you mean something different?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/21/2022 09:20 pm
Why is anybody suggesting a depot boost, pushing the lander, rather than simply boosting the depot and lander in parallel then transferring fuel?
Does that mean that you'd attach the nose of the depot to the engines of the lander and then just push it? Is the idea that both the lander and the depot would be fully fueled so that at the end of it you'd have a fully fueled lander in some higher orbit with no need to transfer fuel?

Or do you mean something different?
That's what I think other people mean.  It seems a heck of a thing to do to avoid having to transfer fuel.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/21/2022 09:21 pm
Why is anybody suggesting a depot boost, pushing the lander, rather than simply boosting the depot and lander in parallel then transferring fuel?

The ideal trajectory is one where the depot and the final spacecraft use exactly half of their fuel to boost to an elliptical orbit, and then transfer the fuel to the final spacecraft and do the Oberth burn with a full load of fuel.   In that case the "depot" needs to have aerobraking capability to return to LEO (or land)

I've been noodling on that idea for about 9 months and for some reason I didn't realize you could boost to the depot and final spacecraft to elliptical together, thus minimizing rendezvous hassles.

Durrr.  thanks!

BTW two stage Oberth isn't necessary.  The refuel in elliptical is effectively equivalent to staging.  You've boosted a bunch of fuel to a higher energy, and then use the fuel at maximum kinetic energy/minimum potential energy, taking full advantage of Mr Oberth

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/21/2022 09:38 pm
Why is anybody suggesting a depot boost, pushing the lander, rather than simply boosting the depot and lander in parallel then transferring fuel?
Does that mean that you'd attach the nose of the depot to the engines of the lander and then just push it? Is the idea that both the lander and the depot would be fully fueled so that at the end of it you'd have a fully fueled lander in some higher orbit with no need to transfer fuel?

Or do you mean something different?
That's what I think other people mean.  It seems a heck of a thing to do to avoid having to transfer fuel.
Oh! I didn't think that was what anyone was proposing. I thought they wanted to put the depot into a highly eccentric Earth orbit (HEEO) rather than a low-Earth orbit (LEO) in order to pick up an extra 3 kps for the lunar Starship. There was a proposal to send a group of ships together to avoid the problems of figuring out how to rendezvous, but I don't think anyone was suggesting actually docking the ships and then having the depot drag them into a higher orbit! At least, I hope not. :-)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/21/2022 09:46 pm
BTW two stage Oberth isn't necessary.  The refuel in elliptical is effectively equivalent to staging.  You've boosted a bunch of fuel to a higher energy, and then use the fuel at maximum kinetic energy/minimum potential energy, taking full advantage of Mr Oberth
Two stage Oberth would be an advanced technique I don't think will happen soon.  It would get you higher v than a single fully fueled ship starting in high eliptical orbit.

Put two fully fueled SS in high elliptical orbit., connect them together.  At perigee fully burn one, separate and then burn the other. 

You cannot replace this by burning half the fuel in each then refueling because once you've burned half your two SS are in hyperbolic orbits and are not coming back, so Herr Oberth can't get a crack at the rest of the fuel.*  Also you don't get your booster back.  So this is a rare example of where staging and refueling cannot be made equivalent.

* Unless you can dock and transfer the fuel in a few minutes before leaving the Earths vicinity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/21/2022 09:57 pm
BTW two stage Oberth isn't necessary.  The refuel in elliptical is effectively equivalent to staging.  You've boosted a bunch of fuel to a higher energy, and then use the fuel at maximum kinetic energy/minimum potential energy, taking full advantage of Mr Oberth
Two stage Oberth would be an advanced technique I don't think will happen soon.  It would get you higher v than a single fully fueled ship starting in high eliptical orbit.

Put two fully fueled SS in high elliptical orbit., connect them together.  At perigee fully burn one, separate and then burn the other. 

You cannot replace this by burning half the fuel in each then refueling because once you've burned half your two SS are in hyperbolic orbits and are not coming back, so Herr Oberth can't get a crack at the rest of the fuel.*  Also you don't get your booster back.  So this is a rare example of where staging and refueling cannot be made equivalent.

* Unless you can dock and transfer the fuel in a few minutes before leaving the Earths vicinity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/21/2022 09:57 pm
We need to talk about prop losses.  (Disclaimer:  I think this is on-topic, especially to the whole "how big a tank are you refilling from the depot" discussion we've been having.  But most of this has to do with looking at the margins you need for the LSS, which will tell you if 1200t of prop actually works.)

There seem to be four basic mechanisms that render prop in the tanks unusable for generating mission delta-v.  They are:

1) Ullage gas losses.  This is the amount of prop that has to be turned into gas to maintain the flight pressure of the tanks.  With 1200t tanks at O:F = 3.6:1 and flight pressure at 6bar, I worked this out to be 2.4t of GCH4 and 6.3t of GOX.  (Methodology:  figure out the mass of each liquid, divide by its density to get a tank volume for each side, then use PV=nRT and the molecular weight of each compound to derive the mass of gas at the flight pressure.  I used T=300K, CH4=12g/mol, and O2=32g/mol.)

Update:  Aarrrgh!  CH4=16g/mol.  Fortunately, I was smarter when I put it into the spreadsheet than when I wrote this.  No change in the result.

2) Sump losses.  I don't know the proper terminology for this, so the name may be odd.  This is the minimum amount of prop you need to have in the bottom of the tank when the engine shuts down for the last time.  The problem to avoid is that you absolutely, positively must never uncover the turbopump inlets and let ullage gas blow in, or your engines will rip themselves to shreds and get all explodey.  So this is dependent on how much slosh you can get, how much area there is on the prop surface when it's close to the outlet, how violent your attitude control maneuvers are near shutdown, and how extreme your baffling is.  I'd guess that this value is smaller for header tanks than the mains (less area in the header tanks), but I'd also guess that neither the depot nor the LSS has header tanks. Beyond that... 1t (500kg for each main tank)?

3) Boil-off losses.  This is dependent on the environment that the depot or LSS spends most of its time.  If you assume that a depot is more like an LSS than it is a lift tanker, and that we take edzieba's "minimum viable" design constraint to heart, then we probably have something like solar white paint for both depot and LSS.  If both vehicles spend as much time as possible with their noses pointed at the Sun, then depot boil-off is mostly driven by Earthshine (quite a bit), and LSS boil-off is mostly driven by Moonshine (not very much).  However, LSS boil-off is also driven by sitting on the lunar surface for a while, receiving almost-perpendicular sun on one side, and lambertian scattering from the regolith all around.  As a SWAG... 500kg/day for the depot, and (average) 200kg/day for the LSS?

But wait!  There's more!  I would assume that the depot would get topped off fairly close to when the LSS launches, but you have to allow for at least a week between top-off and LSS fueling.  Meanwhile, per the HLS source selection statement, SpaceX is designing the LSS to have a 100 day loiter time.  So depot boil-off loss:  3.5t.  LSS boil-off loss: 20t.

4) Flight performance reserve.  This is the amount of extra prop you need in case your engines underperform, your trajectory goes wacky and needs serious correction, you have an off-nominal RPOD, etc.

This one's a big deal, and I've kinda fallen and can't get up on computing it.  I want to specify this as a percentage, but the big question is whether it should be a percentage of the total prop load, or a percentage of the expected delta-v budget.  Any opinions?

The other big question is what reasonable values are for the depot and the LSS.  If the depot stays in VLEO, its FPR is 0%, but if it boosts into some eccentric orbit to reduce the LSS's delta-v budget for LEOish-NRHO-LS-NRHO, then it'll arrive less than full, and it probably needs... 0.5% FPR?  You also need some amount of prop to allow it to return to VLEO, but the LSS needs to count on a certain amount of prop being in the depot when it gets there.  Note, however, that, even if you use barley's "fill the depot in VLEO and then boost to some degree of eccentricity" approach, you always have the option of sending one last lift tanker to the eccentric orbit to top it off.  If you do that, FPR can be set at effectively zero.

For the LSS, I'm guessing that you need at least 2%, because it's a crew-rated system.  But the "delta-v vs. prop" computation method looms large here.  With a delta-v budget somewhere near 8330m/s for the eccentric refueling option, the 2% FPR by prop mass would remove 24t of prop, while 2% by delta-v (increasing it to 8500m/s) would be closer to 51t of prop.

Now let's crank this through.  Lots of assumptions:

A) Dry mass of the LSS is 95t (120t - TPS - elonerons + landing legs + solar paint), but the crew module + crew + payload adds 20t, for a dry-ish mass of 115t.  1200t prop.  Isp=378s.

B) Ullage gas loss = 8.7t

C) Sump loss = 1.0t

D) Boil-off loss @ 100 day loiter, 200kg/day = 20t.

E) FPR, using the "by prop mass" method @ 2% = 24t.

F) We use whatever tanker magic is required so that the LSS is completely topped off just before TLI.

G) Assume a 300km x 2000km refueling orbit, with a TLI to BLT of C3=-1.5km²/s² (which seems to be what the BLT geeks use).  If I did all the math right, that requires a delta-v budget of 8330m/s to do the complete HLS mission.  (300x200 to BLT=2581m/s, BLT-NRHO=80, NRHO-LS=2855, LS-NRHO=2810.)

All of the various types of losses put 54t of prop off-limits.  Even worse, they effectively become part of the inert mass, which, when added to the dry mass, becomes 169t, while the usable prop on departure becomes 1145t.  That leaves the LSS with an available delta-v of 7600m/s, which isn't nearly enough.

FPR and boil-off are clearly the big-ticket assumptions here.  But note that this still assumes all the operational weirdness of the depot boosting up into a mildly eccentric refueling orbit.  You can obviously go much, much higher than that, but I chose this one because it avoids toasting the electronics on a crew-rated ship in the VA belts, and it's not horrifically tanker-intensive.

For good or ill, Option A is the mission for which the depot will be designed.  Until you understand the demands of its conops, speculation about the depot, however fun it is, has a big hole in it.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/21/2022 10:01 pm
I still think you can reduce dry mass lower than that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/21/2022 10:15 pm
BTW two stage Oberth isn't necessary.  The refuel in elliptical is effectively equivalent to staging.  You've boosted a bunch of fuel to a higher energy, and then use the fuel at maximum kinetic energy/minimum potential energy, taking full advantage of Mr Oberth
Two stage Oberth would be an advanced technique I don't think will happen soon.  It would get you higher v than a single fully fueled ship starting in high eliptical orbit.

Put two fully fueled SS in high elliptical orbit., connect them together.  At perigee fully burn one, separate and then burn the other. 

You cannot replace this by burning half the fuel in each then refueling because once you've burned half your two SS are in hyperbolic orbits and are not coming back, so Herr Oberth can't get a crack at the rest of the fuel.*  Also you don't get your booster back.  So this is a rare example of where staging and refueling cannot be made equivalent.

* Unless you can dock and transfer the fuel in a few minutes before leaving the Earths vicinity.

Half the fuel gets you to about Moon SOI.  Anything beyond that the orbital timing gets wonky.

This was worked out extensively in the extra-solar probe thread, with the nice addition here of simpler rendezvous.

So here it is again:

1.  Fuel two starships to max fuel in LEO, per normal refueling procedures
2.  Burn about half the fuel of each to get an elliptical orbit.  Side by side, seperated by a few km, not connected.
3.  Transfer the fuel from the first starship to the second, giving the second one a full load of fuel.  The nearly empty starship needs landing deltaV and a tiny bit more to get it to enter the atmosphere.  Fully reusable.
4.  The fully fueled starship now has an extra 3kps on a full load of fuel.  Transferring to Moon SOI (for example) requires almost no deltaV - a small burn at apoapsis to orient the elliptical orbit towards the Moon intercept, and a very small burn at periapsis to get to Moon's SOI.   This starship now has enough deltaV to land on the Moon and return to earth's atmosphere, for a hot entry but otherwise normal Starship landing.   Obviates over half of Artemis project...

Full set of calcs here:   

https://docs.google.com/spreadsheets/d/18EEOrUn7T_JUXHeNxYsa4M5896ogaUgBfOaNANfdrck/edit?usp=sharing
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/21/2022 10:34 pm
I still think you can reduce dry mass lower than that.

Sigh.  OK, let's do it, now that I have a loss budget and a delta-v budget.  Let's set total prop to 1200t, Isp=378s.  Prop loss, from above, is 54t.  Also from above, delta-v from 300x2000 is 8330m/s.

MR = exp(8330/378/9.807) = 9.46
MR = (prop + maxDry) / (maxDry + propLoss)
MR(maxDry + propLoss) = prop + maxDry
MR*propLoss = prop + maxDry(1-MR)
maxDry = (MR*propLoss - prop) / (1-MR) = (9.46*54 - 1200) / (1-9.46) = 81.4t.

Think you can get that low?

For the record, I think that managing a depot in an eccentric orbit is nuts.  I admit that you can arbitrarily shave off however much you want from the delta-v budget, but the operational problems are horrific.

Meanwhile, I think we both agree that moving the domes and bulkheads forward, trading some payload bay space for more prop, is easy, and makes the conops stupidly simple.  But if you keep the depot in, say, 300x300, then your delta-v budget is 8720m/s (MR=10.51).  Using the algebra above, and sticking with the same total prop and prop losses, yields a maximum dry mass of 66.5t.  Nope.

Now let's try a VLEO depot with a 1500t prop load.  Prop loss would then be 60t, and max dry mass is 91.4t.  That's still not great, and almost certainly indicates that I'm being too conservative on the prop losses.  But it's at least in the ball park.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/21/2022 10:43 pm
I still think you can reduce dry mass lower than that.

Sigh.  OK, let's do it, now that I have a loss budget and a delta-v budget.  Let's set total prop to 1200t, Isp=378s.  Prop loss, from above, is 54t.  Also from above, delta-v from 300x2000 is 8330m/s.

MR = exp(8330/378/9.807) = 9.46
MR = (prop + maxDry) / (maxDry + propLoss)
MR(maxDry + propLoss) = prop + maxDry
MR*propLoss = prop + maxDry(1-MR)
maxDry = (MR*propLoss - prop) / (1-MR) = (9.46*54 - 1200) / (1-9.46) = 81.4t.

Think you can get that low?

For the record, I think that managing a depot in an eccentric orbit is nuts.  I admit that you can arbitrarily shave off however much you want from the delta-v budget, but the operational problems are horrific.

Meanwhile, I think we both agree that moving the domes and bulkheads forward, trading some payload bay space for more prop, is easy, and makes the conops stupidly simple.  But if you keep the depot in, say, 300x300, then your delta-v budget is 8720m/s (MR=10.51).  Using the algebra above, and sticking with the same total prop and prop losses, yields a maximum dry mass of 66.5t.  Nope.

Now let's try a VLEO depot with a 1500t prop load.  Prop loss would then be 60t, and max dry mass is 91.4t.  That's still not great, and almost certainly indicates that I'm being too conservative on the prop losses.  But it's at least in the ball park.
Yes, I do. That's comparable to the EDL-capable tanker mass.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/21/2022 10:51 pm
maxDry = (MR*propLoss - prop) / (1-MR) = (9.46*54 - 1200) / (1-9.46) = 81.4t.

Think you can get that low?
Yes, I do. That's comparable to the EDL-capable tanker mass.

An EDL-capable tanker is a Starship.  Did I miss something about the 120t baseline being reduced?  And then you need to add the crew module.

I think you can get a non-EDL Starship down to 85t-95t if 120t is the baseline.  And I'm willing to believe that the crew module could be as small as 10t, although that would basically require SpaceX to re-engineer it for Option B.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/21/2022 11:11 pm
Look at your boil off losses over the course of the mission.  The ship is gradually getting lighter, so you need less fuel for the final burns.  This are sensitive to when the loitering, and boiloff, occur.  I don't think this closes your delta V budget, but it should get a few hundred m/s closer.

For example on the final burn from lunar surface to NRHO if 20t of fuel has boiled off then you're 20t lighter, or if it hasn't occurred you've got more fuel.


If you're feeling lucky you could do the same with the FPR.  It gives me the heebie-jeebies, but airlines do it on Tokyo-New York non-stop.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 02:59 am
Look at your boil off losses over the course of the mission.  The ship is gradually getting lighter, so you need less fuel for the final burns.  This are sensitive to when the loitering, and boiloff, occur.  I don't think this closes your delta V budget, but it should get a few hundred m/s closer.

For example on the final burn from lunar surface to NRHO if 20t of fuel has boiled off then you're 20t lighter, or if it hasn't occurred you've got more fuel.


If you're feeling lucky you could do the same with the FPR.  It gives me the heebie-jeebies, but airlines do it on Tokyo-New York non-stop.

I agree that might reduce the reserves a bit.  ISTM that you'd want to look at the timing of your burns, the environment the ship is in before each one, and compute boil-off piecewise.

Probably the same thing with FPR:  Figure out the possible underperformance on each burn, the consequences of underperformance (which could be anything from no big deal to a mission abort to a loss of crew), and then figure them out piecewise again.  That would probably militate more toward the "percentage of delta-v" method.

It's a messy spreadsheet, though.

On a related topic:  I've been working off of a set of delta-v numbers from an old trade study that NASA did when evaluating the early HLS architectures.  The slide with most of the juicy numbers is attached.  It occurred to me, however, that I don't know if these numbers already have FPR baked into them or not.  Does anybody know?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/22/2022 04:44 am
Look at your boil off losses over the course of the mission.  The ship is gradually getting lighter, so you need less fuel for the final burns.  This are sensitive to when the loitering, and boiloff, occur.  I don't think this closes your delta V budget, but it should get a few hundred m/s closer.

For example on the final burn from lunar surface to NRHO if 20t of fuel has boiled off then you're 20t lighter, or if it hasn't occurred you've got more fuel.


If you're feeling lucky you could do the same with the FPR.  It gives me the heebie-jeebies, but airlines do it on Tokyo-New York non-stop.
Yup, and boil-off losses can actually be used propulsively. Lower Isp and thrust, but not a total waste.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/22/2022 04:48 am
maxDry = (MR*propLoss - prop) / (1-MR) = (9.46*54 - 1200) / (1-9.46) = 81.4t.

Think you can get that low?
Yes, I do. That's comparable to the EDL-capable tanker mass.

An EDL-capable tanker is a Starship.  Did I miss something about the 120t baseline being reduced? ...
Yeah, some of the early figures for ITS/BFR/Starship show like 85t dry mass for an EDL capable tanker. Earliest ships aren't going to get there, but by the time HLS happens, it's possible especially without the heatshield or flaps (or heavy actuators and batteries for that).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 10:30 am
Yeah, some of the early figures for ITS/BFR/Starship show like 85t dry mass for an EDL capable tanker. Earliest ships aren't going to get there, but by the time HLS happens, it's possible especially without the heatshield or flaps (or heavy actuators and batteries for that).

The 85t ITS was based on an abandoned structural technology. Elon’s talked about getting vanilla Starship down to 100t eventually, which might get an LSS down to 75t-80t, but you’d still have to add in a crew module.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/22/2022 02:14 pm
Yeah, some of the early figures for ITS/BFR/Starship show like 85t dry mass for an EDL capable tanker. Earliest ships aren't going to get there, but by the time HLS happens, it's possible especially without the heatshield or flaps (or heavy actuators and batteries for that).

The 85t ITS was based on an abandoned structural technology. Elon’s talked about getting vanilla Starship down to 100t eventually, which might get an LSS down to 75t-80t, but you’d still have to add in a crew module.
The 85t ITS tanker was also much, much larger and included TPS and aero surfaces.

I’m just pointing out there are three ways of addressing the delta-v requirements, not just your one idea of stretching the tanks. They can also try reducing dry mass or operational changes like when/where they refuel. Don’t get too married to your own concept.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/22/2022 02:39 pm
Yup, and boil-off losses can actually be used propulsively. Lower Isp and thrust, but not a total waste.
Please explain.  Are you suggesting using the boiloff as it occurs or compressing and storing for later?

The former would mean you have to accurately predict when the boiloff occurs and require interesting trajectory planning.  And what do you do when loitering at the gateway, which may be most of the boiloff?

The later requires extra equipment and heat management that might be better employed as a chiller.

You can use a little of the boiloff for attitude control and fine tuning, but I can't see how you can use a large fraction of it.



Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/22/2022 03:46 pm
Yup, and boil-off losses can actually be used propulsively. Lower Isp and thrust, but not a total waste.
And what do you do when loitering at the gateway, which may be most of the boiloff?
I think the intention is to coat the vehicle with something like Solar White (https://iopscience.iop.org/article/10.1088/1757-899X/1240/1/012001/pdf), which should have an equilibrium temperature of 88 K. The idea is that the vehicles will be passively cooled. At least to start with.

Quote
If a coating can be fabricated with about a 0.01 α/ε ratio and if this is placed on a sphere at uniform temperature, far from other heat sources, then the sphere will come to a steady state temperature of about 88 K, sufficient to passively store liquid oxygen. One reason this works is the sphere emits infrared from its entire surface, which is 4 times larger than its cross-sectional area, which determines the amount of solar power absorbed.

Development of a thermal control coating optimized for cryogenic space applications; A Krenn et al; 2022; IOP Conf. Ser.: Mater. Sci. Eng. 1240 012001
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 04:12 pm
Yeah, some of the early figures for ITS/BFR/Starship show like 85t dry mass for an EDL capable tanker. Earliest ships aren't going to get there, but by the time HLS happens, it's possible especially without the heatshield or flaps (or heavy actuators and batteries for that).

The 85t ITS was based on an abandoned structural technology. Elon’s talked about getting vanilla Starship down to 100t eventually, which might get an LSS down to 75t-80t, but you’d still have to add in a crew module.
The 85t ITS tanker was also much, much larger and included TPS and aero surfaces.

I’m just pointing out there are three ways of addressing the delta-v requirements, not just your one idea of stretching the tanks. They can also try reducing dry mass or operational changes like when/where they refuel. Don’t get too married to your own concept.

I've been restricting my thinking to the relatively near-term, more specifically to the Artemis timeframe.  That marries me pretty strongly to concepts that can be implemented using more-or-less today's Starship or simple variants of it.  IMO, tank-stretching (or, more accurately, tank rearrangement, to swap payload space for more prop space without changing the outer mould line) is a simple variant.  Converting most of the ship to carbon composites is not.

Refueling in eccentric orbits should definitely be on the table.  But they come with a large amount of operational complexity.  And if you want to shave thousands, not hundreds, of m/s off the delta-v budget, then you incur even more operational complexity, and some radiation risk to your equipment to boot.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 04:30 pm
And what do you do when loitering at the gateway, which may be most of the boiloff?

Note that loitering at the gateway is different than loitering near the gateway.  In the former, the gateway constrains your attitude control.  In the latter, the LSS is free to point its nose at the Sun, which is likely to be the single most significant reduction of thermal flux available to you.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/22/2022 04:38 pm
Has anyone discussed the possibility of putting a depot in NRHO next to Gateway? I was wondering what SpaceX might do with the lunar Starships once they'd completed their missions, and it occurred to me that the only way to make them be reusable would be if they could refuel at Gateway. Also, since Gateway is nicely distant from both Earth and the moon, you ought to be able to use Solar White to achieve zero boiloff (or close to it). Finally, if it turns out a lunar Starship (freshly arrived from Earth) needed a bit more fuel to get to the moon and back, this would be the place to get it.

If a depot had dry mass of 100 tons (metric) and held 1800 tons of fuel, then assuming ∆v=3650 m/s from LEO to NRHO and specific impulse of 3750 m/s (382 s), I figure it would arrive at NRHO with about 617 tons of fuel. If ∆v=520 m/s for TEI and all other deceleration into LEO can be done via aerocapture, then the depot only needs 15 tons of fuel to return to LEO, so it could dispense 600 tons of fuel at NRHO and still get back to Earth to refuel.

If ∆v=5620 m/s to go from NRHO to the moon and back, I figure 600 tons of fuel would support a 72-ton payload. That doesn't seem too shabby.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/22/2022 04:54 pm
Note that loitering at the gateway is different than loitering near the gateway.  In the former, the gateway constrains your attitude control.  In the latter, the LSS is free to point its nose at the Sun, which is likely to be the single most significant reduction of thermal flux available to you.
I suppose tying them together with a long cable would be a bad idea . . . :-)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/22/2022 05:07 pm
And what do you do when loitering at the gateway, which may be most of the boiloff?

Note that loitering at the gateway is different than loitering near the gateway.  In the former, the gateway constrains your attitude control.  In the latter, the LSS is free to point its nose at the Sun, which is likely to be the single most significant reduction of thermal flux available to you.

This brings into question the methodology (but not necessarily the result) of your estimate of boiloff.  It should probably be something like x kg/day loitering in earth orbit, y kg/day docked to gateway z tonne during lunar orbit/landing rather than a flat 200kg/day.  I realize it's a SWAG, but even a SWAG can be influence by mechinism.  It also makes some aspects of mission planning easier if for example almost all the boiloff occurs on the lunar surface.

3) Boil-off losses.  This is dependent on the environment that the depot or LSS spends most of its time.  If you assume that a depot is more like an LSS than it is a lift tanker, and that we take edzieba's "minimum viable" design constraint to heart, then we probably have something like solar white paint for both depot and LSS.  If both vehicles spend as much time as possible with their noses pointed at the Sun, then depot boil-off is mostly driven by Earthshine (quite a bit), and LSS boil-off is mostly driven by Moonshine (not very much).  However, LSS boil-off is also driven by sitting on the lunar surface for a while, receiving almost-perpendicular sun on one side, and lambertian scattering from the regolith all around.  As a SWAG... 500kg/day for the depot, and (average) 200kg/day for the LSS?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 05:10 pm
Has anyone discussed the possibility of putting a depot in NRHO next to Gateway? I was wondering what SpaceX might do with the lunar Starships once they'd completed their missions, and it occurred to me that the only way to make them be reusable would be if they could refuel at Gateway. Also, since Gateway is nicely distant from both Earth and the moon, you ought to be able to use Solar White to achieve zero boiloff (or close to it). Finally, if it turns out a lunar Starship (freshly arrived from Earth) needed a bit more fuel to get to the moon and back, this would be the place to get it.

If a depot had dry mass of 100 tons (metric) and held 1800 tons of fuel, then assuming ∆v=3650 m/s from LEO to NRHO and specific impulse of 3750 m/s (382 s), I figure it would arrive at NRHO with about 617 tons of fuel. If ∆v=520 m/s for TEI and all other deceleration into LEO can be done via aerocapture, then the depot only needs 15 tons of fuel to return to LEO, so it could dispense 600 tons of fuel at NRHO and still get back to Earth to refuel.

If ∆v=5620 m/s to go from NRHO to the moon and back, I figure 600 tons of fuel would support a 72-ton payload. That doesn't seem too shabby.

Yeah, there are discussions about his up-thread.  A few things to note:

1) If your operational act is together, then just-in-time delivery of prop for missions doesn't require a depot.  Just send a lift tanker that's been filled up before departing to cislunar.  When it's filled the target vehicle, it can return straight to EDL.

2) Even if you need a depot, either to provide some operational breathing room for delays, or because you're filling up lots of small missions over an extended period of time, you still need the lift tankers bringing the prop out to cislunar.    So the capability for tanker direct EDL from cislunar speeds is still essential.  Otherwise, you'll slowly accumulate a fleet of dead, useless tankers.

3) I agree that repurposing a few LSSes as depots is an interesting use case.  However, note that the LSS will need to be refitted with a "depot kit" to make this work.  This is yet another reason why development of such a kit makes a lot of sense, and why being able to slap it on any random Starship will be very handy.

4) Finally, note that the age of LSSes that loiter in cislunar between missions is likely to be short.  It makes much, much more sense to stage LSSes from LEO, where their crews can be ferried up via any ol' commercial crew system, rather than a $4.2B/launch white elephant.  This requires refueling in cislunar with the crew present after they've ascended from the surface, but that seems like something that should be on a steep path for risk reduction. 

That may or may not require depots, depending on how comfy everybody gets with just-in-time lift tankers ferrying stuff around.  Of course, it doesn't take very many cislunar depots to meet demand, so maybe a couple of Option A's and an Option B or two will do the trick, at least until they die of old age.  After that, there will likely be a better solution.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 05:16 pm
Note that loitering at the gateway is different than loitering near the gateway.  In the former, the gateway constrains your attitude control.  In the latter, the LSS is free to point its nose at the Sun, which is likely to be the single most significant reduction of thermal flux available to you.
I suppose tying them together with a long cable would be a bad idea . . . :-)

The temptation to spin the system up so you could fling the Gateway into the Sun might become overwhelming.

This brings into question the methodology (but not necessarily the result) of your estimate of boiloff.  It should probably be something like x kg/day loitering in earth orbit, y kg/day docked to gateway z tonne during lunar orbit/landing rather than a flat 200kg/day.  I realize it's a SWAG, but even a SWAG can be influence by mechinism.  It also makes some aspects of mission planning easier if for example almost all the boiloff occurs on the lunar surface.

Yes, this is what I was alluding to up-thread.  Piecewise estimates are the way to go, but they require extensive spreadsheet-smithing.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/22/2022 07:34 pm
Even if you need a depot, either to provide some operational breathing room for delays, or because you're filling up lots of small missions over an extended period of time, you still need the lift tankers bringing the prop out to cislunar.    So the capability for tanker direct EDL from cislunar speeds is still essential.  Otherwise, you'll slowly accumulate a fleet of dead, useless tankers.
Since it takes so little ∆v for a Starship to return from NRHO (assuming aerocapture is viable), I don't think there's a lot of excuse for accumulating a fleet of dead Starships there--of any flavor.

I could certainly see using tankers to refill the NRHO depot, but it bothers me that they've got all that extra mass for EDL. Of course it would also bother me to create a special kind of tanker that doesn't have that mass and which simply shuttles between LEO and NRHO, but maybe that's really the best way to do this.

So you'd end up with:
1) A fleet of regular tankers that ferry fuel from Earth to an LEO depot.
2) A fleet of "lunar" tankers that ferry fuel from the LEO depot to an NRHO depot.

A new lunar Starship would launch from Earth, refuel at the LEO depot, fly to NRHO, refuel again (if needed), go to the moon and back, and refuel at the LEO depot--either to return to LEO or ferry another mission to the lunar surface.

This raises (for me) a big question: how will SpaceX service vehicles that cannot return to Earth? If they decide a lunar Starship needs to have an engine replaced, how would they do that? (This is probably a good subject for another thread, though--if it doesn't already exist.)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/22/2022 07:38 pm

A new lunar Starship would launch from Earth, refuel at the LEO depot, fly to NRHO, refuel again (if needed), go to the moon and back, and refuel at the LEO depot--either to return to LEO or ferry another mission to the lunar surface.

Thanks for the succinct summary of the process.

How is this less complicated than an elliptical orbit refuel where both a full tanker and a full lander burn half fuel each into the elliptical trajectory side by side (but not connected) and then fill the lander with an Oberth burn for the lander?  (aka the "three burn method")?

It sounds more complicated, not less.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 08:20 pm
Since it takes so little ∆v for a Starship to return from NRHO (assuming aerocapture is viable), I don't think there's a lot of excuse for accumulating a fleet of dead Starships there--of any flavor.

I could certainly see using tankers to refill the NRHO depot, but it bothers me that they've got all that extra mass for EDL. Of course it would also bother me to create a special kind of tanker that doesn't have that mass and which simply shuttles between LEO and NRHO, but maybe that's really the best way to do this.

You're fairly far away from "minimum viable product" territory here.  Unless a bare-metal non-EDL ship can actually aerobrake (remember: no aerosurfaces either), then you have to do propulsive re-insertion to LEO, and that is expensive.  There is some possibility that you can get an LSS-style system to survive aerobraking 30m/s at a time, which gets it back to LEO in about 3 months.  But it's far from a done deal.

A little reality check, using structural mass coefficent ε=dryMass/(dryMass+propMass) and payload ratio λ=payload/(dryMass+propMass).

Let's assume 95t dry for LSS-style ships, 120t for vanilla, all with 1200t tankage, all with Isp=378s.  All outbound trips are via BLT, but all returns are via fast, simply because I don't understand return via BLT, and because I'm too lazy to update my delta-v table.  I'm also going to assume 20t blanket prop loss, again out of laziness.

LSS-style Starship with propulsive return to LEO:
Transferrable prop to NRHO: 223t.
ε=7.3%
λ=17.2%

LSS-style Starship, magical aerobraking return to LEO:
Transferrable prop: 401t
ε=7.3%
λ=31.0%

Vanilla Starship, direct to EDL:
Transferrable prop: 373t
ε=9.1%
λ=28.2%

Propulsive return to LEO kills you, but the difference between a vanilla Starship that's probably pretty easy to get working for direct EDL and a bare-metal LSS that does multi-pass aerobraking via mythical attitude control really isn't that big.

Quote
This raises (for me) a big question: how will SpaceX service vehicles that cannot return to Earth? If they decide a lunar Starship needs to have an engine replaced, how would they do that? (This is probably a good subject for another thread, though--if it doesn't already exist.)

Most likely answer:  they won't.  This is a major reason why I'm extremely fond of tankers (and depots!) that can do direct EDL.  It's also a pretty awesome reason to move away from LSS as quickly as possible to a direct-EDL crew system, but that's way O/T. 
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/22/2022 08:44 pm
You're fairly far away from "minimum viable product" territory here.
But it was such a short walk! ;-)

LSS-style Starship, magical aerobraking return to LEO:
Transferrable prop: 401t
ε=7.3%
λ=31.0%

Vanilla Starship, direct to EDL:
Transferrable prop: 373t
ε=9.1%
λ=28.2%

Propulsive return to LEO kills you, but the difference between a vanilla Starship that's probably pretty easy to get working for direct EDL and a bare-metal LSS that does multi-pass aerobraking via mythical attitude control really isn't that big.
Yeah, if the dry mass differences are that small, I agree there's not much reason to favor the one over the other, in which case, the one that can EDL is the one to go with.

Quote
This raises (for me) a big question: how will SpaceX service vehicles that cannot return to Earth? If they decide a lunar Starship needs to have an engine replaced, how would they do that? (This is probably a good subject for another thread, though--if it doesn't already exist.)

Most likely answer:  they won't.  This is a major reason why I'm extremely fond of tankers (and depots!) that can do direct EDL.  It's also a pretty awesome reason to move away from LSS as quickly as possible to a direct-EDL crew system, but that's way O/T. 
Makes sense. I don't see any way in the near term that SpaceX could arrange to do in-orbit maintenance and repairs. Doing spacewalks seems to be ridiculously difficult because of the clumsy spacesuits, remote-operated equipment seems too restrictive, and building a huge space dock that can hold a Starship and yet be pressurized seems way out of budget.

Still not sold on depots that do EDL, but, in that case, they shouldn't use their engines again once they're in their final orbits.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 08:46 pm

A new lunar Starship would launch from Earth, refuel at the LEO depot, fly to NRHO, refuel again (if needed), go to the moon and back, and refuel at the LEO depot--either to return to LEO or ferry another mission to the lunar surface.

Thanks for the succinct summary of the process.

How is this less complicated than an elliptical orbit refuel where both a full tanker and a full lander burn half fuel each into the elliptical trajectory side by side (but not connected) and then fill the lander with an Oberth burn for the lander?  (aka the "three burn method")?

It sounds more complicated, not less.

HEEO refuelings require a (best case) extra pass through the Van Allen Belts, which is a big deal for a crew.¹

I don't think the whole "tandem burn" thing works, because if the crew burn succeeds and the tanker burn aborts, that's a loss of mission.  (You can't leave the crew in the VA Belts while you debug the tanker, and they have to return to LEO.)

You can put the tanker/depot in the HEEO well before committing the crew to launch, but there are problems there:

1) The tanker/depot will have to endure tens if not hundreds of passes through the VA belts, which will be hard on the electronics.

2) If the crew refuels but has a glitch at TLI, how many orbits can the crew endure before they've blown through their radiation budget?  (My guess:  one.)

3) Even with a one-orbit delay, the argument of perigee for the HEEO is substantially sub-optimal the next time around, so your delta-v budget for the LSS has to accommodate that.  And even if you allowed multi-orbit stays in HEEO for the crew, after the first orbit, things go downhill very quickly.

HEEOs are a pain.

NRHO is no picnic, but it's outside the VA belts, so the radiation environment, while not good, is considerably better.  And while your windows to get to and from NRHO are far from infinite, I suspect there are pretty substantial games you can play with lunar flyby to make a fairly wide window in both directions.  (This is a hunch on my part; can anybody confirm?)

____________________
¹For cargo-only missions, unless the retail tanker launch cost is well below $10M, it's cheaper to use an expendable cargo LSS that stays on the surface.

PS:  If you're refueling in NRHO, doing it before lunar descent is much more expensive in terms of prop than doing it after ascent.  Indeed it's so expensive that you likely need two tankers in NRHO, not one.

On the other hand, it's easy and safe to abort back to LEO if a pre-descent refueling goes wrong.  Post-ascent, you have yourself a life-threatening problem.  However, you can mitigate that risk if the Gateway is available.  (This is one of the only things I like about having a Gateway...)
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/22/2022 08:49 pm
How is this less complicated than an elliptical orbit refuel where both a full tanker and a full lander burn half fuel each into the elliptical trajectory side by side (but not connected) and then fill the lander with an Oberth burn for the lander?  (aka the "three burn method")?

It sounds more complicated, not less.
It's definitely more complicated, but the system you're talking about seems very limited. For example, it offers no way to refuel a lander for reuse/return.

Also, I seem to have forgotten what the point of it was. If you've already got a fully fueled lander, doesn't that already have enough fuel to complete the mission? What's the point of the tanker?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/22/2022 09:18 pm
Also, I seem to have forgotten what the point of it was. If you've already got a fully fueled lander, doesn't that already have enough fuel to complete the mission? What's the point of the tanker?

1200t of prop can't do VLEO-NRHO-LS-NRHO on one tank of gas.  HEEO-NRHO-LS-NRHO can, but it's a pain.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/22/2022 09:22 pm
Also, I seem to have forgotten what the point of it was. If you've already got a fully fueled lander, doesn't that already have enough fuel to complete the mission? What's the point of the tanker?

1200t of prop can't do VLEO-NRHO-LS-NRHO on one tank of gas.  HEEO-NRHO-LS-NRHO can, but it's a pain.
…unless you reduce the dry mass somehow. It’s the mass ratio and Isp that determine delta-v.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/22/2022 10:22 pm

A new lunar Starship would launch from Earth, refuel at the LEO depot, fly to NRHO, refuel again (if needed), go to the moon and back, and refuel at the LEO depot--either to return to LEO or ferry another mission to the lunar surface.

Thanks for the succinct summary of the process.

How is this less complicated than an elliptical orbit refuel where both a full tanker and a full lander burn half fuel each into the elliptical trajectory side by side (but not connected) and then fill the lander with an Oberth burn for the lander?  (aka the "three burn method")?

It sounds more complicated, not less.
I don't think the whole "tandem burn" thing works, because if the crew burn succeeds and the tanker burn aborts, that's a loss of mission.  (You can't leave the crew in the VA Belts while you debug the tanker, and they have to return to LEO.)

You leave enough fuel reserve to enter atmosphere and land normally.  That's not loss of mission, merely an abort of mission.   The periapsis only needs to be dropped 200km in order to get to atmosphere reeentry, that's a trivial amount of fuel when the burn is done at apoasis on a HEEO.

There are equally a number of pathways to mission abort in the other scenarios being talked about.  Possibly more.

I agree it's one extra pass through the Van Allen belts.  I wonder if the 50T of extra cargo space that NASA doesn't need would allow mass to make up for that or not.   Especially if it is useful cargo like water.  200 square meters of crew surface area means a water depth of 0.25 meters all around the crew area.

Here's a study on that:  https://digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=3795&context=theses
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/23/2022 03:43 am
I don't think the whole "tandem burn" thing works, because if the crew burn succeeds and the tanker burn aborts, that's a loss of mission.  (You can't leave the crew in the VA Belts while you debug the tanker, and they have to return to LEO.)

You leave enough fuel reserve to enter atmosphere and land normally.  That's not loss of mission, merely an abort of mission.   The periapsis only needs to be dropped 200km in order to get to atmosphere reeentry, that's a trivial amount of fuel when the burn is done at apoasis on a HEEO.

It's an LSS; it can't reenter.  But it should have enough prop on board to propulsively lower the apogee back to the original insertion LEO.  Or you can leave it where it is while you debug the tanker, but then it's soakin' up the rads.

We can argue semantics about whether "mission abort" = "loss of mission".  You haven't lost the vehicle, which is different from your garden-variety LoM.  Let's say "major contingency likely involving at least one additional tanker launch".  It's not great.  Not a disaster, but not great.

Quote
There are equally a number of pathways to mission abort in the other scenarios being talked about.  Possibly more.

This one is avoidable, though:  simply boost the tanker first, and have it waiting for the LSS when it boosts later.  No need for this tandem boost stuff.  Same argument applies for higher, more eccentric orbits.

Quote
I agree it's one extra pass through the Van Allen belts.  I wonder if the 50T of extra cargo space that NASA doesn't need would allow mass to make up for that or not.   Especially if it is useful cargo like water.  200 square meters of crew surface area means a water depth of 0.25 meters all around the crew area.

Here's a study on that:  https://digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=3795&context=theses

Given that the way we went down this rabbit hole in the first place was arguing about whether an LSS with 1200t of tankage was adequate to do the LEO-NRHO-LS-NRHO mission on one tank, throwing an extra 50t of water at the problem--and making your crew module independently-hulled in the process, with something that could leak between it and the fairing--doesn't seem like a step in the right direction.

In terms of the early Artemis conops, you don't need this shielding, because the LSS is uncrewed before it gets to NRHO.  SpaceX could certainly harden the avionics to be VA Belt-proof for a lot less than 50t.  But that (like a water-shielded crew module) is more complexity, and less in common between the LSS and the vanilla Starship.

I'd much rather use that extra payload bay space and fill it up with lovely, lovely propellant.  That is, admittedly, a structural change, but it should be a surprisingly easy one, given how Starship is constructed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/23/2022 04:45 am

It's an LSS; it can't reenter.  But it should have enough prop on board to propulsively lower the apogee back to the original insertion LEO.  Or you can leave it where it is while you debug the tanker, but then it's soakin' up the rads.

I'd much rather use that extra payload bay space and fill it up with lovely, lovely propellant.  That is, admittedly, a structural change, but it should be a surprisingly easy one, given how Starship is constructed.

Sorry, I was dropping all the way back to "just use standard Mars-capable Starships to land on the moon", no special LSS, so aerobraking is back on the table.  This is how powerful the Oberth effect is.

There's also enough spare deltaV to handle 50T of water leaving 50T of payload.  No super-light ships.  And with the standard 1200T fuel (though one extra ring of fuel might be just as nice to have as it would be for Mars).

Adding custom things is yet another complication one can avoid.  Even the 50T of water makes me cringe, though I have to wonder how radiation for 3 months to Mars compares to to an extra pass through the Van Allen belts.   The problem might be similar.

The whole LSS, tanker, and refueling in Lunar orbit seems far more complicated than using Mars-capable stock Starships and stock LEO fueling ships.

Almost any time one adds a custom thing and finds more things turning custom as a result, the system design is being done wrong.   I can't put that in numbers, just 30+ years of system design experience.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/23/2022 06:41 am

I don't think the whole "tandem burn" thing works, because if the crew burn succeeds and the tanker burn aborts, that's a loss of mission.  (You can't leave the crew in the VA Belts while you debug the tanker, and they have to return to LEO.)

They can be in fairly loose formation.  You can almost certainly stagger them by tens of seconds, enough to make sure the tanker burn starts before committing the crew, without taking a significant performance hit on the rendezvous.

But there's lots of things that can go wrong in any system.  If you had 2000 tonne of fuel in a single vehicle you'd still have the problem of what happens if your engines shut down.  Why is it so different just because half your engines are on a different ship?

In my experience people are bad at predicting what is likely to fail and tend to fixate on a just so story.  A lot of hand wringing should be retired during early testing.  If the engines don't start on cue it will show up on reentry and landing.

Also an (aerobraking) starship can do LEO-NRHO-LEO without refueling.  So you could send the crew direct separately from the lander.  It would still be a just a bit cheaper than SLS.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 09/23/2022 08:28 am
But there's lots of things that can go wrong in any system.  If you had 2000 tonne of fuel in a single vehicle you'd still have the problem of what happens if your engines shut down.  Why is it so different just because half your engines are on a different ship?
Because if you have all your propellant on one vehicle, if half the engines shut down the other half can burn for twice as long (slightly more, using some margin, due to reduction in Oberth gains). If you have two vehicles, if half your engines fail then half your propellant is left behind and completing the burn is impossible.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Bryan Hayward on 09/23/2022 02:11 pm
...snip...

Makes sense. I don't see any way in the near term that SpaceX could arrange to do in-orbit maintenance and repairs. Doing spacewalks seems to be ridiculously difficult because of the clumsy spacesuits, remote-operated equipment seems too restrictive, and building a huge space dock that can hold a Starship and yet be pressurized seems way out of budget.

Still not sold on depots that do EDL, but, in that case, they shouldn't use their engines again once they're in their final orbits.
Agreed, and if a depot doesn't EDL, why not fit it with the 6 engines that it will use in space (I assume RVacs) and use one at a time until it fails, then use another? I'm guessing that will make it last a long time. Or, if it is better to rotate them, do that. Spread the "hours used" out as much as possible. When it is finished, land/splat it on Luna as raw material for whatever the base needs.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/23/2022 03:05 pm
Agreed, and if a depot doesn't EDL, why not fit it with the 6 engines that it will use in space (I assume RVacs) and use one at a time until it fails, then use another? I'm guessing that will make it last a long time. Or, if it is better to rotate them, do that. Spread the "hours used" out as much as possible. When it is finished, land/splat it on Luna as raw material for whatever the base needs.
At this point, I'm pretty sure the depots will just be modified versions of the basic Starship where they've removed the cargo area and stretched the tanks about as far as they can go. Since the basic Starship holds 1200 tons of propellant, it's a fair guess that a depot could hold 1800-2000. (I haven't seen anyone seriously try to estimate this yet though.) Dry mass ought to be lower than the 100 tons for a basic Starship, but I haven't seen a serious estimate for how much lower it could be. Maybe 80 tons?

Since it's really just a Starship, it'll launch like any other Starship, which means it'll need those engines (all six of them) to get into orbit. If it's going to be an LEO depot, you won't need the engines again; any propulsion after that will be from the ullage engines. If it's going to NRHO, then it'll need to get refilled by tankers and then it'll use its engines again to reach NRHO. Again, after that, it shouldn't ever use the big engines again.

After thinking about it, I've reached the point where I'm okay with the idea of using tankers capable of EDL to fill a depot at NRHO rather than trying to get depots to shuttle back and forth. The ability to service them on the ground is clearly more important than being able to lug a bit more fuel out to NRHO.

So what to do with the used lunar Starships? Well, here's an idea: if you've got a fuel depot in NRHO, you don't need a lunar Starship; just use a regular one (designed for crew). A basic Starship could launch to LEO, fill up from the LEO depot, head to NRHO, refill from the NRHO depot, go to the moon and back, and (I think) still have enough fuel to return to Earth without needing an additional refill.

That would be very neat and clean; just four Starship variants: cargo, crew, tanker, and depot. And only depot can't do EDL.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/23/2022 03:46 pm
But there's lots of things that can go wrong in any system.  If you had 2000 tonne of fuel in a single vehicle you'd still have the problem of what happens if your engines shut down.  Why is it so different just because half your engines are on a different ship?
Because if you have all your propellant on one vehicle, if half the engines shut down the other half can burn for twice as long (slightly more, using some margin, due to reduction in Oberth gains). If you have two vehicles, if half your engines fail then half your propellant is left behind and completing the burn is impossible.
Each ship still has six engines.  Your down to binomial calculations comparing say four engines failing on one ship v. two failing on each ship.

In any case the proper comparison is a multi-stage rocket, the stages are not redundant.  That is done all the time.  Refueling is being held to a higher standard in things that have nothing to do with the refueling.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/23/2022 03:54 pm
HEEO refuelings require a (best case) extra pass through the Van Allen Belts, which is a big deal for a crew.¹

I looked it up.  A pass through the Van Allen belts was 1.1mSv on the lightly shielded Apollo.   It was a little less than half of the total dose the Apollo astronauts received.

Time in deep space (e.g. to a journey Mars) is 1 to 1.5mS/day  (unprotected)

So whatever type of ship has humans in it will need extensive shielding to get to Mars.   Might as well reuse that for going to the Moon.

An LRHO tanker is needed solely for the Artemis architecture, which said architecture will be obsolete in 5-10 years when Mars landers are being mass produced.   In 10 years going to the Moon won't require fancy custom craft, just a refuel in HEEO and a standard Mars lander.    In a quirk of astrophysics (no air on Moon) it will require more fuel than going to Mars.

sources: 

https://www.lehman.edu/academics/education/middle-high-school-education/documents/mars.pdf
https://www.nasa.gov/sites/default/files/files/SMIII_Problem7.pdf
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 09/23/2022 03:55 pm
But there's lots of things that can go wrong in any system.  If you had 2000 tonne of fuel in a single vehicle you'd still have the problem of what happens if your engines shut down.  Why is it so different just because half your engines are on a different ship?
Because if you have all your propellant on one vehicle, if half the engines shut down the other half can burn for twice as long (slightly more, using some margin, due to reduction in Oberth gains). If you have two vehicles, if half your engines fail then half your propellant is left behind and completing the burn is impossible.
Two ships could be a crew ship and an accumulator ship still locked together after refueling. In this particular case if half of the engines were lost on one ship, the other ship could pump fuel to it (if properly equipped in advance).
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/23/2022 06:00 pm

It's an LSS; it can't reenter.  But it should have enough prop on board to propulsively lower the apogee back to the original insertion LEO.  Or you can leave it where it is while you debug the tanker, but then it's soakin' up the rads.

Sorry, I was dropping all the way back to "just use standard Mars-capable Starships to land on the moon", no special LSS, so aerobraking is back on the table.  This is how powerful the Oberth effect is.

It would be really helpful if everybody would describe the use case(s) they're discussing for refueling.  I'd like to propose some reference missions we need to cover in the near and mid terms, in chronological order:

1) Option A:
Timeframe:  2023-2026
Rationale:  NASA ordered it.
Rules:
a) Uses LSS (i.e., Starship variant with no EDL capability, thermal improvements for medium-term prop storage, and lunar landing capability.)
b) One refueling for entire mission from Earth orbit, with no crew during refueling.  (Clarification:  one refueling from the depot to the LSS.  The depot obviously receives multiple lift tankers.)
c) 90 day loiter in NRHO. 
d) 5 days on lunar surface.
e) Crew-certified only for NRHO-LS-NRHO.
f) Crew of 2.
g) Total delta-v from 300x300: 8965m/s, but can be reduced through use of higher-energy eccentric orbits for refueling.

2) Option B:
Timeframe: 2025-2028ish
Rationale:  NASA solicitation for sustainable HLS.
Rules: 
a) Uses LSS. 
b) Same as Option A for first mission.
c) Subsequent missions are refueled in NRHO, and require NRHO-LS-NRHO in one fueling. 
d) Still no crew on board during refueling. 
e) Up to two week stays on surface.  Loiter time in NRHO is... the same?
f) Crew of 4.
g) Must dock with Gateway, and assumes its existence.
h) Total delta-v from NRHO: 5665m/s

3) LSS staged from Earth orbit.
Timeframe: 2025-2030ish
Rationale: Eliminates need for SLS/Orion, but can be implemented before Starship is crew-certified for launch and EDL.
Rules: 
a) Crewed mission launched on commercial crew system, rendezvous with LSS in LEO
b) No HEEO rendezvous, due to limitations of CCP systems and radiation.
c) LEO-NRHO-LS-NRHO-refuel-LEOpropulsive.  Delta-v: 13,040m/s from 300x300
d) Alternate conops: LEO-LS-NRHO-refuel-LEOpropulsive.  Delta-v: 12,485m/s
e) No aerobraking, because it's an LSS.
f) Requires cislunar refueling with crew on board (unless they move to the Gateway).
g) Requires two RPODs with CCP system in LEO.
h) No requirement for long loiter in cislunar.
i) Other requirements probably Option B.
j) Crew of at least 4, but could be more.

4) Two LSS with separate transit and lunar access legs.
Rationale: Same as #3, but to avoid crewed refueling in cislunar, one LSS does LEO-NRHO-LEOpropulsive, while one does NRHO-LS-NRHO (essentially Option B).
Rules:
a) No refueling with crew on board.
b) Must support LSS-to-LSS docking.
c) Otherwise, see #3.

5) Starship (not LSS), crew-certified for launch/EDL, lunar mission:
Timeframe: 2027ish to end of Starship life
Rationale: Once Starship is crew-certified for launch and EDL, this is the way to go.
Rules:
a) Probably two crewed refuelings, one in VLEO and one in either HEEO or some cislunar orbit.
b) Crew at least 10.
c) Must be able to dock with and service Gateway, but need not include NRHO in all missions.
d) LEO-LS-LEO delta-v: 9174m/s from VLEO.

6) Earth to Mars from Earth orbit:
Timeframe:  2030ish-end of Starship life
Rationale:  Mars exploration, followed by Mars colonization.
Rules:
a) Amorphous at best.
b) Crew launch from Earth and EDL on Mars using Starship.
c) Short enough transit to be tolerable.
d) Either direct EDL at Mars, or aerocapture followed by EDL.
e) Delta-v highly variable, traded with transit time.  With a full 1200t tank from VLEO, it looks like a 100-day transit is possible in 2031 window.
f) Crew of at least 10 to start, ramping up to about 100.

Ideally, you could design a depot and/or lift tanker to support all of these.  However, given the timeframe, I'd expect depots to become considerably more sophisticated--and larger--before crewed flights to Mars begin.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Ionmars on 09/23/2022 06:16 pm
Thanks for laying out the roadmap. But why is my head swirling?

Without NASA involvement we could probably skip to #6.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/23/2022 06:32 pm
I looked it up.  A pass through the Van Allen belts was 1.1mSv on the lightly shielded Apollo.   It was a little less than half of the total dose the Apollo astronauts received.

It would be more than that.  Apollo transited the VA belts at close to max speed.  In an HEEO, the part of the orbit near the apogee takes up most of the orbital period.

Quote
So whatever type of ship has humans in it will need extensive shielding to get to Mars.   Might as well reuse that for going to the Moon.

You don't design the lunar version based on the Mars requirements.  Mars is at least a decade away.  Option A is (hopefully) four years away.

Quote
An LRHO tanker is needed solely for the Artemis architecture, which said architecture will be obsolete in 5-10 years when Mars landers are being mass produced.   In 10 years going to the Moon won't require fancy custom craft, just a refuel in HEEO and a standard Mars lander.    In a quirk of astrophysics (no air on Moon) it will require more fuel than going to Mars.

It doesn't matter what happens in ten years.  It matters what happens in four years.

You're basically complaining that the refueling architecture for Artemis isn't elegant.  I'd respond that it's as elegant as it can be, given the constraints.

The thing that would make refueling considerably more elegant is bigger tanks in the LSS.  Then all the nonsense about who boosts to what orbit first, or who's tethered/docked/formation-flying to whom during the boost, goes away.  Given that some of that is IMO seriously crazy, that's a good thing.

I'll be very impressed if we see crews launching and doing EDL in Starships before 2030.  So it would be nice to have some options for the latter parts of the Artemis campaign that don't rely on SLS and Orion.  If that requires a bit more refueling complexity to reduce the per-mission cost by more than $3B, I'm OK with that.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/23/2022 07:09 pm
I don't think the whole "tandem burn" thing works, because if the crew burn succeeds and the tanker burn aborts, that's a loss of mission.  (You can't leave the crew in the VA Belts while you debug the tanker, and they have to return to LEO.)

You leave enough fuel reserve to enter atmosphere and land normally.  That's not loss of mission, merely an abort of mission.   The periapsis only needs to be dropped 200km in order to get to atmosphere reeentry, that's a trivial amount of fuel when the burn is done at apoasis on a HEEO.

It's an LSS; it can't reenter.  But it should have enough prop on board to propulsively lower the apogee back to the original insertion LEO.  Or you can leave it where it is while you debug the tanker, but then it's soakin' up the rads.

We can argue semantics about whether "mission abort" = "loss of mission".  You haven't lost the vehicle, which is different from your garden-variety LoM.  Let's say "major contingency likely involving at least one additional tanker launch".  It's not great.  Not a disaster, but not great.

Quote
There are equally a number of pathways to mission abort in the other scenarios being talked about.  Possibly more.

This one is avoidable, though:  simply boost the tanker first, and have it waiting for the LSS when it boosts later.  No need for this tandem boost stuff.  Same argument applies for higher, more eccentric orbits.

Quote
I agree it's one extra pass through the Van Allen belts.  I wonder if the 50T of extra cargo space that NASA doesn't need would allow mass to make up for that or not.   Especially if it is useful cargo like water.  200 square meters of crew surface area means a water depth of 0.25 meters all around the crew area.

Here's a study on that:  https://digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=3795&context=theses (https://digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=3795&context=theses)

Given that the way we went down this rabbit hole in the first place was arguing about whether an LSS with 1200t of tankage was adequate to do the LEO-NRHO-LS-NRHO mission on one tank, throwing an extra 50t of water at the problem--and making your crew module independently-hulled in the process, with something that could leak between it and the fairing--doesn't seem like a step in the right direction.

In terms of the early Artemis conops, you don't need this shielding, because the LSS is uncrewed before it gets to NRHO.  SpaceX could certainly harden the avionics to be VA Belt-proof for a lot less than 50t.  But that (like a water-shielded crew module) is more complexity, and less in common between the LSS and the vanilla Starship.

I'd much rather use that extra payload bay space and fill it up with lovely, lovely propellant.  That is, admittedly, a structural change, but it should be a surprisingly easy one, given how Starship is constructed.
Changing tank volume is utterly trivial, especially compared to other changes needed for LSS. A higher upper dome means the landing engines are a tad higher and can be aimed a tad more vertical.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/23/2022 07:22 pm
Agreed, and if a depot doesn't EDL, why not fit it with the 6 engines that it will use in space (I assume RVacs) and use one at a time until it fails, then use another? I'm guessing that will make it last a long time. Or, if it is better to rotate them, do that. Spread the "hours used" out as much as possible. When it is finished, land/splat it on Luna as raw material for whatever the base needs.
At this point, I'm pretty sure the depots will just be modified versions of the basic Starship where they've removed the cargo area and stretched the tanks about as far as they can go. Since the basic Starship holds 1200 tons of propellant, it's a fair guess that a depot could hold 1800-2000. (I haven't seen anyone seriously try to estimate this yet though.) Dry mass ought to be lower than the 100 tons for a basic Starship, but I haven't seen a serious estimate for how much lower it could be. Maybe 80 tons?

Since it's really just a Starship, it'll launch like any other Starship, which means it'll need those engines (all six of them) to get into orbit. If it's going to be an LEO depot, you won't need the engines again; any propulsion after that will be from the ullage engines. If it's going to NRHO, then it'll need to get refilled by tankers and then it'll use its engines again to reach NRHO. Again, after that, it shouldn't ever use the big engines again.

After thinking about it, I've reached the point where I'm okay with the idea of using tankers capable of EDL to fill a depot at NRHO rather than trying to get depots to shuttle back and forth. The ability to service them on the ground is clearly more important than being able to lug a bit more fuel out to NRHO.

So what to do with the used lunar Starships? Well, here's an idea: if you've got a fuel depot in NRHO, you don't need a lunar Starship; just use a regular one (designed for crew). A basic Starship could launch to LEO, fill up from the LEO depot, head to NRHO, refill from the NRHO depot, go to the moon and back, and (I think) still have enough fuel to return to Earth without needing an additional refill.

That would be very neat and clean; just four Starship variants: cargo, crew, tanker, and depot. And only depot can't do EDL.
This assumes they can get rid of the landing engines. Elon wants to. Doesn't mean it'll happen. Later, when they have a prepared surface or chopsticks, it'll be a done deal.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/23/2022 07:32 pm
At this point, I'm pretty sure the depots will just be modified versions of the basic Starship where they've removed the cargo area and stretched the tanks about as far as they can go. Since the basic Starship holds 1200 tons of propellant, it's a fair guess that a depot could hold 1800-2000. (I haven't seen anyone seriously try to estimate this yet though.) Dry mass ought to be lower than the 100 tons for a basic Starship, but I haven't seen a serious estimate for how much lower it could be. Maybe 80 tons?

I've estimated this if the LCH4 dome is moved forward only to consume the cylindrical portion of the payload bay, so there's no fiddling with engineering the tank to intrude up into the ogive.

1620t.

Note that you can't launch with that amount of prop.  But even if there's just one version for both lift tanker and depot, the lift tankers can launch not quite full and be fine.  Note also that lift tankers need the nose header tanks, which would further complicate using the ogive.  You could get more than 1620t, but it'll increase complexity.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/23/2022 07:47 pm
All well and good, but more likely is that SpaceX will reduce dry mass or pick a different staging orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/23/2022 08:15 pm
Changing tank volume is utterly trivial, especially compared to other changes needed for LSS. A higher upper dome means the landing engines are a tad higher and can be aimed a tad more vertical.

Given how Starship is constructed by stacking combinations of ring segments, domes, bulkheads, and nose pieces, changing the stacking order doesn't seem too difficult.  But I wouldn't go so far as to say that it's "utterly trivial."  Moving the domes changes the lengths of a lot of plumbing and electrical harnesses, which isn't trivial.  However, it sure looks like the GSE was designed with this as a possibility.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/23/2022 08:29 pm
All well and good, but more likely is that SpaceX will reduce dry mass or pick a different staging orbit.
It's been awhile since we saw hard weight numbers during an SS lift or any weight statements from Elon. My guess is that dry mass has gone up but with little effort in mass optimization. Once they actually launch that sucker they'll take a harder look at mass. It'd be great if they can shave 2-3 tens of tons. Till then, everything's on the table.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/23/2022 08:52 pm
Changing tank volume is utterly trivial, especially compared to other changes needed for LSS. A higher upper dome means the landing engines are a tad higher and can be aimed a tad more vertical.

Given how Starship is constructed by stacking combinations of ring segments, domes, bulkheads, and nose pieces, changing the stacking order doesn't seem too difficult.  But I wouldn't go so far as to say that it's "utterly trivial."  Moving the domes changes the lengths of a lot of plumbing and electrical harnesses, which isn't trivial.  However, it sure looks like the GSE was designed with this as a possibility.
Trivial in a relative sense. Trivial compared to doing it in another rocket and trivial compared to other design issues SS faces.


Extending plumbing shouldn't be difficult but would call for a close look during a live fire test and everything leading up to it. AIUI, automotive wiring harnesses are hand work, not automation. In the grand scheme of things extending the harnesses would not be much of an issue. Power cables might need to be a larger gauge. Maybe.


All in all, you'd have to change the scale setting to see the 'difficulty' meter budge.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/24/2022 12:04 am
I've estimated this if the LCH4 dome is moved forward only to consume the cylindrical portion of the payload bay, so there's no fiddling with engineering the tank to intrude up into the ogive.

1620t.

Note that you can't launch with that amount of prop.  But even if there's just one version for both lift tanker and depot, the lift tankers can launch not quite full and be fine.  Note also that lift tankers need the nose header tanks, which would further complicate using the ogive.  You could get more than 1620t, but it'll increase complexity.
If you look at figure 2 in NASA’s Initial Artemis Human Landing System (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf), the tanker does seem to be designed as you describe. But the depot is maybe 20% longer than that. That's why I'm thinking the depot might hold as much as 2000 tons of prop, even if it only launches about half-full.

Of course, that's basing a lot on a drawing that isn't even attributed to anyone, but . . .
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/24/2022 04:19 am
I've estimated this if the LCH4 dome is moved forward only to consume the cylindrical portion of the payload bay, so there's no fiddling with engineering the tank to intrude up into the ogive.

1620t.

Note that you can't launch with that amount of prop.  But even if there's just one version for both lift tanker and depot, the lift tankers can launch not quite full and be fine.  Note also that lift tankers need the nose header tanks, which would further complicate using the ogive.  You could get more than 1620t, but it'll increase complexity.
If you look at figure 2 in NASA’s Initial Artemis Human Landing System (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf), the tanker does seem to be designed as you describe. But the depot is maybe 20% longer than that. That's why I'm thinking the depot might hold as much as 2000 tons of prop, even if it only launches about half-full.

Of course, that's basing a lot on a drawing that isn't even attributed to anyone, but . . .

Let's take the artwork at face value for a moment and think about the implications.  TL;DR:  Numbers will be tortured below, but it's kinda eerie how close this comes to working out to something reasonable.  For those of you whose eyes simply refuse to glaze over, let's forge on:

Looks to me that the beginning of the ogive (i.e., the end of the cylindrical section) is about level with the nose of the other flavors of Starship.  Given that the ogive and nosecone are about 12m in length, that would add about 760m³ of volume.  Average density of 3.6:1 methalox is about 830kg/m³, so we're looking at a depot that has about 1830t of prop when the depot is full.  (Note that I'm assuming that the LCH4 dome is at the top of the cylindrical section, and the ogive contains no prop except for the header tanks.)

First hypothesis:  The stretch is to handle boiloff.

Ullage prop at 6bar would be 13.2t.  Use our 1t sump losses and, from LEO you have 0 FPR.  If we use our previous SWAG of 500kg/day for boiloff in LEO, that would be 3.4 years before you couldn't fill a 1200t LSS, and 1.7 years before you couldn't fill a 1500t LSS.

Both of those loiter times are pretty extreme, and that's a really expensive depot to fill:  13 tanker launches if they can carry 150t to LEO.  If we believe the artwork (which is a given for this little exercise), then the stretch isn't just to handle boiloff.

Second hypothesis:  It's just enough to get the LSS delta-v budget to close for an HEEO-NRHO-LS-NRHO conops.

I computed 54t of prop losses for a 1200t LSS a ways up-thread, and I couldn't get that to close at a refueling orbit of 300x2000km.  So what would close?

95t dry mass + 20t crew module + 54t unusable prop gives you 7920m/s of delta-v.  That would give you 2010m/s to get from the refueling orbit to a C3=-1.5km²/s² departure needed to get to BLT.

That requires that the LSS has to boost by about 1290m/s from a 300x300 orbit.  That leaves it in a 300x7900km orbit, which seems optimized to subject it to just about as much radiation as possible, for as long as possible.  Furthermore, it would leave the depot with way, way more prop available for transfer than was needed.  (See below.)  So this doesn't sound like the right answer either.

Third hypothesis (and I promise the pain will end after this):  The LSS has some spare prop in the tank when it gets to LEO.  If we use that prop to boost as high as possible, subject to conservative prop loss estimates, maybe the depot is sized correctly to completely fill the LSS for the mission.  This is more prop than it'll need, but NASA is nothing if not conservative.  Let's try it.

If we only want to refuel the LSS once (as a risk reduction), then the orbit can't be any higher that the prop the LSS has when it inserts into VLEO.  For a 95t dry +20t crew LSS, based on a 120t dry Starship that can deliver 150t to LEO, that should be about 120-115+150 = 155t of prop.  Take out 9t for ullage gas, 1t for sump, nothing for boiloff, and about 4t for FPR, and you have 129t inert mass for 141t of remaining usable prop, which gives you a delta-v of 2738m/s.  Let's call it 2700m/s.

Let's figure out what the depot would have available if it boosted up to that energy, then returned to VLEO when it was done with the mission.

First, dry mass:  We've added 12m of sheet metal and longerons to the ~95t of an LSS-like dry mass:  As a SWAG, makes the dry depot... about 110t?  (OK:  allocate 35t to thrust structure and engines, leaving the tankage and other stuff at 60t, then multiply by (50+12)/50.  Maybe only a WAG, instead of a SWAG.)  So, with 20t for crew module, we're at 130t.

We already computed 14t for ullage gas and sump above.  We'd like to get the depot back, but it's not the end of the world if the return fails, so let's allocate 0.5% FPR on 1830t: 9t.  It's dumb to let the depot sit in orbit full for very long, but the LSS could have a glitch before launch that required a month or two of debugging, so allocate 60 days @ 500kg/day = 30t for boil-off.  Total unusable prop:  53t.

So we have 130t + 53t = 183t of inert mass, with 1830t - 53t = 1777t usable prop.  (Note that this assumes the depot is chock-full from tankers fueling it before departure to HEEO.)

To do a 2700m/s return from HEEO to LEO with 183t inert mass requires 196t of prop.  Add that to the inert mass for the boost up to HEEO (379t), and you need 406t of prop.  That leaves us with 1777 - 196 - 406 = 1175t of prop available for transfer.

So, this kinda feels like I've fallen into a conspiracy theory involving numerology, but it's hard to deny that I just tortured a bunch of numbers and came within 25t of a full tank of prop for a 1200t LSS, sitting in an LEO+2700 HEEO.  (That's about 300 x 62,000km, which has an orbital period of about 20 hours, for the peanut gallery.  And that's high enough that VA belt radiation exposure should be fairly low.)

We certainly haven't disproven the null hypothesis (i.e., that the artwork is fanciful), but we also haven't disproven a semi-plausible hypothesis for why it might be that size.

Hmmm.  Hrrrr.   Mumble, mumble...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/24/2022 05:36 am
... snipping a lot...

Hmmm.  Hrrrr.   Mumble, mumble...

Did you make sure to count the Oberth effect since you have elliptical orbits and I think you are boosting at periapsis?

It's on the order of 2x the deltaV, it's worth a lot.

I find you have to do the C3 calcs to get it correct.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 09/24/2022 06:33 am
We don't know what the first flight proven Starship will weigh, we don't know what the final version will weigh, and we don't know which generation in between will fly the HLS mission.

At some point they will ask "How can I fly HLS with the Starship I have?" and design the mission around the ship rather than the other way around.  But not yet, because it hasn't reached orbit yet, and it hasn't refueled anything yet.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/24/2022 08:52 am
... snipping a lot...

Hmmm.  Hrrrr.   Mumble, mumble...

Did you make sure to count the Oberth effect since you have elliptical orbits and I think you are boosting at periapsis?

It's on the order of 2x the deltaV, it's worth a lot.

I find you have to do the C3 calcs to get it correct.

Yes. They’re all vis-viva from 300km perigee.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/24/2022 09:37 am
We don't know what the first flight proven Starship will weigh, we don't know what the final version will weigh, and we don't know which generation in between will fly the HLS mission.

At some point they will ask "How can I fly HLS with the Starship I have?" and design the mission around the ship rather than the other way around.  But not yet, because it hasn't reached orbit yet, and it hasn't refueled anything yet.

We know it isn't going to be vanilla Starship, because:

1) It's too heavy.
2) It doesn't have the right thermal characteristics.
3) Any payload is more like cargo than microgravity-deployable stuff.
4) Crew ingress/egress is a big deal.
5) The crew module itself, and its docking systems, require ripping out the headers.
6) It has to land on a rough surface with very high T/W if they use the Raptors.
7) It doesn't have the delta-v unless you're willing to spend prop for HEEO refueling.
8) It has to be operationally simple enough to fend off Blue Origin crying "fraud!"  (IIRC the "one refueling in LEO" was introduced in NASA's response to the appeal.)

Now:  You're tearing off elonerons, not installing the TPS, painting it with funny coatings.  You're adding landing legs, you're probably adding landing thrusters, you're ripping out header tanks (and the downcomers that go with them), installing docking systems, and last but hardly least installing decks, controls, airlocks, elevators, and ECLSS for the crew system.  It's just not a vanilla Starship.  How much more work is it to move the LCH4 dome and LCH4/LOX bulkhead forward a bit--especially if it simplifies the conops?

All that said, I couldn't get things to close from VLEO, even at 1500t prop.  I suspect that's because I've been too conservative with prop loss, something that I'm going to try to refine by re-kludging my model.  And you can't go much further than 1500t unless you're willing to change the outer mould line, which may or may not be a big deal.

They can absolutely use an HEEO for refueling, although it's more complicated, more expensive in tankers, and it doesn't extend cleanly if they want to eventually stage crews out of LEO, which I'm almost certain they do.  As has been pointed out many times on this thread, we'll all guessing.  So we'll see.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Paul451 on 09/24/2022 01:32 pm
reduce dry mass
Quote
reduce the dry mass
Quote
reduce the dry mass

"Dry mass" in this context includes mission mass. Payload. What's the point of having the largest lander if you can't land the largest payloads? What's the point of Starship at all if you have to scratch and scrape to make delta-v budgets close?

If the answer is "reduce the dry mass", then SpaceX has, IMO, failed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: DanClemmensen on 09/24/2022 01:47 pm

All that said, I couldn't get things to close from VLEO, even at 1500t prop.  I suspect that's because I've been too conservative with prop loss, something that I'm going to try to refine by re-kludging my model.  And you can't go much further than 1500t unless you're willing to change the outer mould line, which may or may not be a big deal.

They can absolutely use an HEEO for refueling, although it's more complicated, more expensive in tankers, and it doesn't extend cleanly if they want to eventually stage crews out of LEO, which I'm almost certain they do.  As has been pointed out many times on this thread, we'll all guessing.  So we'll see.
Refuel twice if needed. I don't think a second Depot will add all that much to the cost, and it's reusable for many additional missions. This allows for one LEO Depot and one HEEO Depot. Probably requires more tanker flights, but tanker flights are supposed to be cheap. Using two Deports should allow for HLS to carry lots of downmass and still get back to up NRHO. I don't know if HLS can get all the way back to HEEO, but if it can, then it can refuel there on the way back before it goes back to LEO.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/24/2022 02:31 pm
reduce dry mass
Quote
reduce the dry mass
Quote
reduce the dry mass

"Dry mass" in this context includes mission mass. Payload. What's the point of having the largest lander if you can't land the largest payloads? What's the point of Starship at all if you have to scratch and scrape to make delta-v budgets close?

If the answer is "reduce the dry mass", then SpaceX has, IMO, failed.
It is mostly reducing dry mass of the non-payload portions. SpaceX built the early stages like battleships. They’ve been continuously testing manufacturing techniques, different thicknesses of tank skin, etc.

I don’t see why this topic is so hard for some to understand. This whole argument that a tank stretch HAS to happen rests on basically made up dry mass figures. If you just assume they’re getting better dry mass numbers than ones you or TheRadicalModerate just made up, then it’s no longer a challenge.

And no, Starship hasn’t failed just because they use an efficient dry mass figure for early Artemis missions. That’s literally the opposite.

It’s funny how TheRadicalModerate’s made up dry mass figures have sort of become accepted as truth on this forum just because he’s a prolific poster on this topic (not that I’m one to talk). Am I really being shouted down because I don’t trust the dry mass figures he pulled out of his rear? LOL
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/24/2022 03:50 pm
It’s funny how TheRadicalModerate’s made up dry mass figures have sort of become accepted as truth on this forum just because he’s a prolific poster on this topic (not that I’m one to talk). Am I really being shouted down because I don’t trust the dry mass figures he pulled out of his rear? LOL

You shouldn't trust my rectally-extracted dry mass numbers for the LSS--nor should you trust the similarly-extracted mass numbers for the crew module.  But you should trust the maximum dry mass calculation, whose inputs are the delta-v budget, the Isp, and maximum usable propellant.

Even there, there's room for argument about what portion of the prop is unusable, i.e., lost to ullage gas, sump losses, boil-off, and FPR.  If we revisit that calculation (which is here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2410896#msg2410896)) but use a max prop loss of 30t instead of 54t, then the max dry mass (which is really max dry + payload/crew module--I chose a bad name), for the 300x2000 refueling orbit becomes 108t.

That's getting closer to credible, but it assumes that I'm either pretty far off on the base LSS dry mass (95t, which is derived from Elon's estimate of 120t, less TPS, less elonerons, plus the solar white coating, plus landing legs, plus maybe landing thrusters), or the crew module, with its attendant airlocks, surface access hardware, and docking hardware, is much smaller than 20t.

If it's really going to be a 1200t prop load, by far the easier way to meet that is to jack up the refueling orbit to GTO-like energies and eccentricities.  But then you really do need the giant depot to get enough prop to the refueling orbit to be worthwhile, and you have to be very, very conservative with the amount of prop on hand and usable after the LSS has launched into its VLEO parking orbit.

After doing that horrible set of calculations last night, I'm more convinced that the depot artwork in the MSFC paper is representative of the actual plan, and that the plan does indeed involve refueling the LSS in HEEO, not in VLEO.  I'm surprised by that, because that plan sucks if you want to start staging crews out of LEO instead of NRHO.  The problem with that is, as Gloria Estefan says, the radiation is gonna get you. 

But maybe SpaceX is a lot more confident that they'll be able to do single-orbit refuelings with no contingencies, in which case maybe they've judged the radiation to be acceptable.  That's certainly more in keeping with what you'd have to do if you wanted to launch crews in almost-vanilla Starship to the lunar surface and then return them direct to EDL.  Or maybe they're planning on doing crewed refueling in cislunar before lunar descent, which isn't as efficient as doing it after ascent, but does have the nice property of giving you a clean abort back to EDL if the refueling goes wrong.

There are lots of ways to skin this particular cat.  I still think that swapping payload height for more prop is the cleanest, but I admit that's not the only way to go.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/24/2022 03:54 pm
I'm not really meaning to attack you or even claim your numbers are for sure wrong about dry mass. HOWEVER, I wanted to inject some uncertainty to these conversations.

You're exploring just one branch of the probability tree, and it's not at all certain that it's the most likely one.

(The radiation risk without crew is no problem. And with crew in an emergency, the extra dose from traveling through the Van Allen belts is pretty low, other than what you typically get from being in deep space.)

Van Allen belt radiation is pretty easy to shield against, BTW.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/24/2022 03:59 pm
What's worse, loss of 50t of propellant out of 1200t, or loss of 50t of propellant out of 50t?

What's worse, loss of propellant at a high energy orbit or a low energy orbit?

when and where you lose the propellant is important.   And it hasn't been entirely clear from me when/where this is occurring or optimizing around those facts.

I think any analysis needs to show each conic and when/where the fuel was lost.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/24/2022 06:40 pm
What's worse, loss of 50t of propellant out of 1200t, or loss of 50t of propellant out of 50t?

What's worse, loss of propellant at a high energy orbit or a low energy orbit?

when and where you lose the propellant is important.   And it hasn't been entirely clear from me when/where this is occurring or optimizing around those facts.

I think any analysis needs to show each conic and when/where the fuel was lost.

I agree. I’ve been thinking about adding boiloff and fpr to each maneuver block in my model. Fpr would be a percentage of the upcoming delta-v and boiloff would be a rate and time before the maneuver. Do you think that’s good enough?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 09/24/2022 10:08 pm
What's worse, loss of 50t of propellant out of 1200t, or loss of 50t of propellant out of 50t?

What's worse, loss of propellant at a high energy orbit or a low energy orbit?

when and where you lose the propellant is important.   And it hasn't been entirely clear from me when/where this is occurring or optimizing around those facts.

I think any analysis needs to show each conic and when/where the fuel was lost.

I agree. I’ve been thinking about adding boiloff and fpr to each maneuver block in my model. Fpr would be a percentage of the upcoming delta-v and boiloff would be a rate and time before the maneuver. Do you think that’s good enough?
From a practical math standpoint you likely get to the same answers if you just deal with the time spent between main engine thrust events as a simple calculation of prop loss for that segment. VS trying to do more precises integration equation math models that are not going to yield a better answer because of the accuracy of the input data of just 2 or 3 digits.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/24/2022 10:26 pm
It’s funny how TheRadicalModerate’s made up dry mass figures have sort of become accepted as truth on this forum just because he’s a prolific poster on this topic (not that I’m one to talk). Am I really being shouted down because I don’t trust the dry mass figures he pulled out of his rear? LOL

You shouldn't trust my rectally-extracted dry mass numbers for the LSS--nor should you trust the similarly-extracted mass numbers for the crew module.  But you should trust the maximum dry mass calculation, whose inputs are the delta-v budget, the Isp, and maximum usable propellant.

Even there, there's room for argument about what portion of the prop is unusable, i.e., lost to ullage gas, sump losses, boil-off, and FPR.  If we revisit that calculation (which is here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2410896#msg2410896)) but use a max prop loss of 30t instead of 54t, then the max dry mass (which is really max dry + payload/crew module--I chose a bad name), for the 300x2000 refueling orbit becomes 108t.

That's getting closer to credible, but it assumes that I'm either pretty far off on the base LSS dry mass (95t, which is derived from Elon's estimate of 120t, less TPS, less elonerons, plus the solar white coating, plus landing legs, plus maybe landing thrusters), or the crew module, with its attendant airlocks, surface access hardware, and docking hardware, is much smaller than 20t.

If it's really going to be a 1200t prop load, by far the easier way to meet that is to jack up the refueling orbit to GTO-like energies and eccentricities.  But then you really do need the giant depot to get enough prop to the refueling orbit to be worthwhile, and you have to be very, very conservative with the amount of prop on hand and usable after the LSS has launched into its VLEO parking orbit.

After doing that horrible set of calculations last night, I'm more convinced that the depot artwork in the MSFC paper is representative of the actual plan, and that the plan does indeed involve refueling the LSS in HEEO, not in VLEO.  I'm surprised by that, because that plan sucks if you want to start staging crews out of LEO instead of NRHO.  The problem with that is, as Gloria Estefan says, the radiation is gonna get you. 

But maybe SpaceX is a lot more confident that they'll be able to do single-orbit refuelings with no contingencies, in which case maybe they've judged the radiation to be acceptable.  That's certainly more in keeping with what you'd have to do if you wanted to launch crews in almost-vanilla Starship to the lunar surface and then return them direct to EDL.  Or maybe they're planning on doing crewed refueling in cislunar before lunar descent, which isn't as efficient as doing it after ascent, but does have the nice property of giving you a clean abort back to EDL if the refueling goes wrong.

There are lots of ways to skin this particular cat.  I still think that swapping payload height for more prop is the cleanest, but I admit that's not the only way to go.
This has probably been done already but here's a sanity check on at least one of the numbers, the full up Dragon with crew and cargo weighs 11.5t, carries 1.3t of propellant and 3.3t of cargo. It is unclear if that 3.3t includes crew. I suspect it does. In choppers we figured 250lb per combat equipped soldier so figure the crew at .5t, unless they're packing claymores.


They're no reason for the crew module to weigh any more than a crew dragon. [size=78%]https://en.m.wikipedia.org/wiki/SpaceX_Dragon_2 (https://en.m.wikipedia.org/wiki/SpaceX_Dragon_2)[/size]


Total 11.5t - 1.3t propellant = 10.2t
10.2t - 3.3t cargo = 6.9t
6.9t + .5t crew = 7.4t
7.5t - 1.0t* ablative heat shield = 8.5t.
8.5t + 1.0t* additional ECLSS = 9.5t.


The NASA required cargo is somewhere near that 3.3t. dragon already has a docking port but it would probably need reinforcement. IIRC plans are for two airlocks and one of them would probably be collocated with the docking ring which, with a little cunning SX engineering, would double as at least some of the docking reinforcement.


So, add in the cargo weight, add in the airlocks and add in the elevator and other cargo handling equipment, and you have the total delivered mass. Did I miss anything? 20t looks like a top end number. Might be 2-3t less.


* WAG. If anybody has better numbers, plug them in.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/25/2022 03:47 am
They're no reason for the crew module to weigh any more than a crew dragon.

We're probably a little too far O/T on this one.  However, it's a good point.  Maybe 10t-15t is a better guess than 20t.

We do need to get a handle on prop loss to figure out depot orbits, though.  To that extent, some kind of reasonable estimate of inert mass is important.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/25/2022 02:56 pm
They're no reason for the crew module to weigh any more than a crew dragon.

We're probably a little too far O/T on this one.  However, it's a good point.  Maybe 10t-15t is a better guess than 20t.

We do need to get a handle on prop loss to figure out depot orbits, though.  To that extent, some kind of reasonable estimate of inert mass is important.
Pay no attention to the balkanization fiends. To look at refueling calls for a look at mission parameters. This is all on topic.


Or, we can split this out into threads for LEO refueling, HEEO refueling, NRHO refueling and LLO refueling, then double for a stretch depot, then double again for tankers with bulkheads moved up. :o
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/25/2022 05:13 pm
They're no reason for the crew module to weigh any more than a crew dragon.

We're probably a little too far O/T on this one.  However, it's a good point.  Maybe 10t-15t is a better guess than 20t.

We do need to get a handle on prop loss to figure out depot orbits, though.  To that extent, some kind of reasonable estimate of inert mass is important.
Pay no attention to the balkanization fiends. To look at refueling calls for a look at mission parameters. This is all on topic.


Or, we can split this out into threads for LEO refueling, HEEO refueling, NRHO refueling and LLO refueling, then double for a stretch depot, then double again for tankers with bulkheads moved up. :o

I agree you need a number.  FWIW, I've adjusted my crew module mass down to 15t.  We don't need to completely balkanize, but a crew module mass thread would probably be a good idea.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/26/2022 01:45 pm
They're no reason for the crew module to weigh any more than a crew dragon.

We're probably a little too far O/T on this one.  However, it's a good point.  Maybe 10t-15t is a better guess than 20t.

We do need to get a handle on prop loss to figure out depot orbits, though.  To that extent, some kind of reasonable estimate of inert mass is important.
Pay no attention to the balkanization fiends. To look at refueling calls for a look at mission parameters. This is all on topic.


Or, we can split this out into threads for LEO refueling, HEEO refueling, NRHO refueling and LLO refueling, then double for a stretch depot, then double again for tankers with bulkheads moved up. :o

I agree you need a number.  FWIW, I've adjusted my crew module mass down to 15t.  We don't need to completely balkanize, but a crew module mass thread would probably be a good idea.
I think we just got a reasonable first order approximation until new data shows up. No reason to go further.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/27/2022 12:03 am
OK, I think I've re-kludged my model to handle prop losses.  Not perfect, but better.  I'm just going to include the link here (https://docs.google.com/spreadsheets/d/17NNg4iHZDn98zW8Zq_4JmG04unBC5fR31-C_5Z1F3cM/edit?usp=sharing), in the interests of not dumping zillions of cryptic tables in PNG format.

Some takeaways:

1) LSS with 1200t prop definitely doesn't work without boosting to some kind of moderate HEEO, along with the depot.  I left the depot at an even 1500t, which is adequate to get the LSS to do LEO+1670-NRHO-LS-NRHO with a non-trivial amount of cargo.  Note that a launched LSS arrives in LEO with enough prop to boost to LEO+1670 (which is in the gap between the VA belts), so this still satisfies the "only one refueling for the LSS" requirement.

2) An LSS with tanks rearranged to hold 1500t can make it from a 300x300 VLEO after refueling and do the whole VLEO-NRHO-LS-NRHO mission, complete with a 90-day loiter.  Sure seems a lot simpler to me.

3) Option B reused LSSes need to be serviced by a tanker with 1500t tanks if you only want one EDL-capable tanker to do the deed.  If the tanker is 1200t, you need two refuelings of the LSS, or two refuelings of an NRHO-based depot.  Again, 1500t solves a lot of problems.

4) I've included a couple of cases where crews are staged out of LEO, to show how they work.  I've assumed 1500t prop for these LSSes.  One case needs an NRHO refueling after lunar ascent, and the other refuels before lunar descent.  The former is quite reasonable, while the latter is horrific.  It's not good to haul a bunch of prop down to the lunar surface if you don't have to.  But pre-descent fueling gives you a clean abort (i.e., return to LEO propulsive) if something goes wrong.

5) I've also included a couple of expendable cargo LSS missions to show how cheap CLPS services can be.

6) I think I've chosen reasonable values for sump losses (1t everywhere) and ullage gas loss (based on whether it's a 1200t or 1500t tank).  I'm less certain about boiloff.  In general, vehicles with solar white boiloff-reduction have lower rates, and things that can keep the nose pointed at the sun have lower rates.  Lunar surface rates are assumed to be 3-4x higher than in-space rates for the LSS.

7) Finally, FPR.  Any maneuver with a crew has an FPR of 2% of delta-v for that maneuver.  Any tanker maneuver uses 0.5%.  Any uncrewed LSS maneuver uses 0.75%.  This isn't perfect, because early maneuvers get weighted more heavily than late maneuvers, but it should model consistent underperformance more-or-less correctly.

8) Mission roll-ups:
1200t LSS Option A:  10 tanker launches with 27t cargo.
1500t LSS Option A:  9 tanker launches with 15t cargo.
Reused Option B with 1500t tanker:  10 tanker launches, 5t cargo brought on tanker.
Reused Option B with 1200t tanker: 12 tanker launches, two refuelings of LSS.
Crewed LSS staged from LEO (1500t), post-lunar refueling: 15 tankers, 70t cargo.
Crewed LSS staged from LEO (1500t), pre-lunar refueling:  28 tankers, horror.
Expendable LSS CLPS:  3 tankers, 33t payload.
Expendable LSS CLPS: 8 tankers, 175t payload (LSS is lighter, more payload to LEO).

Note that the weird payload numbers represent my attempt to get things to come out to an integral number of tanker launches.

Probably a few stupid mistakes; let me know what you think.  I may be slow to respond.  I live just north of Tampa, so I'll be spending tomorrow rolling all my outdoor furniture and plants back to the VAB.  Happy hurricane!

Update:  The 1200t LSS model had a problem:  By going up to LEO+1670, there wasn't enough prop left in the LSS to return to LEO if the refueling aborted.  In the spreadsheet, I've done the following:

- Reduced cargo (non-crew-module) mass down to 5t, which ought to be more than enough for Options A/B.

- Dropped down to LEO+1420 (300x9250), which leaves enough prop from launch to return back to 300x300 if necessary.  (See Twark's comment about the possibility of extremely gentle aerobraking to recover the LSS if there's a contingency--I remain skeptical.)

- The result of these two reductions is that you have an easy-to-abort refueling that only requires about 1350t of prop from the depot, which means that the mission can be done with 9 lift tankers.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/27/2022 12:55 am
OK, I think I've re-kludged my model to handle prop losses.  Not perfect, but better.  I'm just going to include the link here (https://docs.google.com/spreadsheets/d/17NNg4iHZDn98zW8Zq_4JmG04unBC5fR31-C_5Z1F3cM/edit?usp=sharing), in the interests of not dumping zillions of cryptic tables in PNG format.

Some takeaways:

1) LSS with 1200t prop definitely doesn't work without boosting to some kind of moderate HEEO, along with the depot.  I left the depot at an even 1500t, which is adequate to get the LSS to do LEO+1670-NRHO-LS-NRHO with a non-trivial amount of cargo.  Note that a launched LSS arrives in LEO with enough prop to boost to LEO+1670 (which is in the gap between the VA belts), so this still satisfies the "only one refueling for the LSS" requirement.

2) An LSS with tanks rearranged to hold 1500t can make it from a 300x300 VLEO after refueling and do the whole VLEO-NRHO-LS-NRHO mission, complete with a 90-day loiter.  Sure seems a lot simpler to me.

3) Option B reused LSSes need to be serviced by a tanker with 1500t tanks if you only want one EDL-capable tanker to do the deed.  If the tanker is 1200t, you need two refuelings of the LSS, or two refuelings of an NRHO-based depot.  Again, 1500t solves a lot of problems.

4) I've included a couple of cases where crews are staged out of LEO, to show how they work.  I've assumed 1500t prop for these LSSes.  One case needs an NRHO refueling after lunar ascent, and the other refuels before lunar descent.  The former is quite reasonable, while the latter is horrific.  It's not good to haul a bunch of prop down to the lunar surface if you don't have to.  But pre-descent fueling gives you a clean abort (i.e., return to LEO propulsive) if something goes wrong.

5) I've also included a couple of expendable cargo LSS missions to show how cheap CLPS services can be.

6) I think I've chosen reasonable values for sump losses (1t everywhere) and ullage gas loss (based on whether it's a 1200t or 1500t tank).  I'm less certain about boiloff.  In general, vehicles with solar white boiloff-reduction have lower rates, and things that can keep the nose pointed at the sun have lower rates.  Lunar surface rates are assumed to be 3-4x higher than in-space rates for the LSS.

7) Finally, FPR.  Any maneuver with a crew has an FPR of 2% of delta-v for that maneuver.  Any tanker maneuver uses 0.5%.  Any uncrewed LSS maneuver uses 0.75%.  This isn't perfect, because early maneuvers get weighted more heavily than late maneuvers, but it should model consistent underperformance more-or-less correctly.

8) Mission roll-ups:
1200t LSS Option A:  10 tanker launches with 27t cargo.
1500t LSS Option A:  9 tanker launches with 15t cargo.
Reused Option B with 1500t tanker:  10 tanker launches, 5t cargo brought on tanker.
Reused Option B with 1200t tanker: 12 tanker launches, two refuelings of LSS.
Crewed LSS staged from LEO (1500t), post-lunar refueling: 15 tankers, 70t cargo.
Crewed LSS staged from LEO (1500t), pre-lunar refueling:  28 tankers, horror.
Expendable LSS CLPS:  3 tankers, 33t payload.
Expendable LSS CLPS: 8 tankers, 175t payload (LSS is lighter, more payload to LEO).

Note that the weird payload numbers represent my attempt to get things to come out to an integral number of tanker launches.

Probably a few stupid mistakes; let me know what you think.  I may be slow to respond.  I live just north of Tampa, so I'll be spending tomorrow rolling all my outdoor furniture and plants back to the VAB.  Happy hurricane!
You used the same dry mass for both the unstretched and the stretched lunar Starship (cells B7 and F7). How do you plan on stretching tanks without increasing dry mass?

Also, I think dry mass could plausibly be reduced for the unstretched version.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/27/2022 02:10 am
You used the same dry mass for both the unstretched and the stretched lunar Starship (cells B7 and F7). How do you plan on stretching tanks without increasing dry mass?

Also, I think dry mass could plausibly be reduced for the unstretched version.

You’re trading some cylindrical payload bay height for more prop. Just swap the order of ring and dome segments. You still wind up with 2m of cylindrical height and an unspecified amount of ogive space (depending on how big the crew module is). But the dry mass is the same.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/27/2022 02:10 am
1) LSS with 1200t prop definitely doesn't work without boosting to some kind of moderate HEEO, along with the depot

Why boost the depot itself?

Why not use the tanker, which is conveniently right there in orbitafter the delivery of its last propellant load?

Presumably the tanker will have lower dry mass (an optimized tanker design deleting the forward nosecone and flaps in favor of a fixed CoM design), and thereafter it can quickly perform a direct entry with only a tiny delta-v burn. The vehicle is immediately available for reuse, unlike a depot aerobraking back to LEO.

Since both mission architectures (boosting the depot vs. boosting the tanker) are readily available, it seems like we should at least compare and contrast the two options.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/27/2022 04:05 am
1) LSS with 1200t prop definitely doesn't work without boosting to some kind of moderate HEEO, along with the depot

Why boost the depot itself?

Why not use the tanker, which is conveniently right there in orbitafter the delivery of its last propellant load?

Presumably the tanker will have lower dry mass (an optimized tanker design deleting the forward nosecone and flaps in favor of a fixed CoM design), and thereafter it can quickly perform a direct entry with only a tiny delta-v burn. The vehicle is immediately available for reuse, unlike a depot aerobraking back to LEO.

Since both mission architectures (boosting the depot vs. boosting the tanker) are readily available, it seems like we should at least compare and contrast the two options.

It's a good idea, but I fooled with it a bit and it doesn't really do anything meaningful.

Going to an EDL-capable tanker saves you some return delta-v, but it doesn't save you huge amounts of prop, because the tanker's wet mass is small after it's transferred the prop to the LSS.  ISTM that either of two "winners" would justify using the EDL tanker:

1) If you could make it fit into 1200t, then you have One Tank Configuration to Rule Them All (1200t).  That's an R&D win.

2) If you could reduce the number of lift tankers you have from 10 to 9.  That's an ops win.

There's one additional problem:  I had an oversight in the first version of the "use HEEO" conops:  If you really boosted the LSS to LEO+1670, there's not enough prop to abort back to the original VLEO (I chose 300x300) if the refueling fails.  That could be a loss of vehicle for a very expensive LSS asset.  At the very least, it requires yet another depot, filled to capacity, which is also very expensive.

To fix the non-abortability problem, I chopped the LSS back to a 5t cargo (more than enough for Options A and B) but even then the LSS can't exceed LEO+1420 (which is 300x9250km, nicely in the VA belt gap, which helps a bit).  But with that low an energy, you can't get by with a 1200t tanker, and you still need 10 launches.

So no win with the EDL tanker.

You can argue, possibly successfully, that I've over-constrained the problem.  If you're willing to do two refuelings on the LSS, one in VLEO and one in HEEO, you can boost to pretty much any energy you want.  But NASA made a fairly big deal about only requiring one refueling against the LSS itself.  This actually came up in the BO lawsuit, where NASA responded to their complaint on this topic.  So going back on that opens up a can of worms that would be better left closed.

BTW, I expect it to be quite a while before you see a lift tanker with a different outer mould line from the vanilla Starship.  So I'd expect the tanker to have the vanilla dry mass (still 120t, although this number is extremely stale), while the depot, devoid of TPS and elonerons, should come in pretty close to the LSS mass.  I keep getting 95t without the crew module but, as robotbeat has noted, there's a lot of guesswork in that number.

The biggest problem with the depot shuttling back and forth to/from LEO+1420 is that you're putting more restarts on the engines than you would otherwise.  But the Raptors are designed to handle lots of restarts, and you're only adding two a mission, which should allow the depot plenty of design life.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/27/2022 04:08 am
You used the same dry mass for both the unstretched and the stretched lunar Starship (cells B7 and F7). How do you plan on stretching tanks without increasing dry mass?

Also, I think dry mass could plausibly be reduced for the unstretched version.

You’re trading some cylindrical payload bay height for more prop. Just swap the order of ring and dome segments. You still wind up with 2m of cylindrical height and an unspecified amount of ogive space (depending on how big the crew module is). But the dry mass is the same.
Sounds like to me you have a bunch of room to reduce dry mass of the standard LSS if you wanted to... Belly-flop aerodynamics are irrelevant for LSS.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/27/2022 04:38 am
If you really boosted the LSS to LEO+1670, there's not enough prop to abort back to the original VLEO (I chose 300x300) if the refueling fails.  That could be a loss of vehicle for a very expensive LSS asset.

Surely it could aerobrake at some altitude down to a lower apogee, no?


 
At the very least, it requires yet another depot, filled to capacity

Not true. You can do less. You can partially fill the tanker and LSS and get "in-between" performance.


BTW, I expect it to be quite a while before you see a lift tanker with a different outer mould line from the vanilla Starship.

I agree, with the caveat that "quite a while" in SpaceX time is equivalent to lightning speed in the Old World Order.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/27/2022 04:55 am

Going to an EDL-capable tanker saves you some return delta-v, but it doesn't save you huge amounts of prop, because the tanker's wet mass is small after it's transferred the prop to the LSS. [...]

Doesn't this presume that the depot isn't "pre-loaded" with propellant for some later mission?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/27/2022 04:56 am
If you really boosted the LSS to LEO+1670, there's not enough prop to abort back to the original VLEO (I chose 300x300) if the refueling fails.  That could be a loss of vehicle for a very expensive LSS asset.

Surely it could aerobrake at some altitude down to a lower apogee, no?

I think we have to assume not.  There's no TPS.  There are no aerosurfaces.  And if there's a coating, it'll get cooked.  It's not designed for aero-anything, except for close to zero angle of attack when bolted to a SuperHeavy.  There might be something really cool they've figured out, but there's no evidence that I know of.


Quote
At the very least, it requires yet another depot, filled to capacity

Not true. You can do less. You can partially fill the tanker and LSS and get "in-between" performance.

I don't think so.  Either you complete the LEO-NRHO-LS-NRHO cycle on one refueling, or you don't.  The problem has multiple constraints.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/27/2022 04:59 am

Going to an EDL-capable tanker saves you some return delta-v, but it doesn't save you huge amounts of prop, because the tanker's wet mass is small after it's transferred the prop to the LSS. [...]

Doesn't this presume that the depot isn't "pre-loaded" with propellant for some later mission?

Yes.  But unless you have high mission cadence (and Artemis certainly does not, so it would have to come from CLPS or commercial flights), the depot's gonna boil dry.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/27/2022 05:09 am
If you really boosted the LSS to LEO+1670, there's not enough prop to abort back to the original VLEO (I chose 300x300) if the refueling fails.  That could be a loss of vehicle for a very expensive LSS asset.

Surely it could aerobrake at some altitude down to a lower apogee, no?

I think we have to assume not.  There's no TPS.  There are no aerosurfaces.  And if there's a coating, it'll get cooked.

This is why you choose "some altitude" that results in no cooking.

Remember that aerobraking can and has been done by regular satellites without any special heat shield. The selected altitude must be within the vehicle's thermal & structural limits.

At the very least, it requires yet another depot, filled to capacity

Not true. You can do less. You can partially fill the tanker and LSS and get "in-between" performance.

I don't think so.  Either you complete the LEO-NRHO-LS-NRHO cycle on one refueling, or you don't.  The problem has multiple constraints.

You misunderstand. I mean that the tanker/depot and the LSS don't have to be both fully refilled in VLEO. You have the option to fill them both somewhere above half full, and the performance scales nicely with the total number of tanker flights.

You can fill both vehicles entirely full, but that's not "the least" you can do. It's also not the most you can do either, since you can always use a HEEO ladder that begins with more than one tanker in VLEO.

I did all the spreadsheet-smithing here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2229093#msg2229093) (including support for partial fills AND different dry masses for tankers and the mission Starship), but I suspect I'm the only human on planet Earth who can use that spreadsheet correctly.  :o :o
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/27/2022 11:11 am
You misunderstand. I mean that the tanker/depot and the LSS don't have to be both fully refilled in VLEO. You have the option to fill them both somewhere above half full, and the performance scales nicely with the total number of tanker flights.

You can fill both vehicles entirely full, but that's not "the least" you can do. It's also not the most you can do either, since you can always use a HEEO ladder that begins with more than one tanker in VLEO.

I did all the spreadsheet-smithing here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2229093#msg2229093) (including support for partial fills AND different dry masses for tankers and the mission Starship), but I suspect I'm the only human on planet Earth who can use that spreadsheet correctly.  :o :o

I suppose you can ladder the tanker side of the equation to assemble a single, full refill for a 1200t LSS at the lowest possible orbit, and that does indeed do the job with 1200t tankers only.  But you're increasing mission complexity on the tanker side, with fairly heavily-laden tankers refueling each other in weird orbits.

Beyond that, the same two questions:

1) Can you do everything with only 1200t tankage?  If you're willing to incur the complexity, then yes.

2) Does it reduce the number of tanker launches?  I haven't cranked this through, but I'd guess that it's an extra fraction of a tanker launch to do.

Frankly, this is why I like the 1500t LSS.  All of this complexity falls away, and all refueling is done in the lowest-possible energy orbit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/27/2022 04:29 pm
You used the same dry mass for both the unstretched and the stretched lunar Starship (cells B7 and F7). How do you plan on stretching tanks without increasing dry mass?

Also, I think dry mass could plausibly be reduced for the unstretched version.

You’re trading some cylindrical payload bay height for more prop. Just swap the order of ring and dome segments. You still wind up with 2m of cylindrical height and an unspecified amount of ogive space (depending on how big the crew module is). But the dry mass is the same.
Last we saw into the ogive, it was tight but had some room. A dragon crew module proxy wouldn't need to be a conic section. No need for any of the fin mechanism and related reinforcing. IIRC, there should be room.


If the landing engines end up staying in the plan the 2m above the top dome would be dedicated to them and set a design restraint on size. This screws with cargo storage and the elevator. Another argument for a stretched hull - if the landing engines stay.


For you and everybody else down that way, stay safe and ride it out.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/27/2022 08:43 pm
Last we saw into the ogive, it was tight but had some room. A dragon crew module proxy wouldn't need to be a conic section. No need for any of the fin mechanism and related reinforcing. IIRC, there should be room.

If you used most of the ogive (using the static envelope specs from the Starship User Guide, from height 17.24m down to 11m, with radius 1m to 3.27m) as pressurized volume for a crew, that would give you something like 98m³ of volume--more than 10x the pressurized volume of a D2.

And with 1500t tanks, you'd still have 2.3m of cylindrical payload height, plus a couple of meters of (wide) ogive height.  That's an unimaginable amount of cargo to be co-manifested on a crewed mission.  If you need more than that, use an uncrewed mission.

Quote
If the landing engines end up staying in the plan the 2m above the top dome would be dedicated to them and set a design restraint on size. This screws with cargo storage and the elevator. Another argument for a stretched hull - if the landing engines stay.

I'd expect the thrusters to mostly be around the waist of the LCH4 dome.  And even if they extend up above that, they'll only take up a small amount of space around the circumference of the payload deck.

Even the depot/tanker variants will want some space for depot-specific hardware.  I wouldn't want to eat more than 6m of the cylindrical payload space, for either the tanker/depot or the LSS. That would limit the max prop stretch to 1520t.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 09/28/2022 05:09 pm
Last we saw into the ogive, it was tight but had some room. A dragon crew module proxy wouldn't need to be a conic section. No need for any of the fin mechanism and related reinforcing. IIRC, there should be room.

If you used most of the ogive (using the static envelope specs from the Starship User Guide, from height 17.24m down to 11m, with radius 1m to 3.27m) as pressurized volume for a crew, that would give you something like 98m³ of volume--more than 10x the pressurized volume of a D2.

And with 1500t tanks, you'd still have 2.3m of cylindrical payload height, plus a couple of meters of (wide) ogive height.  That's an unimaginable amount of cargo to be co-manifested on a crewed mission.  If you need more than that, use an uncrewed mission.

Quote
If the landing engines end up staying in the plan the 2m above the top dome would be dedicated to them and set a design restraint on size. This screws with cargo storage and the elevator. Another argument for a stretched hull - if the landing engines stay.

I'd expect the thrusters to mostly be around the waist of the LCH4 dome.  And even if they extend up above that, they'll only take up a small amount of space around the circumference of the payload deck.

Even the depot/tanker variants will want some space for depot-specific hardware.  I wouldn't want to eat more than 6m of the cylindrical payload space, for either the tanker/depot or the LSS. That would limit the max prop stretch to 1520t.
We don't have a physical size for the landing engines so staking out territory is a bit iffy. Don't forget plumbing, work space and thrust structure. The higher the feed pressure the smaller the engines can be for a given thrust. Maybe high pressure COPV's are a good trade here. As long as there's enough cylindrical space for a usable hatch it'll work out.


OTOH, maybe they'll go away. KFDI, but maybe.


On the upside, it's looking like Ian is cutting south of you and there'll be less storm surge. Talk about looking for a bright spot in a bad situation...
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 09/28/2022 06:04 pm
We don't have a physical size for the landing engines so staking out territory is a bit iffy. Don't forget plumbing, work space and thrust structure. The higher the feed pressure the smaller the engines can be for a given thrust. Maybe high pressure COPV's are a good trade here. As long as there's enough cylindrical space for a usable hatch it'll work out.

You can figure out their thrust pretty well.  95t dry + 15t crew + maybe 15t co-manifested cargo + 170t prop for ascent to NRHO = 295t.  For hover in lunar 1.62m/s² (which might not be necessary, but let's assume), that's 478kN.  Figure eight thrusters, and you're below SuperDraco scale, albeit with methalox or methox and igniters.

I wouldn't be surprised to see the landing thrusters combined with the RCS system.  They're not going to be huge.

For heavier payloads, assume that there are landing pads and nav support.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/30/2022 07:29 pm
You misunderstand. I mean that the tanker/depot and the LSS don't have to be both fully refilled in VLEO. You have the option to fill them both somewhere above half full, and the performance scales nicely with the total number of tanker flights.

You can fill both vehicles entirely full, but that's not "the least" you can do. It's also not the most you can do either, since you can always use a HEEO ladder that begins with more than one tanker in VLEO.

I did all the spreadsheet-smithing here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2229093#msg2229093) (including support for partial fills AND different dry masses for tankers and the mission Starship), but I suspect I'm the only human on planet Earth who can use that spreadsheet correctly.  :o :o

I suppose you can ladder the tanker side of the equation to assemble a single, full refill for a 1200t LSS at the lowest possible orbit, and that does indeed do the job with 1200t tankers only.  But you're increasing mission complexity on the tanker side, with fairly heavily-laden tankers refueling each other in weird orbits.

"Heavily laden" and "weird orbits" seems to be nothing but FUD phraseology. No technical justification is offered for why any of this would be impractical.

Beyond that, the same two questions:

1) Can you do everything with only 1200t tankage?  If you're willing to incur the complexity, then yes.

My spreadsheet allows you to specify the tank volume of tankers and mission vehicles independently, so this use case is supported.

2) Does it reduce the number of tanker launches?  I haven't cranked this through, but I'd guess that it's an extra fraction of a tanker launch to do.

In almost all cases, yes. Again, all the calculations are available in the spreadsheet. Just plug in your assumptions.

Frankly, this is why I like the 1500t LSS.  All of this complexity falls away, and all refueling is done in the lowest-possible energy orbit.

Ironically, with my harmonic refueling it also happens in the lowest "possible" orbit too. That's the driving principle behind the optimization, in fact!  :D
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/01/2022 06:13 pm
Beyond that, the same two questions:

1) Can you do everything with only 1200t tankage?  If you're willing to incur the complexity, then yes.

My spreadsheet allows you to specify the tank volume of tankers and mission vehicles independently, so this use case is supported.

2) Does it reduce the number of tanker launches?  I haven't cranked this through, but I'd guess that it's an extra fraction of a tanker launch to do.

In almost all cases, yes. Again, all the calculations are available in the spreadsheet. Just plug in your assumptions.

Frankly, this is why I like the 1500t LSS.  All of this complexity falls away, and all refueling is done in the lowest-possible energy orbit.

Ironically, with my harmonic refueling it also happens in the lowest "possible" orbit too. That's the driving principle behind the optimization, in fact!  :D

I believe that laddering makes sense when extreme C3 outweighs the need for mission simplicity.  However, for both lunar and Mars missions, that's simply not the case.  In the case of an Artemis mission, you'll want to limit the LSS to a single refueling, simply to reduce risk.  That requires limiting the maximum refueling energy to something that the LSS can reach with its launch prop.  In the case of a Mars mission, with crew already on board, a single refueling in VLEO is more than adequate.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Barley on 10/02/2022 12:59 am
I believe that laddering makes sense when extreme C3 outweighs the need for mission simplicity.  However, for both lunar and Mars missions, that's simply not the case.  In the case of an Artemis mission, you'll want to limit the LSS to a single refueling, simply to reduce risk.  That requires limiting the maximum refueling energy to something that the LSS can reach with its launch prop.  In the case of a Mars mission, with crew already on board, a single refueling in VLEO is more than adequate.
E = 1/2*mv2

The other type of high energy mission is sending a lot of mass.  Laddering through a high energy Earth orbit significantly* increase the mass that can be sent to Mars in a single ship.  Fewer ships tied up in transit and fewer ships that need refueling at Mars.

I think we differ in that you think refueling is complex, risky and difficult.  I think refueling will become routine and easy.  Up to now almost all of the rendezvous and docking has been in human space flight.  Not only are human pilots bad at this but the risk to humans has resulted in extreme conservatism and overly complex procedures.  I expect SpaceX to try new procedures, blow up a few robots and end up with something far faster, safer and more reliable.


* By a factor of 13 if you ladder through a trans-moon injection orbit and use aero braking from Mars transfer orbit down to low Martian orbit or the surface.  This allows an absurd mass of almost 8300t of cargo to Mars in a single SS.  This would allow for example a storm shelter with 10m of water shielding.  (Fitting 10m of water in a 9m SS is left as an exercise, perhaps it would be better to use aluminum.)  Laddering through GTO "only" increase the cargo by a factor of 4.75, which is still very useful.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/02/2022 04:03 am
I believe that laddering makes sense when extreme C3 outweighs the need for mission simplicity.  However, for both lunar and Mars missions, that's simply not the case.  In the case of an Artemis mission, you'll want to limit the LSS to a single refueling, simply to reduce risk.  That requires limiting the maximum refueling energy to something that the LSS can reach with its launch prop.  In the case of a Mars mission, with crew already on board, a single refueling in VLEO is more than adequate.
E = 1/2*mv2

The other type of high energy mission is sending a lot of mass.  Laddering through a high energy Earth orbit significantly* increase the mass that can be sent to Mars in a single ship.  Fewer ships tied up in transit and fewer ships that need refueling at Mars.

I think we differ in that you think refueling is complex, risky and difficult.  I think refueling will become routine and easy.  Up to now almost all of the rendezvous and docking has been in human space flight.  Not only are human pilots bad at this but the risk to humans has resulted in extreme conservatism and overly complex procedures.  I expect SpaceX to try new procedures, blow up a few robots and end up with something far faster, safer and more reliable.


* By a factor of 13 if you ladder through a trans-moon injection orbit and use aero braking from Mars transfer orbit down to low Martian orbit or the surface.  This allows an absurd mass of almost 8300t of cargo to Mars in a single SS.  This would allow for example a storm shelter with 10m of water shielding.  (Fitting 10m of water in a 9m SS is left as an exercise, perhaps it would be better to use aluminum.)  Laddering through GTO "only" increase the cargo by a factor of 4.75, which is still very useful.

Four objections:

1) Unless you're using magic at Mars, there's a limit to how much mass you can land using aerodynamic EDL.  I'd frankly be a bit surprised if it's as much as 150t, and anything more seems like you'll run into ballistic coefficient problems.

2) Ridiculously high C3's to Mars need to be offset by a lot of propulsive braking before EDL.  I can't remember who explained this to me (apologies for no citation), but the quick-and-dirty version of this is to look at the centripetal acceleration needed to keep the Starship in the atmosphere at a particular altitude, so it can kill its speed efficiently.  If you want to keep the acceleration below 8g at 100km altitude, you can't hit periapse at more than 16.5km/s--even if you had the unobtanium TPS that allowed that kind of thermal load.

3) Finally, you can ladder as much as you want up to just a smidge below Earth escape speed, but then things get weird.  Any additional tankers have to leave in roughly the same orbit as the Starship.  That gives you two unpalatable choices:

a) You can fill them all completely full in C3=-somethingSmall, but to get maximum Oberth effect, the Starship wants to use as much of its prop as possible, which means that the tankers have to use pretty much the same amount (unless they're bigger).  Net result:  Even if you plan to ladder from here, there's very little prop to ladder with.

b) You can reduce the Oberth effect and ladder at any n:m ratio you want.  But when I worked this out, the C3 you lose from the less-aggressive perigee burn is hard to make up even with efficient laddering.  Sometimes as much as 3:2 makes sense, but you need to be really screaming before it does--and it's still a massive amount of prop to LEO.

We slogged through some of this over on the extrasolar missions thread.  It kinda starts about here (https://forum.nasaspaceflight.com/index.php?topic=55550.msg2337534#msg2337534), but you have to rummage through lots of posts on other topics to figure it out.

4) It's not that I think that refueling in and of itself is hard, although it certainly will have risks that need to be retired before anybody conservative (e.g. NASA) will think about anything other than a single uncrewed refueling of an expensive asset like an LSS.  The real problem is that everything is harder in an eccentric orbit.  Fiddling with arguments of perigee (or perihelion) is expensive.  Minor RAAN errors are hard to correct.  Phasing takes a long time, because orbital periods are longer.  There are the radiation problems.  Missed-burn contingencies become more and more nightmarish.  There's a lot of operational art that doesn't exist, which will need to be developed.

I'm perfectly happy with a depot in a circular VLEO.  I'm somewhat less happy with a depot or tanker in NRHO, because CR3BP orbits are weird and nobody's done rendezvous in them yet, but you have plenty of time to fool around without things getting too weird.  But HEEO or its heliocentric cousin seems like a lot of trouble for not a lot of gain.  (Again, the obvious exception is a mission that can really benefit from super-high C3.  But Mars ain't it.)

I admit to thinking no more than 10 years out on refueling, and there's nothing but cislunar and maybe a few Mars trips in that timeframe.  At some point in the dim distant future, the operational art will exist, and fancier things will have firm business cases attached to them.  But it's science fiction for right now.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/02/2022 12:25 pm
Exactly how much mass does need to be sent to Mars? It seems like we've worked through the numbers for the lunar case pretty thoroughly, but I haven't seen anything comparable for Mars.

For the moon, I think the conclusion is that you need a depot in LEO that gets filled with ~1500 metric tons of propellant via multiple flights by tanker Starships. Then you stretch the lunar Starship enough to hold 1500 metric tons of propellant. Even though it can't launch with that much, you can fill it up once it gets to the depot. From there, it can do the full mission to NRHO, to the moon, and back to NRHO. We haven't worked out how you might reuse a lunar Starship, but NASA clearly doesn't plan on that.

So what's the equivalent process for a Mars Starship? Surely 1500 mt of prop in LEO isn't enough to take you to Mars! Is it? It seems like you'd want another depot much higher up to give the Mars Starship a second refill--maybe in NRHO--and do something like a gravitational assist off the moon to do a deep dive close to the Earth to maximize the Oberth burn. (I'm just making this up as I go--I'm hoping someone has a better-thought-out plan.)

Apologies if this has already been thoroughly debated elsewhere.

Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 10/02/2022 12:34 pm
You need less propellant to go to Mars than land on the Moon, let alone do the ~7-8km/s mission that Starship HLS needs. So no 1200t depot required. 3.7-3.9km/s or so (depending on opportunity… 3.5km/s for TMI, 200-400m/s for landing) is about the minimum amount. 500t would even be enough.

Also, quit taking TheRadicalModerate’s figures for the Moon as gospel.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 10/02/2022 12:47 pm
Pretty annoying when people who post long pages about what “MUST” be required for lunar missions are taken as gospel because they post a lot, while making unqualified assumptions about dry mass that drive their conclusions.

…yes, yes, glass houses and stones, but you shouldn’t take my prognostications as gospel either!
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/02/2022 01:37 pm
You need less propellant to go to Mars than land on the Moon, let alone do the ~7-8km/s mission that Starship HLS needs. So no 1200t depot required. 3.7-3.9km/s or so (depending on opportunity… 3.5km/s for TMI, 200-400m/s for landing) is about the minimum amount. 500t would even be enough.

Ah, so the key things that make this possible (correct me if I'm wrong) are a) The Mars trip is one-way b) using the Mars atmosphere for braking saves you a lot of delta-v c) Although the trip to Mars takes much longer, the delta-v to go from LEO to a minimum energy TMI isn't really very much more than what's required for TLI. Is that a fair summary?

Also, quit taking TheRadicalModerate’s figures for the Moon as gospel.

Pretty annoying when people who post long pages about what “MUST” be required for lunar missions are taken as gospel because they post a lot, while making unqualified assumptions about dry mass that drive their conclusions.

…yes, yes, glass houses and stones, but you shouldn’t take my prognostications as gospel either!

Actually, my perception was that a) his logic and his math seem reasonable to me, given his assumptions (I looked over his spreadsheet at length) b) other people's disagreements were all complaining his assumptions were too conservative (e.g. his dry-mass estimate could be too high) but c) even with those assumptions, he sees a fairly simple way to make the system work.

So it seems plausible to me that the system SpaceX will end up using (particularly in light of the Marshall paper) will be something close to what he describes. Doubtless there will be differences, and it may be able to carry more payload than he thinks and/or need fewer refueling flights, but the basic idea seems sound to me.

Do you genuinely feel that this is grossly wrong? Or is your concern just that, despite the precise numbers in his spreadsheet, no one should think that all the details are correct?


Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Robotbeat on 10/02/2022 02:23 pm
If you use overly conservative assumptions, your conclusions about what is optimal to do is going to be wrong. This is partly why NASA assumed a 3 stage lunar lander but SpaceX is using one stage. Conservative dry mass leaves you with a whole bunch of stages.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/02/2022 04:07 pm
If you use overly conservative assumptions, your conclusions about what is optimal to do is going to be wrong. This is partly why NASA assumed a 3 stage lunar lander but SpaceX is using one stage. Conservative dry mass leaves you with a whole bunch of stages.
So do you have a simpler approach?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/02/2022 06:20 pm
If you use overly conservative assumptions, your conclusions about what is optimal to do is going to be wrong. This is partly why NASA assumed a 3 stage lunar lander but SpaceX is using one stage. Conservative dry mass leaves you with a whole bunch of stages.

I never did get around to cranking the very low inert mass and 1200t tanks through the model--an oversight.  If the combined dry + crew module mass is 95t or less, 1200t will work straight from VLEO.  You could take that up to 100t if you were willing to forgo any non-crew payload, which sounds like a mistake.

This isn't a completely unreasonable number, especially if you're willing to skimp on the crew module.  If they manage to get the vanilla Starship down to 100t, then the LSS version might get down to 75t, and you'd be in business.  But there's basically zero design margin, and you're left with having to do something exotic¹ if you ever want to stage the LSS out of VLEO instead of NRHO.

Model added to the others (https://docs.google.com/spreadsheets/d/17NNg4iHZDn98zW8Zq_4JmG04unBC5fR31-C_5Z1F3cM/edit?usp=sharing).

______________
¹Update:  The "something exotic" statement is unwarranted.  If you limit total non-crew cargo to 5t, then you could have a 105t dry+crew configuration that would work fine.  I was thinking (without computing) that the requirement to do fast transit instead of a BLT would do you in, but that's more than offset by the fact that you can do LEO-LLO-LS-NRHO when staging out of LEO, eliminating the extra delta-v to get from NRHO down to LLO.

105t is only about 5t lighter than my current "conservative" estimate for dry+crew.  That might have enough margin to be viable.

Note that staging from LEO still requires a refueling, post-lunar-ascent, before returning to LEO propulsively.  The only way you can avoid refueling in cislunar is if you refuel in some kind of fairly energetic HEEO.  I maintain that would be exotic, because you'd wind up subjecting the crew to VA Belt radiation.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: redneck on 10/03/2022 01:15 am
If two fully fueled ships leave LEO docked with crossover plumbing, the attached tanker can empty in a very few minutes. Tanker with 1,500 tons of fuels and LSS with 1,200 tons of fuels at ignition. Tanker empty and detached at about 2,500 m/s while fully fueled LSS continues on to the Lunar mission. Tanker falls back from apogee reenters to base.   

Skimmed through a lot of this discussion without seeing this possibility suggested.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TomH on 10/03/2022 02:07 am
You need less propellant to go to Mars than land on the Moon, let alone do the ~7-8km/s mission that Starship HLS needs. So no 1200t depot required. 3.7-3.9km/s or so (depending on opportunity… 3.5km/s for TMI, 200-400m/s for landing) is about the minimum amount. 500t would even be enough.

Ah, so the key things that make this possible (correct me if I'm wrong) are a) The Mars trip is one-way b) using the Mars atmosphere for braking saves you a lot of delta-v c) Although the trip to Mars takes much longer, the delta-v to go from LEO to a minimum energy TMI isn't really very much more than what's required for TLI. Is that a fair summary?

Regarding a), some Mars trips will be one way, while others are round trip. This does differ from Luna, however, in that propellant for the return trip, both fuel and oxidizer, will be manufactured on Mars. Lunar trips will offer no aerobraking and also will require return prop to be carried to the surface and re-launched from the surface. (I am a supporter of leaving some prop in LLO and repropping there prior to TEI.) So even though the moon is closer, it has two significant disadvantages when contrasted with Mars.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Craigles on 10/03/2022 03:07 am
As an aside but still topical for this in-orbit refueling discussion: We are focusing here on good strategies for orbits, ΔV, proximity operations, and fuel settling, all of which is fine. But the risk that worries me is the risk of chopsticks landing accidents. Fireballs have bad optics. Falcon 9 boosters land superbly, but soon SH/SS refueling needs even higher precision landings.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/03/2022 05:12 am
If two fully fueled ships leave LEO docked with crossover plumbing, the attached tanker can empty in a very few minutes. Tanker with 1,500 tons of fuels and LSS with 1,200 tons of fuels at ignition. Tanker empty and detached at about 2,500 m/s while fully fueled LSS continues on to the Lunar mission. Tanker falls back from apogee reenters to base.   

Skimmed through a lot of this discussion without seeing this possibility suggested.

This architecture does get discussed from time to time.  The biggest problems are that it's very odd structurally, which makes it very odd from a control standpoint (i.e., keeping both ships pointed in the same direction with no significant torques), and the cross-feed plumbing, which has to pump 6.2t/s of mass flow, isn't straightforward.  Note that you'd also either have to double the tanker's amount of prop diverted to autogenous pressurization, or engineer a path to get pressurant from the cross-fed ship back to the tanker supplying the prop.

As an alternative, if you could swing the nose cone of the tanker away and dock it with the tail of your target Starship, everything works like a regular 2-stage rocket.  Then you can do the following conops:

1) Tanker launches and is completely refueled by a depot.
2) Tanker swings nosecone out of the way, exposing docking interface.
3) Tanker docks nose-to-tail with the target Starship.
4) Tanker pumps as much prop into the Starship as possible.
5) While still docked, tanker starts TLI (or TMI) burn.
6) When tanker is down to about 300m/s of delta-v, it shuts down and separates.
7) Target Starship immediately continues burn, does rest of mission.
8) Tanker coasts to apogee, does short burn to lower perigee to reentry altitude.
9) Tanker closes nose cone and does EDL.

But this requires engineering a nose-to-tail docking mechanism, which isn't straightforward.  I suspect that there were good reasons why SpaceX gave up on tail-to-tail docking, and this would be worse.  However, I also suspect that this would be easier to engineer than side-by-side burns with cross-feed.

Update:  Here's (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579) a three-year-old thread with the idea in a slightly different form.  And I'll add that this is nowhere near as simple as stretching the LSS tanks a bit.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/03/2022 05:34 am
As an aside but still topical for this in-orbit refueling discussion: We are focusing here on good strategies for orbits, ΔV, proximity operations, and fuel settling, all of which is fine. But the risk that worries me is the risk of chopsticks landing accidents. Fireballs have bad optics. Falcon 9 boosters land superbly, but soon SH/SS refueling needs even higher precision landings.

A more general way of capturing this is that there's a requirement for some average minimum cadence for the lift tankers over the course of the mission's refueling campaign.  If you blow up a Mechazilla, it goes out of service for repairs.  If it's the only Mechazilla, this is... bad... but if there's a second one, then your cadence degrades a lot more smoothly.  If you have a high risk of blowing up all of the landing systems within the course of a single campaign, then you have a more serious problem.

I think the bottom line on the risk is that SpaceX tried landing legs and didn't like the result.  It is absolutely a requirement that landing be sorta-kinda reliable.  If the chopsticks aren't reliable, Plan B is probably landing legs.  But I suspect that somebody did the math before moving forward with Plan A.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/03/2022 08:25 am
My thought on the side to side docked and cross feed is along the lines of the heavy versions of Delta and Falcon with two cores instead of three. If it were three, the mass ratio would put it beyond escape and getting the boosters back would be a problem. Plus the extra dry mass of another core. 

The tanker pressurant would be the same in either case to empty the tank of liquid whether it is in x minutes or 1/2x minutes.

It seems the tanker would not necessarily need a burn at apogee to put it on a reentry trajectory as long as the TLI trajectory was creating an HEEO with perigee at reentry altitude.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: tbellman on 10/03/2022 03:08 pm
But this requires engineering a nose-to-tail docking mechanism, which isn't straightforward.  I suspect that there were good reasons why SpaceX gave up on tail-to-tail docking, and this would be worse.  However, I also suspect that this would be easier to engineer than side-by-side burns with cross-feed.

The reason given by Elon/SpaceX for giving up on tail-to-tail refilling, was that they realized that they needed a stabilizing arm at the bottom of the ship when it was standing on the launch pad.  They then realized that they could use that arm to fill the tanks of the ship, instead of having plumbing through the entire length of SuperHeavy, filling the ship through the same piping as the booster.  This probably saved them a couple of tonnes on SuperHeavy, and a bunch of valves and other complexity (moving that to stage zero).  And when they had propellant connectors on the side of the ship for filling it when on the launch pad, then they would of course use that when filling it on orbit as well.

I think Elon said this during Tim Dodd's first interview/tour of Starbase (https://forum.nasaspaceflight.com/index.php?topic=54457.0), but I don't have time to listen through it at the moment.  So the above is extracted just from my memory.

EDIT: Yes, in part 2 of the interview (https://youtube.com/watch?v=SA8ZBJWo73E) at 30:02 Elon talks about orbital refilling not being butt-to-butt, and starting at 34:57 they talk about the quick disconnect-arm being used to stabilize the stack.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/03/2022 08:55 pm
If two fully fueled ships leave LEO docked with crossover plumbing, the attached tanker can empty in a very few minutes. Tanker with 1,500 tons of fuels and LSS with 1,200 tons of fuels at ignition. Tanker empty and detached at about 2,500 m/s while fully fueled LSS continues on to the Lunar mission. Tanker falls back from apogee reenters to base.   

Skimmed through a lot of this discussion without seeing this possibility suggested.
How fast props can be transferred is an open question. IIUC, you're suggesting the two ships be locked together while thrusting. This is done with FH at launch but my gut says that this is different enough to need some serious testing. Not a deal breaker but it is something new in the mix.


I've been noodling an elliptical orbit with apogee at L1. It only takes a puff to stop and it's literally down hill in either direction.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/03/2022 08:59 pm
As an aside but still topical for this in-orbit refueling discussion: We are focusing here on good strategies for orbits, ΔV, proximity operations, and fuel settling, all of which is fine. But the risk that worries me is the risk of chopsticks landing accidents. Fireballs have bad optics. Falcon 9 boosters land superbly, but soon SH/SS refueling needs even higher precision landings.
We need to understand the mission before we can judge what details even need to be considered. All on topic, IMO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/03/2022 09:36 pm
My thought on the side to side docked and cross feed is along the lines of the heavy versions of Delta and Falcon with two cores instead of three. If it were three, the mass ratio would put it beyond escape and getting the boosters back would be a problem. Plus the extra dry mass of another core.

Applying lots of shear stress to something that docked, rather than something that was pre-stressed on the ground with heavy equipment, seems pretty challenging.  The only way I can think that a tandem burn would work is if the accelerations of both ships are identical.

Quote
The tanker pressurant would be the same in either case to empty the tank of liquid whether it is in x minutes or 1/2x minutes.

It's the "x minutes vs. 1/2x minutes" that's the problem.  1/2x implies double the pressurant mass flow (rate), and it all has to go in the tanker if you want to preserve the prop in the target Starship.

That pressurant is tapped off from the Raptor pumps, fed through a heat exchanger, and then piped up to the ullage space in each tank.  If you're pumping out prop at twice the rate (for both the tanker and the target Starship), either the engines on the tanker need to handle twice the tap-off flow (a decent-sized design change) or you need to import hot ullage gas from the engines on the target Starship (more complex inter-ship plumbing).

Quote
It seems the tanker would not necessarily need a burn at apogee to put it on a reentry trajectory as long as the TLI trajectory was creating an HEEO with perigee at reentry altitude.

If you do a burn from a circular orbit or from perigee, the TLI can't have a lower perigee; it can only have a higher  apogee.

If you apply 2500m/s of delta-v to a pair of mated spacecraft in a 300x300km orbit, you wind up with the tanker in a 300x40,800km orbit.  If you then want to lower your 300km perigee to, say, 70km for entry, you need to do an apogee burn with a retrograde delta-v of 22m/s.  Not a lot, but not zero.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: edzieba on 10/04/2022 08:20 am
If two fully fueled ships leave LEO docked with crossover plumbing, the attached tanker can empty in a very few minutes. Tanker with 1,500 tons of fuels and LSS with 1,200 tons of fuels at ignition. Tanker empty and detached at about 2,500 m/s while fully fueled LSS continues on to the Lunar mission. Tanker falls back from apogee reenters to base.   

Skimmed through a lot of this discussion without seeing this possibility suggested.
How fast props can be transferred is an open question. IIUC, you're suggesting the two ships be locked together while thrusting. This is done with FH at launch but my gut says that this is different enough to need some serious testing. Not a deal breaker but it is something new in the mix.
Falcon Heavy does not perform propellant cross-feed. This was dropped as being 'too complex' for insufficient benefit in actual performance.
As far as I am aware, no vehicle has flown with cross-feed between stages both under power, with the closest being feeding from a dedicated tank structure to a dedicated engine structure with the two being in-flight separable (STS, stage-and-a-half Atlas).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/04/2022 09:06 am
What I would like is an extremely lightweight drop tank (20 ton tank 1,200 ton prop???) mated to the LSS as it leaves LEO. I don't see that being feasible short term due to the logistics of getting it in orbit. Plus development, attachment and several other issues. Possibly not even long term as it might not make any sense with Lunar ISRU. Trying to not engage in too much handwavium engineering.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 10/04/2022 05:46 pm
What I would like is an extremely lightweight drop tank (20 ton tank 1,200 ton prop???) mated to the LSS as it leaves LEO. I don't see that being feasible short term due to the logistics of getting it in orbit. Plus development, attachment and several other issues. Possibly not even long term as it might not make any sense with Lunar ISRU. Trying to not engage in too much handwavium engineering.
If it's extremely light weight, there is less need to actually drop the tank.  This becomes roughly equivalent to a stretched Starship with 2400 tonne capacity. 

I think it would be best to wait until there is a working Starship to see how far the design can be pushed before either scheme is attempted.  Note that the current F9 is almost double the mass of V9 v1.0.  This does not say that SS can do the same, but it doesn't say it can't
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/04/2022 10:16 pm

[/quote]
If it's extremely light weight, there is less need to actually drop the tank.  This becomes roughly equivalent to a stretched Starship with 2400 tonne capacity. 

I think it would be best to wait until there is a working Starship to see how far the design can be pushed before either scheme is attempted.  Note that the current F9 is almost double the mass of V9 v1.0.  This does not say that SS can do the same, but it doesn't say it can't
[/quote]

You triggered this vision of a tankerish Starship launched with extremely extended tanks that reaches orbit empty and with little payload due to extreme stretch. It is docked to the true payload and refueled with 2,400+ tons of propellant for the Lunar mission. A somewhat more feasible method than what I had suggested.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: LMT on 10/04/2022 10:32 pm

If a tanker had 12-hour turnaround, and it delivered 1/5 of a load, it could fill an inflatable depot for ~ 300 crews each synod, by itself. 

Otherwise, a fleet of tankers would be needed.  E.g., given a 14-day window and preloading of depot tankers at synod start, at least 46 tankers would be needed. 

Scale up ~ 10x (https://aeon.co/essays/elon-musk-puts-his-case-for-a-multi-planet-civilisation) for settlement cargo flights.

The obvious idea is to use the Starships themselves as the "depot." This eliminates the entire depot fleet, which is a massive savings.

Instead you just transfer the passengers at the last minute. You can use 500 passenger P2P-derived vehicles to do that. This further increases the efficiency of launching within that 14-day window, because each launch serves five Starships instead of one.

Delete depots, get taxis.

No, you wouldn't want to haul all depot hw (ruggedized ZBO and propellant transfer hw) on every ship, to Mars and back.  It cuts cargo, and of course requires duplicate crewed systems.  There's little gain -- and little margin for Mars return.

Passive depot inflatables would be a big win, but the efficient fleet depot requires creative thought.

Also, ZBO hw makes more sense on the notional LEO/Deimos tanker ISRU loop, where it eliminates not the depot, but Earth-launch tankers and their SH boosters.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/05/2022 04:53 am
Quote
If it's extremely light weight, there is less need to actually drop the tank.  This becomes roughly equivalent to a stretched Starship with 2400 tonne capacity. 

I think it would be best to wait until there is a working Starship to see how far the design can be pushed before either scheme is attempted.  Note that the current F9 is almost double the mass of V9 v1.0.  This does not say that SS can do the same, but it doesn't say it can't

You triggered this vision of a tankerish Starship launched with extremely extended tanks that reaches orbit empty and with little payload due to extreme stretch. It is docked to the true payload and refueled with 2,400+ tons of propellant for the Lunar mission. A somewhat more feasible method than what I had suggested.

There are sharply diminishing returns to adding more propellant.  Remember that more tankage means more dry mass, so that the mass ratio changes slower than you think as you add prop.  In addition, a Starship with a heavy payload already has a mass ratio of 5.4.  Even if your dry mass doesn't change at all (which is unlikely), doubling the propellant only adds 35% more delta-v.  And if the structural mass fraction stays above 6% (which is pretty aggressive for a vehicle like Starship, which, at 120t dry and 1200t prop, has an SMF of 9.09%), then doubling the prop only gets you 30%.

Instead of just blindly throwing prop at Starship, you need to ask what you're trying to achieve.  When you narrow it down to a variety of round-trip cislunar missions and one-way Mars missions, I haven't found a case where stretching the tanks past 1500t helps (although that extra 300t helps a lot, just because of how the delta-v budget to and from cislunar works out), nor have I found any cases where pusher/depot, tandem burns, or even drop tanks help very much.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/05/2022 04:03 pm
There are sharply diminishing returns to adding more propellant. . . . Instead of just blindly throwing prop at Starship, you need to ask what you're trying to achieve.  When you narrow it down to a variety of round-trip cislunar missions and one-way Mars missions, I haven't found a case where stretching the tanks past 1500t helps (although that extra 300t helps a lot, just because of how the delta-v budget to and from cislunar works out), nor have I found any cases where pusher/depot, tandem burns, or even drop tanks help very much.
I find myself thinking that a special cargo "tug" with a solar-powered (or nuclear) ion drive that slowly chugs from LEO to (say) NRHO would be worth it at some point. Yeah, it's slow and not very energy-efficient, but if it could move a full depot in (say) a month or two, that'd probably be well worth it at some point, since, with specific impulse of ~5000s it wouldn't need to be refueled very often.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/05/2022 09:36 pm
There are sharply diminishing returns to adding more propellant. . . . Instead of just blindly throwing prop at Starship, you need to ask what you're trying to achieve.  When you narrow it down to a variety of round-trip cislunar missions and one-way Mars missions, I haven't found a case where stretching the tanks past 1500t helps (although that extra 300t helps a lot, just because of how the delta-v budget to and from cislunar works out), nor have I found any cases where pusher/depot, tandem burns, or even drop tanks help very much.

I find myself thinking that a special cargo "tug" with a solar-powered (or nuclear) ion drive that slowly chugs from LEO to (say) NRHO would be worth it at some point. Yeah, it's slow and not very energy-efficient, but if it could move a full depot in (say) a month or two, that'd probably be well worth it at some point, since, with specific impulse of ~5000s it wouldn't need to be refueled very often.

At ~ 30 mN/kW (https://forum.nasaspaceflight.com/index.php?topic=52883.msg2211000#msg2211000), and allowing for LEO night and drag, a 2-month transfer of 1500 t would need > 80 MWe.

That's a square of PV > half a km on a side.

How many ion engines would you need?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/05/2022 10:01 pm
How many ion engines would you need?
I'm told there's a quantity discount. :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/06/2022 08:37 pm
] nor have I found any cases where pusher/depot, tandem burns, or even drop tanks help very much.

You get free deltaV (~2x) with Oberth maneuvers, so getting fuel into a higher energy orbit (HEEO) effectively doubles Isp.

(or alternatively 2.72^2 = 7.4x more fuel).

Is why that maneuver is so valuable, and why ion drives are effectively obsolete with orbital refueling.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/06/2022 08:41 pm
There are sharply diminishing returns to adding more propellant. . . . Instead of just blindly throwing prop at Starship, you need to ask what you're trying to achieve.  When you narrow it down to a variety of round-trip cislunar missions and one-way Mars missions, I haven't found a case where stretching the tanks past 1500t helps (although that extra 300t helps a lot, just because of how the delta-v budget to and from cislunar works out), nor have I found any cases where pusher/depot, tandem burns, or even drop tanks help very much.
I find myself thinking that a special cargo "tug" with a solar-powered (or nuclear) ion drive that slowly chugs from LEO to (say) NRHO would be worth it at some point. Yeah, it's slow and not very energy-efficient, but if it could move a full depot in (say) a month or two, that'd probably be well worth it at some point, since, with specific impulse of ~5000s it wouldn't need to be refueled very often.

This isn't going to happen any time soon.  For it to make any sense at all, the cadence would have to be high enough to warrant a major development effort to produce such a huge SEP tug.  And if the cadence is high, then you'll have a whole bunch of fleet costs with a whole bunch of tugs in transit.  Those fleet costs will likely swamp the cost of adding a few additional (again, high-cadence) lift tanker launches to move prop to cislunar via brute force.

Maybe to focus the discussion a bit, I think there are three¹ major milestones in the use of tankers to handle cislunar operations:

1) Refueling for the Option A and B LSS.  This is important because it's the key to Artemis's continued existence.

2) Refueling for an LSS staged with crew out of LEO, and returning propulsively to LEO, with an F9/D2 or Starliner bringing the crew to/from LEO.  This is important because it does away with the use of SLS/Orion and therefore reduces the cost per mission by $2B-$3B, even before Starship is crew-certified for launch and EDL.

3) Refueling for a not-quite-vanilla Starship, which is capable of landing on the Moon and returning directly to EDL on Earth.  This is almost certainly SpaceX's endgame, but it requires that all-important crew-certification of launch and EDL.

Finding the proper conops for each of these requires balancing:

- Operational complexity
- Risk of refueling and extra RPODs
- The fact that refueling risk is reduced over time
- Tolerance of risk for crewed refueling operations
- Tolerance of risk for uncrewed operations involving spacecraft that will later be crewed (and are therefore expensive)
- Tolerance of risk for dumb tankers and depots
- And finally the cost (mostly in lift tanker launches) of the various conops. 

It is, to put it mildly, a rich trade space.

After fooling with these for a while, I think you need to follow these guidelines:

1) Option A and B missions have to be as simple as possible, and the LSS itself can't be refueled more than once per mission, nor can it be refueled with crew on board.

2) Staging LSS from LEO for eventual propulsive return requires two refuelings per mission (one in LEO and one... somewhere else).  The second refueling has two problems:
a) The crew is on board (unless you do weird stuff using the Gateway).
b) The refueling has to occur post-lunar-ascent, which makes abort contingency planning very dicey.  (There is an option to do pre-descent refueling, but the number of lift tankers needed almost doubles.  Carrying the prop to return to LEO down to the lunar surface and back is expensive.)

3) When crews are able to be launched and landed via Starship, there are two crewed refuelings instead of one (one in LEO, one at some higher energy), but pre-descent refueling becomes less expensive (since the Starship can return to EDL, rather than propulsive LEO), which gives you a lot more wiggle-room on abort contingencies.

It is possible that crew certification for Starship launch and EDL comes early enough that milestone #2 never needs to be dealt with.  I'd be surprised if that were true, but I've been surprised before.

_________________
¹Note that I'm not talking about Mars, not because I don't think it'll happen, but rather because it's easy:  just refuel once in VLEO and you're good to go.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/06/2022 09:24 pm
nor have I found any cases where pusher/depot, tandem burns, or even drop tanks help very much.

You get free deltaV (~2x) with Oberth maneuvers, so getting fuel into a higher energy orbit (HEEO) effectively doubles Isp.

(or alternatively 2.72^2 = 7.4x more fuel).

Is why that maneuver is so valuable, and why ion drives are effectively obsolete with orbital refueling.

Well... not really.

It's certainly true that if prop were magically to appear in a high-energy eccentric orbit, it's worth a lot more to climb up to that orbit to refuel than to refuel at low energy.  But prop doesn't magically appear, at least not until somebody can fling it off the Moon using super-cheap ISRU (aka science fiction).  You have to haul it up through the same gravity well as your target Starship.

This is not to say that there aren't missions enabled by refueling in HEEO; there are.  Getting a full tank of prop at high energy and high eccentricity means that you've got more delta-v to play with at the end of your mission, which is important.  But the cost of that mission, in terms of Starship lift tanker launches, doesn't change for using the HEEO.

Furthermore, if you're trying to get to some target orbit that requires Δvtarget, and you boost up to an HEEO that requires expending Δvheeo, then to get to the target from the HEEO still requires Δvremainder = Δvtarget - Δvheeo.  It can be worse than this if you don't burn at perigee, but it can't be better.

This is all fully consistent with the Oberth Effect, but there's no free lunch, unless you're falling into a gravity well that you didn't pay to get out of in the first place.

My recommendation:  Use vis-viva and you'll never go wrong.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/06/2022 10:00 pm
nor have I found any cases where pusher/depot, tandem burns, or even drop tanks help very much.

You get free deltaV (~2x) with Oberth maneuvers, so getting fuel into a higher energy orbit (HEEO) effectively doubles Isp.

(or alternatively 2.72^2 = 7.4x more fuel).

Is why that maneuver is so valuable, and why ion drives are effectively obsolete with orbital refueling.

Well... not really.

It's certainly true that if prop were magically to appear in a high-energy eccentric orbit, it's worth a lot more to climb up to that orbit to refuel than to refuel at low energy.  But prop doesn't magically appear, at least not until somebody can fling it off the Moon using super-cheap ISRU (aka science fiction).  You have to haul it up through the same gravity well as your target Starship.

This is not to say that there aren't missions enabled by refueling in HEEO; there are.  Getting a full tank of prop at high energy and high eccentricity means that you've got more delta-v to play with at the end of your mission, which is important.  But the cost of that mission, in terms of Starship lift tanker launches, doesn't change for using the HEEO.

Furthermore, if you're trying to get to some target orbit that requires Δvtarget, and you boost up to an HEEO that requires expending Δvheeo, then to get to the target from the HEEO still requires Δvremainder = Δvtarget - Δvheeo.  It can be worse than this if you don't burn at perigee, but it can't be better.

This is all fully consistent with the Oberth Effect, but there's no free lunch, unless you're falling into a gravity well that you didn't pay to get out of in the first place.

My recommendation:  Use vis-viva and you'll never go wrong.

The HEEO refuel only costs one additional fully refueled standard Starship (said Starship reusable).   You use up so little fuel in the Oberth burn you have enough left over to complete the entire rest of the Moon mission (full round trip).

Which is cheaper than developing a custom depot stuck in Lunar orbit.

That's 2400t of fuel to LEO.   How much do the other mission concepts cost in fuel to LEO?   Getting fuel to Lunar orbit isn't going to be cheap.

Granted, the "one starship round trip with HEEO refueling" mission concept does require a Starship capable of  Lunar landing and an aerobraked return to Earth.   Same requirements as Mars, IOTW (but at double the fuel cost).

So won't happen with crew until Starship is crew certified.

My bet is by 2035 this will be the standard way to get to the Moon.   All the other stuff (e.g. Artemis) is a place holder.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tbellman on 10/06/2022 10:06 pm
I find myself thinking that a special cargo "tug" with a solar-powered (or nuclear) ion drive that slowly chugs from LEO to (say) NRHO would be worth it at some point. Yeah, it's slow and not very energy-efficient, but if it could move a full depot in (say) a month or two, that'd probably be well worth it at some point, since, with specific impulse of ~5000s it wouldn't need to be refueled very often.

In addition to the large solar panels as LMT pointed out, the propellant costs for your electrical engines are likely to be rather large.

Delta-v for trans-lunar injection from LEO is 3.2 km/s, if you are doing an impulsive maneuver (i.e. a short, high-thrust burn).  For a low-thrust continous burn over a two month period, we are talking about roughly double the Δv.  Let's be generous and say 6 km/s.  To accelerate a 1500 tonne craft by 6 km/s using engines with specific impulse of 5000 s, requires

        mprop = mfinal * (eΔv / (Isp*g₀) - 1) =
            = 1500 t * (e6 km/s / 49 km/s - 1) ≈
            ≈ 195 tonne

propellant.

Most ion engines use xenon as propellant.  I only found xenon prices from 1999 in a quick search, but then xenon cost $1800/kg when buying in small quantites, so the propellant cost for a single such delivery would be about $350M.  And unfortunately we are unlikely to get a quantity discount, because 195 tonnes is several years worth of xenon production (we produce roughly 50 tonnes per year), so we are likely to drive up the world prices quite significantly.

Krypton is another viable propellant for ion engines.  SpaceX use that for Starlink, because it is so much cheaper than xenon.  The disadvantage is that you need more electrical power for the same thrust, but we can ignore that for this analysis.  Back in 1999, krypton cost $290/kg, so 195 tonnes would cost us $56M.  And krypton is much more abundant than xenon, so we might not affect the world prices much.  At least as long as we only do a very small handful of such 1500t deliveries every year.  We are still talking about maybe $20M in just the cost of krypton propellant, so even with krypton, electric propulsion might not be much of a win for this application...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/06/2022 10:40 pm
I find myself thinking that a special cargo "tug" with a solar-powered (or nuclear) ion drive that slowly chugs from LEO to (say) NRHO would be worth it at some point. Yeah, it's slow and not very energy-efficient, but if it could move a full depot in (say) a month or two, that'd probably be well worth it at some point, since, with specific impulse of ~5000s it wouldn't need to be refueled very often.

In addition to the large solar panels as LMT pointed out, the propellant costs for your electrical engines are likely to be rather large.

Delta-v for trans-lunar injection from LEO is 3.2 km/s, if you are doing an impulsive maneuver (i.e. a short, high-thrust burn).  For a low-thrust continous burn over a two month period, we are talking about roughly double the Δv.  Let's be generous and say 6 km/s.  To accelerate a 1500 tonne craft by 6 km/s using engines with specific impulse of 5000 s, requires

        mprop = mfinal * (eΔv / (Isp*g₀) - 1) =
            = 1500 t * (e6 km/s / 49 km/s - 1) ≈
            ≈ 195 tonne

propellant.

Most ion engines use xenon as propellant.  I only found xenon prices from 1999 in a quick search, but then xenon cost $1800/kg when buying in small quantites, so the propellant cost for a single such delivery would be about $350M.  And unfortunately we are unlikely to get a quantity discount, because 195 tonnes is several years worth of xenon production (we produce roughly 50 tonnes per year), so we are likely to drive up the world prices quite significantly.

Krypton is another viable propellant for ion engines.  SpaceX use that for Starlink, because it is so much cheaper than xenon.  The disadvantage is that you need more electrical power for the same thrust, but we can ignore that for this analysis.  Back in 1999, krypton cost $290/kg, so 195 tonnes would cost us $56M.  And krypton is much more abundant than xenon, so we might not affect the world prices much.  At least as long as we only do a very small handful of such 1500t deliveries every year.  We are still talking about maybe $20M in just the cost of krypton propellant, so even with krypton, electric propulsion might not be much of a win for this application...

It's even worse.   The static mass of a modern X3 ion thruster is 561kg/engine (including the required energy source/sink), which can only supply 35M Newton-seconds of propellant momentum and 626GJ of energy over 45 days.   A Raptor 2, OTOH, only weighs twice that amount and supplies that energy over 10 minutes at 1.6G Newton-seconds of propellant momentum.   That's 20 times the momentum efficiency of the X3 thruster.

Most of the mass of an ion drive thruster isn't tossed out the back as fuel, it's a parasitic mass remaining when the fuel is exhausted.

In the era of refueled chemical rockets, ion drives are obsolete for any conceivable use beyond maintaining orbits.

https://docs.google.com/spreadsheets/d/1mrlKefLWzGxsYtBXWF92a0cXqsPZoKLGlPbQizKHI3g/edit#gid=0
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/07/2022 12:33 am
It's even worse.   The static mass of a modern X3 ion thruster is 561kg/engine (including the required energy source/sink), which can only supply 35M Newton-seconds of propellant momentum and 626GJ of energy over 45 days.   A Raptor 2, OTOH, only weighs twice that amount and supplies that energy over 10 minutes at 1.6G Newton-seconds of propellant momentum.   That's 20 times the momentum efficiency of the X3 thruster.

Most of the mass of an ion drive thruster isn't tossed out the back as fuel, it's a parasitic mass remaining when the fuel is exhausted.
Thanks for the time and effort you guys put into this. I started to try to do this myself, but got frustrated when I couldn't easily find answers to questions like a) what's the best ion engine today b) what are its specific impulse and thrust c) what is the mass of that engine? d) what would the mass be of the power supply (whether solar or nuclear)?

Of course, nothing says that there couldn't be some breakthrough that made for a light-weight ion engine with a light-weight power supply, but, as TheRadicalModerate suggests above, we're trying to talk about things SpaceX might plausibly attempt in the next few years (next decade at the most). I'm now convinced that an ion tug isn't going to be feasible in that time frame--if it ever is. Thanks again for the demonstration.

In the era of refueled chemical rockets, ion drives are obsolete for any conceivable use beyond maintaining orbits.

Even if it's not useful for a tug, what about a high Δv mission--to Neptune, say? Do you really think ion drives have no use whatsoever?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/07/2022 02:05 am
...what about a high Δv mission--to Neptune, say? Do you really think ion drives have no use whatsoever?

If the objective is a crewed Starship mission to an outer planet, you have to get creative with propulsion.  Refuel at both ends, yes, and more than that.  Unfortunately, ion engines can't deliver the thrust needed for the big outer-planet maneuvers.

What's the best use of deployed film?  If PV film can't give the thrust, you can consider a dielectric reflector film instead, for laser propulsion.

Case:  a 2-year crewed mission to Callisto:

1.  A PV GW-scale laser array at the LEO depot boosts the ship to Jupiter.

2.  A comparable retrograde electrodynamic laser array at Jupiter drops the ship into Jovian orbit. 

3.  Land and refuel on icy Callisto.

4.  Reapply engines and lasers for Earth-return delta-v.

Basic delta-v calcs in the DE-STAR thread (https://forum.nasaspaceflight.com/index.php?topic=51510.0).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/07/2022 03:52 am
The HEEO refuel only costs one additional fully refueled standard Starship (said Starship reusable).   You use up so little fuel in the Oberth burn you have enough left over to complete the entire rest of the Moon mission (full round trip).

You use exactly the same amount of delta-v, irrespective of the energy of the HEEO.  If you're trying to get to LEO+3200 and you stop in an LEO+2500 HEEO, then that HEEO to LEO+3200 costs 700m/s.  It's simple addition, nothing more--as long as you always burn at perigee.  There are no Oberth freebies.  However, prop consumption will obviously increase if you refueled more than you needed to.

You'll always be the most prop-efficient if you refuel in the lowest energy orbit that still leaves you with enough prop to meet the remaining mission delta-v requirements.  Things get weird if no HEEO exists that leaves you with enough prop to finish the mission--that's where something like laddering comes in--but you need a pretty extreme conops for that to happen.

Don't get me wrong:  if you can't do the mission with a full tank from VLEO, then an HEEO of just enough energy is the most efficient thing, and will result in the lowest amount of prop needing to be delivered to orbit.  But prop minimization isn't the only constraint.

Quote
Which is cheaper than developing a custom depot stuck in Lunar orbit.

Who said anything about a custom depot?  If you need prop in cislunar and your logistics work out, you send a regular lift tanker, refuel, and the tanker returns to EDL.  For a crewed mission that relies on prop in cislunar to get home, you can't do quite this well, because it may be weeks before the actual mission shows up.  In that case, you don't need a custom depot but you may need a regular VLEO depot that hangs around forever, simply to manage the boil-off.

Quote
That's 2400t of fuel to LEO.   How much do the other mission concepts cost in fuel to LEO?   Getting fuel to Lunar orbit isn't going to be cheap.

Assuming a vanilla Starship can get 150t to LEO (which would make 2400t of prop roughly 16 tankers):

1) Option A 1200t LSS, refueling once in LEO+1420 HEEO (no refueling in VLEO): 9 tankers.  (The HEEO energy is limited by how high the LSS can boost itself after launch without refueling in VLEO; the assumption is that only one refueling RPOD will be allowed this early in the program.)

2) Option A 1500t LSS, refueling only in VLEO: 9 tankers.

3) Option B 1200t LSS already in NRHO, refueled by 1500t lift tanker: 10 tankers.

4) LSS staged with crew from VLEO, second refuel post-lunar-ascent, propulsive return to LEO: 15 tankers.  (For all of the remaining conops, note that the first refuel is always in VLEO.)

5) LSS staged crewed from VLEO, second refuel pre-lunar-descent, propulsive return to LEO: 28 tankers.  (No, really!  It's terrible!  And it requires more than one tanker to NRHO.  But that's a "Doctor, Doctor, it hurts when I did this" kind of problem.)

6) Launch/EDL crew-certified Starship with surface landing, second refuel in HEEO, direct EDL on return: 13 tankers.

7) Launch/EDL crew-certified Starship, second refuel pre-descent in LLO, direct EDL on return: 15 tankers

All of these have anywhere between 5t and 50t of non-crew deployable cargo, depending on how things even out to an integral number of lift tankers.  All get the tankers back to EDL.  They all assume LSS is 95t dry, crew module is 15t.  For EDL-capable lunar versions, Starship is 120t dry, crew module 15t.  Isp=378s everywhere.  FPR, boil-off, and fixed prop losses included.

Quote
Granted, the "one starship round trip with HEEO refueling" mission concept does require a Starship capable of  Lunar landing and an aerobraked return to Earth.   Same requirements as Mars, IOTW (but at double the fuel cost).

So won't happen with crew until Starship is crew certified.

My bet is by 2035 this will be the standard way to get to the Moon.   All the other stuff (e.g. Artemis) is a place holder.

I agree, although I wouldn't be surprised to see the second refueling in cislunar.  That's sub-optimal from a propellant efficiency standpoint, but I still think that HEEO for crewed missions comes with a lot of baggage that may make the cislunar version worth the extra 2 lift tanker launches.

And "placeholder" may turn out to be a dismissive word.  If you don't get the launch/EDL crew certification for ten years, it's not a placeholder; it's the early version of your conops.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/07/2022 03:57 am
The HEEO refuel only costs one additional fully refueled standard Starship (said Starship reusable).   You use up so little fuel in the Oberth burn you have enough left over to complete the entire rest of the Moon mission (full round trip).

You use exactly the same amount of delta-v, irrespective of the energy of the HEEO.  If you're trying to get to LEO+3200 and you stop in an LEO+2500 HEEO, then that HEEO to LEO+3200 costs 700m/s.  It's simple addition, nothing more--as long as you always burn at perigee. 

You need to do C3 calculations, not deltaV calculations.

In the end, you end up with far less errors if you always using conservation of momentum and conservation of energy when performing conic section calculations.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/07/2022 04:01 am

You'll always be the most prop-efficient if you refuel in the lowest energy orbit that still leaves you with enough prop to meet the remaining mission delta-v requirements.  Things get weird if no HEEO exists that leaves you with enough prop to finish the mission--that's where something like laddering comes in--but you need a pretty extreme conops for that to happen.


When fuel costs $30-80/kg, "prop efficient" is a minor consideration.

Completing a mission with the least cost, including development costs of custom components, is the most important metric.  Reliability is another important metric, and doing almost the same thing hundreds of times is far more reliable than almost any custom component.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/07/2022 10:53 am
The HEEO refuel only costs one additional fully refueled standard Starship (said Starship reusable).   You use up so little fuel in the Oberth burn you have enough left over to complete the entire rest of the Moon mission (full round trip).

You use exactly the same amount of delta-v, irrespective of the energy of the HEEO.  If you're trying to get to LEO+3200 and you stop in an LEO+2500 HEEO, then that HEEO to LEO+3200 costs 700m/s.  It's simple addition, nothing more--as long as you always burn at perigee. 

You need to do C3 calculations, not deltaV calculations.

In the end, you end up with far less errors if you always using conservation of momentum and conservation of energy when performing conic section calculations.

Spacecraft have a delta-v budget, not a C3 budget.  And vis-viva is conservation of energy.

vperi² = μ(2/rperi - 1/a)

Let's set μ=1 and start in a circular orbit, i.e., r=a.  All maneuvers will be performed at rperiapse=1, which will therefore stay the same (i.e., set to 1 in our normalized units), as rapoapse increases.

If I prove that Δva=1→a=2 = Δva=1→a=1.5 + Δva=1.5→a=2, will you concede?

Here goes:

va=1² = (2 - 1) = 1, so vperi,a=1 = 1
va=2² = (2 - 1/2) = 1.5, so vperi,a=2 = 1.22
va=1.5² = (2 - 1/1.5) = 1.33, so vperi,a=1.5 = 1.15

Δva=1→a=2 = vperi,a=2 - vperi,a=1 = 1.22 - 1 = 0.22.
Δva=1→a=1.5 = vperi,a=1.5 - vperi,a=1 = 1.15 - 1 = 0.15
Δva=1.5→Δva=2 = vperi,a=2 = vperi,a=1.5 = 1.22 - 1.15 = 0.07

Δva=1→a=1.5 + Δva=1.5→Δva=2 = 0.15 + 0.07 = 0.22 = Δva=1→a=2
QED

Again, all maneuvers are performed at periapse, where you'd get the maximum Oberth effects.

Things are different if you transition from an elliptical orbit to a hyperbolic one, but we're not doing that.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 10/07/2022 05:17 pm

1) Option A and B missions have to be as simple as possible, and the LSS itself can't be refueled more than once per mission, nor can it be refueled with crew on board.

This is a silly constraint.  By the time a crew certified LSS leaves LEO there will be there will be an order of magnitude more experience refueling than launching SLS or reentering Orion or a great many other parts of the planned mission.  If that's not enough refueling experience, you can get a lot more fairly cheaply and quickly.  Since refueling is an essential part of any BLEO use of Starship you have to do that eventually, so the best plan is to expedite testing refueling and not place unneeded constraints on the eventual mission.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/07/2022 07:40 pm

1) Option A and B missions have to be as simple as possible, and the LSS itself can't be refueled more than once per mission, nor can it be refueled with crew on board.

This is a silly constraint.  By the time a crew certified LSS leaves LEO there will be there will be an order of magnitude more experience refueling than launching SLS or reentering Orion or a great many other parts of the planned mission.  If that's not enough refueling experience, you can get a lot more fairly cheaply and quickly.  Since refueling is an essential part of any BLEO use of Starship you have to do that eventually, so the best plan is to expedite testing refueling and not place unneeded constraints on the eventual mission.

It's a fair criticism, and one I expected.  Here's my reasoning:

1) Pez-dispenser Starships are probably unsuitable for refueling RPOD and transfer tests, so SpaceX is limited to however much hardware they decide to take out of Starlink service to perform the development.

2) Things will be more schedule-bound than you think.  They always are.  I'd also expect depot and refueling R&D to lag launch, EDL, and re-use R&D, simply because refueling testing is expensive without reusability being fairly far along.  And NASA is going to be fairly impatient with delays.

3) An LSS will be close to an order of magnitude more expensive than a vanilla Starship, simply because it has to be crew-certified.  It's an asset for which you want to aggressively minimize the risks you can.  Limiting LSS to a single refueling is a simple way to minimize risks.

4) NASA made a fairly big deal about the refueling complexity in the source selection statement, and the Blue Origin lawsuit picked up on that, forcing NASA to defend the conops as having almost all refueling off the critical path.  I'd expect that anything that reduced that complexity has very nice CYA properties for NASA.

Requiring the LSS to have multiple refuelings clearly isn't a complete deal-breaker.  But I suspect that minimization of conops complexity will be high on the list of things that NASA wants to see.

The fact that SLS and Orion will have less flight experience is irrelevant to the LSS.  Both SpaceX and NASA will do whatever they can to minimize risks, especially if it's cheap or easy to do so.  This is one that's both.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/07/2022 08:04 pm
I'm still curious what SpaceX might do with the used LSS's that will accumulate at Gateway. It might make sense to try to refuel them there (e.g. via a second depot at Gateway or just from regular tankers) and fly them down to the moon and back without crew just to demonstrate they could be reused.

That's assuming, of course, that they can be reused without needing maintenance. And it sure would help if someone built some paved landing pads on the moon!
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/07/2022 08:41 pm
I'm still curious what SpaceX might do with the used LSS's that will accumulate at Gateway. It might make sense to try to refuel them there (e.g. via a second depot at Gateway or just from regular tankers) and fly them down to the moon and back without crew just to demonstrate they could be reused.

That's assuming, of course, that they can be reused without needing maintenance. And it sure would help if someone built some paved landing pads on the moon!

There should only be two.  After that, you're on to Option B, and they get refueled until they reach end of life.

I don't think they need a depot.  In effect, they are depots, or at least spacecraft with good passive resistance to boil-off.  So you can refuel them long before the crew shows up, using plain ol' tankers that go out to NRHO via BLT and straight back to EDL.

I expect Option B (and Appendix P) HLSes to have fairly short lives, due to dust contamination.  It may be that SpaceX would do better using an EDL-capable LSS, if for no other reason that you can completely clean and refurbish it on the ground.  (You can also integrate new heavy cargoes into it, which an Option B LSS can't do.)  But this definitely requires refueling in HEEO or cislunar.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/07/2022 09:59 pm
A few days ago I mentioned the concept of tanker and cargo Starship mated for the burn from LEO towards the moon. Mainly I brought it up because I hadn't noticed it in this discussion. It didn't seem to be a popular idea both on the technical side and even whether it would be useful. I concede that the technical end may be more difficult than most would like. Possibly to the point of not being worthwhile.

As for the usefulness if it could be implemented, I didn't make the time to check my assumptions until this afternoon. Did a bit of BOTE to see where I would end up.    The way I see it, if the tanker would detach at ~2,500 m/s and fall back to reenter while the cargo Starship kept thrusting. The Starship would have a mass ratio of about 1.5 remaining to NHRO rendezvous. It seems to me that reaching that rendezvous with an extra 500 tons of propellant on board would be far from useless.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/07/2022 10:32 pm
A few days ago I mentioned the concept of tanker and cargo Starship mated for the burn from LEO towards the moon. Mainly I brought it up because I hadn't noticed it in this discussion. It didn't seem to be a popular idea both on the technical side and even whether it would be useful. I concede that the technical end may be more difficult than most would like. Possibly to the point of not being worthwhile.

As for the usefulness if it could be implemented, I didn't make the time to check my assumptions until this afternoon. Did a bit of BOTE to see where I would end up.    The way I see it, if the tanker would detach at ~2,500 m/s and fall back to reenter while the cargo Starship kept thrusting. The Starship would have a mass ratio of about 1.5 remaining to NHRO rendezvous. It seems to me that reaching that rendezvous with an extra 500 tons of propellant on board would be far from useless.

Why couple things mechanically (added mass, very bad) when you can couple them with software?  (mass free except for a proximity radar which Starship probably already has).

Two spacecraft can boost simultaneously and adjust to stay relatively close to each other.  It's not rocket science ;)

(Seriously the software wouldn't be very complex).

That, essentially, is what HEEO refuel option is.    Whether they boost at the same time or rendezvous later is an implementation detail (though at the same time is probably simpler)

A full load of fuel with an additional energy of 2.5km/sec beyond LEO orbital speeds is worth a lot, and effectively the same thing you are talking about.

As RadicalModerate points out, this won't be a working solution until LSS is crew-certified and there's been so many refuelings that it's as routine as refueling a B-52 bomber was in ~1960.


Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/07/2022 10:39 pm
The HEEO refuel only costs one additional fully refueled standard Starship (said Starship reusable).   You use up so little fuel in the Oberth burn you have enough left over to complete the entire rest of the Moon mission (full round trip).

You use exactly the same amount of delta-v, irrespective of the energy of the HEEO.  If you're trying to get to LEO+3200 and you stop in an LEO+2500 HEEO, then that HEEO to LEO+3200 costs 700m/s.  It's simple addition, nothing more--as long as you always burn at perigee. 

You need to do C3 calculations, not deltaV calculations.

In the end, you end up with far less errors if you always using conservation of momentum and conservation of energy when performing conic section calculations.

Spacecraft have a delta-v budget, not a C3 budget.  And vis-viva is conservation of energy.

vperi² = μ(2/rperi - 1/a)

Let's set μ=1 and start in a circular orbit, i.e., r=a.  All maneuvers will be performed at rperiapse=1, which will therefore stay the same (i.e., set to 1 in our normalized units), as rapoapse increases.

If I prove that Δva=1→a=2 = Δva=1→a=1.5 + Δva=1.5→a=2, will you concede?

Here goes:

va=1² = (2 - 1) = 1, so vperi,a=1 = 1
va=2² = (2 - 1/2) = 1.5, so vperi,a=2 = 1.22
va=1.5² = (2 - 1/1.5) = 1.33, so vperi,a=1.5 = 1.15

Δva=1→a=2 = vperi,a=2 - vperi,a=1 = 1.22 - 1 = 0.22.
Δva=1→a=1.5 = vperi,a=1.5 - vperi,a=1 = 1.15 - 1 = 0.15
Δva=1.5→Δva=2 = vperi,a=2 = vperi,a=1.5 = 1.22 - 1.15 = 0.07

Δva=1→a=1.5 + Δva=1.5→Δva=2 = 0.15 + 0.07 = 0.22 = Δva=1→a=2
QED

Again, all maneuvers are performed at periapse, where you'd get the maximum Oberth effects.

Things are different if you transition from an elliptical orbit to a hyperbolic one, but we're not doing that.

You are correct for elliptical orbits. I plead being sleepy and tired and having seen naive deltaV calcs fail for hyperbolic orbits in previous work.

The real gain comes from getting a full load of fuel that has +2.5km/sec velocity beyond LEO.   1/2mv^2 with v being higher and m being higher.  That's what enables full lunar round trips without remote depots, and exploration of the entire solar system and slightly beyond  (hyperbolic orbits) , making a lot of non-chemical methods technological dead ends.

But won't happen until LSS is EDL crew-certified and refuelings are as routine as they were on B-52s circa 1960.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 10/08/2022 12:07 am
The fact that SLS and Orion will have less flight experience is irrelevant to the LSS.  Both SpaceX and NASA will do whatever they can to minimize risks, especially if it's cheap or easy to do so.  This is one that's both.
Avoiding a second refueling avoids some risks but it adds others.

It requires a custom extended tank, which adds risk.

It also reduces margin.   The extra refueling lets you throw mass at other risks.  It also reduces schedule risk, and in particular a risk of SS underperforming; if it gets to orbit it can complete the mission with two refuelings.  There will be no delay trying to make weight by making the windows so thin they are at risk from a finger tap.

Also a second refueling never risks the crew.  This cannot be said of everything that might be done to avoid a second refueling.

Avoiding a refueling is not a no brainer, you have to evaluate all the risks being added against the risk of a refueling, unless you just assign each refueling infinite risk by fiat.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/08/2022 04:56 am
The fact that SLS and Orion will have less flight experience is irrelevant to the LSS.  Both SpaceX and NASA will do whatever they can to minimize risks, especially if it's cheap or easy to do so.  This is one that's both.
Avoiding a second refueling avoids some risks but it adds others.

It requires a custom extended tank, which adds risk.

I don't think it's a custom extended tank.  There are even more good reasons to go with a 1500t tank for lift tankers than there are for the LSS, not the least of which is that it doesn't require sending two tankers to NRHO for Option B.

Also, remember that "extended" really means "with ring segments, intertank bulkhead, and LCH4 dome rearranged."  That's not nothing, but as modifications go, it's not as bad as a lot of the mods that SpaceX has to make for LSS no matter what.

Quote
It also reduces margin.   The extra refueling lets you throw mass at other risks.  It also reduces schedule risk, and in particular a risk of SS underperforming; if it gets to orbit it can complete the mission with two refuelings.

Your argument about margin is well-taken--if you're talking about a 1200t LSS.  But even there, when I added in FPR and fixed prop losses, things worked out OK in a low HEEO.  I wouldn't want to use that method with a crew, but it's got pretty good margin for an uncrewed pre-positioning of the LSS in NRHO.

However, it's a lot more janky than just going with 1500t tanks.  Even after adding FPR, fixed losses, and boil-off allowances, the 1500t LSS gets back from the surface to NRHO with several tonnes of usable prop.

Quote
There will be no delay trying to make weight by making the windows so thin they are at risk from a finger tap.

Well, I've got other people on the thread complaining that my dry mass and crew module numbers are too conservative.  And again, 1500t of prop hides a multitude of sins.  There's a lot of mass margin here.

Quote
Avoiding a refueling is not a no brainer, you have to evaluate all the risks being added against the risk of a refueling, unless you just assign each refueling infinite risk by fiat.

That's fair, but assigning high risk to a procedure with which you have little operational experience is appropriate.  NB:  Even if there turns out to be a lot of experience with tankers fueling depots in VLEO, there will be considerably less experience refueling stuff in HEEO.  Could SpaceX load up on dummy refueling missions to work this out?  Sure.  But why bother if there's a better way to manage the risk?

The other factor here is risk tolerance.  In a perfect world, this wouldn't change over time.  However, HLS is a high enough profile project, both for NASA and SpaceX, that reducing tolerance to ensure that both Option A flights go well is a good political move--especially since NASA wrung their hands about operational complexity in the source selection statement, and Blue Origin used the conops to fling FUD around during the protest and appeal process.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/08/2022 08:20 am
A few days ago I mentioned the concept of tanker and cargo Starship mated for the burn from LEO towards the moon. Mainly I brought it up because I hadn't noticed it in this discussion. It didn't seem to be a popular idea both on the technical side and even whether it would be useful. I concede that the technical end may be more difficult than most would like. Possibly to the point of not being worthwhile.

As for the usefulness if it could be implemented, I didn't make the time to check my assumptions until this afternoon. Did a bit of BOTE to see where I would end up.    The way I see it, if the tanker would detach at ~2,500 m/s and fall back to reenter while the cargo Starship kept thrusting. The Starship would have a mass ratio of about 1.5 remaining to NHRO rendezvous. It seems to me that reaching that rendezvous with an extra 500 tons of propellant on board would be far from useless.

Why couple things mechanically (added mass, very bad) when you can couple them with software?  (mass free except for a proximity radar which Starship probably already has).

Two spacecraft can boost simultaneously and adjust to stay relatively close to each other.  It's not rocket science ;)

(Seriously the software wouldn't be very complex).

That, essentially, is what HEEO refuel option is.    Whether they boost at the same time or rendezvous later is an implementation detail (though at the same time is probably simpler)

A full load of fuel with an additional energy of 2.5km/sec beyond LEO orbital speeds is worth a lot, and effectively the same thing you are talking about.

As RadicalModerate points out, this won't be a working solution until LSS is crew-certified and there's been so many refuelings that it's as routine as refueling a B-52 bomber was in ~1960.

You don't get exactly the same results with software though. With the HEEO refueling, you still have docking, and propellant transfer before resuming boost. At a minimum, this gets another couple of trips through the Van Allens. It may be a better option to do it as you say, but it is not a 1 to 1 match. 
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/08/2022 02:51 pm
A few days ago I mentioned the concept of tanker and cargo Starship mated for the burn from LEO towards the moon. Mainly I brought it up because I hadn't noticed it in this discussion. It didn't seem to be a popular idea both on the technical side and even whether it would be useful. I concede that the technical end may be more difficult than most would like. Possibly to the point of not being worthwhile.

As for the usefulness if it could be implemented, I didn't make the time to check my assumptions until this afternoon. Did a bit of BOTE to see where I would end up.    The way I see it, if the tanker would detach at ~2,500 m/s and fall back to reenter while the cargo Starship kept thrusting. The Starship would have a mass ratio of about 1.5 remaining to NHRO rendezvous. It seems to me that reaching that rendezvous with an extra 500 tons of propellant on board would be far from useless.
Just to be sure I properly understand you: You're talking about refueling while accelerating, right? That is, you would:

a) launch a tanker and an LSS. (Each with 1500t fuel capacity.)
b) fully fuel both of them in LEO from a (rather large) depot. (Or two depots.)
c) couple the tanker to the LSS for refueling.
d) both of them fire together in formation, with the tanker continuously keeping the LSS topped up.
e) when the tanker is almost dry, it stops firing, disconnects, and reels in the fuel line.
f) at apogee, the tanker fires just a little bit to lower perigee to enable reentry.
g) the LSS goes on firing until it reaches TLI--arriving at the moon with about 500t extra fuel.

But I think we previously figured that a fully fueled tanker in LEO could reach the moon with about 500t of fuel still in the tank, so (if two refueling operations were allowed), you'd only be saving a little bit over having both ships accelerate to TLI independently and then have them rendezvous for refueling at any point thereafter. (And have the tanker do a free return or something like it.) Is that right?

Anyway, if NASA is worried about the risks of refueling in LEO, I think the risks of refueling while accelerating will really send them into orbit. (So to speak.) :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/08/2022 03:03 pm
I don't think they need a depot.  In effect, they are depots, or at least spacecraft with good passive resistance to boil-off.  So you can refuel them long before the crew shows up, using plain ol' tankers that go out to NRHO via BLT and straight back to EDL.
I know. I'm just liking the model that says Starships mate with depots, but not with each other, since it makes all the plumbing very unambiguous. It also lets the refueling be completely asynchronous. In this vision, a) a steady stream of tankers keeps the LEO depot full, b) occasionally a tanker fills up from the depot, flies to the moon, and pumps ~500t of prop into that depot (which really does achieve ZBO), so c) whenever an LSS needs to refuel, there's always fuel for it either in LEO or at Gateway.

I expect Option B (and Appendix P) HLSes to have fairly short lives, due to dust contamination.  It may be that SpaceX would do better using an EDL-capable LSS, if for no other reason that you can completely clean and refurbish it on the ground.  (You can also integrate new heavy cargoes into it, which an Option B LSS can't do.)  But this definitely requires refueling in HEEO or cislunar.
This leads me to wonder whether it's possible to do that at Gateway. Or, conversely, what is the minimum it would take to do the necessary maintenance and checkout at Gateway, rather than returning the LSS to Earth? (Assuming you can skip the static fire.) :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/08/2022 03:15 pm
Two concerns I thought of with Highly Elliptical Earth Orbit (HEEO) refueling:

1) These orbits have apogee's further out than the orbit of the moon. That implies they'll eventually have problems with lunar perturbations. Simplest fix is probably to make the period equal to half the lunar period and arrange it so the moon is at 90 degrees to the vector from the depot to the Earth. That makes for a two-week period, though.

2) How long would it take to fill one of these HEEO depots? If it takes 13 tanker trips, assuming you can just do one refueling per perigee, that mean it'd take 3 months to fill the thing. That makes for only four trips a year to the moon. By contrast, a depot in LEO can be refueled in just two weeks, assuming one tanker launch per day.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/08/2022 04:41 pm
Two concerns I thought of with Highly Elliptical Earth Orbit (HEEO) refueling:

1) These orbits have apogee's further out than the orbit of the moon. That implies they'll eventually have problems with lunar perturbations. Simplest fix is probably to make the period equal to half the lunar period and arrange it so the moon is at 90 degrees to the vector from the depot to the Earth. That makes for a two-week period, though.

2) How long would it take to fill one of these HEEO depots? If it takes 13 tanker trips, assuming you can just do one refueling per perigee, that mean it'd take 3 months to fill the thing. That makes for only four trips a year to the moon. By contrast, a depot in LEO can be refueled in just two weeks, assuming one tanker launch per day.

The optimal approach is for both Starships to use half their fuel so that one full Starship can do the perigee burn.. I calculate that gives an apogee of ~19,000 kilometers, so well inside Lunar orbit and well outside Luna's SOI.  It's almost but not quite  a standard GTO.

The 16 tanker trips is done in LEO, so it doesn't take 3 months.   It's only about 4 times the refueling trips for a min-deltaV Hohmann to Mars.   16 tanker trips (with 2 Starships being refueled) is 8 trips each so about 1.5 days at 4 hours per refuel (that's two orbits per refuel).

Once both Starships are full, they both simultaneously boost to the same HEEO orbit using half their fuel each.  Apogee 19,000 kilometers, Perigee 200-300km.

The final fuel transfer is done at just before Apogee of the HEEO(outside the Van Allen belts) and there's plenty of time to do that out there.  A small adjustment burn is done at apgoee by the full Starship to line up its perigee burn so it'll end up at the Moon (or Mars, or wherever), and the almost empty fuel carrier Starship does a slight apogee burn to initiate re-entry.

The reason to use it, though it may not be as optimal as a 1500t fuel dept in Lunar Orbit is that this maneuver gets you anywhere you want to go in the Solar system, at deltaVs so high the online porkchop calculators can't handle it.

There's only one extra trip through the Van Allens (coming back to perigee to make the Oberth burn).

No special measures for keeping the propellant from boiling are needed.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 10/08/2022 05:13 pm
Two concerns I thought of with Highly Elliptical Earth Orbit (HEEO) refueling:

1) These orbits have apogee's further out than the orbit of the moon. That implies they'll eventually have problems with lunar perturbations. Simplest fix is probably to make the period equal to half the lunar period and arrange it so the moon is at 90 degrees to the vector from the depot to the Earth. That makes for a two-week period, though.

2) How long would it take to fill one of these HEEO depots? If it takes 13 tanker trips, assuming you can just do one refueling per perigee, that mean it'd take 3 months to fill the thing. That makes for only four trips a year to the moon. By contrast, a depot in LEO can be refueled in just two weeks, assuming one tanker launch per day.
I think you are being over specific in what you mean by HEEO.

By some usage (https://en.wikipedia.org/wiki/Highly_elliptical_orbit) a Molniya or Geostationary transfer orbit would qualify as HEEO.  ΔV wise these are a little less than halfway to the Moon's surface, have periods close to 12 hours and are reasonably stable*.  A somewhat higher 400x70000km with a period of about a day would also be a HEEO.   Something like this would address the issues you raise

* You don't need absolute stability.  For example, a depot near L2 in a James Webb style halo orbit would require very little station keeping, as long as you are paying attention, there's probably no reason to do that, but you could.  Many orbits near the boundary of Earth's influence are like this.  They are unstable because small perturbations send them off into the void, but they are potentially usable because slightly less small control inputs bring them back.



Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/08/2022 07:09 pm
...this maneuver gets you anywhere you want to go in the Solar system, at deltaVs so high the online porkchop calculators can't handle it.

Getting to the outer solar system is straightforward.  Getting back isn't.

As with a notional 2-year Starship Jupiter mission (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2416683#msg2416683), delta-v far exceeds depot capacity, whether in LEO, HEEO, or elsewhere.  Depots must be supplemented, somehow.  In that mission:

- (Directional) laser propulsion was based in LEO and LJO.

- (Omnidirectional) methalox depots were based in LEO and on Callisto.

Together, and only together, they gave thrust vectors for mission success.

Some such network of pre-positioned directional and omnidirectional propulsion solutions would seem necessary for crewed missions to the outer solar system.

In practice, mission planners would optimize use of the network, in the spirit of Ishimatsu et al. 2016.  That is, applying an extended network flow model to the logistics system.

Q:  To support outer solar system Starship missions, where might you base the next directional and omnidirectional propulsion solutions?  What are some further good moves on the logistics chessboard?

Refs.

Ishimatsu, T., de Weck, O.L., Hoffman, J.A., Ohkami, Y. and Shishko, R., 2016. Generalized multicommodity network flow model for the earth–moon–mars logistics system. (https://dspace.mit.edu/bitstream/handle/1721.1/99360/JSR_Final_Manuscript_Ishimatsu.pdf?sequence=1&isAllowed=y) Journal of Spacecraft and Rockets, 53(1), pp.25-38.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/08/2022 08:54 pm
...this maneuver gets you anywhere you want to go in the Solar system, at deltaVs so high the online porkchop calculators can't handle it.

Getting to the outer solar system is straightforward.  Getting back isn't.

As with a notional 2-year Starship Jupiter mission (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2416683#msg2416683), delta-v far exceeds depot capacity, whether in LEO, HEEO, or elsewhere.  Depots must be supplemented, somehow.  In that mission:

- (Directional) laser propulsion was based in LEO and LJO.

- (Omnidirectional) methalox depots were based in LEO and on Callisto.

Together, and only together, they gave thrust vectors for mission success.

Some such network of pre-positioned directional and omnidirectional propulsion solutions would seem necessary for crewed missions to the outer solar system.

In practice, mission planners would optimize use of the network, in the spirit of Ishimatsu et al. 2016.  That is, applying an extended network flow model to the logistics system.

Q:  To support outer solar system Starship missions, where might you base the next directional and omnidirectional propulsion solutions?  What are some further good moves on the logistics chessboard?

Refs.

Ishimatsu, T., de Weck, O.L., Hoffman, J.A., Ohkami, Y. and Shishko, R., 2016. Generalized multicommodity network flow model for the earth–moon–mars logistics system. (https://dspace.mit.edu/bitstream/handle/1721.1/99360/JSR_Final_Manuscript_Ishimatsu.pdf?sequence=1&isAllowed=y) Journal of Spacecraft and Rockets, 53(1), pp.25-38.

By "anywhere in the solar system" I meant one way flybys.

The only two way trip that is worthwhile, AFAICT, is Mars and the Moon.  Mars is very well covered, and the Moon is the tough one, which is why all the debate.

Beyond Earth/Moon starting at LEO is IMHO a fairly obsolete technique when for less than 1/10 the cost of the SLS you can refuel a Starship in ~GTO and get anywhere else with Vinf of 15km/sec.  For example it drops Mars down to about 68 days, assuming Starship can actually slow down in the Mars atmosphere on the far end.  (in general if you want to *stop* somewhere the 15km/sec is problematic)

Trying to stay on topic, one can extend the fuel depot idea via "laddering" to e.g. an Oberth/gravity turn & burn at Jupiter, but the benefit is marginal.   That topic was explored on the extra-solar probe thread.

So are there missions beyond Luna/Mars that need a return and would be helped with a fuel depot?  Would it change the requirements that have been detailed in this thread?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/08/2022 10:02 pm
A few days ago I mentioned the concept of tanker and cargo Starship mated for the burn from LEO towards the moon. Mainly I brought it up because I hadn't noticed it in this discussion. It didn't seem to be a popular idea both on the technical side and even whether it would be useful. I concede that the technical end may be more difficult than most would like. Possibly to the point of not being worthwhile.

As for the usefulness if it could be implemented, I didn't make the time to check my assumptions until this afternoon. Did a bit of BOTE to see where I would end up.    The way I see it, if the tanker would detach at ~2,500 m/s and fall back to reenter while the cargo Starship kept thrusting. The Starship would have a mass ratio of about 1.5 remaining to NHRO rendezvous. It seems to me that reaching that rendezvous with an extra 500 tons of propellant on board would be far from useless.
Just to be sure I properly understand you: You're talking about refueling while accelerating, right? That is, you would:

a) launch a tanker and an LSS. (Each with 1500t fuel capacity.)
b) fully fuel both of them in LEO from a (rather large) depot. (Or two depots.)
c) couple the tanker to the LSS for refueling.
d) both of them fire together in formation, with the tanker continuously keeping the LSS topped up.
e) when the tanker is almost dry, it stops firing, disconnects, and reels in the fuel line.
f) at apogee, the tanker fires just a little bit to lower perigee to enable reentry.
g) the LSS goes on firing until it reaches TLI--arriving at the moon with about 500t extra fuel.

But I think we previously figured that a fully fueled tanker in LEO could reach the moon with about 500t of fuel still in the tank, so (if two refueling operations were allowed), you'd only be saving a little bit over having both ships accelerate to TLI independently and then have them rendezvous for refueling at any point thereafter. (And have the tanker do a free return or something like it.) Is that right?

Anyway, if NASA is worried about the risks of refueling in LEO, I think the risks of refueling while accelerating will really send them into orbit. (So to speak.) :-)

I'm say hard couple the ships together to control the variations. Formation flying while refueling would be even riskier. The tanker would crossfeed the cargo vehicle during the boost. If it works out that fuel cannot be transferred between ships at that rate, a lessor acceleration by both ships would reduce the required cross flow.

I think it works out that one ship would reach NHRO at 500 tons less dry mass. This would allow reaching it with 1,000 tons less the same dry mass. Approximately 380 tons net vs 880 tons net.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/08/2022 10:08 pm
Two concerns I thought of with Highly Elliptical Earth Orbit (HEEO) refueling:


There's only one extra trip through the Van Allens (coming back to perigee to make the Oberth burn).

No special measures for keeping the propellant from boiling are needed.



I get two extra trips through the Van Allens. Coming back to perigee and then again on the way out.   Not saying it's a showstopper. Depends on the risk, or perceived risk, of transiting them.

Patience please, still learning to use the quotes on this site.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/08/2022 10:33 pm
Quote
I get two extra trips through the Van Allens. Coming back to perigee and then again on the way out.   Not saying it's a showstopper. Depends on the risk, or perceived risk, of transiting them.

Patience please, still learning to use the quotes on this site.

A direct flight from  LEO w/ return is 2 passes through VA belts.

A ~GTO orbit takes one orbit to get back to perigee.  The boost out and the final return is one effective orbit.

At 2 passes through VA  belts per orbit, that's two extra passes for a round trip.

Thanks for the correction.

There's so much more mass and space on Starship to protect crew from VA radiation that I don't think the risk is that great.

You literally wrap the crew area in plastic.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/08/2022 10:39 pm


I'm say hard couple the ships together to control the variations. Formation flying while refueling would be even riskier.

Formation flying while refueling is what US Air Force has been doing since the late 1940s.

The trick is to not be under relative acceleration.  That's the default for spacecraft not under thrust, whether it's HEEO, LEO, or Luna orbit.

Other than radiation, there's no difference between a refueling procedure in orbit around Luna, HEEO, or LEO, although the refueling has to be not controlled by Earth because of lag problems in the first two cases. 

Since Dragon is docking autonomously I think they'll figure out how to do refueling > 0.01 light second away from Earth.

If you are worried about formation refueling, then LEO refueling won't work either, and the entire architecture of Starship is doomed.  Somehow I don't think so.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/08/2022 10:58 pm
...this maneuver gets you anywhere you want to go in the Solar system, at deltaVs so high the online porkchop calculators can't handle it.

Getting to the outer solar system is straightforward.  Getting back isn't.

As with a notional 2-year Starship Jupiter mission (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2416683#msg2416683), delta-v far exceeds depot capacity, whether in LEO, HEEO, or elsewhere.  Depots must be supplemented, somehow.  In that mission:

- (Directional) laser propulsion was based in LEO and LJO.

- (Omnidirectional) methalox depots were based in LEO and on Callisto.

Together, and only together, they gave thrust vectors for mission success.

Some such network of pre-positioned directional and omnidirectional propulsion solutions would seem necessary for crewed missions to the outer solar system.

In practice, mission planners would optimize use of the network, in the spirit of Ishimatsu et al. 2016.  That is, applying an extended network flow model to the logistics system.

Q:  To support outer solar system Starship missions, where might you base the next directional and omnidirectional propulsion solutions?  What are some further good moves on the logistics chessboard?

Refs.

Ishimatsu, T., de Weck, O.L., Hoffman, J.A., Ohkami, Y. and Shishko, R., 2016. Generalized multicommodity network flow model for the earth–moon–mars logistics system. (https://dspace.mit.edu/bitstream/handle/1721.1/99360/JSR_Final_Manuscript_Ishimatsu.pdf?sequence=1&isAllowed=y) Journal of Spacecraft and Rockets, 53(1), pp.25-38.

By "anywhere in the solar system" I meant one way flybys.

Yes, with many tankers and SH launches, understood.  Orbit and return maneuvers are vastly harder, but vital for most useful missions. 

Also, your "turn & burn at Jupiter" isn't a remotely realistic Starship maneuver.  Other propulsion is needed.

...for less than 1/10 the cost of the SLS you can refuel a Starship in ~GTO and get anywhere else with Vinf of 15km/sec.  For example it drops Mars down to about 68 days, assuming Starship can actually slow down in the Mars atmosphere...

It's a false assumption.  Your entry speed is north of 30 km/s.  With such immense KE, EDL is not possible.

-- Unless you add a propulsion node to the logistics system.  But where?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/09/2022 12:36 am
Yes, with many tankers and SH launches, understood.  Orbit and return maneuvers are vastly harder, but vital for most useful missions. 

Orbit is achieved by aerobraking anywhere there is atmosphere.   I also wasn't saying to use all 15km/sec of Vinf except for flybys.  The fuel saved for say a Hohmann transfer to Jupiter allows propulsive braking on the far end.  Obviates the need for a several hundred million in yet another "deep space" rocket engine, all of which, as near as I can tell, are custom to each deep space mission.

Quote
Also, your "turn & burn at Jupiter" isn't a remotely realistic Starship maneuver.  Other propulsion is needed.

In what way? Physics?  Logistics?  Cost?  ROI?  Something else?

Compared to billion dollar missions on SLS to get anywhere and billion dollar custom deep space probes, it can't be cost.

I'm not sure we need 150t of cargo leaving the solar system at 60-80km/sec, so ROI may be the best argument against it.

This extreme example of fuel laddering is to show what is possible.

Quote
It's a false assumption.  Your entry speed is north of 30 km/s.  With such immense KE, EDL is not possible.

I'm not sure I get how Vinf 15km/sec leaving Earth SOI gives you 30km/sec at Mars atmosphere.

Traveling to Mars subtracts 1km/sec relative. Dropping from Mars SOI to Mars atmosphere adds 1.5km/sec, so net 15.5km/sec at atmospheric entry.

Not sure if 15.5km/sec is possible in Mars atmosphere, but it's more possible than 30km/sec.

If say only 12.5km/sec is possible, then leave Earth SOI at 13.5km/sec, reserve 1.5km/sec for braking at Mars, and enter Mars atmosphere 12.5km/sec.  Only adds a few days to transit.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/09/2022 03:03 am
I'm not sure I get how Vinf 15km/sec leaving Earth SOI gives you 30km/sec at Mars atmosphere.

You're taking the relative velocity vector magnitude at Mars, for a trajectory like Jupiter DE-STAR.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/09/2022 06:55 am
Just to be sure I properly understand you: You're talking about refueling while accelerating, right? That is, you would:

a) launch a tanker and an LSS. (Each with 1500t fuel capacity.)
b) fully fuel both of them in LEO from a (rather large) depot. (Or two depots.)
c) couple the tanker to the LSS for refueling.
d) both of them fire together in formation, with the tanker continuously keeping the LSS topped up.
e) when the tanker is almost dry, it stops firing, disconnects, and reels in the fuel line.
f) at apogee, the tanker fires just a little bit to lower perigee to enable reentry.
g) the LSS goes on firing until it reaches TLI--arriving at the moon with about 500t extra fuel.

There's better version than this, and I think it's probably a winner:

a) Launch a tanker, but it's really not a tanker; it's just a pusher, i.e., it's a first stage that can dock to a "pushee" in VLEO.

b) Dock it to a depot, then leave it docked there until its target (its pushee) is on-orbit, then fill the pusher however full you need it (more in a moment).  This just minimizes boil-off.

c) Launch the EDL-capable lunar Starship, i.e., a vanilla Starship that can land on the Moon, hereinafter "EDLC-LSS".  (I know--catchy, eh?  It sounds like an X.25 standard that never quite caught on.)

d) The EDLC-LSS (with crew), gets filled completely full at a depot.

e) Immediately after the EDLC-LSS undocks, the pusher/tanker does an RPOD, leaving it docked nose-to-tail with the EDLC-LSS.  NOTE!!!!  No prop transfer occurs!!!  It's only a pusher!!!!

f) The pusher adds just enough delta-v to get the EDLC-LSS to go HEEO-LS-EDL on its 1500t of prop.

g) The pusher detaches and the EDLC-LSS immediately burns to TLI.  Note that even though the push technically put it in an HEEO, it's still basically at perigee, just going faster than it was, so there will only be one transit of the VA Belts.

h) EDLC-LSS immediately transits to the Moon by finishing the TLI, does LOI, DOI, PDI, landing, hangs out while astronauts cavort on the surface, then ascent, TEI, and finally EDL, delivering the crew to terra firma.

i) The pusher, separate from the EDLC-LSS, rises to the HEEO apogee, does a minor burn to initiate reentry, and does EDL (long before the EDLC-LSS does).

Below are two conops.  The first one is the HEEO-based refueling from a tanker.  (Yeah, it could be a depot.  Yeah, maybe only depots have refueling equipment.  Yeah, then the depot would have to get back to LEO propulsively.  I don't care.  The result will be pretty much the same.)  The second one is the conops I outlined above.  Several takeaways:

1) In terms of propellant usage, the two conops are, for all intents and purposes, identical.¹

2) If you think about it, this isn't surprising.  You have roughly the same aggregate wet mass and the same delta-v.  The fact that the aggregate "rocket" is flying in two pieces in one case and one two-stage piece in another doesn't really change the mass ratio.

3) The big, big, big deal here is that there's only one refueling for the EDLC-LSS, even though there's a second RPOD as the pusher attaches.  But the crew will spend no time at apogee in the middle of the VA Belts, because the TLI burn, even though it's in two stages, it more-or-less continuous.

4) Obvious problems/risks:
a) If the pusher doesn't detach cleanly, you have yourself a problem.  But this is really well-understood tech.
b) Obviously, a nose-to-tail docking that doesn't mess up the aero and thermal characteristics of the pusher is non-trivial.  But this is a clear winner from a risk-reduction standpoint.
c) If the EDLC-LSS doesn't fire after separate, it'll go to apogee, which makes crew absorb a big bag o' protons, but they can burn at the next perigee.

5) Yeah, you could build yourself a li'l tiny depot that did the pusher's job more efficiently.

6) I'm pretty sure this will work with 1200t EDLC-LSS and tankers, but I haven't cranked it through.

7) I tried this with an LEO-push-HEEO-LS-LEOpropulsive conops (which would eliminate Orion and let D2 bring the crew to/from LEO), and it doesn't work.  You'd still need another refueling in cislunar somewhere.


____________
¹The "refuel in HEEO" conops need contingency prop for loiter, in case the EDLC-LSS gets delayed.  With the pusher version, it can sit in VLEO, hooked up to a depot, where boiloff is minimized until just before RPOD.  This is basically a nit.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/09/2022 09:06 am


I'm say hard couple the ships together to control the variations. Formation flying while refueling would be even riskier.

Formation flying while refueling is what US Air Force has been doing since the late 1940s.

The trick is to not be under relative acceleration.  That's the default for spacecraft not under thrust, whether it's HEEO, LEO, or Luna orbit.

Other than radiation, there's no difference between a refueling procedure in orbit around Luna, HEEO, or LEO, although the refueling has to be not controlled by Earth because of lag problems in the first two cases. 

Since Dragon is docking autonomously I think they'll figure out how to do refueling > 0.01 light second away from Earth.

If you are worried about formation refueling, then LEO refueling won't work either, and the entire architecture of Starship is doomed.  Somehow I don't think so.


Given that, it seems like the original suggestion about transferring the propellant during the boost would work out.  Get the full value of the extra propellant burned during the impulse burn from LEO.

I don't think you were suggesting complete transfer during max boost. Transferring during acceleration does address settling problems as opposed to microgravity transfer. One question is how low can the acceleration be without gravity losses eating too much into the potential gains?  RL10 class with deep throttling?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: volker2020 on 10/09/2022 12:37 pm


I'm say hard couple the ships together to control the variations. Formation flying while refueling would be even riskier.

Formation flying while refueling is what US Air Force has been doing since the late 1940s.

The trick is to not be under relative acceleration.  That's the default for spacecraft not under thrust, whether it's HEEO, LEO, or Luna orbit.

Other than radiation, there's no difference between a refueling procedure in orbit around Luna, HEEO, or LEO, although the refueling has to be not controlled by Earth because of lag problems in the first two cases. 

Since Dragon is docking autonomously I think they'll figure out how to do refueling > 0.01 light second away from Earth.

If you are worried about formation refueling, then LEO refueling won't work either, and the entire architecture of Starship is doomed.  Somehow I don't think so.


Given that, it seems like the original suggestion about transferring the propellant during the boost would work out.  Get the full value of the extra propellant burned during the impulse burn from LEO.

I don't think you were suggesting complete transfer during max boost. Transferring during acceleration does address settling problems as opposed to microgravity transfer. One question is how low can the acceleration be without gravity losses eating too much into the potential gains?  RL10 class with deep throttling?

I see some problems with the equation of air formation flying with space operation. In the air, the fuel connection between the planes is flexible, so there is no spinning the other vehicle, when trust gets out of sync, and the air around does stabilize the flight.

All that would require that the trust control on both vehicles must be synchronized, so that you get into a feed back loop. Clearly doable, but a lot more complex than air tanking.

The most obvious solution would be a hard dock on the central axis. 
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/09/2022 03:48 pm
...No prop transfer occurs!!!  It's only a pusher!!!!

f) The pusher adds just enough delta-v to get the EDLC-LSS to go HEEO-LS-EDL on its 1500t of prop.

g) The pusher detaches and the EDLC-LSS immediately burns to TLI...

Alternately, a cis-lunar "pusher" or "tug", could run on ASCENT (https://afresearchlab.com/technology/aerospace/successstories/advanced-spacecraft-energetic-non-toxic-ascent-propellant/) propellant:  [NH3OH] [NO3]

Notably, ASCENT can be manufactured onboard the LEO depot; 96% of its mass can be trawled (https://forum.nasaspaceflight.com/index.php?topic=44542.msg1818140#msg1818140) from the thermosphere, which is almost entirely oxygen and nitrogen.  Lower a trawler's baseline LOX cooler/compressor operating temperature by 13°, to liquefy and store both gases.  Only 4% of ASCENT mass would be launched cargo:  LH2. 

ASCENT is non-toxic and storable in room-temperature inflatables.  Also, it's bimodal.  You can use it as a chemical monopropellant (Isp 250 s) for high-thrust Starship tug maneuvers, or as an electrospray ion engine propellant (Isp 1500 s) on other spacecraft / payloads, for long-duration, low-thrust missions.  Colón and Lightsey 2021.

Manufacture of ionic liquid from cation and anion can be straightforward (video).  How might you produce ASCENT's constituent NH3OH and NO3 from trawled LOX and LN2 plus cargo LH2?

Refs.

Colón, B.J. and Lightsey, E.G., 2021.  Spectre – Design of a Bimodal Propulsion System. (https://www.ssdl.gatech.edu/sites/default/files/ssdl-files/papers/mastersProjects/ColonB_AE8900.pdf)
 

https://www.youtube.com/watch?v=avP7EHsZFX0
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/09/2022 05:50 pm
The fact that SLS and Orion will have less flight experience is irrelevant to the LSS.  Both SpaceX and NASA will do whatever they can to minimize risks, especially if it's cheap or easy to do so.  This is one that's both.
Avoiding a second refueling avoids some risks but it adds others.

It requires a custom extended tank, which adds risk.

I don't think it's a custom extended tank.  There are even more good reasons to go with a 1500t tank for lift tankers than there are for the LSS, not the least of which is that it doesn't require sending two tankers to NRHO for Option B.

Also, remember that "extended" really means "with ring segments, intertank bulkhead, and LCH4 dome rearranged."  That's not nothing, but as modifications go, it's not as bad as a lot of the mods that SpaceX has to make for LSS no matter what.

Quote
It also reduces margin.   The extra refueling lets you throw mass at other risks.  It also reduces schedule risk, and in particular a risk of SS underperforming; if it gets to orbit it can complete the mission with two refuelings.

Your argument about margin is well-taken--if you're talking about a 1200t LSS.  But even there, when I added in FPR and fixed prop losses, things worked out OK in a low HEEO.  I wouldn't want to use that method with a crew, but it's got pretty good margin for an uncrewed pre-positioning of the LSS in NRHO.

However, it's a lot more janky than just going with 1500t tanks.  Even after adding FPR, fixed losses, and boil-off allowances, the 1500t LSS gets back from the surface to NRHO with several tonnes of usable prop.

Quote
There will be no delay trying to make weight by making the windows so thin they are at risk from a finger tap.

Well, I've got other people on the thread complaining that my dry mass and crew module numbers are too conservative.  And again, 1500t of prop hides a multitude of sins.  There's a lot of mass margin here.

Quote
Avoiding a refueling is not a no brainer, you have to evaluate all the risks being added against the risk of a refueling, unless you just assign each refueling infinite risk by fiat.

That's fair, but assigning high risk to a procedure with which you have little operational experience is appropriate.  NB:  Even if there turns out to be a lot of experience with tankers fueling depots in VLEO, there will be considerably less experience refueling stuff in HEEO.  Could SpaceX load up on dummy refueling missions to work this out?  Sure.  But why bother if there's a better way to manage the risk?

The other factor here is risk tolerance.  In a perfect world, this wouldn't change over time.  However, HLS is a high enough profile project, both for NASA and SpaceX, that reducing tolerance to ensure that both Option A flights go well is a good political move--especially since NASA wrung their hands about operational complexity in the source selection statement, and Blue Origin used the conops to fling FUD around during the protest and appeal process.
Why would HEEO refueling add novelty? What is added that is not demonstrated at LEO? The utility of the operation is another question and I have no opinion. Just trying to follow the arguments.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 10/09/2022 05:57 pm
This thread has drifted far off the topic of how to do refueling. It's now almost exclusively about all the neat missions that are enabled by refueling. Maybe we need two threads?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/09/2022 06:21 pm
NB:  Even if there turns out to be a lot of experience with tankers fueling depots in VLEO, there will be considerably less experience refueling stuff in HEEO.
Why would HEEO refueling add novelty? What is added that is not demonstrated at LEO? The utility of the operation is another question and I have no opinion. Just trying to follow the arguments.

HEEO rendezvous procedures are going to be different.  Detecting and quantifying insertion errors will be new.  Phasing takes longer.  Thermal characteristics are different.  Avionics failure rates based on VA Belt exposure will be less well-quantified.  Insertion errors cascade forward into departure window uncertainty, which leads to bigger TLI insertion errors.  Contingency planning for docking difficulty or engine burn aborts is more complicated.

None of this is insurmountable.  But if you're a mission assurance engineer plugging values into your failure tree, you'll be assigning larger uncertainties to all of this stuff, and as a result you'll get more uncertainty when the model spits out loss-of-mission probabilities than it would if you were just going round 'n' round in VLEO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/09/2022 06:21 pm
A few days ago I mentioned the concept of tanker and cargo Starship mated for the burn from LEO towards the moon. Mainly I brought it up because I hadn't noticed it in this discussion. It didn't seem to be a popular idea both on the technical side and even whether it would be useful. I concede that the technical end may be more difficult than most would like. Possibly to the point of not being worthwhile.

As for the usefulness if it could be implemented, I didn't make the time to check my assumptions until this afternoon. Did a bit of BOTE to see where I would end up.    The way I see it, if the tanker would detach at ~2,500 m/s and fall back to reenter while the cargo Starship kept thrusting. The Starship would have a mass ratio of about 1.5 remaining to NHRO rendezvous. It seems to me that reaching that rendezvous with an extra 500 tons of propellant on board would be far from useless.
Just to be sure I properly understand you: You're talking about refueling while accelerating, right? That is, you would:

a) launch a tanker and an LSS. (Each with 1500t fuel capacity.)
b) fully fuel both of them in LEO from a (rather large) depot. (Or two depots.)
c) couple the tanker to the LSS for refueling.
d) both of them fire together in formation, with the tanker continuously keeping the LSS topped up.
e) when the tanker is almost dry, it stops firing, disconnects, and reels in the fuel line.
f) at apogee, the tanker fires just a little bit to lower perigee to enable reentry.
g) the LSS goes on firing until it reaches TLI--arriving at the moon with about 500t extra fuel.

But I think we previously figured that a fully fueled tanker in LEO could reach the moon with about 500t of fuel still in the tank, so (if two refueling operations were allowed), you'd only be saving a little bit over having both ships accelerate to TLI independently and then have them rendezvous for refueling at any point thereafter. (And have the tanker do a free return or something like it.) Is that right?

Anyway, if NASA is worried about the risks of refueling in LEO, I think the risks of refueling while accelerating will really send them into orbit. (So to speak.) :-)
Well, it does do away with ullage thrust.  ;D


IMO, refueling will never be done unless the ships are physically hooked up. IOW, no hose hose line in a gas station. I'll give reasons if you want but ISTM to be a no brainer.


One problem with refueling while under significant thrust is the changing CoM. Between the necessarily constantly changing thrust vector and/or differential throttling, there will be small differences in thrust, maybe constantly changing, between the two rockets. Nothing big. At minimum, the digital engine controls have incremental steps over a range of 256, 512, 1024... steps. This absolutely can be compensated for and almost completely cancelled out but in the large scale non quantum world, nothing is exact.


The net result would be connection points between the ships experiencing small positive and negative shear stress. Adding robustness will make this a minor issue. The question is, is it worth the effort? It's all in the trades and ISTM the branching possibilities have spread so far it's too premature to do more than put this idea into the toolbox to be pulled out and dusted off if things move in this direction.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 10/09/2022 06:24 pm
On further reflection, I am even more convinced that all of the specialized refuelling hardware should be in the depot and the SSs should be at most minimally modified: the best part is no part.

All SS already have a mating connection for thrust, namely the big nine-meter ring that connects to the SH. The depot should implement the SH side of this connection beneath a disposable fairing at the nose. After the fairing is discarded the depot is a cylinder whose top surface looks like the top surface of an SH. The depot will dock nose-to-tail with the SS. The fuel transfer will be via an extendable QD mechanism that is stored beneath the mating ring and is also covered by the disposable fairing. It extends out, up, and around to mate with the SS QD connection. Fuel transfer is done while under a small amount of thrust provided by the depot. Exactly how docking and locking will be accomplished will require actual engineering instead of hand-waving and will depend on the current design of the existing interface. Worst case: use a nine-meter version of the "soft docking" hardware from the IDSS that provides six degrees of freedom through a limited range of motion to complete the soft docking. I strongly suspect this is overkill and real engineers will come up with a simpler system.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/09/2022 07:04 pm
I don't think they need a depot.  In effect, they are depots, or at least spacecraft with good passive resistance to boil-off.  So you can refuel them long before the crew shows up, using plain ol' tankers that go out to NRHO via BLT and straight back to EDL.
I know. I'm just liking the model that says Starships mate with depots, but not with each other, since it makes all the plumbing very unambiguous. It also lets the refueling be completely asynchronous. In this vision, a) a steady stream of tankers keeps the LEO depot full, b) occasionally a tanker fills up from the depot, flies to the moon, and pumps ~500t of prop into that depot (which really does achieve ZBO), so c) whenever an LSS needs to refuel, there's always fuel for it either in LEO or at Gateway.

I expect Option B (and Appendix P) HLSes to have fairly short lives, due to dust contamination.  It may be that SpaceX would do better using an EDL-capable LSS, if for no other reason that you can completely clean and refurbish it on the ground.  (You can also integrate new heavy cargoes into it, which an Option B LSS can't do.)  But this definitely requires refueling in HEEO or cislunar.
This leads me to wonder whether it's possible to do that at Gateway. Or, conversely, what is the minimum it would take to do the necessary maintenance and checkout at Gateway, rather than returning the LSS to Earth? (Assuming you can skip the static fire.) :-)
It would be great to be able to do repair at gateway and as much as I'd like it, it doesn't seem practical.


Drawing a not exact parallel with aircraft and ground vehicles, servicing consists of changing or checking levels of fluids, changing out filters, tensioning belts, and checking critical play. On gas engines it would include plugs and ignition wires. Except maybe for filters and checking for play, none of this would apply. Anything beyond this, including fixing excess play, would be repair.


Can you imagine changing out your cars water pump in a space suit in zero g? They do EVA stuff like this on the ISS only after modeling and scripting every move on the ground and still run into problems.


One optimization we've never seen on a rocket is on orbit maintenance and repair. The day is coming but probably not for another generation or two of rockets. I'd guess that there will be minor tweaks in this direction but the big change will be when they move to on orbit assembly.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/09/2022 07:22 pm
On further reflection, I am even more convinced that all of the specialized refuelling hardware should be in the depot and the SSs should be at most minimally modified: the best part is no part.

All SS already have a mating connection for thrust, namely the big nine-meter ring that connects to the SH. The depot should implement the SH side of this connection beneath a disposable fairing at the nose. After the fairing is discarded the depot is a cylinder whose top surface looks like the top surface of an SH. The depot will dock nose-to-tail with the SS. The fuel transfer will be via an extendable QD mechanism that is stored beneath the mating ring and is also covered by the disposable fairing. It extends out, up, and around to mate with the SS QD connection. Fuel transfer is done while under a small amount of thrust provided by the depot. Exactly how docking and locking will be accomplished will require actual engineering instead of hand-waving and will depend on the current design of the existing interface. Worst case: use a nine-meter version of the "soft docking" hardware from the IDSS that provides six degrees of freedom through a limited range of motion to complete the soft docking. I strongly suspect this is overkill and real engineers will come up with a simpler system.
Doesn't this result in a lot more plumbing, though? (At least, on the depot.) I've been assuming that the depot is little more than a Starship with extended tanks, meaning it fills/drains through a QD port at the bottom. That port would get replaced with a gender-swapped hose of some kind (I'm fuzzy on exactly how this would work, but something a lot like the hoses in the existing ground support equipment that fills them up on the pad) but the interior of the depot would be almost identical to that of a Starship. That means they have to be side-by-side to refuel OR the hose needs to be really long, which (I'm told) introduces a lot of problems.

I do kind of like the idea of the depot securely attaching to the vehicles it refuels. I'm just wondering how the plumbing would work.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/09/2022 07:35 pm
On further reflection, I am even more convinced that all of the specialized refuelling hardware should be in the depot and the SSs should be at most minimally modified: the best part is no part.

All SS already have a mating connection for thrust, namely the big nine-meter ring that connects to the SH. The depot should implement the SH side of this connection beneath a disposable fairing at the nose. After the fairing is discarded the depot is a cylinder whose top surface looks like the top surface of an SH. The depot will dock nose-to-tail with the SS. The fuel transfer will be via an extendable QD mechanism that is stored beneath the mating ring and is also covered by the disposable fairing. It extends out, up, and around to mate with the SS QD connection. Fuel transfer is done while under a small amount of thrust provided by the depot. Exactly how docking and locking will be accomplished will require actual engineering instead of hand-waving and will depend on the current design of the existing interface. Worst case: use a nine-meter version of the "soft docking" hardware from the IDSS that provides six degrees of freedom through a limited range of motion to complete the soft docking. I strongly suspect this is overkill and real engineers will come up with a simpler system.

Yes, we discussed this in the original pusher/tanker stuff.  If you do this, you lose the ability to go straight back to EDL, but 800m/s or so for a near-empty depot isn't exactly a massive propellant requirement.  If you go with 1200t tanks instead of 1500t, it's a bit more expensive (about 50t of prop), but still reasonable.

I don't think there's a prayer of scaling up an IDSS-like soft-capture ring to 9m.  You'd have soft-capture petals covering up the engines on the tail.  The alignment, translational, and rotational velocity errors would also have to be much, much smaller than those supported by the existing IDSS.  And getting all the hard-capture latches to engage would be harder, too.

I thought that three IDSS-like soft capture rings, spaced 120º apart around the tail, might work pretty well.  The sensors and software to get them to work together to damp out errors is a lot more complicated, but... hey--software. 

You could also go with grapples, like we discussed in the side-to-side docking.  But then you really do need a robust hard capture in either case.  Quick back-of-napkin:
Fully fueled (1500t) EDLC-LSS (EDL-capable LSS) with 25t payload:  1660t. 
Pusher mass near burnout, with return prop: 190t. 
Total wet mass of docked system: 1850t. 
Thrust of 3 RVacs throttled down to 70%: 4620kN. 
Burnout acceleration: 2.50m/s². 
Load on the docking system at burnout: 1660t * 2.50m/s² = 4150kN (423tf). 

That's a lot of force on a docking system.  So any hard-capture probably has to be able to align the ships to use the launch load points.

In the conops I did just up-thread, I assumed that there was no prop transfer at all, and that both the EDLC-LSS and its pusher had refueled at one or more depots before RPOD.  There is of course nothing¹ that prevents a depot from fulfilling the role of pusher, as long as you're willing to forgo reusing some or all of the QD hardware.  In that case, the EDLC-LSS and the depot would do the nose-to-tail RPOD, the depot would transfer prop to the EDLC-LSS until it was full, and then the depot would use the remainder of its prop to provide the extra delta-v to allow the EDLC-LSS to complete the whole mission without cislunar refueling.  The depot would then have to make its own way back into VLEO, which is a little more prop-intensive, but not much.

In either case, you're looking at a significant mod to the target Starship's configuration to support pushing.  Irrespective of the style of docking system you use, it's a whole bunch of new stuff in the engine compartment, with some interesting plumbing if you want pusher/depot functionality in one vehicle.

But a pusher/depot only adds a single RPOD to an EDLC-LSS crew's mission, and it gives them minimal exposure to VA belt radiation, and almost all of your contingency planning occurs in a stable, circular VLEO.  There are some kinda gnarly post-push contingencies (failure to undock, timely engine start failure on the EDLC-LSS), but I'd think that they'd have pretty low probabilities.

___________
¹Well, maybe one thing:  The depot would have to be able to hold a bit more than 1900t of prop to do a mission with 25t of cargo in addition to the crew module.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/09/2022 07:37 pm
It would be great to be able to do repair at gateway and as much as I'd like it, it doesn't seem practical.
I know, but it wasn't that long ago that we all thought having an empty first-stage return to the launch pad wasn't practical either. :-) You're probably right, of course, but I'm just wondering how much thought anyone has given to it lately.

Can you imagine changing out your cars water pump in a space suit in zero g? They do EVA stuff like this on the ISS only after modeling and scripting every move on the ground and still run into problems.
Sure, but I don't think anyone has given thought to how you'd set up an orbital repair station if you thought you were going to have to do the same set of things over and over. E.g. bespoke hardware to let you swap out engines. Or maybe a giant balloon big enough to hold a whole Starship so people could work on it in shirtsleeves (probably wearing oxygen masks). Or (more likely) something I haven't even thought about.

One optimization we've never seen on a rocket is on orbit maintenance and repair. The day is coming but probably not for another generation or two of rockets. I'd guess that there will be minor tweaks in this direction but the big change will be when they move to on orbit assembly.
Again, you're likely right, although I wonder if it wouldn't be easier to attempt repair before you try to do complete assembly.

(The connection this has with the refueling thread is that I don't think you can usefully refuel anything that can't EDL. Not past the initial mission, anyway.)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/09/2022 07:42 pm
I do kind of like the idea of the depot securely attaching to the vehicles it refuels. I'm just wondering how the plumbing would work.

The depot always has to securely attach to the vehicles it refuels.  The real question is whether it has to transmit main engine thrust loads to them.  If it's a pusher, it does.

Just to reiterate:  There is almost no difference, from a propellant efficiency standpoint, between a depot that refuels something in HEEO and a pusher that provides the delta-v to get to HEEO before detaching, allowing the target Starship to go straight to TLI.  The difference is in the operational details, especially for crewed flights.

The pusher is a more complex piece of hardware, and a pusher/depot is even more complex, with more complex plumbing.  But the conops is quite a bit simpler.

So many trades, so little time...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/09/2022 07:48 pm
Actual cryogenic depot tech to be aware of:  Eta Space LOXSAT 1 (https://www.nasa.gov/sites/default/files/atoms/files/kortes_perseverancecfm_tagged.pdf), and its successor, Cryo-Dock (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404365#msg2404365).

What decisions can be inferred from the designs?

Quote
LOXSAT 1 will test a suite of CFM technologies previously developed and ground tested. CFM technologies include:

Active and passive thermal control
Cryogenic chill down and transfer
Pressure control
Ground densification
Fluid surface visualization tools
Autogenous vs. helium pressurization
Liquid acquisition devices (LAD)
Zero boil off (ZBO) with pump mixing
High capacity 90K cryocoolers
Ground to flight insulation
Low conductivity supports
Zero-g chill down and transfer
Cryogenic quick disconnects
Ground densification for thermal energy storage
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 10/09/2022 07:56 pm
On further reflection, I am even more convinced that all of the specialized refuelling hardware should be in the depot and the SSs should be at most minimally modified: the best part is no part.

All SS already have a mating connection for thrust, namely the big nine-meter ring that connects to the SH. The depot should implement the SH side of this connection beneath a disposable fairing at the nose. After the fairing is discarded the depot is a cylinder whose top surface looks like the top surface of an SH. The depot will dock nose-to-tail with the SS. The fuel transfer will be via an extendable QD mechanism that is stored beneath the mating ring and is also covered by the disposable fairing. It extends out, up, and around to mate with the SS QD connection. Fuel transfer is done while under a small amount of thrust provided by the depot. Exactly how docking and locking will be accomplished will require actual engineering instead of hand-waving and will depend on the current design of the existing interface. Worst case: use a nine-meter version of the "soft docking" hardware from the IDSS that provides six degrees of freedom through a limited range of motion to complete the soft docking. I strongly suspect this is overkill and real engineers will come up with a simpler system.
Doesn't this result in a lot more plumbing, though? (At least, on the depot.) I've been assuming that the depot is little more than a Starship with extended tanks, meaning it fills/drains through a QD port at the bottom. That port would get replaced with a gender-swapped hose of some kind (I'm fuzzy on exactly how this would work, but something a lot like the hoses in the existing ground support equipment that fills them up on the pad) but the interior of the depot would be almost identical to that of a Starship. That means they have to be side-by-side to refuel OR the hose needs to be really long, which (I'm told) introduces a lot of problems.

I do kind of like the idea of the depot securely attaching to the vehicles it refuels. I'm just wondering how the plumbing would work.
Yes, lots of complex plumbing, but It's all in the depot. That's the whole idea. Move the entire design problem into the depot so all the other SS variants are minimally affected. Although this configuration can probably be used as a pusher, that's a side effect, not a design goal. Yes, this depot is a whole lot more complicated than a simple tank, because it must implement the entire fuel transfer system with no change to the other SS. The fuel transfer QD is separate from the depot's standard QD, which is still there at the tail. I'm not a rocket engineer, so I don't know if you run the pipes up the outside of the depot or if you run them inside the tanks, but they extend from the "bottom" to the "top" to connect the tanks to the refuelling QD. Since the depot does not EDL, they can probably be on the outside.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/09/2022 08:11 pm


I'm say hard couple the ships together to control the variations. Formation flying while refueling would be even riskier.

Formation flying while refueling is what US Air Force has been doing since the late 1940s.

The trick is to not be under relative acceleration.  That's the default for spacecraft not under thrust, whether it's HEEO, LEO, or Luna orbit.

Other than radiation, there's no difference between a refueling procedure in orbit around Luna, HEEO, or LEO, although the refueling has to be not controlled by Earth because of lag problems in the first two cases. 

Since Dragon is docking autonomously I think they'll figure out how to do refueling > 0.01 light second away from Earth.

If you are worried about formation refueling, then LEO refueling won't work either, and the entire architecture of Starship is doomed.  Somehow I don't think so.
Sometimes atmosphere is as much a friend as an enemy. Air to air refueling is in a viscus environment that both damps and inputs small variations. Drogue systems have inherent stability and flexibility in six degrees. Boom systems have a third control system (boom operator). Impulse from fluid flow is a tiny force that is less than random atmospheric burbling. It's a very different environment. Not impossible, but very different.


The idea of rigidly bonding two aircraft together in flight has been demonstrated but not widely used because it's a ticklish PITA in atmosphere. Bonding two spacecraft together in vacuum is a routine operation. I just can't see snaking two propellant hoses, two ullage gas hoses, maybe a nitrogen hose and maybe a power line between ships except as the back side of an extendable QD plate on the depot. Maybe .5-1m.


Maybe they can do without additional bonding points and I fully expect this to be way they'll first try it. My gut says that between the changing CoM and the unpredictable impulse from fluid flowing into the receiving tanks, its dicy. A slow transfer rate will keep fluid flow more controlled but this might not be viable when moving from proof of concept to operations.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 10/09/2022 08:19 pm
On further reflection, I am even more convinced that all of the specialized refuelling hardware should be in the depot and the SSs should be at most minimally modified: the best part is no part.

All SS already have a mating connection for thrust, namely the big nine-meter ring that connects to the SH. The depot should implement the SH side of this connection beneath a disposable fairing at the nose. After the fairing is discarded the depot is a cylinder whose top surface looks like the top surface of an SH. The depot will dock nose-to-tail with the SS. The fuel transfer will be via an extendable QD mechanism that is stored beneath the mating ring and is also covered by the disposable fairing. It extends out, up, and around to mate with the SS QD connection. Fuel transfer is done while under a small amount of thrust provided by the depot. Exactly how docking and locking will be accomplished will require actual engineering instead of hand-waving and will depend on the current design of the existing interface. Worst case: use a nine-meter version of the "soft docking" hardware from the IDSS that provides six degrees of freedom through a limited range of motion to complete the soft docking. I strongly suspect this is overkill and real engineers will come up with a simpler system.

I don't think there's a prayer of scaling up an IDSS-like soft-capture ring to 9m.  You'd have soft-capture petals covering up the engines on the tail.  The alignment, translational, and rotational velocity errors would also have to be much, much smaller than those supported by the existing IDSS.  And getting all the hard-capture latches to engage would be harder, too.

That's a lot of force on a docking system.  So any hard-capture probably has to be able to align the ships to use the launch load points.

I should not have mentioned IDSS: it was intended as an analogy.  It's an example of what some CNC folks call a "hexapod" design. You use six actuators to get six degrees of freedom. There is no effective limit to the size and strength at which they can be implemented except for mass. This SCS does not use a upscaled set of IDSS petals. It uses the same hard aligment system used for the SH/SS mateing.

We know that you can mate an SS with an SH: we have seen it done by the chopsticks.

Like the IDSS, the hexapod soft capture system (SCS) only corrects small errors. The depot must maneuver very carefully to get into the capture window of the SCS. velocities are measured in millimeters per second. This is just like the tolerances for an IDSS docking or for the chopsticks mating the SS and SH.

Yes, the system is intended to use the launch load pins. It's basically doing the same thing that is done to mount the SS atop an SH at the pad.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 10/09/2022 08:24 pm
Maybe they can do without additional bonding points and I fully expect this to be way they'll first try it. My gut says that between the changing CoM and the unpredictable impulse from fluid flowing into the receiving tanks, its dicy. A slow transfer rate will keep fluid flow more controlled but this might not be viable when moving from proof of concept to operations.
If they're not rigidly bonded together, why does center of mass matter? Each one would be adjusting thrust so as to have the same acceleration (as near as possible) and to stay close enough not to put tension on the hoses, but that's not dependent on the center of mass of the combined system. Are you thinking of some other factor that would change the center of mass of each vehicle separately? Also, if we're talking about an ullage burn at 1 milli-g, it's not like the system needs microsecond response times.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/09/2022 10:14 pm
On further reflection, I am even more convinced that all of the specialized refuelling hardware should be in the depot and the SSs should be at most minimally modified: the best part is no part.

All SS already have a mating connection for thrust, namely the big nine-meter ring that connects to the SH. The depot should implement the SH side of this connection beneath a disposable fairing at the nose. After the fairing is discarded the depot is a cylinder whose top surface looks like the top surface of an SH. The depot will dock nose-to-tail with the SS. The fuel transfer will be via an extendable QD mechanism that is stored beneath the mating ring and is also covered by the disposable fairing. It extends out, up, and around to mate with the SS QD connection. Fuel transfer is done while under a small amount of thrust provided by the depot. Exactly how docking and locking will be accomplished will require actual engineering instead of hand-waving and will depend on the current design of the existing interface. Worst case: use a nine-meter version of the "soft docking" hardware from the IDSS that provides six degrees of freedom through a limited range of motion to complete the soft docking. I strongly suspect this is overkill and real engineers will come up with a simpler system.
LOL. Dan, what you describe is except for the orientation of one ship, the original SX plan. Butt to butt instead of a fore end adapter, but otherwise the same.


I think the last Elon comment on the matter said side by side. It's been a while and it's hit its 'use by' date, but it's all we have.


Here's what I expect is happening. A preliminary design study has been done and has not been looked at in months. They have more immediate things to deal with. Like us, they noodle these ideas among themselves but with tighter assumptions. Like us, they wish they had better numbers. When they have a solid (sub) orbital flight under their belt all that work will be gone over with better numbers, and maybe they stick with side to side or they pivot in another direction.


Till then, side by side is what we got. Oh, to be a fly in the SX wall.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/09/2022 10:35 pm
Yes, lots of complex plumbing, but It's all in the depot. That's the whole idea. Move the entire design problem into the depot so all the other SS variants are minimally affected. Although this configuration can probably be used as a pusher, that's a side effect, not a design goal. Yes, this depot is a whole lot more complicated than a simple tank, because it must implement the entire fuel transfer system with no change to the other SS.


The tradeoff for this is that you can't send an EDL-capable tanker to cislunar for refueling ops.

If you're going to be doing enough stuff in cislunar to support a depot, then that's fine; the tanker boosts up to NRHO or LLO or wherever, transfers the prop to the depot, and goes home.  So the extra cost is that you've got a depot in cislunar that's never coming home, and you have ops costs associated with it long-term.

That may be a perfectly reasonable trade--and it also gives you the operational flexibility to ship the prop to cislunar long before the crewed mission.

However, note that you don't need a depot for an Option B LSS refueling, because for all intents and purposes the LSS likely has all the same boiloff management tech on it in the first place.  And you don't need it if you've progressed to the point where you're landing crews with an EDLC-LSS, because it's either refueled in HEEO or it's pushed to HEEO energies before immediately burning the rest of the way to TLI.  It'll have enough prop on board to return to direct EDL without refueling in cislunar.

The cases where a cislunar depot is interesting are the medium-term ones where you have a (non-EDLC) LSS that's staged with crew from LEO, and then returns the crew after the mission to LEO, propulsively, where they transfer to a D2 for EDL.  All of those cases require cislunar refueling, and a depot is handy to get your ducks in a row before committing a crew to a mission.

Quote
The fuel transfer QD is separate from the depot's standard QD, which is still there at the tail. I'm not a rocket engineer, so I don't know if you run the pipes up the outside of the depot or if you run them inside the tanks, but they extend from the "bottom" to the "top" to connect the tanks to the refuelling QD. Since the depot does not EDL, they can probably be on the outside.

So this is actually an important question that I'd like to pose for those of you who understand hypersonic flow:  How big a bump / chine / raceway / whateveryoucallit can you put on the dorsal surface of a Starship without having to substantially rework the max-q launch control system?

Quite a ways up-thread (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404058#msg2404058), I tried to draw a picture of a depot "kit" that was a payload, which could get deployed on the outside of any Starship, turning it into the active side of a depot.  It had in it:

1) Two active grapples, spaced as far apart as possible, to provide berthing.
2) The male-male adapter that allowed two female QD's to interconnect.
3) A cryocooler, which tapped off boiloff from the depot, via the QD, to re-liquify.
4) A big bag o' solar panels, which provided the power for the cryocooler and anything else that required extra power for long loiter.
5) Radiators for the cryocooler.

The kludgy part was that, because it lived inside the payload bay, it sorta had to crawl out on the dorsal side during deployment and make its way down to the tail to hook onto the QD.  If you can package all that stuff into a chine that launches on the outside of any Starship, then you have almost all of the advantages of the "kit" with a considerably less far-fetched deployment strategy.

If you can do this so that it runs all the way from the tail of the depot to the top of the cylindrical portion of the payload bay, then minor extensions of this idea could work for a pusher as well.

Extra credit would be if the kit could survive EDL, but fallback strategies where you'd still come out ahead would be:

a) Attach it only to depots, i.e., ships that can't do EDL.
b) Jettison the whole thing before EDL.

But you have to be able to cram all of that stuff into something that can survive all the flow regimes without having a boo-boo.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/09/2022 10:39 pm
On further reflection, I am even more convinced that all of the specialized refuelling hardware should be in the depot and the SSs should be at most minimally modified: the best part is no part.

All SS already have a mating connection for thrust, namely the big nine-meter ring that connects to the SH. The depot should implement the SH side of this connection beneath a disposable fairing at the nose. After the fairing is discarded the depot is a cylinder whose top surface looks like the top surface of an SH. The depot will dock nose-to-tail with the SS. The fuel transfer will be via an extendable QD mechanism that is stored beneath the mating ring and is also covered by the disposable fairing. It extends out, up, and around to mate with the SS QD connection. Fuel transfer is done while under a small amount of thrust provided by the depot. Exactly how docking and locking will be accomplished will require actual engineering instead of hand-waving and will depend on the current design of the existing interface. Worst case: use a nine-meter version of the "soft docking" hardware from the IDSS that provides six degrees of freedom through a limited range of motion to complete the soft docking. I strongly suspect this is overkill and real engineers will come up with a simpler system.
Doesn't this result in a lot more plumbing, though? (At least, on the depot.) I've been assuming that the depot is little more than a Starship with extended tanks, meaning it fills/drains through a QD port at the bottom. That port would get replaced with a gender-swapped hose of some kind (I'm fuzzy on exactly how this would work, but something a lot like the hoses in the existing ground support equipment that fills them up on the pad) but the interior of the depot would be almost identical to that of a Starship. That means they have to be side-by-side to refuel OR the hose needs to be really long, which (I'm told) introduces a lot of problems.

I do kind of like the idea of the depot securely attaching to the vehicles it refuels. I'm just wondering how the plumbing would work.
The approach that makes the most sense to me is to modify the QD on the depot to be an exact copy of the GSE QD plate on the tower. This means it can mate with any ship just as the towers GSE QD can. Unfortunately, this means it can't interface with the tower GSE QD plate for its own launch. The proposed solution that I prefer is to mount an adapter on the towers GSE QD plate that will allow interface and remove it after launch. Adapter used for depot launches only. The extra adapter mass stays on the ground.


Some think that it would work 'as is' and some think the depot QD plate should extend some short distance out, fixed or actively extending, to keep a bit more space between the two ships. From this came discussion of additional bracing.


Some disagree with this architecture entirely. The opinions and views expressed above are mine and do not reflect yada, yada.  8)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/09/2022 11:24 pm
...modify the QD on the depot to be an exact copy of the GSE QD plate on the tower. This means it can mate with any ship just as the towers GSE QD can.

...mount an adapter on the towers GSE QD plate that will allow interface and remove it after launch. Adapter used for depot launches only.

...some think the depot QD plate should extend some short distance out, fixed or actively extending, to keep a bit more space between the two ships.

Or separate and simplify a depot's two-fluid cryo connection, much as Eta Space has done with Cryo-Dock (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404365#msg2404365). 

They animate the depot atop an F9.  Auditioning?

A Starship depot adaptation would retract further, to pull hw out of slipstream exposure.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/09/2022 11:41 pm
Yes, lots of complex plumbing, but It's all in the depot. That's the whole idea. Move the entire design problem into the depot so all the other SS variants are minimally affected. Although this configuration can probably be used as a pusher, that's a side effect, not a design goal. Yes, this depot is a whole lot more complicated than a simple tank, because it must implement the entire fuel transfer system with no change to the other SS.


The tradeoff for this is that you can't send an EDL-capable tanker to cislunar for refueling ops.

If you're going to be doing enough stuff in cislunar to support a depot, then that's fine; the tanker boosts up to NRHO or LLO or wherever, transfers the prop to the depot, and goes home.  So the extra cost is that you've got a depot in cislunar that's never coming home, and you have ops costs associated with it long-term.

That may be a perfectly reasonable trade--and it also gives you the operational flexibility to ship the prop to cislunar long before the crewed mission.

However, note that you don't need a depot for an Option B LSS refueling, because for all intents and purposes the LSS likely has all the same boiloff management tech on it in the first place.  And you don't need it if you've progressed to the point where you're landing crews with an EDLC-LSS, because it's either refueled in HEEO or it's pushed to HEEO energies before immediately burning the rest of the way to TLI.  It'll have enough prop on board to return to direct EDL without refueling in cislunar.

The cases where a cislunar depot is interesting are the medium-term ones where you have a (non-EDLC) LSS that's staged with crew from LEO, and then returns the crew after the mission to LEO, propulsively, where they transfer to a D2 for EDL.  All of those cases require cislunar refueling, and a depot is handy to get your ducks in a row before committing a crew to a mission.

Quote
The fuel transfer QD is separate from the depot's standard QD, which is still there at the tail. I'm not a rocket engineer, so I don't know if you run the pipes up the outside of the depot or if you run them inside the tanks, but they extend from the "bottom" to the "top" to connect the tanks to the refuelling QD. Since the depot does not EDL, they can probably be on the outside.

So this is actually an important question that I'd like to pose for those of you who understand hypersonic flow:  How big a bump / chine / raceway / whateveryoucallit can you put on the dorsal surface of a Starship without having to substantially rework the max-q launch control system?

Quite a ways up-thread (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404058#msg2404058), I tried to draw a picture of a depot "kit" that was a payload, which could get deployed on the outside of any Starship, turning it into the active side of a depot.  It had in it:

1) Two active grapples, spaced as far apart as possible, to provide berthing.
2) The male-male adapter that allowed two female QD's to interconnect.
3) A cryocooler, which tapped off boiloff from the depot, via the QD, to re-liquify.
4) A big bag o' solar panels, which provided the power for the cryocooler and anything else that required extra power for long loiter.
5) Radiators for the cryocooler.

The kludgy part was that, because it lived inside the payload bay, it sorta had to crawl out on the dorsal side during deployment and make its way down to the tail to hook onto the QD.  If you can package all that stuff into a chine that launches on the outside of any Starship, then you have almost all of the advantages of the "kit" with a considerably less far-fetched deployment strategy.

If you can do this so that it runs all the way from the tail of the depot to the top of the cylindrical portion of the payload bay, then minor extensions of this idea could work for a pusher as well.

Extra credit would be if the kit could survive EDL, but fallback strategies where you'd still come out ahead would be:

a) Attach it only to depots, i.e., ships that can't do EDL.
b) Jettison the whole thing before EDL.

But you have to be able to cram all of that stuff into something that can survive all the flow regimes without having a boo-boo.
Any maxQ problems would depend on the size of the whatever. Any imbalance, up to some unknown point, could be overcome by thrust vectoring or differential thrust. There's a lot of propellant flexibility because in worst case the depot would reach orbit with no propellant of its own to kick off the campaign.


Why not build depots to stay in orbit? The only downside is a ship that can have no other use and may have a short operational life. The lifetime will only become more robust as the SS matures. After all, a major design goal is fast turnaround. With only one launch and atmospheric transit, combined with non coking engines, a one campaign lifetime at the beginning will eventually extend to more as the design matures.


They're planning on turning these things out like sausages. A few special builds that are force multipliers and never come back will have little impact on production. A self deploying universal conversion kit is more of a complication than a help. If a depot is needed cislunar, send a depot. Let it stay there. If it lives long enough it'll be there for future use. If not, well, it did its job. The next one will last longer.


There is the issue of need. Consensus is that space traffic will only get heavier. Having a depot on orbit, LEO, cislunar or elsewhere, incurs little expense. In exchange, it offers flexibility and opportunity. Infrastructure of any sort is a force multiplier.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 10/09/2022 11:46 pm
If you're going to be doing enough stuff in cislunar to support a depot, then that's fine; the tanker boosts up to NRHO or LLO or wherever, transfers the prop to the depot, and goes home.  So the extra cost is that you've got a depot in cislunar that's never coming home, and you have ops costs associated with it long-term.

That may be a perfectly reasonable trade--and it also gives you the operational flexibility to ship the prop to cislunar long before the crewed mission.

The cases where a cislunar depot is interesting are the medium-term ones where you have a (non-EDLC) LSS that's staged with crew from LEO, and then returns the crew after the mission to LEO, propulsively, where they transfer to a D2 for EDL.  All of those cases require cislunar refueling, and a depot is handy to get your ducks in a row before committing a crew to a mission.
Even though this depot is fairly complicated, it's still not a huge expense. Just build it and park it in cislunar space forever. Even if depots are not yet in series production to support Mars, building a second one alongside the Artemis one is not a large extra expense.
Quote
Quote
The fuel transfer QD is separate from the depot's standard QD, which is still there at the tail. I'm not a rocket engineer, so I don't know if you run the pipes up the outside of the depot or if you run them inside the tanks, but they extend from the "bottom" to the "top" to connect the tanks to the refuelling QD. Since the depot does not EDL, they can probably be on the outside.
So this is actually an important question that I'd like to pose for those of you who understand hypersonic flow:  How big a bump / chine / raceway / whateveryoucallit can you put on the dorsal surface of a Starship without having to substantially rework the max-q launch control system?
For a non-EDL depot, these whatists should pose a lot less of a challenge than the EDL control surfaces on an EDL SS. They do not need to be on the dorsal side. I think you have two of them, one on each side, and they are aerodynamically identical even if the actual plumbing is different.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/10/2022 12:26 am
...modify the QD on the depot to be an exact copy of the GSE QD plate on the tower. This means it can mate with any ship just as the towers GSE QD can.

...mount an adapter on the towers GSE QD plate that will allow interface and remove it after launch. Adapter used for depot launches only.

...some think the depot QD plate should extend some short distance out, fixed or actively extending, to keep a bit more space between the two ships.

Or separate and simplify a depot's two-fluid cryo connection, much as Eta Space has done with Cryo-Dock (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2404365#msg2404365). 

They animate the depot atop an F9.  Auditioning?

A Starship depot adaptation would retract further, to pull hw out of slipstream exposure.
One plan calls for interconnecting ullage spaces for pressure equalization while pumps handle the transfer and another calls for venting the receiving ships ullage and letting the other ships ullage pressure force the transfer. N2 may be used used for a while despite the goal of getting rid of it. And argument can be made for power interconnection. Some advocate for an 'in line' arrangement as you show. It can handle more connections if necessary. One school of thought is that whatever is used for transfer should interface with the connections that are already there.


The cryo-dock is targeted at transfers quite a bit smaller than what SX is looking at. I'm not sure what scaling issues might be driving the design considerations. Maybe none at all.


I'm picturing a maybe half meter deep doghouse on the depot, and maybe all ships, with a roll up door that drops into place to protect the QD plate after launch disconnect. The depot QD would extend maybe another half meter out of the doghouse when it gets to work.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/10/2022 02:13 am
N2 may be used used for a while despite the goal of getting rid of it.

SpaceX on-orbit cryo demo targets "autogenous pressurization and propellant transfer".

N2 isn't mentioned in the NASA doc (https://www.nasa.gov/sites/default/files/atoms/files/kortes_perseverancecfm_tagged.pdf).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/10/2022 04:06 pm
N2 may be used used for a while despite the goal of getting rid of it.

SpaceX on-orbit cryo demo targets "autogenous pressurization and propellant transfer".

N2 isn't mentioned in the NASA doc (https://www.nasa.gov/sites/default/files/atoms/files/kortes_perseverancecfm_tagged.pdf).
That looks like 12/21, 10 months ago. It states aspirations and most probably contract requirements. Some aspirations may not be contract requirements. N2 isn't mentioned either way. It has uses other than ullage so as I said, there is a 'use by' date on public SX aspirations. We have surmises but less than perfect knowledge. It'll be interesting no matter what they do.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/10/2022 09:50 pm
https://selenianboondocks.com/2008/01/an-insane-but-interesting-idea-fleet-launched-orbital-craft/

If  you think refueling in space while under acceleration is tough, try this one from Jon Goff.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/10/2022 09:58 pm
https://selenianboondocks.com/2008/01/an-insane-but-interesting-idea-fleet-launched-orbital-craft/

If  you think refueling in space while under acceleration is tough, try this one from Jon Goff.

That's not as insane as refueling under acceleration, and it's not nearly as insane as refueling under acceleration without rigid, load-bearing couplings.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/11/2022 08:41 am
https://selenianboondocks.com/2008/01/an-insane-but-interesting-idea-fleet-launched-orbital-craft/

If  you think refueling in space while under acceleration is tough, try this one from Jon Goff.

That's not as insane as refueling under acceleration, and it's not nearly as insane as refueling under acceleration without rigid, load-bearing couplings.

And yet if you think about it, the Shuttle orbiter was 'refueling' from the external tank on every launch.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/11/2022 06:00 pm
https://selenianboondocks.com/2008/01/an-insane-but-interesting-idea-fleet-launched-orbital-craft/ (https://selenianboondocks.com/2008/01/an-insane-but-interesting-idea-fleet-launched-orbital-craft/)

If  you think refueling in space while under acceleration is tough, try this one from Jon Goff.

That's not as insane as refueling under acceleration, and it's not nearly as insane as refueling under acceleration without rigid, load-bearing couplings.

And yet if you think about it, the Shuttle orbiter was 'refueling' from the external tank on every launch.
Shuttle was rigidly mounted to the ET. The connector and structural mating was done on the ground. For the tank, it was a one night stand. Spending a hundred tech hours (a total WAG) on shuttle side inspection and refurb would not even have been a rounding error on shuttle turnaround. Other than that...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/11/2022 07:12 pm
https://selenianboondocks.com/2008/01/an-insane-but-interesting-idea-fleet-launched-orbital-craft/

If  you think refueling in space while under acceleration is tough, try this one from Jon Goff.

That's not as insane as refueling under acceleration, and it's not nearly as insane as refueling under acceleration without rigid, load-bearing couplings.

And yet if you think about it, the Shuttle orbiter was 'refueling' from the external tank on every launch.

Yeah, and that was, if not insane, at least kinda dumb, as it turned out.  More importantly for this discussion:  It required a large team of humans in a shirtsleeve environment to align the bipods, connect them, tension them, and--last but hardly least--arm their pryrotechnic frangible bolts.  That's a silly--and likely impossible--design for in-space automated operations.

Let's make sure we're all talking about the same things here.  In terms of accelerating while refueling and docked / berthed / tethered / formation-flying / whatever, we have the following:

1) Two spacecraft boosting independently to the same HEEO so that RPOD (rendezvous, proximity ops, and docking) is relatively quick once in the HEEO, allowing refueling that should take less than an orbit if everything goes right.  This is what we've mostly talked about.  No coupled acceleration involved. 

Note that the most general case of this is what we've been calling "laddering", which involves n ships refueling m ships in orbit 1, then the m ships refueling p ships in orbit 2, etc. It's similar to the thing Jon Goff was describing.  It can make sense for achieving extremely high delta-v, but its practical utility for cislunar or even martian ops is effectively zero.

2) Coupled ullage accelerations, on the order of 1mm/s² - 1cm/s².  Most on-orbit refueling schemes require this to settle the prop, and if you're doing side-by-side refueling, the accelerations need to be coordinated on the two vehicles, or transmitted from one vehicle to another via something able to take loads of a few newtons.  I assert that this requires at least semi-rigid docking, although this is a debatable point.

3) Coordinated boost accelerations while connected, anywhere from 3-12m/s².  This is insane for a side-by-side configuration that's docked on orbit.  It's doubly insane because it's unnecessary.  Prop consumption for boosting two ships to an HEEO in a coupled system is the same as prop consumption for boosting the two ships separately.¹  If your objection here is, "But what about cross-feed?" the only thing cross-feed would get you is higher thrust, which will only save you single-digit m/s to make an orbital burn slightly more impulsive.

4) Pusher configurations, where the tanker/depot acts as a first stage for the target Starship, then separates, allowing the target Starship to continue the burn, with the whole staging process costing low-single-digit m/s in non-impulsive losses.  This is not a case of accelerating while refueling; it's just a staging operation.  The pushing depot/tanker may or may not have previously refueled the target Starship, but that's really a separate operation.

#1, using #2 to transfer prop, is a viable option, as is #4, which can either transfer prop using ullage acceleration from the pusher, or where the pusher and target can separately use #2 in VLEO from separate tankers/depots.

I propose we stop talking about #3.

___________
¹Proof:

A particular delta-v requires a fixed mass ratio MR. 
MR = exp( Δv / Isp / 9.8 ) = (inert + propInitial) / (inert + propFinal)
This is just the rocket equation.

We want to find the prop consumed in executing the delta-v maneuver.

propFinal = propInitial - propConsumed
MR = (inert + propInitial) / (inert + propInitial - propConsumed)
MR(inert + propInitial) - MR*propConsumed = inert + propInitial
MR*propConsumed = (inert + propInitial)(MR - 1)
propConsumed = (inert + propInitial)(1 - 1/MR)

Now suppose we have two spacecraft, with masses inert1+propInitial1 and inert2+propInitial2.  It should now be obvious that the prop consumption is the same whether they're boosted separately or together:

(inert1 + propInitial1)(1 - 1/MR) + (inert2 + propInitial2)(1 - 1/MR) =
(inert1 + propInitial1 + inert2 + propInitial2)(1 - 1/MR)

Remember, MR is a constant, dependent solely on the delta-v needed to reach the refueling orbit and the Isp.  Things would get weird if you had two different Isp's, but that's not what we're talking about here.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/12/2022 08:42 pm
You can argue, possibly successfully, that I've over-constrained the problem.  If you're willing to do two refuelings on the LSS, one in VLEO and one in HEEO, you can boost to pretty much any energy you want.  But NASA made a fairly big deal about only requiring one refueling against the LSS itself.  This actually came up in the BO lawsuit, where NASA responded to their complaint on this topic.  So going back on that opens up a can of worms that would be better left closed.

Frankly, this is why I like the 1500t LSS.  All of this complexity falls away, and all refueling is done in the lowest-possible energy orbit.

Ironically, with my harmonic refueling it also happens in the lowest "possible" orbit too. That's the driving principle behind the optimization, in fact!  :D

I believe that laddering makes sense when extreme C3 outweighs the need for mission simplicity.  However, for both lunar and Mars missions, that's simply not the case.  In the case of an Artemis mission, you'll want to limit the LSS to a single refueling, simply to reduce risk.  That requires limiting the maximum refueling energy to something that the LSS can reach with its launch prop.

I missed this particular bit of silliness the first time around.

If you are only 'permitted' one filling, then the math says that VLEO is the only sensible choice.

So essentially your argument is "if we assume no laddering, then there's no laddering." Imagine that...  ::)

In all my laddering calculations I assume the vehicle is fully fueled in VLEO first (ie the sensible approach), so apparently we were talking/calculating past each-other the entire time.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/12/2022 08:58 pm

If a tanker had 12-hour turnaround, and it delivered 1/5 of a load, it could fill an inflatable depot for ~ 300 crews each synod, by itself. 

Otherwise, a fleet of tankers would be needed.  E.g., given a 14-day window and preloading of depot tankers at synod start, at least 46 tankers would be needed. 

Scale up ~ 10x (https://aeon.co/essays/elon-musk-puts-his-case-for-a-multi-planet-civilisation) for settlement cargo flights.

The obvious idea is to use the Starships themselves as the "depot." This eliminates the entire depot fleet, which is a massive savings.

Instead you just transfer the passengers at the last minute. You can use 500 passenger P2P-derived vehicles to do that. This further increases the efficiency of launching within that 14-day window, because each launch serves five Starships instead of one.

Delete depots, get taxis.

No, you wouldn't want to haul all depot hw (ruggedized ZBO and propellant transfer hw) on every ship, to Mars and back

Maybe, maybe not. It depends on how much mass.

Many people seem to be [over-]engineering elaborate complex systems for both. This is a mistake IMO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/12/2022 09:24 pm
Since both mission architectures (boosting the depot vs. boosting the tanker) are readily available, it seems like we should at least compare and contrast the two options.

It's a good idea, but I fooled with it a bit and it doesn't really do anything meaningful.

Going to an EDL-capable tanker saves you some return delta-v, but it doesn't save you huge amounts of prop, because the tanker's wet mass is small after it's transferred the prop to the LSS.



Which is cheaper than developing a custom depot stuck in Lunar orbit.

Who said anything about a custom depot?  If you need prop in cislunar and your logistics work out, you send a regular lift tanker, refuel, and the tanker returns to EDL.


So which one is it, TheRadicalModerate?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/12/2022 09:36 pm

1) Option A and B missions have to be as simple as possible, and the LSS itself can't be refueled more than once per mission, nor can it be refueled with crew on board.

This is a silly constraint.  By the time a crew certified LSS leaves LEO there will be there will be an order of magnitude more experience refueling than launching SLS or reentering Orion or a great many other parts of the planned mission.  If that's not enough refueling experience, you can get a lot more fairly cheaply and quickly.  Since refueling is an essential part of any BLEO use of Starship you have to do that eventually, so the best plan is to expedite testing refueling and not place unneeded constraints on the eventual mission.

It's a fair criticism, and one I expected.  Here's my reasoning:

1) Pez-dispenser Starships are probably unsuitable for refueling RPOD and transfer tests, so SpaceX is limited to however much hardware they decide to take out of Starlink service to perform the development.

2) Things will be more schedule-bound than you think.  They always are.  I'd also expect depot and refueling R&D to lag launch, EDL, and re-use R&D, simply because refueling testing is expensive without reusability being fairly far along.  And NASA is going to be fairly impatient with delays.

Delays with Artemis??? Inconceivable!

I don't see anyone nipping at SpaceX's heels, so NASA can get as "impatient" as they want. There's no faster alternative.

3) An LSS will be close to an order of magnitude more expensive than a vanilla Starship, simply because it has to be crew-certified.  It's an asset for which you want to aggressively minimize the risks you can.  Limiting LSS to a single refueling is a simple way to minimize risks.


It's a "simple" to be sure. But what really matters is the cost/benefit tradeoff.

4) NASA made a fairly big deal about the refueling complexity in the source selection statement, and the Blue Origin lawsuit picked up on that, forcing NASA to defend the conops as having almost all refueling off the critical path.  I'd expect that anything that reduced that complexity has very nice CYA properties for NASA.

I prefer when rocket scientists design rockets, not politicians.

Requiring the LSS to have multiple refuelings clearly isn't a complete deal-breaker.


Given that admission, it's weird that you said you "need" only one refilling about eight times...  :-\
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/12/2022 09:50 pm
The fact that SLS and Orion will have less flight experience is irrelevant to the LSS.  Both SpaceX and NASA will do whatever they can to minimize risks, especially if it's cheap or easy to do so.  This is one that's both.
Avoiding a second refueling avoids some risks but it adds others.

It requires a custom extended tank, which adds risk.

I don't think it's a custom extended tank.  There are even more good reasons to go with a 1500t tank for lift tankers than there are for the LSS, not the least of which is that it doesn't require sending two tankers to NRHO for Option B.

Also, remember that "extended" really means "with ring segments, intertank bulkhead, and LCH4 dome rearranged."  That's not nothing, but as modifications go, it's not as bad as a lot of the mods that SpaceX has to make for LSS no matter what.

This also moves the center-of-gravity higher, increasing landing risk.

Landing a tall top-heavy rocket on the Moon is, as you put it, "a procedure with which we have little operational experience," so we should (in your words) "assign high risk to the procedure."
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/12/2022 09:57 pm
I don't think they need a depot.  In effect, they are depots, or at least spacecraft with good passive resistance to boil-off.  So you can refuel them long before the crew shows up, using plain ol' tankers that go out to NRHO via BLT and straight back to EDL.
I know. I'm just liking the model that says Starships mate with depots, but not with each other, since it makes all the plumbing very unambiguous.

Is "ambiguous plumbing" somehow a real risk? How, exactly?

It also lets the refueling be completely asynchronous. In this vision, a) a steady stream of tankers keeps the LEO depot full, b) occasionally a tanker fills up from the depot, flies to the moon, and pumps ~500t of prop into that depot (which really does achieve ZBO), so c) whenever an LSS needs to refuel, there's always fuel for it either in LEO or at Gateway.

You don't need non-androgynous refilling to "let" that idea work (it works just fine either way), so that's not really an argument in favor.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/12/2022 10:13 pm
If you are only 'permitted' one filling, then the math says that VLEO is the only sensible choice.

So essentially your argument is "if we assume no laddering, then there's no laddering." Imagine that...  ::)

In all my laddering calculations I assume the vehicle is fully fueled in VLEO first (ie the sensible approach), so apparently we were talking/calculating past each-other the entire time.

I don't think the "single refueling only" thing is a long-term constraint.  But I'll bet it is for Option A.  Having Artemis III fail, even if it didn't jeopardize the crew, would probably do the program in.  NASA's risk tolerance will be lower on that mission than on subsequent missions.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/12/2022 10:24 pm
Just to be sure I properly understand you: You're talking about refueling while accelerating, right? That is, you would:

a) launch a tanker and an LSS. (Each with 1500t fuel capacity.)
b) fully fuel both of them in LEO from a (rather large) depot. (Or two depots.)
c) couple the tanker to the LSS for refueling.
d) both of them fire together in formation, with the tanker continuously keeping the LSS topped up.
e) when the tanker is almost dry, it stops firing, disconnects, and reels in the fuel line.
f) at apogee, the tanker fires just a little bit to lower perigee to enable reentry.
g) the LSS goes on firing until it reaches TLI--arriving at the moon with about 500t extra fuel.

There's better version than this, and I think it's probably a winner:

a) Launch a tanker, but it's really not a tanker; it's just a pusher

Ahh yes, the return of the StarPusher!! :D

https://forum.nasaspaceflight.com/index.php?topic=49167.msg2006787#msg2006787

https://forum.nasaspaceflight.com/index.php?topic=52033.msg2140950#msg2140950



Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/12/2022 10:46 pm
This also moves the center-of-gravity higher, increasing landing risk.

Landing a tall top-heavy rocket on the Moon is, as you put it, "a procedure with which we have little operational experience," so we should (in your words) "assign high risk to the procedure."

That's a good point.

I have an old spreadsheet (many assumptions and guesses) that looks at CoG for various tank sizes.  Looks like it would move CoG up by 10% (from 17.2m for 1200t to 18.9m for 1500t) to go from 1200t to 1500t. 

That's not nothing.  And landing risk goes towards loss-of-crew, while refueling risk just goes towards loss-of-mission.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/12/2022 11:06 pm
Prop consumption for boosting two ships to an HEEO in a coupled system is the same as prop consumption for boosting the two ships separately.¹  If your objection here is, "But what about cross-feed?" the only thing cross-feed would get you is higher thrust

This is wrong. Crossfeed will change the effective mass ratio of the stages (since they'll have different burnout times), which directly increases delta-v, just like staging.

Mind you, I still think the overall idea is insane.  ;D
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/13/2022 05:14 am
Prop consumption for boosting two ships to an HEEO in a coupled system is the same as prop consumption for boosting the two ships separately.¹  If your objection here is, "But what about cross-feed?" the only thing cross-feed would get you is higher thrust

This is wrong. Crossfeed will change the effective mass ratio of the stages (since they'll have different burnout times), which directly increases delta-v, just like staging.

Mind you, I still think the overall idea is insane.  ;D

You don't care about the effective mass ratio; it's the same based on where you decide to put the energy of your HEEO.  If you're trying to be as prop-efficient as possible, that energy should be just high enough that the fully fueled target can complete the mission.

All three of these are equivalent in terms of total prop to LEO:

1) A target and a tanker independently boost to LEO+x HEEO, then the tanker completely fills the target, so it has just enough delta-v available to finish the mission.

2) The fully fueled target and a linked, cross-fed tank or tanker boost to LEO+x, then the tank or tanker is jettisoned, leaving the target with a full tank, giving it just enough delta-v to finish the mission.

3) A target is topped off, then a pusher boosts it to LEO+x and is jettisoned, leaving the target with just enough delta-v to finish the mission.

The only difference between them is that there will be slightly different amounts of non-impulsive gravity losses, depending on how much total thrust is generated by the two vehicles.  I'd be surprised if they varied by more than 20m/s.

Cross-feed for launch is actually the same, but the reduction in gravity loss can be much more substantial, because you have more high-thrust prop after the cross-feeding boosters are jettisoned.  Still, I'd be surprised if the difference was more than a couple hundred m/s.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 10/14/2022 08:36 am
Prop consumption for boosting two ships to an HEEO in a coupled system is the same as prop consumption for boosting the two ships separately.¹  If your objection here is, "But what about cross-feed?" the only thing cross-feed would get you is higher thrust

This is wrong. Crossfeed will change the effective mass ratio of the stages (since they'll have different burnout times), which directly increases delta-v, just like staging.

Mind you, I still think the overall idea is insane.  ;D

You don't care about the effective mass ratio

...

Cross-feed for launch is actually the same

This is obviously wrong.

If we imagine "perfect" crossfeed, the feeding stage would burn 100% of the consumed propellant before dropping off. At that point the math "degenerates" to conventional staging.

You're not suggesting that conventional staging has no delta-v benefits (vs a monolithic stage w the same mass ratio), are you?  :-\
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 10/14/2022 02:34 pm
Prop consumption for boosting two ships to an HEEO in a coupled system is the same as prop consumption for boosting the two ships separately.¹  If your objection here is, "But what about cross-feed?" the only thing cross-feed would get you is higher thrust

This is wrong. Crossfeed will change the effective mass ratio of the stages (since they'll have different burnout times), which directly increases delta-v, just like staging.

Mind you, I still think the overall idea is insane.  ;D

You don't care about the effective mass ratio

...

Cross-feed for launch is actually the same

This is obviously wrong.

If we imagine "perfect" crossfeed, the feeding stage would burn 100% of the consumed propellant before dropping off. At that point the math "degenerates" to conventional staging.

You're not suggesting that conventional staging has no delta-v benefits (vs a monolithic stage w the same mass ratio), are you?  :-\
He was explicit about what he is comparing.  You are mistaken.

Compare your continuous cross feed system with one that transfers the fuel instantaneously at the end of the first burn.  It should be obvious that it does not matter when the transfer occurs.  Refueling works out the same in "flat" space or if you use a HEEO so that all burns have the same Oberth effect. 

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/14/2022 05:44 pm
If we imagine "perfect" crossfeed, the feeding stage would burn 100% of the consumed propellant before dropping off. At that point the math "degenerates" to conventional staging.

You're not suggesting that conventional staging has no delta-v benefits (vs a monolithic stage w the same mass ratio), are you?

With the same mass ratio?  No, that definitely wouldn't have any benefits.  But the whole idea behind staging is that the average mass ratio is higher.

But this is all a red herring for the problem at hand, which is to minimize prop to orbit.  I assume that you agree that minimal prop will be achieved by:

1) Having the target Starship fully fueled on the first (and only) rung of the ladder.
2) Having the energy of that first rung only as high as needed so that the target Starship uses up its entire delta-v budget accomplishing the mission.

That means that the aggregate mass ratio needed to achieve the first rung is a constant, irrespective of the staging architecture:  independent tanker/target, cross-fed tank(er)/target, or target pushed by a stage.  They all require exactly the same amount of propellant, mod insignificant differences in non-impulsive gravity loss.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/14/2022 05:53 pm

But this is all a red herring for the problem at hand, which is to minimize prop to orbit.  I assume that you agree that minimal prop will be achieved by:

Why is that a major goal?  At $100/kg it's still not worth optimizing something that is 1/10 to 1/100 the development cost of something that will be made 3-4 times.

Airlines only optimize fuel because there's nothing else left to optimize.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 10/14/2022 06:09 pm
Requiring the LSS to have multiple refuelings clearly isn't a complete deal-breaker.  But I suspect that minimization of conops complexity will be high on the list of things that NASA wants to see.

I assert a conops with refilling of HLS Starship in the lunar vicinity reduces mission risk because it increases HLS tolerance of Orion schedule uncertainty.

Put differently, no one knows in which year (or decade) Orion will arrive to rendezvous with HLS. And although some Starships may have low or zero propellant boil-off, HLS likely won't. Having a low boil-off depot Starship in the cis-lunar vicinity allows propellant accumulation potentially years before the propellant is needed. And HLS Starship can be there years early too, patiently waiting for Orion to roll out to the pad. Days before the Orion launch window opens HLS fills from the depot and maneuvers to the rendezvous orbit.

How could NASA not like that plan?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 10/14/2022 07:14 pm
Requiring the LSS to have multiple refuelings clearly isn't a complete deal-breaker.  But I suspect that minimization of conops complexity will be high on the list of things that NASA wants to see.

I assert a conops with refilling of HLS Starship in the lunar vicinity reduces mission risk because it increases HLS tolerance of Orion schedule uncertainty.

Put differently, no one knows in which year (or decade) Orion will arrive to rendezvous with HLS. And although some Starships may have low or zero propellant boil-off, HLS likely won't. Having a low boil-off depot Starship in the cis-lunar vicinity allows propellant accumulation potentially years before the propellant is needed. And HLS Starship can be there years early too, patiently waiting for Orion to roll out to the pad. Days before the Orion launch window opens HLS fills from the depot and maneuvers to the rendezvous orbit.

How could NASA not like that plan?
Who pays for this? I guess you mean a modification to the milestone payments of the HLS contract? I'd think SpaceX would also want additional compensation for the fairly major extension to the agreed-to loiter time.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/14/2022 09:17 pm

But this is all a red herring for the problem at hand, which is to minimize prop to orbit.  I assume that you agree that minimal prop will be achieved by:

Why is that a major goal?  At $100/kg it's still not worth optimizing something that is 1/10 to 1/100 the development cost of something that will be made 3-4 times.

Airlines only optimize fuel because there's nothing else left to optimize.

For the time being, the figure of merit is not $/kg; it's launches/mission.  Eventually, you'll be able to model the process as one or more depots that provides a continuous process, but it's gonna be discrete until cadence is such that the depot doesn't boil dry between missions, which may be tens of months apart.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 10/14/2022 11:55 pm
Who pays for this? I guess you mean a modification to the milestone payments of the HLS contract? I'd think SpaceX would also want additional compensation for the fairly major extension to the agreed-to loiter time.

Short answer: yes. I believe funding will be available, both NASA-provided and SpaceX-internal, for a lunar vicinity Starship depot.

Caveats: (1) The puzzle has a lot of moving pieces. (2) My crystal ball is sometimes hazy.

Long answer:
I predict Artemis II will launch NET 2025, probably NET 3Q2025. Further, there will be considerable change to the way the Artemis program is conducted after the conclusion of that mission. Probably the most important change will be that around 4Q2025 Artemis management will have conducted a formal schedule risk assessment, and will have shared that with NASA leadership and Congress. Expectations will be that Artemis III launches NET 1Q2027, with considerable likelihood it might launch in 2H2027, or even after that.

That will be the assessment in 4Q2025. By that time lots of other puzzle pieces will have moved.
- The PPE/HALO Gateway hardware will be nowhere near its destination orbit.
- Starship propellant transfer in LEO will be routine.
- An uncrewed Starship will have landed on the surface of the Moon.
- A different uncrewed Starship, launched in 2024, will have performed a major propulsive maneuver in close proximity to Mars, having prevented propellant boil-off during the trans-Mars cruise.

Based on that last point, SpaceX will have a proven system for use as a lunar vicinity depot. Combined with the then-acknowledged Artemis III schedule uncertainty and the desire to "buy down risk," the decision to use a lunar vicinity depot for the crewed HLS Starship becomes easy.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/15/2022 03:16 am
Requiring the LSS to have multiple refuelings clearly isn't a complete deal-breaker.  But I suspect that minimization of conops complexity will be high on the list of things that NASA wants to see.

I assert a conops with refilling of HLS Starship in the lunar vicinity reduces mission risk because it increases HLS tolerance of Orion schedule uncertainty.

Put differently, no one knows in which year (or decade) Orion will arrive to rendezvous with HLS. And although some Starships may have low or zero propellant boil-off, HLS likely won't.

According to the source selection statement, LSS (aka HLS Starship) has 100 days of loiter without refueling.  To me, that implies that the prop management tech in LSS is roughly the same as that in a depot.

Quote
Having a low boil-off depot Starship in the cis-lunar vicinity allows propellant accumulation potentially years before the propellant is needed. And HLS Starship can be there years early too, patiently waiting for Orion to roll out to the pad. Days before the Orion launch window opens HLS fills from the depot and maneuvers to the rendezvous orbit.

You don't accumulate prop in cislunar; it's insanely expensive.  You accumulate it in LEO, then send it to cislunar shortly before the mission.  Even if the Orion gets massively delayed, topping off the depot in LEO is much, much cheaper than topping it off in cislunar.

However, there are a lot of different conops, only a few of which involve an Orion, or even a transit vehicle that's separate from the lander/ascender.  Before you even thought about a cislunar depot, you'd have to commit to one of the conops that required it.

As it happens, if you ever want to use LSS to stage crews out of LEO, before some form of lunar Starship can be crew-certified for launch and EDL, then you do need to commit to lunar refueling.  That requires at least an EDL-capable lift tanker to go to cislunar.  Whether it transfers prop to a depot or just transfers it straight to the LSS depends a lot on timing, and how boil-off-tolerant your mission is.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/15/2022 08:49 am


You don't accumulate prop in cislunar; it's insanely expensive.  You accumulate it in LEO, then send it to cislunar shortly before the mission.  Even if the Orion gets massively delayed, topping off the depot in LEO is much, much cheaper than topping it off in cislunar.


That depends largely on the operating realities of the company. If tankers are sitting idle, and actual launch costs are as low as some speculate, then keeping them busy with propellant launches may make business sense. Somewhat as Falcon launches Starlinks instead of sitting idle. Or as a cargo truck, plane, or ship is an expense sitting idle at the terminal. 

If more propellant than is needed is accumulating in LEO, and tankers are idle,  then sending some of it to cislunar can make sense. There are many missions enabled by large quantities of propellant in HEO. Any high energy mission would benefit by leaving HEO with full tanks.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 10/15/2022 07:03 pm
[...] accumulate[d] prop in cislunar [is] insanely expensive.

You and I agree on a lot, and we are in complete agreement on that point.

Quote
You accumulate it in LEO, then send it to cislunar ...

Again, complete agreement.

Quote
... shortly before the mission.

Sometime between now and 2026 one or the other of us will see this differently than they see it now!

Propellant in cis-lunar space is certainly costly. It is also valuable. The question is simply under what circumstances is the value greater than the cost?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/16/2022 09:37 pm
That depends largely on the operating realities of the company. If tankers are sitting idle, and actual launch costs are as low as some speculate, then keeping them busy with propellant launches may make business sense. Somewhat as Falcon launches Starlinks instead of sitting idle. Or as a cargo truck, plane, or ship is an expense sitting idle at the terminal. 

It's always cheaper to let something sit idle than it is to send it out for no reason.  Yes, it will always make sense to launch full tankers, even if the mission only requires a fractional tanker.  But you're not just going to launch prop to LEO depots for no reason.

You should also consider that if things are really that cheap, two things will happen:

1) Demand will spike, which will first soak up spare capacity and then increase prices.

2) The market will be hot enough that Starship will get some competition.  That may reduce prices somewhat, but it will also make SpaceX want to be as ops-efficient as possible.

In both cases, nobody's gonna be launching prop just for grins and giggles.

Quote
If more propellant than is needed is accumulating in LEO, and tankers are idle,  then sending some of it to cislunar can make sense.

No.  It never makes sense, absent a conops with sufficient cadence to generate the demand.  It's always better to leave excess prop in LEO until you need it.  Even if there's boil-off, the boiled-off prop costs a fraction of what it costs to replace the equivalent amount of prop in cislunar.¹

Quote
There are many missions enabled by large quantities of propellant in HEO. Any high energy mission would benefit by leaving HEO with full tanks.

Which missions do you have in mind?  And in what HEEO, with what RAAN and argument of perigee?  Are they crewed missions?  What are their abort contingencies?

One of the nice things about using cislunar depots for high-energy missions is that, unlike an HEEO, which is in a near-constant sidereal reference frame, the cis-lunar ones are in an Earth-Moon rotating reference frame, which means that you have a not-terrible window to any departure asymptote once a month, instead of once a year.

I do think that if refueling in HEEO is really the best way to do lunar missions, then an HEEO depot might make sense.  But even then, you don't move the prop to it until you need it, for exactly the same reason you don't move prop to cislunar depots until you need it:  boil-off costs more at high energy than at low energy.

____________
¹Weasel words:  If all you have is passive boil-off management, then cislunar is colder than LEO.  Whether it's enough colder that boil-off is cheaper would require some figuring.  But if there's enough traffic going to cislunar to make this at all useful, I'm pretty sure that SpaceX will have a zero boil-off depot technology, and LEO will then be just as good as anywhere else to store excess prop.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: redneck on 10/17/2022 08:44 am
It's always cheaper to let something sit idle than it is to send it out for no reason.  Yes, it will always make sense to launch full tankers, even if the mission only requires a fractional tanker.  But you're not just going to launch prop to LEO depots for no reason.

Actually it is not cheaper to let equipment and personnel sit idle if there is any revenue, or potential revenue work for them to do. Idle personnel cost as much as busy ones, and building inventory in good locations is potential revenue work. The alternatives are to eat the expense for losses, or lay them off. Neither is good for the long term if something useful can be done.

Which missions do you have in mind?  And in what HEEO, with what RAAN and argument of perigee?  Are they crewed missions?  What are their abort contingencies?

Anyone here can think of missions enabled by more propellant in a convenient location. Dropping down for an Oberth burn at perigee with a 20/1 mass ratio can get 18 km/sec Earth relative at infinity. Most planets and asteroids can be reached without gravity assist. HEO is different from HEEO. The most convenient orbits can be used based on demand.

One of the nice things about using cislunar depots for high-energy missions is that, unlike an HEEO, which is in a near-constant sidereal reference frame, the cis-lunar ones are in an Earth-Moon rotating reference frame, which means that you have a not-terrible window to any departure asymptote once a month, instead of once a year.

I do think that if refueling in HEEO is really the best way to do lunar missions, then an HEEO depot might make sense.  But even then, you don't move the prop to it until you need it, for exactly the same reason you don't move prop to cislunar depots until you need it:  boil-off costs more at high energy than at low energy.


Boil off in HEO can be eliminated with a fairly simple sun shade. In LEO it requires more technology.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: daavery on 10/17/2022 12:26 pm
I suspect that any refueling plan will ( for now)  require the 2 vehicles be in the same orbit, not just meeting at one point common between a circular orbit and an elliptical orbit. I will be very surprised if the fuel transfer takes less than 1 orbit. right now it takes over an hour to fuel a starship using 100s of HP driving the fueling pumps.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/17/2022 03:53 pm
I suspect that any refueling plan will ( for now)  require the 2 vehicles be in the same orbit, not just meeting at one point common between a circular orbit and an elliptical orbit.

to anyone who's played KSP for more than an hour would say 'now and forever'.

Or simply calculate the periapsis speeds of a circular and then an elliptic orbit.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 10/17/2022 05:36 pm
Clearly it needs to be the same orbit.

But KSP (which is fun, don’t get me wrong!) is a really bad guide for what is practical or not. It makes building new, many-stage launch vehicles super easy. It makes rendezvous and docking really hard (even though we’ve done this literally hundreds of times with ISS, etc). And it makes vertical landing recovery of multiple stages for reuse nearly impossible without modding, even though the vast majority of US rocket launches use this approach without problem. It also barely makes reuse worth it even when you do manage to make it happen. (And your point is valid, just wanted to mention that.)

There are also other methods that could allow the vehicles to be in separate orbits, like tethers. So “for now” is defensible.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/17/2022 06:41 pm
I suspect that any refueling plan will ( for now)  require the 2 vehicles be in the same orbit, not just meeting at one point common between a circular orbit and an elliptical orbit. I will be very surprised if the fuel transfer takes less than 1 orbit. right now it takes over an hour to fuel a starship using 100s of HP driving the fueling pumps.

Always the same orbit.

As for refueling time:  If you're in a circular VLEO, it doesn't really matter.  If you're in, say, LEO+2500m/s, that's about a 12hr orbit.  I doubt that two Starships that have boosted independently to HEEO can do rendezvous and docking in less than an orbit--even a 12hr orbit. 

If we allocate 2hr for undock and checkout before the remainder of the TLI burn, that would give us 10hr to pump up to 1200t.  That would require 33kg/s.  That's not an unreasonable transfer rate--especially in microgravity, with a few bar of ullage pressure and a set of pretty modest pumps.

On the other hand, if you want to do it in one orbit, and RPOD and undock/checkout can be reduced to, say, 8hr, then you'd need 83kg/s.  That's probably getting up there.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/17/2022 07:02 pm
That depends largely on the operating realities of the company. If tankers are sitting idle, and actual launch costs are as low as some speculate, then keeping them busy with propellant launches may make business sense. Somewhat as Falcon launches Starlinks instead of sitting idle. Or as a cargo truck, plane, or ship is an expense sitting idle at the terminal. 

It's always cheaper to let something sit idle than it is to send it out for no reason.  Yes, it will always make sense to launch full tankers, even if the mission only requires a fractional tanker.  But you're not just going to launch prop to LEO depots for no reason.

You should also consider that if things are really that cheap, two things will happen:

1) Demand will spike, which will first soak up spare capacity and then increase prices.

2) The market will be hot enough that Starship will get some competition.  That may reduce prices somewhat, but it will also make SpaceX want to be as ops-efficient as possible.

In both cases, nobody's gonna be launching prop just for grins and giggles.

Quote
If more propellant than is needed is accumulating in LEO, and tankers are idle,  then sending some of it to cislunar can make sense.

No.  It never makes sense, absent a conops with sufficient cadence to generate the demand.  It's always better to leave excess prop in LEO until you need it.  Even if there's boil-off, the boiled-off prop costs a fraction of what it costs to replace the equivalent amount of prop in cislunar.¹

Quote
There are many missions enabled by large quantities of propellant in HEO. Any high energy mission would benefit by leaving HEO with full tanks.

Which missions do you have in mind?  And in what HEEO, with what RAAN and argument of perigee?  Are they crewed missions?  What are their abort contingencies?

One of the nice things about using cislunar depots for high-energy missions is that, unlike an HEEO, which is in a near-constant sidereal reference frame, the cis-lunar ones are in an Earth-Moon rotating reference frame, which means that you have a not-terrible window to any departure asymptote once a month, instead of once a year.

I do think that if refueling in HEEO is really the best way to do lunar missions, then an HEEO depot might make sense.  But even then, you don't move the prop to it until you need it, for exactly the same reason you don't move prop to cislunar depots until you need it:  boil-off costs more at high energy than at low energy.

____________
¹Weasel words:  If all you have is passive boil-off management, then cislunar is colder than LEO.  Whether it's enough colder that boil-off is cheaper would require some figuring.  But if there's enough traffic going to cislunar to make this at all useful, I'm pretty sure that SpaceX will have a zero boil-off depot technology, and LEO will then be just as good as anywhere else to store excess prop.
Wellll, zero boiloff in LEO calls for more massive cooling hardware and PV panels than HEEO or cislunar. Same operating costs but not the same overall costs.


I agree that squeezing a tight launch schedule to do propellant launches for speculative use makes no sense. OTOH, if it's known that propellant will be needed within a known timeframe and there is a real honest to gosh lull in the launch schedule, it can make sense. If spaceflight is ever going to become routine, infrastructure has to be available on an ad hoc basis. Build it, and they will come.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: launchwatcher on 10/17/2022 07:05 pm
It makes rendezvous and docking really hard (even though we’ve done this literally hundreds of times with ISS, etc).
MechJeb does rendezvous & docking quite nicely.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/17/2022 09:25 pm
Wellll, zero boiloff in LEO calls for more massive cooling hardware and PV panels than HEEO or cislunar. Same operating costs but not the same overall costs.

Amortized over hundreds of launches?  I'd be surprised if beefier power and heat rejection moved the fully burdened cost by more than 1%.  The same can't be said for forward-deploying prop in cislunar without it being earmarked to a mission.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 10/18/2022 05:34 am
[...] forward-deploying prop in cislunar without it being earmarked to a mission.

Again we agree. I've not noticed anyone suggesting that though, so you might be fighting a straw man.

I think we likely agree the value of propellant is only realized when it is used for propulsion of a vehicle conducting a mission. And obviously for that it needs to be in the tanks of the mission vehicle.

For the class of missions (and it's quite large) where all the propellant needed for the mission can be loaded into the vehicle's tanks in LEO, that's by far the most efficient way to proceed. I just don't see SpaceX forever limiting itself solely to missions in that class.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 10/18/2022 02:38 pm
It's always cheaper to let something sit idle than it is to send it out for no reason.  Yes, it will always make sense to launch full tankers, even if the mission only requires a fractional tanker.  But you're not just going to launch prop to LEO depots for no reason.

Actually it is not cheaper to let equipment and personnel sit idle if there is any revenue, or potential revenue work for them to do. Idle personnel cost as much as busy ones, and building inventory in good locations is potential revenue work. The alternatives are to eat the expense for losses, or lay them off. Neither is good for the long term if something useful can be done.
Only if someone is paying you to sit loaded propellant depots in orbit.
Otherwise, all you have done is moved from groundside unsold raw materials (unlaunched Starship missions) that could be used for multiple services (prop launches, satellite launches, lunar missions, etc) serving a large market into an unsold in-orbit product (orbiting propellant) with a small or possibly non-existent market.

For using "spare" Starships to speculatively fill prop depots specifically, you're not using idle assets to generate revenue but instead spending money to build up idle inventory with no buyer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/18/2022 03:36 pm
It's always cheaper to let something sit idle than it is to send it out for no reason.  Yes, it will always make sense to launch full tankers, even if the mission only requires a fractional tanker.  But you're not just going to launch prop to LEO depots for no reason.

Actually it is not cheaper to let equipment and personnel sit idle if there is any revenue, or potential revenue work for them to do. Idle personnel cost as much as busy ones, and building inventory in good locations is potential revenue work. The alternatives are to eat the expense for losses, or lay them off. Neither is good for the long term if something useful can be done.
Only if someone is paying you to sit loaded propellant depots in orbit.
Otherwise, all you have done is moved from groundside unsold raw materials (unlaunched Starship missions) that could be used for multiple services (prop launches, satellite launches, lunar missions, etc) serving a large market into an unsold in-orbit product (orbiting propellant) with a small or possibly non-existent market.

For using "spare" Starships to speculatively fill prop depots specifically, you're not using idle assets to generate revenue but instead spending money to build up idle inventory with no buyer.

Great arguments if it's a business.

There's no profit to be had on the Moon, it's all government.

IF government had any sense, they'd replace the billions they are spending on SLS and custom lightweight space probes/satellites with Starship and heavy off-the-shelf components for probes/satellites, and use the saved billions for logistics instead.

Logistics was a large part of the western forts system cost in the 19th century America, for example, and it worked out well.

But modern government has no sense, so I'm just pipe dreaming.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/18/2022 04:43 pm
Wellll, zero boiloff in LEO calls for more massive cooling hardware and PV panels than HEEO or cislunar. Same operating costs but not the same overall costs.

Amortized over hundreds of launches?  I'd be surprised if beefier power and heat rejection moved the fully burdened cost by more than 1%.  The same can't be said for forward-deploying prop in cislunar without it being earmarked to a mission.
This is the amorphous time frame problem that we've both complained about. Both of us owe Mia Culpa's.


Mea Culpa.


I expect early depots to have one or two campaign lifetimes. The mission pace of Artemus will be low. If a HEO or cislunar depot is called for, with boiloff management easier in the high energy orbits, why not stockpile well in advance and keep operations smooth (see earlier comments on opportunistic tanker launches)?


If a depot is going to see hundreds of transfers this will most certainly be well down the road. Under these circumstances I seriously doubt keeping a depot filled will be a mission specific event, and inventory turnover will be high enough that the trade equations change in ways impossible to predict this early.


When SX has a system that works end to end, demonstrating launch, propellant transfer and EDL, we'll have a clearer idea of what the future holds. Wanna see a launch! Wanna se a launch!
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 10/18/2022 06:26 pm
This is the amorphous time frame problem that we've both complained about. Both of us owe Mia Culpa's.

Mea Culpa.

I expect early depots to have one or two campaign lifetimes. The mission pace of Artemus will be low. If a HEO or cislunar depot is called for, with boiloff management easier in the high energy orbits, why not stockpile well in advance and keep operations smooth (see earlier comments on opportunistic tanker launches)?

If a depot is going to see hundreds of transfers this will most certainly be well down the road. Under these circumstances I seriously doubt keeping a depot filled will be a mission specific event, and inventory turnover will be high enough that the trade equations change in ways impossible to predict this early.

When SX has a system that works end to end, demonstrating launch, propellant transfer and EDL, we'll have a clearer idea of what the future holds. Wanna see a launch! Wanna se a launch!

There's going to be an Era of Low BEO Cadence, and an Era of High BEO Cadence.  At low cadence, you should assume that anything left in orbit will boil dry, and timing lift tanker launches for specific missions is important.  At high cadence, prop becomes a lot more fungible, and you manage prop to LEO as a logistical problem that's separate from (albeit dependent on) the variety of LEO and BEO missions that need to be satisfied.

I'm concentrating on the Era of Low Cadence for the time being.  I don't think anybody knows how the market shakes out at high cadence.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/23/2022 05:21 pm
This is the amorphous time frame problem that we've both complained about. Both of us owe Mia Culpa's.

Mea Culpa.

I expect early depots to have one or two campaign lifetimes. The mission pace of Artemus will be low. If a HEO or cislunar depot is called for, with boiloff management easier in the high energy orbits, why not stockpile well in advance and keep operations smooth (see earlier comments on opportunistic tanker launches)?

If a depot is going to see hundreds of transfers this will most certainly be well down the road. Under these circumstances I seriously doubt keeping a depot filled will be a mission specific event, and inventory turnover will be high enough that the trade equations change in ways impossible to predict this early.

When SX has a system that works end to end, demonstrating launch, propellant transfer and EDL, we'll have a clearer idea of what the future holds. Wanna see a launch! Wanna se a launch!

There's going to be an Era of Low BEO Cadence, and an Era of High BEO Cadence.  At low cadence, you should assume that anything left in orbit will boil dry, and timing lift tanker launches for specific missions is important.  At high cadence, prop becomes a lot more fungible, and you manage prop to LEO as a logistical problem that's separate from (albeit dependent on) the variety of LEO and BEO missions that need to be satisfied.

I'm concentrating on the Era of Low Cadence for the time being.  I don't think anybody knows how the market shakes out at high cadence.
Yup. Back in the day, folks probably had to order gasoline from Hiram's Hardware Store a month in advance. When volume got high enough Hiram put a pump out front.


It's not a perfect analogy (none are) but following it a bit further, unless a farmers needs met some threshold, it made more sense to go out of route for gas than to install a pump out on the farm. The tradeoffs work different for depot locations but it's the same problem with the same decision points.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 10/25/2022 04:39 pm
This is the amorphous time frame problem that we've both complained about. Both of us owe Mia Culpa's.

Mea Culpa.

I expect early depots to have one or two campaign lifetimes. The mission pace of Artemus will be low. If a HEO or cislunar depot is called for, with boiloff management easier in the high energy orbits, why not stockpile well in advance and keep operations smooth (see earlier comments on opportunistic tanker launches)?

If a depot is going to see hundreds of transfers this will most certainly be well down the road. Under these circumstances I seriously doubt keeping a depot filled will be a mission specific event, and inventory turnover will be high enough that the trade equations change in ways impossible to predict this early.

When SX has a system that works end to end, demonstrating launch, propellant transfer and EDL, we'll have a clearer idea of what the future holds. Wanna see a launch! Wanna se a launch!

There's going to be an Era of Low BEO Cadence, and an Era of High BEO Cadence.  At low cadence, you should assume that anything left in orbit will boil dry, and timing lift tanker launches for specific missions is important.  At high cadence, prop becomes a lot more fungible, and you manage prop to LEO as a logistical problem that's separate from (albeit dependent on) the variety of LEO and BEO missions that need to be satisfied.

I'm concentrating on the Era of Low Cadence for the time being.  I don't think anybody knows how the market shakes out at high cadence.
Yup. Back in the day, folks probably had to order gasoline from Hiram's Hardware Store a month in advance. When volume got high enough Hiram put a pump out front.


It's not a perfect analogy (none are) but following it a bit further, unless a farmers needs met some threshold, it made more sense to go out of route for gas than to install a pump out on the farm. The tradeoffs work different for depot locations but it's the same problem with the same decision points.

Another similar situation was coal fired ships in the 19th century.  The British Empire had a network of refueling stations all around the world on the empire where "the sun never set".

How they built out that network of coal stations would be an interesting piece of history to research.

We still have a few remnants of that culture:

https://www.azlyrics.com/lyrics/nathanevans/wellermanseashanty.html
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/25/2022 07:21 pm
This is the amorphous time frame problem that we've both complained about. Both of us owe Mia Culpa's.

Mea Culpa.

I expect early depots to have one or two campaign lifetimes. The mission pace of Artemus will be low. If a HEO or cislunar depot is called for, with boiloff management easier in the high energy orbits, why not stockpile well in advance and keep operations smooth (see earlier comments on opportunistic tanker launches)?

If a depot is going to see hundreds of transfers this will most certainly be well down the road. Under these circumstances I seriously doubt keeping a depot filled will be a mission specific event, and inventory turnover will be high enough that the trade equations change in ways impossible to predict this early.

When SX has a system that works end to end, demonstrating launch, propellant transfer and EDL, we'll have a clearer idea of what the future holds. Wanna see a launch! Wanna se a launch!

There's going to be an Era of Low BEO Cadence, and an Era of High BEO Cadence.  At low cadence, you should assume that anything left in orbit will boil dry, and timing lift tanker launches for specific missions is important.  At high cadence, prop becomes a lot more fungible, and you manage prop to LEO as a logistical problem that's separate from (albeit dependent on) the variety of LEO and BEO missions that need to be satisfied.

I'm concentrating on the Era of Low Cadence for the time being.  I don't think anybody knows how the market shakes out at high cadence.
Yup. Back in the day, folks probably had to order gasoline from Hiram's Hardware Store a month in advance. When volume got high enough Hiram put a pump out front.


It's not a perfect analogy (none are) but following it a bit further, unless a farmers needs met some threshold, it made more sense to go out of route for gas than to install a pump out on the farm. The tradeoffs work different for depot locations but it's the same problem with the same decision points.

Another similar situation was coal fired ships in the 19th century.  The British Empire had a network of refueling stations all around the world on the empire where "the sun never set".

How they built out that network of coal stations would be an interesting piece of history to research.

We still have a few remnants of that culture:

https://www.azlyrics.com/lyrics/nathanevans/wellermanseashanty.html (https://www.azlyrics.com/lyrics/nathanevans/wellermanseashanty.html)
And there's a bit of a feedback loop too. Where you go impacts where refueling is needed. Where refueling is available impacts where you can go. It's not tightly bound but it is there.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Asteroza on 10/25/2022 11:04 pm
This is the amorphous time frame problem that we've both complained about. Both of us owe Mia Culpa's.

Mea Culpa.

I expect early depots to have one or two campaign lifetimes. The mission pace of Artemus will be low. If a HEO or cislunar depot is called for, with boiloff management easier in the high energy orbits, why not stockpile well in advance and keep operations smooth (see earlier comments on opportunistic tanker launches)?

If a depot is going to see hundreds of transfers this will most certainly be well down the road. Under these circumstances I seriously doubt keeping a depot filled will be a mission specific event, and inventory turnover will be high enough that the trade equations change in ways impossible to predict this early.

When SX has a system that works end to end, demonstrating launch, propellant transfer and EDL, we'll have a clearer idea of what the future holds. Wanna see a launch! Wanna se a launch!

There's going to be an Era of Low BEO Cadence, and an Era of High BEO Cadence.  At low cadence, you should assume that anything left in orbit will boil dry, and timing lift tanker launches for specific missions is important.  At high cadence, prop becomes a lot more fungible, and you manage prop to LEO as a logistical problem that's separate from (albeit dependent on) the variety of LEO and BEO missions that need to be satisfied.

I'm concentrating on the Era of Low Cadence for the time being.  I don't think anybody knows how the market shakes out at high cadence.
Yup. Back in the day, folks probably had to order gasoline from Hiram's Hardware Store a month in advance. When volume got high enough Hiram put a pump out front.


It's not a perfect analogy (none are) but following it a bit further, unless a farmers needs met some threshold, it made more sense to go out of route for gas than to install a pump out on the farm. The tradeoffs work different for depot locations but it's the same problem with the same decision points.

Another similar situation was coal fired ships in the 19th century.  The British Empire had a network of refueling stations all around the world on the empire where "the sun never set".

How they built out that network of coal stations would be an interesting piece of history to research.

We still have a few remnants of that culture:

https://www.azlyrics.com/lyrics/nathanevans/wellermanseashanty.html (https://www.azlyrics.com/lyrics/nathanevans/wellermanseashanty.html)
And there's a bit of a feedback loop too. Where you go impacts where refueling is needed. Where refueling is available impacts where you can go. It's not tightly bound but it is there.

If there were a few more nodes in the problem, simulating a solution using a slime mold network to look for solutions may be interesting.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 10/25/2022 11:54 pm
If there were a few more nodes in the problem, simulating a solution using a slime mold network to look for solutions may be interesting.

Or you might setup the nodes, depots, and campaigns in SpaceNet (https://forum.nasaspaceflight.com/index.php?topic=54991.msg2311191#msg2311191), and share your results and example apps (http://spacenet.mit.edu/applications.php) in forum.  That would be useful.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 11/02/2022 05:34 pm

From Ars Technica https://arstechnica.com/science/2022/11/spacex-is-now-building-a-raptor-engine-a-day-nasa-says/ (https://arstechnica.com/science/2022/11/spacex-is-now-building-a-raptor-engine-a-day-nasa-says/)

Quote
SpaceX plans to keep its low-Earth orbit propellant depots topped off with fuel for missions other than Artemis, Kirasich added. "So it's not like every time we go to the Moon we're going to start with an empty depot," he said.

The key points in this paragraph are 'depots' plural, and depots will be kept topped off for multiple uses.

It's not clear if the plural means finite lifetime with replacement or multiple depots on orbit at the same time. I'd go with multiple on orbit with the understanding that not right from the get go.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/02/2022 05:51 pm
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: wes_wilson on 11/03/2022 11:10 am
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.

That's slick.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: AbuSimbel on 11/03/2022 11:41 am
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.
Woah I think you just found the thing lol
The thing being SpaceX/Elon’s signature stroke of economic genius through which they can develop new capabilities while still competing in an actual market and not using the government as a cash cow.

Goes to show how big things are often written off as an overkill when the real problem is a lack of vision to leverage them
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 11/03/2022 12:57 pm
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.
That depends on the orbit of the depot and the target orbit of the payload. Regardless of payload mass you need to cart ~100 tonnes of Starship dry mass (plus a few thousand tonnes of prop) around for every burn, so plane changes would become extremely expensive (regardless of whether you moved Starship to match the depot or vice versa). You could well end up in a situation where you can only offload net a few hundred kg of propellant from a chuck-a-cubesat-out-the-side mission after you account for the propellant used to move between the depot orbit and the deployment orbit (and/or move the depot from its previous orbit to a matching one).
I could see a Transporter-esque service where you can fly your payload for cheap if and only if you are happy being dropped off into an orbit a low energy burn away from a depot orbit, but not a general load-prop-on-every-launch setup.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/03/2022 01:25 pm
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.
That depends on the orbit of the depot and the target orbit of the payload. Regardless of payload mass you need to cart ~100 tonnes of Starship dry mass (plus a few thousand tonnes of prop) around for every burn, so plane changes would become extremely expensive (regardless of whether you moved Starship to match the depot or vice versa). You could well end up in a situation where you can only offload net a few hundred kg of propellant from a chuck-a-cubesat-out-the-side mission after you account for the propellant used to move between the depot orbit and the deployment orbit (and/or move the depot from its previous orbit to a matching one).
I could see a Transporter-esque service where you can fly your payload for cheap if and only if you are happy being dropped off into an orbit a low energy burn away from a depot orbit, but not a general load-prop-on-every-launch setup.
I have not been following the transporter business. Is there a small set of customer-preferred planes? If so, put a depot in each plane. Now the problem is how to move the depots to where they will be needed after they are filled.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 11/03/2022 01:37 pm
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.
That depends on the orbit of the depot and the target orbit of the payload. Regardless of payload mass you need to cart ~100 tonnes of Starship dry mass (plus a few thousand tonnes of prop) around for every burn, so plane changes would become extremely expensive (regardless of whether you moved Starship to match the depot or vice versa). You could well end up in a situation where you can only offload net a few hundred kg of propellant from a chuck-a-cubesat-out-the-side mission after you account for the propellant used to move between the depot orbit and the deployment orbit (and/or move the depot from its previous orbit to a matching one).
I could see a Transporter-esque service where you can fly your payload for cheap if and only if you are happy being dropped off into an orbit a low energy burn away from a depot orbit, but not a general load-prop-on-every-launch setup.
I have not been following the transporter business. Is there a small set of customer-preferred planes? If so, put a depot in each plane. Now the problem is how to move the depots to where they will be needed after they are filled.
Transporter launches fly to dropoff orbits near Starlink deployment orbits, plus whatever orbits can be reached with one or two additional burns (usually only for headline customers paying extra) e.g. an apogee raise and circularisation.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: volker2020 on 11/03/2022 03:05 pm
I found another part of this article more interesting:
SpaceX plans to keep its low-Earth orbit propellant depots topped off with fuel for missions other than Artemis,

I see that as confirmation, that SpaceX will have many propellant depots in different planes, and will always have a easily accessible depot in reach, regardless when they start.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 11/03/2022 04:02 pm
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.

That's slick.

It's a good idea, but alas, SpaceX only makes about 20-24 non-SpaceX non-NASA launches per year, and at 2 / month that's maybe enough to top off from venting related to cooling but not enough to stock a depot.

https://en.wikipedia.org/wiki/List_of_Falcon_9_and_Falcon_Heavy_launches#2022_2

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/03/2022 04:30 pm
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.

That's slick.

It's a good idea, but alas, SpaceX only makes about 20-24 non-SpaceX non-NASA launches per year, and at 2 / month that's maybe enough to top off from venting related to cooling but not enough to stock a depot.

https://en.wikipedia.org/wiki/List_of_Falcon_9_and_Falcon_Heavy_launches#2022_2
Who says they can’t be SpaceX or NASA launches? Starlink may be volume constrained as Starship performance is dialed in, plus they may want to limit the number of Starlink satellites per launch because of plane constraints/etc.

And 20-24 launches per year is CURRENT rate, and that number has increased massively over time and no reason to think it won’t grow in the future. 20-24 is also a pretty large number. Extra propellant from them alone may be enough for an Artemis mission.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/03/2022 04:44 pm
The most obvious candidate for launches with extra propellant is smallsat rideshares. The destination is LEO (granted, often at higher inclinations than you might like), the total payload is very small, and to be competitive it has to occur regularly even if demand is relatively low and uneven. These customers are very cost conscious (or why not get a dedicated launch?), and there should be many opportunities.

Smallsat rideshare on tanker flights is a pretty obvious idea.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 11/03/2022 04:51 pm
Pretty interesting. So they’d be able to use really lightweight missions that would normally launch on Falcon 9 (or even Falcon 1), offer it for basically the marginal cost of launch, then the “profit” would be propellant they can load into a depot for other missions.

That's slick.

It's a good idea, but alas, SpaceX only makes about 20-24 non-SpaceX non-NASA launches per year, and at 2 / month that's maybe enough to top off from venting related to cooling but not enough to stock a depot.

https://en.wikipedia.org/wiki/List_of_Falcon_9_and_Falcon_Heavy_launches#2022_2
Who says they can’t be SpaceX or NASA launches? Starlink may be volume constrained as Starship performance is dialed in, plus they may want to limit the number of Starlink satellites per launch because of plane constraints/etc.

And 20-24 launches per year is CURRENT rate, and that number has increased massively over time and no reason to think it won’t grow in the future. 20-24 is also a pretty large number. Extra propellant from them alone may be enough for an Artemis mission.

24 launches *per year*.  I doubt unused prop is going to sit in orbit for an average of 4-5 months, which at 2/month is how long it will take to fill a depot with 20t payload flights that happen to have 120t of fuel with them.   What's the prop boiling rate again?   I suspect a depot turnover time will need to be less than a month.

The non-NASA non-SpaceX flights haven't been growing.  The existing space market is dominated by very expensive low volume satellites, and Falcon-9 has already lowered their launch costs to rounding error (the insurance probably costs more than the expendable 2nd stage for example)

Falcon-9 is picking up contracts because of schedule considerations, not cost, if I read the ESA wins correctly.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 11/03/2022 04:54 pm
Have we reached a consensus on what a depot turnover time might look like?

IOTW, how long can a depot sit in LEO with 1200t of fuel before it needs to be topped off with a 150t refueling due to fuel evaporation and its venting?

My guess is 1-3 weeks, not months.   But this thread is very long, so I might have missed something.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 11/04/2022 03:10 pm
Have we reached a consensus on what a depot turnover time might look like?

IOTW, how long can a depot sit in LEO with 1200t of fuel before it needs to be topped off with a 150t refueling due to fuel evaporation and its venting?

My guess is 1-3 weeks, not months.   But this thread is very long, so I might have missed something.
IIRC, real world boil off rates are one of the boxes yet to be checked. And what mitigation might be used is an unknown. There would also be unknown losses from RCS. It's hard to get a consensus when all of the important numbers are unknown. But then, I'm constantly impressed by the number of people who have strong opinions on things they have no understanding of - and equate opinions with knowledge. Thankfully not so much here at NSF.


Side note: I've suggested CMG's or reaction wheels for the depot in the past with no traction. In theory active cooling can achieve zero boiloff but there would still be RCS losses.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/04/2022 03:47 pm
Cryo cooled space telescopes (some in LEO) have liquid helium (sometimes even superfluid, all the way to 2 Kelvin) that lasts for a year to years. The boiloff rate is just gonna depend on what SpaceX decides is acceptable. There’s no reason they couldn’t make it last for months between topping off, even in LEO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/04/2022 03:49 pm
Boiloff gases can and almost certainly will be used to assist stationkeeping.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 11/04/2022 04:15 pm
Cryo cooled space telescopes (some in LEO) have liquid helium (sometimes even superfluid, all the way to 2 Kelvin) that lasts for a year to years. The boiloff rate is just gonna depend on what SpaceX decides is acceptable. There’s no reason they couldn’t make it last for months between topping off, even in LEO.
Those telescopes are also designed and built around the Helium dewar, and have to deal with a lot of work in eliminating thermal bridges to approach that performance.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/04/2022 05:44 pm
Cryo cooled space telescopes (some in LEO) have liquid helium (sometimes even superfluid, all the way to 2 Kelvin) that lasts for a year to years. The boiloff rate is just gonna depend on what SpaceX decides is acceptable. There’s no reason they couldn’t make it last for months between topping off, even in LEO.
Those telescopes are also designed and built around the Helium dewar, and have to deal with a lot of work in eliminating thermal bridges to approach that performance.
Of course. It takes aggressive engineering to get a very low boiloff rate. This is why I said the boiloff rate will be set to whatever SpaceX can tolerate.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 11/05/2022 02:20 pm
Boiloff gases can and almost certainly will be used to assist stationkeeping.
Yup. Boosting orbit a tad takes propellant in one form or another. Controlling attitude is another thing. I've no idea how much would be lost to this. If it's significant then CMGs or reaction wheels start to make sense. The higher the loss and the longer the expected depot lifetime, the more sense.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: volker2020 on 11/05/2022 03:07 pm
Cryo cooled space telescopes (some in LEO) have liquid helium (sometimes even superfluid, all the way to 2 Kelvin) that lasts for a year to years. The boiloff rate is just gonna depend on what SpaceX decides is acceptable. There’s no reason they couldn’t make it last for months between topping off, even in LEO.
Those telescopes are also designed and built around the Helium dewar, and have to deal with a lot of work in eliminating thermal bridges to approach that performance.
Of course. It takes aggressive engineering to get a very low boiloff rate. This is why I said the boiloff rate will be set to whatever SpaceX can tolerate.
Yes, it should be self evident that SpaceX can add as much isolation as they feel fit, they could add reflectors that shield the ship from heat radiation and or the heat radiation from earth, worst case they could even install a cooler (strongly doubt they would need it). There is no reason, why a tanker need to be full, so basically they could use 100t for extra equipment if they wanted. The real question is, what level of cooling makes economical sense.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Nomadd on 11/16/2022 03:51 am
 I hadn't seen this before. Is it in one of the other threads?
 The depot looks big enough for two starship loads.
 *Teslarati*
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/16/2022 04:27 am
what level of cooling makes economical sense.


One (incomplete) datapoint: cryocooler mass.

This PDF (https://ntrs.nasa.gov/api/citations/20160000337/downloads/20160000337.pdf) gives a scaling law on page 14, which simplifies (https://www.wolframalpha.com/input?i=simplify+%280.1773+x+%2B+14.223+%2B+0.0011+x+%2B+6.3577%29+%2B+.0493+%280.1773+x+%2B+14.223+%2B+0.0011+x+%2B+6.3577%29+-+0.492) to:

Quote
mass = 0.187195 kg/W * (cooling power) + 21.1033 kg

I believe this does not include the radiator, but it's hard to tell.






Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/16/2022 04:39 am
I hadn't seen this before. Is it in one of the other threads?
 The depot looks big enough for two starship loads.
 *Teslarati*

https://forum.nasaspaceflight.com/index.php?topic=50157.msg2406262#msg2406262

Depot seems to be roughly 61.5 meters (https://old.reddit.com/r/SpaceXLounge/comments/xa0ab2/nasa_has_released_a_new_paper_about_starship/inqyui7/) tall.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 11/16/2022 03:28 pm
I hadn't seen this before. Is it in one of the other threads?
 The depot looks big enough for two starship loads.

Not quite. But (as discussed throughout this thread) it is enough for the depot to refuel in circular-LEO, then push itself into the highest (safe) eccentric orbit which LSS can reach from launch. (Even a LEO/GEO eccentric orbit should be possible. If we ignore the VA belts.)

The benefit is that LSS takes on a full load of propellant that "contains" the higher energy of that eccentric orbit. (Or rather, when the LSS does a perigee burn to TLI, it gets an Oberth boost, reducing the propellant cost to TLI.) This solves a lot of the concerns about mass budgets for the lunar landing architecture; especially the "sustainable" model.

From a mission architecture POV, it doesn't greatly increase risks (again ignoring the VA belts), because LSS still only does one RPOD for refuelling. Plus all depot-filling risks are completed and stable before LSS launches, which is before SLS/Orion launches, so the risk of mission delay isn't drastically increased.

[edit: words are hard]
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Bryan Hayward on 11/16/2022 04:09 pm

From a mission architecture POV, it doesn't greatly increase risks (again ignoring the VA belts), because LSS still only does one RPOD for refuelling. Plus all depot-filling risks are completed and stable before LSS launches, which is before SLS/Orion launches, so the risk of mission delay isn't drastically increased.


Re: VA belts - there is a spot between the belts that has a much lower radiation load. I suspect that is where they will do this "high" orbit refuel. It doesn't make sense to do it inside an Allen Belt if you don't have to.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 11/16/2022 05:51 pm

From a mission architecture POV, it doesn't greatly increase risks (again ignoring the VA belts), because LSS still only does one RPOD for refuelling. Plus all depot-filling risks are completed and stable before LSS launches, which is before SLS/Orion launches, so the risk of mission delay isn't drastically increased.


Re: VA belts - there is a spot between the belts that has a much lower radiation load. I suspect that is where they will do this "high" orbit refuel. It doesn't make sense to do it inside an Allen Belt if you don't have to.

The sweet spot is about 4 Earth Radii (4 ER), which is about 20,000km - 25,000km.   That gets you 2km/sec of deltaV boost at periapsis (added to LEO), which is about what is needed.

One still traverses the lower belts an additional time.

OTOH, there's plenty of mass for Starship to add radiation protection far in excess of what Orion's mass budget can do.

https://docs.google.com/spreadsheets/d/18EEOrUn7T_JUXHeNxYsa4M5896ogaUgBfOaNANfdrck/

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/16/2022 09:00 pm
I hadn't seen this before. Is it in one of the other threads?
 The depot looks big enough for two starship loads.
 *Teslarati*

That comes from an IAC talk that some of the NASA people gave (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf), and it is referenced up-thread here somewhere.

I'm still skeptical that the depot is actually stretched, because it doesn't need to be stretched.¹  Mod whatever amount of boiloff is to be expected, it'll never need to be able to transfer more prop than however big the target's main tanks are.

Something that stretched has implications for the GSE.  Unless you can move the hard points down into the center of the tanks somewhere, I don't think the chopsticks can raise it up high enough to stack it on the booster.  Seems like a lot of work to make mid-tank hard points structurally sound enough to carry the (somewhat larger) dry mass.

______________
¹A few weasel words required here:
a) If you need to do two quick refills in a row, a big depot is handy.
b) If you have so much traffic that you have continuous transfers in and out of the depot, bigger is probably better.

However, I don't see either of those scenarios occurring until well after Option A.  There is a case in Option B refueling where, if your lift tankers only have 1200t tanks, you need two of them to haul prop out to NRHO for reuse, and doing so in rapid succession may have some operational simplicity. But 1200t tankers are kinda silly, because rearranging ring segments, domes, and inter-tank bulkheads is pretty easy, which yields a 1650t tanker with the same dry mass.  Then, even if you can't fill it completely full at launch, you can in LEO, and you can do the entire service mission to NRHO to get the LSS topped up for the next Artemis mission in one LEO-NRHO-transferProp-EDL pass.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/16/2022 09:20 pm
From a mission architecture POV, it doesn't greatly increase risks (again ignoring the VA belts), because LSS still only does one RPOD for refuelling. Plus all depot-filling risks are completed and stable before LSS launches, which is before SLS/Orion launches, so the risk of mission delay isn't drastically increased.

From my soi disant "HEEO hater" perch, the VA belts¹ aren't the only issue.  If you're not circular, then your argument of perigee is actually important.  If you stick a depot up into that eccentric orbit, you dramatically narrow the set of departure windows it can service.  So, yeah, you've kept your RPOD risks low, but you've introduced window risks, which in turn increase your boiloff risks. 

On top of that, any RAAN misalignments become more painful, because you'll have to do plane changes for TLI at higher speeds, which increases their delta-v consumption.

There is a way to minimize some of this:  Put the depot in LEO, load it up, then have it boost itself into the target HEEO just before the LSS is due to launch--or even after it's launched and sitting in a VLEO parking orbit.  The the LSS can top itself off, and the depot can propulsively return to LEO for the next mission.

That takes some of the risk out, but if there's a failed RPOD, you consume a lot more prop re-kludging the ω and RAAN than if you just dealt with giving the LSS big enough tanks to do LEO-NRHO-LS-NRHO in the first place.  And you can rearrange segments to get a 1500t tankage on the LSS and still have a 2m-high cylindrical payload bay for airlocks and hatches, plus whatever ogive space the crew module doesn't take up.

However, Twark has mentioned that up-thread that re-jiggering the tanks on the LSS this way does indeed raise the landing center of mass.  I think I got about 2m higher.  That increases lunar landing risk.

As usual, it's a very rich trade space.

__________
¹I got a lot less certain about VA belt risks when I noticed that Arty 2 plans to put the crew into a 300x60,000km HEEO for checkout, before continuing on to the free-return using the Orion OMS pod.  If NASA's OK with dipping their astronauts four times through the VA (once for HEEO insertion, once pre-TLI, once post-TLI, and once for TEI-to-EDL), then maybe this isn't such a big deal.  Of course, exposing one crew to these risks, which are balanced against the risks of an ICPS underperformance leaving them with bad abort options if TLI goes wrong, is different from exposing every crew to them.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/16/2022 09:25 pm
I hadn't seen this before. Is it in one of the other threads?
 The depot looks big enough for two starship loads.
 *Teslarati*

https://forum.nasaspaceflight.com/index.php?topic=50157.msg2406262#msg2406262

Depot seems to be roughly 61.5 meters (https://old.reddit.com/r/SpaceXLounge/comments/xa0ab2/nasa_has_released_a_new_paper_about_starship/inqyui7/) tall.

So, while you've got your pixel-counting hat on, how much clearance is there before the chopsticks hit the top of their travel?  And if it's not enough, can you move the depot's chopstick hard points so that they're attached to a spot where there's nothing but sheet metal and stringers?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/16/2022 09:38 pm

From a mission architecture POV, it doesn't greatly increase risks (again ignoring the VA belts), because LSS still only does one RPOD for refuelling. Plus all depot-filling risks are completed and stable before LSS launches, which is before SLS/Orion launches, so the risk of mission delay isn't drastically increased.


Re: VA belts - there is a spot between the belts that has a much lower radiation load. I suspect that is where they will do this "high" orbit refuel. It doesn't make sense to do it inside an Allen Belt if you don't have to.

If all you're doing is refueling an uncrewed LSS, then it matter--much--if your apogee is inside VA #1.

I got pelted with various brickbats when I suggested that having a single refueling RPOD might be a constraint for an Option A mission.  Most assume that you'd refuel once in LEO, then top off in HEEO, before continuing on to do HEEO-NRHO-LS-NRHO.  If that's the case, you might as well refuel in something GTO-like an have done with it.  However, RPODs have risks attached to them, and if you do want to limit things to a single refueling for Option A, then you're left with an apogee that's constrained by how high the LSS can get on the prop that it launched with.  That may easily be inside VA #1.

As for crewed refueling, you may be able to refuel in the gap, as long as you don't have RPOD problems.  If you do, then you dip your crew through the VA #1 twice for every additional orbit, until they decide to abort.  And even if everything goes peachy, you're still increasing crew exposure from a pair of VA #1 transits to four.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 11/16/2022 10:27 pm
If NASA's OK with dipping their astronauts four times through the VA (once for HEEO insertion, once pre-TLI, once post-TLI, and once for TEI-to-EDL), then maybe this isn't such a big deal.  Of course, exposing one crew to these risks, which are balanced against the risks of an ICPS underperformance leaving them with bad abort options if TLI goes wrong, is different from exposing every crew to them.
As for crewed refueling,

Not sure why you keep bringing this up. Crew xfer is in NRHO. LSS refuelling is uncrewed.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/17/2022 04:18 am
If NASA's OK with dipping their astronauts four times through the VA (once for HEEO insertion, once pre-TLI, once post-TLI, and once for TEI-to-EDL), then maybe this isn't such a big deal.  Of course, exposing one crew to these risks, which are balanced against the risks of an ICPS underperformance leaving them with bad abort options if TLI goes wrong, is different from exposing every crew to them.
As for crewed refueling,

Not sure why you keep bringing this up. Crew xfer is in NRHO. LSS refuelling is uncrewed.

Option A/B LSS is almost certainly the first application, but this is a general Starship refueling thread.  And beyond Option A and B, crewed LSS missions staged out of LEO, and eventually EDL-enabled missions launched direct from the ground, are all going to need a refueling architecture.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TomH on 11/17/2022 04:55 am

So, while you've got your pixel-counting hat on, how much clearance is there before the chopsticks hit the top of their travel?  And if it's not enough, can you move the depot's chopstick hard points so that they're attached to a spot where there's nothing but sheet metal and stringers?

In that the tower is modular, how difficult would it be to remove the top, add another section, and put the top back on? If there are multiple pads in the future, how difficult would it be to simply make the newer towers taller and launch variants of differing height from those taller towers?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/17/2022 04:59 am
Some reflection and reconsideration of Starship refilling now that Artemis 1 is through TLI seems timely.

The success of Artemis 1 will likely mean NASA will continue to drink the Kool-Aid awhile longer regarding the Artemis 3 schedule, so they will now want SpaceX to demonstrate Starship refilling ASAP, since in NASA's mind refilling might be the long pole for an Artemis 3 landing. That pressure could increase the priority at SpaceX of refilling as compared to the priority of Starlink launches. (Personally I would like that outcome, even without fully believing the reasoning is correct.)

HLS Starship is going to require some special modifications that require work, even for an uncrewed landing demonstration mission. Thus I prognosticate a mission requiring refilling before LEO departure will appear on the Starship manifest prior to the launch of the first HLS Starship. (Again personally, I'm hoping that demonstration mission is Mars-bound in 2024.)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/17/2022 05:06 am
Some reflection and reconsideration of Starship refilling now that Artemis 1 is through TLI seems timely.

The success of Artemis 1 will likely mean NASA will continue to drink the Kool-Aid awhile longer regarding the Artemis 3 schedule, so they will now want SpaceX to demonstrate Starship refilling ASAP, since in NASA's mind refilling might be the long pole for an Artemis 3 landing. That pressure could increase the priority at SpaceX of refilling as compared to the priority of Starlink launches. (Personally I would like that outcome, even without fully believing the reasoning is correct.)

HLS Starship is going to require some special modifications that require work, even for an uncrewed landing demonstration mission. Thus I prognosticate a mission requiring refilling before LEO departure will appear on the Starship manifest prior to the launch of the first HLS Starship. (Again personally, I'm hoping that demonstration mission is Mars-bound in 2024.)
Recall that we have seen several allusions to a demonstration of on-orbit fuel transfer to be done between two tanks on the same ship, so we might see that before we see actual ship-to-ship transfer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/17/2022 01:03 pm
Surely if you were going to lower the lifting hardpoints, the "natural" location would be the common dome seam. This area is already stronger, so it would require the least reinforcing mass.


SpaceX has shown that they're willing to bring in a portable crane for one-time (or otherwise "rare") operations, so I wouldn't lose any sleep over it. If altering the Starship lift points is somehow insurmountable (unlikely IMO), they already have a ready-made backup plan.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/17/2022 02:20 pm
Surely if you were going to lower the lifting hardpoints, the "natural" location would be the common dome seam. This area is already stronger, so it would require the least reinforcing mass.


SpaceX has shown that they're willing to bring in a portable crane for one-time (or otherwise "rare") operations, so I wouldn't lose any sleep over it. If altering the Starship lift points is somehow insurmountable (unlikely IMO), they already have a ready-made backup plan.
Also recall that Depot won't land, so these lift points are only used to lift the empty Depot gently and under controlled conditions. They are not used to catch a descending Depot. Since Depot launches are rare, they can also rig an external harness to allow the chopsticks to do the lift.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/17/2022 02:37 pm
If NASA's OK with dipping their astronauts four times through the VA (once for HEEO insertion, once pre-TLI, once post-TLI, and once for TEI-to-EDL), then maybe this isn't such a big deal.  Of course, exposing one crew to these risks, which are balanced against the risks of an ICPS underperformance leaving them with bad abort options if TLI goes wrong, is different from exposing every crew to them.
As for crewed refueling,

Not sure why you keep bringing this up. Crew xfer is in NRHO. LSS refuelling is uncrewed.

Option A/B LSS is almost certainly the first application, but this is a general Starship refueling thread.  And beyond Option A and B, crewed LSS missions staged out of LEO, and eventually EDL-enabled missions launched direct from the ground, are all going to need a refueling architecture.
By then there will have been dozens, perhaps hundreds, of refuelings.  Maybe the handwringing over RPODs will go away.  Just like the handwringing over load-and-go.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 11/17/2022 06:15 pm
Some reflection and reconsideration of Starship refilling now that Artemis 1 is through TLI seems timely.

The success of Artemis 1 will likely mean NASA will continue to drink the Kool-Aid awhile longer regarding the Artemis 3 schedule, so they will now want SpaceX to demonstrate Starship refilling ASAP, since in NASA's mind refilling might be the long pole for an Artemis 3 landing. That pressure could increase the priority at SpaceX of refilling as compared to the priority of Starlink launches. (Personally I would like that outcome, even without fully believing the reasoning is correct.)

HLS Starship is going to require some special modifications that require work, even for an uncrewed landing demonstration mission. Thus I prognosticate a mission requiring refilling before LEO departure will appear on the Starship manifest prior to the launch of the first HLS Starship. (Again personally, I'm hoping that demonstration mission is Mars-bound in 2024.)
Recall that we have seen several allusions to a demonstration of on-orbit fuel transfer to be done between two tanks on the same ship, so we might see that before we see actual ship-to-ship transfer.
How would we know if this happens if it isn't announced? Speculation has it that the first test would be between the headers and main tanks. There'd be no special hardware and the only external ops difference would be ullage settling. I doubt this would be apparent to a casual observer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/17/2022 06:49 pm
Some reflection and reconsideration of Starship refilling now that Artemis 1 is through TLI seems timely.

The success of Artemis 1 will likely mean NASA will continue to drink the Kool-Aid awhile longer regarding the Artemis 3 schedule, so they will now want SpaceX to demonstrate Starship refilling ASAP, since in NASA's mind refilling might be the long pole for an Artemis 3 landing. That pressure could increase the priority at SpaceX of refilling as compared to the priority of Starlink launches. (Personally I would like that outcome, even without fully believing the reasoning is correct.)

HLS Starship is going to require some special modifications that require work, even for an uncrewed landing demonstration mission. Thus I prognosticate a mission requiring refilling before LEO departure will appear on the Starship manifest prior to the launch of the first HLS Starship. (Again personally, I'm hoping that demonstration mission is Mars-bound in 2024.)
Recall that we have seen several allusions to a demonstration of on-orbit fuel transfer to be done between two tanks on the same ship, so we might see that before we see actual ship-to-ship transfer.
How would we know if this happens if it isn't announced? Speculation has it that the first test would be between the headers and main tanks. There'd be no special hardware and the only external ops difference would be ullage settling. I doubt this would be apparent to a casual observer.
If it's a formal milestone of the HLS contract, then we might not "see" it, but we will eventually see evidence of the progress payment. SpaceX or NASA might even say something about it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/17/2022 07:18 pm
By then there will have been dozens, perhaps hundreds, of refuelings.  Maybe the handwringing over RPODs will go away.  Just like the handwringing over load-and-go.

But that's why you want to keep crewed flights in mind when you're figuring this out.  It's not just RPOD hand-wringing.  There's radiation hand-wringing involved, too.

And even if the prox ops and docking become completely bulletproof, I'm not sure what happens with rendezvous.  AFAIK, eccentric orbit rendezvous, especially single-orbit eccentric rendezvous (again, radiation and crews--bad!) isn't exactly a thing right now.  I suspect that phasing is relatively straightforward, but you'll also have ω and RAAN adjustments that will be more interesting.  (Note:  ω--argument of perigee--isn't a thing for a circular orbit, but it is for an eccentric orbit.)

The big trade, which has to be made pretty early in the program, is whether sticking with 1200t mains for LSS and HEEO refueling is acceptable, or whether it's better to go to 1500t+ mains and circular VLEO refueling.  If they decide to go down the latter path, they don't need to worry about HEEO refueling for a long time, if ever.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 11/25/2022 02:06 am
By then there will have been dozens, perhaps hundreds, of refuelings.  Maybe the handwringing over RPODs will go away.  Just like the handwringing over load-and-go.

But that's why you want to keep crewed flights in mind when you're figuring this out.  It's not just RPOD hand-wringing.  There's radiation hand-wringing involved, too.

And even if the prox ops and docking become completely bulletproof, I'm not sure what happens with rendezvous.  AFAIK, eccentric orbit rendezvous, especially single-orbit eccentric rendezvous (again, radiation and crews--bad!) isn't exactly a thing right now.  I suspect that phasing is relatively straightforward, but you'll also have ω and RAAN adjustments that will be more interesting.  (Note:  ω--argument of perigee--isn't a thing for a circular orbit, but it is for an eccentric orbit.)

The big trade, which has to be made pretty early in the program, is whether sticking with 1200t mains for LSS and HEEO refueling is acceptable, or whether it's better to go to 1500t+ mains and circular VLEO refueling.  If they decide to go down the latter path, they don't need to worry about HEEO refueling for a long time, if ever.
In one of Eric's articles on the Artemus launch he mentioned that angle changed throughout the launch window. I think they have the mechanics down pat but might need tight restrictions on the exact launch time for any particular HEEO. That said, big tanks and VLEO sounds like a better idea.


Remind me. With all the variations discussed, I've lost track. If they do the big tanks and VLEO, how far can they get? Refuel on lunar orbit before or after landing?  Which lunar orbit?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/26/2022 01:57 am
Remind me. With all the variations discussed, I've lost track. If they do the big tanks and VLEO, how far can they get? Refuel on lunar orbit before or after landing?  Which lunar orbit?

If I assume a 95t dry LSS, 15t crew module, 7t deployable payload, and Isp=368s (with at least two R2SL's running), a 1500t LSS can just do VLEO-NRHOviaBLT-100dayLoiter-crewXfer-LS-10dayMission-NRHO-crewXfer.

That also assumes 150kg/day boiloff in NRHO, and 500kg/day boiloff on the lunar surface near the poles, and 2% delta-v FPR for all crewed segments.

The mission actually doesn't close if you do fast transit to NRHO.

You can easily do a mission with 1200t tanks if you're willing to refuel both in LEO and something like an LEO+2000m/s HEEO.  But AFAICT, that's not the plan.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/26/2022 10:51 am
15t crew module, 7t deployable payload

That's gonna look really weird with a giant almost-entirely-empty space ship.


Besides the bad optics, the shockingly small dry-mass-to-useful-payload ratio suggests strongly that it will be more efficient (as measured by launch mass per payload mass) to send more payload per vehicle.



To illustrate what I mean, take the extreme example: if only 1% of your landed mass were payload, then you could double your payload for only a ~1% increase in launch mass.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/26/2022 07:44 pm
15t crew module, 7t deployable payload

That's gonna look really weird with a giant almost-entirely-empty space ship.

Besides the bad optics, the shockingly small dry-mass-to-useful-payload ratio suggests strongly that it will be more efficient (as measured by launch mass per payload mass) to send more payload per vehicle.

To illustrate what I mean, take the extreme example: if only 1% of your landed mass were payload, then you could double your payload for only a ~1% increase in launch mass.

Four things:

1) It can look however it wants but it more than satisfies the Option A and B requirements, which are what LSS is intended to satisfy.

2) You can make it land heavy stuff easily if you're willing to do more than one refueling.  We've had that discussion up-thread.  I'm still assuming that minimization of conops complexity is a big deal for NASA.

3) I'll note that "a giant almost-entirely-empty space ship" is a pretty good description of any spacecraft that's designed to hold humans for anything other than a few days.  If you have almost the entire ogive for pressurized crew spaces, it's actually easier for them to get down to the cargo deck (where the elevator is) if the cylindrical portion of the payload bay is only a couple of meters high instead of 8m.

4) NASA clearly thinks that heavy cargo gets landed through CLPS, or the App. P Human-Class Delivery Lander (HDL) variant.  Using the same 95t dry and average Isp=368s assumptions, an expendable LSS can land 26t of cargo on the surface for 3 lift tankers, and 175t for 9 lift tankers.  (The latter assumes that a vanilla 120t-dry Starship can lift 150t to LEO.)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/26/2022 07:53 pm
Remind me. With all the variations discussed, I've lost track. If they do the big tanks and VLEO, how far can they get? Refuel on lunar orbit before or after landing?  Which lunar orbit?

If I assume a 95t dry LSS, 15t crew module, 7t deployable payload, and Isp=368s (with at least two R2SL's running), a 1500t LSS can just do VLEO-NRHOviaBLT-100dayLoiter-crewXfer-LS-10dayMission-NRHO-crewXfer.

That also assumes 150kg/day boiloff in NRHO, and 500kg/day boiloff on the lunar surface near the poles, and 2% delta-v FPR for all crewed segments.

The mission actually doesn't close if you do fast transit to NRHO.

You can easily do a mission with 1200t tanks if you're willing to refuel both in LEO and something like an LEO+2000m/s HEEO.  But AFAICT, that's not the plan.
AFAICT, your 1500t LSS isn't the plan, either.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/26/2022 08:38 pm
AFAICT, your 1500t LSS isn't the plan, either.

True.  So we're not seeing the whole plan yet.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 11/27/2022 12:20 am
Remind me. With all the variations discussed, I've lost track. If they do the big tanks and VLEO, how far can they get? Refuel on lunar orbit before or after landing?  Which lunar orbit?

If I assume a 95t dry LSS, 15t crew module, 7t deployable payload, and Isp=368s (with at least two R2SL's running), a 1500t LSS can just do VLEO-NRHOviaBLT-100dayLoiter-crewXfer-LS-10dayMission-NRHO-crewXfer.

That also assumes 150kg/day boiloff in NRHO, and 500kg/day boiloff on the lunar surface near the poles, and 2% delta-v FPR for all crewed segments.

The mission actually doesn't close if you do fast transit to NRHO.

You can easily do a mission with 1200t tanks if you're willing to refuel both in LEO and something like an LEO+2000m/s HEEO.  But AFAICT, that's not the plan.
IOW, it works if all the gods smile. Tight.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/27/2022 04:19 am
Remind me. With all the variations discussed, I've lost track. If they do the big tanks and VLEO, how far can they get? Refuel on lunar orbit before or after landing?  Which lunar orbit?

If I assume a 95t dry LSS, 15t crew module, 7t deployable payload, and Isp=368s (with at least two R2SL's running), a 1500t LSS can just do VLEO-NRHOviaBLT-100dayLoiter-crewXfer-LS-10dayMission-NRHO-crewXfer.

That also assumes 150kg/day boiloff in NRHO, and 500kg/day boiloff on the lunar surface near the poles, and 2% delta-v FPR for all crewed segments.

The mission actually doesn't close if you do fast transit to NRHO.

You can easily do a mission with 1200t tanks if you're willing to refuel both in LEO and something like an LEO+2000m/s HEEO.  But AFAICT, that's not the plan.
IOW, it works if all the gods smile. Tight.

Yeah, pretty tight.  But there is boil-off and some delta-v margin built in.  And of course we don't know what dry mass and crew module mass will be.  95t and 15t are guesses.  If it's more, then it's two refuelings for sure.  If it's less, things'll be fine.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/30/2022 04:16 am
15t crew module, 7t deployable payload

That's gonna look really weird with a giant almost-entirely-empty space ship.

Besides the bad optics, the shockingly small dry-mass-to-useful-payload ratio suggests strongly that it will be more efficient (as measured by launch mass per payload mass) to send more payload per vehicle.

To illustrate what I mean, take the extreme example: if only 1% of your landed mass were payload, then you could double your payload for only a ~1% increase in launch mass.

Four things:

1) It can look however it wants but it more than satisfies the Option A and B requirements, which are what LSS is intended to satisfy.

2) You can make it land heavy stuff easily if you're willing to do more than one refueling.  We've had that discussion up-thread.  I'm still assuming that minimization of conops complexity is a big deal for NASA.

How silly of me. I forgot this is a flags-and-footprints program, not a cost-per-kilogram program.

3) I'll note that "a giant almost-entirely-empty space ship" is a pretty good description of any spacecraft that's designed to hold humans for anything other than a few days.

Only in movies. ;)

  If you have almost the entire ogive for pressurized crew spaces, it's actually easier for them to get down to the cargo deck (where the elevator is) if the cylindrical portion of the payload bay is only a couple of meters high instead of 8m.

That results from shoehorning in an existing vehicle. It's a historically contingent accident, so we should be cautious generalizing it to "any spacecraft."
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/01/2022 03:48 am
Four things:

1) It can look however it wants but it more than satisfies the Option A and B requirements, which are what LSS is intended to satisfy.

2) You can make it land heavy stuff easily if you're willing to do more than one refueling.  We've had that discussion up-thread.  I'm still assuming that minimization of conops complexity is a big deal for NASA.

How silly of me. I forgot this is a flags-and-footprints program, not a cost-per-kilogram program.

So you quote both the point that it satisfies Option A/B and the one that says that you can make it as heavy as you want with a bit of extra complexity,¹ but your response is that it's only good for flags and footprints?

___________
¹Or not:  You can use an expendable LSS and land as much heavy stuff as you want via CLPS or HDL.  And unless tanker flights are priced at well under $10M, it's actually cheaper than trying to reuse a cargo Starship.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 12/04/2022 11:06 am
Four things:

1) It can look however it wants but it more than satisfies the Option A and B requirements, which are what LSS is intended to satisfy.

2) You can make it land heavy stuff easily if you're willing to do more than one refueling.  We've had that discussion up-thread.  I'm still assuming that minimization of conops complexity is a big deal for NASA.

How silly of me. I forgot this is a flags-and-footprints program, not a cost-per-kilogram program.

So you quote both the point that it satisfies Option A/B and the one that says that you can make it as heavy as you want with a bit of extra complexity,¹ but your response is that it's only good for flags and footprints?

___________
¹Or not:  You can use an expendable LSS and land as much heavy stuff as you want via CLPS or HDL.  And unless tanker flights are priced at well under $10M, it's actually cheaper than trying to reuse a cargo Starship.

Of course. That's all Artemis is, and that's what shaped those Option A/B requirements.

If you're doing any real work on the Moon, you "make it as heavy as you want you need to be economical." Using an extremely inefficient payload fraction ain't it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 12/04/2022 03:03 pm
Can someone write a succinct summary of Options A and B or else post a link to one? I hunted and hunted, but the most I could find was that Option A is pretty much what I already think of as HLS while option B is "the same thing, but sustainable."
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/04/2022 03:39 pm
Can someone write a succinct summary of Options A and B or else post a link to one? I hunted and hunted, but the most I could find was that Option A is pretty much what I already think of as HLS while option B is "the same thing, but sustainable."
It's a bit crazy. Technically, the formal requirements for Option A and Option B are embedded in NASA's RFP for the HLS, which I think is described as "Appendix H of NextSTEP-2". The Option A requirements envisioned a lander something like NASA's three-stage reference design, sort of a glorified Apollo LM, so the requirements are not very ambitious at all. Instead, SpaceX bid a Starship variant, which so grossly exceeded the requirements as to make them seem silly. Starship HLS has so much extra margin that then can add "sustainability" and meet the Option B requirements with (I believe) very little effort. The Option B sustainability requirements are (I think) the same as those for the Appendix P (second source) landers.

(Going from memory here, I may have messed it up.) The sustainability requirements are to support a crew of four, increase the mission duration to a full month, and increase the downmass and upmass from the tiny Option A contract requirements to the slightly less tiny Option B requirements. But "sustainable" is mostly about reduced cost per mission to make the Artemis program more sustainable. Reuse is not a requirement, but was given as an example of a way to reduce cost.

Since Starship HLS Option A will already almost meet Option B, I wonder if it would be cheaper for SpaceX to just build the initial HLS to meet Option B in the first place.

Edited to add:
    The actual reguirements documents are reachable from
   https://sam.gov/opp/d5460a204ab23cc0035c088dcc580d17/view
There is a whole set of PDFs in a zipped folder for "attachment F". Within the folder, the main doc is
     HLS-RQMT-001 SRD Rev R 20190927.pdf
It contains both the Option A and the Option B requirements, which I will now re-read to see how badly I misrepresented them in this post.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tbellman on 12/04/2022 08:04 pm
Can someone write a succinct summary of Options A and B or else post a link to one? I hunted and hunted, but the most I could find was that Option A is pretty much what I already think of as HLS while option B is "the same thing, but sustainable."

Attached is the requirements document from the RFP; this is a slightly newer version than the one you find at sam.gov that Dan Clemmensen pointed to.  Chapter 4 is the interresting stuff, with 4.2 containing the higher requirements for the sustained phase (i.e, for Option B).

As Dan mentioned, Option B requires transport of four crew, a bit more mass both up and down, and longer stays on the surface.  But also:
• Be able to dock to the Lunar Gateway (for Option A, providers could chose if they wanted to be able to dock to Orion or to Gateway).
• Be able to reach anywhere on the Moon, not just between 84°S and 90°S (this requires more Δv).
• Survive continous darkness for at least 50 hours (with a goal of up to 191 hours).
• Support longer EVA excursions.

But on the flip side, for the longer stays and when carrying four crew, the provider can assume that there will be a separate habitat waiting on the surface, so the lander does not need to be the habitat for the entire stay.

"Sustainability" here is not just about costs, but also about being more capable.  And also, as the requirements document says, "Beyond the initial HLS missions, the HLS will support sustainable presence on the Moon by providing a regular cadence of reliable transportation services for humans and cargo".

I believe that the Starship provided for Artemis III will not comply with all the requirements for Option B.  I think SpaceX (and NASA) will view that Starship as a kind of prototype to learn what works well and what works less good.  And since they will then plan on throwing away that ship anyway, they will not spend money and effort on outfitting it with features not needed for that first landing.  In particular, I suspect dealing with the lunar night, the extra Δv needed for global access, and procedures for refilling of propellant in NRHO (so the ship can be reused) will not be developed and ready for Artemis III.  (But this depends on how late Artemis III will be compared to how late Starship will be.)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/04/2022 08:46 pm

I believe that the Starship provided for Artemis III will not comply with all the requirements for Option B.  I think SpaceX (and NASA) will view that Starship as a kind of prototype to learn what works well and what works less good.  And since they will then plan on throwing away that ship anyway, they will not spend money and effort on outfitting it with features not needed for that first landing.  In particular, I suspect dealing with the lunar night, the extra Δv needed for global access, and procedures for refilling of propellant in NRHO (so the ship can be reused) will not be developed and ready for Artemis III.  (But this depends on how late Artemis III will be compared to how late Starship will be.)
Even if SpaceX does not implement all of Option B prior to Artemis III, they may design for it and then not actually install all of the required HW. In particular, they might choose to handle the night by adding a big battery. If so, they will decide prior to Artemis III where that battery will reside within the HLS.

The Δv needed for global access is already accounted for, I think. The difference is the number of tanker flights.

HLS reuse is not a specific requirement. SpaceX already intends to reuse "elements of the HLS system", namely the depot and tankers. This is better reuse than what was contemplated by NASA prior to the award, where they discussed reusing only the ascent module of a three-part HLS.

I think all instances of systems with very slow cadence are prototypes, with lessons learned with each use resulting in changes in the next instance. So sure, the uncrewed demo is a prototype for the Crewed Artemis III HLS, and the Artemis III HLS is a prototype for the Option B HLS. But Option B will be a design tweak, not a redesign and certainly not a new design. Caveat: it might be a big change at the mission level if they decide to put a depot into NRHO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/05/2022 12:35 am
Can someone write a succinct summary of Options A and B or else post a link to one? I hunted and hunted, but the most I could find was that Option A is pretty much what I already think of as HLS while option B is "the same thing, but sustainable."

Attached is the requirements document from the RFP; this is a slightly newer version than the one you find at sam.gov that Dan Clemmensen pointed to.  Chapter 4 is the interresting stuff, with 4.2 containing the higher requirements for the sustained phase (i.e, for Option B).

As Dan mentioned, Option B requires transport of four crew, a bit more mass both up and down, and longer stays on the surface.  But also:
• Be able to dock to the Lunar Gateway (for Option A, providers could chose if they wanted to be able to dock to Orion or to Gateway).
• Be able to reach anywhere on the Moon, not just between 84°S and 90°S (this requires more Δv).
• Survive continous darkness for at least 50 hours (with a goal of up to 191 hours).
• Support longer EVA excursions.

But on the flip side, for the longer stays and when carrying four crew, the provider can assume that there will be a separate habitat waiting on the surface, so the lander does not need to be the habitat for the entire stay.

"Sustainability" here is not just about costs, but also about being more capable.  And also, as the requirements document says, "Beyond the initial HLS missions, the HLS will support sustainable presence on the Moon by providing a regular cadence of reliable transportation services for humans and cargo".

I believe that the Starship provided for Artemis III will not comply with all the requirements for Option B.  I think SpaceX (and NASA) will view that Starship as a kind of prototype to learn what works well and what works less good.  And since they will then plan on throwing away that ship anyway, they will not spend money and effort on outfitting it with features not needed for that first landing.  In particular, I suspect dealing with the lunar night, the extra Δv needed for global access, and procedures for refilling of propellant in NRHO (so the ship can be reused) will not be developed and ready for Artemis III.  (But this depends on how late Artemis III will be compared to how late Starship will be.)

I agree that there are lots of lessons to be learned in Option A that will necessitate at least some tweaks/redesign before they have an Option B-compliant offering.

One obvious missing capability to turn the Option A LSS into Option B is the ability to refuel in cislunar, which is how the LSS becomes reusable.  That capability actually can't be exercised until the second mission (presumably Arty 6) of the Option B LSS (i.e., the one for Arty 4).

That refueling capability also makes it easy for the LSS to have access to all lunar locations, and may even provide a cheap way to survive darkness, by using an APU and however much methalox you need to provide the needed power.

One of the things we haven't discussed a lot is that the existence of both cislunar and LEO refueling not only enables Option B, but it also dramatically reduces the cost for other providers to provide Appendix P / SLT services.  This obviously requires other providers to accept SpaceX/Starship methalox refueling as a first option, probably with some kind of backup system (FH, Vulcan, New Glenn...) in case Starship goes down for a while.  However, in general, App. P bidders can provide much more capable offerings if they can buy methalox from SpaceX in NRHO.

This does require that SpaceX be willing to offer their refueling interconnects and any docking/stabilization mechanisms as an open system, and be willing to contract with third parties to prove that depot-to-3rd-party refueling works cleanly.  But the upside of doing so is massive for the App. P providers and NASA.

The question is how aggressively SpaceX might want to go into the open refueling business.  But my guess is that it's a very nice, very profitable business, with close to double the number of tanker launches they'd get if they're only responsible for their own HLS/HDL birds, which will presumably get only half the missions, per NASA policy.  And everybody wins.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 12/05/2022 05:10 pm
One obvious missing capability to turn the Option A LSS into Option B is the ability to refuel in cislunar, which is how the LSS becomes reusable. 
Don't they also need some way to inspect/test the vehicle before reuse? That seems to me to be the biggest missing piece--although it provides a great reason to have a crew at Gateway. :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/05/2022 05:43 pm
One obvious missing capability to turn the Option A LSS into Option B is the ability to refuel in cislunar, which is how the LSS becomes reusable. 
Don't they also need some way to inspect/test the vehicle before reuse? That seems to me to be the biggest missing piece--although it provides a great reason to have a crew at Gateway. :-)
There is no hard requirement in Option B for reuse of the HLS lander. The requirement is for "sustainable" access to the moon, meaning (more or less) one or more times a year for not too much money. The original NASA reference lander has three elements: transit, descent, and ascent, and only the ascent element was reusable in its Option B. The Starship HLS system has three elements: Depot, Tanker, and HLS. The tanker and Depot are reusable and the HLS is not very expensive, so even if it is expended  the system still meets the sustainability requirement. It is probably cheaper to expend it than to refuel it, but the big problems for reuse are reprovisioning and loading new large cargo. Provisioning and cargo are easy to load on a new HLS on Earth, much harder to transfer in space (cislunar, LEO, or other).

HLS is "not very expensive" because much of it is built from standard Starship parts in a high-production SpaceX factory.

Starship HLS also increases sustainability by eliminating most separate cargo missions. Cargo missions have generally been assumed to be one-way anyway.

It might make sense to reuse Starship HLS if SLS/Orion is retired and HLS conveys the crew from LEO to the Lunar surface and back.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 12/05/2022 06:49 pm
One obvious missing capability to turn the Option A LSS into Option B is the ability to refuel in cislunar, which is how the LSS becomes reusable. 
Don't they also need some way to inspect/test the vehicle before reuse? That seems to me to be the biggest missing piece--although it provides a great reason to have a crew at Gateway. :-)
There is no hard requirement in Option B for reuse of the HLS lander. The requirement is for "sustainable" access to the moon, meaning (more or less) one or more times a year for not too much money. The original NASA reference lander has three elements: transit, descent, and ascent, and only the ascent element was reusable in its Option B. The Starship HLS system has three elements: Depot, Tanker, and HLS. The tanker and Depot are reusable and the HLS is not very expensive, so even if it is expended  the system still meets the sustainability requirement. It is probably cheaper to expend it than to refuel it, but the big problems for reuse are reprovisioning and loading new large cargo. Provisioning and cargo are easy to load on a new HLS on Earth, much harder to transfer in space (cislunar, LEO, or other).

HLS is "not very expensive" because much of it is built from standard Starship parts in a high-production SpaceX factory.

Starship HLS also increases sustainability by eliminating most separate cargo missions. Cargo missions have generally been assumed to be one-way anyway.

It might make sense to reuse Starship HLS if SLS/Orion is retired and HLS conveys the crew from LEO to the Lunar surface and back.
The cost of the LSS is the crew habitat and other LSS crew related addons to the basic SS. A basic SS hardware may be easily as low as $50M. But an LSS hardware is likely to run at greater than $150M. NOTE here is that a depot and a BEO tanker would be at the close to a basic SS costs. Thus such to use an LSS at least 3 times will gain significant savings overall for surface missions of $50 to >$100M on each mission. Which can say a lot about a stripped down oneway cargo hardware to cost ~$75M for the cargo SS that delivers about 75 to 100t of cargo to the surface.

Even using a throw away tanker vs a Lunar Depot you would still save money by just reusing the LSS 3 to 5 times. You could even refuel the LSS for return to LEO so it could be loaded for next mission and still save money. Such that initially only Tankers are thrown away until its possible to return EDL the Tanker from Lunar orbit. Once that happens an additional ~$50M is saved per mission.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/06/2022 05:54 pm
One obvious missing capability to turn the Option A LSS into Option B is the ability to refuel in cislunar, which is how the LSS becomes reusable. 
Don't they also need some way to inspect/test the vehicle before reuse? That seems to me to be the biggest missing piece--although it provides a great reason to have a crew at Gateway. :-)
There is no hard requirement in Option B for reuse of the HLS lander. The requirement is for "sustainable" access to the moon, meaning (more or less) one or more times a year for not too much money. The original NASA reference lander has three elements: transit, descent, and ascent, and only the ascent element was reusable in its Option B. The Starship HLS system has three elements: Depot, Tanker, and HLS. The tanker and Depot are reusable and the HLS is not very expensive, so even if it is expended  the system still meets the sustainability requirement. It is probably cheaper to expend it than to refuel it, but the big problems for reuse are reprovisioning and loading new large cargo. Provisioning and cargo are easy to load on a new HLS on Earth, much harder to transfer in space (cislunar, LEO, or other).

HLS is "not very expensive" because much of it is built from standard Starship parts in a high-production SpaceX factory.

Starship HLS also increases sustainability by eliminating most separate cargo missions. Cargo missions have generally been assumed to be one-way anyway.

It might make sense to reuse Starship HLS if SLS/Orion is retired and HLS conveys the crew from LEO to the Lunar surface and back.
The cost of the LSS is the crew habitat and other LSS crew related addons to the basic SS. A basic SS hardware may be easily as low as $50M. But an LSS hardware is likely to run at greater than $150M. NOTE here is that a depot and a BEO tanker would be at the close to a basic SS costs. Thus such to use an LSS at least 3 times will gain significant savings overall for surface missions of $50 to >$100M on each mission. Which can say a lot about a stripped down oneway cargo hardware to cost ~$75M for the cargo SS that delivers about 75 to 100t of cargo to the surface.

Even using a throw away tanker vs a Lunar Depot you would still save money by just reusing the LSS 3 to 5 times. You could even refuel the LSS for return to LEO so it could be loaded for next mission and still save money. Such that initially only Tankers are thrown away until its possible to return EDL the Tanker from Lunar orbit. Once that happens an additional ~$50M is saved per mission.
I question the $50m price tag on SS. An F9 full stack has been estimated at $30m and never hit the production numbers that SS is being designed for. Maybe something in the $10-25m range? A cost that low might skew the numbers against reuse although the landing engines would bump the costs.

A point to remember is the need for later landings to host a crew of four and stay alive through extended darkness. If folding this into the first build bumps the cost enough to justify reuse, so be it, but is does crimp redesign from 'lessons learned'.

In principal I am in favor of as much reuse as possible but I just can't make the case close in this instance. The build rate, the build cost, the expanding mission requirements, the inability to redesign for lessons learned, all seem to work against it.

OTOH, when it time for the 4 crew
overnight version, reuse might make sense. 
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Anguy on 12/06/2022 06:25 pm
I'd like to ask how effective are current TPS tiles as long time insulation for supply depot? Most proposals suggest stripping the depot bare, but wouldn't it be more effective to do the opposite a cover it all around with tiles? They could be dual used as simple omnidirectional insulation (so Earth radiation would be blocked too) instead of some deployable sunshield and the depot can also be used for Lunar missions and return to LEO by aerobraking, saving a lot of propellant.
Adding 20+tons to ship dry mass is not that big problem, if they can be used this way...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/06/2022 06:50 pm
One obvious missing capability to turn the Option A LSS into Option B is the ability to refuel in cislunar, which is how the LSS becomes reusable. 
Don't they also need some way to inspect/test the vehicle before reuse? That seems to me to be the biggest missing piece--although it provides a great reason to have a crew at Gateway. :-)
There is no hard requirement in Option B for reuse of the HLS lander. The requirement is for "sustainable" access to the moon, meaning (more or less) one or more times a year for not too much money. The original NASA reference lander has three elements: transit, descent, and ascent, and only the ascent element was reusable in its Option B. The Starship HLS system has three elements: Depot, Tanker, and HLS. The tanker and Depot are reusable and the HLS is not very expensive, so even if it is expended  the system still meets the sustainability requirement.

Mostly agree.  I assume that by "expendable HLS" you mean one that returns to NRHO but isn't reused.  Leaving it on the surface is obviously hard on the crew...

Quote
HLS is "not very expensive" because much of it is built from standard Starship parts in a high-production SpaceX factory.

Probably true from a bill-of-materials standpoint, but almost certainly not true from an assembly and certification standpoint.  Crew-rated ships are harder simply because the quality control and documentation requirements are massive.

Quote
Starship HLS also increases sustainability by eliminating most separate cargo missions. Cargo missions have generally been assumed to be one-way anyway.

I think it's highly unlikely that most crewed missions will carry heavy unpressurized cargo, for the following reasons:

1) For new HLS ships, the LEO-NRHO-LS-NRHO conops is sufficiently tight that heavy cargoes often will blow your delta-v budget.  This can be avoided if you're willing to have a crew do a refuel in an HEEO but you don't want that.  It's operationally complex, risks the crew during refueling ops, and the crew gets exposed to more radiation than necessary.

2) Doing on-orbit transfer of heavy cargo through the HLS hatch is an unsolved problem, which is likely to be really hard.  The cargo you'd transfer is incredibly expensive, and subjecting it to the risks of unstowing it from the supply ship, getting a robot arm to precision-position it in the target ship, batten it down for transport, and then essentially re-integrate it, simply subjects it to too much risk.

3) Heavy cargo launches on a dumb, expendable (left on the surface) LSS are cheap.  No crew involved, so manufacturing costs are low, and you can get almost 40t of cargo on an LSS CLPS or HDL mission for a measly 3 tankers of prop.

4) There won't be any such thing as extremely heavy cargo on an HLS mission, because any payload constructed will likely have to be compatible with whatever the other Appendix P offering is--and it won't be capable of handling heavy cargo.  AIUI, the requirement is only for pressurized cargo that can be passed through an IDA tunnel.

Quote
It might make sense to reuse Starship HLS if SLS/Orion is retired and HLS conveys the crew from LEO to the Lunar surface and back.

It makes all kinds of sense.  But it doesn't help the cargo situation.  microgravity is microgravity, and vacuum integration is vacuum integration.

Side note (and pet peeve):  One of the other App P vendors can make an absolute killing if they design their ship to be transported to NRHO on a vanilla Starship, then recaptured and returned to EDL on the same Starship.  Not only does this allow heavy cargo to be integrated in a regular payload processing facility, but it also allows the vendor to thoroughly clean and refurbish the HLS, which will almost certainly extend the number of missions it can perform.  They'd obviously need a Plan B in case Starship dropped out of service for a while, but the advantages (and cost savings) are huge.  I doubt Blue would do this, but Dynetics might.  And so might Lockmart.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 12/06/2022 07:20 pm
One obvious missing capability to turn the Option A LSS into Option B is the ability to refuel in cislunar, which is how the LSS becomes reusable. 
Don't they also need some way to inspect/test the vehicle before reuse? That seems to me to be the biggest missing piece--although it provides a great reason to have a crew at Gateway. :-)
There is no hard requirement in Option B for reuse of the HLS lander. The requirement is for "sustainable" access to the moon, meaning (more or less) one or more times a year for not too much money. The original NASA reference lander has three elements: transit, descent, and ascent, and only the ascent element was reusable in its Option B. The Starship HLS system has three elements: Depot, Tanker, and HLS. The tanker and Depot are reusable and the HLS is not very expensive, so even if it is expended  the system still meets the sustainability requirement. It is probably cheaper to expend it than to refuel it, but the big problems for reuse are reprovisioning and loading new large cargo. Provisioning and cargo are easy to load on a new HLS on Earth, much harder to transfer in space (cislunar, LEO, or other).

HLS is "not very expensive" because much of it is built from standard Starship parts in a high-production SpaceX factory.

Starship HLS also increases sustainability by eliminating most separate cargo missions. Cargo missions have generally been assumed to be one-way anyway.

It might make sense to reuse Starship HLS if SLS/Orion is retired and HLS conveys the crew from LEO to the Lunar surface and back.
The cost of the LSS is the crew habitat and other LSS crew related addons to the basic SS. A basic SS hardware may be easily as low as $50M. But an LSS hardware is likely to run at greater than $150M. NOTE here is that a depot and a BEO tanker would be at the close to a basic SS costs. Thus such to use an LSS at least 3 times will gain significant savings overall for surface missions of $50 to >$100M on each mission. Which can say a lot about a stripped down oneway cargo hardware to cost ~$75M for the cargo SS that delivers about 75 to 100t of cargo to the surface.

Even using a throw away tanker vs a Lunar Depot you would still save money by just reusing the LSS 3 to 5 times. You could even refuel the LSS for return to LEO so it could be loaded for next mission and still save money. Such that initially only Tankers are thrown away until its possible to return EDL the Tanker from Lunar orbit. Once that happens an additional ~$50M is saved per mission.
I question the $50m price tag on SS. An F9 full stack has been estimated at $30m and never hit the production numbers that SS is being designed for. Maybe something in the $10-25m range? A cost that low might skew the numbers against reuse although the landing engines would bump the costs.

A point to remember is the need for later landings to host a crew of four and stay alive through extended darkness. If folding this into the first build bumps the cost enough to justify reuse, so be it, but is does crimp redesign from 'lessons learned'.

In principal I am in favor of as much reuse as possible but I just can't make the case close in this instance. The build rate, the build cost, the expanding mission requirements, the inability to redesign for lessons learned, all seem to work against it.

OTOH, when it time for the 4 crew
overnight version, reuse might make sense.
My near term SS cost number for a Tanker equivalent is the ~$50M number. In the 2030 time-frame that number does decrease in my cost modeling to a little lower at the $30-40M level. The cost model is derived from some observed data about the superstructure manufacturing costs based on manpower during the initial SS suborbital builds and tests. I still use this data since it was easier to come to reasonable value for this cost since the manpower totals possible for this work was much more evident than now. Since then a significant amount of the build has been automated but so has the complexity due to swapping wall thickness for use of stringers to lower mass.

The bottom line here on the Depot and refueling is that having the depots do EDL may not be that important. Since the cost of a Depot that is likely to become obsolete in a couple of years for it to EDL so it can be scrapped is so low as in significantly less than $100M if even anywhere near that. And by not having the EDL equipment added it would save on cost for that while adding some additional equipment for long duration orbital operational stay. Such that a Depot that is used for 5 missions becomes ~$10M cost per mission that may have as many as 8 Tankers in one mission. As such that then would possibly make total prop cost for a mission to start at a value even less than $150M (good possibility that eventually a cost of $80-100M) to no higher than $400M when the Tankers have yet to be reused.

So evaluate design speculations by looking at cost of the various options. Since SpaceX will go for the ultimate mission lowest cost while still being able to have high number of missions usability rate/year.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/06/2022 07:34 pm
I'd like to ask how effective are current TPS tiles as long time insulation for supply depot? Most proposals suggest stripping the depot bare, but wouldn't it be more effective to do the opposite a cover it all around with tiles? They could be dual used as simple omnidirectional insulation (so Earth radiation would be blocked too) instead of some deployable sunshield and the depot can also be used for Lunar missions and return to LEO by aerobraking, saving a lot of propellant.
Adding 20+tons to ship dry mass is not that big problem, if they can be used this way...

TL;DR version: the TPS will help a little bit, but not enough.

The longer version is complicated.  TPS really does three things:

1) It's an insulator:  it has a very low heat conductivity.  If you put a big heat pulse on one side of it, it takes a long time for the pulse to reach the other side.  But eventually, all things being in equilibrium, it will transfer just as much heat, at the same rate, once it reaches equilibrium.  Not great for long-term storage.  However:

2) It also has high radiant emittance.  The amount of heat re-radiated into space is εσT⁴, where ε is the emissivity (loosely, how much the material behaves like a black body), σ is a constant, 5.76E-8 W/m²/K⁴, and T is the temperature difference between the surface and the environment.

3) It's a reflector.  TPS does this very poorly, but any radiation that just bounces off obviously doesn't heat up anything.

During reentry, the high emissivity property dominates, because as the temperature rises, the emittance increases as the fourth power.  The reflectivity is relatively unimportant.  But the insulation is very important because reentry is short, so not much heat can soak through the insulation.  (Note that there's a non-trivial problem after a spacecraft lands, because there can be a big enough heat pulse making its way through the insulation to damage the spacecraft, and there's no way to remove it--because it's a good insulator.)

For a coating to keep cryogens cold, TPS isn't great.  Once the system reaches equilibrium, the insulation doesn't really help you.  And high emissivity doesn't help a lot, because space isn't very hot compared to reentry, so that T⁴ term doesn't help very much.  But reflectivity is quite important.

What you really want is a coating with extremely high reflectivity and emissivity.  There's an experimental coating that's called Solar White, which has very good emissivity in infrared wavelengths and very good reflectivity in optical wavelengths.  As passive coatings go, this is probably the best you can do.  You can also obviously put up sun shades and active cryocooling, but indications are that the LSS will likely go with some version of white, at least as the minimum viable product.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/06/2022 08:06 pm
My near term SS cost number for a Tanker equivalent is the ~$50M number. In the 2030 time-frame that number does decrease in my cost modeling to a little lower at the $30-40M level. The cost model is derived from some observed data about the superstructure manufacturing costs based on manpower during the initial SS suborbital builds and tests. I still use this data since it was easier to come to reasonable value for this cost since the manpower totals possible for this work was much more evident than now. Since then a significant amount of the build has been automated but so has the complexity due to swapping wall thickness for use of stringers to lower mass.

This jibes pretty well with how I do this, which is that a naked Starship's costs (no crew) is somewhere between 25%-50% engines.  If you conservatively estimate the cost of Raptors at $2M/engine, that comes out to $24M-$48M.  I'd guess that SuperHeavy is closer to 75% engine cost, which would be $88M.

Quote
The bottom line here on the Depot and refueling is that having the depots do EDL may not be that important. Since the cost of a Depot that is likely to become obsolete in a couple of years for it to EDL so it can be scrapped is so low as in significantly less than $100M if even anywhere near that. And by not having the EDL equipment added it would save on cost for that while adding some additional equipment for long duration orbital operational stay. Such that a Depot that is used for 5 missions becomes ~$10M cost per mission that may have as many as 8 Tankers in one mission. As such that then would possibly make total prop cost for a mission to start at a value even less than $150M (good possibility that eventually a cost of $80-100M) to no higher than $400M when the Tankers have yet to be reused.

So evaluate design speculations by looking at cost of the various options. Since SpaceX will go for the ultimate mission lowest cost while still being able to have high number of missions usability rate/year.

Depots in LEO that can't do EDL are a no-brainer.  The question is whether to have a depot in cislunar and, if so, whether it's expendable.

Ideally, a plain ordinary lift tanker can top off at the LEO depot, go to NRHO, use an included "depot kit" to do RPOD with the LSS, refuel it, and go straight back to EDL.  Since the LSS itself has good prop storage properties, this is probably cheapest.

If the lift tanker can't carry a depot kit for whatever reason, or if it's not capable of carrying enough prop in one trip, then a depot in cislunar may make sense.  That way the depot kit is on the depot, so it can be an intermediary between the lift tanker and the LSS.

The only case where you'd have an expendable depot in cislunar is if there's some really compelling reason to carry the prop from LEO to cislunar in the depot itself (something you'd probably do on the first trip anyway, but then it'll need to be refilled), or if there's something particularly complex about the lift tanker-depot-LSS intermediation.  That seems unlikely to me.

Another big plus for having a reusable depot in cislunar:  The other App P winner can use it, as long as they don't do something stupid like use hydrolox.  This would seem to militate very strongly toward making the depot docking and transfer interfaces an open system, so SpaceX can sell (or even buy!) methalox to/from third parties.  Even if the Starship transport system were to go down for an extended period (due to an accident, for example), the depot would still be there, able to be refueled by third parties, until there was enough prop in it for an App P mission.  That's obviously more expensive than hauling the prop in a Starship lift tanker, but there are ways to jigger the price to defray third parties' risk if Starship isn't available.¹

______________
¹BTW, if Starship flights cost $50M/launch, the cost of prop delivered to a depot in NRHO is roughly $1M/t.  The cost of prop delivered by an FHE would be roughly $10M/t.  The cost of prop delivered on a VC-6 Heavy would be $17M/t.  SpaceX can assign risks to the possibility that Starship might not be available, forcing it to use an FHE, and set a price that defrays that risk while still providing a high likelihood of making a tidy profit.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 12/06/2022 08:52 pm
______________
¹BTW, if Starship flights cost $50M/launch, the cost of prop delivered to a depot in NRHO is roughly $1M/t.  The cost of prop delivered by an FHE would be roughly $10M/t.  The cost of prop delivered on a VC-6 Heavy would be $17M/t.  SpaceX can assign risks to the possibility that Starship might not be available, forcing it to use an FHE, and set a price that defrays that risk while still providing a high likelihood of making a tidy profit.

TFW throwing away Starships is still 10x cheaper than the competition.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Anguy on 12/06/2022 09:18 pm
The cost of the LSS is the crew habitat and other LSS crew related addons to the basic SS. A basic SS hardware may be easily as low as $50M. But an LSS hardware is likely to run at greater than $150M. NOTE here is that a depot and a BEO tanker would be at the close to a basic SS costs. Thus such to use an LSS at least 3 times will gain significant savings overall for surface missions of $50 to >$100M on each mission. Which can say a lot about a stripped down oneway cargo hardware to cost ~$75M for the cargo SS that delivers about 75 to 100t of cargo to the surface.

So what if just crew section is reusable and prop section is expendable? After each mission CS can detach on orbit and wait for brand new PS from Earth that will dock to it...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 12/06/2022 09:54 pm
I'm still wondering what sort of testing/refurbishing will be done either at Gateway or on the lunar surface. Or is that just not interesting if you aren't planning very frequent trips to the moon?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/06/2022 10:11 pm
I'm still wondering what sort of testing/refurbishing will be done either at Gateway or on the lunar surface. Or is that just not interesting if you aren't planning very frequent trips to the moon?
And I still wonder how to reprovision and load new cargo. I don't think the single IDSS port suffices.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 12/06/2022 10:32 pm
I'm still wondering what sort of testing/refurbishing will be done either at Gateway or on the lunar surface. Or is that just not interesting if you aren't planning very frequent trips to the moon?
And I still wonder how to reprovision and load new cargo. I don't think the single IDSS port suffices.
Use of 2 berthing ports inside the cargo bay of the LSS where each airlock has 1. To then berth a container to one of them or even 2 of them. Each container would be 2.5mX2.5mX3m in size ~18m^3 in volume. Sized so that they fit not only through the cargo hatch but also on the elevator. Such that a permanently landed LSS could be reprovisioned. Can also be done in orbit. And would travel with vehicle down and can be offloaded on the surface with the trash. Or loaded with the items to be returned to Earth and returned to orbit and then transferred as needed to eventually get it back to Earth.

Is that a solution you could work with?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/06/2022 10:55 pm
I'm still wondering what sort of testing/refurbishing will be done either at Gateway or on the lunar surface. Or is that just not interesting if you aren't planning very frequent trips to the moon?
And I still wonder how to reprovision and load new cargo. I don't think the single IDSS port suffices.
Use of 2 berthing ports inside the cargo bay of the LSS where each airlock has 1. To then berth a container to one of them or even 2 of them. Each container would be 2.5mX2.5mX3m in size ~18m^3 in volume. Sized so that they fit not only through the cargo hatch but also on the elevator. Such that a permanently landed LSS could be reprovisioned. Can also be done in orbit. And would travel with vehicle down and can be offloaded on the surface with the trash. Or loaded with the items to be returned to Earth and returned to orbit and then transferred as needed to eventually get it back to Earth.

Is that a solution you could work with?
I think you are headed in the right direction. That probably suffices for provisioning and pressurized cargo, but not for large unpressurized cargo like rovers. It also requires SpaceX to implement functionality on the Option B lander that is not an Option B (or appendix P) requirement, Which means NASA would need a different mission plan for Appendix P, Unless you think the Appendix P bidders will implement the same interface?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sebk on 12/06/2022 11:12 pm
1) It's an insulator:  it has a very low heat conductivity.  If you put a big heat pulse on one side of it, it takes a long time for the pulse to reach the other side.  But eventually, all things being in equilibrium, it will transfer just as much heat, at the same rate, once it reaches equilibrium.  Not great for long-term storage. 

This is incorrect. It wont transfer heat at the same rate. If you use higher thermal resistivity material, you'd get slower heat transfer.

Of course you'd still have boiloff unless you used active cooling. But you'd get much slower boiloff and you could use lower power active cooling to stop the boiloff.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 12/07/2022 12:02 am
I'm still wondering what sort of testing/refurbishing will be done either at Gateway or on the lunar surface. Or is that just not interesting if you aren't planning very frequent trips to the moon?
And I still wonder how to reprovision and load new cargo. I don't think the single IDSS port suffices.
Use of 2 berthing ports inside the cargo bay of the LSS where each airlock has 1. To then berth a container to one of them or even 2 of them. Each container would be 2.5mX2.5mX3m in size ~18m^3 in volume. Sized so that they fit not only through the cargo hatch but also on the elevator. Such that a permanently landed LSS could be reprovisioned. Can also be done in orbit. And would travel with vehicle down and can be offloaded on the surface with the trash. Or loaded with the items to be returned to Earth and returned to orbit and then transferred as needed to eventually get it back to Earth.

Is that a solution you could work with?
I think you are headed in the right direction. That probably suffices for provisioning and pressurized cargo, but not for large unpressurized cargo like rovers. It also requires SpaceX to implement functionality on the Option B lander that is not an Option B (or appendix P) requirement, Which means NASA would need a different mission plan for Appendix P, Unless you think the Appendix P bidders will implement the same interface?
For unpressurized cargo as long as it fits in a 2.5m W by 3m L by 2.5m T form factor that will fit on the elevator it would be transferred into the cargo bay by the Gateway CANDARM or other like arm in LEO. That size is about the max size that will fit on the current design size for the elevator.

For a rover that would be a large rover at ~4X the size of the Apollo rovers. Is that big enough? Various framing and pallets to be able to hold and move enmasse as a large group of smaller items in a very lightweight frame. Also strong enough to be handled and moved around and secured against acceleration.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/07/2022 02:31 am
I'm still wondering what sort of testing/refurbishing will be done either at Gateway or on the lunar surface. Or is that just not interesting if you aren't planning very frequent trips to the moon?
And I still wonder how to reprovision and load new cargo. I don't think the single IDSS port suffices.
Use of 2 berthing ports inside the cargo bay of the LSS where each airlock has 1. To then berth a container to one of them or even 2 of them. Each container would be 2.5mX2.5mX3m in size ~18m^3 in volume. Sized so that they fit not only through the cargo hatch but also on the elevator. Such that a permanently landed LSS could be reprovisioned. Can also be done in orbit. And would travel with vehicle down and can be offloaded on the surface with the trash. Or loaded with the items to be returned to Earth and returned to orbit and then transferred as needed to eventually get it back to Earth.

Is that a solution you could work with?
I think you are headed in the right direction. That probably suffices for provisioning and pressurized cargo, but not for large unpressurized cargo like rovers. It also requires SpaceX to implement functionality on the Option B lander that is not an Option B (or appendix P) requirement, Which means NASA would need a different mission plan for Appendix P, Unless you think the Appendix P bidders will implement the same interface?
For unpressurized cargo as long as it fits in a 2.5m W by 3m L by 2.5m T form factor that will fit on the elevator it would be transferred into the cargo bay by the Gateway CANDARM or other like arm in LEO. That size is about the max size that will fit on the current design size for the elevator.

For a rover that would be a large rover at ~4X the size of the Apollo rovers. Is that big enough? Various framing and pallets to be able to hold and move enmasse as a large group of smaller items in a very lightweight frame. Also strong enough to be handled and moved around and secured against acceleration.
But my basic problem is at the system level. Someone must deliver this stuff from Earth to NRHO. It may very well be cheaper to just send a new HLS and discard the old one.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tbellman on 12/07/2022 11:15 am
But my basic problem is at the system level. Someone must deliver this stuff from Earth to NRHO. It may very well be cheaper to just send a new HLS and discard the old one.

The real basic problem is that there is no heavy cargo that NASA wants to send on the HLS lander, and thus there is no problem with cargo transfer to be solved.

The Lunar Terrain Vehicle is to be delivered to the lunar surface by the LTV provider, and probably months in advance of the first crew to use it.  The LTV provider might contract SpaceX to do the delivery on a Starship, but that ain't gonna be the HLS lander.

The foundation surface habitat is currently just a vague hope in NASA's eyes, and if and when that gets built, it's not going to be delivered on the HLS lander, but on dedicated cargo landers.

The pressurised rover that is tentatively assumed to be built by JAXA is something that NASA could potentially want to be delivered together with the crew that's going to use it, but I believe it is much more likely that it will be sent on a dedicated cargo lander (and it may be that JAXA will be responsible for delivering it to the Moon as well, but that is very uncertain).

The nuclear reactor that NASA wants to test on the Moon—no way that's going to be sent on the same lander as the crew!  Dedicated cargo lander for certain.

Any other cargo NASA wants on the HLS lander can easily be transfered through the IDSS docking tunnel: tools, sample containers, food, clothes, spacesuits (with at least backpack detached, possibly arms and legs as well), water, air.  The IDSS port is what they use for reprovisioning ISS, after all.  (OK, they use the larger CBM as well, but the Cygnus implementation is not that much larger than the IDSS.)

Maybe NASA will come up with some large cargo that they want on the HLS lander, but that's going to be 2035 at the earliest.

Stop worrying about problems that no-one have!
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 12/07/2022 01:43 pm
I'd like to ask how effective are current TPS tiles as long time insulation for supply depot? Most proposals suggest stripping the depot bare, but wouldn't it be more effective to do the opposite a cover it all around with tiles? They could be dual used as simple omnidirectional insulation (so Earth radiation would be blocked too) instead of some deployable sunshield and the depot can also be used for Lunar missions and return to LEO by aerobraking, saving a lot of propellant.
Adding 20+tons to ship dry mass is not that big problem, if they can be used this way...

The other option is to cover it with white blankets ala Shuttle. Might not survive intense aerobraking (moderate aerobraking should be fine), but it should mass a lot less than tiles.

I presume R&D could share commonality with HLS. Blanket performance should be substantially superior to paint, which buys down thermal risks in contingency planning.



Edit: Shuttle FRSI blankets massed 1.6 kg/m2 (https://depts.washington.edu/matseed/mse_resources/Webpage/Space%20Shuttle%20Tiles/Space%20Shuttle%20Tiles.htm), so roughly 1/8th to 1/6th the mass of Starship tiles.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/07/2022 02:30 pm
But my basic problem is at the system level. Someone must deliver this stuff from Earth to NRHO. It may very well be cheaper to just send a new HLS and discard the old one.

The real basic problem is that there is no heavy cargo that NASA wants to send on the HLS lander, and thus there is no problem with cargo transfer to be solved.

Maybe NASA will come up with some large cargo that they want on the HLS lander, but that's going to be 2035 at the earliest.

Stop worrying about problems that no-one have!
I know that the current mission concepts depend on one-way cargo landers to deliver large cargo to the Moon. But those concepts predate the selection of Starship as the HLS. Option B is supposed to be about "sustainability": i.e., allowing for an increased cadence by reducing costs. Starship HLS has many tons of downmass capability, not the tiny amount in the Option B and Appendix P requirements. This capability could be used to reduce or eliminate the costs of one-way cargo missions. This total cost of delivery of the cargo on a single-mission HLS may very well be lower than the cost of reuse of HLS plus the cargo missions. Or not. It should be analyzed.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tbellman on 12/07/2022 04:44 pm
I know that the current mission concepts depend on one-way cargo landers to deliver large cargo to the Moon. But those concepts predate the selection of Starship as the HLS. Option B is supposed to be about "sustainability": i.e., allowing for an increased cadence by reducing costs. Starship HLS has many tons of downmass capability, not the tiny amount in the Option B and Appendix P requirements. This capability could be used to reduce or eliminate the costs of one-way cargo missions. This total cost of delivery of the cargo on a single-mission HLS may very well be lower than the cost of reuse of HLS plus the cargo missions. Or not. It should be analyzed.

It Does. Not. Matter. that the HLS Starship has tonnes of more cargo capability than required, because NASA does not have the ability to make use of it.  They do not have any such payloads now, they do not have any such payloads on their (vague) roadmaps, and they don't even have any such payloads in their wildest dreams.  Before NASA can make use of the cargo capabilities of the HLS Starship, they need to:

• Figure out what they want.  And despite the Artemis program being five years in, they haven't really figured out what kind of science they would want to do with even minimal cargo capacity, so how long do you think it will take them to figure out how to do even more (while still being limited to four crew once a year at best)?

• Secure financing for new payloads.  If not hard promises, at least reasonable indications that the US president, the OMB and the US congress won't just laugh when NASA tells them what they want.

• Procure such payloads.

• Wait for those payloads to actually be designed, manufactured and tested.

Before worrying about transferring large cargo to the HLS Starship, you need to pass at least the first point above: figuring out the relevant cargo.  And then you only need to worry about cargo transfer if those payloads actually benefit from being delivered on the HLS lander together with the crew, as opposed to being delivered on dedicated cargo flights directly from Earth (without passing NRHO and LOP-G).



I also think you are naïve in your interpretation of "sustainability".  For example, the  requirements document in the HLS RFP says this about HLS Sustainability:

Quote from: HLS-RQMT-001 chapter 3, HLS-Obj-005
Beyond the initial HLS missions, the HLS will support sustainable presence on the Moon by providing a regular cadence of reliable transportation services for humans and cargo.

(My bolding.)  Note the words "regular" and "reliable"; nothing about increased cadence, just regular.  Once every 12 months is regular.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/07/2022 05:12 pm
My near term SS cost number for a Tanker equivalent is the ~$50M number. In the 2030 time-frame that number does decrease in my cost modeling to a little lower at the $30-40M level. The cost model is derived from some observed data about the superstructure manufacturing costs based on manpower during the initial SS suborbital builds and tests. I still use this data since it was easier to come to reasonable value for this cost since the manpower totals possible for this work was much more evident than now. Since then a significant amount of the build has been automated but so has the complexity due to swapping wall thickness for use of stringers to lower mass.

This jibes pretty well with how I do this, which is that a naked Starship's costs (no crew) is somewhere between 25%-50% engines.  If you conservatively estimate the cost of Raptors at $2M/engine, that comes out to $24M-$48M.  I'd guess that SuperHeavy is closer to 75% engine cost, which would be $88M.

Quote
The bottom line here on the Depot and refueling is that having the depots do EDL may not be that important. Since the cost of a Depot that is likely to become obsolete in a couple of years for it to EDL so it can be scrapped is so low as in significantly less than $100M if even anywhere near that. And by not having the EDL equipment added it would save on cost for that while adding some additional equipment for long duration orbital operational stay. Such that a Depot that is used for 5 missions becomes ~$10M cost per mission that may have as many as 8 Tankers in one mission. As such that then would possibly make total prop cost for a mission to start at a value even less than $150M (good possibility that eventually a cost of $80-100M) to no higher than $400M when the Tankers have yet to be reused.

So evaluate design speculations by looking at cost of the various options. Since SpaceX will go for the ultimate mission lowest cost while still being able to have high number of missions usability rate/year.

Depots in LEO that can't do EDL are a no-brainer.  The question is whether to have a depot in cislunar and, if so, whether it's expendable.

Ideally, a plain ordinary lift tanker can top off at the LEO depot, go to NRHO, use an included "depot kit" to do RPOD with the LSS, refuel it, and go straight back to EDL.  Since the LSS itself has good prop storage properties, this is probably cheapest.

If the lift tanker can't carry a depot kit for whatever reason, or if it's not capable of carrying enough prop in one trip, then a depot in cislunar may make sense.  That way the depot kit is on the depot, so it can be an intermediary between the lift tanker and the LSS.

The only case where you'd have an expendable depot in cislunar is if there's some really compelling reason to carry the prop from LEO to cislunar in the depot itself (something you'd probably do on the first trip anyway, but then it'll need to be refilled), or if there's something particularly complex about the lift tanker-depot-LSS intermediation.  That seems unlikely to me.

Another big plus for having a reusable depot in cislunar:  The other App P winner can use it, as long as they don't do something stupid like use hydrolox.  This would seem to militate very strongly toward making the depot docking and transfer interfaces an open system, so SpaceX can sell (or even buy!) methalox to/from third parties.  Even if the Starship transport system were to go down for an extended period (due to an accident, for example), the depot would still be there, able to be refueled by third parties, until there was enough prop in it for an App P mission.  That's obviously more expensive than hauling the prop in a Starship lift tanker, but there are ways to jigger the price to defray third parties' risk if Starship isn't available.¹

______________
¹BTW, if Starship flights cost $50M/launch, the cost of prop delivered to a depot in NRHO is roughly $1M/t.  The cost of prop delivered by an FHE would be roughly $10M/t.  The cost of prop delivered on a VC-6 Heavy would be $17M/t.  SpaceX can assign risks to the possibility that Starship might not be available, forcing it to use an FHE, and set a price that defrays that risk while still providing a high likelihood of making a tidy profit.
A depot kit would be handy but tough (not impossible) to self deploy. Does it make dV sense to spot one at the gateway for visiting tankers? It would only launch once, the gateway arm could help in assembly and deployment, and it gives gateway a reason to exist.


BTW, I like your cost comparisons. Keep a lid on it or you might find multiple space programs fighting for a chance to wack you.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/07/2022 06:32 pm
I'm still wondering what sort of testing/refurbishing will be done either at Gateway or on the lunar surface. Or is that just not interesting if you aren't planning very frequent trips to the moon?
And I still wonder how to reprovision and load new cargo. I don't think the single IDSS port suffices.
Use of 2 berthing ports inside the cargo bay of the LSS where each airlock has 1. To then berth a container to one of them or even 2 of them. Each container would be 2.5mX2.5mX3m in size ~18m^3 in volume. Sized so that they fit not only through the cargo hatch but also on the elevator. Such that a permanently landed LSS could be reprovisioned. Can also be done in orbit. And would travel with vehicle down and can be offloaded on the surface with the trash. Or loaded with the items to be returned to Earth and returned to orbit and then transferred as needed to eventually get it back to Earth.

Is that a solution you could work with?
I think you are headed in the right direction. That probably suffices for provisioning and pressurized cargo, but not for large unpressurized cargo like rovers. It also requires SpaceX to implement functionality on the Option B lander that is not an Option B (or appendix P) requirement, Which means NASA would need a different mission plan for Appendix P, Unless you think the Appendix P bidders will implement the same interface?
For unpressurized cargo as long as it fits in a 2.5m W by 3m L by 2.5m T form factor that will fit on the elevator it would be transferred into the cargo bay by the Gateway CANDARM or other like arm in LEO. That size is about the max size that will fit on the current design size for the elevator.

For a rover that would be a large rover at ~4X the size of the Apollo rovers. Is that big enough? Various framing and pallets to be able to hold and move enmasse as a large group of smaller items in a very lightweight frame. Also strong enough to be handled and moved around and secured against acceleration.
Um, uh, am I missing something? Doesn't the big stuff get delivered direct to the lunar surface? When there's big bulky unpressurized cargo to deliver, a berthing hatch would be an unnecessary self imposed limit. Same for a 'standard' elevator. If another company needs access give them the keys and a set of instructions.


If the ship is a one way it's no big thing. If it comes back, aero brake and store it on orbit (probably near a depot) until another overdimentional load shows up on the manifest.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/07/2022 06:53 pm
But my basic problem is at the system level. Someone must deliver this stuff from Earth to NRHO. It may very well be cheaper to just send a new HLS and discard the old one.

The real basic problem is that there is no heavy cargo that NASA wants to send on the HLS lander, and thus there is no problem with cargo transfer to be solved.

Maybe NASA will come up with some large cargo that they want on the HLS lander, but that's going to be 2035 at the earliest.

Stop worrying about problems that no-one have!
I know that the current mission concepts depend on one-way cargo landers to deliver large cargo to the Moon. But those concepts predate the selection of Starship as the HLS. Option B is supposed to be about "sustainability": i.e., allowing for an increased cadence by reducing costs. Starship HLS has many tons of downmass capability, not the tiny amount in the Option B and Appendix P requirements. This capability could be used to reduce or eliminate the costs of one-way cargo missions. This total cost of delivery of the cargo on a single-mission HLS may very well be lower than the cost of reuse of HLS plus the cargo missions. Or not. It should be analyzed.
To use a trucking analogy, wind turbine blades are well within the weight limits of a standard flatbed but will not fit. They use a specialized trailer that will never find a backhaul. These trailers are deadheaded back for another blade, and that's part of the cost of doing business.


Like all analogies, it can only go so far. It may be more economical to just run one way with big items. One way or return, if it's too big to fit on an unmodified crewed ship, it needs an adequate cargo carrier.


If the bulk of cargo can be carried on crewed ships (an untested assumption) a specialized cargo ship will absolutely add costs. But, it adds value that the crewed ship can't. They can explore only so far on foot.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/07/2022 10:44 pm
It Does. Not. Matter. that the HLS Starship has tonnes of more cargo capability than required, because NASA does not have the ability to make use of it.  They do not have any such payloads now, they do not have any such payloads on their (vague) roadmaps, and they don't even have any such payloads in their wildest dreams.

I agree, with one caveat:  There's a class of "commodity" payloads that could be planned and slotted into any surface architecture very quickly.  Examples:

1) 150t of solar panels.
2) 150t of batteries, fuel cells, or APUs.
3) A big tank of LOX.
4) A big tank of LCH4.
5) A big tank of water.
6) 150t of the equivalent of portland cement for regolith.

But these are all payloads where their most important property is scale, and HLS can't scale its payloads large enough to matter.

The HLS LSS, at least with its current conops, is a lot more marginal than most people think.  An expendable CLPS or HDL LSS, on the other hand, is not.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tbellman on 12/08/2022 12:17 am
I agree, with one caveat:  There's a class of "commodity" payloads that could be planned and slotted into any surface architecture very quickly.  Examples:

1) 150t of solar panels.
2) 150t of batteries, fuel cells, or APUs.
3) A big tank of LOX.
4) A big tank of LCH4.
5) A big tank of water.
6) 150t of the equivalent of portland cement for regolith.

NASA could probably also buy ten thousand rubber ducks and send to the Moon.  But why?

Several megawatts worth of solar panels, or 150 tonnes of water, does not do you (or more specifically NASA) any good on their own.  For them to be useful, NASA would need other payloads that can use those resources.  And those are unlikely to pop up for NASA before mid next decade.

And even if or when they do need solar panels, batteries, water or cement in non-trivial amounts on the Moon, why would they send them to the Lunar Gateway and transfer them over to the HLS lander, instead of sending a cargo ship directly from Earth to the Moon?  Remember, my complaint was about the handwringing over how to transfer large cargo to the HLS Starship on orbit.

Quote from: TheRadicalModerate
But these are all payloads where their most important property is scale, and HLS can't scale its payloads large enough to matter.

The HLS LSS, at least with its current conops, is a lot more marginal than most people think.  An expendable CLPS or HDL LSS, on the other hand, is not.

And without the current conops, it would not be the HLS Starship...  For example, HLS Integrated Landers must be self-sufficient; they can't rely on pre-emplaced assets.  This means the landers must carry solar panels that work on the surface of the Moon, and work regardless of the lander's orientation to the sun.  That in turn effectively prohibits HLS Starship from carrying TPS tiles and thus from returning to Earth.

Similarly, the requirements to use SLS, Orion and LOP-G, forces the HLS Starship to pick up crew and supplies in NRHO instead of on Earth, and in practice forces it to be able to stay in NRHO for a year or two between each sortie down to the Moon.

A lunar architecture that really takes advantage of Starship, would send a cargo ship with solar panels or other power sources (e.g. a small nuclear reactor), to the intended landing spot, and deploy them robotically.  Then the crewed Starship would depart from Earth, and after landing on the Moon it would connect to the power sources already there.  (You want batteries on board for several hours of operation, and a small APU is probably also a good idea.)  Perhaps pre-emplaced cryo-coolers can also be connected to?  When the crew are tired of the Moon, they disconnect from the surface assets, and return all the way through EDL down to the surface of Earth.  Since that lunar Starship can carry TPS tiles and air control surfaces, it can EDL on Earth, where it can be inspected, repaired and updated, and new supplies, medium sized cargo and crew can be loaded for the next mission to the Moon.

Now I only need to find a way to connect this to refilling of propellant to make this comment not totally off-topic for this thread. :)

Right, the crew Starship (at least) would need to be refilled with propellant twice; once in LEO, and once in some later orbit; my favourite is during the trans-lunar coast.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/08/2022 04:35 am
NASA could probably also buy ten thousand rubber ducks and send to the Moon.  But why?

If you know that large commodity payloads are available, then your architectural planning gets a lot easier, the lead times on development are shorter, and the whole enterprise is cheaper.  It doesn't solve the problem of NASA figuring out what they actually want to do, but it does make it easier to do it once they have figured it out.

Quote
And even if or when they do need solar panels, batteries, water or cement in non-trivial amounts on the Moon, why would they send them to the Lunar Gateway and transfer them over to the HLS lander, instead of sending a cargo ship directly from Earth to the Moon?

That's what I meant when I said you'd send them via CLPS or HDL, which don't require a detour to NRHO.  (I'm assuming that there's nothing associated with HDL that requires it be staged from the Gateway, is there?)

Quote
Remember, my complaint was about the handwringing over how to transfer large cargo to the HLS Starship on orbit.

Yes, and I agreed with you.  The only thing I wanted to point out is that if you can get big commodity payloads in place (and you can, assuming that Starship works as advertised), then other hunks of mission planning and DDT&E become a lot easier.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 12/08/2022 02:08 pm
Is there any evidence that anyone at NASA is thinking about what to do with the capability to deliver large payloads to the moon?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 12/08/2022 06:13 pm
Is there any evidence that anyone at NASA is thinking about what to do with the capability to deliver large payloads to the moon?
Yes.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/08/2022 07:28 pm
Is there any evidence that anyone at NASA is thinking about what to do with the capability to deliver large payloads to the moon?
Even without hard evidence I find it hard to believe most people at NASA under 30 years old aren't dreamers. The bureaucracy probably tamps it down over time but it would never go away completely. Some of those people are salivating over the potential SX is offering.


Wait for SLS to die a natural fiscal death and the budget and program momentum will open up new opportunities. 


 





Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 12/08/2022 09:15 pm
Is there any evidence that anyone at NASA is thinking about what to do with the capability to deliver large payloads to the moon?
Yes.
Anything you'd care to share?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/08/2022 09:28 pm
Is there any evidence that anyone at NASA is thinking about what to do with the capability to deliver large payloads to the moon?
Yes.
Anything you'd care to share?

Here. (https://surveygizmoresponseuploads.s3.amazonaws.com/fileuploads/623127/5489366/111-381503be1c5764e533d2e1e923e21477_HeldmannJenniferL.pdf)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: georgegassaway on 12/08/2022 10:49 pm
At the risk of being called "off topic" regarding the recent discussions...

Has SpaceX indicated any testing whatsoever of On-Orbit refueling hardware or anything else involved with on-orbit refueling?

I do not mean 3rd party speculators, I mean SpaceX.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 12/08/2022 11:43 pm
Is there any evidence that anyone at NASA is thinking about what to do with the capability to deliver large payloads to the moon?
Yes.
Anything you'd care to share?

Here. (https://surveygizmoresponseuploads.s3.amazonaws.com/fileuploads/623127/5489366/111-381503be1c5764e533d2e1e923e21477_HeldmannJenniferL.pdf)
There's not much substance to that, though. I guess it's a proof of existence--there is at least one person at NASA (I think) who's thinking about what to do with Starship's capabilities, but there's nothing concrete in there. It's more like an appeal begging someone else to think about it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/09/2022 03:27 am
There's not much substance to that, though. I guess it's a proof of existence--there is at least one person at NASA (I think) who's thinking about what to do with Starship's capabilities, but there's nothing concrete in there. It's more like an appeal begging someone else to think about it.

Yeah, and it's a person at Ames.  Bah!  If you don't work at JSC or MSFC, you're not a steely-eyed missileperson! (I refuse to use smileys--except accidentally when I make lists with more than eight items--but if I did, there'd be one here...)

I think your assessment is largely correct.  I also suspect that the mission planners aren't quite ready to drink the Starship Kool-Aid.  Until they are, they're gonna keep using design paradigms that they know work in the absence of obscene amounts of throw weight.  Show 'em a successful refueling test, and there'll be a stampede.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 12/09/2022 10:43 am
Designing payloads that can only be transported via Starship also means locking out Option P landers. And given that the Option P lander contract even exists in the first place, we can be that NASA suggesting such payloads would result in congress similarly pitching a fit (and a completely coincidental lack of budget being allocated to development of such payloads). That basically limits you to payloads that are of low enough mass to be transported by all prospective landers but are modular enough that you could cram more onto a Starship lander for a reasonable and useful benefit.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/09/2022 10:23 pm
Designing payloads that can only be transported via Starship also means locking out Option P landers. And given that the Option P lander contract even exists in the first place, we can be that NASA suggesting such payloads would result in congress similarly pitching a fit (and a completely coincidental lack of budget being allocated to development of such payloads). That basically limits you to payloads that are of low enough mass to be transported by all prospective landers but are modular enough that you could cram more onto a Starship lander for a reasonable and useful benefit.

I'm not sure what the relationship is between Artemis, CLPS, and the HDL (the App. P-requested cargo version of whatever HLS gets picked).  But CLPS mission planners are free to pick whatever platform will get their job done.  And if Artemis planners can rely on a CLPS LSS, they can make a payload--especially a commodity payload--as big as they want.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/10/2022 06:47 pm
Designing payloads that can only be transported via Starship also means locking out Option P landers. And given that the Option P lander contract even exists in the first place, we can be that NASA suggesting such payloads would result in congress similarly pitching a fit (and a completely coincidental lack of budget being allocated to development of such payloads). That basically limits you to payloads that are of low enough mass to be transported by all prospective landers but are modular enough that you could cram more onto a Starship lander for a reasonable and useful benefit.

I'm not sure what the relationship is between Artemis, CLPS, and the HDL (the App. P-requested cargo version of whatever HLS gets picked).  But CLPS mission planners are free to pick whatever platform will get their job done.  And if Artemis planners can rely on a CLPS LSS, they can make a payload--especially a commodity payload--as big as they want.
Is there any reason that SX has to work through NASA to deliver a lunar science package? If SX were to announce intent to deliver science packages to the moon NASA would then be only one of many customers. IIUC NASA suggested that Artemus bidders have other uses for their hardware to make the program more sustainable.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 12/10/2022 08:52 pm
Designing payloads that can only be transported via Starship also means locking out Option P landers. And given that the Option P lander contract even exists in the first place, we can be that NASA suggesting such payloads would result in congress similarly pitching a fit (and a completely coincidental lack of budget being allocated to development of such payloads). That basically limits you to payloads that are of low enough mass to be transported by all prospective landers but are modular enough that you could cram more onto a Starship lander for a reasonable and useful benefit.
this isn’t true.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/10/2022 09:29 pm
Is there any reason that SX has to work through NASA to deliver a lunar science package? If SX were to announce intent to deliver science packages to the moon NASA would then be only one of many customers. IIUC NASA suggested that Artemus bidders have other uses for their hardware to make the program more sustainable.

They can obviously host any kind of non-NASA package, irrespective of whether it's for science, commercial development, an ITAR-acceptable foreign government, etc.  It just won't be part of CLPS or SLT.

"Commercial approach" is a sub-factor of the "management approach" evaluation in the HLS source selection methodology.  It's not a requirement.

In practice, I suspect this whole thing has the tail wagging the dog.  NASA's commitment to CLPS and SLT makes an LSS more attractive to non-NASA customers, because NASA's contracts make the financial and development risks associated with using the LSS smaller.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/12/2022 03:10 am
Something that's popped up on other threads but hasn't been discussed here (at least for a while) is the possibility of SpaceX selling prop to third parties in the near term.  An obvious potential third party would be the SLD winner.

If this were to happen, numerous folks would get their panties in a wad about dissimilar redundancy.  However, if SpaceX were to publish an open interface to the depot RPOD and prop transfer mechanisms, dissimilar redundancy is possible:

1) If Starship tankers are down for some reason, a medium-heavy launcher could fling stuff into TLI, 12-15t at a time.  The prop payloads so flung would need to be able to do RPODs and transfers to the depot.  (One such launcher is obviously Falcon Heavy.)

2) If the depot itself is down, that's more serious, but still possible to work out:  If the prop payload can emulate the depot-side interfaces, it can directly fuel the target vehicles.

So, three questions:

a) How much work would SpaceX have to do to drive an open depot RPOD/transfer standard?  Since they don't even have a proprietary version that works yet, it's obviously non-trivial. 

b) Could SpaceX short-circuit the usual technical/marketing deathmatch (anybody who's ever been involved with the IETF will understand the dynamic) by offering to accept prop into the depot and pass it through at the supplier's price, with some modest handling markup?

c) Could SpaceX do deals with third-party consumers (e.g. the other SLD service provider, a variety of heavy lunar robotic probes, or even interplanetary missions that didn't mind incurring a bit of extra cost to stage from NRHO) such that they could guarantee a price?  I would expect that price to be based on the likelihood that Starship tankers meet some particular cost, plus some premium that factored in the possibility that, if Starship tankers had a problem, SpaceX would have to provide the prop 15t at a time via Falcon Heavy.

My personal soapbox:  Dynetics should have just gone for it, negotiated a price for 60-100t of prop from a depot, and supersized their SLD architecture to use it.  If SpaceX guaranteed delivery via whatever platform works, it would have satisfied dissimilar redundancy for NASA.  If NASA believed it would work, they'd kill the BO/NT bid deader than a stone.  However, I'd put the odds of this having happened at less than 10%, since it would have leaked if they'd put that into their bid.

However, apparently Dynetics said in a recent presentation (https://forum.nasaspaceflight.com/index.php?topic=57816.msg2439944#msg2439944) that their bid would be able to refuel from Starship.  Whether it was from a Starship tanker or a payload launched on a Starship was unclear.

We usually assume that third-party prop sales, especially in cislunar, are further out than the timeframe for the discussion we have here.  I'm not so sure that's a good assumption.  This could be a case where SpaceX could make a very generous market to dramatically increase demand in the short- to medium-term.  Such a market would not only be potentially lucrative but it would also cover the cost of a lot of flight heritage for tankers and depots.

But it all hinges on being able to standardize the RPOD/transfer interfaces.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 12/12/2022 04:52 am
Something that's popped up on other threads but hasn't been discussed here (at least for a while) is the possibility of SpaceX selling prop to third parties in the near term.  An obvious potential third party would be the SLD winner.

If this were to happen, numerous folks would get their panties in a wad about dissimilar redundancy.  However, if SpaceX were to publish an open interface to the depot RPOD and prop transfer mechanisms, dissimilar redundancy is possible:

1) If Starship tankers are down for some reason, a medium-heavy launcher could fling stuff into TLI, 12-15t at a time.  The prop payloads so flung would need to be able to do RPODs and transfers to the depot.  (One such launcher is obviously Falcon Heavy.)

2) If the depot itself is down, that's more serious, but still possible to work out:  If the prop payload can emulate the depot-side interfaces, it can directly fuel the target vehicles.

So, three questions:

a) How much work would SpaceX have to do to drive an open depot RPOD/transfer standard?  Since they don't even have a proprietary version that works yet, it's obviously non-trivial. 

b) Could SpaceX short-circuit the usual technical/marketing deathmatch (anybody who's ever been involved with the IETF will understand the dynamic) by offering to accept prop into the depot and pass it through at the supplier's price, with some modest handling markup?

c) Could SpaceX do deals with third-party consumers (e.g. the other SLD service provider, a variety of heavy lunar robotic probes, or even interplanetary missions that didn't mind incurring a bit of extra cost to stage from NRHO) such that they could guarantee a price?  I would expect that price to be based on the likelihood that Starship tankers meet some particular cost, plus some premium that factored in the possibility that, if Starship tankers had a problem, SpaceX would have to provide the prop 15t at a time via Falcon Heavy.

My personal soapbox:  Dynetics should have just gone for it, negotiated a price for 60-100t of prop from a depot, and supersized their SLD architecture to use it.  If SpaceX guaranteed delivery via whatever platform works, it would have satisfied dissimilar redundancy for NASA.  If NASA believed it would work, they'd kill the BO/NT bid deader than a stone.  However, I'd put the odds of this having happened at less than 10%, since it would have leaked if they'd put that into their bid.

However, apparently Dynetics said in a recent presentation (https://forum.nasaspaceflight.com/index.php?topic=57816.msg2439944#msg2439944) that their bid would be able to refuel from Starship.  Whether it was from a Starship tanker or a payload launched on a Starship was unclear.

We usually assume that third-party prop sales, especially in cislunar, are further out than the timeframe for the discussion we have here.  I'm not so sure that's a good assumption.  This could be a case where SpaceX could make a very generous market to dramatically increase demand in the short- to medium-term.  Such a market would not only be potentially lucrative but it would also cover the cost of a lot of flight heritage for tankers and depots.

But it all hinges on being able to standardize the RPOD/transfer interfaces.

Perhaps, but it has taken Tesla at least 5 years to open up to non-Tesla cars.   There's both technical and competitive reasons to delay as long as possible.   

Technical meaning "we control both sides of the software and can thus iterate and break APIs".   Much harder to do if there's a fixed standard.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 12/12/2022 01:45 pm
The reason it took Tesla so long is also because of competitors being stubborn (or thinking more traditional solutions were more mature), like OneWeb. Like OneWeb, though, circumstances could change.

Until a few years ago, most EV Competitors to Tesla didn’t have production cars which could charge at the 100kW minimum that would be needed to make sense, let alone 200-250kW that Supercharger v3 can do.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/12/2022 06:00 pm
Something that's popped up on other threads but hasn't been discussed here (at least for a while) is the possibility of SpaceX selling prop to third parties in the near term.  An obvious potential third party would be the SLD winner.

If this were to happen, numerous folks would get their panties in a wad about dissimilar redundancy.  However, if SpaceX were to publish an open interface to the depot RPOD and prop transfer mechanisms, dissimilar redundancy is possible:

1) If Starship tankers are down for some reason, a medium-heavy launcher could fling stuff into TLI, 12-15t at a time.  The prop payloads so flung would need to be able to do RPODs and transfers to the depot.  (One such launcher is obviously Falcon Heavy.)

2) If the depot itself is down, that's more serious, but still possible to work out:  If the prop payload can emulate the depot-side interfaces, it can directly fuel the target vehicles.

So, three questions:

a) How much work would SpaceX have to do to drive an open depot RPOD/transfer standard?  Since they don't even have a proprietary version that works yet, it's obviously non-trivial. 

b) Could SpaceX short-circuit the usual technical/marketing deathmatch (anybody who's ever been involved with the IETF will understand the dynamic) by offering to accept prop into the depot and pass it through at the supplier's price, with some modest handling markup?

c) Could SpaceX do deals with third-party consumers (e.g. the other SLD service provider, a variety of heavy lunar robotic probes, or even interplanetary missions that didn't mind incurring a bit of extra cost to stage from NRHO) such that they could guarantee a price?  I would expect that price to be based on the likelihood that Starship tankers meet some particular cost, plus some premium that factored in the possibility that, if Starship tankers had a problem, SpaceX would have to provide the prop 15t at a time via Falcon Heavy.

My personal soapbox:  Dynetics should have just gone for it, negotiated a price for 60-100t of prop from a depot, and supersized their SLD architecture to use it.  If SpaceX guaranteed delivery via whatever platform works, it would have satisfied dissimilar redundancy for NASA.  If NASA believed it would work, they'd kill the BO/NT bid deader than a stone.  However, I'd put the odds of this having happened at less than 10%, since it would have leaked if they'd put that into their bid.

However, apparently Dynetics said in a recent presentation (https://forum.nasaspaceflight.com/index.php?topic=57816.msg2439944#msg2439944) that their bid would be able to refuel from Starship.  Whether it was from a Starship tanker or a payload launched on a Starship was unclear.

We usually assume that third-party prop sales, especially in cislunar, are further out than the timeframe for the discussion we have here.  I'm not so sure that's a good assumption.  This could be a case where SpaceX could make a very generous market to dramatically increase demand in the short- to medium-term.  Such a market would not only be potentially lucrative but it would also cover the cost of a lot of flight heritage for tankers and depots.

But it all hinges on being able to standardize the RPOD/transfer interfaces.
Here are counter questions. How much prop is at issue, and how many different vehicles would need refueling services? Short term.


Nothing out there is using props at the scale LSS needs and there would most probably be only one or two different ships besides SS/LSS needing refueling. Short term.


The idea of flinging small loads and doing direct transfers if the depot is down suggests morfing to a custom package for each customer and letting the depot do what it's designed to do - support SX ops. Customers will each have unique needs and limits that might not easily adapt to the SX system even if a more generic QD suite were available. Worse, even if a customer were using methalox that doesn't mean they won't need helium or some other fluid that the SX system is not designed to handle. Short term.


SX is so mass and volume rich that they can potentially do things almost as an afterthought that other rocketeers would see as a major design and operational goal. Launch a starship and deploy tankage and OTV from the cargo bay, designed to support one specific customer and mission. Learn lessons  that would point to a universal replenishment standard. Short term.


Long Medium term, build a more efficient universal long term solution.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/13/2022 08:40 pm
Here are counter questions. How much prop is at issue, and how many different vehicles would need refueling services? Short term.

If an SLD provider were to drink the Kool-Aid, quite a bit.  If not, then probably nothing for quite a while.

Quote
The idea of flinging small loads and doing direct transfers if the depot is down suggests morfing to a custom package for each customer and letting the depot do what it's designed to do - support SX ops. Customers will each have unique needs and limits that might not easily adapt to the SX system even if a more generic QD suite were available. Worse, even if a customer were using methalox that doesn't mean they won't need helium or some other fluid that the SX system is not designed to handle. Short term.

Even hydrolox customers could benefit quite handsomely from depot-provided LOX.

And yes, you are correct that this doesn't solve the consumables problem in general.  But non-propellant consumables are a fart in a windstorm compared to the stuff coming out the flamey end.  However, if the SLD provider actually got to the point where the only reason they needed to launch a servicing mission was to replenish those consumables, I'd think that would be an excellent reason for SpaceX to provide payload support in the ogive of a tanker.

I don't disagree that getting a standard that would work for everybody would be fairly difficult, but SpaceX will be the 500lb gorilla here.  And frankly, if two Starships can nuzzle up to each other to feed, then a gnat-sized vehicle could probably find a way to work things out.

All that said, after listening to the Dynetics preso, it sounds like they are indeed building their own tanker, which would be launchable as payload on a Starship.

Maybe this is a possible compromise:  build an "adapter" tank reference design, which can provide depot compatibility to transfer prop (in either direction), but ensure that the form factor is modest enough to be able to join up with any specific needs for the target vehicle.  For example, instead of Dynetics launching a tank of methalox on whatever launcher they choose, they'd launch an empty tank, which looked like their tank on one end, but looked like a Starship on the other.  That way, if the depot is available, they can launch a bunch of empty tanks on one launcher.  But if the depot is down, they can still launch full tanks.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 12/14/2022 12:00 pm
Designing payloads that can only be transported via Starship also means locking out Option P landers. And given that the Option P lander contract even exists in the first place, we can be that NASA suggesting such payloads would result in congress similarly pitching a fit (and a completely coincidental lack of budget being allocated to development of such payloads). That basically limits you to payloads that are of low enough mass to be transported by all prospective landers but are modular enough that you could cram more onto a Starship lander for a reasonable and useful benefit.

I'm not sure what the relationship is between Artemis, CLPS, and the HDL (the App. P-requested cargo version of whatever HLS gets picked).  But CLPS mission planners are free to pick whatever platform will get their job done.  And if Artemis planners can rely on a CLPS LSS, they can make a payload--especially a commodity payload--as big as they want.
Is there any reason that SX has to work through NASA to deliver a lunar science package? If SX were to announce intent to deliver science packages to the moon NASA would then be only one of many customers. IIUC NASA suggested that Artemus bidders have other uses for their hardware to make the program more sustainable.
No reason they have to, but developers and builders of science payloads are not set up to do things that way. Red Dragon had some tacit NASA backing to start with (as an EDL demonstrator) but very little takeup even at bargain basement prices for science payloads. Neither Dragon 1 nor Dragon 2 have flown a free-flying science mission, and none are on the books (despite takeup for the more expensive manned free-flyer missions). CLPS and VADR both involve NASA as the middleman soliciting payloads for commercial missions.
i.e. there's nothing to stop SpaceX from landing independent science missions, but there aren't any independent science missions looking for a ride.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/14/2022 10:10 pm
Designing payloads that can only be transported via Starship also means locking out Option P landers. And given that the Option P lander contract even exists in the first place, we can be that NASA suggesting such payloads would result in congress similarly pitching a fit (and a completely coincidental lack of budget being allocated to development of such payloads). That basically limits you to payloads that are of low enough mass to be transported by all prospective landers but are modular enough that you could cram more onto a Starship lander for a reasonable and useful benefit.

I'm not sure what the relationship is between Artemis, CLPS, and the HDL (the App. P-requested cargo version of whatever HLS gets picked).  But CLPS mission planners are free to pick whatever platform will get their job done.  And if Artemis planners can rely on a CLPS LSS, they can make a payload--especially a commodity payload--as big as they want.
Is there any reason that SX has to work through NASA to deliver a lunar science package? If SX were to announce intent to deliver science packages to the moon NASA would then be only one of many customers. IIUC NASA suggested that Artemus bidders have other uses for their hardware to make the program more sustainable.
No reason they have to, but developers and builders of science payloads are not set up to do things that way. Red Dragon had some tacit NASA backing to start with (as an EDL demonstrator) but very little takeup even at bargain basement prices for science payloads. Neither Dragon 1 nor Dragon 2 have flown a free-flying science mission, and none are on the books (despite takeup for the more expensive manned free-flyer missions). CLPS and VADR both involve NASA as the middleman soliciting payloads for commercial missions.
i.e. there's nothing to stop SpaceX from landing independent science missions, but there aren't any independent science missions looking for a ride.
There's a healthy market for cubesats. It's a big jump from cubesats to SS but not unimaginable.


First that sucker has to fly. Once it's demonstrated refueling and has done a demo lunar landing it will have demonstrated that it's the real thing.


Science packages are traditionally long lead time items so only he more adventurous that really understand the meaning of 'payload rich' will have shorter lead time packages ready for manifest within a couple of years of that first landing.


As "Quantity has a quality all its own" (J. Stalin, at Yalta) in reference to armies, so too does SS in deliverable mass. As a mind experiment, imagine the James Webb if designed to fly on SS. It could easily grow from 6.2t to 20t with simplified PV and sunshade and enough propellant to last until everybody looses interest in continued funding. It would have been less expensive to design and less expensive to build. How much is hard to say but IMO, significantly. Optical systems are inherently high dollar items. Other types of packages would show greater savings.


Between the private sector and universities looking for prestige programs it's not unthinkable that SX will run private science packages to the moon.


Is anybody working on cubesat mongo? Something based on maybe half meter cubes.



Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/15/2022 03:18 am
One other possible near- to medium-term market for third-party prop sales using a depot in LLO:  Point-to-point lunar missions with some fairly modest-sized hopper.  Hops on the Moon are expensive, because every mission has four big suborbital burns.  Rather than lowering prop for them onto the lunar surface, the difference between a long suborbital hop and going all the way to LLO on one of the hops isn't very much.  Stopping to refuel on the way back to base, the hopper would be ready to go for the next mission.

Making hydrolox out of lunar water would be great--eventually.  But the economics of doing lunar exploration this way will be far superior to capitalizing high-scale water mining and electrolysis.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 12/15/2022 03:33 am
One other possible near- to medium-term market for third-party prop sales using a depot in LLO:  Point-to-point lunar missions with some fairly modest-sized hopper.  Hops on the Moon are expensive, because every mission has four big suborbital burns.  Rather than lowering prop for them onto the lunar surface, the difference between a long suborbital hop and going all the way to LLO on one of the hops isn't very much.  Stopping to refuel on the way back to base, the hopper would be ready to go for the next mission.

That's good out-of-the-box thinking. LLO is not generally a stable orbital regime. Maybe a depot in polar LLO could facilitate transfers between the north and south lunar poles?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/15/2022 04:23 am
One other possible near- to medium-term market for third-party prop sales using a depot in LLO:  Point-to-point lunar missions with some fairly modest-sized hopper.  Hops on the Moon are expensive, because every mission has four big suborbital burns.  Rather than lowering prop for them onto the lunar surface, the difference between a long suborbital hop and going all the way to LLO on one of the hops isn't very much.  Stopping to refuel on the way back to base, the hopper would be ready to go for the next mission.

That's good out-of-the-box thinking. LLO is not generally a stable orbital regime. Maybe a depot in polar LLO could facilitate transfers between the north and south lunar poles?

It can facilitate transfers to anywhere, as long as you're only going to touch LLO once.  There's a frozen LLO very near 90º, so you can have the following conops (which isn't prop-optimal, but has good abort properties):

1) Assume hopper is just full enough to return to the depot from a lunar base at the poles.  When the depot will cross the longitude of the target mission (which it will do every two weeks), launch to the depot and completely refuel.

2) De-orbit from the depot to land at the target.

3) Do the mission.

4) When the mission is over or at any time there's an abort, hop (suborbitally) back to the base.  Repeat.

Even if you only want a 5t crew module, this isn't a small amount of prop.  But it's a small amount of prop compared to a mission from Earth.  So if you got a serious exploration program going, this would be a pretty good market.

Again, Dynetics as it probably will be for SLD would be almost perfect for this role.  It's a lot smaller than Starship, but for expeditions of a few days, an ALPACA that can refuel at a Starship depot, using Starship prop logistics, might easily outperform the LSS itself.

It'd be nice if you could point to something that the other provider had that's actually superior to LSS for some purpose.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/16/2022 12:49 am
One other possible near- to medium-term market for third-party prop sales using a depot in LLO:  Point-to-point lunar missions with some fairly modest-sized hopper.  Hops on the Moon are expensive, because every mission has four big suborbital burns.  Rather than lowering prop for them onto the lunar surface, the difference between a long suborbital hop and going all the way to LLO on one of the hops isn't very much.  Stopping to refuel on the way back to base, the hopper would be ready to go for the next mission.

That's good out-of-the-box thinking. LLO is not generally a stable orbital regime. Maybe a depot in polar LLO could facilitate transfers between the north and south lunar poles?

It can facilitate transfers to anywhere, as long as you're only going to touch LLO once.  There's a frozen LLO very near 90º, so you can have the following conops (which isn't prop-optimal, but has good abort properties):

1) Assume hopper is just full enough to return to the depot from a lunar base at the poles.  When the depot will cross the longitude of the target mission (which it will do every two weeks), launch to the depot and completely refuel.

2) De-orbit from the depot to land at the target.

3) Do the mission.

4) When the mission is over or at any time there's an abort, hop (suborbitally) back to the base.  Repeat.

Even if you only want a 5t crew module, this isn't a small amount of prop.  But it's a small amount of prop compared to a mission from Earth.  So if you got a serious exploration program going, this would be a pretty good market.

Again, Dynetics as it probably will be for SLD would be almost perfect for this role.  It's a lot smaller than Starship, but for expeditions of a few days, an ALPACA that can refuel at a Starship depot, using Starship prop logistics, might easily outperform the LSS itself.

It'd be nice if you could point to something that the other provider had that's actually superior to LSS for some purpose.
Let's step back from purely a technical discussion and add in a bit of timing.


AIUI, Dynetics would have a mission ~2 years following the LSS crewed mission. Is it reasonable to expect a depot used for the LSS mission to have not boiled off? Neither we nor SX has a realistic boil off model based the hardware to be used. If the depot is dry, tankage of some sort needs a ride up there.


Integrating the Dynetics and depot operations, while theoretically routine, will not be simple. It's two different companies and different cultures. It can no doubt be done but at the cost of X hours of effort by both companies.


By SS standards it's not all that much propellant. If Dynetics is planning on buying a ride on SS with their own mini tanker/depot why bother with a dicy SS depot at all? The work would be less as would the revenue, but the risk would also be less.


If it were firmly known that the LLO depot would be alive and kicking it would be the basis for collaboration discussions. The only way I can see Dynetics going for it is if they hit a snag with their own plans.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/16/2022 03:01 am
AIUI, Dynetics would have a mission ~2 years following the LSS crewed mission. Is it reasonable to expect a depot used for the LSS mission to have not boiled off? Neither we nor SX has a realistic boil off model based the hardware to be used. If the depot is dry, tankage of some sort needs a ride up there.

Even if the depot is (mostly) dry, as long as it remains functional, it can be replenished with cheap propellant.  Just send the prop along in a Starship tanker, which returns from NRHO to EDL when it has transferred the prop.

The better question is whether a dry depot is still functional or not.  If it has solar panels and enough propellant for station keeping and attitude control, I suspect it's functional.  It may be able to store enough supercritical GCH4 and GOX in COPVs to accomplish those goals.

Another question is whether something like this would tip the balance between passive and active cooling.  If an Option B or SLT LSS only needs a non-integral number of tankers' worth of prop, it's possible that it wouldn't require any additional tanker launches at all.  But then you'd really need to have a ZBO depot.

Quote
Integrating the Dynetics and depot operations, while theoretically routine, will not be simple. It's two different companies and different cultures. It can no doubt be done but at the cost of X hours of effort by both companies...

I agree that this is a non-trivial amount of work.  From what we've seen recently, it's not happening right away.  But it also seems to be the case that Dynetics' tanks/tankers will fly on Starship as payloads, so they're getting quite a bit of cost reduction from that if things go well.

Note that handling cryogenic payloads in any payload bay is a new feature.  Two ways to do this:

1) Just-in-time filling of tanks on the launch pad, with the needed venting.

2) Launch the tanks dry (albeit pressurized for rigidity), then fill them on-orbit, before deploying them.

That second method may actually be the real answer here:  If you have a dry tank that's hooked up, via a PAF, to the Starship propellant system, then you've managed to forgo the docking requirements that you'd need to use the depot, and piggybacking off of most of the tanker's prop transfer plumbing makes the adaptation about as easy as possible.

This does bring up a question, though:  Assuming that Dynetics wants to second-source their tanks on Vulcan, how would this work then?  Centaur is hydrolox (as is New Glenn's second stage), so flowing through its GSE doesn't work, either on the pad or in orbit.  How does Dynetics expect this to work?

Quote
If it were firmly known that the LLO depot would be alive and kicking it would be the basis for collaboration discussions. The only way I can see Dynetics going for it is if they hit a snag with their own plans.

Don't confuse a hopper architecture with an HLS architecture.  SLD/SLT is going to have to refuel in NRHO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 12/16/2022 03:52 pm
It's at least theoretically possible to fabricate a tank that will hold liquid oxygen or methane indefinitely. Not in LEO, but certainly at NRHO.
Quote
If a coating can be fabricated with about a 0.01 α/ϵ ratio and if this is placed on a sphere at uniform temperature, far from other heat sources, then the sphere will come to a steady state temperature of about 88 K, sufficient to passively store liquid oxygen.
Development of a thermal control coating optimized for cryogenic space applications (https://www.researchgate.net/scientific-contributions/Robert-C-Youngquist-6154658) (A Krenn et al 2022 IOP Conf. Ser.: Mater. Sci. Eng. 1240 012001)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 12/16/2022 03:54 pm
And, of course, we can do much better than a sphere. JWST is one such example. Or even just a long cylinder pointed toward the Sun.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 12/16/2022 03:56 pm
The biggest issue seems to be that the coating that can actually achieve that is a bit thick--5 mm or so. However, if the tanks are meant to be used over and over, perhaps it wouldn't be a big deal to send them up empty to start with.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/16/2022 05:57 pm
It's at least theoretically possible to fabricate a tank that will hold liquid oxygen or methane indefinitely. Not in LEO, but certainly at NRHO.
Quote
If a coating can be fabricated with about a 0.01 α/ϵ ratio and if this is placed on a sphere at uniform temperature, far from other heat sources, then the sphere will come to a steady state temperature of about 88 K, sufficient to passively store liquid oxygen.
Development of a thermal control coating optimized for cryogenic space applications (https://www.researchgate.net/scientific-contributions/Robert-C-Youngquist-6154658) (A Krenn et al 2022 IOP Conf. Ser.: Mater. Sci. Eng. 1240 012001)

The issue here is TRL.  They know what they need, but I'm not sure that's what they have yet.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 12/16/2022 07:18 pm
It's at least theoretically possible to fabricate a tank that will hold liquid oxygen or methane indefinitely. Not in LEO, but certainly at NRHO.
Quote
If a coating can be fabricated with about a 0.01 α/ϵ ratio and if this is placed on a sphere at uniform temperature, far from other heat sources, then the sphere will come to a steady state temperature of about 88 K, sufficient to passively store liquid oxygen.
Development of a thermal control coating optimized for cryogenic space applications (https://www.researchgate.net/scientific-contributions/Robert-C-Youngquist-6154658) (A Krenn et al 2022 IOP Conf. Ser.: Mater. Sci. Eng. 1240 012001)

The issue here is TRL.  They know what they need, but I'm not sure that's what they have yet.
JWST shows high TRL even for passive cooling to like LH2 temperatures far from Earth…

But SpaceX probably wants to keep far away from deployable MLI if they can. (But they shouldn’t be so afraid of it.)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/16/2022 09:32 pm
The issue here is TRL.  They know what they need, but I'm not sure that's what they have yet.
JWST shows high TRL even for passive cooling to like LH2 temperatures far from Earth…

But SpaceX probably wants to keep far away from deployable MLI if they can. (But they shouldn’t be so afraid of it.)

Webb only has to shade against one source in ES-L2.  There are two (at least) in NRHO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 12/17/2022 12:04 pm
It's at least theoretically possible to fabricate a tank that will hold liquid oxygen or methane indefinitely. Not in LEO, but certainly at NRHO.
Quote
If a coating can be fabricated with about a 0.01 α/ϵ ratio and if this is placed on a sphere at uniform temperature, far from other heat sources, then the sphere will come to a steady state temperature of about 88 K, sufficient to passively store liquid oxygen.
Development of a thermal control coating optimized for cryogenic space applications (https://www.researchgate.net/scientific-contributions/Robert-C-Youngquist-6154658) (A Krenn et al 2022 IOP Conf. Ser.: Mater. Sci. Eng. 1240 012001)

The issue here is TRL.  They know what they need, but I'm not sure that's what they have yet.
JWST shows high TRL even for passive cooling to like LH2 temperatures far from Earth…

But SpaceX probably wants to keep far away from deployable MLI if they can. (But they shouldn’t be so afraid of it.)

Not passive, or not "like" LH2 (<30 K, or <20 K at 1 bar). The only part of JWST that drops below the critical point of hydrogen requires active cooling.


https://old.reddit.com/r/jameswebb/comments/s4yfs8/jwst_temperature_v_time_cooldown_profile_from/
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 12/18/2022 12:36 am
AIUI, Dynetics would have a mission ~2 years following the LSS crewed mission. Is it reasonable to expect a depot used for the LSS mission to have not boiled off? Neither we nor SX has a realistic boil off model based the hardware to be used. If the depot is dry, tankage of some sort needs a ride up there.

Even if the depot is (mostly) dry, as long as it remains functional, it can be replenished with cheap propellant.  Just send the prop along in a Starship tanker, which returns from NRHO to EDL when it has transferred the prop.

The better question is whether a dry depot is still functional or not.  If it has solar panels and enough propellant for station keeping and attitude control, I suspect it's functional.  It may be able to store enough supercritical GCH4 and GOX in COPVs to accomplish those goals.

Another question is whether something like this would tip the balance between passive and active cooling.  If an Option B or SLT LSS only needs a non-integral number of tankers' worth of prop, it's possible that it wouldn't require any additional tanker launches at all.  But then you'd really need to have a ZBO depot.

Quote
Integrating the Dynetics and depot operations, while theoretically routine, will not be simple. It's two different companies and different cultures. It can no doubt be done but at the cost of X hours of effort by both companies...

I agree that this is a non-trivial amount of work.  From what we've seen recently, it's not happening right away.  But it also seems to be the case that Dynetics' tanks/tankers will fly on Starship as payloads, so they're getting quite a bit of cost reduction from that if things go well.

Note that handling cryogenic payloads in any payload bay is a new feature.  Two ways to do this:

1) Just-in-time filling of tanks on the launch pad, with the needed venting.

2) Launch the tanks dry (albeit pressurized for rigidity), then fill them on-orbit, before deploying them.

That second method may actually be the real answer here:  If you have a dry tank that's hooked up, via a PAF, to the Starship propellant system, then you've managed to forgo the docking requirements that you'd need to use the depot, and piggybacking off of most of the tanker's prop transfer plumbing makes the adaptation about as easy as possible.

This does bring up a question, though:  Assuming that Dynetics wants to second-source their tanks on Vulcan, how would this work then?  Centaur is hydrolox (as is New Glenn's second stage), so flowing through its GSE doesn't work, either on the pad or in orbit.  How does Dynetics expect this to work?

Quote
If it were firmly known that the LLO depot would be alive and kicking it would be the basis for collaboration discussions. The only way I can see Dynetics going for it is if they hit a snag with their own plans.

Don't confuse a hopper architecture with an HLS architecture.  SLD/SLT is going to have to refuel in NRHO.
I'm not sure I understand that last paragraph. I was alluding to a snag in Dynetics refueling plans.


The idea of the Dynetics tanker up in the cargo bay being loaded from the SS main tanks would be easily done using existing plumbing IF the SS depot has the QD up top. We have a difference of opinion on this.


Most all payloads need some custom TLC while mounted and waiting to launch. Keeping propellants topped off would be just another one of those things. A cryo cooler on board or a few holes in a fairing half for feeds and vents. Just have to be careful not to vent methane and O2 at the same time.


The Dynetics ship needs a ride to LEO. Is it a done deal that SS also takes it to LLO? The 1600/1550t tanks would make it a tight fit. Does anybody make an OTV that would physically fit in a standard cargo SS and make it from LEO to LLO?


Wouldn't that be a hoot? A Dynetics tanker launched on a SpaceX ship with a Neutron kicker from Rocket Labs. All done on a commercial basis.


Maybe we should wait and see how much propellant we're talking about. :(
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/18/2022 04:23 am
I'm not sure I understand that last paragraph. I was alluding to a snag in Dynetics refueling plans.

A Dynetics HLS will have to fuel in NRHO, not LLO.  LLO only makes sense if they're going to use it has a point-to-point hopper as well as a descent-ascent element.

Quote
The idea of the Dynetics tanker up in the cargo bay being loaded from the SS main tanks would be easily done using existing plumbing IF the SS depot has the QD up top. We have a difference of opinion on this.

Even without the QD, it should be easy to flow prop up through the PAF into the Dynetics tank(er)s--on a Starship.  But if they do that, how do they get a second source via Vulcan or New Glenn?  Neither of those second stages have methane on board--nor will their GSE on the pad above the first stage.

You can pre-load the LCH4, but it's a freakin' nightmare keeping it cold from the time it leaves payload processing to launch time.  Unless you're willing to travel with a big cryocooler and proper venting, you can't do it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: cAsE-sEnSlTivE on 12/19/2022 01:05 pm
what if they make the entire fuel depot rotate and transfer the fuel to a docking vehicle through a rotating seal? That way the vehicle that docks doesn't have to rotate
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: SkyRate on 12/19/2022 02:14 pm
what if they make the entire fuel depot rotate and transfer the fuel to a docking vehicle through a rotating seal? That way the vehicle that docks doesn't have to rotate
A long-term storage depot will want to reflect in the ecliptic plane and radiate out of it, and to be easy to dock to, so some kind of stabilization is clearly needed.
But I think that thrusters/momentum wheels have to be the lower-risk option compared to an entirely new (and mission-critical) solution.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Jim on 01/06/2023 05:53 pm

And yet if you think about it, the Shuttle orbiter was 'refueling' from the external tank on every launch.

Shuttle was rigidly mounted to the ET. The connector and structural mating was done on the ground. For the tank, it was a one night stand. Spending a hundred tech hours (a total WAG) on shuttle side inspection and refurb would not even have been a rounding error on shuttle turnaround. Other than that...

Yeah, and that was, if not insane, at least kinda dumb, as it turned out.  More importantly for this discussion:  It required a large team of humans in a shirtsleeve environment to align the bipods, connect them, tension them, and--last but hardly least--arm their pryrotechnic frangible bolts.  That's a silly--and likely impossible--design for in-space automated operations.

How the vehicles were mated doesn't negate the analogy.   It was the fact that the shuttle was sucking out the propellant with pumps.  In any refueling scenario, it is the tanker pushing fluid to the receiver.  The receiver is not taking low pressure fluid and pumping it up into high pressure tanks.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/06/2023 06:13 pm

And yet if you think about it, the Shuttle orbiter was 'refueling' from the external tank on every launch.

Shuttle was rigidly mounted to the ET. The connector and structural mating was done on the ground. For the tank, it was a one night stand. Spending a hundred tech hours (a total WAG) on shuttle side inspection and refurb would not even have been a rounding error on shuttle turnaround. Other than that...
Yeah, and that was, if not insane, at least kinda dumb, as it turned out.  More importantly for this discussion:  It required a large team of humans in a shirtsleeve environment to align the bipods, connect them, tension them, and--last but hardly least--arm their pryrotechnic frangible bolts.  That's a silly--and likely impossible--design for in-space automated operations.

How the vehicles were mated doesn't negate the analogy.   It was the fact that the shuttle was sucking out the propellant with pumps.  In any refueling scenario, it is the tanker pushing fluid to the receiver.  The receiver is not taking low pressure fluid and pumping it up into high pressure tanks.
As risky as it is to argue against Jim:
The Shuttle did not have any in-tank boost pumps. Liquid propellants were pushed out of the tanks and over the tank-to-vehicle connections and into the engine pumpheads by pressure from the ullage gas, not suction from the pump heads. Being autogenously pressurised that gas was supplied by the SSMEs, but could just as well have come from gas bottles inside the tanks (terrible idea though, no mass margins). Disconnect the tank-to-vehicle lines, and that ullage pressure would have equally well pushed propellants out the end of the lines into free space. Plumb those lines into a tank with a hole in the top, and they would have been pumped into that second tan (until it overflowed and propellants leaked out, or the first tank emptied and your ullage gas started bubbling across the link and making a mess, or you forgot to do all this under acceleration and just had a messy mixed flow shooting everywhere).

STS demonstrated that you can move cryogens over a pipe then disconnect the pipe and leave it permanently disconnected until remoted by hand on the ground. But Atlas also demonstrated that decades earlier with the LOX feeds to the booster engines. The hard part will be a cryogenic fluid coupler that can mate in orbit, demate, and remate again, multiple times, in orbit, whilst still allowing the receiving end of that connector to seal against flight pressures. That's been done for storable propellants, but not for cryogens, and cryogenic sealing and connectors are a notorious pain.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/06/2023 06:25 pm
STS demonstrated that you can move cryogens over a pipe then disconnect the pipe and leave it permanently disconnected until remoted by hand on the ground. But Atlas also demonstrated that decades earlier with the LOX feeds to the booster engines. The hard part will be a cryogenic fluid coupler that can mate in orbit, demate, and remate again, multiple times, in orbit, whilst still allowing the receiving end of that connector to seal against flight pressures. That's been done for storable propellants, but not for cryogens, and cryogenic sealing and connectors are a notorious pain.
What are the differences between doing this on the ground (as in the SS QD arm) and doing it in orbit, and do these differences make the job harder or easier? Clearly, the presence of a 1g gravity field on the ground is a big difference. What about others?

It's also true that SpaceX's current experience with its QD arm is limited and they can also do manual inspections and adjustments as needed, but I think they have already done some connect/reconnect tests.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Jim on 01/06/2023 06:29 pm

The Shuttle did not have any in-tank boost pumps. Liquid propellants were pushed out of the tanks and over the tank-to-vehicle connections and into the engine pumpheads by pressure from the ullage gas, not suction from the pump heads. Being autogenously pressurised that gas was supplied by the SSMEs, but could just as well have come from gas bottles inside the tanks (terrible idea though, no mass margins). Disconnect the tank-to-vehicle lines, and that ullage pressure would have equally well pushed propellants out the end of the lines into free space.

Yes, of course and head pressure.   The point was where the propellants were going and pumps were aiding in the pressure differential.  The supply tank doesn't have to keep increasing pressure to over come back pressure.

But you can go back even further with the connection: the GSE and umbilicals disconnects that filled the tanks have been used on almost all cryogenic vehicles.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 01/06/2023 06:59 pm

The Shuttle did not have any in-tank boost pumps. Liquid propellants were pushed out of the tanks and over the tank-to-vehicle connections and into the engine pumpheads by pressure from the ullage gas, not suction from the pump heads. Being autogenously pressurised that gas was supplied by the SSMEs, but could just as well have come from gas bottles inside the tanks (terrible idea though, no mass margins). Disconnect the tank-to-vehicle lines, and that ullage pressure would have equally well pushed propellants out the end of the lines into free space.

Yes, of course and head pressure.   The point was where the propellants were going and pumps were aiding in the pressure differential.  The supply tank doesn't have to keep increasing pressure to over come back pressure.

But you can go back even further with the connection: the GSE and umbilicals disconnects that filled the tanks have been used on almost all cryogenic vehicles.
The primary item about Starship is that it is not using cryo Hydrogen! That is the one that has shown too be the difficult and still finicky set of QDs. Note that while the LH QD on SLS kept having leaks the LOX ones were fairly good to go.

SpaceX so far has not had much leak problems with the QD plates and automated connect and disconnect and then connect over and over on the same vehicle. So the primary concern is the alignment prior to trying to dock and not damaging the QDs. Also not damaging the QDs during undock. All while on orbit.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 01/06/2023 09:27 pm
Also, lack of frost buildup in vacuum helps stuff from sticking quite as much. Vacuum also means leaks are not usually as much of a major fire/explosion hazard (as you’d need BOTH fuel AND oxidizer present at decent pressure or no ignition).

Vacuum helps insulate everything much better.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/06/2023 09:37 pm
Yes, of course and head pressure.   The point was where the propellants were going and pumps were aiding in the pressure differential.  The supply tank doesn't have to keep increasing pressure to over come back pressure.

But you can go back even further with the connection: the GSE and umbilicals disconnects that filled the tanks have been used on almost all cryogenic vehicles.

I'm not sure I'm getting your point here, but tanks can be filled by their internal pressure differential or filled at the same pressure by small pumps.  In either case, it will be the responsibility of the receiver to vent to keep pressure more-or-less static as prop flows in, and the sender to pressurize as prop flows out.

My favorite version of this is simply to connect the two ullage spaces together and use very small pumps (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2368599#msg2368599) to move the prop from one tank to the other.  There are no vented ullage gas losses this way.  In contrast, if you have an open system, where the receiver vents to the outside and the sender does something autogenous to keep the ullage space from becoming progressively lower pressure, you're wasting the receiver's ullage and doing something complicated and/or heavy on the sender.

Another reason to use pumps and no pressure differential:  If you're pressure-feeding the coupled tank system and you have an ullage oopsie where you uncover the high-pressure sump, the system will equalize more-or-less instantaneously, which will be a bummer.  It's a recoverable bummer, but one that can cost you a couple of tonnes of vented ullage gas.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 01/06/2023 10:19 pm
Yes, of course and head pressure.   The point was where the propellants were going and pumps were aiding in the pressure differential.  The supply tank doesn't have to keep increasing pressure to over come back pressure.

But you can go back even further with the connection: the GSE and umbilicals disconnects that filled the tanks have been used on almost all cryogenic vehicles.

I'm not sure I'm getting your point here, but tanks can be filled by their internal pressure differential or filled at the same pressure by small pumps.  In either case, it will be the responsibility of the receiver to vent to keep pressure more-or-less static as prop flows in, and the sender to pressurize as prop flows out.

My favorite version of this is simply to connect the two ullage spaces together and use very small pumps (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2368599#msg2368599) to move the prop from one tank to the other.  There are no vented ullage gas losses this way.  In contrast, if you have an open system, where the receiver vents to the outside and the sender does something autogenous to keep the ullage space from becoming progressively lower pressure, you're wasting the receiver's ullage and doing something complicated and/or heavy on the sender.

Another reason to use pumps and no pressure differential:  If you're pressure-feeding the coupled tank system and you have an ullage oopsie where you uncover the high-pressure sump, the system will equalize more-or-less instantaneously, which will be a bummer.  It's a recoverable bummer, but one that can cost you a couple of tonnes of vented ullage gas.
All of that vented ullage is actually needed to provide acceleration to maintain the prop settling. So it is not wasted.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: BT52 on 01/07/2023 07:59 pm
How much is minimum acceleration needed?

I assume minimum as surface tension no?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 01/07/2023 08:46 pm
How much is minimum acceleration needed?

I assume minimum as surface tension no?
SpaceX knows the answer. But not sure if we ever became privy to that value for SS. Without knowing that value with significant margins the SS on orbit would have some difficulties. Such as performing a circularizing burn at apogee after a drift in a preliminary elliptical transfer orbit.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/07/2023 09:18 pm
STS demonstrated that you can move cryogens over a pipe then disconnect the pipe and leave it permanently disconnected until remoted by hand on the ground. But Atlas also demonstrated that decades earlier with the LOX feeds to the booster engines. The hard part will be a cryogenic fluid coupler that can mate in orbit, demate, and remate again, multiple times, in orbit, whilst still allowing the receiving end of that connector to seal against flight pressures. That's been done for storable propellants, but not for cryogens, and cryogenic sealing and connectors are a notorious pain.
What are the differences between doing this on the ground (as in the SS QD arm) and doing it in orbit, and do these differences make the job harder or easier? Clearly, the presence of a 1g gravity field on the ground is a big difference. What about others?
Not just gravity, but atmospheric pressure, and atmospheric temperature. That affects seals and joints that can no longer assume a benign environment and occasional dips to cryogenic temperatures, but instead need to operate (flexing is the big ticket issue) in the long term at cryogenic temperatures (cold bridging coupled with vacuum insulation) and/or hot-soaking from insolation.
Not insurmountable challenges, but because it's not been attempted there will be a big pile of unknown-unknowns to discover first.
How much is minimum acceleration needed?

I assume minimum as surface tension no?
From long duration coast studies with Centaur, micro-G accelerations are sufficient settling thrust to ensure inlets remain covered ready or engine startup, so that is likely the range needed to keep those inlets covered for the much less vigorous propellant transfer process.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/07/2023 09:46 pm
How much is minimum acceleration needed?

I assume minimum as surface tension no?
SpaceX knows the answer. But not sure if we ever became privy to that value for SS. Without knowing that value with significant margins the SS on orbit would have some difficulties. Such as performing a circularizing burn at apogee after a drift in a preliminary elliptical transfer orbit.

I think ULA had done successful experiments with about 1mm/s² for long-term settling.

Note that settling for pre-ignition of engines is different from settling for prop transfer.  The former can be relatively high-thrust, because it's only active for a couple of seconds.  The latter has to be as low-thrust as possible, because it'll be active for minutes or even hours.  But your engines don't blow up if you accidentally uncover the sump.

All of that vented ullage is actually needed to provide acceleration to maintain the prop settling. So it is not wasted.

I just don't see how they're gonna use cold or warm gas thrusters for settling for prop transfer.  Isp is too low, which requires way too much mass flow to hit the proper thrust. 

Let's make the following assumptions:

Settling acceleration:  5mm/s² (I'm being conservative).
Prop transfer time: 2000s (a bit more than half an hour).
Cold (or maybe warm) gas Isp: 70s.
Methox or methalox combusting gas Isp: 300s
Depot dry mass: 90t.  Prop capacity: 1600t.
Tanker dry mass: 120t: Prop payload: 150t.

Then the lightest coupled system is an empty depot with a tanker transferring the first load of prop:  90t + 120t + 150t = 360t.  Using cold-to-warm ullage gas for the target acceleration requires 1800N, and the target transfer time would therefore require 5.2t of ullage gas.  Not great, but OK.

But the heaviest system is a depot receiving its last load of prop:  90t + 1450t + 120t + 150t = 1810t.  Now cold gas requires 9050N of thrust, and 26.4t of gas for the full transfer.  That's not only unacceptable from a prop efficiency standpoint, but there isn't enough cold ullage gas available in the first place.

In contrast, the heavy case with combusting methox or methalox would only require 6.2t of prop.

We've rehashed this elsewhere about a zillion times, but a set of redundant COPVs solve an awful lot of problems if they have the following duty cycle:

1) Vent the empty COPV to some very low pressure, into which it's easy to pump liquid prop.

2) Pump LCH4 or LOX into it until full.

3) Seal the vent and the pump inlet.

4) Heat it with electric heaters until it's supercritical at flight pressure (300-500bar).

5) Use it for whatever you need:
a) Bringing cold main or header tanks up to flight pressure.
b) Pushing liquids into pressure-fed methalox engines.
c) Driving monopropellant hot-gas thrusters.
d) Driving methox combustion thrusters.
e) spin-up gas for air- or space-restarts of the Starship Raptors.

(Edit: added #e above.)

6) When the COPV drops below minimum flight pressure, switch over to another redundant COPV and start the cycle over.

If you have three sets of tanks each for LCH4 and LOX, one can be active, one can be in redundant standby, and one can be refilling, heating, and pressurizing.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 01/08/2023 06:26 am

If you have three sets of tanks each for LCH4 and LOX, one can be active, one can be in redundant standby, and one can be refilling, heating, and pressurizing.

How much do the COPVs weigh dry?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/08/2023 09:18 am
Let's make the following assumptions:

Settling acceleration:  5mm/s² (I'm being conservative).
Prop transfer time: 2000s (a bit more than half an hour).
Cold (or maybe warm) gas Isp: 70s.
Methox or methalox combusting gas Isp: 300s
Depot dry mass: 90t.  Prop capacity: 1600t.
Tanker dry mass: 120t: Prop payload: 150t.

Then the lightest coupled system is an empty depot with a tanker transferring the first load of prop:  90t + 120t + 150t = 360t.  Using cold-to-warm ullage gas for the target acceleration requires 1800N, and the target transfer time would therefore require 5.2t of ullage gas.  Not great, but OK.

But the heaviest system is a depot receiving its last load of prop:  90t + 1450t + 120t + 150t = 1810t.  Now cold gas requires 9050N of thrust, and 26.4t of gas for the full transfer.  That's not only unacceptable from a prop efficiency standpoint, but there isn't enough cold ullage gas available in the first place.

In contrast, the heavy case with combusting methox or methalox would only require 6.2t of prop.

Thanks for this.

Possibly off topic for this thread, but I'm working on animating a spin-G setup aimed at eliminating the prop losses and orbit change due to settling acceleration.

If there are ~12 transfer cycles to fill the depot, the final heavy case you've mentioned there is 6.2t, and the first light case (I think) requires  ~1.2t. Is that implying that even with combusting methalox, that we're expending around 44t of prop just to fill a depot if we're relying on linear acceleration to settle the prop?

Also, does that imply that as long as less propellant than this (1.2t-6.2t/cycle) is used to perform a spin up/down cycle for each transfer, then any (reasonable) amount of extra mass dedicated to spin-G infrastructure on a depot will eventually pay itself off in prop mass savings over time?



Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/08/2023 08:12 pm
How much is minimum acceleration needed?

I assume minimum as surface tension no?
SpaceX knows the answer. But not sure if we ever became privy to that value for SS. Without knowing that value with significant margins the SS on orbit would have some difficulties. Such as performing a circularizing burn at apogee after a drift in a preliminary elliptical transfer orbit.

I think ULA had done successful experiments with about 1mm/s² for long-term settling.

Note that settling for pre-ignition of engines is different from settling for prop transfer.  The former can be relatively high-thrust, because it's only active for a couple of seconds.  The latter has to be as low-thrust as possible, because it'll be active for minutes or even hours.  But your engines don't blow up if you accidentally uncover the sump.

All of that vented ullage is actually needed to provide acceleration to maintain the prop settling. So it is not wasted.

I just don't see how they're gonna use cold or warm gas thrusters for settling for prop transfer.  Isp is too low, which requires way too much mass flow to hit the proper thrust. 

Let's make the following assumptions:

Settling acceleration:  5mm/s² (I'm being conservative).
Prop transfer time: 2000s (a bit more than half an hour).
Cold (or maybe warm) gas Isp: 70s.
Methox or methalox combusting gas Isp: 300s
Depot dry mass: 90t.  Prop capacity: 1600t.
Tanker dry mass: 120t: Prop payload: 150t.

Then the lightest coupled system is an empty depot with a tanker transferring the first load of prop:  90t + 120t + 150t = 360t.  Using cold-to-warm ullage gas for the target acceleration requires 1800N, and the target transfer time would therefore require 5.2t of ullage gas.  Not great, but OK.

But the heaviest system is a depot receiving its last load of prop:  90t + 1450t + 120t + 150t = 1810t.  Now cold gas requires 9050N of thrust, and 26.4t of gas for the full transfer.  That's not only unacceptable from a prop efficiency standpoint, but there isn't enough cold ullage gas available in the first place.

In contrast, the heavy case with combusting methox or methalox would only require 6.2t of prop.

We've rehashed this elsewhere about a zillion times, but a set of redundant COPVs solve an awful lot of problems if they have the following duty cycle:

1) Vent the empty COPV to some very low pressure, into which it's easy to pump liquid prop.

2) Pump LCH4 or LOX into it until full.

3) Seal the vent and the pump inlet.

4) Heat it with electric heaters until it's supercritical at flight pressure (300-500bar).

5) Use it for whatever you need:
a) Bringing cold main or header tanks up to flight pressure.
b) Pushing liquids into pressure-fed methalox engines.
c) Driving monopropellant hot-gas thrusters.
d) Driving methox combustion thrusters.

6) When the COPV drops below minimum flight pressure, switch over to another redundant COPV and start the cycle over.

If you have three sets of tanks each for LCH4 and LOX, one can be active, one can be in redundant standby, and one can be refilling, heating, and pressurizing.
There was a report of a small engine test at MacGregor but no details. IIRC, all that was visible on the video was the heat plume. Maybe an ullage thruster. Maybe a lunar landing engine. Maybe something else.


Small, mid range ISP engines do seem to be handy things to have in your tool box.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/08/2023 09:01 pm

If you have three sets of tanks each for LCH4 and LOX, one can be active, one can be in redundant standby, and one can be refilling, heating, and pressurizing.

How much do the COPVs weigh dry?

You'd need to do a budget for the max amount of supercritical gas you'd need in a short period (i.e., in a period shorter than it took to pump more liquid into the COPV and heat it to flight enthalpy), then pick a max pressure.

The Akin mass estimation (https://spacecraft.ssl.umd.edu/academics/791S16/791S16L08.MERsx.pdf) for COPVs (which I assume is implicitly at 300bar) is (115.3*V + 3)kg, where V is in m³. 

John had linked an old paper a while back (https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=49622.0;attach=1998495;sess=54584) that had mass proportional to pV, specifically:

M = C(⍴/Σ)pV

Where ⍴ is the average density of the COPV walls, Σ is the strength of the COPV material, and C is a coefficient that's 3 for the most straightforward designs.  This'll work as an estimate for whatever pressure and volume you choose.

Note that I left off an important application for high-pressure gas: 

e) spin-up gas for air- or space-restarts of the Starship Raptors.

Also, note that b) up-thread, the pressure-fed methalox thrusters, may also be the LSS landing thrusters, which could consume more prop, and therefore need bigger COPVs, than ullage or attitude control applications that used methalox thrusters.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/08/2023 10:48 pm
Possibly off topic for this thread, but I'm working on animating a spin-G setup aimed at eliminating the prop losses and orbit change due to settling acceleration.

If there are ~12 transfer cycles to fill the depot, the final heavy case you've mentioned there is 6.2t, and the first light case (I think) requires  ~1.2t. Is that implying that even with combusting methalox, that we're expending around 44t of prop just to fill a depot if we're relying on linear acceleration to settle the prop?

Also, does that imply that as long as less propellant than this (1.2t-6.2t/cycle) is used to perform a spin up/down cycle for each transfer, then any (reasonable) amount of extra mass dedicated to spin-G infrastructure on a depot will eventually pay itself off in prop mass savings over time?

First, I did a better estimate of the exact amount of ullage prop lost to maintain a particular acceleration.  Estimates for ullage accelerations of 5mm/s², 1.0mm/s², and 0.1mm/s², at Isps of both 70s and 300s, are attached.  The spreadsheet they came from is here (https://docs.google.com/spreadsheets/d/1MUkdFeYbZp8wlTXGPiapRMv4fTWCFodr8UrVPwyYXNI/edit?usp=sharing); if you want to fool with it, you can make a copy.  Note that the max prop in the depot jumps around a bit, because I rejiggered it so that it comes out to an integral number of tankers.

Performance is highly dependent on the actual ullage acceleration required.  At 5mm/s², it takes an extra tanker if you're using cold gas thrusters for ullage acceleration.  But if you want to use the same number of tankers and still get more than 1600t of prop in the depot with cold/warm gas, you need an ullage acceleration of 1.4m/s² or lower.

If you tend to fill a depot for a specific mission, then number of tankers is really the right metric, and overall prop efficiency isn't very important.  But if you're continuously filling and partially drawing down a depot, as you would in very high cadence ops, then the efficiency becomes more important.  Note also that increasing the power of the transfer pumps, and thereby shortening the transfer time, can also change things.  As usual, it's a pretty rich trade space.

So, to answer your question:  In high cadence, there's probably a number where rotational settling makes more sense than ullage thrust.  But it doesn't make any sense early on, when low cadence will almost certainly boil the depot dry between missions.  And even at high cadence, you'll have to show your work on why the extra complexity is worth it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/08/2023 10:57 pm
There was a report of a small engine test at MacGregor but no details. IIRC, all that was visible on the video was the heat plume. Maybe an ullage thruster. Maybe a lunar landing engine. Maybe something else.

Small, mid range ISP engines do seem to be handy things to have in your tool box.

If there was a heat plume, then it's definitely not a cold gas thruster.  Even a warm gas thruster wouldn't show much of a plume, because adiabatic expansion will drop the temperature a lot.

The problem with small methox or methalox combustion thrusters is that they need igniters, which are a lot more complex, and likely have bigger impulse bits, than you'd like for something doing attitude control, prox ops, or docking.  But my guess is that they'll be essential for lunar landing, or even small orbital maneuvers.

And (to stay nominally on-topic) they make a big difference if ullage accelerations have to be anything over about 1.5-2.0mm/s².
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 01/09/2023 05:09 am

If you have three sets of tanks each for LCH4 and LOX, one can be active, one can be in redundant standby, and one can be refilling, heating, and pressurizing.

How much do the COPVs weigh dry?

You'd need to do a budget for the max amount of supercritical gas you'd need in a short period (i.e., in a period shorter than it took to pump more liquid into the COPV and heat it to flight enthalpy), then pick a max pressure.

The Akin mass estimation (https://spacecraft.ssl.umd.edu/academics/791S16/791S16L08.MERsx.pdf) for COPVs (which I assume is implicitly at 300bar) is (115.3*V + 3)kg, where V is in m³. 

John had linked an old paper a while back (https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=49622.0;attach=1998495;sess=54584) that had mass proportional to pV, specifically:

M = C(⍴/Σ)pV

Where ⍴ is the average density of the COPV walls, Σ is the strength of the COPV material, and C is a coefficient that's 3 for the most straightforward designs.  This'll work as an estimate for whatever pressure and volume you choose.

Note that I left off an important application for high-pressure gas: 

e) spin-up gas for air- or space-restarts of the Starship Raptors.

Also, note that b) up-thread, the pressure-fed methalox thrusters, may also be the LSS landing thrusters, which could consume more prop, and therefore need bigger COPVs, than ullage or attitude control applications that used methalox thrusters.

The reason I'm asking is that the additional weight of the COPVs has to be put into the solution.

If you have 10t of COPvs and they only save 5t of propellant, it's a clear loss.   Even at 1:2 the COPvs have to be landed so he real calculation is the weight of the COPvs + the fuel it takes to land them.

It at 422kg/m3 it requires about 5 cubic meters to get 2t of propellant in each copV (for CH4 side).

that's 115*5 + 3 = 578kg per COPV

need 3 of these, that's 1.5t of COPVs, plus another 0.5t of propellant to land that, so 2.0t

From your spreadsheet at 1mm/s2 that eats up all the savings compared to ullage thrusters.

TL;DR - hot gas thrusters don't have any advantage over ullage thrusters as long as acceleration is low.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: volker2020 on 01/09/2023 06:08 am
So, to answer your question:  In high cadence, there's probably a number where rotational settling makes more sense than ullage thrust.  But it doesn't make any sense early on, when low cadence will almost certainly boil the depot dry between missions.  And even at high cadence, you'll have to show your work on why the extra complexity is worth it.

I am not sure about this. First, once the fuel is settled, you don't a lot of force to keep it that way, surface tension is your friend here. So I would love first to understand which force would distribute the fuel back, before I calculate which acceleration I need to counter it. Additionally, if you choose the orientation of the ships correctly, the process of pumping mass from one ship to another will create a little acceleration in itself, by moving the center of gravity. That could be even enough.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: livingjw on 01/09/2023 03:30 pm

If you have three sets of tanks each for LCH4 and LOX, one can be active, one can be in redundant standby, and one can be refilling, heating, and pressurizing.

How much do the COPVs weigh dry?

You'd need to do a budget for the max amount of supercritical gas you'd need in a short period (i.e., in a period shorter than it took to pump more liquid into the COPV and heat it to flight enthalpy), then pick a max pressure.

The Akin mass estimation (https://spacecraft.ssl.umd.edu/academics/791S16/791S16L08.MERsx.pdf) for COPVs (which I assume is implicitly at 300bar) is (115.3*V + 3)kg, where V is in m³. 

John had linked an old paper a while back (https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=49622.0;attach=1998495;sess=54584) that had mass proportional to pV, specifically:

M = C(⍴/Σ)pV

Where ⍴ is the average density of the COPV walls, Σ is the strength of the COPV material, and C is a coefficient that's 3 for the most straightforward designs.  This'll work as an estimate for whatever pressure and volume you choose.

Note that I left off an important application for high-pressure gas: 

e) spin-up gas for air- or space-restarts of the Starship Raptors.

Also, note that b) up-thread, the pressure-fed methalox thrusters, may also be the LSS landing thrusters, which could consume more prop, and therefore need bigger COPVs, than ullage or attitude control applications that used methalox thrusters.

Atkins COPV tank mass trend closely matches NASA CR-287 figure 20 if you assume a COPV design pressure of 410 bar (~6000psi) with a W/PV ~ 7x10^7 pci/psi. Converts to: Mass = .281 kg/bar/m^3. These seem like reasonable, though it would have been great to have Atkin's COPV pressures for completeness. Tanks are available at 6000psi and above.

The weight equation as a function of pressure(assumed ~6000psi) and volume correlated to Atkin's data is:   
       W = 7x10^-7 x P x V  lb/psi/in^3             or               M = .281 x P x V  kg/bar/m^3

For example:  for V = 1m^3 and P = 410 bar     M = 115.22 kg which matches Atkin's weight relation.

Edit: We need more large aerospace grade COPV data to nail this down.  Anyone know of sources?

John
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/09/2023 06:15 pm
There was a report of a small engine test at MacGregor but no details. IIRC, all that was visible on the video was the heat plume. Maybe an ullage thruster. Maybe a lunar landing engine. Maybe something else.

Small, mid range ISP engines do seem to be handy things to have in your tool box.

If there was a heat plume, then it's definitely not a cold gas thruster.  Even a warm gas thruster wouldn't show much of a plume, because adiabatic expansion will drop the temperature a lot.

The problem with small methox or methalox combustion thrusters is that they need igniters, which are a lot more complex, and likely have bigger impulse bits, than you'd like for something doing attitude control, prox ops, or docking.  But my guess is that they'll be essential for lunar landing, or even small orbital maneuvers.

And (to stay nominally on-topic) they make a big difference if ullage accelerations have to be anything over about 1.5-2.0mm/s².
Oh, it was definitely a combustion engine at MacGregor.


If the issue is propellant consumption (ISP really) then setting the bar at 1.5-2.9mm/s^2 doesn't come into it. If cold/warm gas thrusters use too much propellant at 1.0mm/s^2, a simple low power pressure fed gas/gas engine looks attractive.


A gas/gas engine somewhat addresses the ignition complexity. For a few bucks you can buy a dependable long lived piezo ignited cigar lighter. Might need just a tad of optimization...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: BT52 on 01/09/2023 08:28 pm
Hmm dint dude from Masten space joined SpaceX?

If yes then then they got small nozzle ISP solutions at their hands already. 
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/09/2023 09:13 pm
TL;DR - hot gas thrusters don't have any advantage over ullage thrusters as long as acceleration is low.

I still don't think this works.

I guessed on Isp for cold/warm gas.  It's obviously going to be dependent on main tank pressure if you're just using ullage gas.  I'd assume we're using methane, since it's lighter than molecular oxygen.

I'm way too stupid right now to figure out what the pressure in the depot and tanker methane tanks will be.  Let's assume that the tanker hasn't undergone condensation collapse yet, and the depot has.  I'm gonna guess that, together, they're at something like 3bar.  What does that yield for a reasonable nozzle exit velocity?  I'll bet it isn't Isp=70s.  Maybe... 20s?  (Please, somebody who has these pressure and Isp spreadsheets, chime in.  Let's assume that T=260K for the depot and T=300K for the tanker.)

Worst case, we have a tanker that's (1200t - 150t)/1200t = 87.5% empty, and a depot that's 150t/1614t = 9.3% empty.  Assuming O:F=3.6, that should have about 722m³ of methane ullage.  Using 3bar common tank pressure and T=280K common temperature (I'm guessing again, but things ought to mix pretty well when a warm tanker with lots of empty space and higher pressure gets connected to a cold depot with not very much space), then, by Boyle's Law:

300,000Pa * 722m³ = 8.314n280K
n = 93,044mol
startUllageMass = 1490kg

But wait!  There's more!  We can't use all of that mass, because there's some minimum pressure where you won't have choked flow anymore, and Isp will drop into the low single digits.  Again, too stupid to calculate, so let's guess it's at 1.5bar.  That will leave:

150,000Pa * 722m³ = 8.314n260K (I'm guess on the adiabatic temperature)
n = 50,100mol
endUllageMass = 802kg

usableUllageMass = 688kg

Using Isp=20s, xferTime=2000s, and depotProp=1600t, that limits your max acceleration to a=0.0058mm/s².

That's... pretty small...

NB:  There's a fair amount of hand-waving going on here.  If anybody can provide better numbers for Isp and before-and-after temperatures, please do!
_______________

On top of all that, there are other reasons why you need lots of high-pressure gas, with the biggest one being spin-up gas.  So, once you've gone to the expense of developing a renewable pressurization system for some bag of COPVs, why wouldn't you at least use warm gas being fed from supercritical reservoirs to drive non-combusting thrusters?

And if it turns out that the LSS really does need waist landing thrusters, which will almost certainly have to be combusting gas, why wouldn't you use those for ullage burns?  They'd have to throttle down pretty low, but that doesn't seem like a terrible problem for pressure-fed methox.

Mind you, I still think you'll need non-combusting warm gas thrusters for prox ops and docking, but if they feed off of the COPV high-pressure manifolds, they're easy, too.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/09/2023 09:16 pm
Hmm dint dude from Masten space joined SpaceX?

If yes then then they got small nozzle ISP solutions at their hands already.

Unless he and SpaceX are willing to get sued by Astrobotic for theft of intellectual property, they don't.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: BT52 on 01/09/2023 09:24 pm
Hmm dint dude from Masten space joined SpaceX?

If yes then then they got small nozzle ISP solutions at their hands already.

Unless he and SpaceX are willing to get sued by Astrobotic for theft of intellectual property, they don't.

Well they cant get sueed by knowledge of building small nozzle engine. Ahh i see. I was under impression they bought also some designs with him. But i guess not.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/09/2023 09:53 pm
Hmm dint dude from Masten space joined SpaceX?

If yes then then they got small nozzle ISP solutions at their hands already.

Unless he and SpaceX are willing to get sued by Astrobotic for theft of intellectual property, they don't.

Well they cant get sueed by knowledge of building small nozzle engine. Ahh i see. I was under impression they bought also some designs with him. But i guess not.

I don't think SpaceX has any lack of knowledge about how to build a small engine.  So unless this guy came with plans and test results (which absolutely would be theft of intellectual property), he doesn't help very much.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 01/09/2023 10:43 pm
I mentioned way back in the thread that SpaceX tested a small methane/GOx engine with the initial intent of using them in place of conventional nitrogen cold-gas RCS. Musk later said they're sticking with cold-gas for RCS. But there's no reason to assume they'd do an long ullage burn in a stupid way for no good reason.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 01/09/2023 10:56 pm
It turns out one might be able to do zero-g siphon tricks:

https://twitter.com/astro_pettit/status/719490255682842626?lang=en
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/10/2023 06:58 am
But there's no reason to assume they'd do an long ullage burn in a stupid way for no good reason.

Agreed.  My point was that it doesn't look like the mass of CH4 in ullage space will be adequate for the ullage impulse.  (I didn't do the computation including the O2 ullage, but things were bad enough that roughly doubling the mass wouldn't be adequate either.)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 01/10/2023 02:47 pm
But there's no reason to assume they'd do an long ullage burn in a stupid way for no good reason.

Agreed.  My point was that it doesn't look like the mass of CH4 in ullage space will be adequate for the ullage impulse.  (I didn't do the computation including the O2 ullage, but things were bad enough that roughly doubling the mass wouldn't be adequate either.)
If you add the O2 ullage the total available mass is a factor of 4X for mass over that of the CH4 alone.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/10/2023 04:42 pm
But there's no reason to assume they'd do an long ullage burn in a stupid way for no good reason.

Agreed.  My point was that it doesn't look like the mass of CH4 in ullage space will be adequate for the ullage impulse.  (I didn't do the computation including the O2 ullage, but things were bad enough that roughly doubling the mass wouldn't be adequate either.)
If you add the O2 ullage the total available mass is a factor of 4X for mass over that of the CH4 alone.

But not the impulse.

[Second Update:  My first cut at this was completely wrong, with massive confusion of extensive and intensive properties.  Just to save space (and, yeah, a little bit of embarrassment), I'm just deleting it.  Here's the new version:]

Since O2 has exactly double the molecular weight of CH4, O2 cold gas exit speed vs. CH4 speed should be sqrt(1/2)vch4 = 0.71 of the CH4 cold gas speed.

The LOX tank should have 29% more volume than the LCH4 tank, and the same pressure.  So the O2 ullage mass should be 2*1.29 = 2.6x that of the CH4 mass.

Therefore, the total impulse of the LOX ullage (mo2*vo2) should be 1.8x that of the CH4 ullage, and the total impulse of both tanks is 2.8x what I calculated up-thread, which means that the acceleration should go from 57 microgee to 159 microgee.  Note that I'm assuming that the cold gas thrusters can accept two different mass flows at different times during the ullage acceleration.  Shouldn't be a problem.

159 microgee might be close to a viable ullage acceleration, but it's still a really small acceleration.  Still seems to me that cold gas ullage thrust is pretty iffy.

Update #3:  From the 2008 Kutter-Zegler paper on turning Centaur into a prop depot (https://www.ulalaunch.com/docs/default-source/extended-duration/a-practical-affordable-cryogenic-propellant-depot-based-on-ula's-flight-experience.pdf):

Quote
Through improved understanding of low-g fluid behavior Centaur has reduced the standard parking orbit settling from 10-3 g to 10-4 g realizing a significant performance enhancement while maintaining adequate propellant control. In the quest for even more performance and longer mission duration, Centaur has demonstrated effective propellant control at accelerations down to 10-5 g...

Now, these numbers are for settling during storage, for the purpose of ensuring that they'd only vent gas instead of gas/liquid.  I can't think of a reason why prop transfer would have different properties, as long as vibration in the system is minimal.  If that's the case, then >10 microgee should be fine, and cold gas ullage acceleration will work.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 01/10/2023 10:41 pm
Can capillary action be used to transfer fuel?

https://m.facebook.com/Astro.Pettit/photos/a.241804595873073/998651346855057/
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/11/2023 07:04 pm
But there's no reason to assume they'd do an long ullage burn in a stupid way for no good reason.

Agreed.  My point was that it doesn't look like the mass of CH4 in ullage space will be adequate for the ullage impulse.  (I didn't do the computation including the O2 ullage, but things were bad enough that roughly doubling the mass wouldn't be adequate either.)
If you add the O2 ullage the total available mass is a factor of 4X for mass over that of the CH4 alone.

But not the impulse.

[Second Update:  My first cut at this was completely wrong, with massive confusion of extensive and intensive properties.  Just to save space (and, yeah, a little bit of embarrassment), I'm just deleting it.  Here's the new version:]

Since O2 has exactly double the molecular weight of CH4, O2 cold gas exit speed vs. CH4 speed should be sqrt(1/2)vch4 = 0.71 of the CH4 cold gas speed.

The LOX tank should have 29% more volume than the LCH4 tank, and the same pressure.  So the O2 ullage mass should be 2*1.29 = 2.6x that of the CH4 mass.

Therefore, the total impulse of the LOX ullage (mo2*vo2) should be 1.8x that of the CH4 ullage, and the total impulse of both tanks is 2.8x what I calculated up-thread, which means that the acceleration should go from 57 microgee to 159 microgee.  Note that I'm assuming that the cold gas thrusters can accept two different mass flows at different times during the ullage acceleration.  Shouldn't be a problem.

159 microgee might be close to a viable ullage acceleration, but it's still a really small acceleration.  Still seems to me that cold gas ullage thrust is pretty iffy.

Update #3:  From the 2008 Kutter-Zegler paper on turning Centaur into a prop depot (https://www.ulalaunch.com/docs/default-source/extended-duration/a-practical-affordable-cryogenic-propellant-depot-based-on-ula's-flight-experience.pdf):

Quote
Through improved understanding of low-g fluid behavior Centaur has reduced the standard parking orbit settling from 10-3 g to 10-4 g realizing a significant performance enhancement while maintaining adequate propellant control. In the quest for even more performance and longer mission duration, Centaur has demonstrated effective propellant control at accelerations down to 10-5 g...

Now, these numbers are for settling during storage, for the purpose of ensuring that they'd only vent gas instead of gas/liquid.  I can't think of a reason why prop transfer would have different properties, as long as vibration in the system is minimal.  If that's the case, then >10 microgee should be fine, and cold gas ullage acceleration will work.
ISTM that much depends on how transferred fluids are introduced into their new tank. For example, if the inlet is at the top of the tank the propellant will have any velocity imparted by the transfer force plus the acceleration of ullage thrust. When it hits the surface it splashes, and in the case of the LOX tank, the transfer tube forces it to hit asymmetrically and the ship picks up a wiggle.


If introduced at the bottom the transfer force will have to overcome whatever head pressure the already settled propellant will impose. On the first transfer, with little or no propellant already in the tank, where would be no pressure head. On the last transfer, there's a lot of head pressure.


Conclusion: bottom fill works best and whatever provides the transfer pressure needs to be variable.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/11/2023 08:55 pm
Conclusion: bottom fill works best and whatever provides the transfer pressure needs to be variable.
Is this news, though? I thought it was already established that there's a large pipe at the bottom of each tank to fill/drain the cryogenic fluid and a small one at the top for the ullage gas. (This is reflected in the QD adaptor.) The system already works on Earth, and pressure head ought to be less in space.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/11/2023 10:33 pm
Conclusion: bottom fill works best and whatever provides the transfer pressure needs to be variable.

This was why we stopped thinking very much about dorsal-to-dorsal, nose-to-tail configurations--or inline tail-to-tail configurations for that matter.

I still think that equalizing ullage pressures (by physically connecting the two ullage spaces) and using a very low power pump solves all the transfer pressure management problems.

One other problem, which is sort of related to your "don't splash" problem:  Prop will "geyser" into the receiving tank, even if it's pumped from the bottom.  At microgee accelerations, that geysering effect is going to cause all kinds of sloshing.  That may be OK, because uncovering the pump outlet on the receiving tank shouldn't be a problem.  But if you get blobs of prop bouncing off the receiving tank walls¹, they may mess up microgee ullage accelerations enough to cause sloshing on the sending side, which could uncover the inlet.

If you're using low-power pumps with equalized ullage pressures, the pump can probably be made robust enough to work with the sending side's inlet uncovered temporarily.  The pump will have to be able to re-prime itself with tiny head pressures, though.

Note that this is yet another reason not to use ullage pressure differences to transfer prop:  If you ever uncover the sending side inlet, the higher pressure ullage gas will instantly blow through the line, equalizing pressures, and then you have yourself a problem.  You can recover from this, but it'll require some combination of venting the receiving side and heating the sending side to restore the pressure differential.  It's wasteful, slow, and could potentially happen so often that the system wouldn't work at all.

_____________
¹Yet another problem related to geysering:  If you have unsettled blobs on the receiving side, they'll occasionally get sucked into the ullage pressure equalization line.  That could be made to be OK, but you may wind up pumping some of the prop through the system multiple times.  Presumably, as the tank gets full it will geyser less.  That's important, because otherwise you could have a case where the recirculated prop problem gets worse just before you're completely full.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/11/2023 11:01 pm

One other problem, which is sort of related to your "don't splash" problem:  Prop will "geyser" into the receiving tank, even if it's pumped from the bottom.  At microgee accelerations, that geysering effect is going to cause all kinds of sloshing.  That may be OK, because uncovering the pump outlet on the receiving tank shouldn't be a problem.  But if you get blobs of prop bouncing off the receiving tank walls¹, they may mess up microgee ullage accelerations enough to cause sloshing on the sending side, which could uncover the inlet.

If this is the only use for the fill pipe, they can put diverter such as a mushroom cap over the outlet.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/12/2023 03:15 am

One other problem, which is sort of related to your "don't splash" problem:  Prop will "geyser" into the receiving tank, even if it's pumped from the bottom.  At microgee accelerations, that geysering effect is going to cause all kinds of sloshing.  That may be OK, because uncovering the pump outlet on the receiving tank shouldn't be a problem.  But if you get blobs of prop bouncing off the receiving tank walls¹, they may mess up microgee ullage accelerations enough to cause sloshing on the sending side, which could uncover the inlet.

If this is the only use for the fill pipe, they can put diverter such as a mushroom cap over the outlet.

At the kind of accelerations and head pressures we're talking about, even turbulence after the diversion is likely to cause some slosh.  But it might be enough to prevent blobs of prop from slamming into stuff with enough force to cause slosh in the sending tank.  And a mushroom cap would probably help for the "nearly full" case, too.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/12/2023 07:28 am
If you intend to fill a depot for a specific mission, then number of tankers is really the right metric, and overall prop efficiency isn't very important.  But if you're continuously filling and partially drawing down a depot, as you would in very high cadence ops, then the efficiency becomes more important.  Note also that increasing the power of the transfer pumps, and thereby shortening the transfer time, can also change things.  As usual, it's a pretty rich trade space.

So, to answer your question:  In high cadence, there's probably a number where rotational settling makes more sense than ullage thrust.  But it doesn't make any sense early on, when low cadence will almost certainly boil the depot dry between missions.  And even at high cadence, you'll have to show your work on why the extra complexity is worth it.

I didn't respond to this earlier post, but seeing this, as well as your comment about 5mm/s2 being conservative and finding the Kutter research paper (https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf) had me all but give up on the spin-G prop transfer concept I was noodling with, because I think it probably only comes out on top of linear acceleration in scenarios where you actually need considerably more than 5mm/s2 (like say 1% of G = 100mm/s2).


One other problem, which is sort of related to your "don't splash" problem:  Prop will "geyser" into the receiving tank, even if it's pumped from the bottom.  At microgee accelerations, that geysering effect is going to cause all kinds of sloshing.  That may be OK, because uncovering the pump outlet on the receiving tank shouldn't be a problem.  But if you get blobs of prop bouncing off the receiving tank walls¹, they may mess up microgee ullage accelerations enough to cause sloshing on the sending side, which could uncover the inlet.

If this is the only use for the fill pipe, they can put diverter such as a mushroom cap over the outlet.

At the kind of accelerations and head pressures we're talking about, even turbulence after the diversion is likely to cause some slosh.  But it might be enough to prevent blobs of prop from slamming into stuff with enough force to cause slosh in the sending tank.  And a mushroom cap would probably help for the "nearly full" case, too.

...But then I see these posts, and I wonder if 5mm/s2 is really that conservative all things considered? The environment inside those prop tanks sounds like it will be quite dynamic if you want to pump propellant at any significant rate, and fluids are gonna fluid. In your earlier post you were assuming 150 ton of prop transferred in 2000s (75kg/s) which seems more than enough to disrupt surface tension if those 75kg only weigh ~40g, but still have the inertia of 75kg. You wouldn't want to get up there and find you can only pump at 15kg/s if you want to avoid chaotic effects. Playing with your spreadsheet to increase transfer time to 10,000s really does blow out the number of tanker trips.

My mental image was of one of those "incoming prop" geysers interacting with the receiver tank prop in such a way as to induce a vortex in the receiver tank, potentially torquing the entire structure in chaotic ways each time.  BoE says 1400 ton of prop swirling at 1 rpm is roughly equivalent to a reaction wheel with 1,800,000 Nms of angular momentum, so if that kind of induced swirling is a real problem, then that's a heck of a design challenge for your pump connectors. Presumably there are simple ways to baffle/disrupt these effects, but it could mean pumping slower or increasing tank mass. Maybe you have to pump slower anyway, and not expending a prohibitive amount of prop on linear G acceleration would be the problem a spin-G prop transfer scheme would be solving. I think I'll noodle a bit more.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: skyflyer81 on 01/12/2023 02:10 pm
The tolerances would have to be pretty tight for something like this to work, but it is completely different than standard pumping in gravity or spinning ships around
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/12/2023 03:45 pm
The tolerances would have to be pretty tight for something like this to work, but it is completely different than standard pumping in gravity or spinning ships around
The image looks cool, but I can't quite make sense of it. Can you elaborate a little? :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: skyflyer81 on 01/12/2023 03:53 pm
The tolerances would have to be pretty tight for something like this to work, but it is completely different than standard pumping in gravity or spinning ships around
The image looks cool, but I can't quite make sense of it. Can you elaborate a little? :-)

Just thinking that some baffles that start on one side of the interior of a tank could rotate around to the other side and "squeeze" out propellant
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/12/2023 04:38 pm
My mental image was of one of those "incoming prop" geysers interacting with the receiver tank prop in such a way as to induce a vortex in the receiver tank, potentially torquing the entire structure in chaotic ways each time.  BoE says 1400 ton of prop swirling at 1 rpm is roughly equivalent to a reaction wheel with 1,800,000 Nms of angular momentum, so if that kind of induced swirling is a real problem, then that's a heck of a design challenge for your pump connectors. Presumably there are simple ways to baffle/disrupt these effects, but it could mean pumping slower or increasing tank mass. Maybe you have to pump slower anyway, and not expending a prohibitive amount of prop on linear G acceleration would be the problem a spin-G prop transfer scheme would be solving. I think I'll noodle a bit more.
I just had an off-the-wall thought; what if you induced the swirling on purpose? If the propellant were swirling around the walls of the tank, at least you'd know where it was; instead of sloshing randomly, it'd be held to the walls by centrifugal force. Between that and the ullage acceleration, that ought to let the depot avoid sucking liquid out of the ullage pipe until the tank really was full.

To get the fluid to swirl, I'm visualizing replacing the nozzle at the bottom of the propellant tanks with something like a three-arm lawn sprinkler head (https://cdn.shopify.com/s/files/1/0595/2477/6095/products/ux_a17032300ux1505_ux_g03_1800x1800.jpg), except that it wouldn't spin.

This would cause the connected ships to counterrotate, of course, until the pumping stopped and the liquids settled down. To reduce this effect, you could have the methane tank swirl in the opposite direction from the oxygen tank, but, since most of the mass is oxygen, this would only help a little. You could also have the source tanks swirl in the opposite direction from the depot tanks (e.g., via matching sprinkler heads at the bottoms). There would still be net torque, since the center of mass wouldn't usually be symmetrical (the depot usually will have a lot more mass, if only because it's usually partly full), but at least it would reduce the amount of it. As a possible plus, when fueling is almost complete, the joined vehicles will be rotating almost on the depot's main axis (until the propellants settle down), so the last dregs of prop in the tanker should settle to the side and bottom. If one of the nozzles ended there, it would make it easier to suck out the last few drops. (Assuming it were okay for the other two to suck ullage gas.)

Apologies if this is unclear. Extra apologies if someone else already thought of this and I just see it!
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/12/2023 07:21 pm
I just had an off-the-wall thought...

Well, it's really an on-the-wall thought...

As a meta-observation, note that what we're doing here is trading some combination of ullage propellant consumption and thruster complexity against some combination of transfer time (a proxy for flow rate and the momentum that comes with it) and complexity (in terms of funky inlet and outlet schemes and attitude stability). 

Seems to me that the KISS solution is to start with higher ullage accelerations and make the flow patterns be either axial and laminar or, alternatively, so turbid that the pool winds up with very few surface tensions breaks that could cause blobs on the receiving side and/or inlet uncovering on the sending side.

Things we haven't yet tried to compute and/or don't know:

1) What are the actual dynamics of the pool around the inlets and outlets as a function of flow rate? 
1a) Would knowing the maximum forward velocity that a blob could leave the pool without hitting the top of tank be a useful figure of merit?
1b) Does anybody know how to compute what's needed to break surface tension as a function of laminar flow speed and surface area of the flow?  (This doesn't help very much if the flow is highly turbulent, but might be interesting if it stays mostly laminar coming out of the outlet.)

2) Are there going to be combusting gas thrusters available?

3) A random one:  How much hydrostatic head does a pump need to function reliably?  I'm tempted to say that the answer to this is "zero", but I'm pump-illiterate.  Based on some calculations, zero is a pretty good approximation to the head at the bottom of both tanks, even at 5mm/s².

Given that we have some evidence that SpaceX is doing something new with combusting gas thrusters, and that combusting gas, if it's available in low enough thrusts or impulse bits, solves the problem, that seems the most likely solution to me.  But I'd guess that flow dynamics are a big unknown, and likely one that can only really be known via on-orbit experiments.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/12/2023 11:43 pm
Well, it's really an on-the-wall thought...
It's so off-the-wall it's actually on-the-wall! :-)

Given that we have some evidence that SpaceX is doing something new with combusting gas thrusters, and that combusting gas, if it's available in low enough thrusts or impulse bits, solves the problem, that seems the most likely solution to me.  But I'd guess that flow dynamics are a big unknown, and likely one that can only really be known via on-orbit experiments.
That leads me to wonder why SpaceX isn't already trying to do experiments to settle these questions. They put something in orbit every week. You'd think they could find room for a few payloads that could answer these questions. It seems to make a lot more sense than learning them using entire Starship upper stages.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 01/13/2023 12:14 am
It seems to make a lot more sense than learning them using entire Starship upper stages.

They need to launch those Starships anyway in order to test everything else.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 01/13/2023 12:37 am
The tolerances would have to be pretty tight for something like this to work, but it is completely different than standard pumping in gravity or spinning ships around

Surely this could be simplified from two moveable baffles down to one moveable baffle and one fixed baffle.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Thrustpuzzle on 01/13/2023 01:15 am
I just had an off-the-wall thought; what if you induced the swirling on purpose? If the propellant were swirling around the walls of the tank, at least you'd know where it was; instead of sloshing randomly, it'd be held to the walls by centrifugal force. Between that and the ullage acceleration, that ought to let the depot avoid sucking liquid out of the ullage pipe until the tank really was full.
Centrifugal seperation can work in zero G, (https://ntrs.nasa.gov/citations/19910000545) but has losses due to wall friction so it's really useful if the walls are smooth to reduce friction and turbulence.  There's a nifty benefit in that the pumps (and power generation) can all be on the depot and not in tankers, so you don't have to transport that machinery mass every launch, just once for the depot.

My feeling is that this could work but might have high energy expense needed to accelerate and maintain that momentum. Of course maybe that's more acceptable compared that to tons of ullage gas used to give microthrust setting. It would be fun to determine the mechanical stability of two coupled spacecraft with huge liquid gyroscope flywheels inside each.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: skyflyer81 on 01/13/2023 01:36 pm
The tolerances would have to be pretty tight for something like this to work, but it is completely different than standard pumping in gravity or spinning ships around

Surely this could be simplified from two moveable baffles down to one moveable baffle and one fixed baffle.

lol, so obvious. definitely would simplify it a ton.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/13/2023 01:52 pm
Centrifugal seperation can work in zero G, (https://ntrs.nasa.gov/citations/19910000545) but has losses due to wall friction so it's really useful if the walls are smooth to reduce friction and turbulence.  There's a nifty benefit in that the pumps (and power generation) can all be on the depot and not in tankers, so you don't have to transport that machinery mass every launch, just once for the depot.

My feeling is that this could work but might have high energy expense needed to accelerate and maintain that momentum. Of course maybe that's more acceptable compared that to tons of ullage gas used to give microthrust setting. It would be fun to determine the mechanical stability of two coupled spacecraft with huge liquid gyroscope flywheels inside each.
I couldn't find the actual paper(s) from that link, but it gave me search terms that turned up this publicly available paper: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5884133/

I've only skimmed it so far, but I think it contains the equations we'd want to actually calculate whether this will work or not. I did notice that the paper is only concerned with very small systems, but it does work out a general formula that includes both swirling and microgravity.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: frith01 on 01/13/2023 03:23 pm
The tolerances would have to be pretty tight for something like this to work, but it is completely different than standard pumping in gravity or spinning ships around
The image looks cool, but I can't quite make sense of it. Can you elaborate a little? :-)

Just thinking that some baffles that start on one side of the interior of a tank could rotate around to the other side and "squeeze" out propellant

Interesting idea for a zero-G door-hinge type fluid pusher, but this would only work if the tanks were symmetrical, and didn't have anti-slosh baffles, pipes, and other stuff routing through the tanks.   Cryogenic temperatures would prevent the use of any flexible surface from being used on the outer edges.

Could be used in cargo area for fuel Blivet  type usage though.

https://www.atlinc.com/images/atl_dropdrum-group.png (https://www.atlinc.com/images/atl_dropdrum-group.png)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 01/13/2023 11:52 pm
The tolerances would have to be pretty tight for something like this to work, but it is completely different than standard pumping in gravity or spinning ships around

Surely this could be simplified from two moveable baffles down to one moveable baffle and one fixed baffle.

lol, so obvious. definitely would simplify it a ton.

Yeah that's basically all I'm good for. Call me Captain Apparent!  ;)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/14/2023 10:52 am
Even ignoring all the cryogenic sliding seal issues, plunger mass, plunger stiffness and jamming, etc, issues: radial plungers would mean that halfway through a burn, your vehicle would have 85% of its mass one one side and 5% on the other.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/14/2023 06:54 pm
Conclusion: bottom fill works best and whatever provides the transfer pressure needs to be variable.

This was why we stopped thinking very much about dorsal-to-dorsal, nose-to-tail configurations--or inline tail-to-tail configurations for that matter.

I still think that equalizing ullage pressures (by physically connecting the two ullage spaces) and using a very low power pump solves all the transfer pressure management problems.

One other problem, which is sort of related to your "don't splash" problem:  Prop will "geyser" into the receiving tank, even if it's pumped from the bottom.  At microgee accelerations, that geysering effect is going to cause all kinds of sloshing.  That may be OK, because uncovering the pump outlet on the receiving tank shouldn't be a problem.  But if you get blobs of prop bouncing off the receiving tank walls¹, they may mess up microgee ullage accelerations enough to cause sloshing on the sending side, which could uncover the inlet.

If you're using low-power pumps with equalized ullage pressures, the pump can probably be made robust enough to work with the sending side's inlet uncovered temporarily.  The pump will have to be able to re-prime itself with tiny head pressures, though.

Note that this is yet another reason not to use ullage pressure differences to transfer prop:  If you ever uncover the sending side inlet, the higher pressure ullage gas will instantly blow through the line, equalizing pressures, and then you have yourself a problem.  You can recover from this, but it'll require some combination of venting the receiving side and heating the sending side to restore the pressure differential.  It's wasteful, slow, and could potentially happen so often that the system wouldn't work at all.

_____________
¹Yet another problem related to geysering:  If you have unsettled blobs on the receiving side, they'll occasionally get sucked into the ullage pressure equalization line.  That could be made to be OK, but you may wind up pumping some of the prop through the system multiple times.  Presumably, as the tank gets full it will geyser less.  That's important, because otherwise you could have a case where the recirculated prop problem gets worse just before you're completely full.
Yup. Geysering (good word) is why variable pump pressure seem a good idea. Pumping into a dry tank would face minimum head with geysering at max risk. As the head increases the pump pressure would need to be higher, but not so high as to geyser. Any hiccup that allows the sending outlet to ingest gas and send it on to the receiver just kicks off a positive feedback loop.


You've got me three quarters convinced that equalizing ullage pressure is the way to go. There is something about not equalizing ullage and venting from the receiver that's attractive. ISTM that vented ullage thrust would take too long for initial settling but might work well to maintain a settled state. But then, the lower the g (minimizing consumption) the easier things splash.


Counter to this is that during later transfers, as the sending ship is going empty and the receiving ship is filling up, the head pressure is high and pressure fed transfer would slow and maybe even stall out. Down the road the time it takes for transfer will become important. That probably means active pumping and if ya gotta pump, embrace KISS and don't screw around with two ways to power the transfer.


Once their testing advances to transfer between two ships, pressure fed would be the standard SpaceX 'just good enough' method but maybe they'll have a pump in parallel and move propellants back and forth and try it every which way.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/14/2023 07:04 pm
Pumping into a dry tank would face minimum head with geysering at max risk. As the head increases the pump pressure would need to be higher, but not so high as to geyser.
Is the head pressure going to increase? If the ullage acceleration is just a few tens of micro-g, it doesn't seem like there'll be very much holding the propellants in place. That's why I was thinking about deliberately making them swirl. If you know they're going to slow around, at least be sure they're sloshing the way you expect them to.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/14/2023 07:16 pm
If you intend to fill a depot for a specific mission, then number of tankers is really the right metric, and overall prop efficiency isn't very important.  But if you're continuously filling and partially drawing down a depot, as you would in very high cadence ops, then the efficiency becomes more important.  Note also that increasing the power of the transfer pumps, and thereby shortening the transfer time, can also change things.  As usual, it's a pretty rich trade space.

So, to answer your question:  In high cadence, there's probably a number where rotational settling makes more sense than ullage thrust.  But it doesn't make any sense early on, when low cadence will almost certainly boil the depot dry between missions.  And even at high cadence, you'll have to show your work on why the extra complexity is worth it.

I didn't respond to this earlier post, but seeing this, as well as your comment about 5mm/s2 being conservative and finding the Kutter research paper (https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf) had me all but give up on the spin-G prop transfer concept I was noodling with, because I think it probably only comes out on top of linear acceleration in scenarios where you actually need considerably more than 5mm/s2 (like say 1% of G = 100mm/s2).


One other problem, which is sort of related to your "don't splash" problem:  Prop will "geyser" into the receiving tank, even if it's pumped from the bottom.  At microgee accelerations, that geysering effect is going to cause all kinds of sloshing.  That may be OK, because uncovering the pump outlet on the receiving tank shouldn't be a problem.  But if you get blobs of prop bouncing off the receiving tank walls¹, they may mess up microgee ullage accelerations enough to cause sloshing on the sending side, which could uncover the inlet.

If this is the only use for the fill pipe, they can put diverter such as a mushroom cap over the outlet.

At the kind of accelerations and head pressures we're talking about, even turbulence after the diversion is likely to cause some slosh.  But it might be enough to prevent blobs of prop from slamming into stuff with enough force to cause slosh in the sending tank.  And a mushroom cap would probably help for the "nearly full" case, too.

...But then I see these posts, and I wonder if 5mm/s2 is really that conservative all things considered? The environment inside those prop tanks sounds like it will be quite dynamic if you want to pump propellant at any significant rate, and fluids are gonna fluid. In your earlier post you were assuming 150 ton of prop transferred in 2000s (75kg/s) which seems more than enough to disrupt surface tension if those 75kg only weigh ~40g, but still have the inertia of 75kg. You wouldn't want to get up there and find you can only pump at 15kg/s if you want to avoid chaotic effects. Playing with your spreadsheet to increase transfer time to 10,000s really does blow out the number of tanker trips.

My mental image was of one of those "incoming prop" geysers interacting with the receiver tank prop in such a way as to induce a vortex in the receiver tank, potentially torquing the entire structure in chaotic ways each time.  BoE says 1400 ton of prop swirling at 1 rpm is roughly equivalent to a reaction wheel with 1,800,000 Nms of angular momentum, so if that kind of induced swirling is a real problem, then that's a heck of a design challenge for your pump connectors. Presumably there are simple ways to baffle/disrupt these effects, but it could mean pumping slower or increasing tank mass. Maybe you have to pump slower anyway, and not expending a prohibitive amount of prop on linear G acceleration would be the problem a spin-G prop transfer scheme would be solving. I think I'll noodle a bit more.
I'm assuming that an 'ideal' mature transfer scenario allows 6 hours (21,600 s) for transfer. The depots orbit can be matched up every 12 hours, assuming a single launch site. Figure three hours for approach and docking and the same to undock and clear.


I'm also assuming that only the depot will have active cooling and that PV and maybe radiators will have to tuck away during transfer ops. These assumptions point to wanting a loading campaign done either as fast as possible (a tanker every 12 hours) or very relaxed with a chance to literally chill between loads. There is no aspect of Elon that is relaxed so a fast cadence looks like a good bet. Fast also means less wear and tear on the panel deployment mechanism.


There's a lot of assumptions here but you might think about tucking some of them into your noodling.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/14/2023 07:37 pm
Pumping into a dry tank would face minimum head with geysering at max risk. As the head increases the pump pressure would need to be higher, but not so high as to geyser.
Is the head pressure going to increase? If the ullage acceleration is just a few tens of micro-g, it doesn't seem like there'll be very much holding the propellants in place. That's why I was thinking about deliberately making them swirl. If you know they're going to slow around, at least be sure they're sloshing the way you expect them to.

Head pressure doesn't increase very much, but viscosity should become your friend as things get deeper.  The deeper the pool of prop, the more any laminar jets coming from the outlet will degrade into turbulent flow and gradually damp out.

This was in the back of my mind when I asked the question about whether there's a formula for looking at what's required to break surface tension, as a function of a flow rate (more-or-less a flow velocity) and the cross section over which that rate was effective.  That's not perfect, because even if flow can't break surface tension, it's going to cause waves, which can then build up to break surface tension in other ways.  But at least it should give you a back-of-napkin estimate of the maximum tolerable rate.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/14/2023 07:43 pm
I'm assuming that an 'ideal' mature transfer scenario allows 6 hours (21,600 s) for transfer. The depots orbit can be matched up every 12 hours, assuming a single launch site. Figure three hours for approach and docking and the same to undock and clear.

Unless you have an extremely well insulated transfer path, you'll have increased boil-off inefficiency during the transfer if you drop to very low rates, because there will be more time for the flow to warm up going from one tank to another.  This is probably solvable with more insulation, but that may increase the coupling complexity.

The other problem, of course, is that longer transfers take longer ullage "burns".  So unless the acceleration requirements decrease sub-linearly with lower flow rates (and I can't see a reason offhand why they would), then longer transfer times don't help.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/14/2023 09:14 pm
I'm assuming that an 'ideal' mature transfer scenario allows 6 hours (21,600 s) for transfer. The depots orbit can be matched up every 12 hours, assuming a single launch site. Figure three hours for approach and docking and the same to undock and clear.

Unless you have an extremely well insulated transfer path, you'll have increased boil-off inefficiency during the transfer if you drop to very low rates, because there will be more time for the flow to warm up going from one tank to another.  This is probably solvable with more insulation, but that may increase the coupling complexity.

The other problem, of course, is that longer transfers take longer ullage "burns".  So unless the acceleration requirements decrease sub-linearly with lower flow rates (and I can't see a reason offhand why they would), then longer transfer times don't help.
Surely you mean sun shielding, not insulation, since the transfer is occurring in a vacuum?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/14/2023 09:26 pm
I just had an off-the-wall thought...

Well, it's really an on-the-wall thought...

As a meta-observation, note that what we're doing here is trading some combination of ullage propellant consumption and thruster complexity against some combination of transfer time (a proxy for flow rate and the momentum that comes with it) and complexity (in terms of funky inlet and outlet schemes and attitude stability). 

Seems to me that the KISS solution is to start with higher ullage accelerations and make the flow patterns be either axial and laminar or, alternatively, so turbid that the pool winds up with very few surface tensions breaks that could cause blobs on the receiving side and/or inlet uncovering on the sending side.

Things we haven't yet tried to compute and/or don't know:

1) What are the actual dynamics of the pool around the inlets and outlets as a function of flow rate? 
1a) Would knowing the maximum forward velocity that a blob could leave the pool without hitting the top of tank be a useful figure of merit?
1b) Does anybody know how to compute what's needed to break surface tension as a function of laminar flow speed and surface area of the flow?  (This doesn't help very much if the flow is highly turbulent, but might be interesting if it stays mostly laminar coming out of the outlet.)

2) Are there going to be combusting gas thrusters available?

3) A random one:  How much hydrostatic head does a pump need to function reliably?  I'm tempted to say that the answer to this is "zero", but I'm pump-illiterate.  Based on some calculations, zero is a pretty good approximation to the head at the bottom of both tanks, even at 5mm/s².

Given that we have some evidence that SpaceX is doing something new with combusting gas thrusters, and that combusting gas, if it's available in low enough thrusts or impulse bits, solves the problem, that seems the most likely solution to me.  But I'd guess that flow dynamics are a big unknown, and likely one that can only really be known via on-orbit experiments.
A couple conclusions and a bit of hands on lore and observations.

A rule of thumb for locating a flow meter is 10x the pipe ID from the last change of direction or obstruction and a minimum of 3x the ID downstream to a change of direction, obstruction or an open termination. The first is to allow the flow to become laminar, the last to avoid back impedance. This is probably all flow rate dependent but is the standard I've adhered to for instrumentation - none of which faced any radical flow rates. If transfer is direct connection between QD plates chances are high that the inlet to the O2 tank will be turbulent but I haven't thought this through for implications.

At the receiving inlet when empty, the head is zero and the only forces keeping the inrushing (in seeping?) propellant from going ballistic is g force and surface tension. Once there is any depth of fluid there will be impedance from viscosity. The impedance will be fluid velocity dependent, and sensitive to the fluid depth above the inlet. The outer annulus of fluid will be impeded the most and the fluid at the centerline of the outlet impeded the least. As fluid velocity increases the differential between the annulus and centerline will increase.

Not being capable of running the numbers I'll go out on a limb with a qualitative assessment. Higher thrust and lower transfer rate at the beginning of the transfer will keep everything playing nice. Lower thrust and higher transfer rate will work fine as the propellant level goes up. The key is to watch that centerline velocity and keep it low enough to keep the plume (right word?) from breaching the surface. The visuals would be a roiling in the propellant above the inlet with no spatter.

When filling the depot, each tanker will face different ullage thrust and transfer rates. This could well shoot down the steady 12 hour cadence I suggested earlier.

BTW, I tried looking up the viscosity of liquid methane and oxygen, and water for comparison. My talents lie elsewhere.


Edit to add: if centrifugal pumps have problems with a low head pressure a positive displacement piston pump might fill the bill.


Edit 2: Rad Mod got to viscosity before I did.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/14/2023 09:33 pm
I'm assuming that an 'ideal' mature transfer scenario allows 6 hours (21,600 s) for transfer. The depots orbit can be matched up every 12 hours, assuming a single launch site. Figure three hours for approach and docking and the same to undock and clear.

I think the single site has two launch opportunities a day, but they are not 12 hours apart.   There are also two landing opportunities a day.  This assumes the Depot orbit crosses the latitude of the launch site. Am I missing something?  If those two launch opportunities are far enough apart, then you still get an average of 12 hours for the full transfer operation. Each tanker will need some time (maybe 24 hours?) to sync with Depot's location in its orbit, and will need maybe the same amount of time after completion of fueling to move to the correct spot along the orbit to start its deorbit.  With all of this, I guess each tanker mission is about three days, so you need six tankers to maintain a 12-hour cadence.  I'm pretty sure you also need separate launch and catch towers.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/14/2023 09:45 pm
I'm assuming that an 'ideal' mature transfer scenario allows 6 hours (21,600 s) for transfer. The depots orbit can be matched up every 12 hours, assuming a single launch site. Figure three hours for approach and docking and the same to undock and clear.

Unless you have an extremely well insulated transfer path, you'll have increased boil-off inefficiency during the transfer if you drop to very low rates, because there will be more time for the flow to warm up going from one tank to another.  This is probably solvable with more insulation, but that may increase the coupling complexity.

The other problem, of course, is that longer transfers take longer ullage "burns".  So unless the acceleration requirements decrease sub-linearly with lower flow rates (and I can't see a reason offhand why they would), then longer transfer times don't help.
To make things worse, the very act of pumping will raise the propellant temp. How does this scale with pumping rate?


Sooo many things to trade. It almost like going into a big box hardware store to pick out a new drill.  :D
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/14/2023 11:17 pm
I'm assuming that an 'ideal' mature transfer scenario allows 6 hours (21,600 s) for transfer. The depots orbit can be matched up every 12 hours, assuming a single launch site. Figure three hours for approach and docking and the same to undock and clear.

I think the single site has two launch opportunities a day, but they are not 12 hours apart.   There are also two landing opportunities a day.  This assumes the Depot orbit crosses the latitude of the launch site. Am I missing something?  If those two launch opportunities are far enough apart, then you still get an average of 12 hours for the full transfer operation. Each tanker will need some time (maybe 24 hours?) to sync with Depot's location in its orbit, and will need maybe the same amount of time after completion of fueling to move to the correct spot along the orbit to start its deorbit.  With all of this, I guess each tanker mission is about three days, so you need six tankers to maintain a 12-hour cadence.  I'm pretty sure you also need separate launch and catch towers.
I'm pretty sure it's close enough to 12 hours to make no difference. One ascending node and one descending node.


In theory, when the target orbit is the same inclination as the launch sites latitude, they can match orbits near dead on with no chase time. In theory. IIRC they have hooked up with the ISS with a six hour chase.


Fast turnaround for the tankers is not on the immediate horizon. Until they get some experience under their belt I'd expect to see a fleet of tankers, each doing only one transfer per campaign. They promise to be the simplest and least expensive variant.


Artemus will need two depots or one depot and one super tanker that will need some refilling. Each would be a separate campaign using the same tanker fleet. If the depot is near zero boiloff they can launch it well in advance, then launch the second depot or supertanker, refill as necessary and move it to where it needs to be. Once all the pieces are in place, crew launches.


Back to refueling: If the chase is six hours they can still launch every 12 hours and have a six hour deep pipeline bringing another tanker to the depot every 12 hours. Assuming the slowest transfer can be done in 12 hours.


With a full fleet of tankers that can collectively service one campaign, the timely return of the tankers is less of a concern. The biggest limitation on return is boiling dry.
Dropping only a couple of km after the transfer, the tankers would continually be opening the distance to the depot, but not so low that decay would be immanent. Bring 'em back as convenient. That said, yes, multiple pads would be a good thing (he said, wondering how OLM and tower will look after that first launch).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/15/2023 01:16 am
Surely you mean sun shielding, not insulation, since the transfer is occurring in a vacuum?

I was thinking more of the actual transfer plumbing, which will heat up radiatively but transfer heat to the prop flowing through it conductively.  If it's always shaded or insulated well enough that conductive transfer negligible, then it's not an issue.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/15/2023 03:11 am
I'm assuming that an 'ideal' mature transfer scenario allows 6 hours (21,600 s) for transfer. The depots orbit can be matched up every 12 hours, assuming a single launch site. Figure three hours for approach and docking and the same to undock and clear.

I'm also assuming that only the depot will have active cooling and that PV and maybe radiators will have to tuck away during transfer ops. These assumptions point to wanting a loading campaign done either as fast as possible (a tanker every 12 hours) or very relaxed with a chance to literally chill between loads. There is no aspect of Elon that is relaxed so a fast cadence looks like a good bet. Fast also means less wear and tear on the panel deployment mechanism.

There's a lot of assumptions here but you might think about tucking some of them into your noodling.

Thanks, and agreed that it's a huge trade space. I've also been assuming that only the depot will have active cooling, and will in fact have a number of optimisations for keeping prop cool that it doesn't make sense to install on every Starship (OT: whether that would lead to sending a single depot with each fleet of Starships that makes the Earth-Mars transit is an interesting question to me: the ability to keep one ship uncrewed and cold the whole trip seems like a good energy saving measure, then also, you have the capability to top up the rest of the fleet's header tanks with fresh cryo prop in the days before landing).

Just attaching a screenshot of what happens if you plug 21,600s into TheRadicalModerate's spreadsheet (https://docs.google.com/spreadsheets/d/1MUkdFeYbZp8wlTXGPiapRMv4fTWCFodr8UrVPwyYXNI/edit?usp=sharing). It now requires 15 tanker trips to fill a depot, and thats with 300s hot gas thrusters going the whole time to create 5mm/s2. I somehow doubt that losing nearly 40% of the propellant launched is really acceptable. With 70s ullage thrust, the transfers lose so much prop that the process never completes. If you go to 1mm/s2 and 300s thrust you get to 5.7% losses over 12 tanker loads which is (maybe) acceptable.

My key takeaway/question is whether propellant loss due to milli-G acceleration is the only reason to do prop transfers as fast as the system allows. Everything else (I think) pushes you towards taking your time - you probably wouldn't launch a tanker more than once a day (at least at first) so a 6+ hour prop transfer isn't going to be the limiting factor, and lower transfer speed means less sloshing/swirling in the tank.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/15/2023 03:39 am
If we need nearly continuous low acceleration and we think this will waste too much propellant, perhaps we need to look a high efficiency thrusters. Specifically, ion thrusters of some sort. Something like the Gateway PPE. This would be part of the Depot, not the tankers, so the cost of lifting it to orbit is only incurred once. Ideally, we would use some magical thruster that uses CH4 or O2 instead of Xenon.

Question: how much thrust is needed to maintain 0.5 mm/s2 for a max-loaded Depot/tanker combo?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/15/2023 04:23 am
My key takeaway/question is whether propellant loss due to milli-G acceleration is the only reason to do prop transfers as fast as the system allows. Everything else (I think) pushes you towards taking your time - you probably wouldn't launch a tanker more than once a day (at least at first) so a 6+ hour prop transfer isn't going to be the limiting factor, and lower transfer speed means less sloshing/swirling in the tank.

For the parameters you have set here, the transfer mass flow is only 6.9kg/s.  Also, the longer you take at a particular acceleration, the more delta-v you wind up spending.  If you take 21,600s at 5mm/s², you're expending more than 1600m/s of delta-v.  Your depot/tanker will wind up in a much higher orbit, and both vehicles will have to spend a bunch of delta-v to get back to the depot's parking orbit.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/15/2023 04:33 am
My key takeaway/question is whether propellant loss due to milli-G acceleration is the only reason to do prop transfers as fast as the system allows. Everything else (I think) pushes you towards taking your time - you probably wouldn't launch a tanker more than once a day (at least at first) so a 6+ hour prop transfer isn't going to be the limiting factor, and lower transfer speed means less sloshing/swirling in the tank.

For the parameters you have set here, the transfer mass flow is only 6.9kg/s.  Also, the longer you take at a particular acceleration, the more delta-v you wind up spending.  If you take 21,600s at 5mm/s², you're expending more than 1600m/s of delta-v.  Your depot/tanker will wind up in a much higher orbit, and both vehicles will have to spend a bunch of delta-v to get back to the depot's parking orbit.
I think the maneuver during propellant transfer is a big circle, so the net change in position is zero. This introduces a very small lateral acceleration which should not affect the transfer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/15/2023 04:43 am
If we need nearly continuous low acceleration and we think this will waste too much propellant, perhaps we need to look a high efficiency thrusters. Specifically, ion thrusters of some sort. Something like the Gateway PPE. This would be part of the Depot, not the tankers, so the cost of lifting it to orbit is only incurred once. Ideally, we would use some magical thruster that uses CH4 or O2 instead of Xenon.

Question: how much thrust is needed to maintain 0.5 mm/s2 for a max-loaded Depot/tanker combo?

F = ma, baby.  m = 95t + 1450t + 120t + 150t = 1,815,000kg.  a = 0.005m/s² 0.0005m/s².  Thrust = 9075N 908N.

Electric propulsion usually generates 60mN/kW.  So you'd need 151MW 15.1MW of power to drive the coupled system.  Nope.

Update:  Thanks to tbellman for catching that I'd dropped a decimal--with no change in the conclusion.

I think the maneuver during propellant transfer is a big circle, so the net change in position is zero. This introduces a very small lateral acceleration which should not affect the transfer.

OK, fair point.  But... 1600m/s spent on ullage acceleration?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/15/2023 10:41 am
On geysering and slosh reduction: The early Jupiter missiles (https://forum.nasaspaceflight.com/index.php?topic=57685.msg2435702#msg2435702) used a system of floating bodies inside the tank covering the propellant/ullage gas interface to reduce propellant slosh (after observation that transporting water tanks with lumber floating in them significantly reduced slosh) prior to the introduction of anti-slosh and anti-vortex baffles to the tank walls. I've previously suggested this floating body system may be repurposed as an insulator layer for reduction of ullage gas heating, but it could also serve to reduce the effects of geysering from propellant pumping, by absorbing stream and droplet energy in trying to move the more massive (in discrete terms) floats rather than allowing droplets to continue unimpeded.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tbellman on 01/15/2023 02:08 pm
Question: how much thrust is needed to maintain 0.5 mm/s2 for a max-loaded Depot/tanker combo?

F = ma, baby.  m = 95t + 1450t + 120t + 150t = 1,815,000kg.  a = 0.005m/s².  Thrust = 9075N.

Electric propulsion usually generates 60mN/kW.  So you'd need 151MW of power to drive the coupled system.  Nope.

Dan asked about half a millimeter per second squared, while you calculated for five.  So you would only need 15 megawatt.

"Only"...

So your conclusion is correct anyway; that is still a "Nope", probably even a "No way in h***".

Heck, even an empty Starship at 120 tonne would need 60 N thrust for that acceleration, which would require 1 MW electric power, which is probably about an order of magnitude more than you can expect from solar panels carried on a Starship...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/15/2023 02:34 pm
Electric propulsion usually generates 60mN/kW.  So you'd need 151MW of power to drive the coupled system.  Nope.
Is that really out of the question, though? I just skimmed "Temperature and Power Specific Mass Scaling for LEU Closed-Cycle Brayton Systems for Space Surface Power and Nuclear Electric Propulsion (http://anstd.ans.org/NETS-2019-Papers/Track-6--Energy-Conversion-Technology-and-Development/abstract-126-0.pdf) (Christopher G. Morrison, 2019), and it suggests that a 150 MW reactor would have a mass of about 150 tons (looking at figure 8 on page 5). Admittedly, that'd be a very advanced, high-temperature reactor, but it's not totally nuts.

Of course, if you can put that in orbit, you might as well use it to transport all payload except maybe human beings.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/15/2023 05:48 pm
Question: how much thrust is needed to maintain 0.5 mm/s2 for a max-loaded Depot/tanker combo?

F = ma, baby.  m = 95t + 1450t + 120t + 150t = 1,815,000kg.  a = 0.005m/s².  Thrust = 9075N.

Electric propulsion usually generates 60mN/kW.  So you'd need 151MW of power to drive the coupled system.  Nope.

Dan asked about half a millimeter per second squared, while you calculated for five.  So you would only need 15 megawatt.

"Only"...

So your conclusion is correct anyway; that is still a "Nope", probably even a "No way in h***".

Heck, even an empty Starship at 120 tonne would need 60 N thrust for that acceleration, which would require 1 MW electric power, which is probably about an order of magnitude more than you can expect from solar panels carried on a Starship...
Yep, TheRadicalModerate answered the question I should have asked instead of the question I asked.  :)

Apparently, ISS has 2500 m3 of panels to generate an average system power of 125 kW. That's 2m3/kw at the system level, so we would need 3.02 km2 of solar panels to generate 151 MW. You would construct this nightmare using a whole lot of Starship flights and then use it for thirty years. You don't want to accelerate the panels or batteries, so by the time you have done all the engineering you probably should be looking at a different solution.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/15/2023 09:56 pm
I'm assuming that an 'ideal' mature transfer scenario allows 6 hours (21,600 s) for transfer. The depots orbit can be matched up every 12 hours, assuming a single launch site. Figure three hours for approach and docking and the same to undock and clear.

I'm also assuming that only the depot will have active cooling and that PV and maybe radiators will have to tuck away during transfer ops. These assumptions point to wanting a loading campaign done either as fast as possible (a tanker every 12 hours) or very relaxed with a chance to literally chill between loads. There is no aspect of Elon that is relaxed so a fast cadence looks like a good bet. Fast also means less wear and tear on the panel deployment mechanism.

There's a lot of assumptions here but you might think about tucking some of them into your noodling.

Thanks, and agreed that it's a huge trade space. I've also been assuming that only the depot will have active cooling, and will in fact have a number of optimisations for keeping prop cool that it doesn't make sense to install on every Starship (OT: whether that would lead to sending a single depot with each fleet of Starships that makes the Earth-Mars transit is an interesting question to me: the ability to keep one ship uncrewed and cold the whole trip seems like a good energy saving measure, then also, you have the capability to top up the rest of the fleet's header tanks with fresh cryo prop in the days before landing).

Just attaching a screenshot of what happens if you plug 21,600s into TheRadicalModerate's spreadsheet (https://docs.google.com/spreadsheets/d/1MUkdFeYbZp8wlTXGPiapRMv4fTWCFodr8UrVPwyYXNI/edit?usp=sharing). It now requires 15 tanker trips to fill a depot, and thats with 300s hot gas thrusters going the whole time to create 5mm/s2. I somehow doubt that losing nearly 40% of the propellant launched is really acceptable. With 70s ullage thrust, the transfers lose so much prop that the process never completes. If you go to 1mm/s2 and 300s thrust you get to 5.7% losses over 12 tanker loads which is (maybe) acceptable.

My key takeaway/question is whether propellant loss due to milli-G acceleration is the only reason to do prop transfers as fast as the system allows. Everything else (I think) pushes you towards taking your time - you probably wouldn't launch a tanker more than once a day (at least at first) so a 6+ hour prop transfer isn't going to be the limiting factor, and lower transfer speed means less sloshing/swirling in the tank.
Um, uh... We'll, maybe we should think of 21,600 seconds as a window. Running simple minded numbers that even I can handle, using the whole window would be a transfer of only ~7kg/sec. Divvied up, it's ~1.4kg methane and ~5.5kg O2. Someone else can figure out the volume.


With a lot of hand waving, I'd say that this would be somewhere near what's needed when starting with an empty receiving tank, and at the highest settling thrust. Maybe that 5mm/s^2, but lower if possible. As the propellant level rises thrust can drop and transfer rate increase. The first 150t load would most likely be the most expensive, propellant wise and wouldn't need the full window. Each following load would need less settling thrust and less transfer time. Something to keep in mind is that the thrust/transfer rates may be very different for methane and O2.


The .5mm/s^2 needed for Centaur might be higher than needed for SS. Different sizes, different thrusters, different fuel, different team. Whatever is needed, I think we'll be looking at a range, not one number.


Does anybody have an intelligible comparison of the viscosity of water at 20C vs methane and O2 at both boiling and slightly above freezing?



Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/15/2023 10:44 pm
OK, let's start over. Assume we need 5 mm/s2 accelleration, and we want to use centrepital force. We mount the tanker atop the depot and then sling them from a cable with tanker toward the center of rotation, and start spinning. We need another mass at the far end of the cable. let's say that otehr mass is 1/10 the mass of our tanker+depot+propellant. For a cable that is 1100 meters, the tanker+depot CoM will be 100 meters from the center of rotation. The formula is
    accel = (rotational rate squared) * radius
I get a rotational rate of 0.42 RPM.     Is this correct?

Tension in the cable in Newtons is mass the complex times 0.005 m/s2, which is not a lot.

Practical considerations:
    How to connect the cable to the depot so as to still allow the tanker to dock atop it? I think this will require a wide set of spreaders. CoM must stay "below" the attachment point even for a full tanker on an empty depot unless the tanket is stabilized to the cable using the tanker's chopstick catchpoints.
    How to spin up after docking the tanker and spin down before undocking?  I think thrusters on the depot.
    How to keep the cable from tangling when not spinning? I think make it slightly rigid, possibly using a pressurized pipe.
    What is the right length and the right mass for the counterweight? I think 100 meters is too short for CoM to center of spin since the gradient is too high (tanker too light, depot too heavy). A longer cable means a lower rotation rate.
    How much energy is needed to spin up and spin down? ( too much math for me right  now).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/15/2023 11:35 pm
On geysering and slosh reduction: The early Jupiter missiles (https://forum.nasaspaceflight.com/index.php?topic=57685.msg2435702#msg2435702) used a system of floating bodies inside the tank covering the propellant/ullage gas interface to reduce propellant slosh (after observation that transporting water tanks with lumber floating in them significantly reduced slosh) prior to the introduction of anti-slosh and anti-vortex baffles to the tank walls. I've previously suggested this floating body system may be repurposed as an insulator layer for reduction of ullage gas heating, but it could also serve to reduce the effects of geysering from propellant pumping, by absorbing stream and droplet energy in trying to move the more massive (in discrete terms) floats rather than allowing droplets to continue unimpeded.
Any idea what the floats were made of on the O2 side?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/16/2023 12:27 am
OK, let's start over. Assume we need 5 mm/s2 accelleration, and we want to use centrepital force. We mount the tanker atop the depot and then sling them from a cable with tanker toward the center of rotation, and start spinning. We need another mass at the far end of the cable. let's say that otehr mass is 1/10 the mass of our tanker+depot+propellant. For a cable that is 1100 meters, the tanker+depot CoM will be 100 meters from the center of rotation. The formula is
    accel = (rotational rate squared) * radius
I get a rotational rate of 0.42 RPM.     Is this correct?

Tension in the cable in Newtons is mass the complex times 0.005 m/s2, which is not a lot.

Practical considerations:
    How to connect the cable to the depot so as to still allow the tanker to dock atop it? I think this will require a wide set of spreaders. CoM must stay "below" the attachment point even for a full tanker on an empty depot unless the tanket is stabilized to the cable using the tanker's chopstick catchpoints.
    How to spin up after docking the tanker and spin down before undocking?  I think thrusters on the depot.
    How to keep the cable from tangling when not spinning? I think make it slightly rigid, possibly using a pressurized pipe.
    What is the right length and the right mass for the counterweight? I think 100 meters is too short for CoM to center of spin since the gradient is too high (tanker too light, depot too heavy). A longer cable means a lower rotation rate.
    How much energy is needed to spin up and spin down? ( too much math for me right  now).
While the concept looks promising in theory, tethers are hard. Your inflatable tube tether is ingenious and has the seeds of a deployment system embedded in it. Deployment is one of the really hard parts.


Tethers are something that deserve a lot more research than has been put into them. Another way of saying that is: they might be useful in another 5-15 years if somebody invests a lot of time, energy and money.
See: [size=78%]https://en.m.wikipedia.org/wiki/Space_tether (https://en.m.wikipedia.org/wiki/Space_tether)[/size]


BTW, in LEO, a metallic tether normal to earth would be a wire cutting through magnetic field lines - potentially a generator to run some ullage thrusters but again, not ready for prime time.


In the time frame of Artemus, it looks like optimizing to minimize settling thrust is most practical. If it takes a few more tanker flights that's the cost of doing business.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/16/2023 01:26 am
While the concept looks promising in theory, tethers are hard. Your inflatable tube tether is ingenious and has the seeds of a deployment system embedded in it. Deployment is one of the really hard parts.
I don't think this is original with me. I think I saw it on the forum somewhere. I was certainly inspired by something here. There is no reason the inflatable stiffener must also be the tension member, so it's not necessarily an "inflatable tether". The stiffener might be the main part of the deployment mechanism, or not. After deployment it is only needed when the tether is not under tension.

The energy needed to spin the system up is apparently
  energy = mass * radius * acceleration.

I derived this from the flywheel equation.

Its really the energy stored in the counterweight plus the energy stored in the (depot+tanker), but I think the energy is evenly divided so I just doubled the energy stored in the (depot+tanker).
m=1,815,000kg.  a = 0.005m/s².  r = 100 (but probably more)

We therefore need
      e = 1,815,000kg *  100 m * 0.005m/s².
         = 907,500 Joules
for each spin-up and each spin-down. I have no idea how to convert this into expended propellant.

I suspect you are correct and this system cannot be developed in time for the first HLS mission: they will just throw extra tankers at the problem.

I also suspect the professionals will come up with an entirely different solution that is far superior.
             
   
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/16/2023 05:00 am
It occurred to me that, if you engineer your pumping power so that it achieves the needed flow rate but causes the liquid to exit the outlet pipe at exactly ambient pressure, then there's no geysering possible.

You can figure out that pump power via the Poiseuille equation,¹ which tells you the pressure drop from the pump outlet to the ambient fill pipe outlet, and then you can derive the pump power² from the pressure drop.  So it requires a requires a variable-power pump to compensate for the ambient hydrostatic head at the outlet, but otherwise the prop will ooze out of the outlet at just the right rate with basically no kinetic energy.

______________
¹Poiseuille equation:

Δp = 8μLQ/(πR⁴)

Where:
Δp is the pressure drop through the pipe
μ is the dynamic viscosity (5.0E-5 Pa-s for LOX, 1.1E-5 Pa-s for LCH4)
L is the pipe length
Q is the volumetric flow rate (in m³/s, not kg/s)
R is the pipe radius

²The pump power is:

P = (Δp + poutletHead)Q/η

where
P is the power (W)
η is the pump efficiency.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/16/2023 07:09 am
It occurred to me that, if you engineer your pumping power so that it achieves the needed flow rate but causes the liquid to exit the outlet pipe at exactly ambient pressure, then there's no geysering possible.

You can figure out that pump power via the Poiseuille equation,¹ which tells you the pressure drop from the pump outlet to the ambient fill pipe outlet, and then you can derive the pump power² from the pressure drop.  So it requires a requires a variable-power pump to compensate for the ambient hydrostatic head at the outlet, but otherwise the prop will ooze out of the outlet at just the right rate with basically no kinetic energy.

______________
¹Poiseuille equation:

Δp = 8μLQ/(πR⁴)


Interesting, I didn't realise this (/R4!!). Key takeaway being that fat pipes are your friend, since one assumes SpaceX were already going to pick the most efficient pumps possible.

Thought I should show you the current progress of my noodling (not the final product - the revisions of these animations will form part of my next youtube video). I should say that this began as a revision of my spin-G Starship work, but the configuration that I came up with seemed to be applicable to the propellant transfer problem.

This particular spin config - with the offset nose docking and dorsal-ventral rotation - was one that I hadn't considered until recently, but I'm liking it more and more (solar panels/radiators not yet shown, bi-directional ports assumed). It means you don't need to move the header tank at all. Even if you really did want to put the docking port on the nose, you'd have to figure out how to protect the port with the heat shield during EDL.

The offset port at the nose also avoids intermediate axis issues if you rotate in the dorsal-ventral plane. Obviously the pipes on the depot variant are pretty long, but they are all on the leeward side of the vehicle. This animation depicts 4 pipes on the arm (2x Methane and LOX, liquid and ullage) each with a diameter of 20cm. 

https://youtu.be/sN6Zh1P0748
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/16/2023 03:33 pm
It occurred to me that, if you engineer your pumping power so that it achieves the needed flow rate but causes the liquid to exit the outlet pipe at exactly ambient pressure, then there's no geysering possible.

You can figure out that pump power via the Poiseuille equation,¹ which tells you the pressure drop from the pump outlet to the ambient fill pipe outlet, and then you can derive the pump power² from the pressure drop.  So it requires a requires a variable-power pump to compensate for the ambient hydrostatic head at the outlet, but otherwise the prop will ooze out of the outlet at just the right rate with basically no kinetic energy.

______________
¹Poiseuille equation:

Δp = 8μLQ/(πR⁴)

Where:
Δp is the pressure drop through the pipe
μ is the dynamic viscosity (5.0E-5 Pa-s for LOX, 1.1E-5 Pa-s for LCH4)
L is the pipe length
Q is the volumetric flow rate (in m³/s, not kg/s)
R is the pipe radius

²The pump power is:

P = (Δp + poutletHead)Q/η

where
P is the power (W)
η is the pump efficiency.
This looks like an upper limit on a dry receiving tank. Once the fluid depth reaches ~one inlet diameter it can pump a bit harder and let viscosity damp geysering. Given the consumption numbers we're seeing for settling, the transfer time looks critical.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/16/2023 04:00 pm
It occurred to me that, if you engineer your pumping power so that it achieves the needed flow rate but causes the liquid to exit the outlet pipe at exactly ambient pressure, then there's no geysering possible.

You can figure out that pump power via the Poiseuille equation,¹ which tells you the pressure drop from the pump outlet to the ambient fill pipe outlet, and then you can derive the pump power² from the pressure drop.  So it requires a requires a variable-power pump to compensate for the ambient hydrostatic head at the outlet, but otherwise the prop will ooze out of the outlet at just the right rate with basically no kinetic energy.

______________
¹Poiseuille equation:

Δp = 8μLQ/(πR⁴)


Interesting, I didn't realise this (/R4!!). Key takeaway being that fat pipes are your friend, since one assumes SpaceX were already going to pick the most efficient pumps possible.

Thought I should show you the current progress of my noodling (not the final product - the revisions of these animations will form part of my next youtube video). I should say that this began as a revision of my spin-G Starship work, but the configuration that I came up with seemed to be applicable to the propellant transfer problem.

This particular spin config - with the offset nose docking and dorsal-ventral rotation - was one that I hadn't considered until recently, but I'm liking it more and more (solar panels/radiators not yet shown, bi-directional ports assumed). It means you don't need to move the header tank at all. Even if you really did want to put the docking port on the nose, you'd have to figure out how to protect the port with the heat shield during EDL.

The offset port at the nose also avoids intermediate axis issues if you rotate in the dorsal-ventral plane. Obviously the pipes on the depot variant are pretty long, but they are all on the leeward side of the vehicle. This animation depicts 4 pipes on the arm (2x Methane and LOX, liquid and ullage) each with a diameter of 20cm. 

https://youtu.be/sN6Zh1P0748 (https://youtu.be/sN6Zh1P0748)
Very interesting and a nice render. If the depot never comes back that external plumbing is no problem at all.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/16/2023 04:59 pm
We therefore need
      e = 1,815,000kg *  100 m * 0.005m/s².
         = 907,500 Joules
for each spin-up and each spin-down. I have no idea how to convert this into expended propellant.
OK, I did all the elaborate math, and then realized there is a simpler way to think about this to get to the same number. Consider two masses at relative rest. These masses then accelerate on antiparallel courses to reach relative velocities that equal their tangential velocities in the tether system, and they then (conceptually) grab the ends of the tether. This is NOT how the system works in practice, but it consumes the same amount of energy This mathematically converts the energy problem into a delta-V problem. The tangential velocities in the tether system are tiny. For a 100 m radius at .42 rpm (.00707 rps) the tangential velocity is 628 m * .00707/s =  4.44 m/s.

For equal masses, multiply by 2 to get the total energy. I conclude that spin-up and spin-down use much less propellant than the RMOD needed to hook the tanker to the depot.

Because we are now using delta-V, we can easily evaluate the mass expenditure in the usual way.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/16/2023 08:47 pm
This looks like an upper limit on a dry receiving tank. Once the fluid depth reaches ~one inlet diameter it can pump a bit harder and let viscosity damp geysering. Given the consumption numbers we're seeing for settling, the transfer time looks critical.

That's correct.  However, the "duh!" moment I had was realizing that, for any flow rate, there's a pump power that results in that flow arriving at ambient pressure.

FWIW, I'm building a transfer power worksheet into the prop expenditure worksheet.  The current powers are so low (single-digit watts) that I think I have a bug.  However, when I did this computation by hand up-thread, the power was also very low.  I'll post the link when I'm more confident that it's not wrong.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 01/17/2023 12:50 am
OK, let's start over. Assume we need 5 mm/s2 accelleration, and we want to use centrepital force. We mount the tanker atop the depot and then sling them from a cable with tanker toward the center of rotation, and start spinning. We need another mass at the far end of the cable. let's say that otehr mass is 1/10 the mass of our tanker+depot+propellant. For a cable that is 1100 meters, the tanker+depot CoM will be 100 meters from the center of rotation.

I agree that if the xfer time and settling-acceleration are both large, centripetal force is required instead of linear ullage thrust, but...

All the complexity of your cable and counter-weight (and their deployment as part of the docking/spin-up/spin-down/undocking sequence) exists just to allow them to both spin "tail down, nose up" for the prop transfer, while still being side-by-side.

Surely there's less complexity and risk in just sticking an extra set of propellant/gas lines and intakes in the tankers/depot/LSS tanks at the appropriate position for side-by-side rotation?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/17/2023 02:01 am
OK, let's start over. Assume we need 5 mm/s2 accelleration, and we want to use centrepital force. We mount the tanker atop the depot and then sling them from a cable with tanker toward the center of rotation, and start spinning. We need another mass at the far end of the cable. let's say that otehr mass is 1/10 the mass of our tanker+depot+propellant. For a cable that is 1100 meters, the tanker+depot CoM will be 100 meters from the center of rotation.

I agree that if the xfer time and settling-acceleration are both large, centripetal force is required instead of linear ullage thrust, but...

All the complexity of your cable and counter-weight (and their deployment as part of the docking/spin-up/spin-down/undocking sequence) exists just to allow them to both spin "tail down, nose up" for the prop transfer, while still being side-by-side.

Surely there's less complexity and risk in just sticking an extra set of propellant/gas lines and intakes in the tankers/depot/LSS tanks at the appropriate position for side-by-side rotation?
I did not make myself clear. Sorry. I am assuming that the tanker docks atop the depot, like an SS stacked on a booster. They are not side by side. The depot would have an interface similar to the booster interface at its top end: basically a 9 meter diameter dock. The booster is assumed to deploy a QD "arm" from near its top that reaches up to connect to the QD plate on the SS (tanker, LSS, whatever).

I also assume that depot installation is permanent: it will last for many years and perform many, many propellant transfers. The tether and (large) counterweight are deployed (constructed?) once. You can think of the depot, tether, and counterweight as one really long spacecraft.

For each propellant transfer:
     start with no spin
     SS RMOD
     spin
     transfer propellant
     despin
     SS undock and depart.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/17/2023 04:52 am
Does anybody have an intelligible comparison of the viscosity of water at 20C vs methane and O2 at both boiling and slightly above freezing?

Water (https://en.wikipedia.org/w/index.php?title=File:Dynamic_Viscosity_of_Water.png) at 20ºC, μ=8E-4 Pa-s.
For LOX near boiling, μ=5.0E-5 Pa-s
For LCH4 near boiling μ=1.1E-5 Pa-s

I don't have the slushy versions for LOX and LCH4, but they don't matter, because prop on-orbit is almost certainly going to be near boiling.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/17/2023 05:01 am
I agree that if the xfer time and settling-acceleration are both large, centripetal force is required instead of linear ullage thrust, but...

If I'm right that, irrespective of the volumetric flow, you can always adjust pump power to put the outlet pressure at ambient, then transfer time is whatever you want it to be.

I suspect that there will be reasons for settling acceleration to be greater than 1mm/s², but if that's the case, then combusting methox thrusters are about a jillion times easier to engineer than a rotating system.  Rotating systems are terrible to make reliable, especially when you're moving their center of mass around by pumping stuff from point A to point B.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 01/17/2023 07:56 am
we want to use centrepital force. We mount the tanker atop the depot and then sling them from a cable with tanker toward the center of rotation, and start spinning. We need another mass at the far end of the cable. let's say that otehr mass is 1/10 the mass of our tanker+depot+propellant. For a cable that is 1100 meters, the tanker+depot CoM will be 100 meters from the center of rotation.
I did not make myself clear. Sorry. I am assuming that the tanker docks atop the depot, like an SS stacked on a booster. They are not side by side. The depot would have an interface similar to the booster interface at its top end: basically a 9 meter diameter dock. The booster is assumed to deploy a QD "arm" from near its top that reaches up to connect to the QD plate on the SS (tanker, LSS, whatever).
I also assume that depot installation is permanent: it will last for many years and perform many, many propellant transfers. The tether and (large) counterweight are deployed (constructed?) once. You can think of the depot, tether, and counterweight as one really long spacecraft.

Now I'm confused. Where does the counterweight+arm attach to the depot, if the tankers are docking at the depot-nose?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: volker2020 on 01/17/2023 09:17 am
Not sure a nose by nose rotational docking is even possible. The mass center of a docked combination of full depot and an empty starship, will be somewhere in the upper third of the depot oxygen tank. How rotation in this case should help settle something is beyond my expertise. Apart from that, the rotation is poised to become extremely unstable. If SpaceX would make that work, it would be a marvel in itself.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: BT52 on 01/17/2023 10:10 am
I dont wanna be rude or anything. Any rotation is quite stupid way to overengineer.

Linear acceleration is way to go. Its well know phenomena. And furthermore they known its effects from 2nd stage.

Just copy pasta baffle system from main tanks and its almost good. Then they just need pump, power source and manipulator arm for docking adapter plate.

Maybe oversimplified view but i think that way they can reuse as many parts or experiences from ground zero stage as possible.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/17/2023 11:28 am
I wonder if it would be possible to plumb fluid transfer in via the header tanks? With their small enclosed volume, you can pump props in as fast as you can and allow the violent sloshing and geysering to occur inside them. The headers already have more extensive baffling due to the need to minimise slosh during the flip manoeuvre, so even with no additional fluid management hardware and just an open line to the main tank allowing mixed phase flow, a huge amount of energy in the incoming propellant flow should be expended inside the header allowing for gentler flow into the main tank.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/17/2023 02:25 pm
we want to use centrepital force. We mount the tanker atop the depot and then sling them from a cable with tanker toward the center of rotation, and start spinning. We need another mass at the far end of the cable. let's say that otehr mass is 1/10 the mass of our tanker+depot+propellant. For a cable that is 1100 meters, the tanker+depot CoM will be 100 meters from the center of rotation.
I did not make myself clear. Sorry. I am assuming that the tanker docks atop the depot, like an SS stacked on a booster. They are not side by side. The depot would have an interface similar to the booster interface at its top end: basically a 9 meter diameter dock. The booster is assumed to deploy a QD "arm" from near its top that reaches up to connect to the QD plate on the SS (tanker, LSS, whatever).
I also assume that depot installation is permanent: it will last for many years and perform many, many propellant transfers. The tether and (large) counterweight are deployed (constructed?) once. You can think of the depot, tether, and counterweight as one really long spacecraft.

Now I'm confused. Where does the counterweight+arm attach to the depot, if the tankers are docking at the depot-nose?
The tether connects to a spreader. The spreader has rigid cables, one for each side of the depot+tanker. The cables are about 12 m apart and maybe 50 m long and connect to a trapeze that connects to the depot near its "top", near its chopstick lift points.  After Tanker docks to depot, it can also connect to the rigid cables for lateral stability if needed, but the actual centripetal acceleration on tanker comes from its base, which sits on the depot.

The entire RMOD occurs in the de-spun state, so if necessary the depot can rotate 90 degrees to become  perpendicular to the tether so the docking port is completely clear of the tether. Then tanker docks and then depot+tanker swing back in line with the tether before tanker hooks its stabilizers onto the tether. I don't think all of this is needed. I think tanker can RMOD by maneuvering itself between the spread cables: they are farther apart than chopstick arms and RMOD can take as long as needed. Those stabilizers will never have more than a tiny force on them. Are only needed at all if the CoM of tanker+depot is "higher" than the trapeze connection point on depot (i.e., full tanker, empty depot).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/17/2023 08:25 pm
This looks like an upper limit on a dry receiving tank. Once the fluid depth reaches ~one inlet diameter it can pump a bit harder and let viscosity damp geysering. Given the consumption numbers we're seeing for settling, the transfer time looks critical.

That's correct.  However, the "duh!" moment I had was realizing that, for any flow rate, there's a pump power that results in that flow arriving at ambient pressure.

FWIW, I'm building a transfer power worksheet into the prop expenditure worksheet.  The current powers are so low (single-digit watts) that I think I have a bug.  However, when I did this computation by hand up-thread, the power was also very low.  I'll post the link when I'm more confident that it's not wrong.
Ok. Now I get it. Didn't think it through.


Hmmm. If the pump hits exactly at ambient pressure at the top of the inlet, wouldn't the flow stall out? I'm thinking the column of fluid would just make it to the lip of the inlet and no further. One tiny smidge more and it overflows the lip and fills the sump. Once the tank level reaches the top of the inlet it would continue to rise about one smidge worth, then stall out again. How many kWh pump energy in a smidge?


From a practical point of view what is the relationship between pressure and flow rate in a centrifugal pump? If it's pumping into a closed off pipe it's at high pressure but zero flow. Valve the outlet open just a tad (first cousin to a smidge) and flow starts but pressure drops. Once the outlet is fully open both pressure and volume rise and fall with RPM but is it a linear relationship? My gut says it's close. Does the relationship change facing different back pressures?


Am I overthinking this? ???


A random thought. If the volume and pressure are directly related, the transfer op would counterintuitively go faster at 6bar ullage pressure than at .5bar. Would it be fast enough to materially impact propellant consumption? Would the amount of makeup gas from the high pressure COPV's negate any advantage? The makeup gas will be warm and would contract in the tanks. Electric heater on the COPV outlets?


My brain is starting to hurt.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/17/2023 11:12 pm
Ok. Now I get it. Didn't think it through.


Hmmm. If the pump hits exactly at ambient pressure at the top of the inlet, wouldn't the flow stall out? I'm thinking the column of fluid would just make it to the lip of the inlet and no further. One tiny smidge more and it overflows the lip and fills the sump. Once the tank level reaches the top of the inlet it would continue to rise about one smidge worth, then stall out again. How many kWh pump energy in a smidge?


From a practical point of view what is the relationship between pressure and flow rate in a centrifugal pump? If it's pumping into a closed off pipe it's at high pressure but zero flow. Valve the outlet open just a tad (first cousin to a smidge) and flow starts but pressure drops. Once the outlet is fully open both pressure and volume rise and fall with RPM but is it a linear relationship? My gut says it's close. Does the relationship change facing different back pressures?


Am I overthinking this? ???


A random thought. If the volume and pressure are directly related, the transfer op would counterintuitively go faster at 6bar ullage pressure than at .5bar. Would it be fast enough to materially impact propellant consumption? Would the amount of makeup gas from the high pressure COPV's negate any advantage? The makeup gas will be warm and would contract in the tanks. Electric heater on the COPV outlets?


My brain is starting to hurt.

Yeah, it's pretty counterintuitive, and I'm not sure I've understood it correctly.  I'm wondering if Poseuille only applies to static pressure, but that would be weird with an equation that has volumetric flow built into it.  But if you've got the same volumetric flow and it's incompressible, it's hard to see how there isn't dynamic pressure just from the momentum of the flow itself.

Hopefully somebody will chime in and explain it all.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 01/18/2023 12:41 am
I dont wanna be rude or anything. Any rotation is quite stupid way to overengineer.
Linear acceleration is way to go. Its well know phenomena.

The fall back to centripetal settling is a result of the calcs for the amount of prop required for ullage burns during transfer. If you have to burn an entire extra tanker load of propellant just to do prop settling, then the technical complexity of rotation might be worth it. Perhaps not the trapeze act that Dan has proposed, but certainly the cost of adding extra sumps/vents at the required points in the tankers, depot, and LSS.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/18/2023 01:16 am
I dont wanna be rude or anything. Any rotation is quite stupid way to overengineer.
Linear acceleration is way to go. Its well know phenomena.

The fall back to centripetal settling is a result of the calcs for the amount of prop required for ullage burns during transfer. If you have to burn an entire extra tanker load of propellant just to do prop settling, then the technical complexity of rotation might be worth it. Perhaps not the trapeze act that Dan has proposed, but certainly the cost of adding extra sumps/vents at the required points in the tankers, depot, and LSS.
Yes, I was shocked by the amount of propellant needed to use acceleration for settling, so I started investigating rotation again. I was then shocked again by the very small amount of energy needed to use centripetal acceleration. The "trapeze act" is a whacko thought experiment, not really a proposal. To actually implement it, you would probably need an entire separate specialized SS to carry it, deploy it, and then act as the counterweight, roughly doubling the cost of the Depot.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/18/2023 04:36 am
I dont wanna be rude or anything. Any rotation is quite stupid way to overengineer.
Linear acceleration is way to go. Its well know phenomena.

The fall back to centripetal settling is a result of the calcs for the amount of prop required for ullage burns during transfer. If you have to burn an entire extra tanker load of propellant just to do prop settling, then the technical complexity of rotation might be worth it. Perhaps not the trapeze act that Dan has proposed, but certainly the cost of adding extra sumps/vents at the required points in the tankers, depot, and LSS.
Yes, I was shocked by the amount of propellant needed to use acceleration for settling, so I started investigating rotation again. I was then shocked again by the very small amount of energy needed to use centripetal acceleration. The "trapeze act" is a whacko thought experiment, not really a proposal. To actually implement it, you would probably need an entire separate specialized SS to carry it, deploy it, and then act as the counterweight, roughly doubling the cost of the Depot.

It's a shocking amount if you:

1) Use high ullage acceleration.
2) Use very low transfer flow rates.
3) Use cold gas ullage thrusters instead of combusting gas.

All three of these are interrelated.  If there are geysering or splashing problems, then higher acceleration or lower flow rates can mitigate them.  If that combo requires inefficient use of ullage prop, then going from cold gas to combustion gas dramatically increases the efficiency.

I'm still trying to figure out if my reading of the Poiseuille equation means that you can always avoid geysering.  If so, then low transfer rates are dumb, and high transfer rates are easy.  Then the only issue is whether some other kind of vibration or attitude dynamics can cause either the sending or receiving tanks to become unsettled.  If that's the case, then you need higher acceleration, which likely requires higher Isp, which requires combusting gas.

But a combustion thruster is soooooo much easier to implement than a rotating system.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/18/2023 04:50 am
I wonder if it would be possible to plumb fluid transfer in via the header tanks? With their small enclosed volume, you can pump props in as fast as you can and allow the violent sloshing and geysering to occur inside them. The headers already have more extensive baffling due to the need to minimise slosh during the flip manoeuvre, so even with no additional fluid management hardware and just an open line to the main tank allowing mixed phase flow, a huge amount of energy in the incoming propellant flow should be expended inside the header allowing for gentler flow into the main tank.

It might help on the sending side--if there's actually a problem on the sending side.  You don't want the pump to become unprimed by sucking in gas, but in general, the sending side should be more placid.

If you're using equalized pressure, I guess the main cause of unsettled conditions on the sending side would be... sucking liquid through the equalization line and dropping it into the settled prop in the sending tank.  So everything really comes back to ensuring that the receiving side stays settled enough not to suck liquid through the equalization line.

I don't think using the header helps on the receiving side.  If you receive prop into a header tank, then the header has to pump it into the main, where it will splash, which will still cause prop to get into the gas pressure equalization line.

Maybe locate the equalization lines in the header tanks?  It's a pretty hardy blob of prop that can get sucked from the main, into the header, and then into the inter-vehicle equalization.

Another possibility is simply to cover the equalization line with some kind of prop management device that prevents blobs from making it into the line, even if there's splashing.  Seems that this ought to be a pretty tractable problem.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/18/2023 11:18 am
That assumes presence of an equalisation line. Initial transfer (dry receiver) - vent the incoming tank to vacuum, close the vent, and transfer prop from the pressurised sender tank (can use ullage pressure, likely no pump required*). As there is no ullage volume to compress, you can geyser and spray and slosh and vaporise as much as you want with minimal concern for a significant portion of the transfer. Eventually enough prop will boil from this vigorous action and reduced droplet size (and thus increased surface area) for autogenous pressurisation to occur and raise the pressure to match the sending tank, and at that point transfer pauses to allow for prop settling. Once prop has settled, the tank is vented to near-vacuum (low enough to minimise static boiloff, IIRC around 10kPa for CH4 and 0.15kPa for O2) and transfer can continue with the outlet well covered and the ullage vented as required to maintain that low pressure in the receiving tank. This state persists for subsequent propellant loadings. This loses some amount of gaseous prop from venting back down from equilibrium to the target low pressure, but that mass loss is relatively small: e.g. for CH4 assuming a 3 Bar sender tank pressure and the absolute worst case of the entire tank volume to be vented, the maximum mass loss is around 1.5 tonnes. That gives an upper limit for any anti-geysering hardware dry mass or low-geysering-transfer propellant consumption.


*For a 9m diameter tank, every bar of pressure the sending tank is above the receiving tank means roughly 6 MN of force trying to push prop from the sender to receiver tank.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 01/18/2023 11:27 am
.This loses some amount of gaseous prop from venting back down from equilibrium to the target low pressure, but that mass loss is relatively small: e.g. for CH4 assuming a 3 Bar sender tank pressure and the absolute worst case of the entire tank volume to be vented, the maximum mass loss is around 1.5 tonnes.

Except that's not the "absolute worst case," because (as you point out) liquid propellant will be constantly evaporating to replenish the ullage gas.

It's also not the "absolute worst case" because we don't know if it will actually reliably transfer 100% of the propellant. If some is missed that needs to be counted, and even a small amount of liquid could exceed 1.5 tonnes.

You also missed the ullage thrusters. If the technique works but takes a lot longer, then that additional ullage prop mass should also be accounted for.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/18/2023 11:44 am
.This loses some amount of gaseous prop from venting back down from equilibrium to the target low pressure, but that mass loss is relatively small: e.g. for CH4 assuming a 3 Bar sender tank pressure and the absolute worst case of the entire tank volume to be vented, the maximum mass loss is around 1.5 tonnes.

Except that's not the "absolute worst case," because (as you point out) liquid propellant will be constantly evaporating to replenish the ullage gas.

It's also not the "absolute worst case" because we don't know if it will actually reliably transfer 100% of the propellant. If some is missed that needs to be counted, and even a small amount of liquid could exceed 1.5 tonnes.
Only when in a vacuum state. The vents would not be opened again until the tank has reached pressure equilibrium with the sender tank, and then settles, so venting would not lose liquids (settled so no mixed phase) and pressure would not be allowed to drop before the triple point (so no flash boiling).
Quote
You also missed the ullage thrusters. If the technique works but takes a lot longer, then that additional ullage prop mass should also be accounted for.
By transferring at a high flow rate and then settling with a sealed tank, you can minimise the time at the maximum settling thrust needed (to keep the inlet covered on the sender tank) and then switch to the minimum thrust needed for the post-transfer sealed tank settling (equivalent to a cost-phase settling thrust). This is opposed to needing to keep sufficient thrust to both keep the sender tank inlet covered and keep the receiving tank from geysering for the entire transfer duration.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/18/2023 06:01 pm
.This loses some amount of gaseous prop from venting back down from equilibrium to the target low pressure, but that mass loss is relatively small: e.g. for CH4 assuming a 3 Bar sender tank pressure and the absolute worst case of the entire tank volume to be vented, the maximum mass loss is around 1.5 tonnes.

Except that's not the "absolute worst case," because (as you point out) liquid propellant will be constantly evaporating to replenish the ullage gas.

It's also not the "absolute worst case" because we don't know if it will actually reliably transfer 100% of the propellant. If some is missed that needs to be counted, and even a small amount of liquid could exceed 1.5 tonnes.
Only when in a vacuum state. The vents would not be opened again until the tank has reached pressure equilibrium with the sender tank, and then settles, so venting would not lose liquids (settled so no mixed phase) and pressure would not be allowed to drop before the triple point (so no flash boiling).
Quote
You also missed the ullage thrusters. If the technique works but takes a lot longer, then that additional ullage prop mass should also be accounted for.
By transferring at a high flow rate and then settling with a sealed tank, you can minimise the time at the maximum settling thrust needed (to keep the inlet covered on the sender tank) and then switch to the minimum thrust needed for the post-transfer sealed tank settling (equivalent to a cost-phase settling thrust). This is opposed to needing to keep sufficient thrust to both keep the sender tank inlet covered and keep the receiving tank from geysering for the entire transfer duration.
The sender, depot or tanker, will need to retain enough propellant to maneuver away from the receiver. In addition, a tanker needs props to leave orbit and the depot need enough so it doesn't boil dry - maybe until the next campaign. With some judicious baffling there shouldn't be a problem with exposing the outlet.



Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 01/18/2023 06:33 pm
.This loses some amount of gaseous prop from venting back down from equilibrium to the target low pressure, but that mass loss is relatively small: e.g. for CH4 assuming a 3 Bar sender tank pressure and the absolute worst case of the entire tank volume to be vented, the maximum mass loss is around 1.5 tonnes.

Except that's not the "absolute worst case," because (as you point out) liquid propellant will be constantly evaporating to replenish the ullage gas.

It's also not the "absolute worst case" because we don't know if it will actually reliably transfer 100% of the propellant. If some is missed that needs to be counted, and even a small amount of liquid could exceed 1.5 tonnes.
Only when in a vacuum state. The vents would not be opened again until the tank has reached pressure equilibrium with the sender tank, and then settles, so venting would not lose liquids (settled so no mixed phase) and pressure would not be allowed to drop before the triple point (so no flash boiling).
Quote
You also missed the ullage thrusters. If the technique works but takes a lot longer, then that additional ullage prop mass should also be accounted for.
By transferring at a high flow rate and then settling with a sealed tank, you can minimise the time at the maximum settling thrust needed (to keep the inlet covered on the sender tank) and then switch to the minimum thrust needed for the post-transfer sealed tank settling (equivalent to a cost-phase settling thrust). This is opposed to needing to keep sufficient thrust to both keep the sender tank inlet covered and keep the receiving tank from geysering for the entire transfer duration.
The sender, depot or tanker, will need to retain enough propellant to maneuver away from the receiver. In addition, a tanker needs props to leave orbit and the depot need enough so it doesn't boil dry - maybe until the next campaign. With some judicious baffling there shouldn't be a problem with exposing the outlet.
Tankers will presumably have header tanks for their own needs.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 01/18/2023 06:38 pm
What do people mean when they talk about geysering? Cryogenic geysering is as far as I know related to cryogenic liquids in long pipes under enough acceleration and external heating that convection becomes important. I am not sure how this applies to millig propellant transfer?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/18/2023 06:39 pm
.This loses some amount of gaseous prop from venting back down from equilibrium to the target low pressure, but that mass loss is relatively small: e.g. for CH4 assuming a 3 Bar sender tank pressure and the absolute worst case of the entire tank volume to be vented, the maximum mass loss is around 1.5 tonnes.

Except that's not the "absolute worst case," because (as you point out) liquid propellant will be constantly evaporating to replenish the ullage gas.

It's also not the "absolute worst case" because we don't know if it will actually reliably transfer 100% of the propellant. If some is missed that needs to be counted, and even a small amount of liquid could exceed 1.5 tonnes.
Only when in a vacuum state. The vents would not be opened again until the tank has reached pressure equilibrium with the sender tank, and then settles, so venting would not lose liquids (settled so no mixed phase) and pressure would not be allowed to drop before the triple point (so no flash boiling).
Quote
You also missed the ullage thrusters. If the technique works but takes a lot longer, then that additional ullage prop mass should also be accounted for.
By transferring at a high flow rate and then settling with a sealed tank, you can minimise the time at the maximum settling thrust needed (to keep the inlet covered on the sender tank) and then switch to the minimum thrust needed for the post-transfer sealed tank settling (equivalent to a cost-phase settling thrust). This is opposed to needing to keep sufficient thrust to both keep the sender tank inlet covered and keep the receiving tank from geysering for the entire transfer duration.
The sender, depot or tanker, will need to retain enough propellant to maneuver away from the receiver. In addition, a tanker needs props to leave orbit and the depot need enough so it doesn't boil dry - maybe until the next campaign. With some judicious baffling there shouldn't be a problem with exposing the outlet.
Tankers will presumably have header tanks for their own needs.
Is the header used for deorbit? Thought it was only used post fireball.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/18/2023 06:43 pm
What do people mean when they talk about geysering? Cryogenic geysering is as far as I know related to cryogenic liquids in long pipes under enough acceleration and external heating that convection becomes important. I am not sure how this applies to millig propellant transfer?
Rad Mod came up with this to describe the impact of too high a transfer pressure on the receiver. If the inlet plume punches through the settled propellant it would be geyser like.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/18/2023 06:46 pm
That assumes presence of an equalisation line. Initial transfer (dry receiver) - vent the incoming tank to vacuum, close the vent, and transfer prop from the pressurised sender tank (can use ullage pressure, likely no pump required*). As there is no ullage volume to compress, you can geyser and spray and slosh and vaporise as much as you want with minimal concern for a significant portion of the transfer...

That only solves half the problem, though.  The sending tank's pressure will be dropping, increasing the back pressure on the system.  To solve that, you either need to have really big pumps to drop the sending tank's ullage pressure so much that the prop pool boils, or you need a heater to boil prop on the sending side to keep the pressure up. Neither of these is an efficient solution, either energy-wise nor prop efficiency-wise.

In contrast, the equalization line doesn't require any more gas to be wasted as ullage, needs tiny little pumps, doesn't require heaters in the mains, and is less complex in general.  The geysering/splash problem is likely a real one, but if it can't be solved with a bit of clever configuration of the pump outlet, then just putting a PMD around the equalization inlet probably solves it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/18/2023 07:09 pm
Ok. Now I get it. Didn't think it through.


Hmmm. If the pump hits exactly at ambient pressure at the top of the inlet, wouldn't the flow stall out? I'm thinking the column of fluid would just make it to the lip of the inlet and no further. One tiny smidge more and it overflows the lip and fills the sump. Once the tank level reaches the top of the inlet it would continue to rise about one smidge worth, then stall out again. How many kWh pump energy in a smidge?


From a practical point of view what is the relationship between pressure and flow rate in a centrifugal pump? If it's pumping into a closed off pipe it's at high pressure but zero flow. Valve the outlet open just a tad (first cousin to a smidge) and flow starts but pressure drops. Once the outlet is fully open both pressure and volume rise and fall with RPM but is it a linear relationship? My gut says it's close. Does the relationship change facing different back pressures?


Am I overthinking this? ???


A random thought. If the volume and pressure are directly related, the transfer op would counterintuitively go faster at 6bar ullage pressure than at .5bar. Would it be fast enough to materially impact propellant consumption? Would the amount of makeup gas from the high pressure COPV's negate any advantage? The makeup gas will be warm and would contract in the tanks. Electric heater on the COPV outlets?


My brain is starting to hurt.

Yeah, it's pretty counterintuitive, and I'm not sure I've understood it correctly.  I'm wondering if Poseuille only applies to static pressure, but that would be weird with an equation that has volumetric flow built into it.  But if you've got the same volumetric flow and it's incompressible, it's hard to see how there isn't dynamic pressure just from the momentum of the flow itself.

Hopefully somebody will chime in and explain it all.
I've been noodling this from the pumps POV. In looking at the relationship between volume and pressure I could not find a single practical constant to serve as an anchor.


One extreme is pumping against a closed valve and the other extreme is the valve open to an infinite vacuum.


Against a closed valve there would be max pressure and zero flow. Depending on the motor design there would be either a limitation on RPM, a high current draw, or some combination of both.


Opening the valve into an infinite vacuum, the pressure would be minimum, the flow maximum and again depending on motor design, an increase in RPM, a drop in current draw, or some combination.


Current draw could be held constant as could RPM but the two seem to be so intermingled I can't figure out what I'm modeling let alone the practical implications. Maybe assume a synchronous, constant RPM AC motor for modeling despite the DC motor that would probably be used in the ship? Doesn't feel useful.


Yeah, we need a pump engineer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 01/18/2023 07:20 pm
I personally do not expect to see gas transfer/equilization lines. If they do pressure/temperature driven pumping (which I currently expect) then there is obviously no need and for electrical pumping I do not think it is worth the hassle of extra outlets, lines and connections.

If you build up to much pressure differential you let the propellants slosh a bit until it decreases enough and then you continue. If you are in a hurry you might trade a bit of venting against extra RCS propellant.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 01/18/2023 07:26 pm
Yeah, we need a pump engineer.
...or maybe a plumber.  :)

In the mundane world of water piping, a pump is characterized by its flow rate against a particular backpressure. Backpressue is usually quoted in "head", which is the height of a water column.   Plumbers compute the expected backpressure of a pipe system by summing the piecewise backpressures, where each fitting has a backpressure spec and there is a forumula to compute the backpressure of a length of pipe based on its length and diameter.

Clearly the backpressures will be different for other fluids.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Nevyn72 on 01/18/2023 08:48 pm
So what is easier to pump, a cryogenic liquid or a gas?

Is it more efficient to pump the liquid from the tanker to the depot and let ullage gas flow from the depot and return to the tanker to balance volumes?

Or is it better to pump ullage gas from the depot to the tanker and have that 'push' the liquid through?

Note: this presupposes the use of a gas equalisation line.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 01/18/2023 09:04 pm
So what is easier to pump, a cryogenic liquid or a gas?

Is it more efficient to pump the liquid from the tanker to the depot and let ullage gas flow from the depot and return to the tanker to balance volumes?

Or is it better to pump ullage gas from the depot to the tanker and have that 'push' the liquid through?

Note: this presupposes the use of a gas equalisation line.
To first order pumping power is volumetric flow times pressure differential. Gases have ~1000 times larger volume...

But as I stated previously I am not sure you need to move the gas with autogenous pressurization. Once you get past the transients the gas pressure is the vapor pressure at the temperature of the liquid. If you compress the gas on one side it will start condensing into the liquid while an expanding gas volume will be replenished by the liquid boiling. Since the mass of the liquid is generally much higher than that of the ullage gas the resulting temperature changes are relatively small as long as everything mixes well. Splashing and sloshing would be beneficial as long as the liquid outlet is submerged.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/18/2023 10:12 pm
I personally do not expect to see gas transfer/equilization lines. If they do pressure/temperature driven pumping (which I currently expect) then there is obviously no need and for electrical pumping I do not think it is worth the hassle of extra outlets, lines and connections.

If you build up to much pressure differential you let the propellants slosh a bit until it decreases enough and then you continue. If you are in a hurry you might trade a bit of venting against extra RCS propellant.
That last is the detail where the devil resides. How much venting? How much settling propellant? How long to transfer?


No ullage crossover and using thermal/pressure to power the transfer is mechanically simplest and avoids risk of recirculating splatter, but is it a path to acceptable propellant consumption? I really have not a clue.


There are times I wish I were more of a numbers guy.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/19/2023 12:43 pm
That assumes presence of an equalisation line. Initial transfer (dry receiver) - vent the incoming tank to vacuum, close the vent, and transfer prop from the pressurised sender tank (can use ullage pressure, likely no pump required*). As there is no ullage volume to compress, you can geyser and spray and slosh and vaporise as much as you want with minimal concern for a significant portion of the transfer...

That only solves half the problem, though.  The sending tank's pressure will be dropping, increasing the back pressure on the system.  To solve that, you either need to have really big pumps to drop the sending tank's ullage pressure so much that the prop pool boils, or you need a heater to boil prop on the sending side to keep the pressure up. Neither of these is an efficient solution, either energy-wise nor prop efficiency-wise.
Starting from a 0.1 Bar receiver tank and a 3 Bar sender tank, with the sender tank 2/3 full of prop (1/3 ullage volume) and the receiver tank empty of prop, the entire fluid propellant volume could be transferred to the receiver without any pumping or venting (final state: receiver 2/3 full with prop, 0.3 Bar ullage pressure, sender filled entirely with with 1 Bar ullage).
A worse case would be a sender tank almost entirely full of fluid prop with only a small ullage head volume, attempting to fill an empty tank: the equilibrium state with the maximum fluid transfer volume (assuming valves shut at ullage pressure equilibrium to avoid oscillation, the same starting pressures as previously, and tanks of equal volume) would mean that you can have no less than ~20% of the sender tank as ullage volume to complete a full fluid transfer from replenished ullage pressure alone (final equilibrium stage being the sender at 0.6 Bar all ullage, and the receiver 80% full of liquid prop with the ullage at 0.5 Bar).
The key trick is that all one needs to do to transfer more than that 80% capacity that is to vent the receiver again, and you now once again have a pressure differential to work with to complete transfer.

Quote
In contrast, the equalization line doesn't require any more gas to be wasted as ullage, needs tiny little pumps, doesn't require heaters in the mains, and is less complex in general.  The geysering/splash problem is likely a real one, but if it can't be solved with a bit of clever configuration of the pump outlet, then just putting a PMD around the equalization inlet probably solves it.
Whether you pump fluids directly or pump the ullage gas, it still requires a pump to be added, along with a power supply for that pump (power that could also be used for boiling prop to replenish ullage gas). Pumping the ullage gas is also not as simple as it appears, as for the same volumetric gas flow rate you do not get the same flow of liquid prop. It also requires the propellants in the receiver tank be settles and stratified during the transfer process or else the ullage recirc line will ingest liquids as well as gas (resulting in prop that needs to be resettled and repumped), whereas transfer to a sealed receiver allows for slosh and geysering during the transfer process, only the sender tank needs to keep the outlet covered with liquid prop.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: livingjw on 01/19/2023 03:24 pm
Is geysering during propellant transfer really a problem in space?  Heat transfer will be small, transfer pump pressures are low and there is no gravity to accelerate the vapor. Just trying to get a handle on the concerns.

John
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 01/19/2023 05:56 pm
Is geysering during propellant transfer really a problem in space?  Heat transfer will be small, transfer pump pressures are low and there is no gravity to accelerate the vapor. Just trying to get a handle on the concerns.

John
I thought so too  :) Unfortunate word choice:
What do people mean when they talk about geysering? Cryogenic geysering is as far as I know related to cryogenic liquids in long pipes under enough acceleration and external heating that convection becomes important. I am not sure how this applies to millig propellant transfer?
Rad Mod came up with this to describe the impact of too high a transfer pressure on the receiver. If the inlet plume punches through the settled propellant it would be geyser like.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/19/2023 07:36 pm
That assumes presence of an equalisation line. Initial transfer (dry receiver) - vent the incoming tank to vacuum, close the vent, and transfer prop from the pressurised sender tank (can use ullage pressure, likely no pump required*). As there is no ullage volume to compress, you can geyser and spray and slosh and vaporise as much as you want with minimal concern for a significant portion of the transfer...

That only solves half the problem, though.  The sending tank's pressure will be dropping, increasing the back pressure on the system.  To solve that, you either need to have really big pumps to drop the sending tank's ullage pressure so much that the prop pool boils, or you need a heater to boil prop on the sending side to keep the pressure up. Neither of these is an efficient solution, either energy-wise nor prop efficiency-wise.
Starting from a 0.1 Bar receiver tank and a 3 Bar sender tank, with the sender tank 2/3 full of prop (1/3 ullage volume) and the receiver tank empty of prop, the entire fluid propellant volume could be transferred to the receiver without any pumping or venting (final state: receiver 2/3 full with prop, 0.3 Bar ullage pressure, sender filled entirely with with 1 Bar ullage).
A worse case would be a sender tank almost entirely full of fluid prop with only a small ullage head volume, attempting to fill an empty tank: the equilibrium state with the maximum fluid transfer volume (assuming valves shut at ullage pressure equilibrium to avoid oscillation, the same starting pressures as previously, and tanks of equal volume) would mean that you can have no less than ~20% of the sender tank as ullage volume to complete a full fluid transfer from replenished ullage pressure alone (final equilibrium stage being the sender at 0.6 Bar all ullage, and the receiver 80% full of liquid prop with the ullage at 0.5 Bar).
The key trick is that all one needs to do to transfer more than that 80% capacity that is to vent the receiver again, and you now once again have a pressure differential to work with to complete transfer.

Quote
In contrast, the equalization line doesn't require any more gas to be wasted as ullage, needs tiny little pumps, doesn't require heaters in the mains, and is less complex in general.  The geysering/splash problem is likely a real one, but if it can't be solved with a bit of clever configuration of the pump outlet, then just putting a PMD around the equalization inlet probably solves it.
Whether you pump fluids directly or pump the ullage gas, it still requires a pump to be added, along with a power supply for that pump (power that could also be used for boiling prop to replenish ullage gas). Pumping the ullage gas is also not as simple as it appears, as for the same volumetric gas flow rate you do not get the same flow of liquid prop. It also requires the propellants in the receiver tank be settles and stratified during the transfer process or else the ullage recirc line will ingest liquids as well as gas (resulting in prop that needs to be resettled and repumped), whereas transfer to a sealed receiver allows for slosh and geysering during the transfer process, only the sender tank needs to keep the outlet covered with liquid prop.
It looks like your model looks only at the ullage pressure and ignores plumbing losses and head from increasing liquid depth in the receiver. It's a starting point, IIUC.


The worst case would be a 1350t capacity tanker with 150t of propellant pumping the last load into a 1600t capacity depot. IIRC you are a proponent of using a standard tanker as a depot and not stretching the tankers beyond 1200t. Don't remember if you advocate for keeping the tanker at 1200t and separate tanks in the cargo space.


As much as I dislike separate tanks for delivery, the ullage thrust/consumption problem might be a strong argument in its favor.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/19/2023 08:40 pm
If you build up to much pressure differential you let the propellants slosh a bit until it decreases enough and then you continue. If you are in a hurry you might trade a bit of venting against extra RCS propellant.
That last is the detail where the devil resides. How much venting? How much settling propellant? How long to transfer?

No ullage crossover and using thermal/pressure to power the transfer is mechanically simplest and avoids risk of recirculating splatter, but is it a path to acceptable propellant consumption? I really have not a clue.

I think I know why the Poiseuille results are so non-intuitive:  they don't include dynamic pressure.  They're a solution for a flow in equilibrium, but the non-equilibrium case requires the volumetric flow to be established.  So the dynamic pressure of the flow can still cause geysering.

So, two choices for limiting geysering:

1) Limit geysering by starting the pumping operation at low power/high acceleration, then gradually increase power and decrease acceleration as the pool fills.  The pool will then provide viscous damping as the flow rate inceases.

2) Assuming that only some small amount of prop will blob up and hit the equalization inlet, let it.  Soon enough, the pool will be deep enough that viscous damping will prevent geysering.  You wind up wasting a bit of energy re-pumping prop that recirculated, but it's probably a trivial amount.

A final note on how to limit recirculation:  For ullage acceleration a and geyser velocity vg,  a blob will fall back into the pool before hitting the equalization inlet if the height h between the top of the pool and the inlet is greater than vg²/(2a).  This is really just a limiting condition on how much geysering you can tolerate.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/19/2023 09:30 pm
Is geysering during propellant transfer really a problem in space?  Heat transfer will be small, transfer pump pressures are low and there is no gravity to accelerate the vapor. Just trying to get a handle on the concerns.

John

I don't know.  Assuming that there's actually a pump, then there are two potential problems:

1) So many free-floating blobs in the receiver get sucked into the pressure equalization line between the tanks that transfer efficiency drops asymptotically to zero.  I think this is unlikely, unless the pool viscosity is so low that geysering can occur in an almost-full tank, when the geyser and the equalization inlet can be close together.

Solutions:
a) Lower the outlet flow when the receiver pool is shallow, then increase it as viscosity and pool depth conspire to eliminate the geysering.
b) Put the opposite of a PMD around the equalization inlet, designed to guid the occasional blob away from the inlet and down the walls, back to the main pool

2) Free-floating blobs crash into the tank walls with enough momentum to set up some sort of vibration/resonance in the system, causing slosh in the sending tank, which eventually leads to the outlet being uncovered.

Solutions:
a) Create reasonable baffles.
b) If liquid can come through the equalization outlet, make sure it doesn't splash into the sending pool.
c) Deal with the fact that the outlet will occasionally get uncovered, and make sure that the pump can re-prime itself. 

Note:  a self-priming pump may be annoying, but it's nothing like the amount of annoyance that'll occur if a pressure-fed system has the same problem and uncovers the inlet.  Then you're talking about having to re-establish the pressure differential between tanks, which could take hours.

I'm inclined to think that the geysering problem is something that needs some engineering applied to it, but that it's far from an insurmountable problem.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 01/19/2023 09:44 pm
Is geysering during propellant transfer really a problem in space?  Heat transfer will be small, transfer pump pressures are low and there is no gravity to accelerate the vapor. Just trying to get a handle on the concerns.

John

I don't know.  Assuming that there's actually a pump, then there are two potential problems:

1) So many free-floating blobs in the receiver get sucked into the pressure equalization line between the tanks that transfer efficiency drops asymptotically to zero.  I think this is unlikely, unless the pool viscosity is so low that geysering can occur in an almost-full tank, when the geyser and the equalization inlet can be close together.

Solutions:
a) Lower the outlet flow when the receiver pool is shallow, then increase it as viscosity and pool depth conspire to eliminate the geysering.
b) Put the opposite of a PMD around the equalization inlet, designed to guid the occasional blob away from the inlet and down the walls, back to the main pool

2) Free-floating blobs crash into the tank walls with enough momentum to set up some sort of vibration/resonance in the system, causing slosh in the sending tank, which eventually leads to the outlet being uncovered.

Solutions:
a) Create reasonable baffles.
b) If liquid can come through the equalization outlet, make sure it doesn't splash into the sending pool.
c) Deal with the fact that the outlet will occasionally get uncovered, and make sure that the pump can re-prime itself. 

Note:  a self-priming pump may be annoying, but it's nothing like the amount of annoyance that'll occur if a pressure-fed system has the same problem and uncovers the inlet.  Then you're talking about having to re-establish the pressure differential between tanks, which could take hours.

I'm inclined to think that the geysering problem is something that needs some engineering applied to it, but that it's far from an insurmountable problem.
One problem is that cryogenic geysering is an existing well known fenomen that can be a problem for (or even destroy) cryogenic piping on launch vehicles and tank farms. You are talking about the tank inlet jet splashing while filling a tank.

https://youtu.be/gpzO4b-02ZE
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/20/2023 01:19 am
One problem is that cryogenic geysering is an existing well known fenomen that can be a problem for (or even destroy) cryogenic piping on launch vehicles and tank farms. You are talking about the tank inlet jet splashing while filling a tank.

Fair enough.  Let's call it... fountaining?  Something that implies that mass flow from the inlet is coming out with sufficient momentum to break the surface tension of the existing inlet pool.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/20/2023 06:40 am
Or maybe just 'spraying', because aerosolisation is the specific problem to avoid. A nice laminar fountain isn't really a problem, its only once that flow breaks up into small high surface area droplets that it becomes an issue (for a tank with an open inlet and outlet that assumes stratified phases).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/20/2023 06:15 pm
If you build up to much pressure differential you let the propellants slosh a bit until it decreases enough and then you continue. If you are in a hurry you might trade a bit of venting against extra RCS propellant.
That last is the detail where the devil resides. How much venting? How much settling propellant? How long to transfer?

No ullage crossover and using thermal/pressure to power the transfer is mechanically simplest and avoids risk of recirculating splatter, but is it a path to acceptable propellant consumption? I really have not a clue.

I think I know why the Poiseuille results are so non-intuitive:  they don't include dynamic pressure.  They're a solution for a flow in equilibrium, but the non-equilibrium case requires the volumetric flow to be established.  So the dynamic pressure of the flow can still cause geysering.

So, two choices for limiting geysering:

1) Limit geysering by starting the pumping operation at low power/high acceleration, then gradually increase power and decrease acceleration as the pool fills.  The pool will then provide viscous damping as the flow rate inceases.

2) Assuming that only some small amount of prop will blob up and hit the equalization inlet, let it.  Soon enough, the pool will be deep enough that viscous damping will prevent geysering.  You wind up wasting a bit of energy re-pumping prop that recirculated, but it's probably a trivial amount.

A final note on how to limit recirculation:  For ullage acceleration a and geyser velocity vg,  a blob will fall back into the pool before hitting the equalization inlet if the height h between the top of the pool and the inlet is greater than vg²/(2a).  This is really just a limiting condition on how much geysering you can tolerate.
If splatter ingestion is a problem something like the float check valve in a some snorkels might be he answer. Misting would get by but a glob of props would shut it for a moment.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/22/2023 09:59 pm
If splatter ingestion is a problem something like the float check valve in a some snorkels might be he answer. Misting would get by but a glob of props would shut it for a moment.

Kinda hard to use a float check valve in a system where there's essentially no buoyancy.

I really don't think this is an intractable problem.  The way we fell into this rabbit hole was as a result of an analysis of what it took to use cold gas thrusters for ullage acceleration.  That's a "Doctor, Doctor, it hurts when I do this" kind of problem: don't do that. 

With adequate acceleration, you can use high transfer rates while mitigating fountaining, splash, splatter, whatever you want to call it.  High transfer rates and decent thruster Isp result in extremely modest prop losses for ullage thrust.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/24/2023 12:55 am
If splatter ingestion is a problem something like the float check valve in a some snorkels might be he answer. Misting would get by but a glob of props would shut it for a moment.

Kinda hard to use a float check valve in a system where there's essentially no buoyancy.

I really don't think this is an intractable problem.  The way we fell into this rabbit hole was as a result of an analysis of what it took to use cold gas thrusters for ullage acceleration.  That's a "Doctor, Doctor, it hurts when I do this" kind of problem: don't do that. 

With adequate acceleration, you can use high transfer rates while mitigating fountaining, splash, splatter, whatever you want to call it.  High transfer rates and decent thruster Isp result in extremely modest prop losses for ullage thrust.

To be fair, that wasn’t the only reason we fell into this rabbit hole, although I realise I was concerned about a different phenomena than the fountaining/sloshing effect we’re now talking about. That will indeed go away as the depot tanks fill up. I’m talking about the “induced swirling” effect Dan and I were thought experimenting on (that could get worse as the depot fills up).

I had to go back and find this old Dyson air multiplier video, because this shows the “entrained air” concept that I was referencing.

https://youtu.be/Ha1z4juJbYc

Depending how fast you have to pump the propellant (i.e. proportional to how much prop you expend in milli-G acceleration), you might get to a situation where you have to slow down as the depot tank gets full to avoid creating induced vortices in the depot, which could have large angular momentum and torque the whole system around.

Maybe I’m over egging this, but it seems to me that unless you’re going to fill the whole prop tank with baffles, there will be some threshold flow rate over which it’s not safe to pump.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/24/2023 01:52 am
If splatter ingestion is a problem something like the float check valve in a some snorkels might be he answer. Misting would get by but a glob of props would shut it for a moment.

Kinda hard to use a float check valve in a system where there's essentially no buoyancy.

I really don't think this is an intractable problem.  The way we fell into this rabbit hole was as a result of an analysis of what it took to use cold gas thrusters for ullage acceleration.  That's a "Doctor, Doctor, it hurts when I do this" kind of problem: don't do that. 

With adequate acceleration, you can use high transfer rates while mitigating fountaining, splash, splatter, whatever you want to call it.  High transfer rates and decent thruster Isp result in extremely modest prop losses for ullage thrust.
If we take 6 hours as a transfer target, and I haven't boogered the numbers, the 120t of LOX needs to move at a rate of 4.9l/sec and 30t of liquid methane at 3.3l/sec.


Do we have a pixel count of the QD inner diameters? I can't think of a reason to not expect the diameter to carry through the system. If the diameter is 10cm (WAG) we need an average flow velocity of 156cm/sec for the LOX and 105cm/sec for the methane.


That's about my limit on numbers and they really need to be verified. The target transfer time is a bit arbitrary and the plumbing diameter needs verification. Between this and the viscosity numbers, is there enough to rough in the g force needed to avoid geysering erupting starting with an empty depot and ending with completion of a 150t transfer, then at completion of each successive transfer?


From this we can probably do a sanity check on propellant needs and/or ullage pressures and volumes for the various settling schemes.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/24/2023 02:55 am
If splatter ingestion is a problem something like the float check valve in a some snorkels might be he answer. Misting would get by but a glob of props would shut it for a moment.

Kinda hard to use a float check valve in a system where there's essentially no buoyancy.

I really don't think this is an intractable problem.  The way we fell into this rabbit hole was as a result of an analysis of what it took to use cold gas thrusters for ullage acceleration.  That's a "Doctor, Doctor, it hurts when I do this" kind of problem: don't do that. 

With adequate acceleration, you can use high transfer rates while mitigating fountaining, splash, splatter, whatever you want to call it.  High transfer rates and decent thruster Isp result in extremely modest prop losses for ullage thrust.

To be fair, that wasn’t the only reason we fell into this rabbit hole, although I realise I was concerned about a different phenomena than the fountaining/sloshing effect we’re now talking about. That will indeed go away as the depot tanks fill up. I’m talking about the “induced swirling” effect Dan and I were thought experimenting on (that could get worse as the depot fills up).

I had to go back and find this old Dyson air multiplier video, because this shows the “entrained air” concept that I was referencing.

https://youtu.be/Ha1z4juJbYc (https://youtu.be/Ha1z4juJbYc)

Depending how fast you have to pump the propellant (i.e. proportional to how much prop you expend in milli-G acceleration), you might get to a situation where you have to slow down as the depot tank gets full to avoid creating induced vortices in the depot, which could have large angular momentum and torque the whole system around.

Maybe I’m over egging this, but it seems to me that unless you’re going to fill the whole prop tank with baffles, there will be some threshold flow rate over which it’s not safe to pump.
FWIW, because of the methane downcomer transfer tube the propellant inlets will be off center. This doesn't negate vortex formation but I think it would keep it from being a big concentric tank enveloping swirl. Add a 'T' fitting and there would either be two counter swirls or chaotic movement - I think. If that's not enough, a flapper valve in the outlet T could periodically reverse the output direction and nip a big vortex in the bud.


Or, as you say, some max flow rate that the above measures might raise.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/24/2023 05:01 am
To be fair, that wasn’t the only reason we fell into this rabbit hole, although I realise I was concerned about a different phenomena than the fountaining/sloshing effect we’re now talking about. That will indeed go away as the depot tanks fill up. I’m talking about the “induced swirling” effect Dan and I were thought experimenting on (that could get worse as the depot fills up)...

Maybe I’m over egging this, but it seems to me that unless you’re going to fill the whole prop tank with baffles, there will be some threshold flow rate over which it’s not safe to pump.

I don't think you have to fill the whole tank with baffles; you just need enough of them to generate turbulence before any kind of organized laminar flow builds up.  It's not often that tip vortices are your friend, but this is one time they are.

You could probably cage the outlet (inlet?  the place where stuff comes into the receiving tank) with something to break up the flow even more.  Note that I'm implicitly assuming that fill/drain is not the same as the downcomers.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/24/2023 05:16 am
If we take 6 hours as a transfer target, and I haven't boogered the numbers, the 120t of LOX needs to move at a rate of 4.9l/sec and 30t of liquid methane at 3.3l/sec.

That's the kind of time scale where prop losses from ullage thrust are gonna hurt you pretty badly.

I was initially thinking that half an hour would be about right for 150t of prop, but even that might be too leisurely.  Remember that the worst case is actually when the depot transfers all of its prop to a payload Starship.  And you can wind up with orbital constraints on how long you can take, especially if you're doing high-cadence tanker ops where you want to de-orbit for quick turnaround, or if you have a depot that's transferring its whole load in an HEEO.

If fountaining is a non-issue with any appreciable pool depth in the receiving tank, then you should start off slow but ramp up to >100kg/s as soon as possible.  Remember, in microgravity, your pump power can be tiny, even for high flow rates.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/24/2023 06:34 am
If we take 6 hours as a transfer target, and I haven't boogered the numbers, the 120t of LOX needs to move at a rate of 4.9l/sec and 30t of liquid methane at 3.3l/sec.

That's the kind of time scale where prop losses from ullage thrust are gonna hurt you pretty badly.

I was initially thinking that half an hour would be about right for 150t of prop, but even that might be too leisurely.  Remember that the worst case is actually when the depot transfers all of its prop to a payload Starship.  And you can wind up with orbital constraints on how long you can take, especially if you're doing high-cadence tanker ops where you want to de-orbit for quick turnaround, or if you have a depot that's transferring its whole load in an HEEO.

If fountaining is a non-issue with any appreciable pool depth in the receiving tank, then you should start off slow but ramp up to >100kg/s as soon as possible.  Remember, in microgravity, your pump power can be tiny, even for high flow rates.

Looks like yesterday's WDR took about 52 minutes to load Starship with 1200 ton of prop (385kg/s total - so maybe up to 200kg/s per tank). Not sure if that would be representative of an in-space fill sequence, but hard to imagine it going any faster than that.


FWIW, because of the methane downcomer transfer tube the propellant inlets will be off center. This doesn't negate vortex formation but I think it would keep it from being a big concentric tank enveloping swirl. Add a 'T' fitting and there would either be two counter swirls or chaotic movement - I think. If that's not enough, a flapper valve in the outlet T could periodically reverse the output direction and nip a big vortex in the bud.


Or, as you say, some max flow rate that the above measures might raise.

I don't think you have to fill the whole tank with baffles; you just need enough of them to generate turbulence before any kind of organized laminar flow builds up.  It's not often that tip vortices are your friend, but this is one time they are.

You could probably cage the outlet (inlet?  the place where stuff comes into the receiving tank) with something to break up the flow even more.  Note that I'm implicitly assuming that fill/drain is not the same as the downcomers.

Yeah, just been looking at this video of fire suppression sprinklers, where the flow rate is pretty high (google says 50-100 litres per second), and they create a pretty evenly distributed stream of fluid. Maybe even just having a small array of these inside the inlet to the receiver tank would be enough to make the flow sufficiently chaotic.

https://youtu.be/NXAgjKDFc14

Still would be nice to see some CFD of liquid oxygen entering a propellant tank.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 01/24/2023 10:22 pm
Remember, in microgravity, your pump power can be tiny, even for high flow rates.

(Not really. Friction/inertia will instead dominate and create an irreducible minimum.)

But the real limit will be cavitation. That same micro-g that reduces the back-pressure from the "head" also reduces the "tail" pressure, the rate that the fluid can reach the intake. (Normally, gravity in a tank is, in effect, acting as a first-stage "pump" pushing the fluid into the actual pump.)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/24/2023 10:24 pm
Looks like yesterday's WDR took about 52 minutes to load Starship with 1200 ton of prop (385kg/s total - so maybe up to 200kg/s per tank). Not sure if that would be representative of an in-space fill sequence, but hard to imagine it going any faster than that.

That brings up another interesting question:  Transfer LCH4 and LOX simultaneously, or sequentially?  Constraints on pumping simultaneously would be:

1) Pump power/energy.  You're going to be in eclipse for part of the transfer sequence, so you're reliant on batteries for 100% of the energy part of the time, and the solar power has to recharge the batteries as well as run the pumps while you're in sunlight.  Running one pump will consume less power than two.

2) Contingency complexity.  If you had a pump fail or de-prime itself while the other was running, the dynamics will be weird.  I'd be tempted not to think about that weirdness, at least in the early days.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/24/2023 10:39 pm
Remember, in microgravity, your pump power can be tiny, even for high flow rates.

(Not really. Friction/inertia will instead dominate and create an irreducible minimum.)

But the real limit will be cavitation. That same micro-g that reduces the back-pressure from the "head" also reduces the "tail" pressure, the rate that the fluid can reach the intake. (Normally, gravity in a tank is, in effect, acting as a first-stage "pump" pushing the fluid into the actual pump.)

Friction head loss is what Poiseuille will give you.  My mistake up-thread was thinking that it also includes dynamic pressure--I'm now mostly convinced that it doesn't.

But it's an excellent point about being cavitation-limited.  Static pressures at 1mm/s² are only in double-digit Pa, even for full tanks.  The h-P diagrams I've seen don't go low enough to show what the liquid saturation curve looks like at those pressures, but I doubt that it's encouraging.

This may be the best argument I've heard for why the system has to be pressure-fed.

It still might be desirable to connect the two ullage spaces, but have a gaseous pump pulling from the receiver and pumping into the sender.  That way, if you lose your pressure differential (e.g. by uncovering the sender's side of the transfer pipe), there's a nice, simple way to restore the differential--and there's still no venting required.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 01/24/2023 11:18 pm
Pumps will not cavitate as long as the ullage pressure is large enough.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/25/2023 04:01 am
Pumps will not cavitate as long as the ullage pressure is large enough.

Good point.  So you can still have equalized pressure between the two ullage spaces, as long as the absolute pressure is a few bar, correct?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/25/2023 07:03 am
Not sure a nose by nose rotational docking is even possible. The mass center of a docked combination of full depot and an empty starship, will be somewhere in the upper third of the depot oxygen tank. How rotation in this case should help settle something is beyond my expertise. Apart from that, the rotation is poised to become extremely unstable. If SpaceX would make that work, it would be a marvel in itself.

I meant to respond to this earlier. I'm attaching a cross-section of my model, just so it's clear what I mean when I say offset nose-to-nose docking. Offsetting this way (about 5m centreline to centreline) to avoids any need to move the LOX header tank and keeps the entire docking mechanism on the leeward side of Starship, so there's no need to change the heat shield either. It also addresses any intermediate axis instability issues (the offset config increases moment of inertia of this spin axis by ~3x versus the alternative, intermediate spin axis).

You're right that in the extreme case of a 1300-1600 ton depot and empty 95 ton starship, the centre of mass will be somewhere on the LOX side of the common dome (about 25m from the centre of rotation). But in that case you can still settle the LOX, so you start transferring LOX first, until the centre of mass moves into the methane tank, then you transfer all the methane, then you finish by transferring the final bit of LOX.

Lastly, I've realised that having a depot with a "prop transfer arm", as I showed in the video in my earlier post  (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2449741#msg2449741)is completely unnecessary. I'll change my animation to assume we pipe any propellant up alongside the pipe to the LOX header tank, and have the transfer occur through the 4x probe/drogue setup I depicted next to the docking port. This way any two starships can transfer prop to each other (not just to and from the depot, which was a limitation with the previous config).   

I suspect that there will be reasons for settling acceleration to be greater than 1mm/s², but if that's the case, then combusting methox thrusters are about a jillion times easier to engineer than a rotating system.  Rotating systems are terrible to make reliable, especially when you're moving their center of mass around by pumping stuff from point A to point B.

Maybe I'm missing something here, but can you give me an example of why rotating systems are so hard to make reliable?  I must admit I have a bit of a cringe whenever I see tether systems being proposed, but supposing direct docking as in this picture, spun to produce settling acceleration of 1% of G (0.45 rpm), where do you see the complexity arising? Because I'd hope the potential side benefits of this config in a crewed context would be obvious to everyone (that's not a reason to do this if it makes the prop transfer problem harder, but I'm missing how it does that).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/25/2023 08:12 am
Pumps will not cavitate as long as the ullage pressure is large enough.

Good point.  So you can still have equalized pressure between the two ullage spaces, as long as the absolute pressure is a few bar, correct?
If you already have high ullage pressures, then as calculated above you can omit the pumps entirely and move fluids by pressure difference along. Can't cavitate pumps if you don't have any pumps.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/25/2023 09:45 pm
Pumps will not cavitate as long as the ullage pressure is large enough.

Good point.  So you can still have equalized pressure between the two ullage spaces, as long as the absolute pressure is a few bar, correct?
If you already have high ullage pressures, then as calculated above you can omit the pumps entirely and move fluids by pressure difference along. Can't cavitate pumps if you don't have any pumps.

There's a fundamental difference between high ullage pressures and high ullage pressure differences.  The first will eliminate cavitation.  The second will move propellant.  If you're going to pressure-feed, you need to manage the ullage pressures in both tanks, via venting (on the receiver) and heating (on the sender).

I'm still struggling a bit trying to figure out pump power requirements, so really big numbers there could change my mind.  But unless that happens, pressure-feeding sounds insanely more complicated than equalizing pressures, especially since the QD has pre-press plumbing built into it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/26/2023 09:55 am
Pumps will not cavitate as long as the ullage pressure is large enough.

Good point.  So you can still have equalized pressure between the two ullage spaces, as long as the absolute pressure is a few bar, correct?
If you already have high ullage pressures, then as calculated above you can omit the pumps entirely and move fluids by pressure difference along. Can't cavitate pumps if you don't have any pumps.

There's a fundamental difference between high ullage pressures and high ullage pressure differences.  The first will eliminate cavitation.  The second will move propellant.  If you're going to pressure-feed, you need to manage the ullage pressures in both tanks, via venting (on the receiver) and heating (on the sender).

I'm still struggling a bit trying to figure out pump power requirements, so really big numbers there could change my mind.  But unless that happens, pressure-feeding sounds insanely more complicated than equalizing pressures, especially since the QD has pre-press plumbing built into it.
Equalising pressures during pumping requires continuous ullage gas generation for the sender tank, and continuous venting for the receiver tank. This requires continuous closed-loop control of both gas generation and tank venting throughout the entire transfer process.

Pressure transfer requires pressurising the sender tank once at the start of transfer, venting the receiver tank once at the start of transfer, then opening the inter-tank valve(s) to allow fluid to flow. No additional venting or pressurisation is required during transfer. In the event of an extreme fluid volume transfer (e.g. completely full sender to completely empty receiver) then once transfer ceases (pressure equalised) the tanks can be isolated again, the sender repressurised, the receiver vented, and the inter-tank valve opened again to repeat the process. Venting and pressurisation are "run to completion" processes with no direct constraints on pressurisation/venting time or rate.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/26/2023 11:12 pm
Equalising pressures during pumping requires continuous ullage gas generation for the sender tank, and continuous venting for the receiver tank. This requires continuous closed-loop control of both gas generation and tank venting throughout the entire transfer process.

No, it just requires connecting the two ullage spaces together with a small-diameter pipe.  Then, as prop is pumped out of the sending tank and into the receiving tank, that creates higher pressure in the receiver ullage and lower pressure in the sender, so ullage gas naturally flows from the receiver to the sender to equalize the difference.

I've attached a cheesy diagram of what I'm thinking about.  Note that this system will pump prop in either direction, which is a requirement for a depot, may be a requirement for a tanker (because sometimes tankers will receive prop from depots to send it to cislunar, transfer it, and then go straight back to EDL), and probably not a requirement for a target Starship.

The only requirement is that the joint ullage space be pressurized to the necessary level to avoid pump¹ cavitation.  A couple of bar?²

Quote
Pressure transfer requires pressurising the sender tank once at the start of transfer, venting the receiver tank once at the start of transfer, then opening the inter-tank valve(s) to allow fluid to flow. No additional venting or pressurisation is required during transfer. In the event of an extreme fluid volume transfer (e.g. completely full sender to completely empty receiver) then once transfer ceases (pressure equalised) the tanks can be isolated again, the sender repressurised, the receiver vented, and the inter-tank valve opened again to repeat the process. Venting and pressurisation are "run to completion" processes with no direct constraints on pressurisation/venting time or rate.

I'm not saying that pressure-fed transfers won't work; I'm saying that they're complicated, slow, wasteful, and error-prone.

Complicated:  Yes, you can do this only once at the beginning, but in a worst-case transfer (e.g. 1200t from depot to target Starship), and with the assumption that you have 5% ullage space in a full tank and you don't vent down to below half a bar, you wind up with 10.5bar in the receiving tank.  So you're venting multiple times, and you're pressurizing your sending tank to a pretty high value.

Slow:  In a blow-down tank (which is essentially what you've created here), and assuming that all of your pressure difference can be converted to dynamic pressure, then the volumetric flow rate from one tank to the other is proportional to the square root of the pressure difference.  (Δp = ½⍴v², and v = volumetricFlowRate / (π*rPipe²).  This doesn't include static drops due to viscosity, but they're tiny.)  So as the difference drops, so does your flow rate.  I'm too lazy and stupid to integrate this, but the upshot is you need to do, if not closed-loop control, then lots of vents to keep the pressures reasonable (see "complicated" above).

Wasteful:  Unless you're willing to vent all of the receiver ullage into a propulsion system on both ships that delivers the ullage thrust (which coincidentally would require exactly the same plumbing you'd need for equalized ullage pressure, but with more valves), then you're dumping a few tonnes of otherwise perfectly good prop into space.

Error-prone:  Uncover the sender's side of the transfer pipe even for a moment, and you've equalized pressures.  Then you get to start all over, with more complexity and waste.  And of course you have to mitigate whatever consequences result from a high-speed gas flow.  (I hereby coin the term "depot fart" to describe this phenomenon.)

_______________
¹Here's a spreadsheet on flow rates and power (https://docs.google.com/spreadsheets/d/1G_Ujssgnlr2sszbDhL-p8jlW1G7xPEHURpPb8We7yu8/edit?usp=sharing), with a couple of screenshots below.  One is for what I consider the worst case:  a depot-to-target full transfer in HEEO, with a constraint of 3 hours transfer time, so that it's possible to do a one-orbit RPOD, transfer, undock, and checkout before perigee insertion.  That would require close to a 4kW pump.  (Well, two 4kW pumps, running sequentially.)  The second one is for a tanker-to-depot transfer, with a pumps limited to 1kW, which takes about half an hour.

²My assumption for cold-pressurization (i.e. ullage pressurization without the Raptors running) is that you flow liquid prop into COPVs, seal 'em up, and heat the liquid to supercritical temperatures and pressures.  Then, when you need ullage pressure, you release gas from the COPVs.  Starship is going to need sources of GCH4 and GO2 for a variety of purposes:  ullage pressurization, warm gas thrusters, combusting gas thrusters, blow-down pressure for liquid methalox-driven equipment like pressure-fed liquid thrusters or APUs, etc.  So there's some irreducible complexity involved no matter what.  But that whole system is dramatically simplified by pumping liquid, not gas, as its first step.  I suspect that you can make your transfer pumps do double duty here, but I haven't attempted writing down the plumbers' nightmare required to implement it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/27/2023 12:40 am
Maybe I'm missing something here, but can you give me an example of why rotating systems are so hard to make reliable?  I must admit I have a bit of a cringe whenever I see tether systems being proposed, but supposing direct docking as in this picture, spun to produce settling acceleration of 1% of G (0.45 rpm), where do you see the complexity arising? Because I'd hope the potential side benefits of this config in a crewed context would be obvious to everyone (that's not a reason to do this if it makes the prop transfer problem harder, but I'm missing how it does that).
I started off really keen on rotating systems, but I was eventually talked out of it. I can share the reasons that changed my mind. For proof, you'll need input from actual experts.

One of the biggest issues is that they have to minimize the number of "things no one has ever done before." If we're just talking blue-sky stuff that might be in use in 2100, that's different, but we're talking about what SpaceX might do in the next three or four years. That really rules out the most interesting rotational systems. In fact, there has been so little study of rotating systems in space that merely figuring out what you have to do to make a stable system is a nontrivial task.

Another factor I didn't consider is plumbing. Apparently cryogenic plumbing is really hard to make work. Every bend in a pipe and every meter of pipe is a potential problem. So proposals that involve running a lot of cryogens through a lot of pipe are generally dead-on-arrival. Nose-to-nose proposals either require really long external pipes or entirely new piping inside the rocket.

For the configuration you propose, there's the extra challenge that the depot will have anywhere from 0 to 1500 tons of fuel in it, but the tanker will only have 150, so this thing is going to be awfully unbalanced, and the center of rotation is going to change during fueling. There might be a way to make that work, but, again, it falls into the category of "stuff no one has ever done."

By contrast, ullage burns are well-established technology for settling propellant before an engine burn. For that purpose, they work at amazingly low accelerations--the sort of thing you could do for an hour without burning much fuel. It's a bit of a leap of faith that it will work for refueling, but it seems plausible with a lot less risk. Also, refueling can tolerate more sloshing than an engine burn, since it's not the end of the world if a propellant pump occasionally sucks some gas or a gas pump occasionally sucks some fluid--assuming you designed for that.

So when you've just got a few years to make this work, and you've got a choice between a technology that you're pretty sure will work at an acceptable cost vs one that could take ten or twenty years to work the bugs out, it's obvious why only the former gets serious attention.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/27/2023 04:26 am
Another factor I didn't consider is plumbing. Apparently cryogenic plumbing is really hard to make work. Every bend in a pipe and every meter of pipe is a potential problem. So proposals that involve running a lot of cryogens through a lot of pipe are generally dead-on-arrival. Nose-to-nose proposals either require really long external pipes or entirely new piping inside the rocket.

The two biggest problems I can see with this for crewed systems, specifically LSS, are:

1) The way the prop transfer docking is arranged interferes with the way the crew-transfer docking is assumed to work.  LSS has no header tanks in the nose, so putting the IDSS-compliant dock in the nose seems like a no-brainer.  Getting that to play nice with the prop transfer seems difficult.

2) Having cryo piping going through the walls of a crew compartment sounds like something hard to get crew-certified.  LSS doesn't need to do this, because it has no header tanks.  A Starship crew-certified for launch and EDL would need to do this, but that just means it's yet another on a long list of things that will be incredibly hard to get launch/EDL certified.  It's certainly not something you'd want to add to your v1.0 refueling architecture, especially since LSS doesn't work without the v1.0 refueling architecture.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/27/2023 06:39 am
One of the biggest issues is that they have to minimize the number of "things no one has ever done before." If we're just talking blue-sky stuff that might be in use in 2100, that's different, but we're talking about what SpaceX might do in the next three or four years. That really rules out the most interesting rotational systems. In fact, there has been so little study of rotating systems in space that merely figuring out what you have to do to make a stable system is a nontrivial task.

Another factor I didn't consider is plumbing. Apparently cryogenic plumbing is really hard to make work. Every bend in a pipe and every meter of pipe is a potential problem. So proposals that involve running a lot of cryogens through a lot of pipe are generally dead-on-arrival. Nose-to-nose proposals either require really long external pipes or entirely new piping inside the rocket.

For the configuration you propose, there's the extra challenge that the depot will have anywhere from 0 to 1500 tons of fuel in it, but the tanker will only have 150, so this thing is going to be awfully unbalanced, and the center of rotation is going to change during fueling. There might be a way to make that work, but, again, it falls into the category of "stuff no one has ever done."

Just considering what happens during the Starship landing swoop, which was practiced with SN8, SN9, SN10, SN11 and SN15, and which SpaceX will have to nail on at least a number of further Starship landings: Cryogenic oxygen has to travel/be pumped the length of Starship from the nose header tank to the engines, whilst the whole structure is rotating through the same spin direction as I've proposed for prop transfer.

Starship's key goal of full reusability has led to a configuration which will be pretty good at "pumping cryogenic propellant through long pipes while rotating". Sure, it hasn't been demonstrated in space, but the principle has been demonstrated in arguably more demanding conditions than what it would experience in space.

Quote
By contrast, ullage burns are well-established technology for settling propellant before an engine burn. For that purpose, they work at amazingly low accelerations--the sort of thing you could do for an hour without burning much fuel. It's a bit of a leap of faith that it will work for refueling, but it seems plausible with a lot less risk. Also, refueling can tolerate more sloshing than an engine burn, since it's not the end of the world if a propellant pump occasionally sucks some gas or a gas pump occasionally sucks some fluid--assuming you designed for that.

So when you've just got a few years to make this work, and you've got a choice between a technology that you're pretty sure will work at an acceptable cost vs one that could take ten or twenty years to work the bugs out, it's obvious why only the former gets serious attention.

I do hear what you're saying, really. But I'd argue this isn't a first-principles approach to the problem, and the difference in current TRL is not going to be that big a deal to SpaceX (*coughs* landing boost stages, steel rockets, mechazilla).

For the sake of argument, I had a go at calculating how much propellant it would take to spin up a 1850 metric ton Depot/tanker combo docked (offset) nose to nose as per the diagram in my post (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2452126#msg2452126) to 0.5rpm (1% of G).

Assuming 4x 300s hot gas thrusters (455N each), and calculating a total moment of inertia of 1,248,750,000 kg.m2 for the whole system. I got a total spin-up time of 735s (12.2 minutes), and a total propellant mass use of ~460kg per spin-up / spin-down. It should actually be somewhat better than this in some situations, because the offset centre of mass would give one set of thrusters a longer lever arm and better torque, but let's say 1 ton of prop used per prop transfer.

Using the Radical Moderate's spreadsheet calculator, I found that you have to use hot gas thrusters with half hour transfer times and <1.5mm/s settling acceleration before linear acceleration beats this.

In the TRM's "worst case" post above with a depot in a 3 hour (300x8000km) HEEO - I can't see where I'm doing the math wrong, because I can only get 15 ton of payload of prop to the depot in that orbit (assuming isp=380s, 100 ton tanker dry weight, 7.7km/s LEO, then 1.3km/s transfer burn) - Whatever the case, propellant only gets more precious the further away from LEO you are, and it seems a little absurd to throw it away if you don't need to.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/27/2023 06:47 am
Another factor I didn't consider is plumbing. Apparently cryogenic plumbing is really hard to make work. Every bend in a pipe and every meter of pipe is a potential problem. So proposals that involve running a lot of cryogens through a lot of pipe are generally dead-on-arrival. Nose-to-nose proposals either require really long external pipes or entirely new piping inside the rocket.

The two biggest problems I can see with this for crewed systems, specifically LSS, are:

1) The way the prop transfer docking is arranged interferes with the way the crew-transfer docking is assumed to work.  LSS has no header tanks in the nose, so putting the IDSS-compliant dock in the nose seems like a no-brainer.  Getting that to play nice with the prop transfer seems difficult.

2) Having cryo piping going through the walls of a crew compartment sounds like something hard to get crew-certified.  LSS doesn't need to do this, because it has no header tanks.  A Starship crew-certified for launch and EDL would need to do this, but that just means it's yet another on a long list of things that will be incredibly hard to get launch/EDL certified.  It's certainly not something you'd want to add to your v1.0 refueling architecture, especially since LSS doesn't work without the v1.0 refueling architecture.

Admittedly I've never put too much stock in the claim they'll remove the nose header tank in order to place an IDSS compliant dock in the nose, because it's never been clear to me how that would work with the heat-shield for EDL. Any re-entry will still need a skydiver maneuver, ergo the dock has to be offset, ergo you might as well leave the header tank where it is.

LSS is never intended to return, so I can see them going down a path where they remove the tank and put in the nose dock as you say, but then do they have a hinged, heat-shielded, nose-cone arrangement on every other Starship that needs to EDL so they can keep a common heritage across their vehicles?

As for putting cryo through the crew comparment... well there is no shortage of vacuum in space. 
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 01/27/2023 09:44 am
Complicated:  Yes, you can do this only once at the beginning, but in a worst-case transfer (e.g. 1200t from depot to target Starship), and with the assumption that you have 5% ullage space in a full tank and you don't vent down to below half a bar, you wind up with 10.5bar in the receiving tank.  So you're venting multiple times, and you're pressurizing your sending tank to a pretty high value.
Only if you try and one-shot every transfer. Multiple steps with venting in between is not a high complexity operation, and requires bang-bang control of a single vent valve and a single main liquid valve to perform. It is hard to get more simple than that. The 'complexity' of ullage gas regeneration is one that is mandatory to solve anyway for Starship, in order to allow coasts to destinations that require more than the header tank capacity alone at the destination (e.g. HLS) as all autogenous ullage generated during any initial burns would otherwise condense.
Quote
Slow:  In a blow-down tank (which is essentially what you've created here), and assuming that all of your pressure difference can be converted to dynamic pressure, then the volumetric flow rate from one tank to the other is proportional to the square root of the pressure difference.  (Δp = ½⍴v², and v = volumetricFlowRate / (π*rPipe²).  This doesn't include static drops due to viscosity, but they're tiny.)  So as the difference drops, so does your flow rate.  I'm too lazy and stupid to integrate this, but the upshot is you need to do, if not closed-loop control, then lots of vents to keep the pressures reasonable (see "complicated" above).
See also: "not complicated", above for multiple transfer steps.
Quote
Wasteful:  Unless you're willing to vent all of the receiver ullage into a propulsion system on both ships that delivers the ullage thrust (which coincidentally would require exactly the same plumbing you'd need for equalized ullage pressure, but with more valves), then you're dumping a few tonnes of otherwise perfectly good prop into space.
If you utilise a vent routed locally downwards, every vent is some extra free settling thrust. The small (e.g. an entire methane tank at 1 Bar CH4 vented to vacuum is less than 400kg) overall mass loss is not a major concern and is greatly overstated.
Quote
Error-prone:  Uncover the sender's side of the transfer pipe even for a moment, and you've equalized pressures.  Then you get to start all over, with more complexity and waste.  And of course you have to mitigate whatever consequences result from a high-speed gas flow.  (I hereby coin the term "depot fart" to describe this phenomenon.)
An issue shared with pump transfer. If the inlet is uncovered, you also risk having to pick bits of pump out of your teeth. A pressure-based transfer is a single valve closure away from complete system shutdown. A pump-based transfer requires graceful pump shutdown to avoid a catastrophic system failure - resulting in anywhere from loss of prop transfer capability (loss of mission capability) to outright loss of both vehicles.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: colbourne on 01/27/2023 12:11 pm
It still seems to me having read this thread that it would be a lot easier to transfer full tanks of fuel (maybe with their own rocket engines) to the Starship, rather than trying to transfer the fuel. There would be some weight penalty , but the simplicity and time saved would probably compensate for this especially if their is venting of gas during the refueling and storage.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 01/27/2023 07:30 pm
It still seems to me having read this thread that it would be a lot easier to transfer full tanks of fuel (maybe with their own rocket engines) to the Starship, rather than trying to transfer the fuel. There would be some weight penalty , but the simplicity and time saved would probably compensate for this especially if their is venting of gas during the refueling and storage.
I like that idea too, but, again, you're building something new that has to get designed, built, and tested. Something with new, smaller tanks, a new rocket engine, a new guidance system, etc. Nothing about it is reused from the other work on Starship. It has to somehow get back into the cargo hold for the trip back down as well.

At least to start with, it just looks a whole lot easier to make relatively small modifications to Starship to make it serve as both tanker and depot. And if that costs you something in fuel, well, fuel is cheap. Once the whole system is working, then it'll make sense to study it and find ways to optimize it. But it really looks like the quickest, safest way to build something that'll actually work is to just make three Starship variants.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Oersted on 01/27/2023 08:51 pm
The best (only?) way for Starships to dock will be back to back, or dorsal to dorsal if you prefer. The dorsal area is the area not covered in tiles and the area least affected by eventual weak spots or protuberances resulting from the docking ports.

That also means the minimum length of tunnels for crew transfers and pipes for fuel/oxidiser transfers, because the crew compartments and the tanks will all be nestled next to each other.

Will a pitch axis rotation of two ships docked dorsally enable a gravity fuel transfer?   
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/27/2023 08:54 pm
The 'complexity' of ullage gas regeneration is one that is mandatory to solve anyway for Starship, in order to allow coasts to destinations that require more than the header tank capacity alone at the destination (e.g. HLS) as all autogenous ullage generated during any initial burns would otherwise condense.

Yes, I agree completely.  But unless you're planning on running all of the things for which you need gas out of the mains (and you can't get high-pressure gas¹ that way), you're going to have to move liquid into a COPV and heat it.  And you know what you need to do that?  A pump.  Why not use it to transfer prop as well?

Quote
If you utilise a vent routed locally downwards, every vent is some extra free settling thrust. The small (e.g. an entire methane tank at 1 Bar CH4 vented to vacuum is less than 400kg) overall mass loss is not a major concern and is greatly overstated.

It's 1.8t for both tanks, and 1bar is a reasonable number for transfer into the depot.  But a target Starship is going to arrive with tanks close to flight pressure, so at least quadruple that.  Unless you're going to wait for them to undergo condensation collapse before transfer?  Not a great thing to do in HEEO.  Not even a great thing to do in LEO or NRHO if you have a crewed refueling.

I agree that it's not the end of the world.  But why would you do this if you didn't have to?
 
Quote
An issue shared with pump transfer. If the inlet is uncovered, you also risk having to pick bits of pump out of your teeth.

You're thinking about this like a rocket engine turbopump.  It's not:  it's an electric pump that has to generate about 0.1bar of pressure.  It draws as much power as a hairdryer or two.  It's not going to overspin, and even if it did, nothing would happen.  And if you're positing a valve control loop that can mitigate a depot fart, you can certainly come up with an electric servo that's responsive enough stop feeding juice into an unprimed pump.

More importantly, you need this pump no matter what, for filling COPVs to produce high-pressure gas.  They don't weigh very much; build in redundancy.  You'll need it for the high-pressure applications anyway, because if you can't generate supercritical gas then the mission fails.

_________________________________
¹Things off the top of my head that need GCH4 or GOX at pressures higher than 4-6bar:

1) Spin-start gas--quite a bit of it, at well more than 6bar.
2) Cold/warm gas thrusters with Isp > 20s.
3) Blow-down for methox or methalox combustion thrusters.  Might not be needed but I'm betting they will.  Hard to get the needed ullage thrust without them.  And of course there are the LSS landing thrusters, if they happen.
4) Ullage pressurization on demand.
5) I'm betting they'll have an APU onboard by Option B or SLT.  It takes surprisingly little methalox to generate a few kW for long durations in the dark.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/27/2023 09:13 pm
The best (only?) way for Starships to dock will be back to back, or dorsal to dorsal if you prefer. The dorsal area is the area not covered in tiles and the area least affected by eventual weak spots or protuberances resulting from the docking ports.

That also means the minimum length of tunnels for crew transfers and pipes for fuel/oxidiser transfers, because the crew compartments and the tanks will all be nestled next to each other.

I think you're right for crew transfer on launch/EDL crew-certified Starships.  But for Option A/B/SLT lunar Starships, I doubt it.  Those don't have thermal tile issues, and I shudder to think what the Orion proximity ops software would make of trying to figure out what to do when approaching what looks like a 9 meter infinite wall.  You also need to think about the limited docking real estate at the Gateway.

Stuff's gonna stay on-axis for a while.

Also, I don't think high-flow cryogenic fluid transfer in close proximity to a crew tunnel is a feature.  My bet is that prop transfer docking and crew docking are two completely separate systems.

Quote
Will a pitch axis rotation of two ships docked dorsally enable a gravity fuel transfer?

For pitch or yaw axis, you'd have to keep the center of mass forward of the tanks to keep them settled at the bottom.  That'll be hard to do.

Roll axis is possible, but you'll have sump slosh problems and the sumps will be on the TPS side, which isn't great.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/28/2023 07:04 pm
Another factor I didn't consider is plumbing. Apparently cryogenic plumbing is really hard to make work. Every bend in a pipe and every meter of pipe is a potential problem. So proposals that involve running a lot of cryogens through a lot of pipe are generally dead-on-arrival. Nose-to-nose proposals either require really long external pipes or entirely new piping inside the rocket.

The two biggest problems I can see with this for crewed systems, specifically LSS, are:

1) The way the prop transfer docking is arranged interferes with the way the crew-transfer docking is assumed to work.  LSS has no header tanks in the nose, so putting the IDSS-compliant dock in the nose seems like a no-brainer.  Getting that to play nice with the prop transfer seems difficult.

2) Having cryo piping going through the walls of a crew compartment sounds like something hard to get crew-certified.  LSS doesn't need to do this, because it has no header tanks.  A Starship crew-certified for launch and EDL would need to do this, but that just means it's yet another on a long list of things that will be incredibly hard to get launch/EDL certified.  It's certainly not something you'd want to add to your v1.0 refueling architecture, especially since LSS doesn't work without the v1.0 refueling architecture.

Admittedly I've never put too much stock in the claim they'll remove the nose header tank in order to place an IDSS compliant dock in the nose, because it's never been clear to me how that would work with the heat-shield for EDL. Any re-entry will still need a skydiver maneuver, ergo the dock has to be offset, ergo you might as well leave the header tank where it is.

LSS is never intended to return, so I can see them going down a path where they remove the tank and put in the nose dock as you say, but then do they have a hinged, heat-shielded, nose-cone arrangement on every other Starship that needs to EDL so they can keep a common heritage across their vehicles?

As for putting cryo through the crew comparment... well there is no shortage of vacuum in space.
I think you're over interpreting commonality. SH and SS are the best existing example. Ignoring everything on SS from the cargo space on up and everything having to do with EDL, they have high commonality.


They have the same engines with vacuum optimization where needed. They are made of the same materials assembled with with the same tooling to the same diameter. Same rings with some thickness differences. Same upper dome (I think). Different common and bottom dome/thrust structure due to the transfer tube and different engine count. I haven't checked but expect that the stringers are the same design with differing lengths and placement. They are two variations on one general design. Commonality does not mean exactly the same.


Header tanks are doodads that come and go with need. The automotive industry handles bigger differences on an hourly basis at a production rate orders of magnitude higher than anything SS will ever see. They use build sheets for each vehicle that have tentacles that reach from engineering to marketing to purchasing to production. SX will do the same.


I'm about to post to the engineering thread about human rated ships and header tanks.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tbellman on 01/28/2023 10:04 pm
It still seems to me having read this thread that it would be a lot easier to transfer full tanks of fuel (maybe with their own rocket engines) to the Starship, rather than trying to transfer the fuel. There would be some weight penalty , but the simplicity and time saved would probably compensate for this especially if their is venting of gas during the refueling and storage.

Absolutely not.  This is a terrible idea.

Full tanks of Starship is about 260 tonnes of liquid methane, and 940 tonnes of liquid oxygen.  But you need to split that up in multiple smaller tanks that can be lifted to orbit by a tanker Starship.  Just the additional "endcaps" of all those tanks will be several tens of tonnes.  And the dome shape of those endcaps will force the entire ship to be longer (there will be wasted space between the tanks), adding to the mass.

The design of the ship would be entirely different from how it is now.  You would need some kind of backbone truss/beam onto which the tanks are mounted, and each tank also needs "skirts" covering the domes, so the ships profile will still be a cylinder (for aerodynamic purposes), and to carry the TPS tiles.  Those skirts needs to be connected together so there isn't a gap between them letting in hot plasma during EDL.  All this adds to the mass that needs to be lifted by the "tanker" ships (and presumably also brought down back to the surface).

Alternatively, you would design the ship as a large empty tube, with TPS tiles covering one side, and the other side being huge doors through which the smaller tanks can be removed and inserted.  But that will effectively turn the ship into a double-walled ship, which increases its dry mass significantly.

And the mechanics of unmounting empty tanks and mounting new full tanks, and coupling them together, sounds like a nightmare to me.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: colbourne on 01/29/2023 01:43 am
It still seems to me having read this thread that it would be a lot easier to transfer full tanks of fuel (maybe with their own rocket engines) to the Starship, rather than trying to transfer the fuel. There would be some weight penalty , but the simplicity and time saved would probably compensate for this especially if their is venting of gas during the refueling and storage.

Absolutely not.  This is a terrible idea.

Full tanks of Starship is about 260 tonnes of liquid methane, and 940 tonnes of liquid oxygen.  But you need to split that up in multiple smaller tanks that can be lifted to orbit by a tanker Starship.  Just the additional "endcaps" of all those tanks will be several tens of tonnes.  And the dome shape of those endcaps will force the entire ship to be longer (there will be wasted space between the tanks), adding to the mass.

The design of the ship would be entirely different from how it is now.  You would need some kind of backbone truss/beam onto which the tanks are mounted, and each tank also needs "skirts" covering the domes, so the ships profile will still be a cylinder (for aerodynamic purposes), and to carry the TPS tiles.  Those skirts needs to be connected together so there isn't a gap between them letting in hot plasma during EDL.  All this adds to the mass that needs to be lifted by the "tanker" ships (and presumably also brought down back to the surface).

Alternatively, you would design the ship as a large empty tube, with TPS tiles covering one side, and the other side being huge doors through which the smaller tanks can be removed and inserted.  But that will effectively turn the ship into a double-walled ship, which increases its dry mass significantly.

And the mechanics of unmounting empty tanks and mounting new full tanks, and coupling them together, sounds like a nightmare to me.
Starship can lift about 100 tonnes to LEO. Each trip will be using a tank with  endcaps , so you have already paid the weight price. For the trip to Mars there will be some penalty, but these empty tanks can be jettisoned (hopefully to be reused or used as valuable scrap on Mars in the future) and do not need to be streamlined (They would have to be streamlined for the trip to LEO).
I would see only a small Mars lander Starship making the whole trip to Mars.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/29/2023 03:55 am
Starship can lift about 100 tonnes to LEO. Each trip will be using a tank with  endcaps , so you have already paid the weight price. For the trip to Mars there will be some penalty, but these empty tanks can be jettisoned (hopefully to be reused or used as valuable scrap on Mars in the future) and do not need to be streamlined (They would have to be streamlined for the trip to LEO).
I would see only a small Mars lander Starship making the whole trip to Mars.

This is certainly a debatable point, but it shouldn't be debated in a Starship refueling thread.  You're not going to refuel a Starship this way.  The architecture won't support it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Hog on 01/29/2023 07:03 pm
Why are we pumping fluids, when we can be loading blocks of solid propellant, yes solid propellant, "but it's not what you think." 

Let's solidify that O2 and CH4. The Ship pulls up to a refueler and loads a certain quantity of propellants, the Ship backs away, methane/oxygen "melts" into subcooled props and following a quick ullage kick, away she goes firing the Raptors.

What's a kilogram of methane ice worth?  oxygen ice?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/29/2023 07:42 pm
Why are we pumping fluids, when we can be loading blocks of solid propellant, yes solid propellant, "but it's not what you think." 

Let's solidify that O2 and CH4. The Ship pulls up to a refueler and loads a certain quantity of propellants, the Ship backs away, methane/oxygen "melts" into subcooled props and following a quick ullage kick, away she goes firing the Raptors.

What's a kilogram of methane ice worth?  oxygen ice?

So... you're going to put doors on the main tanks?  The densities of solid oxygen and methane are a little bit higher than the liquids, but not by a huge amount.  The blocks of prop are going to be big.  How are you planning on making those big doors reliably leak-proof at 6bar?

Nobody has ever frozen CH4 and O2 at high scale, and they certainly haven't done it in a high-temperature vacuum.  Nor have they frozen it on the ground and launched it as a solid.

Cryogenic pumps aren't exactly new technology.  Adapting them to work in microgravity is trivial compared to developing whole new orbital industrial processes.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 01/30/2023 02:00 pm
Also, I don't think high-flow cryogenic fluid transfer in close proximity to a crew tunnel is a feature.  My bet is that prop transfer docking and crew docking are two completely separate systems.

Not meaning to rehash this excessively, but the main advantage I was aiming to point out was that spin-G settling acceleration means it wouldn’t have to be high flow. You could take 18+ hours to pump it over with at a relatively tiny flow rate, and still be launching tankers daily. I’m picturing the walls of the LOX header downcomer pipe embedded with a set of 4 relatively tiny pipes, then surround the whole structure with a shell pipe that gets vented to vacuum once in space. Also, if you could size your pumping system for those tiny flow rates, how much power would that save? It’s not like on the ground where ambient temperatures will lead to unacceptable levels of off-gassing if you don’t move it fast enough.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 01/30/2023 09:35 pm
Pumps will not cavitate as long as the ullage pressure is large enough.

Good point.  So you can still have equalized pressure between the two ullage spaces, as long as the absolute pressure is a few bar, correct?
If you already have high ullage pressures, then as calculated above you can omit the pumps entirely and move fluids by pressure difference along. Can't cavitate pumps if you don't have any pumps.

There's a fundamental difference between high ullage pressures and high ullage pressure differences.  The first will eliminate cavitation.  The second will move propellant.  If you're going to pressure-feed, you need to manage the ullage pressures in both tanks, via venting (on the receiver) and heating (on the sender).

I'm still struggling a bit trying to figure out pump power requirements, so really big numbers there could change my mind.  But unless that happens, pressure-feeding sounds insanely more complicated than equalizing pressures, especially since the QD has pre-press plumbing built into it.
Equalising pressures during pumping requires continuous ullage gas generation for the sender tank, and continuous venting for the receiver tank. This requires continuous closed-loop control of both gas generation and tank venting throughout the entire transfer process.

Pressure transfer requires pressurising the sender tank once at the start of transfer, venting the receiver tank once at the start of transfer, then opening the inter-tank valve(s) to allow fluid to flow. No additional venting or pressurisation is required during transfer. In the event of an extreme fluid volume transfer (e.g. completely full sender to completely empty receiver) then once transfer ceases (pressure equalised) the tanks can be isolated again, the sender repressurised, the receiver vented, and the inter-tank valve opened again to repeat the process. Venting and pressurisation are "run to completion" processes with no direct constraints on pressurisation/venting time or rate.
ISTM that continuous venting of the receiver during pressure powered transfer would simplify the operation. If parasitic losses cause things to stall out before completion the COPV's on the supply side could add more ullage pressure.


That said, has anyone looked at how this would line up with cold or hot gas consumption for settling thrust? The receiver ullage would be lost anyway whether continuous or pulsed. Might as well make use of it if it adds enough thrust to make the added complexity worth while. I've the echo of a memory of this being asked but no memory of a quantitative answer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 01/30/2023 11:06 pm
.This loses some amount of gaseous prop from venting back down from equilibrium to the target low pressure, but that mass loss is relatively small: e.g. for CH4 assuming a 3 Bar sender tank pressure and the absolute worst case of the entire tank volume to be vented, the maximum mass loss is around 1.5 tonnes.

Except that's not the "absolute worst case," because (as you point out) liquid propellant will be constantly evaporating to replenish the ullage gas.

It's also not the "absolute worst case" because we don't know if it will actually reliably transfer 100% of the propellant. If some is missed that needs to be counted, and even a small amount of liquid could exceed 1.5 tonnes.
Only when in a vacuum state. The vents would not be opened again until the tank has reached pressure equilibrium with the sender tank, and then settles, so venting would not lose liquids (settled so no mixed phase) and pressure would not be allowed to drop before the triple point (so no flash boiling).

"Flash boiling" isn't the only kind of boiling. Lower pressure will always cause an increase in evaporation rate, even before hitting the vapor pressure.

You misunderstood my point about liquid. I wasn't saying that liquid would be vented, but rather that some might be left in the tank.

You also missed the ullage thrusters. If the technique works but takes a lot longer, then that additional ullage prop mass should also be accounted for.
By transferring at a high flow rate and then settling with a sealed tank, you can minimise the time at the maximum settling thrust needed (to keep the inlet covered on the sender tank) and then switch to the minimum thrust needed for the post-transfer sealed tank settling (equivalent to a cost-phase settling thrust). This is opposed to needing to keep sufficient thrust to both keep the sender tank inlet covered and keep the receiving tank from geysering for the entire transfer duration.

Again, you're missing my point. It has yet to be shown that this doesn't take more time vs. using pumps. More time = more ullage propellant used.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 01/31/2023 09:05 pm
Also, if you could size your pumping system for those tiny flow rates, how much power would that save? It’s not like on the ground where ambient temperatures will lead to unacceptable levels of off-gassing if you don’t move it fast enough.

If you're doing a full depot-to-target transfer (1600t depot transferring 1200t of prop to a target with 0t of prop), a 1kW pump can do the transfer sequentially (i.e., first LOX, then methane) in 4 hours 44 minutes.¹  If the same transfer takes 18 hours, the pump only has to be 200W.  If you're solar-panel-limited, that might be important.  If you're battery-limited, it's less important.

Note that there are lots of reasons to do faster transfers:

1) Minimizes prop consumption for ullage thrust (the big advantage of a rotating scheme).

2) Gives you more flexibility in terms of orbital dynamics.  The big one here is probably HEEO-based refueling, where you'd like to do the bulk of refueling near apogee, so the Starship is ready to do a departure burn by perigee.  Note also that a GTO-sized HEEO has a period of about 10 hours, so 18+ likely requires 3 orbits, each of which makes two transits through the Van Allen Belts.  (FWIW, I don't think HEEO refueling will be happening any time soon, but that's been extensively litigated up-thread.  Let's just say that it's an option that you wouldn't want to preclude and move on.)

3) I agree that boil-off isn't a huge issue, but it's not a trivial one, especially if you're transferring prop to a target without good boil-off properties, or if you're transferring to a target whose mission plan needs an amount of prop that's close to needing an integral number of tankers, and a little boil-off could force an extra launch.

4) When you finally get to crewed transfers, minimizing the amount of time that humans are literally bouncing off the walls while strapped to a giant bomb might be nice. 

PS:
_____________
¹Assumes a 15cm pipe and 5mm/s² acceleration.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 02/03/2023 07:31 am
Also, if you could size your pumping system for those tiny flow rates, how much power would that save? It’s not like on the ground where ambient temperatures will lead to unacceptable levels of off-gassing if you don’t move it fast enough.

If you're doing a full depot-to-target transfer (1600t depot transferring 1200t of prop to a target with 0t of prop), a 1kW pump can do the transfer sequentially (i.e., first LOX, then methane) in 4 hours 44 minutes.¹  If the same transfer takes 18 hours, the pump only has to be 200W.  If you're solar-panel-limited, that might be important.  If you're battery-limited, it's less important.

Note that there are lots of reasons to do faster transfers:

1) Minimizes prop consumption for ullage thrust (the big advantage of a rotating scheme).

2) Gives you more flexibility in terms of orbital dynamics.  The big one here is probably HEEO-based refueling, where you'd like to do the bulk of refueling near apogee, so the Starship is ready to do a departure burn by perigee.  Note also that a GTO-sized HEEO has a period of about 10 hours, so 18+ likely requires 3 orbits, each of which makes two transits through the Van Allen Belts.  (FWIW, I don't think HEEO refueling will be happening any time soon, but that's been extensively litigated up-thread.  Let's just say that it's an option that you wouldn't want to preclude and move on.)

3) I agree that boil-off isn't a huge issue, but it's not a trivial one, especially if you're transferring prop to a target without good boil-off properties, or if you're transferring to a target whose mission plan needs an amount of prop that's close to needing an integral number of tankers, and a little boil-off could force an extra launch.

4) When you finally get to crewed transfers, minimizing the amount of time that humans are literally bouncing off the walls while strapped to a giant bomb might be nice. 

PS:
_____________
¹Assumes a 15cm pipe and 5mm/s² acceleration.

Yikes, you weren't kidding about "extensive" - I just read from page 80 up to the present, only having occasionally lurked on this thread up until this month.

Re number 2) HEEO-based refuelling: I think specifically, the conops where you fill up LSS in vLEO, then top it up during a single orbit in vLEO+1600-2000 m/s with a buddy tanker, rendezvousing shortly after the first burn - seems to me the most obvious way to handle that case, and as good a reason as any to set a lower limit on how slowly the system will pump propellant. In that case you would need to transfer 500-600t of prop during a 3-4 hour orbit (~50kg/s). So presumably the pump power can be less than 1kW, but not by very much.

My other thought while reading the thread was around whether the effect of milli-G acceleration - if done in a "spiral in" direction during this single orbit in vLEO+1600-2000m/s - could cause an appreciable reduction of TLI delta V. You could justify significantly higher than 5mm/s² acceleration if it was all adding to your velocity at perigee. You might not be getting the Oberth multiplier, but it wouldn't be "wasted", even so.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/03/2023 07:38 pm
Re number 2) HEEO-based refuelling: I think specifically, the conops where you fill up LSS in vLEO, then top it up during a single orbit in vLEO+1600-2000 m/s with a buddy tanker, rendezvousing shortly after the first burn - seems to me the most obvious way to handle that case, and as good a reason as any to set a lower limit on how slowly the system will pump propellant. In that case you would need to transfer 500-600t of prop during a 3-4 hour orbit (~50kg/s). So presumably the pump power can be less than 1kW, but not by very much.

VLEO+2000 is about a 5.5 hour orbit.  Being able to do RPOD, pre-transfer checkout, post-transfer prep, undocking/prox ops, and pre-insertion checkout in 2.5 hours seems pretty optimistic.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/04/2023 03:41 am
Re number 2) HEEO-based refuelling: I think specifically, the conops where you fill up LSS in vLEO, then top it up during a single orbit in vLEO+1600-2000 m/s with a buddy tanker, rendezvousing shortly after the first burn - seems to me the most obvious way to handle that case, and as good a reason as any to set a lower limit on how slowly the system will pump propellant. In that case you would need to transfer 500-600t of prop during a 3-4 hour orbit (~50kg/s). So presumably the pump power can be less than 1kW, but not by very much.

VLEO+2000 is about a 5.5 hour orbit.  Being able to do RPOD, pre-transfer checkout, post-transfer prep, undocking/prox ops, and pre-insertion checkout in 2.5 hours seems pretty optimistic.

... it seems pretty SpaceX. I approve.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/23/2023 08:52 pm
Random thought from another thread:

It seems reasonable always to point a depot's nose at the Sun.  That way, the ogive portion of the nose can act as a dewar for the LCH4 tank below it, and very little incident radiation should strike the LCH4 or LOX tank walls.  (The Sun has an angular diameter of 0.53º, so the total incident flux should be sin (0.53º/2) = 0.0046 * solarConstant = 6.3W/m², which is almost nothing.)

However, that leaves the albedo radiation reflected/emitted from Earth.  I believe the average flux at mid-latitude inclinations is about 250W/m².

Let's divide the sphere that the depot sees into three sections:

1) The Sun.  Dealt with via pointing at it.
2) The sunlit or nighttime Earth.  250W/m², over some variable viewing angle.
3) Empty space, which is presumably cold.

Inside the depot's tanks, we will have:

a) Conductive heating when a blob of prop touches a hot tank wall.  Tank walls should only be truly hot if they're facing the sunlit Earth.

b) Conductive/convective heating, where the ullage gas picks up heat from a tank wall and transmits it thermally to a blob of liquid prop.

c) Radiative heating, where a hot wall emits IR directly into a blob of prop (bad), or some other section of wall (considerably less bad).

Radiative heating ought to be fairly modest, because any particular spot on a tank wall ought to cycle between fairly hot while directly heated by Earth and cooling when pointed at the Earth nightside or at empty space.  flux = εσT⁴, so even modest declines in the temperature of the inside wall result in much less radiative heating.

So what if we apply continuous, extremely low acceleration toward the Earth?  That should put whatever prop is in the depot as far away from a hot wall as possible, minimizing conductive heating.  This won't help much if the depot is mostly full, but before it's more than half full, this should dramatically reduce heat transfer.

Does this make sense?

Note that continuous radial acceleration in a circular orbit doesn't actually change the shape of your orbit.  It just makes you go around faster, making your true anomaly larger than it would be if you were in freefall.  So the navigational consequences of this are easy to work around.

This also requires pretty fancy RCS management to achieve, since we're holding the attitude of the depot fixed by pointing it at the Sun.  Seems like a reasonable problem, though.  Also, I'd think that whatever boiloff does occur should be able to power cold gas thrusters, both for the radial ullage burn and the burns necessary to counteract Earth's tidal forces, which want the nose to point at the center of the Earth, rather than at the Sun.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 02/23/2023 11:20 pm
Random thought from another thread:

It seems reasonable always to point a depot's nose at the Sun.  That way, the ogive portion of the nose can act as a dewar for the LCH4 tank below it, and very little incident radiation should strike the LCH4 or LOX tank walls.  (The Sun has an angular diameter of 0.53º, so the total incident flux should be sin (0.53º/2) = 0.0046 * solarConstant = 6.3W/m², which is almost nothing.)

However, that leaves the albedo radiation reflected/emitted from Earth.  I believe the average flux at mid-latitude inclinations is about 250W/m².

Let's divide the sphere that the depot sees into three sections:

1) The Sun.  Dealt with via pointing at it.
2) The sunlit or nighttime Earth.  250W/m², over some variable viewing angle.
3) Empty space, which is presumably cold.

Inside the depot's tanks, we will have:

a) Conductive heating when a blob of prop touches a hot tank wall.  Tank walls should only be truly hot if they're facing the sunlit Earth.

b) Conductive/convective heating, where the ullage gas picks up heat from a tank wall and transmits it thermally to a blob of liquid prop.

c) Radiative heating, where a hot wall emits IR directly into a blob of prop (bad), or some other section of wall (considerably less bad).

Radiative heating ought to be fairly modest, because any particular spot on a tank wall ought to cycle between fairly hot while directly heated by Earth and cooling when pointed at the Earth nightside or at empty space.  flux = εσT⁴, so even modest declines in the temperature of the inside wall result in much less radiative heating.

So what if we apply continuous, extremely low acceleration toward the Earth?  That should put whatever prop is in the depot as far away from a hot wall as possible, minimizing conductive heating.  This won't help much if the depot is mostly full, but before it's more than half full, this should dramatically reduce heat transfer.

Does this make sense?

Note that continuous radial acceleration in a circular orbit doesn't actually change the shape of your orbit.  It just makes you go around faster, making your true anomaly larger than it would be if you were in freefall.  So the navigational consequences of this are easy to work around.

This also requires pretty fancy RCS management to achieve, since we're holding the attitude of the depot fixed by pointing it at the Sun.  Seems like a reasonable problem, though.  Also, I'd think that whatever boiloff does occur should be able to power cold gas thrusters, both for the radial ullage burn and the burns necessary to counteract Earth's tidal forces, which want the nose to point at the center of the Earth, rather than at the Sun.

Good and interesting hypothesis.  Now do the math.

There's also a heat shield that can be pointed at the Earth, so the Earth is only heating the tail and the heat shield.  The tiles should absorb all that heat and release it right back at the Earth, and the  insulation will keep the stainless steel cool on that side.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 02/24/2023 01:14 am
It seems reasonable always to point a depot's nose at the Sun.
Is it, though? As we've discussed upthread, if it is far away from Earth or the moon, a depot coated in Solar White tiles will get cold enough to freeze oxygen (or close to it), even though it's otherwise unshielded from the direct sun. What messes this up in LEO is the warmth from the Earth, which comes at inconvenient wavelengths.

In that case, why not make the depot long and thin and keep it pointed at the Earth, not the sun? In that case, if you want some small amount of microgravity, hang a counterweight on a long cable from the nose of the depot in the direction of the Earth. (I haven't figured out how much mass/cable you'd need to get, say, 50 micro-g; it seems to involve solving a cubic, and I'm too sleepy at the moment . . .)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 02/24/2023 03:23 am
Problem is you can't "point" toward the Earth as it's nearly half the sky in LEO. So...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/24/2023 03:25 am
Good and interesting hypothesis.  Now do the math.

There's also a heat shield that can be pointed at the Earth, so the Earth is only heating the tail and the heat shield.  The tiles should absorb all that heat and release it right back at the Earth, and the  insulation will keep the stainless steel cool on that side.

The math is hard.

I doubt that a depot will have a heat shield.  And if it did, it's much more likely to be SOFI or some kind of bolt-on MLI (although I'm still not sure that works) than a refractory shield plus insulation.¹

It seems reasonable always to point a depot's nose at the Sun.
Is it, though? As we've discussed upthread, if it is far away from Earth or the moon, a depot coated in Solar White tiles will get cold enough to freeze oxygen (or close to it), even though it's otherwise unshielded from the direct sun. What messes this up in LEO is the warmth from the Earth, which comes at inconvenient wavelengths.

The depot is going to be in LEO, not in NRHO or some other BEO location.  So Earth radiation is at least a medium deal.

Also, when we say "Solar White", remember that we're talking about a coating that, last I saw, didn't actually exist.  The 90K equilibrium temperature is only achieved if the reflectance is 99% (i.e., the incident power absorbed is 1%) and the emissivity is 90%.  I suspect that there are coatings that are better than any currently in use, but they're unlikely to get down to the 90K level.

Quote
In that case, why not make the depot long and thin and keep it pointed at the Earth, not the sun? In that case, if you want some small amount of microgravity, hang a counterweight on a long cable from the nose of the depot in the direction of the Earth. (I haven't figured out how much mass/cable you'd need to get, say, 50 micro-g; it seems to involve solving a cubic, and I'm too sleepy at the moment . . .)

Because solar irradiance is about 1366W/m², and Earth irradiance maxes out at 250W/m².

_____________
¹Does anybody know if Centaur still uses the bolt-on/blow-off MLI panels?  And if so, what's on the outside of them to keep them aerodynamically intact?

Centaur used to blow those puppies off shortly after fairing jettison using pyrotechnics.  That would obviously be a problem for a vanilla-flavored Starship, but for a depot, you'd just keep them in place forever.  I'm just not sure that a vehicle as big as Starship could keep them in place reliably during launch.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 02/25/2023 03:03 am
Random thought from another thread:

It seems reasonable always to point a depot's nose at the Sun.  That way, the ogive portion of the nose can act as a dewar for the LCH4 tank below it, and very little incident radiation should strike the LCH4 or LOX tank walls.  (The Sun has an angular diameter of 0.53º, so the total incident flux should be sin (0.53º/2) = 0.0046 * solarConstant = 6.3W/m², which is almost nothing.)

However, that leaves the albedo radiation reflected/emitted from Earth.  I believe the average flux at mid-latitude inclinations is about 250W/m².

Let's divide the sphere that the depot sees into three sections:

1) The Sun.  Dealt with via pointing at it.
2) The sunlit or nighttime Earth.  250W/m², over some variable viewing angle.
3) Empty space, which is presumably cold.

Inside the depot's tanks, we will have:

a) Conductive heating when a blob of prop touches a hot tank wall.  Tank walls should only be truly hot if they're facing the sunlit Earth.

b) Conductive/convective heating, where the ullage gas picks up heat from a tank wall and transmits it thermally to a blob of liquid prop.

c) Radiative heating, where a hot wall emits IR directly into a blob of prop (bad), or some other section of wall (considerably less bad).

Radiative heating ought to be fairly modest, because any particular spot on a tank wall ought to cycle between fairly hot while directly heated by Earth and cooling when pointed at the Earth nightside or at empty space.  flux = εσT⁴, so even modest declines in the temperature of the inside wall result in much less radiative heating.

So what if we apply continuous, extremely low acceleration toward the Earth?  That should put whatever prop is in the depot as far away from a hot wall as possible, minimizing conductive heating.  This won't help much if the depot is mostly full, but before it's more than half full, this should dramatically reduce heat transfer.

Does this make sense?

Note that continuous radial acceleration in a circular orbit doesn't actually change the shape of your orbit.  It just makes you go around faster, making your true anomaly larger than it would be if you were in freefall.  So the navigational consequences of this are easy to work around.

This also requires pretty fancy RCS management to achieve, since we're holding the attitude of the depot fixed by pointing it at the Sun.  Seems like a reasonable problem, though.  Also, I'd think that whatever boiloff does occur should be able to power cold gas thrusters, both for the radial ullage burn and the burns necessary to counteract Earth's tidal forces, which want the nose to point at the center of the Earth, rather than at the Sun.
I'm quite fond of CMG's (Control Moment Gyro) for attitude control. They weigh a bit but use propellant only when desaturating. Desaturation can be done as convenient, allowing boiloff time to build enough pressure to make hot gas more practical. With luck and some creative hardware, boiloff might be low enough that the higher ISP of hot gas would be welcome.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/25/2023 07:04 am
It seems reasonable always to point a depot's nose at the Sun.
Is it, though? As we've discussed upthread, if it is far away from Earth or the moon, a depot coated in Solar White tiles will get cold enough to freeze oxygen (or close to it), even though it's otherwise unshielded from the direct sun.

Vera Rubin is going to love that....

One good approach is to keep the nose slightly "nose up" with respect to the Sun. Some additional insolation falls on the insulated tiles, but it's worth it to shade the uninsulated stainless backside.



Problem is you can't "point" toward the Earth as it's nearly half the sky in LEO. So...

You can do a pretty good job, regardless.

When the Sun is at a low angle, you point the black tiles at the Earth.

As the Sun climbs higher (and your nose pitches up to track it), you roll, keeping the black tiles aimed at the more sunlit half of the Earth—rolling clockwise when the the Sun is on the left side of the sky, anti-clock when it's on the right.

This attitude minimizes both thermal/albedo heating from Earth and nuisance reflections, because the brightest/warmest part of the Earth is seeing the black insulated tiles.

After half an orbit the Sun is now "behind" you, and the vehicle has rolled 180°, so the black tiles are again pointing at the Earth. The nose remains pointed at the Sun.



CMGs are great here. The ISS CMG would be more than sufficient, and (with decades-old technology) it masses about 1 metric ton.


You can use the same CMG during the trip to Mars. During a solar storm the main radiation direction "wiggles around" a lot, and if you can continuously align the vehicle in that direction you get improved radiation protection for the same mass. The motion is not fast (<0.1 °/s), so it shouldn't cause discomfort.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 02/26/2023 02:07 am
It seems reasonable always to point a depot's nose at the Sun.
Is it, though? As we've discussed upthread, if it is far away from Earth or the moon, a depot coated in Solar White tiles will get cold enough to freeze oxygen (or close to it), even though it's otherwise unshielded from the direct sun.

Vera Rubin is going to love that....

One good approach is to keep the nose slightly "nose up" with respect to the Sun. Some additional insolation falls on the insulated tiles, but it's worth it to shade the uninsulated stainless backside.



Problem is you can't "point" toward the Earth as it's nearly half the sky in LEO. So...

You can do a pretty good job, regardless.

When the Sun is at a low angle, you point the black tiles at the Earth.

As the Sun climbs higher (and your nose pitches up to track it), you roll, keeping the black tiles aimed at the more sunlit half of the Earth—rolling clockwise when the the Sun is on the left side of the sky, anti-clock when it's on the right.

This attitude minimizes both thermal/albedo heating from Earth and nuisance reflections, because the brightest/warmest part of the Earth is seeing the black insulated tiles.

After half an orbit the Sun is now "behind" you, and the vehicle has rolled 180°, so the black tiles are again pointing at the Earth. The nose remains pointed at the Sun.



CMGs are great here. The ISS CMG would be more than sufficient, and (with decades-old technology) it masses about 1 metric ton.


You can use the same CMG during the trip to Mars. During a solar storm the main radiation direction "wiggles around" a lot, and if you can continuously align the vehicle in that direction you get improved radiation protection for the same mass. The motion is not fast (<0.1 °/s), so it shouldn't cause discomfort.
The ISS has four, so figure four tons. Ahh, but ISS masses 419t, and SS depot variant clocks on at 85t (dry) + 1600t propellant + 20t (guesstimate for cryo cooling, shades, PV and a hefty battery pack). Call it 1705t. Have I slipped a decimal somewhere? A loaded SS depot is 4x the mass of the ISS?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/26/2023 08:35 pm

The ISS has four, so figure four tons.

Each CMG is 600 lbs (https://www.nasa.gov/vision/space/workinginspace/exp9_eva_advancer.html) (of that, the spinning mass itself is 220 lb (https://ntrs.nasa.gov/api/citations/20100021932/downloads/20100021932.pdf)). Accounting for sig figs, that's a total system mass of 1.0-1.2 metric tons.

Ahh, but ISS masses 419t, and SS depot variant clocks on at 85t (dry) + 1600t propellant + 20t (guesstimate for cryo cooling, shades, PV and a hefty battery pack). Call it 1705t. Have I slipped a decimal somewhere? A loaded SS depot is 4x the mass of the ISS?

Angular momentum is what really matters, not mass. That's a product of the moment of inertia and the maximum rate of rotation.

The ISS is over 100 meters long, so it has a large moment of inertia per mass. For Starship most of the mass is concentrated in the tanks, so it's nice and compact.

A depot is all tanks, of course, but it's still half as long (you said ~1700 tonnes so no stretch tanker). The way moment of inertia scales, that means ~1/4 the moment of inertia per mass.

And of course we're rolling the Starship, which is the axis with the lowest moment of inertia.


Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/26/2023 08:52 pm

The ISS has four, so figure four tons.

Each CMG is 600 lbs (https://www.nasa.gov/vision/space/workinginspace/exp9_eva_advancer.html) (of that, the spinning mass itself is 220 lb (https://ntrs.nasa.gov/api/citations/20100021932/downloads/20100021932.pdf)). Accounting for sig figs, that's a total system mass of 1.0-1.2 metric tons.

Ahh, but ISS masses 419t, and SS depot variant clocks on at 85t (dry) + 1600t propellant + 20t (guesstimate for cryo cooling, shades, PV and a hefty battery pack). Call it 1705t. Have I slipped a decimal somewhere? A loaded SS depot is 4x the mass of the ISS?

Angular momentum is what really matters, not mass. That's a product of the moment of inertia and the maximum rate of rotation.

The ISS is over 100 meters long, so it has a large angular momentum per mass. For Starship most of the mass is concentrated in the tanks, so it's nice and compact for a low moment of inertia.

A depot is all tanks, of course, but it's still half as long (you said ~1700 tonnes so no stretch tanker). The way moment of inertia scales, that means ~1/4 the moment of inertia per mass.

Can you use a CMG in a system with a center of mass that changes radically?  There's a pretty big difference between an empty depot and a full one.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 02/27/2023 12:02 am
Can you use a CMG in a system with a center of mass that changes radically?

Three words: ISS construction sequence.  :)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 02/27/2023 07:46 am
Can you use a CMG in a system with a center of mass that changes radically?

Three words: ISS construction sequence.  :)
Yes, a CMG does not care about the rest of the spacecraft - it will provide the same torques for the same durations regardless. Spacecraft mass properties will just determine how far that torque gets you with regard to rotations.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 02/28/2023 09:21 am

The ISS has four, so figure four tons.

Each CMG is 600 lbs (https://www.nasa.gov/vision/space/workinginspace/exp9_eva_advancer.html) (of that, the spinning mass itself is 220 lb (https://ntrs.nasa.gov/api/citations/20100021932/downloads/20100021932.pdf)). Accounting for sig figs, that's a total system mass of 1.0-1.2 metric tons.

Ahh, but ISS masses 419t, and SS depot variant clocks on at 85t (dry) + 1600t propellant + 20t (guesstimate for cryo cooling, shades, PV and a hefty battery pack). Call it 1705t. Have I slipped a decimal somewhere? A loaded SS depot is 4x the mass of the ISS?

Angular momentum is what really matters, not mass. That's a product of the moment of inertia and the maximum rate of rotation.

The ISS is over 100 meters long, so it has a large moment of inertia per mass. For Starship most of the mass is concentrated in the tanks, so it's nice and compact.

A depot is all tanks, of course, but it's still half as long (you said ~1700 tonnes so no stretch tanker). The way moment of inertia scales, that means ~1/4 the moment of inertia per mass.

And of course we're rolling the Starship, which is the axis with the lowest moment of inertia.
Didn't realize you were quoting system mass, so yeah, ~a ton.

CMB's CMG,smight might work out during transfer ops too. How tight the two ships would be physically bound is an open question as is how RCS and ullage settling would be implemented. One ship to rule them all or a cooperative effort? If they're lashed up tight and only the depot supplies the off axis settling thrust, a set of CMB's CMG's would allow attitude control without thrusters. The hard 'blap' from the F9 cold gas RCS strikes me as a bit too impulsey for a lashup.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 02/28/2023 09:34 am

The ISS has four, so figure four tons.

Each CMG is 600 lbs (https://www.nasa.gov/vision/space/workinginspace/exp9_eva_advancer.html) (of that, the spinning mass itself is 220 lb (https://ntrs.nasa.gov/api/citations/20100021932/downloads/20100021932.pdf)). Accounting for sig figs, that's a total system mass of 1.0-1.2 metric tons.

Ahh, but ISS masses 419t, and SS depot variant clocks on at 85t (dry) + 1600t propellant + 20t (guesstimate for cryo cooling, shades, PV and a hefty battery pack). Call it 1705t. Have I slipped a decimal somewhere? A loaded SS depot is 4x the mass of the ISS?

Angular momentum is what really matters, not mass. That's a product of the moment of inertia and the maximum rate of rotation.

The ISS is over 100 meters long, so it has a large angular momentum per mass. For Starship most of the mass is concentrated in the tanks, so it's nice and compact for a low moment of inertia.

A depot is all tanks, of course, but it's still half as long (you said ~1700 tonnes so no stretch tanker). The way moment of inertia scales, that means ~1/4 the moment of inertia per mass.

Can you use a CMG in a system with a center of mass that changes radically?  There's a pretty big difference between an empty depot and a full one.
Don't see why not. Just have to expect any specific input to not always give the same results.


With a continuously updated mass distribution model the depot (I'm assuming CMG's are depot specific) would have a rough idea of what input is needed and any inaccuracies in the model would only call for a bit of trim. Things might move slower as the depot fills up but that would be countered by the lashup center of mass migrating closer to the depots stand alone center of mass as it fills up.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 02/28/2023 09:50 am
While writing the two posts above it struck me that if the depot and other ship are physically tightly bound the control systems had also better be tightly bound.


Would it be best if the Depot completely took over and controlled both ships? Every cooperative arrangement that I come up with for settling thrust and attitude control seems to get more complicated the deeper I get into it. The last thing we want is the two ships fighting each other because of minor calibration issues.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: robot_enthusiast on 02/28/2023 02:24 pm
While writing the two posts above it struck me that if the depot and other ship are physically tightly bound the control systems had also better be tightly bound.


Would it be best if the Depot completely took over and controlled both ships? Every cooperative arrangement that I come up with for settling thrust and attitude control seems to get more complicated the deeper I get into it. The last thing we want is the two ships fighting each other because of minor calibration issues.
It seems like it would be a relatively straightforward thing to use the IMU's and star trackers from one ship to feed the control computers in both. AFAIK, they use Ethernet protocol networking to communicate within the vehicle, getting one of the flight computers to poll the network on the other ship shouldn't be that difficult. They generally seem to prefer to keep the control computers as close as physically possible to the thing they'll be controlling (for example, the engine controller is mounted directly to the side of each engine), so I think it's unlikely they would want to send commands across the ship-ship link. For simplicity they could also just have one ship be operating dumb, just providing a fixed amount of ullage thrust while the other handles orientation.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 02/28/2023 04:26 pm
A SpaceX example of the Master/Slaving of the Guidance computer sets is the F9/D2 stack. The Dragon is the master of everything until separation. Once a stage separates away it is then its own master.

The SS and SH will likely operate similar to F9 in the implementation of the software. The question would be which of the two vehicles Tanker/Depot would have all the added software loaded to run as a super master of a pair of SS with the other having software to be SS slave. The SS operating as a slave would not be a normal as it is currently software set on a SS. That software would probably look very similar to the software on the SH to enable it to be a slave to the SS.

SpaceX has successfully done it before on F9 so doing it for the Depot/Tanker case should not prove that difficult. Especially once they prove out the software for the Master/Slave full stack Starship launch software packages.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/28/2023 08:25 pm
While writing the two posts above it struck me that if the depot and other ship are physically tightly bound the control systems had also better be tightly bound.

Would it be best if the Depot completely took over and controlled both ships? Every cooperative arrangement that I come up with for settling thrust and attitude control seems to get more complicated the deeper I get into it. The last thing we want is the two ships fighting each other because of minor calibration issues.

It will certainly be true that the coupled pair of ships (tanker/depot, depot/target, or even tanker/target) will form a single controlled system.  Whether one ship acts as the master or they operate cooperatively is too far in the weeds even for me.

However, this has a direct impact on the physical coupling of the pair.  I doubt that it's advisable to have the pair tandem free-flying and get this to work--the control loop is just too diffuse, and even the kinematics of the pair will get weird.  Even a loosely coupled but flexible berthing will induce unpredictable oscillations that may be hard to damp out.

This is why I expect the berthing to be capable of handling decent-sized loads between the two ships with minimal flex and oscillation.



Just to review the bidding on things we've talked about for attitude control, prox ops, and ullage acceleration.  From smallest torques/translations to largest:

1) CMGs (I think you guys have convinced me that this might be useful).
2) Cold gas RCS, fed from ullage pressure only.
3) Cold gas RCS, fed from supercritical COPVs.
4) Combusting gas RCS/thrusters, fed from supercritical COPVs.

I'm pretty sure you need #4 no matter what, if for no other reason than ullage acceleration needs more impulse than can be readily generated without the enthalpy of combustion, and SpaceX likely has to develop them for lunar landing and takeoff ops.  Then the question becomes what combination of #1-#3 is needed for fine attitude control, both coupled (during refueling) and uncoupled (during attitude maintenance to minimize boil-off).

Just remember that the minimum impulse bit for any thrusters can be relatively large for Starship, because the total mass and moments of inertia are so large.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 02/28/2023 08:32 pm
While writing the two posts above it struck me that if the depot and other ship are physically tightly bound the control systems had also better be tightly bound.


Would it be best if the Depot completely took over and controlled both ships? Every cooperative arrangement that I come up with for settling thrust and attitude control seems to get more complicated the deeper I get into it. The last thing we want is the two ships fighting each other because of minor calibration issues.
It seems like it would be a relatively straightforward thing to use the IMU's and star trackers from one ship to feed the control computers in both. AFAIK, they use Ethernet protocol networking to communicate within the vehicle, getting one of the flight computers to poll the network on the other ship shouldn't be that difficult. They generally seem to prefer to keep the control computers as close as physically possible to the thing they'll be controlling (for example, the engine controller is mounted directly to the side of each engine), so I think it's unlikely they would want to send commands across the ship-ship link. For simplicity they could also just have one ship be operating dumb, just providing a fixed amount of ullage thrust while the other handles orientation.
Yeah, it's not the actual implementation (RCS or thrust) that needs coordination, it's the decision on what needs to be done. The other ship would best do what it's told while the depot, designed for just this, makes the decisions.


The depot star tracker would be the one in active use and the absolute global coordinate system would be immaterial for the other ship. Once it's under depot control it would get commands like 'rotate Z axis -3deg'.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 02/28/2023 08:40 pm
A SpaceX example of the Master/Slaving of the Guidance computer sets is the F9/D2 stack. The Dragon is the master of everything until separation. Once a stage separates away it is then its own master.

The SS and SH will likely operate similar to F9 in the implementation of the software. The question would be which of the two vehicles Tanker/Depot would have all the added software loaded to run as a super master of a pair of SS with the other having software to be SS slave. The SS operating as a slave would not be a normal as it is currently software set on a SS. That software would probably look very similar to the software on the SH to enable it to be a slave to the SS.

SpaceX has successfully done it before on F9 so doing it for the Depot/Tanker case should not prove that difficult. Especially once they prove out the software for the Master/Slave full stack Starship launch software packages.
ISTM that depot should be the master. Presumably it's the one with mission specific hardware. Why do differently for the mission specific software? The mission in this case is propellant transfer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/28/2023 09:15 pm
A SpaceX example of the Master/Slaving of the Guidance computer sets is the F9/D2 stack. The Dragon is the master of everything until separation. Once a stage separates away it is then its own master.

The SS and SH will likely operate similar to F9 in the implementation of the software. The question would be which of the two vehicles Tanker/Depot would have all the added software loaded to run as a super master of a pair of SS with the other having software to be SS slave. The SS operating as a slave would not be a normal as it is currently software set on a SS. That software would probably look very similar to the software on the SH to enable it to be a slave to the SS.

SpaceX has successfully done it before on F9 so doing it for the Depot/Tanker case should not prove that difficult. Especially once they prove out the software for the Master/Slave full stack Starship launch software packages.

Note that this is a pretty straightforward piece of software engineering.  Master/slave negotiations are used for all kinds of distributed computing.  Typically, they're just states in the finite state machine that runs your system.  After the negotiation, one side is the master and does most of the computation, while the other side is the slave and simply receives commands (events) from the master to do stuff.  How you divide up the command protocol is more art than science, and often depends on the organization of your GN&C software.

BTW:  Is the D2 in charge of most of the ascent process?  Normally, the F9 S2 has the main computing system, and is responsible for guidance and nav, but it likely relies on the core stage for control during ascent.  Following staging, it then has full control of GN&C for itself and the payload, while the core then owns all of its own GN&C through landing.

D2 can obviously decide to send an abort event to the S2, but I'd think that even then, the S2 was responsible for engine shutdown and a handshake back to the D2 indicating that it was good to do separation and escape.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 02/28/2023 09:18 pm
While writing the two posts above it struck me that if the depot and other ship are physically tightly bound the control systems had also better be tightly bound.

Would it be best if the Depot completely took over and controlled both ships? Every cooperative arrangement that I come up with for settling thrust and attitude control seems to get more complicated the deeper I get into it. The last thing we want is the two ships fighting each other because of minor calibration issues.

It will certainly be true that the coupled pair of ships (tanker/depot, depot/target, or even tanker/target) will form a single controlled system.  Whether one ship acts as the master or they operate cooperatively is too far in the weeds even for me.

However, this has a direct impact on the physical coupling of the pair.  I doubt that it's advisable to have the pair tandem free-flying and get this to work--the control loop is just too diffuse, and even the kinematics of the pair will get weird.  Even a loosely coupled but flexible berthing will induce unpredictable oscillations that may be hard to damp out.

This is why I expect the berthing to be capable of handling decent-sized loads between the two ships with minimal flex and oscillation.



Just to review the bidding on things we've talked about for attitude control, prox ops, and ullage acceleration.  From smallest torques/translations to largest:

1) CMGs (I think you guys have convinced me that this might be useful).
2) Cold gas RCS, fed from ullage pressure only.
3) Cold gas RCS, fed from supercritical COPVs.
4) Combusting gas RCS/thrusters, fed from supercritical COPVs.

I'm pretty sure you need #4 no matter what, if for no other reason than ullage acceleration needs more impulse than can be readily generated without the enthalpy of combustion, and SpaceX likely has to develop them for lunar landing and takeoff ops.  Then the question becomes what combination of #1-#3 is needed for fine attitude control, both coupled (during refueling) and uncoupled (during attitude maintenance to minimize boil-off).

Just remember that the minimum impulse bit for any thrusters can be relatively large for Starship, because the total mass and moments of inertia are so large.
I'm pretty much in agreement but think about decision by committee. Committees are good at finding the questions to ask (and a lot of irrelevant questions) and defining responses that are unacceptable, but making decisions? LoL. Everything would boil off before the two ships reach consensus.


CMG's vs thrusters. This looks like a good idea but there are so many trades it really needs to be treated as an option pending operational experience. Thrusters absolutely for translation and hot gas for ISP. HP vs LP cold gas? That gets us back to the trades. We can noodle some of this but on orbit experience will inform on what's really needed vs a hypothetical capability.


An empty depot will have a lot of mass at one end and near nothing at the other. A small vent or burn at the bottom would be a slap up side the head at the top. One trade to look at is translation while empty. Low thrust hot gas at the bottom and hypothetically, CMG's keeping the top aligned so it stays translation and not rotation. Then there is every state up to and including completely full. It can all be modeled but we don't know exactly what will be the most useful without experience. It's good to be able to do everything at high efficiency (by some arbitrary definition) but in this real world maybe the low use behaviors don't need all that much optimization if it's at he expense of the high use maneuvers.


Let's go get some experience, Yeah!
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 02/28/2023 09:31 pm
ISTM that depot should be the master. Presumably it's the one with mission specific hardware. Why do differently for the mission specific software? The mission in this case is propellant transfer.

Shouldn't make a bit of difference.  It's a computer network.  Which bits get twiddled by which CPU is an almost arbitrary choice.  If I were doing it, I'd let both CPUs be either master or slave.  No doubt you make one the preferred choice, but either one should be able to take over in the event of a failure.

The more interesting choices are how you separate the CPUs from the various I/O systems.  Do you let a master on one vehicle reach across and tweak I/O (i.e., directly command thrusters or CMGs) on the other vehicle, or do you relay commands through the slaved CPU?  Keeping everybody in sync, so any element can take over in the event of a failure, is hard.

I'm pretty much in agreement but think about decision by committee. Committees are good at finding the questions to ask (and a lot of irrelevant questions) and defining responses that are unacceptable, but making decisions? LoL. Everything would boil off before the two ships reach consensus.

It's not a committee.  It's a set of coupled FSMs that are carefully coordinated by defining a protocol containing the events needed to keep the system behaving properly.  Yes, master/slave is often a very simple (and very effective) basis for writing your state machines and defining the events.  But think about this more like cloud computing than two people arguing over the best course of action.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 02/28/2023 10:54 pm
ISTM that depot should be the master. Presumably it's the one with mission specific hardware. Why do differently for the mission specific software? The mission in this case is propellant transfer.

Shouldn't make a bit of difference.  It's a computer network.  Which bits get twiddled by which CPU is an almost arbitrary choice.  If I were doing it, I'd let both CPUs be either master or slave.  No doubt you make one the preferred choice, but either one should be able to take over in the event of a failure.

The more interesting choices are how you separate the CPUs from the various I/O systems.  Do you let a master on one vehicle reach across and tweak I/O (i.e., directly command thrusters or CMGs) on the other vehicle, or do you relay commands through the slaved CPU?  Keeping everybody in sync, so any element can take over in the event of a failure, is hard.

I'm pretty much in agreement but think about decision by committee. Committees are good at finding the questions to ask (and a lot of irrelevant questions) and defining responses that are unacceptable, but making decisions? LoL. Everything would boil off before the two ships reach consensus.

It's not a committee.  It's a set of coupled FSMs that are carefully coordinated by defining a protocol containing the events needed to keep the system behaving properly.  Yes, master/slave is often a very simple (and very effective) basis for writing your state machines and defining the events.  But think about this more like cloud computing than two people arguing over the best course of action.
In the software guidance systems it is more of a hierarchical assignment of takings. At the top is mission -> navigation -> attitude control -> engine controls/aerodynamic surfaces control. In the case of a F9/D2 the D2 is doing the first two and sometimes maybe a little of the third as well. The stages are each under command executing the next 2 tasking levels. Note is that some information like IMU data traverses up through all the layers since even some of the actions at the lowest layers need that data without delay.

In the pair of SS case it is correct to think that either computer set could do the work of the first 2 levels but only the local computers can do the last 2 levels since the control loops need to be tight and reliable. It is that which keeps the general initial stability of the pair. The higher levels must deal with control loop interference patters by 2 vehicles doing active stabilization. Fortunately it may be possible to cheat some here in that the torques involved are mostly small and slow to build to being problems. Allowing the higher level to then dampen them out.

Active control systems are a headache.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/01/2023 03:01 am
In the pair of SS case it is correct to think that either computer set could do the work of the first 2 levels but only the local computers can do the last 2 levels since the control loops need to be tight and reliable. It is that which keeps the general initial stability of the pair. The higher levels must deal with control loop interference patters by 2 vehicles doing active stabilization. Fortunately it may be possible to cheat some here in that the torques involved are mostly small and slow to build to being problems. Allowing the higher level to then dampen them out.

Active control systems are a headache.

One thing to remember is that highly precise attitude control of the coupled system after berthing/docking isn't nearly as important as highly precise attitude control of the uncoupled system during prox ops.  But the uncoupled system, rather than having some kind of master/slave relationship, has an active/passive relationship to simplify it.  One vehicle just needs to null out its rotational rates wrt the orbital frame of reference, and the other one does all the active translation and rotation to make everything line up.

Once they're linked, you can use a pretty forgiving control loop to maintain a safe attitude.  But the amount of boil-off during the coupled period is small enough that you don't need to hold the precise attitude that the depot would need to minimize direct insolation.

Still a headache, but a manageable one.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 03/02/2023 06:07 pm
A SpaceX example of the Master/Slaving of the Guidance computer sets is the F9/D2 stack.


I would say a closer analogy would be FH side boosters and core. These are both complete stages with (largely) identical avionics, flying in formation in a side-by-side configuration. They experience much greater loads than refilling, but they also have stronger connections, but both are 'no stronger than necessary.'

The Dragon is the master of everything until separation. Once a stage separates away it is then its own master.

Do we have a source for this?

I would have thought the second stage avionics would perform ascent GNC, for commonality with non-Dragon F9 flights.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 03/16/2023 05:00 pm
. . . think about decision by committee. Committees are good at finding the questions to ask (and a lot of irrelevant questions) and defining responses that are unacceptable, but making decisions? LoL. Everything would boil off before the two ships reach consensus.
When you use "committees" of computers, it's called distributed computing. There has been a great deal of work done to resolve the problem of making such a system reach consensus. In the most general case, this is a fiendishly difficult problem. Algorithms like Paxos (https://people.cs.rutgers.edu/~pxk/417/notes/paxos.html) are used to solve it, and they are not easy to wrap your head around.

They also tend to be relatively slow, so you normally only use these to let a set of contending processors elect a leader. For example, when I was at Microsoft on the project now called Bing, we had to coordinate thousands of servers which individually had half-lives of a year or so. To oversimplify a bit, we used Paxos to let them elect a leader, and, if the leader failed, to replace it. For all the problems we had with Bing over the years, Paxos never failed us. Individual servers failed, but the ensemble was immortal.

With one or more Starships trying to do formation flying, it doesn't seem like there would be all that many processors involved, so something like Paxos would probably be overkill, but it proves there is at least a solution that is known to work.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/16/2023 07:23 pm
. . . think about decision by committee. Committees are good at finding the questions to ask (and a lot of irrelevant questions) and defining responses that are unacceptable, but making decisions? LoL. Everything would boil off before the two ships reach consensus.
When you use "committees" of computers, it's called distributed computing. There has been a great deal of work done to resolve the problem of making such a system reach consensus. In the most general case, this is a fiendishly difficult problem. Algorithms like Paxos (https://people.cs.rutgers.edu/~pxk/417/notes/paxos.html) are used to solve it, and they are not easy to wrap your head around.

They also tend to be relatively slow, so you normally only use these to let a set of contending processors elect a leader. For example, when I was at Microsoft on the project now called Bing, we had to coordinate thousands of servers which individually had half-lives of a year or so. To oversimplify a bit, we used Paxos to let them elect a leader, and, if the leader failed, to replace it. For all the problems we had with Bing over the years, Paxos never failed us. Individual servers failed, but the ensemble was immortal.

With one or more Starships trying to do formation flying, it doesn't seem like there would be all that many processors involved, so something like Paxos would probably be overkill, but it proves there is at least a solution that is known to work.

As you said, figuring out which node is authoritative is an extensively solved problem.

What's considerably more difficult is the distributed control problem.  The difference between a properly controlled system and one that isn't is often determined by command and feedback lag.  That lag may often be determined by how many layers of mediation the control loop has to go through to execute a command.

If you can get a master on one Starship to directly control thrusters and read sensors on the other Starship, without having a slaved controller to mediate the commands, the control loop can be tightened up substantially.  But you need to have a very good handle on things like dropped packets and failover for that to be a viable strategy.

There was a discussion over on the Artemis thread (https://forum.nasaspaceflight.com/index.php?topic=58212.msg2466667#msg2466667) about the Gateway Docking System Specification using time-triggered ethernet (https://www.sae.org/standards/content/as6802/), which allows a synchronous, congestion-free channel to co-exist with vanilla switched ethernet.  Something like that would be a key enabler in letting one master reach deep into the I/O network of another spacecraft to shorten control lags.

I still think formation flying is nuts.  But the same problem applies to pairs of Starships that are docked and need to coordinate ullage and attitude burns.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 03/16/2023 09:48 pm
Why are people acting like Quorum computing (byzantine fault tolerant decisionmaking using multiple parallel systems) is not decades old and already in active use on Dragon - among many other non-SpaceX uses? Don't reinvent the wheel.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 03/16/2023 11:27 pm
Do flocks of geese fly together by controlling each other or by having a central authority control them all?

No, they follow the vortices of the goose in front of them.   That in turn naturally makes a V formation.

No reason why such emergent behavior can't be done for mating fuel tanks in orbit.  The math for this is still a bit obscure but I bet SpaceX knows it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 03/18/2023 07:13 pm
. . . think about decision by committee. Committees are good at finding the questions to ask (and a lot of irrelevant questions) and defining responses that are unacceptable, but making decisions? LoL. Everything would boil off before the two ships reach consensus.
When you use "committees" of computers, it's called distributed computing. There has been a great deal of work done to resolve the problem of making such a system reach consensus. In the most general case, this is a fiendishly difficult problem. Algorithms like Paxos (https://people.cs.rutgers.edu/~pxk/417/notes/paxos.html) are used to solve it, and they are not easy to wrap your head around.

They also tend to be relatively slow, so you normally only use these to let a set of contending processors elect a leader. For example, when I was at Microsoft on the project now called Bing, we had to coordinate thousands of servers which individually had half-lives of a year or so. To oversimplify a bit, we used Paxos to let them elect a leader, and, if the leader failed, to replace it. For all the problems we had with Bing over the years, Paxos never failed us. Individual servers failed, but the ensemble was immortal.

With one or more Starships trying to do formation flying, it doesn't seem like there would be all that many processors involved, so something like Paxos would probably be overkill, but it proves there is at least a solution that is known to work.

As you said, figuring out which node is authoritative is an extensively solved problem.

What's considerably more difficult is the distributed control problem.  The difference between a properly controlled system and one that isn't is often determined by command and feedback lag.  That lag may often be determined by how many layers of mediation the control loop has to go through to execute a command.

If you can get a master on one Starship to directly control thrusters and read sensors on the other Starship, without having a slaved controller to mediate the commands, the control loop can be tightened up substantially.  But you need to have a very good handle on things like dropped packets and failover for that to be a viable strategy.

There was a discussion over on the Artemis thread (https://forum.nasaspaceflight.com/index.php?topic=58212.msg2466667#msg2466667) about the Gateway Docking System Specification using time-triggered ethernet (https://www.sae.org/standards/content/as6802/), which allows a synchronous, congestion-free channel to co-exist with vanilla switched ethernet.  Something like that would be a key enabler in letting one master reach deep into the I/O network of another spacecraft to shorten control lags.

I still think formation flying is nuts.  But the same problem applies to pairs of Starships that are docked and need to coordinate ullage and attitude burns.
Yeah, formation flying is one screwup from LoM. It's necessary until the two ships are lashed up, but no further. IMO, best if one ship autonomously holds position while the other maneuvers to lash up.


I initially thought that the depot would best take the active role but if they use CMG's this might not be true. There is danger of RCS impingement. There's no point in putting CMG's on anything but the depot so it would be in the best position to react to any impingement from the other, maneuvering, ship.


Once things are lashed up and coms latency becomes important it's essentially a point to point connection. If each ship controls its own hardware the com link would be unrouted and only pass instructions like "rotate 2degrees Y+. NETBIOS over ARCnet? It'd be PDQ.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 03/18/2023 07:15 pm
Do flocks of geese fly together by controlling each other or by having a central authority control them all?

No, they follow the vortices of the goose in front of them.   That in turn naturally makes a V formation.

No reason why such emergent behavior can't be done for mating fuel tanks in orbit.  The math for this is still a bit obscure but I bet SpaceX knows it.
Get back with us when geese start mating on the fly :D


edit: that would be insertive behavior.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/18/2023 09:08 pm
Once things are lashed up and coms latency becomes important it's essentially a point to point connection. If each ship controls its own hardware the com link would be unrouted and only pass instructions like "rotate 2degrees Y+. NETBIOS over ARCnet? It'd be PDQ.

Jeez, you must be as old as I am.  The 80's called and they want their PHY and transport API back.  (I wrote an ARCnet driver for CP/M, and I implemented a NetBIOS that used an XNS transport.  How's that for a geezer smackdown?)

Seems like everything's switched ethernet today, and there's this time-triggered ethernet which will provide isochronous performance.

But I don't think you can message something like "rotate x degrees".  First of all, that command doesn't mean anything if the other ship is rotating x degrees at the same time.  Second of all, you need closed-loop control with a very, very short delay between sensor input and thruster output.  Otherwise, you'll oscillate, which consumes lots of prop if you're lucky and kills you if you're unlucky.

This is why I was suggesting that, while distributed quorum-sensing is a necessary condition, I doubt it's sufficient.  If I were doing this, I'd use quorum-sensing to designate a master GN&C CPU to seize control of the relevant I/O strings on both spacecraft, and to read sensors and command thrusters as if it were all one big system.

I don't know how tight the control loop needs to be, though.  If it's ~100ms, you can pretty much do anything you want.  But if it's <1ms, dropped packets (not an uncommon occurrence in a noisy environment) can make things weird.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TomH on 03/20/2023 12:34 am
IIRC, STS had 5 computers. Four were similar and worked in concert. If there was no consensus, decision making was shifted to the 5th computer which was dissimilar and completely independent. I am sure someone else will have better recall than I.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 03/20/2023 01:38 am
Once things are lashed up and coms latency becomes important it's essentially a point to point connection. If each ship controls its own hardware the com link would be unrouted and only pass instructions like "rotate 2degrees Y+. NETBIOS over ARCnet? It'd be PDQ.

Jeez, you must be as old as I am.  The 80's called and they want their PHY and transport API back.  (I wrote an ARCnet driver for CP/M, and I implemented a NetBIOS that used an XNS transport.  How's that for a geezer smackdown?)

Seems like everything's switched ethernet today, and there's this time-triggered ethernet which will provide isochronous performance.

But I don't think you can message something like "rotate x degrees".  First of all, that command doesn't mean anything if the other ship is rotating x degrees at the same time.  Second of all, you need closed-loop control with a very, very short delay between sensor input and thruster output.  Otherwise, you'll oscillate, which consumes lots of prop if you're lucky and kills you if you're unlucky.

This is why I was suggesting that, while distributed quorum-sensing is a necessary condition, I doubt it's sufficient.  If I were doing this, I'd use quorum-sensing to designate a master GN&C CPU to seize control of the relevant I/O strings on both spacecraft, and to read sensors and command thrusters as if it were all one big system.

I don't know how tight the control loop needs to be, though.  If it's ~100ms, you can pretty much do anything you want.  But if it's <1ms, dropped packets (not an uncommon occurrence in a noisy environment) can make things weird.
OMG! CPM Arcnet? I yield. You be the mastah of the geezers. What are the chances that what we been thinking was a slot for ejecting starlinks is really where they insert the boot floppy?

And I'm sorry, but a physical layer that is not probabilistic ain't ethernet. Calling it ethernet is just marketing bs and is disrespectful to an honest data storm.

I've spent the last week resurrecting a 98SE tower. The only I/O was an IDE HDD with a USB/IDE dongle for the laptop end, then plug it into the tower on the second IDE channel to copy the files it needs. Putting Lantastic on it is tempting but it's got nothing to talk to.

I think the project has been getting to me.


Edit: know anybody who needs a lightly used Osborn?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/20/2023 02:18 am
IIRC, STS had 5 computers. Four were similar and worked in concert. If there was no consensus, decision making was shifted to the 5th computer which was dissimilar and completely independent. I am sure someone else will have better recall than I.

The crew had to manually switch to the backup if things went horribly wrong.  The idea was that a completely different implementation was unlikely to have the same bug in the same place.

The set of 4 identical CPUs would all receive the same input, compute the outputs, and exchange state vectors every 40ms.  If one of the state vectors didn't agree with the others, the other 3 would take over the bad one's output duties.  Crude but effective.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 03/20/2023 02:23 am
Once things are lashed up and coms latency becomes important it's essentially a point to point connection. If each ship controls its own hardware the com link would be unrouted and only pass instructions like "rotate 2degrees Y+. NETBIOS over ARCnet? It'd be PDQ.

Jeez, you must be as old as I am.  The 80's called and they want their PHY and transport API back.  (I wrote an ARCnet driver for CP/M, and I implemented a NetBIOS that used an XNS transport.  How's that for a geezer smackdown?)

Seems like everything's switched ethernet today, and there's this time-triggered ethernet which will provide isochronous performance.

But I don't think you can message something like "rotate x degrees".  First of all, that command doesn't mean anything if the other ship is rotating x degrees at the same time.  Second of all, you need closed-loop control with a very, very short delay between sensor input and thruster output.  Otherwise, you'll oscillate, which consumes lots of prop if you're lucky and kills you if you're unlucky.

This is why I was suggesting that, while distributed quorum-sensing is a necessary condition, I doubt it's sufficient.  If I were doing this, I'd use quorum-sensing to designate a master GN&C CPU to seize control of the relevant I/O strings on both spacecraft, and to read sensors and command thrusters as if it were all one big system.

I don't know how tight the control loop needs to be, though.  If it's ~100ms, you can pretty much do anything you want.  But if it's <1ms, dropped packets (not an uncommon occurrence in a noisy environment) can make things weird.
In a more serious vein, the data environment before and after lash up are very different. Pre lash up, ISTM that keeping one ship passive while the other maneuvers keeps the risk low. Passive in this case doesn't mean dead. It means holding attitude but not otherwise maneuvering.


After lash up the rules change. Both ships need to (probably) work together for ullage settling and attitude control. But...


If the depot has CMG's and can thrust off axis with TVC on the settling engine, it should be able to handle everything. And, once they're lashed up they can have a physical low noise data connection. As you suggested, the second ships sensors would become an extension of the depot. Control systems could be tied in too if necessary.


Some questions:
- What are the chances the sensors and controls on SS are a data buss sharing network?
- Can off axis thrust be made a virtue during transfer? It imposes a certain type of order on slosh.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/20/2023 05:46 pm
In a more serious vein, the data environment before and after lash up are very different. Pre lash up, ISTM that keeping one ship passive while the other maneuvers keeps the risk low. Passive in this case doesn't mean dead. It means holding attitude but not otherwise maneuvering.

After lash up the rules change. Both ships need to (probably) work together for ullage settling and attitude control. But...

If the depot has CMG's and can thrust off axis with TVC on the settling engine, it should be able to handle everything. And, once they're lashed up they can have a physical low noise data connection. As you suggested, the second ships sensors would become an extension of the depot. Control systems could be tied in too if necessary.

Some questions:
- What are the chances the sensors and controls on SS are a data buss sharing network?
- Can off axis thrust be made a virtue during transfer? It imposes a certain type of order on slosh.

The data bus is likely 1000BaseT ethernet.  I assume that it's switched, because everything is these days.  So it should be quite easy to make one network out of two Starships.

I obviously don't know the exact architecture of the network.  Never stops me from guessing, though.  For long runs (e.g. across the prop tanks), I'd use redundant wiring, with switches at either end.  From those switches, which no doubt have all sorts of vibration and noise hardening, you'd have fairly short runs to other distribution switches, which would feed individual engine controllers, gimbals, thrusters, heaters, coolers, IMUs, CMGs, etc.

The trick would be to minimize mass.  A switch almost certainly weighs less than two 20m Cat 6 cables, but it probably doesn't weigh less than 3 pairs of 2m cables.  No doubt somebody spent a lot of time of topology.  But once you get everything wired up, it's... just another LAN.  Give each Starship its own private subnet and they'll happily talk to one another.  Then it's just a question of how you allocated the computational resources.

I don't think off-axis thrust is ever a virtue.  But with full access to all the ullage thrusters, it's not difficult to keep things from rotating.  Even if the depot has CMGs, I don't think you want to load them as much as they'd have to be loaded to balance a 30-60 minute ullage burn during transfer.

BTW, here's the wikipedia article on time-triggered ethernet (https://en.wikipedia.org/wiki/TTEthernet).  It's not really synchronous, but it's effectively isochronous for certain traffic types.  Kinda clever.  This is supposed to be what's used in the Gateway Docking System Specification, which is pressure- and latch-compatible with IDSS, but has different electrical and fluid connectors.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: oldAtlas_Eguy on 03/20/2023 06:43 pm
I was thinking about some additional sensors to be used for refining the GNC to be able to make the second SS fly very precisely along with the primary SS. And that is nothing more than a bunch of strain gauges that report 3 axis load values at the hard point or even semi-soft point connections between the 2 SS. By trying to maintain as low of a 3 axis loads through these connection points the secondary SS adjusts its actions to basically fly precisely along with the primary SS. NOTE is that as prop flows from one to the other this method inherently can provide accurate feedback to maintain flight and stability.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/20/2023 08:36 pm
I was thinking about some additional sensors to be used for refining the GNC to be able to make the second SS fly very precisely along with the primary SS. And that is nothing more than a bunch of strain gauges that report 3 axis load values at the hard point or even semi-soft point connections between the 2 SS. By trying to maintain as low of a 3 axis loads through these connection points the secondary SS adjusts its actions to basically fly precisely along with the primary SS. NOTE is that as prop flows from one to the other this method inherently can provide accurate feedback to maintain flight and stability.

You'd need torsional or bending loads as well as tension/compression.  I still think you're probably better with IMUs and/or visual trackers, and centralized control of both ships as a single system.  The number of degrees of freedom needed to manage them as a two-body system that needs to maintain translational and rotational coherence gets out of hand pretty quickly.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 03/21/2023 07:38 am
I was thinking about some additional sensors to be used for refining the GNC to be able to make the second SS fly very precisely along with the primary SS. And that is nothing more than a bunch of strain gauges that report 3 axis load values at the hard point or even semi-soft point connections between the 2 SS. By trying to maintain as low of a 3 axis loads through these connection points the secondary SS adjusts its actions to basically fly precisely along with the primary SS. NOTE is that as prop flows from one to the other this method inherently can provide accurate feedback to maintain flight and stability.

You'd need torsional or bending loads as well as tension/compression.
6 strain gauges (plus redundancy), then.
Quote
I still think you're probably better with IMUs and/or visual trackers, and centralized control of both ships as a single system.  The number of degrees of freedom needed to manage them as a two-body system that needs to maintain translational and rotational coherence gets out of hand pretty quickly.
6 DoF, not excessive. And since the goal is to minimise mechanical loads on the coupler, then directly measuring mechanical loads on the coupler is the gold standard compared to IMUs/RADAR/LIDAR/etc.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/21/2023 09:01 pm
6 DoF, not excessive. And since the goal is to minimise mechanical loads on the coupler, then directly measuring mechanical loads on the coupler is the gold standard compared to IMUs/RADAR/LIDAR/etc.

Hmm, I seem to have lost my response to this.  Trying again:

You need at least nine degrees of freedom:

1) 3 for translation.
2) 3 for vehicle #1 rotation.
3) 3 for vehicle #2 rotation.

I don't think figuring everything via strain gauges will work.  Consider two cases:

a) Two ships drifting apart along the y-axis (x = fore/aft, y=left/right, z=up/down).
b) Two ships rotating about a common axis of inertia that happens to be parallel with the x-axis.

These should have identical strains, but they certainly don't have identical control actions to null the rates.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 03/22/2023 08:10 pm
In a more serious vein, the data environment before and after lash up are very different. Pre lash up, ISTM that keeping one ship passive while the other maneuvers keeps the risk low. Passive in this case doesn't mean dead. It means holding attitude but not otherwise maneuvering.

After lash up the rules change. Both ships need to (probably) work together for ullage settling and attitude control. But...

If the depot has CMG's and can thrust off axis with TVC on the settling engine, it should be able to handle everything. And, once they're lashed up they can have a physical low noise data connection. As you suggested, the second ships sensors would become an extension of the depot. Control systems could be tied in too if necessary.

Some questions:
- What are the chances the sensors and controls on SS are a data buss sharing network?
- Can off axis thrust be made a virtue during transfer? It imposes a certain type of order on slosh.

The data bus is likely 1000BaseT ethernet.  I assume that it's switched, because everything is these days.  So it should be quite easy to make one network out of two Starships.

I obviously don't know the exact architecture of the network.  Never stops me from guessing, though.  For long runs (e.g. across the prop tanks), I'd use redundant wiring, with switches at either end.  From those switches, which no doubt have all sorts of vibration and noise hardening, you'd have fairly short runs to other distribution switches, which would feed individual engine controllers, gimbals, thrusters, heaters, coolers, IMUs, CMGs, etc.

The trick would be to minimize mass.  A switch almost certainly weighs less than two 20m Cat 6 cables, but it probably doesn't weigh less than 3 pairs of 2m cables.  No doubt somebody spent a lot of time of topology.  But once you get everything wired up, it's... just another LAN.  Give each Starship its own private subnet and they'll happily talk to one another.  Then it's just a question of how you allocated the computational resources.

I don't think off-axis thrust is ever a virtue.  But with full access to all the ullage thrusters, it's not difficult to keep things from rotating.  Even if the depot has CMGs, I don't think you want to load them as much as they'd have to be loaded to balance a 30-60 minute ullage burn during transfer.

BTW, here's the wikipedia article on time-triggered ethernet (https://en.wikipedia.org/wiki/TTEthernet).  It's not really synchronous, but it's effectively isochronous for certain traffic types.  Kinda clever.  This is supposed to be what's used in the Gateway Docking System Specification, which is pressure- and latch-compatible with IDSS, but has different electrical and fluid connectors.
Yeah, off axis is not the best of things. Just wondering if there's any way of making lemonade.

In another post you set a goal of minimizing load through the lash up mechanism and followed it up with an expansion from 6 to 9 DoF to reach that goal.

Not too much mass gives a lot of structural strength. If a couple hundred kilos of mass potentially simplifies control it shouldn't be dismissed out of hand. A lot depends on control latency and adverse feedback loops between ships and the actual forces needed in the worst case.

Worst case for off axis settling thrust would be a near empty depot topping off a near full receiving ship. And whatever there is from slosh, which IIRC, is still an open question. Off axis guarantees slosh but it makes direction, if not magnitude, predictable. This is where a way to make lemonade might be handy.


Edit: added 'off axis'
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/23/2023 02:47 am
Not too much mass gives a lot of structural strength. If a couple hundred kilos of mass potentially simplifies control it shouldn't be dismissed out of hand. A lot depends on control latency and adverse feedback loops between ships and the actual forces needed in the worst case.

Worst case for off axis settling thrust would be a near empty depot topping off a near full receiving ship. And whatever there is from slosh, which IIRC, is still an open question. Off axis guarantees slosh but it makes direction, if not magnitude, predictable. This is where a way to make lemonade might be handy.

I don't think this is particularly hard with a centralized processor and a distributed control bus (i.e., a LAN that can talk to everything on both ships).  It's only hard if you insist on treating it as a distributed control problem.  It doesn't have to be--other than designating a master and a failover, which is a solved problem.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: talltim007 on 03/30/2023 05:03 pm
I may have missed it earlier in this thread, but why not use baffles or diffusers to deal with sloshing in the receiving tank?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 03/30/2023 08:07 pm
I may have missed it earlier in this thread, but why not use baffles or diffusers to deal with sloshing in the receiving tank?
IANARS, but what I think I see is that other than speculation, we can't characterize the slosh. There are too many options on how things will be done and no firm ground on which to base informed speculation.


IIU slosh correctly, baffles can be general purpose or specific to an angle and velocity. Might as well throw in fluid depth. Speculation: general purpose masses more and overall dampens less than task specific.


One thing that sets NSF apart from many sites is we sometimes recognize when the speculation is getting too rarified, and we shy away. Unfortunately, these speculations often get ruled out instead of getting a 'hold that thought' label.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 03/31/2023 11:52 am
I may have missed it earlier in this thread, but why not use baffles or diffusers to deal with sloshing in the receiving tank?
Baffles and diffusers are already present to manage sloshing under thrust. The problem is that baffles and diffusers do not do much for propellant movement in microgravity; that's the realm of Propellant Management Devices. Baffles are also less effective as the tank drains, as that is when the surface-area/volume grows rapidly, the number of baffles immersed in the propellant drops, and the distance the surface neds to move in order to uncover the inlet reaches a minimum.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 03/31/2023 07:23 pm
I don't think the orbital fuel transfer system will work by acceleration.

It'll work by tricks with surface tension:

https://www.youtube.com/watch?v=1FxYleYSWHM
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 03/31/2023 07:24 pm
I bet it will.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: BT52 on 04/01/2023 04:45 am
I bet it will.

Sorry to bother but in witch sense. I think acceleration is most straight forward way to generate similar situation as knowledge from second stage tank simulation can be taking into account.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 04/01/2023 04:53 am
I bet it will.

Sorry to bother but in witch sense. I think acceleration is most straight forward way to generate similar situation as knowledge from second stage tank simulation can be taking into account.
Yeah. I meant that I think gentle acceleration will be used.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 04/11/2023 06:02 pm
https://m.youtube.com/watch?v=921VbEMAwwY (https://m.youtube.com/watch?v=921VbEMAwwY)


It looks like crew SS and (presumably) the depot are doing full length contact. Of course the perennial question arises of the art departments veracity. My guess is they show what the current engineering thoughts are with some license but no contradictions.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/12/2023 03:07 am
6 DoF, not excessive. And since the goal is to minimise mechanical loads on the coupler, then directly measuring mechanical loads on the coupler is the gold standard compared to IMUs/RADAR/LIDAR/etc.

...You need at least nine degrees of freedom:

1) 3 for translation.
2) 3 for vehicle #1 rotation.
3) 3 for vehicle #2 rotation.

The system has 9 DoF, but the stresses on the connector still only have 6 DoF (3-axis linear force + 3 axis torques). Therefore it only requires 6 strain gauge channels, plus redundancy if needed.


I don't think figuring everything via strain gauges will work.

Who does?

Has anyone proposed getting rid of gyroscopes, accelerometers, star trackers, and GPS and using only the strain gauges? I haven't seen such a proposal.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 04/19/2023 11:24 pm
Has anyone proposed getting rid of gyroscopes, accelerometers, star trackers, and GPS and using only the strain gauges? I haven't seen such a proposal.

No, nobody was proposing gutting the GN&C sensor suite.  But the argument was whether you could implement distributed control using just the strain gauges.  This all fell out from a discussion about distributed vs. centralized control for the docked (or loosely-coupled) system.

I'm in the centralized control camp, because gluing the two avionics networks together via an ethernet (xBaseT) connection is easy, and master/slave determination (oops, sorry--leader/follower determination) is a solved problem.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/20/2023 03:42 am
Has anyone proposed getting rid of gyroscopes, accelerometers, star trackers, and GPS and using only the strain gauges? I haven't seen such a proposal.

No, nobody was proposing gutting the GN&C sensor suite.  But the argument was whether you could implement distributed control using just the strain gauges.

I think you're just misreading.

The first post to mention strain gauges says:

I was thinking about some additional sensors to be used for refining the GNC to be able to make the second SS fly very precisely along with the primary SS. And that is nothing more than a bunch of strain gauges that report 3 axis [sic] load values at the... connections

I think you read "nothing more" and misinterpreted it as discarding all the other sensors, despite the bold text. But Atlas just meant that those are the only additional sensors and nothing more beyond that.

Agreed, centralized control is the way.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 04/20/2023 04:21 am
Hey, has anybody looked carefully at the refueling animation on the SpaceX Starship page?  S24 has two raceways on its dorsal surface, but they don't look exactly like the ones in the animation.  Are there any details to be gleaned on docking refueling connectors from the animation?

It's possible, maybe even likely, that the animation is notional or fanciful, but there might be some hints in there.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/20/2023 04:37 am
As seen on TV:

https://www.youtube.com/watch?v=921VbEMAwwY#t=145s

Looks like a pair of twin "standpipes" on the dorsal midline, going from the top of the fuel tank to the bottom of the LOX tank. Features are also seen at the top of the LOX tank, so we may be looking at 3-4 external pipes not just 2.

Refilling fluid connections between ships are made at the top and bottom of the standpipes. Minimal design.

Above the standpipes looks like a combined zero-thrust fuel vent + pair of RCS thrusters.

Below, it looks like the Stage 0 umbilical port. I don't see anything on either side that would connect over, so it looks like it's unused during refilling.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 04/20/2023 04:46 am
As seen on TV:

Thanks, that's better than the web page.

The first thing to note is that it's a double raceway in the refueling scene and a single raceway in the scene immediately following it.  That would seem to require one to increase one's "fanciful" priors.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 04/20/2023 05:30 am
It's not uncommon to model a "highlight" feature in higher fidelity for the scene it's being highlighted in, then use a lower-fidelity model for other shots. But the fact that they did model a more detailed version is probably not an accident.

This design makes a lot of sense to me. The LOX and CH4 ports are maximally far apart, minimizing the chances of detonation in case of a leak.

The "standpipes" bring ullage GOX down from the top of the oxygen tank, and bring liquid fuel up from the bottom of the methane tank. So we're actually looking at multiple pipes (3-4), even though they may not be modeled as such.

The docking procedure is nice and symmetrical, without any large torque or slosh concerns. The docking hardware is located maximally in the lee of the superheated reentry plasma, minimizing wear and tear. The docking hardpoints are located far apart to efficiently transfer torque, and they're both located at a dome-wall interface which is already stronger to begin with. Very nice.

The only thing I'm not sure about is, why run the standpipes on the outside? It seems like they would be more protected inside. It's the same propellant inside the pipe as in the tank, so it's not about separating fuel and oxidizer. So why?  ???



Edit: Ok, I'm an idiot. All the pipes I mentioned are located inside the tanks.

The standpipes are actually the contingency transfer pipes. If one of the two cryo fluid + ullage gas connectors fails, they can simply purge the connector to vacuum, throw a couple valves in each ship, and use the other connector to transfer the stranded propellant.  :o  It takes over twice as long, but this way the entire tanker mission won't be a waste. Genius!

This explains why they're on the outside, and why each appears to be one continuous pipe directly connecting the fuel and oxidizer ports, and why there are two (liquid and gas).

By doing this they can minimize mission risk during the development of their cryogenic propellant transfer technology.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 04/30/2023 05:24 pm
EVLEO on-orbit ISRU manufacture of ASCENT propellant (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2417457#msg2417457) (formerly AF-M315E) for tugs, payloads, etc. could be accomplished within a closed electrochemical nitrogen cycle.  Notably, ASCENT vacuum-manufacture is easier in space than on the ground.

Production of ammonia and nitrate is only part of the greater challenge, but it's remarkable that high efficiency is reported in recent experiments, e.g., nearly 100% efficiency in electrochemical ammonia synthesis (https://www.researchgate.net/publication/362200475_Electroreduction_of_nitrogen_at_almost_100_current-to-ammonia_efficiency). 

Q:  What is the most efficient electrochemical nitrate synthesis method?

Each constituent of an ASCENT tank load -- pressurant and ionic liquid -- can be sourced in this nitrogen cycle, with 96% of the mass potentially accessible in the thermosphere.

Some fraction of the remaining feedstock (H2) is potentially sourced from docking Starships and other ships.  E.g., H2 can be reduced out of excess, waste Starship gCH4 pressurant, prior to undocking and EDL.

N2 is an abundant feedstock molecule, so in this application only two reactions would be needed from the electrochemical nitrogen cycle: 

1.  nitrogen reduction reaction (N2RR) for ammonia, and

2.  nitrogen oxidation reaction (N2OR) for nitrate.

Refs.

Du, H.L., Chatti, M., Hodgetts, R.Y., Cherepanov, P.V., Nguyen, C.K., Matuszek, K., MacFarlane, D.R. and Simonov, A.N., 2022. Electroreduction of nitrogen with almost 100% current-to-ammonia efficiency. (https://www.researchgate.net/publication/362200475_Electroreduction_of_nitrogen_at_almost_100_current-to-ammonia_efficiency) Nature, 609(7928), pp.722-727.

McLean, C.H., Viesca, S.G., Deininger, W.D., Unruh, B.W., Spores, R.A., Frate, D.T., Yim, J.T., Johnson, W.L., Aggarwal, P.K. and Reed, B.D., 2014, March. Green propellant infusion mission program overview and status. (https://www.semanticscholar.org/paper/Green-Propellant-Infusion-Mission-program-overview-McLean-Viesca/e98186d6ace8e62275f0b82a7391f23bc4b23bfd) In 2014 IEEE Aerospace Conference (pp. 1-20). IEEE.

Yang, X., Mukherjee, S., O'Carroll, T., Hou, Y., Singh, M.R., Gauthier, J.A. and Wu, G., 2023. Achievements, Challenges, and Perspectives on Nitrogen Electrochemistry for Carbon‐Neutral Energy Technologies. (https://www.researchgate.net/publication/366214555_Achievements_Challenges_and_Perspectives_on_Nitrogen_Electrochemistry_for_Carbon-Neutral_Energy_Technologies) Angewandte Chemie, 135(10), p.e202215938.
 
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 08/07/2023 12:45 am
Did anyone consider surface tension methods?

https://en.wikipedia.org/wiki/Propellant_management_device
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/07/2023 02:49 am
Did anyone consider surface tension methods?

https://en.wikipedia.org/wiki/Propellant_management_device

I think a weakness of the prior discussion is that everyone pretty much assumed that the general problem of extracting cryogens from a tank in zero or micro-gravity was already solved, so there hasn't been much said about it. (Or else I somehow missed it.)

Using search terms from the Wikipedia article, I found a more recent conference article, "A Detailed Historical Review of Propellant Management Devices for Low Gravity Propellant Acquisition (https://ntrs.nasa.gov/api/citations/20170000667/downloads/20170000667.pdf)," which is very informative. It suggests that capillary systems are only necessary if the ullage burn is under 1E-4 g. (Of course, you might need it to feed the ullage engine just to get the burn started, I suppose.) Otherwise, "a hole in the bottom of the tank" could be all you'd need.

That article suggests that, of the different surface-tension methods, only the "Screen Channel Liquid Acquisition Devices" aka "Gallery Arms" have the ability to manage the large propellant flows that a depot would require. (p. 15 "C. Advantages and Disadvantages.") The challenge would apparently be to make one that was robust enough for repeated use.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/13/2023 10:35 pm
I think you could get useful amounts of artificial gravity if you simply tethered two depots together with cables no more than 500m long. While ullage burns are probably fine, this does suggest an alternative way to do it.

Suppose you have two depots of equal mass connected by a cable of length 2L such that the cable is always on a line through the center of the Earth. Assume the midpoint of the cable is in a circular orbit at a height, h, of 100 km, so the ensemble has a period of about 85 minutes and the angular velocity, ω, will be about 0.0012 radians/sec.

At distance L from the center of mass, the centrifugal acceleration will be ω²(R+h+L), where R is the radius of the Earth. The acceleration due to gravity will be -μ/(R+h+L)². Using the fact that L is much smaller than R+h, the acceleration at the two depots is very close to 3ω²L. If you wanted 100 micro-gravities, L of 222 would suffice, for a total cable length of 444m. For a milli-g, you'd need a 4.4 km cable, which is still not unreasonable. (Someone should check my math though.)

The rotation should be slow enough that a vehicle could dock with either one without any need to stop the rotation.

Obviously there are stability issues, but if you make one depot heavier than the other, it will tend to stay down. Or just make a counterweight that's heavier than a full depot plus Starship. Maybe that's still too much work just to avoid ullage burns, but it doesn't seem to add all that much complexity. Or so it seems to me.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 08/14/2023 07:40 am
I think you could get useful amounts of artificial gravity if you simply tethered two depots together with cables [...]

1) Depending how you do the cables, you can control the rotation of 1 depot around its CoM independently of its motion about the cable - at least for short periods << 2π/ω, at much lower cost than the docking tanker/ship constantly thrusting to 'hover' and dock. So your ω can be perhaps 6x faster (depending on how long you need to rendezvous and dock)
[EDIT: you lose the favourable tidal forces, but get much better ω². Solving for 10mms⁻² = ω²L - (tidal force with cable horizontal)... For 15 min rotation, ω = π/450. Tidal force is μ/(R+h)² * L/(R+h) i.e. gravity vector misaligned by L/(R+h) radians. 2L ~= 425m.]

2) I feel there should be a much easier solution, just involving the 2 rendezvousing vehicles spinning, and attention paid to where the joint centre of mass is (as it migrates) relative to places in each tank you can pump from. (Or open a valve from, while you pump gas into the tank)
[EDIT: Tankers/depots would need an extra sump/outlet in a certain place in the forward dome and near the top of the aft dome. For 10mms⁻², ω is on the order of 0.03 rad/s (3 min rotations), relevant distances from CoM in the range ~10 - 40m. Note that only rotation about the axis of highest moment of inertia is stable, so cartwheeling over each other]

The surface tension methods are growing on me, but if they need a bit of acceleration to squeeze the last couple of tonnes out of a tanker... well those last tonnes won't be worth it via linear acceleration if the depot is near full. There will be waste.
[EDIT: Job done, the tanker needs those last few tonnes of fuel for EDL anyway.]

1st priority - just get something that works so HLS can progress - linear acceleration
2nd priority - make a simple, reliable system that isn't too wasteful when the depot is near full - surface tension methods
3rd priority - get every last drop out of a tanker as efficiently as possible, tolerating complexity - rotation methods
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 08/14/2023 11:42 am
For spin or gravity-gradient settling, what you gain in not using propellant for settling thrust you may well lose in power needed for pumping. A microgree settling thrust and side-by-side transfer means a very small pressure differential - achievable by ullage venting - is sufficient for total propellant transfer from one tank to another. At the ends of a >200m tether, you now have a non-negligible pump head to deal with, along with an orientation change. There is also the propellant needed to set up for the transfer: for spin-settling, that's spinup and spindown thrust. For gravity-gradient, you still are effectively performing orbit raising/lowering burns to move the vehicles to their respective offset altitudes, though for most transfers you could probably omit return burns (though the next tanker would still be burning incrementally for initial insertion longer as the depot orbit raises).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: rsdavis9 on 08/14/2023 01:46 pm
I think you could get useful amounts of artificial gravity if you simply tethered two depots together with cables [...]

1) Depending how you do the cables, you can control the rotation of 1 depot around its CoM independently of its motion about the cable - at least for short periods << 2π/ω, at much lower cost than the docking tanker/ship constantly thrusting to 'hover' and dock. So your ω can be perhaps 6x faster (depending on how long you need to rendezvous and dock)
[EDIT: you lose the favourable tidal forces, but get much better ω². Solving for 10mms⁻² = ω²L - (tidal force with cable horizontal)... For 15 min rotation, ω = π/450. Tidal force is μ/(R+h)² * L/(R+h) i.e. gravity vector misaligned by L/(R+h) radians. 2L ~= 425m.]

2) I feel there should be a much easier solution, just involving the 2 rendezvousing vehicles spinning, and attention paid to where the joint centre of mass is (as it migrates) relative to places in each tank you can pump from. (Or open a valve from, while you pump gas into the tank)
[EDIT: Tankers/depots would need an extra sump/outlet in a certain place in the forward dome and near the top of the aft dome. For 10mms⁻², ω is on the order of 0.03 rad/s (3 min rotations), relevant distances from CoM in the range ~10 - 40m. Note that only rotation about the axis of highest moment of inertia is stable, so cartwheeling over each other]

The surface tension methods are growing on me, but if they need a bit of acceleration to squeeze the last couple of tonnes out of a tanker... well those last tonnes won't be worth it via linear acceleration if the depot is near full. There will be waste.
[EDIT: Job done, the tanker needs those last few tonnes of fuel for EDL anyway.]

1st priority - just get something that works so HLS can progress - linear acceleration
2nd priority - make a simple, reliable system that isn't too wasteful when the depot is near full - surface tension methods
3rd priority - get every last drop out of a tanker as efficiently as possible, tolerating complexity - rotation methods

Since you only need microgees I like a hub that starships would connect to nose first. Nice and stable and nose first keeps acceleration same as on ground. Hub could support 6 starships all nose in easily. Hosing could be flexible from hub to QD plate. Small flyers with thrust or maybe long RMS (remote manipulator system) like system.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 08/15/2023 01:12 am
For spin or gravity-gradient settling, what you gain in not using propellant for settling thrust you may well lose in power needed for pumping.
True, but also electric power is free

At the ends of a >200m tether, you now have a non-negligible pump head to deal with
Assuming the other end of the tether is simply a spare depot acting as a counterweight, then depending on docking geometry relative to the tether, you could be 'pumping' downhill

There is also the propellant needed to set up for the transfer: for spin-settling, that's spinup and spindown thrust. For gravity-gradient, you still are effectively performing orbit raising/lowering burns
1) For the gravity-gradient scenario being discussed, the only setup is the docking maneuver with one depot in an already set up system, i.e. a small amount of 'hover' thrust on the final approach. (though docking to the top one does mean going to a higher energy orbit)

2) For spin (or pumping along a tether), true there is such a cost. The premise is that propellant transfer is a slow process and set up acceleration << total acceleration i.e. the propellant transfer time t >> 2/ω. Is this a valid assumption?

the next tanker would still be burning incrementally for initial insertion longer as the depot orbit raises

That... is actually very interesting. I had assumed linear acceleration was wasted ∆v, but perhaps not. If propellant transfer is always done prograde at perigee, then that's the direction that fuel was eventually going to go anyway on basically any mission that needs refueling. Some mission profiles actually require a top-up in a very high orbit. So you have depots going through cycles of
- start in a low orbit
- mainly getting filled up and the orbit drifting to a mid-height apogee
- mainly getting depleted and the orbit drifting to a high apogee
- one final top-up for a particularly high ∆v mission, leaving the depot with just enough fuel for
- empty depot orbit lowering, potentially employing aerobraking?
- start the cycle again

There is still the problem of ullage thrust having a horrible ISP (can the depot have an extra tiny methane engine? with electric turbos?), and that the extra orbital energy might not be in the exact direction the missions need. But it still reduces linear thrust settling from a total waste to only a partial waste.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 08/15/2023 12:02 pm
There is still the problem of ullage thrust having a horrible ISP
From the NASA NEA mission proposal (https://ntrs.nasa.gov/citations/20230003852), the hot-gas thruster ISP is 295s. Not the greatest ISP, but not terrible either.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 08/15/2023 12:07 pm
Note that only rotation about the axis of highest moment of inertia is stable, so cartwheeling over each other]
Don't get carried away by the intermediate axis theorem.  You want an engineering definition of stability not a pure math one.  I.e. with time constants and the amount of effort to stabilize "unstable" rotations.  Then compare that the effort for the "stable" alternative.

On the one hand avoiding an instability is good, on the other hand there could be many other factors that are more important than avoiding a weak instability.

For example: the Apollo barbeque roll, around the long axis with a 20-minute period, was "unstable".  This did not appear to cause any angst.  It probably did cause a few reaction control system operations.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 08/15/2023 01:32 pm
Note that only rotation about the axis of highest moment of inertia is stable, so cartwheeling over each other]
[...] For example: the Apollo barbeque roll, around the long axis with a 20-minute period, was "unstable".  This did not appear to cause any angst.  It probably did cause a few reaction control system operations.
That's a good point. Starship tankers will be less rigid body motion and more fluid dynamics, so that complicates things. Also if one ship has a CoM more forward than the other (e.g. because the tanker has header tanks) then I worry the other axes will be very unstable... But I grant you that this only means "I don't know how to analyze it" as opposed to "it's impossible". Maybe there's a way to e.g. deliberately transfer fuel and LOX at different rates to manage ballast and somehow make it all work within acceptable RCS use.

Is there a verdict on whether 1 milli g is enough? Would it be sensitive to vibrations?

Maybe the best result is a combination of subtle gravity and very localized surface tension features. surface tension has a short distance and so good gradient around the outlet to control bubbles, and gravity just makes sure it has enough liquid in the vicinity to work with.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 08/15/2023 01:36 pm
Note that only rotation about the axis of highest moment of inertia is stable
It is only the intermediate axis (if present, which is not always the case for some axisymmetric objects) that is unstable, the other two axes are stable. Spin-stabilised satellite stacks (particularly long narrow ones with a heavy SRM at the bottom, with the CoM moving forwards as the motor burns, and perturbations from the SRM attempting to rotate the stack being the reason for the spin in the first place) work, after all.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/15/2023 01:44 pm
For spin or gravity-gradient settling, what you gain in not using propellant for settling thrust you may well lose in power needed for pumping. A microgree settling thrust and side-by-side transfer means a very small pressure differential - achievable by ullage venting - is sufficient for total propellant transfer from one tank to another. At the ends of a >200m tether, you now have a non-negligible pump head to deal with, along with an orientation change. There is also the propellant needed to set up for the transfer: for spin-settling, that's spinup and spindown thrust. For gravity-gradient, you still are effectively performing orbit raising/lowering burns to move the vehicles to their respective offset altitudes, though for most transfers you could probably omit return burns (though the next tanker would still be burning incrementally for initial insertion longer as the depot orbit raises).
I guess I wasn't very clear in my original description. What I'm discussing might better be called "tidal settling," since the idea is to use a gentle settling force--just like with an ullage burn--but without actually venting any propellant. There would never be a need to spin down, since it's not spinning fast enough to matter. And there's definitely no pumping prop along the tether! Where did you get that idea?

To simplify the idea as much as possible, imagine a large counterweight made of lead with a cable that connects it to the nose of a depot. As long as the counterweight is heavier than the depot, tidal forces will tend to keep it down (towards the Earth) and keep the depot up. Because the system orbits the Earth, it's technically spinning, but only once per orbit. This has none of the instability problems that make rotational settling impractical.

Tidal forces generate a very small amount of gravity in the depot--the same as an ullage burn, if you get the cable length right. A Starship would dock with the depot (not the counter weight) side-by-side and transfer propellant exactly the same way it would during an ullage burn. No extra plumbing.

Obviously there are details to work out. (E.g. unless the counterweight is really huge, there's going to be a significant change in the center of mass whenever a vehicle docks/departs, which will likely cause oscillations. It's unclear whether those can be tolerated or not.) The biggest drawback that I can see is that the cost of lifting the counterweight is probably prohibitive. Although replacing it with a space station might be fun to discuss--probably in a different thread.

Anyway, does that make the idea any clearer?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 08/15/2023 01:57 pm
It is only the intermediate axis [...] that is unstable, the other two axes are stable. Spin-stabilised satellite stacks (particularly long narrow ones with a heavy SRM at the bottom [...] work, after all.

Maybe this is only a semantic quibble, but:
the long axis is stable in a rigid body because it's the highest energy state (given fixed angular momentum). In a system with damping, once you have minor peturbations (or dynamic imbalance?) energy can be lost, allowing angular momentum to bleed into other axes. (though this takes a lot of time, compared to the intermediate axis case).

On the topic of damping, liquid is very damp  :-X
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 08/15/2023 02:08 pm
[tidally stabilised tether]
As long as the counterweight is heavier than the depot, tidal forces will tend to keep it down (towards the Earth) and keep the depot up.

I don't think that's necessary. It will work on any combination of masses, as long as it's vertical to Earth.

Having a heavier counterweight will increase the g-load felt by the depot, reducing the required tether length beyond the mid-point (which reduces the risk of MMOD breakage), but doesn't itself add (let alone cause) tidal stability.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 08/15/2023 06:16 pm
Note that only rotation about the axis of highest moment of inertia is stable
It is only the intermediate axis (if present, which is not always the case for some axisymmetric objects) that is unstable, the other two axes are stable. Spin-stabilised satellite stacks (particularly long narrow ones with a heavy SRM at the bottom, with the CoM moving forwards as the motor burns, and perturbations from the SRM attempting to rotate the stack being the reason for the spin in the first place) work, after all.
The first corollary to the Intermediate axis theorem is that the low moment axis is also unstable on a different time scale, because it is an energy maximum.  As Brigatine suggests this is particularly relevant to a floppy, fluid filled tanker.  Even the high moment axis is suspect if fluid is not rotating as a rigid body. 

IIRC you can also derive most of this from the topology of the phase space.  But the topology does not have a measure of distance or time, so cannot tell you how fast things happen.  Since you really care how long the states are stable, and/or how much active control is needed to stabilize them, you should do the CFD or run experiments rather than relying on abstract theorems that do not address these important questions.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 08/21/2023 02:41 am
I'm just trying to remember, where did this thread get to with the concept of using an accumulation tanker as a booster stage for TLI burns, in scenarios where HEO refilling is currently proposed? I vaguely recall something like this was discussed previously, but I can't find it.

The idea was the prime starship would have been refilled by a (first) accumulation tanker in LEO. Then there would be a second, modified, accumulation tanker, with a removable nose cone, so it can dock to the prime starship in the same position as the superheavy booster.  The TLI burn is two stage and the accumulation booster stage retains enough prop to burn back into LEO for reuse.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 08/21/2023 11:51 am
I'm just trying to remember, where did this thread get to with the concept of using an accumulation tanker as a booster stage for TLI burns, in scenarios where HEO refilling is currently proposed? I vaguely recall something like this was discussed previously, but I can't find it.

The idea was the prime starship would have been refilled by a (first) accumulation tanker in LEO. Then there would be a second, modified, accumulation tanker, with a removable nose cone, so it can dock to the prime starship in the same position as the superheavy booster.  The TLI burn is two stage and the accumulation booster stage retains enough prop to burn back into LEO for reuse.
Why would you do serial staging with a custom stage instead of parallel staging with a standard tanker?

To answer your question it was discussed on the first few pages of this thread, for example:

In my opinion the second stage of development of the concept should be a permanently orbiting 'tank' which can be repositioned in orbit as required.
The prime candidate for this would be a modified SH which would be comparatively easily developed given it's an iteration of an existing design.

I envisige you would replace the SH thrust structure with a SS thrust structure. The SH (tank) could even supply fuel to the SH (launch) to help with the increased takeoff weight. You would then be using the proven SS attachment system for In-Orbit fuel transfers.

This is similar to an idea that Twark_Main and I discussed (thread is here (https://forum.nasaspaceflight.com/index.php?topic=49134.msg2007579#msg2007579)).  I'd pretty much assumed that you just stretched the SS with a few extra tank rings and cannibalized some of the payload bay space, but you could probably start with a set of SH tanks as well.

The most demanding mission is getting a heavy cargo + crew to the lunar surface and back without lunar surface refueling.  You have to avoid refueling in HEEO to keep the crew out of the Van Allen Belts during refueling, and you have to carry all your prop with you for the return.  The "pusher/tanker" described in the link gets launched almost empty, uses regular tankers (whatever those are--likely just no-payload SSes) to acquire however much prop is needed, then docks nose-to-tail with the payload SS.  It transfers some of the prop to the payload SS, but then just acts as a first stage for most of the TLI delta-v.  Once in TLI, it either does a direct or free return to LEO, either aerobraking or propulsively inserting into whatever orbit the next mission needs.

The big deal here is risk reduction for crewed missions.  You only need one rendezvous/docking operation in the relatively low-radiation environment of LEO, and one jettison of the pusher/tanker.  Pusher/tanker isn't quite as prop-efficient as a series of refuelings in LEO and one refueling in HEEO, but prop efficiency probably isn't the most important criterion for a crewed mission.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/21/2023 06:35 pm
I'm just trying to remember, where did this thread get to with the concept of using an accumulation tanker as a booster stage for TLI burns, in scenarios where HEO refilling is currently proposed? I vaguely recall something like this was discussed previously, but I can't find it.
I think we abandoned terminology like "accumulation tanker" when SpaceX released a document to the effect that there would just be "tankers" and "depots" to refuel the HLS. I thought we dropped discussion of HEO refueling when someone pointed out that SpaceX could just make the depot a little bigger and thus avoid problems with the Van Allen Belts. But I might have missed something.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 08/22/2023 11:14 am
Why would you do serial staging with a custom stage instead of parallel staging with a standard tanker?

...because serial staging lets you perform two burns at the (final) perigee, maximizing Oberth in the escape / TMI burn.

With "parallel staging" you have to wait for a fuel transfer process between burns, meaning there's not enough time to perform both burns down low in Earth's gravity well (LEO / perigee).


I thought we dropped discussion of HEO refueling when someone pointed out that SpaceX could just make the depot a little bigger and thus avoid problems with the Van Allen Belts.

The Van Allen belts didn't really "kill" the idea of HEO refueling. It's just yet another trade-off.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 08/22/2023 11:37 am
With "parallel staging" you have to wait for a fuel transfer process between burns, meaning there's not enough time to perform both burns down low in Earth's gravity well (LEO / perigee).
If you wanted to, you could have an entire orbit worth of time for fuel transfer (unless you're already hyperbolic)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 08/22/2023 01:39 pm
The Van Allen belts didn't really "kill" the idea of HEO refueling. It's just yet another trade-off.
HEO refueling has other problems as well. It's a lot more complicated than simply refueling from a depot in a circular LEO. Not worth it if the only reason for it is to avoid making the depot and the HLS a little bit longer. Assuming that doesn't introduce intractable problems itself, of course.

Longer term, assuming there are ever going to be frequent flights to the moon, the right solution is probably a depot in NRHO or the moon itself.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: rsdavis9 on 08/22/2023 01:57 pm
I believe that if you are in a very high earth orbit like/or L1 that to do interplanetary injection it is worth it to retro fire your engines to go into a highly elliptical orbit so you can perigee burn for max oberth effect. Is this true?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 08/22/2023 10:42 pm
I believe that if you are in a very high earth orbit like/or L1 that to do interplanetary injection it is worth it to retro fire your engines to go into a highly elliptical orbit so you can perigee burn for max oberth effect. Is this true?

1) The NEA mission concept used it (see below)
2) Consider the extreme case, where your velocity at apogee tends to zero, so your ∆v to lower Perigee to the optimal height also tends to zero, and the benefit is non-zero

Quote
Earth Departure ∆v = 0.694 km/s
(0.242 km/s HEO Lowering and 0.452 km/s TNI)

(https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=59320.0;attach=2203291;image)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 09/01/2023 08:37 am
Why would you do serial staging with a custom stage instead of parallel staging with a standard tanker?

...because serial staging lets you perform two burns at the (final) perigee, maximizing Oberth in the escape / TMI burn.

Wasn't meaning it to be a one-of-a-kind, non-reusable variant though. An idea I've been noodling over is to try to get two birds with one stone:
1) Rather than doing a "chomper" variant of SS for large/heavy payloads, where issues with operating an 8-meter hatch could make it more trouble than it's worth, it would be simpler to have twin fairings jettisoned and recovered like F9. So the first use of this variant would be delivering high mass/ 8m diameter payloads to LEO. The problem with this concept has always been: how do you reuse it?
2) I was thinking this "booster" reuse mode is the answer to that. Say you have mechanisms on the payload mounting plate that are able to mate to the base of a regular starship, you load propellant to both, then deliver that extra delta-V during the Oberth maneuver. Booster never leaves cis-lunar space, so save enough prop to brake into next rendezvous orbit, rinse and repeat.

Getting a bit Kerbal I'll admit, but serial staging in LEO seems like it could be the missing element that could provide that extra margin for round trips to the moon, or reduced trip times to Mars or NEOs. Also it occurs to me that once you have a booster stage in LEO, you can start to stratify variants on engine number (say 9 in the booster/heavy cargo variant - leave the crew/small-cargo/HSS variants with 6 engines). Or stratify on header tank size, for more optimal cryo propellant storage.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/01/2023 11:55 am

2) I was thinking this "booster" reuse mode is the answer to that. Say you have mechanisms on the payload mounting plate that are able to mate to the base of a regular starship, you load propellant to both, then deliver that extra delta-V during the Oberth maneuver. Booster never leaves cis-lunar space, so save enough prop to brake into next rendezvous orbit, rinse and repeat.

Getting a bit Kerbal
If you want Kerbal

Let the booster aerobrake after separation.  In principle you could be above escape velocity at separation and still get your booster back.  And of course the higher the speed at final separation the higher the total performance.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 09/01/2023 11:03 pm
How feasible would it be to... basically do the same mission plan as above, but with the two ships docked side by side, and doing a high-flow propellant transfer under very high acceleration (full thrust for the heavier ship).

Potential issues:

1) How precisely could thrust in the lighter ship and gimbaling/balance be controlled to manage forces in the docking mechanism?
2) How much propellant could physically be transferred during the burn? For comparison, what's the combined cross-section of the propellant feeds to the 6 or 9 engines?

It probably wouldn't work, especially wrt (almost) fully depleting the tanker and having a full ship at separation - but for what reason and by what margin?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/02/2023 02:55 am
How feasible would it be to... basically do the same mission plan as above, but with the two ships docked side by side, and doing a high-flow propellant transfer under very high acceleration (full thrust for the heavier ship).
You may be overestimating how much higher thrust would help.  A few g for 9 minutes (a guess from Falcon 9) is not that far away from a pure impulse.  In a parabolic orbit the radius will vary by less than 10% over the burn.  The difference in Vinfinity compared to a perfect impulsive burn should be somewhere below 200m/s.  Not nothing to be sure but it does mean you can't add much mass in the search of perfection.

I think I was conservative in my estimates, but I'd be interested if somebody with a simulator could evaluate the integrals.  I'd be even more interested if there was closed form solution, but it seems unlikely.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 09/02/2023 06:30 am
To clarify: the main advantage I was thinking of would be you can have a standard EDL-capable tanker w/ nosecone, and still get 2 stages worth of delta-v in one perigee.

Trying to make a counterfactual against which to compare the flat-top serial-staging orbital booster concept.
What sort of prop transfer rates are realistic?
what sort of acceleration can side-by-side docking safely accommodate?


I accept Barley's assessment of a very marginal benefit from having more than 9 engines burning at once.
TBH, I generally like a 2-perigee 2-burn concept better anyway (intermediate apogee TBD), but for now working on the premise of an impatient crew so single maneuver.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/02/2023 01:58 pm
To clarify: the main advantage I was thinking of would be you can have a standard EDL-capable tanker w/ nosecone, and still get 2 stages worth of delta-v in one perigee.

... but for now working on the premise of an impatient crew so single maneuver.

As Twark_Main said the main advantage of a 2 stage burn at perigee is for the escape burn.  When you reach vescape there will not be another perigee.  In solar orbit the next possibility would be a flyby and Oberth maneuver, but you have to wait years rather than days.

Why would you do serial staging with a custom stage instead of parallel staging with a standard tanker?

...because serial staging lets you perform two burns at the (final) perigee, maximizing Oberth in the escape / TMI burn.

With "parallel staging" you have to wait for a fuel transfer process between burns, meaning there's not enough time to perform both burns down low in Earth's gravity well (LEO / perigee).


Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: mikelepage on 09/03/2023 03:50 am

2) I was thinking this "booster" reuse mode is the answer to that. Say you have mechanisms on the payload mounting plate that are able to mate to the base of a regular starship, you load propellant to both, then deliver that extra delta-V during the Oberth maneuver. Booster never leaves cis-lunar space, so save enough prop to brake into next rendezvous orbit, rinse and repeat.

Getting a bit Kerbal
If you want Kerbal

Let the booster aerobrake after separation.  In principle you could be above escape velocity at separation and still get your booster back.  And of course the higher the speed at final separation the higher the total performance.

Ha! (Although I'm curious now. How low can your perigee realistically go during TLI anyway?).

That said, by my BoE calcs I don't think the booster SS gets anywhere near escape velocity. (I know people have done similar calcs to this before - probably more accurately - but here's my work). A 1320 ton booster SS pushing a 1470 ton fully loaded SS (150 ton cargo, total mass 2790ton) gets us an extra ~2.1km/s delta-v at TLI.

So 8.4km/s delta-v to play with. After separation, SS has ~6.3km/s delta-v, of which 3.57km/s (680ton prop) is used to land on the moon. Once landed on the moon, we assume SS offloads most of its payload, with 20 tons payload returned back to Earth, so our liftoff delta-v remaining (140 ton starship, 520 tons prop) is 5.78 km/s. With 5.67km/s needed to return to Earth.

Even this seemed fairly tight to me, until I remembered someone upthread had worked out that as long as we had at least two HLS Starships tag-teaming the trip from LEO to NRHO and transferring the necessary prop to return to LEO (from NRHO) in NRHO, we could avoid lugging that extra prop down to the lunar surface and back (I think the difference is ~380 tons or so), and that gives us significant margin.

TLDR: Having a booster SS variant avoids having to send tanker starships anywhere but LEO, which seems much more efficient.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/03/2023 05:16 am

2) I was thinking this "booster" reuse mode is the answer to that. Say you have mechanisms on the payload mounting plate that are able to mate to the base of a regular starship, you load propellant to both, then deliver that extra delta-V during the Oberth maneuver. Booster never leaves cis-lunar space, so save enough prop to brake into next rendezvous orbit, rinse and repeat.

Getting a bit Kerbal
If you want Kerbal

Let the booster aerobrake after separation.  In principle you could be above escape velocity at separation and still get your booster back.  And of course the higher the speed at final separation the higher the total performance.

Ha! (Although I'm curious now. How low can your perigee realistically go during TLI anyway?).

That said, by my BoE calcs I don't think the booster SS gets anywhere near escape velocity. (I know people have done similar calcs to this before - probably more accurately - but here's my work). A 1320 ton booster SS pushing a 1470 ton fully loaded SS (150 ton cargo, total mass 2790ton) gets us an extra ~2.1km/s delta-v at TLI.

So 8.4km/s delta-v to play with. After separation, SS has ~6.3km/s delta-v, of which 3.57km/s (680ton prop) is used to land on the moon. Once landed on the moon, we assume SS offloads most of its payload, with 20 tons payload returned back to Earth, so our liftoff delta-v remaining (140 ton starship, 520 tons prop) is 5.78 km/s. With 5.67km/s needed to return to Earth.

Even this seemed fairly tight to me, until I remembered someone upthread had worked out that as long as we had at least two HLS Starships tag-teaming the trip from LEO to NRHO and transferring the necessary prop to return to LEO (from NRHO) in NRHO, we could avoid lugging that extra prop down to the lunar surface and back (I think the difference is ~380 tons or so), and that gives us significant margin.

TLDR: Having a booster SS variant avoids having to send tanker starships anywhere but LEO, which seems much more efficient.

We're probably thinking about two different things.  If you're staying in the Earth-Moon system, you don't want to exceed escape velocity.  If you exceed escape velocity, you will not remain in Earth-Moon system unless you slow down pretty soon.  (You would only do this if time was much more valuable than propellant.  You probably need a torch ship for this to make sense.)  If you are staying within the Moons orbit rendezvous and refueling is sufficient because all orbits repeat (more or less) in less than a month.  So whatever you need to do you can send out a fleet of ships, burn in unison at perigee (or elsewhere) then have time to refuel before the next burn. There might be operational reasons to prefer something different, but it's not strictly necessary.

The situation I was thinking of is something like Parker Solar probe.  You need to get beyond escape velocity by enough to get to Venus.  And if you can go faster you might avoid one or more of the gravity assists at Venus and get to the closest solar approach sooner.  If you can go fast enough (~18km/s!!) you could go directly to a close perihelion without a gravity assist.

For high energy escape like this you would start with the fueled booster and starship in say GEO transfer orbit.  That is only 773m/s from escape so the 2.1km/s burn of the booster puts both booster and SS on an escape trajectory.  Once they separate the SS burns to head fast for Venus, while the booster heads low enough into the Earth's atmosphere to scrub at least 1.4km/s to aerobraking so that it remains near Earth and can be recovered in days or months.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Eer on 09/03/2023 04:41 pm
This discussion of multiple Ships being lashed together to fire simultaneously triggered the thought that though Kerbalesque in the extreme, a Falcon Heavy-like monster assembled in GEO or geo-transfer orbit could impart a huge amount of dv to the center core. The two side ships would be tankers full of propellant with 6-vac+3-gimbaled engines able to push through trans-Martian-or-wherever injection, and then boost back.  Core ship saves propellant for breaking at destination.

Obviously not a first generation configuration, but leverages overall commonality of design to hopefully achieve higher dv transit.

No, I have no factual basis for this idea, and I’ll probably feel better after a nap.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/03/2023 05:56 pm
Why would you do serial staging with a custom stage instead of parallel staging with a standard tanker?

...because serial staging lets you perform two burns at the (final) perigee, maximizing Oberth in the escape / TMI burn.

Wasn't meaning it to be a one-of-a-kind, non-reusable variant though.

Me neither.  :)  IMO both reusable and non-reusable options are available.

You guys do seem to have figured out that StarKicker can do a flip-and-burn after staging so it remains within Earth's sphere-of-influence, then aerobrake back to LEO for reuse (https://forum.nasaspaceflight.com/index.php?topic=52033.msg2140950#msg2140950).

This basic idea was suggested by Robert Zubrin as one of several possible "improvements" upon the Starship architecture.

The Van Allen belts didn't really "kill" the idea of HEO refueling. It's just yet another trade-off.
HEO refueling has other problems as well. It's a lot more complicated than simply refueling from a depot in a circular LEO.

The incremental complexity isn't too great. There's no new incremental hardware R&D, for instance, which is a big win.

In the HEEO refilling plan you just have a standard tanker, which performs a rendezvous and fuel transfer in high orbit. SpaceX wil do lots of tanker missions, so we can expect they'll make that entire process extremely cheap, reliable, and routine. I expect nearly 100% automation here.

No need for hardware, just software.  :)

Not worth it if the only reason for it is to avoid making the depot and the HLS a little bit longer.

If you run the delta-v math on tandem HEEO refilling ladders, it offers a lot more than (the equivalent of) "a little bit longer" prop tanks.

The math (from way back on page 32!  :o ): https://forum.nasaspaceflight.com/index.php?topic=50157.msg2229093#msg2229093

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/03/2023 09:12 pm
Also, don't think of StarKicker as (necessarily) some dinky little thing. Perhaps it shouldn't be surprising, but the optimal size (https://www.projectrho.com/public_html/rocket/multistage.php#optimizedeltav) for StarKicker is about the same size as Super Heavy.  :o

The main difference is you don't need as many Raptors. This frees up area on the thrust puck, and presumably this "free" area would be used to swap in some RVac engines, yielding a nice Isp boost.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 09/03/2023 11:45 pm
Empty mass 200t, actually that can reach orbit... so why aren't all the depots SH-sized? 3,400t fuel storage
(or even stretch to 3,600+margin for a nice round number capacity of 3 missions serviced on the trot)

Possibly something about aerodynamic stability/structural integrity on ascent... that's all I can think of? But that would equally affect starkicker.

(Though how much extra empty mass to add aerobraking capability? Or is it fully propulsive return to LEO?)

This discussion of multiple Ships being lashed together to fire simultaneously triggered the thought that though Kerbalesque in the extreme, a Falcon Heavy-like monster assembled in GEO or geo-transfer orbit could impart a huge amount of dv to the center core. The two side ships would be tankers full of propellant with 6-vac+3-gimbaled engines able to push through trans-Martian-or-wherever injection, and then boost back.  Core ship saves propellant for breaking at destination.

Obviously not a first generation configuration, but leverages overall commonality of design to hopefully achieve higher dv transit.

No, I have no factual basis for this idea, and I’ll probably feel better after a nap.

Until Starship actually goes to space, a Kerbal fantasy is all it's good for. All I want to know is, where is this Raptor-Heavy powered mission going? You might be carrying the centre core with you and end up doing a Callisto-Orbit-Rendezvous with it or something weird. Does that even reach?.
(I'm starting to see why electric propulsion is so popular for asteroid return missions. Also, there's probably a more appropriate thread for mission plans and staging methods that don't use propellant transfer)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/04/2023 05:03 pm

If you run the delta-v math on tandem HEEO refilling ladders, it offers a lot more than (the equivalent of) "a little bit longer" prop tanks.

The math (from way back on page 32!  :o ): https://forum.nasaspaceflight.com/index.php?topic=50157.msg2229093#msg2229093

Not sure what I'm missing, but if I make a single change to your spreadsheet--raising mission Starship prop from 1200 to 1500 tons--then LEO direct works with 14 prop transfers, but tanker slam needs 16, and the rest effectively don't work at all (20 or more transfers).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/04/2023 09:43 pm

If you run the delta-v math on tandem HEEO refilling ladders, it offers a lot more than (the equivalent of) "a little bit longer" prop tanks.

The math (from way back on page 32!  :o ): https://forum.nasaspaceflight.com/index.php?topic=50157.msg2229093#msg2229093

Not sure what I'm missing, but if I make a single change to your spreadsheet--raising mission Starship prop from 1200 to 1500 tons--then LEO direct works with 14 prop transfers

I actually agree. If we're talking specifically about refueling for "standard" lunar missions (ie ~100 tonne payload), I expect that's what we'll see.

Honestly I forgot I had written the descriptions on that spreadsheet in terms of Moon missions. I really should make an updated version where the wording is more generic.

but tanker slam needs 16, and the rest effectively don't work at all (20 or more transfers).

I don't see why 20 tanker launches means "effectively doesn't work at all" in your book. Establishing a city on Mars is going to require a lot more than 20 tanker missions.  ???

You'll always need more launches when using tandem refilling, because lofting more tanker dry mass into higher orbits = more total impulse = more propellant. Can't cheat physics!  :D


To make the math very simple: tandem refilling allows any Starship-compatible vehicle configuration to gain (up to) 3.2 km/s of additional delta-v from LEO. Effectively, your vehicle begins its departure burn at (approaching) Earth escape velocity, instead of a ~circular orbit.


What's maybe getting lost is that this isn't necessarily an either/or choice. I think of it more like RTLS vs droneship landings — for missions that need additional delta-v (ie beyond whatever the "large" configuration offers), it becomes an optional value-add service SpaceX can upsell to the customer.   8)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/05/2023 12:06 am
but tanker slam needs 16, and the rest effectively don't work at all (20 or more transfers).
I don't see why 20 tanker launches means "effectively doesn't work at all" in your book. Establishing a city on Mars is going to require a lot more than 20 tanker missions.  ???
I just mean that, in comparison to a method that only needed 14 refuelings, something that needs 20+ effectively doesn't work. By that same logic, I suppose SLS effectively doesn't work either. :-)

You'll always need more launches when using tandem refilling, because lofting more tanker dry mass into higher orbits = more total impulse = more propellant. Can't cheat physics!  :D
Yeah, I get that part. But couldn't you also get that by just having a second depot in GEO? It seems that the motivation for HEEO refueling is just to satisfy NASA's requirement that a manned mission only refuel one time. Does it have any other advantage?

To be sure we're talking about the same thing, here's what I think we're discussing: The depot itself is in HEEO, with a period of a week or two. As a result, it likely only gets a refueling tanker to visit it once per orbit, so ten refuelings would take ten or twenty weeks. That could mean passive methods to reduce boiloff won't be adequate, so it'll need active cooling. The depot will pass through the Van Allen belts over and over and over, so it'll need to be radiation hardened to some extent. The tankers will be exposed to a good bit more radiation as well (two passes each refueling), so they might need to be tougher as well.

Then there's the question of launching a mission from HEEO. Most likely, it has to rendezvous with the depot at perigee, do a plane change at apogee (since the HEEO is unlikely to be aligned with the mission target), and then do a big Oberth burn at perigee. But is that going to work? Unlike LEO, HEEO precesses rather slowly, so it seems to me that you'd be unlikely to be in the right place with respect to the Earth and the mission target. (The argument of periapsis will be wrong--and expensive to change.)

Contrast this with proposals to use an LEO depot and kick into HEEO to do the plane change. For those missions, you get to pick the orbit so you can be sure the perigee occurs right where you want it. (Or so it seems to me, anyway.) Everything except the mission itself is in LEO, so there's no special radiation protection required, and the mission itself only has to do two passes through the belts and it only does that once, so you have a better shot at timing things so as to avoid the worst parts of the belts.

Maybe I'm just failing to visualize things properly, but these seem like pretty big problems to me. Not small things that one might easily trade off.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/05/2023 02:20 am

To be sure we're talking about the same thing, here's what I think we're discussing: The depot itself is in HEEO, with a period of a week or two. As a result, it likely only gets a refueling tanker to visit it once per orbit, so ten refuelings would take ten or twenty weeks.

Where you can send one ship you can send more than one.  In this case you could gather a fleet of ten tankers in LEO and have them burn in formation so they all rendezvous with the depot during the same orbit and transfer the fuel in less than a week.  You could fly the depot from LEO at the same time as the tankers.   You could also send five tankers at a time and take two orbits, etc.

There are tradeoffs between the number of ships used v. the length of time, but generally the fewer launch windows you have the more you want to send on each one of them.

Then there's the question of launching a mission from HEEO. Most likely, it has to rendezvous with the depot at perigee, do a plane change at apogee (since the HEEO is unlikely to be aligned with the mission target), and then do a big Oberth burn at perigee. But is that going to work? Unlike LEO, HEEO precesses rather slowly, so it seems to me that you'd be unlikely to be in the right place with respect to the Earth and the mission target. (The argument of periapsis will be wrong--and expensive to change.)
I suspect that for many years, and certainly for Artemis, the entire sequence of refueling will be preplanned for a particular mission.  The planning works backwards from the final goal, so everything is where you need it then you need it.  Depots will not usually be left in HEEO for the next mission, since they will be in the wrong orbit.  Depending on design and cost they will either be abandoned in orbit, disposed of by reentry or landed for reuse.

There might be contingencies to deal with occasional failures, such as scheduling 11 tankers when you need 10.  But if too many events miss their windows it's a mission fail, unless some clever clogs can figure an after the fact Hiten style rescue.


But couldn't you also get that by just having a second depot in GEO?
Did you mean GEO or GEO transfer orbit?  GEO itself is an awful place for a depot.  It's expensive to get to and the perigee is far too high for a good Oberth effect.  In the delta-V maps posted above you'll see GEO is a dead end.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/05/2023 02:35 am
I suspect that for many years, and certainly for Artemis, the entire sequence of refueling will be preplanned for a particular mission.  The planning works backwards from the final goal, so everything is where you need it then you need it.  Depots will not usually be left in HEEO for the next mission, since they will be in the wrong orbit.  Depending on design and cost they will either be abandoned in orbit, disposed of by reentry or landed for reuse.
Okay, this is the key point I missed; I assumed the depot was supposed to be reusable. If that's not the case, then I can see how HEEO could work.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 09/05/2023 02:59 am
Depots will not usually be left in HEEO for the next mission, since they will be in the wrong orbit.  Depending on design and cost they will either be abandoned in orbit, disposed of by reentry or landed for reuse.
Or perhaps (?) aerobraked into a vaguely ecliptic-aligned LEO, for re-use of the depot (though not of the high-energy orbit)

If a depot is able to EDL and re-use, what's the difference between that and a tanker?
(I guess it's still optimized to get to orbit with less fuel remaining and more empty tank, so still a valid variant)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/05/2023 10:47 am
Barley did a great job addressing most of the other points. Just a couple things....


To be sure we're talking about the same thing, here's what I think we're discussing: The depot itself is in HEEO, with a period of a week or two. As a result, it likely only gets a refueling tanker to visit it once per orbit, so ten refuelings would take ten or twenty weeks. That could mean passive methods to reduce boiloff won't be adequate, so it'll need active cooling.

Why such a long orbital period? As you yourself point out, making that choice is causing problems for yourself.

Do the math on the extra delta-v gained by going from a 48-72 hour orbital period vs 2 weeks. It's only 0.1-0.2 km/s. So by choosing a more pragmatic (shorter) orbital period, effectively the tradeoff is that you "only" get a boost of 3.0-3.1 km/s, vs 3.2 km/s.


The depot will pass through the Van Allen belts over and over and over, so it'll need to be radiation hardened to some extent. The tankers will be exposed to a good bit more radiation as well (two passes each refueling), so they might need to be tougher as well.

Yet another reason why SpaceX was smart to make Starship so big.  :)

The penalty for radiation hardening gets smaller as the vehicle gets larger.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/05/2023 11:01 am
you could gather a fleet of ten tankers in LEO and have them burn in formation so they all rendezvous with the depot during the same orbit and transfer the fuel in less than a week

Ten tankers burning from LEO to HEEO?? Why so many?

It's very beneficial to consolidate fuel into one tanker in LEO before delivering it to HEEO. It's quite inefficient to have each individual tanker launch deliver directly to a higher orbit.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/05/2023 11:14 am
If a depot is able to EDL and re-use, what's the difference between that and a tanker?

Hence tandem refilling. :)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/05/2023 01:51 pm
you could gather a fleet of ten tankers in LEO and have them burn in formation so they all rendezvous with the depot during the same orbit and transfer the fuel in less than a week

Ten tankers burning from LEO to HEEO?? Why so many?

It's very beneficial to consolidate fuel into one tanker in LEO before delivering it to HEEO. It's quite inefficient to have each individual tanker launch deliver directly to a higher orbit.
I'm just showing the scheduling.  He asked for ten tankers so that's what I gave him.  I agree with consolidating the fuel into a minimal number of tankers as low as practicable.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Corey Mandler on 09/05/2023 01:55 pm
raptor 3 for tankers makes more sense as higher twr means more efficiant
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/05/2023 02:07 pm
If a depot is able to EDL and re-use, what's the difference between that and a tanker?
The role assigned.

The difference between the tanker and depot could be subtle.  Such as having extra capacity for some non-obvious consumable to enable longer on-orbit loiter.

Although they might go through phases.  During prototyping all the ships will be different, so some are slightly better for particular roles.  Then all the ships are the same as mass production kicks in.  Then they start to specialize again.


Note that there may be at least three different roles:
1) Earth to LEO.
2) LEO to HEEO.
3) Accumulation in HEEO.
The second and third roles may be pretty similar, although the last needs longer loiter.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/05/2023 07:03 pm
Why such a long orbital period? As you yourself point out, making that choice is causing problems for yourself.

Do the math on the extra delta-v gained by going from a 48-72 hour orbital period vs 2 weeks. It's only 0.1-0.2 km/s. So by choosing a more pragmatic (shorter) orbital period, effectively the tradeoff is that you "only" get a boost of 3.0-3.1 km/s, vs 3.2 km/s.
Interesting. Yeah, I get the same results, assuming a 200 km perigee. A two-day period gets you just under 3 k/s ∆v and a three-day period gets you just over it, while a 14-day period only gets you 3.15. The shorter periods make the depot spend more time in the Van Allen belts, but it's not as big a difference as I thought it would be; 60% of the time for a two-day orbit vs. 34% for a two-week orbit. (That's assuming the belts are actually spheres.) The absolute time in the belts is actually a good bit less for the shorter periods, which makes it a lot easier on the tankers and the mission. So, I agree, there's no good argument for an HEO with a period over three days.

The only reason I assumed HEO with a one or two-week period was that's what was used in papers I'd read about minimizing the cost of plane changes. Those papers assumed a depot in a circular LEO and a mission that would first boost itself into a two-week HEO, do the plane change at apogee, then do the big Oberth burn at perigee. But plane changes aren't the issue here, and even if they were, a two-week orbit only saves about 500 m/s off a 2-day orbit (300 m/s off a 3-day orbit), and that's in the most extreme case (where you have to cancel both the orbital inclination and the Earth's tilt).

So, given that you're willing to expend the depot after a single mission, it does all seem to work. If you want to reuse the depot, though, I think it pretty much has to be in LEO. Do you agree?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/05/2023 07:40 pm
Huh, that's a very interesting result about the Van Allen belt. Thanks for running the numbers!

Sounds like those papers may have been talking about a Bi-elliptic type combination transfers (https://en.wikipedia.org/wiki/Bi-elliptic_transfer#Versatility_in_combination_maneuvers), so in that case higher is better. I suspect 2 weeks is roughly the highest orbit where you can still ignore lunar perturbation and use "simple" (vs low-energy) analysis and planning.


So, given that you're willing to expend the depot after a single mission, it does all seem to work. If you want to reuse the depot, though, I think it pretty much has to be in LEO. Do you agree?

   A.) It doesn't need to be a specialized depot in HEEO. All you need is a standard tanker to deliver the fuel to HEEO. Afterwards the tanker just deorbits, reenters and gets reused — like normal.

   B.) Even if you want to use a specialized depot, the depot doesn't have to be expended after a single mission. It can be reused by aerobraking it back to LEO, after (optional) apogee burns to change orbital parameters if needed.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Asteroza on 09/05/2023 11:58 pm
I'll just drop a link for Jon Goff's blog post on roving depots associated with HEEO for interplanetary departure, in the context of Jon Goff's 3 burn departure paper, as reference info...

https://selenianboondocks.com/2021/02/an-updated-propellant-depot-taxonomy-part-vi-roving-depots/ (https://selenianboondocks.com/2021/02/an-updated-propellant-depot-taxonomy-part-vi-roving-depots/)

It'd be nice if Jon Goff could weigh in on this discussion, with respect to the HEEO orbital period in light of the sub-2 week option.
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: Brigantine on 09/06/2023 01:42 am
I'll just drop a link for Jon Goff's blog post on roving depots associated with HEEO for interplanetary departure, in the context of Jon Goff's 3 burn departure paper, as reference info...

https://selenianboondocks.com/2021/02/an-updated-propellant-depot-taxonomy-part-vi-roving-depots/ (https://selenianboondocks.com/2021/02/an-updated-propellant-depot-taxonomy-part-vi-roving-depots/)
He talks a lot about "the next time its plane lines up with the low-orbit"... is this based on orbits that naturally precess by a large amount? Is this a significant issue for elliptical earth orbits in the ecliptic plane?
Title: Re: Starship In-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/06/2023 02:02 am
I'll just drop a link for Jon Goff's blog post on roving depots associated with HEEO for interplanetary departure, in the context of Jon Goff's 3 burn departure paper, as reference info...

https://selenianboondocks.com/2021/02/an-updated-propellant-depot-taxonomy-part-vi-roving-depots/ (https://selenianboondocks.com/2021/02/an-updated-propellant-depot-taxonomy-part-vi-roving-depots/)

It'd be nice if Jon Goff could weigh in on this discussion, with respect to the HEEO orbital period in light of the sub-2 week option.

2 weeks seems like a long time.  A while back I calculated that 20,000km apogee was good enough (~6 hour period), most of the benefit is gained for the additional deltaV and for Mr Oberth and it's a lot less time to line up orbits for destinations and for refueling.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2023 04:13 pm
   A.) It doesn't need to be a specialized depot in HEEO. All you need is a standard tanker to deliver the fuel to HEEO. Afterwards the tanker just deorbits, reenters and gets reused — like normal.

   B.) Even if you want to use a specialized depot, the depot doesn't have to be expended after a single mission. It can be reused by aerobraking it back to LEO, after (optional) apogee burns to change orbital parameters if needed.
The trouble, I think, is boiloff. To manage boiloff passively, you want to cover the depot with Solar White tiles. But if you do that, then I think you give up on recovering it; it'll burn up on reentry.

If you use a regular tanker, you can deorbit and reuse it, but those black reentry tiles seem guaranteed to warm up the interior fast.

Of course you could always use active cooling, but SpaceX has already said they plan to do it passively.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2023 05:08 pm
2 weeks seems like a long time.  A while back I calculated that 20,000km apogee was good enough (~6 hour period), most of the benefit is gained for the additional deltaV and for Mr Oberth and it's a lot less time to line up orbits for destinations and for refueling.
I also get 6 hrs for a 20,000 km altitude, but the extra ∆v is just 2.0 km/s vs. the 3.0 km/s you get for a 2-day period. One kps seems like a lot to me. It also spends 60% of its time in the Van Allen belts, almost all of it in the inner belt.

As far as lining up goes, I figure it precesses once every 1.28 years vs. 45 days for a circular orbit 200 km up. To me, that seems like enough precession to be a problem but not enough to be useful.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2023 05:15 pm
He talks a lot about "the next time its plane lines up with the low-orbit"... is this based on orbits that naturally precess by a large amount? Is this a significant issue for elliptical earth orbits in the ecliptic plane?
The trouble with precession is that it means that any Earth orbit "in the ecliptic plane" won't stay in the ecliptic plane. Over time, it'll move. How fast? Here's a table I worked out, assuming orbits are inclined 28.4 degrees to the Earth's equator (latitude of Kennedy Space Center).

Circular orbit (200 km up): 45.5 days to go around once. (Since 28.4 degrees is above the tropic of Cancer, these orbits are never exactly in the ecliptic plane, but they get very close to it once per 45.5-day cycle.)

Elliptic orbits with perigee 200 km up (by period):
6 hrs: 1.3 years
1 day: 2.8 years
2 days: 15 years
3 days: 23 years
1 week: 55 years
2 weeks: 111 years

Now if you had one vehicle in LEO and another in HEO, they should align with each other about once every 45 days--just because the LEO orbit is precessing so fast.

Caveat: someone should check my numbers.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/06/2023 07:33 pm
2 weeks seems like a long time.  A while back I calculated that 20,000km apogee was good enough (~6 hour period), most of the benefit is gained for the additional deltaV and for Mr Oberth and it's a lot less time to line up orbits for destinations and for refueling.
I also get 6 hrs for a 20,000 km altitude, but the extra ∆v is just 2.0 km/s vs. the 3.0 km/s you get for a 2-day period. One kps seems like a lot to me. It also spends 60% of its time in the Van Allen belts, almost all of it in the inner belt.

As far as lining up goes, I figure it precesses once every 1.28 years vs. 45 days for a circular orbit 200 km up. To me, that seems like enough precession to be a problem but not enough to be useful.

There's also the quantum of refueling to consider.  the 2.0km/sec gives you half full starships in the HEO so one full starship to do the boost at perigee.   [technically 2.2km/sec - 3650km/sec * ln(1320/720) ]

Anything else and the logistics get weird. 

3km/sec requires a mass ratio of  2.27 which uses 62% of the fuel.  You'd have to boost two refuelers to the HEO and that'd have to all be coordinated over 2 day orbits.  Yikes.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/06/2023 09:29 pm


3km/sec requires a mass ratio of  2.27 which uses 62% of the fuel.  You'd have to boost two refuelers to the HEO and that'd have to all be co is ordinated over 2 day orbits.  Yikes.

IMHO the yikes is completely inappropriate.  It's not like navigating a sailboat, where wind and tide are at least somewhat unpredictable.  If you do the orbital math right everything ends up exactly where you need it.  The math is absolutely trivial compared to every other part of a rocket program, such as finite element analysis, let alone computational fluid dynamics.  I am often amazed by how much money and effort engineers will expend to avoid simple calculations.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/06/2023 09:52 pm


3km/sec requires a mass ratio of  2.27 which uses 62% of the fuel.  You'd have to boost two refuelers to the HEO and that'd have to all be co is ordinated over 2 day orbits.  Yikes.

IMHO the yikes is completely inappropriate.  It's not like navigating a sailboat, where wind and tide are at least somewhat unpredictable.  If you do the orbital math right everything ends up exactly where you need it.  The math is absolutely trivial compared to every other part of a rocket program, such as finite element analysis, let alone computational fluid dynamics.  I am often amazed by how much money and effort engineers will expend to avoid simple calculations.

Show the plan including all the 200t launches to LEO to make this work.   then tell me it's inappropriate.

Not to mention an additional $30M in fuel.  (1200t * $25k/ton = $30M).  That's an optimistic estimate.  It takes half of that to make a Mars trip, so wherever it is you want to go, it better be worth the extra 1-2 km/sec (it likely isn't).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2023 10:47 pm
There's also the quantum of refueling to consider.  the 2.0km/sec gives you half full starships in the HEO so one full starship to do the boost at perigee.   [technically 2.2km/sec - 3650km/sec * ln(1320/720) ]

Anything else and the logistics get weird. 

3km/sec requires a mass ratio of  2.27 which uses 62% of the fuel.  You'd have to boost two refuelers to the HEO and that'd have to all be coordinated over 2 day orbits.  Yikes.
Okay, I get about the same numbers you do, assuming I'm understanding you correctly. Correct me if I'm wrong, but in your scenario a pair of depots (wet mass, 1200 mt; dry mass, 100 mt) are fueled in LEO and then boosted to HEO such that each is exactly half depleted (8 hour orbit, 2.3 kps ∆v, 27,600 km altitude of apogee). One tops up the other and then returns to Earth. The other waits to fully fuel a mission. (I'm using Raptor's target specific impulse of 3750 m/s.)

Your objection to using higher orbits is that it would take three depots, not just two. Is that right?



Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/06/2023 11:56 pm
Not to mention an additional $30M in fuel.  (1200t * $25k/ton = $30M).  That's an optimistic estimate.  It takes half of that to make a Mars trip, so wherever it is you want to go, it better be worth the extra 1-2 km/sec (it likely isn't).
Far be it from me to defend the HEO depot idea, but, unless I'm seriously confused (always a possibility), the total ∆v for the mission isn't changed. It just moves some of it from the mission vehicle to the depot. The rocket equation can't be fooled.

The original motivation for it, as I recall, was that it appears that an HLS Starship that refueled in LEO before heading to NRHO and then to the moon would find itself just a little short of enough fuel to get back to NRHO. Refueling at NRHO is off the table because NASA won't allow the vehicle to be refueled twice. Hence the idea of a depot in HEEO. In this case, the determining factor is what's the most elliptical orbit an empty HLS can get into on a single load of fuel. That's when you put the depot.

I have to believe, though, that there are better knobs to twist to make this happen just using a single depot in LEO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 09/07/2023 12:58 am
a pair of depots (wet mass, 1200 mt; dry mass, 100 mt)
I had thought the depots would have considerably more than 1200t of fuel capacity. Doesn't a standard ship with a payload section already have 1200t of fuel? And the depots are supposed to be stretched?


Also, re precession, the Near-Earth-Asteroid proposal uses a 7,800 x 113,300 km re-fueling orbit (and subsequently uses 242 m/s to lower perigee before the departure burn). Does that reduce the rate of precession? Either by design or incidentally.
[and more generally, what resources do you use to calculate precession?]
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/07/2023 01:29 am
There's also the quantum of refueling to consider.  the 2.0km/sec gives you half full starships in the HEO so one full starship to do the boost at perigee.   [technically 2.2km/sec - 3650km/sec * ln(1320/720) ]

Anything else and the logistics get weird. 

3km/sec requires a mass ratio of  2.27 which uses 62% of the fuel.  You'd have to boost two refuelers to the HEO and that'd have to all be coordinated over 2 day orbits.  Yikes.
Okay, I get about the same numbers you do, assuming I'm understanding you correctly. Correct me if I'm wrong, but in your scenario a pair of depots (wet mass, 1200 mt; dry mass, 100 mt) are fueled in LEO and then boosted to HEO such that each is exactly half depleted (8 hour orbit, 2.3 kps ∆v, 27,600 km altitude of apogee). One tops up the other and then returns to Earth. The other waits to fully fuel a mission. (I'm using Raptor's target specific impulse of 3750 m/s.)

Your objection to using higher orbits is that it would take three depots, not just two. Is that right?

The second ship *is* the mission.

So only two ships have to boost up to HEO.

Try counting the costs at $25 (or $50) per kilo for each of the scenarios you propose
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/07/2023 02:32 am
Also, re precession, the Near-Earth-Asteroid proposal uses a 7,800 x 113,300 km re-fueling orbit (and subsequently uses 242 m/s to lower perigee before the departure burn). Does that reduce the rate of precession? Either by design or incidentally.
[and more generally, what resources do you use to calculate precession?]
I just used the formulas from the Wikipedia article on Nodal Precession (https://en.wikipedia.org/wiki/Nodal_precession) and made an Excel spreadsheet.

For your NEA proposal, I figure an orbital period of 1.7 days and a precession time of 14.5 years. And are you sure you meant 242 m/s not 24.2 ms/s? Otherwise, I think you'll impact the planet.

You should double-check my math, of course. :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 09/07/2023 03:29 am
Noted. Thanks.

are you sure you meant 242 m/s not 24.2 ms/s? Otherwise, I think you'll impact the planet.
It says 0.242 km/s HEO lowering, I haven't done the maths... I wouldn't be too surprised if there is some degree of plane change included in that figure

(https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=59320.0;attach=2203291;image)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/07/2023 03:54 am
Noted. Thanks.

are you sure you meant 242 m/s not 24.2 ms/s? Otherwise, I think you'll impact the planet.
It says 0.242 km/s HEO lowering, I haven't done the maths... I wouldn't be too surprised if there is some degree of plane change included in that figure

(https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=59320.0;attach=2203291;image)

Why is perigee at 7,800 km?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 09/07/2023 08:08 am
Why is perigee at 7,800 km?
That's a great question, and the best hypothesis I could come up with is that it's to mitigate collision avoidance complications, including in the event of an "anomaly" during refueling. I imagine debris in a e.g. 300 x 113,000 km orbit has the potential to cause inconvenience for a considerable while, unlike LEO. Essentially Pe = 7,800 is an acceptable enough accidental graveyard orbit

If so, perhaps a suborbital Pe would be an acceptable alternative at lower cost... depending on timing and geography

[Further NEA - specific discussion: thread (https://forum.nasaspaceflight.com/index.php?topic=59320.msg2521692#msg2521692) exists]
[Any comments on broader implications for HEO re-fueling in general: still on topic]
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tbellman on 09/07/2023 12:32 pm
The original motivation for it, as I recall, was that it appears that an HLS Starship that refueled in LEO before heading to NRHO and then to the moon would find itself just a little short of enough fuel to get back to NRHO. Refueling at NRHO is off the table because NASA won't allow the vehicle to be refueled twice. Hence the idea of a depot in HEEO. In this case, the determining factor is what's the most elliptical orbit an empty HLS can get into on a single load of fuel. That's when you put the depot.

I have to believe, though, that there are better knobs to twist to make this happen just using a single depot in LEO.

(My bolding.)
I have not heard any such requirement.  I suspect that you are misremembering and mixing a couple of things up:

• First, in the source selection statement for HLS Option A, Kathy Lueders wrote about the Starship propellant refilling that the high number of launches needed (up to 14 tankers, one depot, and one lander, all in just six months) meant a high risk of delays, but that it was tempered by doing them in Earth orbit (presumably LEO) rather than in lunar orbit "where an unexpected event would create a much higher risk to loss of mission".

But this is not an outright ban, it is just an additional risk.  And it is mostly a schedule risk: if refilling in NRHO fails, then it will take longer to send up a new tanker/depot there.  That could e.g. be long enough that all propellant in the lander ship has time to boil off, and the ship then becomes derelict.

This was also written in April 2021.  In the preceeding six months, SpaceX had launched 19 times.  Doing 16 launches in a similar length of time, and doing it in 2024, just to support a lunar landing, was not something NASA could consider a trivial task when they evaluated the HLS bids.

• Second, doing a second refilling in a high elliptic orbit, would entail passing through the Van Allen belts thrice instead of once (on the outgoing leg of the journey).  You probably want to avoid that if the ship is carrying living humans during those passages due to the radiation.  But:
A) A second refilling doesn't have to happen in such an orbit; e.g. at the Lunar Gateway in NRHO, during the trans-lunar coasting, or in low lunar orbit after taking off from the lunar surface, depending on the mission.
B) The HLS lander will not be carrying crew until after rendez-vous with Orion in NRHO, so unless the ship itself becomes significantly radioactive due to passing through the Van Allen belts multiple times, this doesn't matter.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/07/2023 03:02 pm
The original motivation for it, as I recall, was that it appears that an HLS Starship that refueled in LEO before heading to NRHO and then to the moon would find itself just a little short of enough fuel to get back to NRHO. Refueling at NRHO is off the table because NASA won't allow the vehicle to be refueled twice. Hence the idea of a depot in HEEO. In this case, the determining factor is what's the most elliptical orbit an empty HLS can get into on a single load of fuel. That's when you put the depot.
You also have to look at the least elliptical orbit where a fully fuel HLS has to be to complete the mission.  The final refueling can be anywhere between that and the most elliptical orbit you can reach.  If there is no overlap, you cannot complete the mission subject to the single refueling restraint.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/07/2023 03:37 pm


3km/sec requires a mass ratio of  2.27 which uses 62% of the fuel.  You'd have to boost two refuelers to the HEO and that'd have to all be co is ordinated over 2 day orbits.  Yikes.

IMHO the yikes is completely inappropriate.  It's not like navigating a sailboat, where wind and tide are at least somewhat unpredictable.  If you do the orbital math right everything ends up exactly where you need it.  The math is absolutely trivial compared to every other part of a rocket program, such as finite element analysis, let alone computational fluid dynamics.  I am often amazed by how much money and effort engineers will expend to avoid simple calculations.

Show the plan including all the 200t launches to LEO to make this work.   then tell me it's inappropriate.

Not to mention an additional $30M in fuel.  (1200t * $25k/ton = $30M).  That's an optimistic estimate.  It takes half of that to make a Mars trip, so wherever it is you want to go, it better be worth the extra 1-2 km/sec (it likely isn't).
To solve the problem you need to work backwards from the goal.  If you need the extra 1km/s it's priceless. "Yikes  lets not return them safely to Earth" is not going to work.

We don't know the mass of starship.  We don't know the cost to launch.  We don't know how hard or trivial refueling is.  We don't know about boiloff.  Far too early to take parts of the trade space off the table.  Easy refueling with low performance SS leads to a different plan than hard refueling with outstanding performance.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/08/2023 02:54 pm
You'd have to boost two refuelers to the HEO and that'd have to all be coordinated over 2 day orbits.  Yikes.

If you have three full vehicles (two refuelers + mission ship) in LEO, it's wasteful to take all three to your highest orbit.

Instead, you want a lower orbit (requiring less delta-v), such that they arrive two-thirds full. Then you dump one tanker into the other two vehicles, and the nearly empty tanker returns to Earth. Then your two (now full) vehicles repeat as before.


Don't worry about the measly two extra refueling events. If you had three full tankers in LEO, you already did 18 refuelings just for that. So in reality, all this does is is go from 19 refueling events to 21 refueling events (~10% more).


Yes, it's a lot of math. You have to reverse-calculate the height of each elliptical orbit, based on the tanker parameters, the mission vehicle parameters, AND the individual payload mass. That's why I made my big tanker ladder spreadsheet (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2229093#msg2229093): it does all the math for you. :D


P.S.:  Let's not go from arbitrarily assuming a 2 week period HEEO to now arbitrarily assuming a 48 hour period HEEO.  :-\  That would be missing the forest for the trees in my post here (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2521037#msg2521037).

My point is that we should choose the elliptical orbit based on the mission constraints themselves. What we should not do is choose an elliptical orbit by throwing darts at a dartboard ("2 weeks", "48 hours", etc), then only afterwards calculate the relevant mission constraints on that orbit, and finally throw up our hands in defeat because we found exactly one orbit that doesn't work.  ::)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/08/2023 05:54 pm
You'd have to boost two refuelers to the HEO and that'd have to all be coordinated over 2 day orbits.  Yikes.

If you have three full vehicles (two refuelers + mission ship) in LEO, it's wasteful to take all three to your highest orbit.

Instead, you want a lower orbit (requiring less delta-v), such that they arrive two-thirds full. Then you dump one tanker into the other two vehicles, and the nearly empty tanker returns to Earth. Then your two (now full) vehicles repeat as before.

...


There's an interesting middle ground here as well. You can have your tankers do a single tandem refilling, then have the remaining full tanker rendezvous with the mission vehicle in the highest possible orbit.

The advantage is that your mission vehicle only has one refilling event in HEEO, and it limits the number of passes through the Van Allen belt.

This exact refilling ladder can't be calculated using my refilling spreadsheet, but the math is pretty easy. The trick is to remember that for optimal rocket staging each stage provides half the total delta-v. Since tanker ladders are essentially identical (mathematically) to parallel staging, this means that the delta-v from LEO to the intermediate HEEO is exactly half that of the delta-v to the final HEEO.

This allows us to construct an equation:

v2 = 2 * v1

Where v1 is the excess velocity (from LEO) of the intermediate HEEO, and v2 is the excess velocity of the second HEEO.

Solve for v1.  :)

We know we need the tanks to be full at the end, which allows us to construct more equations. Let's call the mass of our final propellant transfer X, and use the rocket equation.

Mission vehicle burn directly to HEEO2:
v2 = Isp * ln(m1/m0) = Isp * ln[ (mission vehicle wet mass ) / (mission vehicle wet mass - X) ]

Tanker burn from LEO --> HEEO1 and HEEO1 --> HEEO2:
v1 = Isp * ln(m1/m0) = Isp * ln[ (tanker wet mass) / (tanker wet mass - (mission vehicle prop capacity - X)) ]


Substitute these into the first equation and simplify:

ln[ (mission vehicle wet mass) / (mission vehicle wet mass - X) ] = 2 * ln[ (tanker wet mass) / (tanker wet mass - (mission vehicle prop capacity - X)) ]


Insert numerical values:

ln[(120 + 150 + 1200) / (120 + 150 + 1200 - X) ] = 2 ln[ (100 + 1200) / (100 + 1200 - (1200 - X)) ]


Get lazy and just use Wolfram|Alpha (https://www.wolframalpha.com/input?i=solve%3A+ln%5B%28120+%2B+150+%2B+1200%29+%2F+%28120+%2B+150+%2B+1200+-+x%29%5D+%3D+2+ln%5B%28100+%2B+1200%29+%2F+%28100+%2B+1200+-+%281200+-+x%29%29%5D)  ;D :

x = 786.47 tonnes


Substitute back into the original equation to find v2 and v1:

v2 = Isp * ln[ (mission vehicle wet mass) / (mission vehicle wet mass - X) ]
v2 = 375 s * 9.80665 m/s^2 * ln[ (120 + 150 + 1200) / (120 + 150 + 1200 - 786.47) ]
v2 = 3677.49375 m/s * ln[ 1470 / 683.53 ]

v2 = 2816 m/s
v1 = 1408 m/s


Sorry for the long post, but hopefully this "worked example" is helpful.




Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/08/2023 07:34 pm

There's an interesting middle ground here as well. You can have your tankers do a single tandem refilling, then have the remaining full tanker rendezvous with the mission vehicle in the highest possible orbit.

The advantage is that your mission vehicle only has one refilling event in HEEO, and it limits the number of passes through the Van Allen belt.

This exact refilling ladder can't be calculated using my refilling spreadsheet, but the math is pretty easy. The trick is to remember that for optimal rocket staging each stage provides half the total delta-v. Since tanker ladders are essentially identical (mathematically) to parallel staging, this means that the delta-v from LEO to the intermediate HEEO is exactly half that of the delta-v to the final HEEO.


Optimal is a widely misused term.  You should always ask optimal with respect to what?
In this case optimal means sizing the two stages.  Since we have existing Star Ships and tankers we are solving a different problem.  If we want to force the use of that "optimal" to size the stages were are constrained by the availability of small integers, and fractions composed of small integers.

Lets start with three ships, two tankers with dry mass 100 tonne, a mission vehicle with dry mass 120+150 tonne, all vehicles hold 1200 tonne of fuel, Isp is 375s so exaust velocity is 3677 m/s

Burn two tankers in parallel using half the fuel.
m0 = 100 + 1200 + 100 + 1200 = 2600
m1 = 100 + 100 + 1200 = 1400

m0/m1 = 1.857
Δv1 = 3677 * ln(1.857 ) = 2276 m/s

Meanwhile we burn the mission ship to the same delta_v  Since the exhaust velocity is the same the mass ratio will be the same so
m1mission = (1200 + 120 + 150)/1.857 = 792

As a sanity check this is between the fully loaded and fully empty masses of the mission ship.

Now we "stage" by moving the remaining 600 tonne of fuel from one tanker into the other and "discarding" the dry weight of the empty tanker.  We will burn the remaining tanker and mission ship in parallel until 1200 tonne of fuel remain to fill the mission ship.

m'0 = 1200+100+792 =2092
m'1 = 1200+100+120+150 = 1570
Δv2 = 3677 * ln( 2092/1570) = 1055

So the intermediate HEEO is at LEO + 2276
And the final HEEO is at LEO + 3331

I have not shown these are optimal (although within the constraints they are) but they are clearly possible, better than your "optimal" and don't need Wolfram alpha, just a log table.


Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Keldor on 09/08/2023 08:07 pm

There's an interesting middle ground here as well. You can have your tankers do a single tandem refilling, then have the remaining full tanker rendezvous with the mission vehicle in the highest possible orbit.

The advantage is that your mission vehicle only has one refilling event in HEEO, and it limits the number of passes through the Van Allen belt.

This exact refilling ladder can't be calculated using my refilling spreadsheet, but the math is pretty easy. The trick is to remember that for optimal rocket staging each stage provides half the total delta-v. Since tanker ladders are essentially identical (mathematically) to parallel staging, this means that the delta-v from LEO to the intermediate HEEO is exactly half that of the delta-v to the final HEEO.


Optimal is a widely misused term.  You should always ask optimal with respect to what?
In this case optimal means sizing the two stages.  Since we have existing Star Ships and tankers we are solving a different problem.  If we want to force the use of that "optimal" to size the stages were are constrained by the availability of small integers, and fractions composed of small integers.

Lets start with three ships, two tankers with dry mass 100 tonne, a mission vehicle with dry mass 120+150 tonne, all vehicles hold 1200 tonne of fuel, Isp is 375s so exaust velocity is 3677 m/s

Burn two tankers in parallel using half the fuel.
m0 = 100 + 1200 + 100 + 1200 = 2600
m1 = 100 + 100 + 1200 = 1400

m0/m1 = 1.857
Δv1 = 3677 * ln(1.857 ) = 2276 m/s

Meanwhile we burn the mission ship to the same delta_v  Since the exhaust velocity is the same the mass ratio will be the same so
m1mission = (1200 + 120 + 150)/1.857 = 792

As a sanity check this is between the fully loaded and fully empty masses of the mission ship.

Now we "stage" by moving the remaining 600 tonne of fuel from one tanker into the other and "discarding" the dry weight of the empty tanker.  We will burn the remaining tanker and mission ship in parallel until 1200 tonne of fuel remain to fill the mission ship.

m'0 = 1200+100+792 =2092
m'1 = 1200+100+120+150 = 1570
Δv2 = 3677 * ln( 2092/1570) = 1055

So the intermediate HEEO is at LEO + 2276
And the final HEEO is at LEO + 3331

I have not shown these are optimal (although within the constraints they are) but they are clearly possible, better than your "optimal" and don't need Wolfram alpha, just a log table.

The optimal refueling altitude is always going to be as low as possible.  This is because otherwise we're just moving extra tank mass around when we don't have to.

For a high energy mission, we'd want to start by topping off the payload vehicle as well as another tanker in low orbit, then both will burn into a high elliptical orbit, where the second tanker will top off the payload vehicle one final time before it leaves Earth orbit for good.  What we don't want to do is try to do all our refueling in high orbit, where we pay to drag dozens of nearly empty tankers all the way up.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/08/2023 09:45 pm
Yes, it's a lot of math. You have to reverse-calculate the height of each elliptical orbit, based on the tanker parameters, the mission vehicle parameters, AND the individual payload mass. That's why I made my big tanker ladder spreadsheet (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2229093#msg2229093): it does all the math for you. :D
The way I like to think of the basic problem is that you start with one fully fueled Starship and one depot in circular LEO with at least some fuel in it, and you want to end up with a fully fueled Starship in HEEO and an empty depot. This means you'll want to accelerate them in tandem until the point where, between the two of them, they have exactly enough fuel for a single Starship. Then you transfer all the remaining fuel from the depot to the Starship. Since the velocities have to match, so do the mass ratios, and that lets you solve for everything.

This ends up being a function of the wet mass of the Starship (S), the wet mass of the Depot (D), the amount of fuel in the Depot (F), and the specific impulse of the Starship (k). Then

∆v = k ln((S+D)/(S+D-F))

Which is quite elegant; the argument of the logarithm is just the mass ratio of the two ships taken as a unit. (Imagine that they're bolted together.)

It doesn't matter how big the payload is (assuming we're counting that as part of the wet mass) nor how much fuel the Starship actually holds. However, if F is too big (about 3000 tons of fuel in my examples), then your fully fueled Starship ends up on a hyperbolic orbit, not a high Earth orbit, and that spoils the plan, which called for doing a big Oberth burn at the next perigee.

For the examples I played with, this required a 2.5 kps burn, putting the vehicles into a 13.6-hour orbit with an apogee of 51,000 km (from Earth's center). For the payloads I was playing with, this amounted to 3.5 kps extra hyperbolic excess velocity (on top of 10.5 kps I got without this trick), which is not too shabby.

By the way, I might have found an error in your spreadsheet. We get the same answers for delta-v but different answers for apogee. They're only different by about 6.5%, but that's a lot since the other numbers are within 0.1%. You're using a very complicated formula for apogee, and that may be the problem. I'm using r_a = (r_p*v_p)^2/(2*mu-r_p*v_p^2) where r_a is the apogee (measured from the center of the Earth), r_p is the perigee, v_p is the velocity at perigee, and mu is the standard gravitational parameter for the Earth.

I did not explore your idea of using two or three extra depots.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/08/2023 10:02 pm
And the final HEEO is at LEO + 3331
Hmm. I compute escape velocity from a circular orbit with an altitude of 200 km to be 3226 m/s, so that's likely to be a problem. :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/08/2023 10:36 pm
Okay, thinking about the multiple-depot case, I think I see the optimal (i.e. max ∆v) way to use a "wagon train" comprising a fully fueled Starship and n fully fueled depots. Let S be the mass of a fully fueled Starship (including payload). Assume all depos have the same wet mass (D) and the same mass of fuel (F). Assume that the specific impulse is the same for all vehicles.

In the case with a Starship and a single depo, we wanted to boost until F fuel was consumed between the two vehicles. At that point, fully fueling the Starship fully drains the depot. That this is optimal probably doesn't need elaborating. The boost that accomplishes this is ∆v = k ln((S+D)/(S+D-F)). (The trick to finding this is to see that the mass ratios for the two vehicles must be equal because both have the same ∆v.)

When you add another depo, the main thing you want to do is to discard excess mass at the first opportunity. Otherwise, you're wasting fuel. Therefore, just as in the single-depot case, you want to do a single burn that consumes F fuel. After that, you can redistribute what's left, and you will discard exactly one depot. Then repeat.

It shouldn't take much thought to see that ∆v_n = k ln((S+n*D)/(S+n*D-F)). Then sum these up to get the total ∆v--note that this does not include the ∆v from the Starship itself.

However, using what I think are reasonable numbers, if you start from a 200-km-altitude orbit, you can't use even just two depots; between the two of them, you'll go 800 m/s over escape velocity. So this trick is likely to be of interest only for a single depot.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/09/2023 12:04 am
The optimal refueling altitude is always going to be as low as possible.  This is because otherwise we're just moving extra tank mass around when we don't have to.

For a high energy mission, we'd want to start by topping off the payload vehicle as well as another tanker in low orbit, then both will burn into a high elliptical orbit, where the second tanker will top off the payload vehicle one final time before it leaves Earth orbit for good.  What we don't want to do is try to do all our refueling in high orbit, where we pay to drag dozens of nearly empty tankers all the way up.
I think all anyone is talking about is putting fuel depos in LEO--as low as practical--and then filling them up with a series of "tanker" flights from the ground. The depots are likely to be "stretched" Starships, holding perhaps 1500 tons of fuel each.

The simplest use case is for a Starship launched from the ground to rendezvous with a single depo (in LEO), refuel itself, and then do a big burn to higher orbit, to the moon, or elsewhere. If that works, that's what you ought to do--no question. (I don't think anyone is debating this point.)

But this doesn't seem to have enough ∆v for some missions.

So assume you've got a fully fueled Starship in LEO and a second, fully fueled depo in LEO. How do you leverage that second depo to add some ∆v to your Starship? This is the problem everyone is trying to solve.

Boosting them both into the same HEO and then transferring the remaining fuel from the depo to the Starship seems like the best way to do that. When the fully fueled Starship comes back to perigee, it does the burn for its final destination. This can buy you as much as 2.5 kps at perigee--3.5 kps of additional hyperbolic excess in the examples I played with.

Potentially, the (almost) empty depo does aerobraking to eventually get back to LEO.

Even then, though, as you say, you want to make the fuel transfer in the lowest possible orbit. Too low and the Starship won't have room for the fuel. Too high, and the depo won't have enough fuel left to fully fuel the Starship.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Bob Shaw on 09/09/2023 12:17 am
If your launch costs are unreasonably cheap then all these edge of case studies don't matter. You accept the losses, fuel up and fire up. If Starship works we're no longer in marginal economics mode - the old paradigms simply don't apply.

We must wait and see.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/09/2023 01:47 am
If your launch costs are unreasonably cheap then all these edge of case studies don't matter. You accept the losses, fuel up and fire up. If Starship works we're no longer in marginal economics mode - the old paradigms simply don't apply.

We must wait and see.
If you run out of fuel before you reach your destination, then it really doesn't matter how cheap launch costs are. Nor can you launch a mission and "wait and see" whether it gets where it's supposed to go.

Cheap launch costs make a lot of things possible, but they do not eliminate the need to plan, to design, and to think ahead.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/09/2023 02:20 am
The depots are likely to be "stretched" Starships, holding perhaps 1500 tons of fuel each.
There doesn't seem to be any need for "stretched" Starships depot or tanker.  A "standard" 1200 tons works just fine at and above LEO.  The hardest and most frequent tanker use is from Earth to LEO.  That's what you need to optimize for, everything else is rounding.  And if all tankers are capable of reentry you don't need to worry about changing planes and such.  Once the fuel is transferred back to Earth goes the empty tanker.

The only thing that might need to be different for a depot or high orbit tanker is the ability to loiter by reducing boil off.

That's not to say StarShips won't get longer, but it will be like Falcon 9.  The standard version gets longer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/09/2023 03:37 am
The depots are likely to be "stretched" Starships, holding perhaps 1500 tons of fuel each.
There doesn't seem to be any need for "stretched" Starships depot or tanker.  A "standard" 1200 tons works just fine at and above LEO.  The hardest and most frequent tanker use is from Earth to LEO.  That's what you need to optimize for, everything else is rounding.  And if all tankers are capable of reentry you don't need to worry about changing planes and such.  Once the fuel is transferred back to Earth goes the empty tanker.

The only thing that might need to be different for a depot or high orbit tanker is the ability to loiter by reducing boil off.

That's not to say StarShips won't get longer, but it will be like Falcon 9.  The standard version gets longer.

If you need a non-integer number of Starships to fuel the final Starship, then a stretched version makes more economic and logistical sense than boosting an extra Starship with less than a full load of fuel.

Two starships half full is an elliptical orbit w/ 2.2km/sec of additional velocity at perigee (and one full Starship after refueling at same orbit).  if you want 3.6km/sec of additional velocity at perigee you need 885t of propellant boosted into that orbit to fill up the starship you are trying to boost.  That just so happens to be exactly a 1500t fuel ship that masses 120t dry. (3.65km/sec exhaust velocity).

If you try that with a 1,200t fuel Starship you'll have to send up two fuelers into that orbit and they won't be full (about 75%), and you'll have boosted up 6 extra engines and tank which is a waste, plus the logistics of yet another refueling in an orbit that crosses Van Allen belts.

I'm pretty sure they are going to build a stretched version when they change to 9 engines because 1500t makes a lot of sense for a lot of missions w/ 9 engines.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 09/09/2023 09:47 am
If you need a non-integer number of Starships to fuel the final Starship, then a stretched version makes more economic and logistical sense than boosting an extra Starship with less than a full load of fuel.
Only a a non-mass-produced vehicle. If you are shipping goods and you just can't quite fit things into one container, do you construct a custom bigger container or do you ship two containers? "Sizing the vehicle to the specific mission" is the sort of thinking that results in SLS and the Atlas & Vulcan 'dial-a-rocket' concepts.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/09/2023 01:19 pm
The depots are likely to be "stretched" Starships, holding perhaps 1500 tons of fuel each.
There doesn't seem to be any need for "stretched" Starships depot or tanker.  A "standard" 1200 tons works just fine at and above LEO.  The hardest and most frequent tanker use is from Earth to LEO.  That's what you need to optimize for, everything else is rounding.  And if all tankers are capable of reentry you don't need to worry about changing planes and such.  Once the fuel is transferred back to Earth goes the empty tanker.

The only thing that might need to be different for a depot or high orbit tanker is the ability to loiter by reducing boil off.

That's not to say StarShips won't get longer, but it will be like Falcon 9.  The standard version gets longer.

If you need a non-integer number of Starships to fuel the final Starship, then a stretched version makes more economic and logistical sense than boosting an extra Starship with less than a full load of fuel.

Two starships half full is an elliptical orbit w/ 2.2km/sec of additional velocity at perigee (and one full Starship after refueling at same orbit).  if you want 3.6km/sec of additional velocity at perigee you need 885t of propellant boosted into that orbit to fill up the starship you are trying to boost.  That just so happens to be exactly a 1500t fuel ship that masses 120t dry. (3.65km/sec exhaust velocity).

If you try that with a 1,200t fuel Starship you'll have to send up two fuelers into that orbit and they won't be full (about 75%), and you'll have boosted up 6 extra engines and tank which is a waste, plus the logistics of yet another refueling in an orbit that crosses Van Allen belts.


That's not how you do it.  Starting from low Earth orbit the final tanker burns 300 tonnes of fuel to an intermediate elliptical orbit where it rendezvouses with the second tanker to fill up.

The intermediate orbit is about 400x5000 km which may avoid the worst of the radiation.  Depending on sun spots.

The fuel needed to raise the empty mass of the second tanker is only 30 tonnes.  Depending on exactly how much a tanker can deliver to LEO from earth this may or may not require one extra flight and refueling.  There is better than an 80% chance it does not.

And this assumes that the 1200 and 1500 capacity tankers have equal dry mass and performance.  Which is not going to be the case if one is production and the other is custom.  The production ship will have better performance for the cost.

There are three things people think are hard.

Refueling in orbit.  This may be impossible, but I'd bet it's at worst landing on a drone ship in the middle of the ocean hard, so SpaceX will nail it and make it routine.

Van Alan Radiation.  I'm not sure this will ever be easy, but I think it is manageable for automated ships.  The main issue is electronics and you can deal with this with redundancy and shielding.

Math.  Math is not hard.  Don't have people who think math is hard for do things that require math.

Also: what edzieba said.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/09/2023 01:35 pm
The depots are likely to be "stretched" Starships, holding perhaps 1500 tons of fuel each.
There doesn't seem to be any need for "stretched" Starships depot or tanker.  A "standard" 1200 tons works just fine at and above LEO.  The hardest and most frequent tanker use is from Earth to LEO.  That's what you need to optimize for, everything else is rounding.  And if all tankers are capable of reentry you don't need to worry about changing planes and such.  Once the fuel is transferred back to Earth goes the empty tanker.

The only thing that might need to be different for a depot or high orbit tanker is the ability to loiter by reducing boil off.

That's not to say StarShips won't get longer, but it will be like Falcon 9.  The standard version gets longer.

If you need a non-integer number of Starships to fuel the final Starship, then a stretched version makes more economic and logistical sense than boosting an extra Starship with less than a full load of fuel.

Having a depot in LEO fixes this. With low-to-zero boiloff, you can just save the propellant for the next mission.

It also means you only have to fill the final Starship in LEO once, rather than N separate tanker refuelings. This reduces the LEO loiter time of your mission vehicle substantially, which is good for MMOD risk and also just random stochastic mechanical failures.

Two starships half full is an elliptical orbit w/ 2.2km/sec of additional velocity at perigee (and one full Starship after refueling at same orbit).  if you want 3.6km/sec of additional velocity at perigee you need 885t of propellant boosted into that orbit...

Let me stop you right there.

No you don't.

You very much do not want to use "that orbit." If you want to introduce an additional refueling in HEEO, you should totally recompute the refueling ladder.

I can see you still haven't grasped the fundamental insight of harmonic refueling ladders.   :-\

you'll have boosted up 6 extra engines and tank which is a waste

This is the fundamental insight.

If you have three vehicles full of fuel in LEO, then you want to ditch one of those ships as low as possible. The lowest orbit is the one where the first tanker dumps all of its remaining fuel (other than the deorbit and landing reserve) into the other two vehicles, which fills the other two vehicles completely.

Mathematically, these two conditions completely constrain the orbital height. If the orbit is any higher, you won't fill the receiving tanks completely, meaning you "boosted up 6 extra engines and a tank" higher than necessary. If the orbit is any lower, there's not enough tank capacity in the receiving vehicles, so you waste all the propellant that can't be transferred.

Now we just have to find the height. If you know the fuel capacity and the vehicle masses, you can use the rocket equation to reverse calculate the height of this orbit. That's what my Big Spreadsheet O' Doom does.

Also: what Barley and edzieba said.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/09/2023 05:22 pm
Two starships half full is an elliptical orbit w/ 2.2km/sec of additional velocity at perigee (and one full Starship after refueling at same orbit).  if you want 3.6km/sec of additional velocity at perigee you need 885t of propellant boosted into that orbit...

Let me stop you right there.

No you don't.

You very much do not want to use "that orbit." If you want to introduce an additional refueling in HEEO, you should totally recompute the refueling ladder.
[snip]

Now we just have to find the height. If you know the fuel capacity and the vehicle masses, you can use the rocket equation to reverse calculate the height of this orbit. That's what my Big Spreadsheet O' Doom does.

Also: what Barley and edzieba said.

I found the perigee velocity, which is equivalent to finding the height.  Same thing, in orbital mechanics.  But velocity is easier as you can rocket equation your way to the answer quickly and analytically, which is more understandable than a  Big Spreadsheet O' Doom.

I think we agree, the example I just quoted was why it's a bad idea to not have the correct height (velocity)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/10/2023 09:04 am
Two starships half full is an elliptical orbit w/ 2.2km/sec of additional velocity at perigee (and one full Starship after refueling at same orbit).  if you want 3.6km/sec of additional velocity at perigee you need 885t of propellant boosted into that orbit...

... You very much do not want to use "that orbit." If you want to introduce an additional refueling in HEEO, you should totally recompute the refueling ladder. ...

Now we just have to find the height. If you know the fuel capacity and the vehicle masses, you can use the rocket equation to reverse calculate the height of this orbit. That's what my Big Spreadsheet O' Doom does.

I found the perigee velocity, which is equivalent to finding the height.  Same thing, in orbital mechanics.  But velocity is easier as you can rocket equation your way to the answer quickly and analytically, which is more understandable than a  Big Spreadsheet O' Doom.

"rocket equation your way to the answer quickly and analytically" is exactly what the Big Spreadsheet O' Doom does. ;) It just performs all the tedious math for you, including the (awkward to handle analytically) discretization of propellant into an integer number of tanker launches.

I think we agree, the example I just quoted was why it's a bad idea to not have the correct height (velocity)

Thanks, sorry, I see that now. In that case consider it as just putting a fine point on your post.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/10/2023 03:36 pm

There's an interesting middle ground here as well. You can have your tankers do a single tandem refilling, then have the remaining full tanker rendezvous with the mission vehicle in the highest possible orbit.

The advantage is that your mission vehicle only has one refilling event in HEEO, and it limits the number of passes through the Van Allen belt.

This exact refilling ladder can't be calculated using my refilling spreadsheet, but the math is pretty easy. The trick is to remember that for optimal rocket staging each stage provides half the total delta-v. Since tanker ladders are essentially identical (mathematically) to parallel staging, this means that the delta-v from LEO to the intermediate HEEO is exactly half that of the delta-v to the final HEEO.


Optimal is a widely misused term.  You should always ask optimal with respect to what?
In this case optimal means sizing the two stages.  Since we have existing Star Ships and tankers we are solving a different problem.  If we want to force the use of that "optimal" to size the stages were are constrained by the availability of small integers, and fractions composed of small integers.

Lets start with three ships, two tankers with dry mass 100 tonne, a mission vehicle with dry mass 120+150 tonne, all vehicles hold 1200 tonne of fuel, Isp is 375s so exaust velocity is 3677 m/s

Burn two tankers in parallel using half the fuel.
m0 = 100 + 1200 + 100 + 1200 = 2600
m1 = 100 + 100 + 1200 = 1400

m0/m1 = 1.857
Δv1 = 3677 * ln(1.857 ) = 2276 m/s

Meanwhile we burn the mission ship to the same delta_v  Since the exhaust velocity is the same the mass ratio will be the same so
m1mission = (1200 + 120 + 150)/1.857 = 792

As a sanity check this is between the fully loaded and fully empty masses of the mission ship.

Now we "stage" by moving the remaining 600 tonne of fuel from one tanker into the other and "discarding" the dry weight of the empty tanker.  We will burn the remaining tanker and mission ship in parallel until 1200 tonne of fuel remain to fill the mission ship.

m'0 = 1200+100+792 =2092
m'1 = 1200+100+120+150 = 1570
Δv2 = 3677 * ln( 2092/1570) = 1055

So the intermediate HEEO is at LEO + 2276
And the final HEEO is at LEO + 3331

I have not shown these are optimal (although within the constraints they are) but they are clearly possible, better than your "optimal" and don't need Wolfram alpha, just a log table.

Yep, well done. Brain fart, I don't know what I was thinking with the equal delta-v idea.

The student has exceeded the master! Always a proud day.  ;D

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/10/2023 03:38 pm
Okay, thinking about the multiple-depot case, I think I see the optimal (i.e. max ∆v) way to use a "wagon train" comprising a fully fueled Starship and n fully fueled depots. Let S be the mass of a fully fueled Starship (including payload). Assume all depos have the same wet mass (D) and the same mass of fuel (F). Assume that the specific impulse is the same for all vehicles.

In the case with a Starship and a single depo, we wanted to boost until F fuel was consumed between the two vehicles. At that point, fully fueling the Starship fully drains the depot. That this is optimal probably doesn't need elaborating. The boost that accomplishes this is ∆v = k ln((S+D)/(S+D-F)). (The trick to finding this is to see that the mass ratios for the two vehicles must be equal because both have the same ∆v.)

When you add another depo, the main thing you want to do is to discard excess mass at the first opportunity. Otherwise, you're wasting fuel. Therefore, just as in the single-depot case, you want to do a single burn that consumes F fuel. After that, you can redistribute what's left, and you will discard exactly one depot. Then repeat.

It shouldn't take much thought to see that ∆v_n = k ln((S+n*D)/(S+n*D-F)). Then sum these up to get the total ∆v--note that this does not include the ∆v from the Starship itself.

However, using what I think are reasonable numbers, if you start from a 200-km-altitude orbit, you can't use even just two depots; between the two of them, you'll go 800 m/s over escape velocity. So this trick is likely to be of interest only for a single depot.

Sounds like you got it exactly right.

This is precisely the mission architecture I used in my Spreadsheet. Now that the idea is catching on, I really ought to make a revised version that's fit for human consumption...  :-\
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/10/2023 06:42 pm

Each 1.82m segment of Starship fuel tank masses 1645kg while holding 116m3 of propellant at 986kg/m3 is 114t of propellant, or 1.44% tank mass, (which is 6x+ better than LH2).

So for an extra 1.6t of dry mass you get 115.6t of wet mass.

That's only true if all your rings are overbuilt.  Ideally* the rings are just strong enough to do the job plus appropriate margin.  For the upper tank you add each new ring at the bottom and it has to be thicker than the one above.  And you need to strengthen the entire lower tank.

An ideal tank may not fail in all modes at once, but it should be pretty close to failing in one of them, so just make it longer without additional reenforcing has limited applicability.

* Probably not on version 1.0 but eventually.  It's pretty easy to get for thinner sheets from a rolling mill.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/10/2023 06:52 pm

There's an interesting middle ground here as well. You can have your tankers do a single tandem refilling, then have the remaining full tanker rendezvous with the mission vehicle in the highest possible orbit.

The advantage is that your mission vehicle only has one refilling event in HEEO, and it limits the number of passes through the Van Allen belt.

This exact refilling ladder can't be calculated using my refilling spreadsheet, but the math is pretty easy. The trick is to remember that for optimal rocket staging each stage provides half the total delta-v. Since tanker ladders are essentially identical (mathematically) to parallel staging, this means that the delta-v from LEO to the intermediate HEEO is exactly half that of the delta-v to the final HEEO.


Optimal is a widely misused term.  You should always ask optimal with respect to what?
In this case optimal means sizing the two stages.  Since we have existing Star Ships and tankers we are solving a different problem.  If we want to force the use of that "optimal" to size the stages were are constrained by the availability of small integers, and fractions composed of small integers.

Lets start with three ships, two tankers with dry mass 100 tonne, a mission vehicle with dry mass 120+150 tonne, all vehicles hold 1200 tonne of fuel, Isp is 375s so exaust velocity is 3677 m/s

Burn two tankers in parallel using half the fuel.
m0 = 100 + 1200 + 100 + 1200 = 2600
m1 = 100 + 100 + 1200 = 1400

m0/m1 = 1.857
Δv1 = 3677 * ln(1.857 ) = 2276 m/s

Meanwhile we burn the mission ship to the same delta_v  Since the exhaust velocity is the same the mass ratio will be the same so
m1mission = (1200 + 120 + 150)/1.857 = 792

As a sanity check this is between the fully loaded and fully empty masses of the mission ship.

Now we "stage" by moving the remaining 600 tonne of fuel from one tanker into the other and "discarding" the dry weight of the empty tanker.  We will burn the remaining tanker and mission ship in parallel until 1200 tonne of fuel remain to fill the mission ship.

m'0 = 1200+100+792 =2092
m'1 = 1200+100+120+150 = 1570
Δv2 = 3677 * ln( 2092/1570) = 1055

So the intermediate HEEO is at LEO + 2276
And the final HEEO is at LEO + 3331

I have not shown these are optimal (although within the constraints they are) but they are clearly possible, better than your "optimal" and don't need Wolfram alpha, just a log table.

I find it a bit odd that you are under utilizing the tankers - they are capable of holding an extra 200t of fuel because fuel is their cargo (and 200t of fuel is merely 2 rings into the nominal cargo space, a small design change).

This changes it to:

m0 = 100 + 1400 + 100 + 1400 = 3000
m1 = 100 + 100 + 1400 = 1600

m0/m1 = 1.875
Δv1 = 3677 * ln(1.857 ) = 2311 m/s

m1mission = (1200 + 120 + 150)/1.875 = 784

Then transferring 700t of fuel to the final fueler and burning until we get 1200t of fuel remaining:

m'0 = 1400 +100+784 =2284
m'1 = 1200+100+120+150 = 1570
Δv2 = 3677 * ln( 2284/1570) = 1378

So the intermediate HEEO is at LEO + 2311
And the final HEEO is at LEO + 3689

Alas, it only takes 3.2km/sec to get to escape velocity from LEO, so under-utilizing fuelers can at most get you 3.2km/sec deltaV, unless one wishes to throw away a fueler.

In another post I'll show how how close one can get to 3.2km/sec by using stretched starships, which unlike the above 3 starship scenario, has only 2 starships.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/10/2023 07:18 pm

I find it a bit odd that you are under utilizing the tankers - they are capable of holding an extra 200t of fuel because fuel is their cargo (and 200t of fuel is merely 2 rings into the nominal cargo space, a small design change).

Mostly didactics.  Don't want to change too many things at once.

But you can use a dry cargo ship as a tanker, leaving the cargo area empty.  You can move a ton of firewood in a couple of trips in a Prius.  An F150 would be better, unless you don't have an F150.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/10/2023 07:58 pm

There's an interesting middle ground here as well. You can have your tankers do a single tandem refilling, then have the remaining full tanker rendezvous with the mission vehicle in the highest possible orbit.

The advantage is that your mission vehicle only has one refilling event in HEEO, and it limits the number of passes through the Van Allen belt.

This exact refilling ladder can't be calculated using my refilling spreadsheet, but the math is pretty easy. The trick is to remember that for optimal rocket staging each stage provides half the total delta-v. Since tanker ladders are essentially identical (mathematically) to parallel staging, this means that the delta-v from LEO to the intermediate HEEO is exactly half that of the delta-v to the final HEEO.


Optimal is a widely misused term.  You should always ask optimal with respect to what?
In this case optimal means sizing the two stages.  Since we have existing Star Ships and tankers we are solving a different problem.  If we want to force the use of that "optimal" to size the stages were are constrained by the availability of small integers, and fractions composed of small integers.

Lets start with three ships, two tankers with dry mass 100 tonne, a mission vehicle with dry mass 120+150 tonne, all vehicles hold 1200 tonne of fuel, Isp is 375s so exaust velocity is 3677 m/s

Burn two tankers in parallel using half the fuel.
m0 = 100 + 1200 + 100 + 1200 = 2600
m1 = 100 + 100 + 1200 = 1400

m0/m1 = 1.857
Δv1 = 3677 * ln(1.857 ) = 2276 m/s

Meanwhile we burn the mission ship to the same delta_v  Since the exhaust velocity is the same the mass ratio will be the same so
m1mission = (1200 + 120 + 150)/1.857 = 792

As a sanity check this is between the fully loaded and fully empty masses of the mission ship.

Now we "stage" by moving the remaining 600 tonne of fuel from one tanker into the other and "discarding" the dry weight of the empty tanker.  We will burn the remaining tanker and mission ship in parallel until 1200 tonne of fuel remain to fill the mission ship.

m'0 = 1200+100+792 =2092
m'1 = 1200+100+120+150 = 1570
Δv2 = 3677 * ln( 2092/1570) = 1055

So the intermediate HEEO is at LEO + 2276
And the final HEEO is at LEO + 3331

I have not shown these are optimal (although within the constraints they are) but they are clearly possible, better than your "optimal" and don't need Wolfram alpha, just a log table.

Here's how to get close to the maximum 3.2km/sec for (easily) recoverable fuel depots using stretched starships, and only one refueling in an HEEO, so far simpler conops, plus less amortization cost on a fueler starship that has to cross van allen belts.

Have to work backwards from the above process since we know the maximum deltaV is 3200m/sec (we could go more but then the model has to handle burn-back of the fueler)).

mratio = e3200/3677 = 2.39

Now we define a stretch factor x that is 2t of dry mass and 114t of fuel mass.

m1mission = (1200 + 116x + 120 + 150)/2.39 =  48.5x + 615

then burn to until there exactly 1200 + 114x tons of fuel left at 3200 m/sec

m'0 = 1400 + 114x + 100 + 615 + 48.5x  = 2115 + 48.5x
m'1 = 1200 + 114x + 100 + 120 + 150 = 1570 + 114x
Δv2 = 3677 * ln((2115 + 48.5x)/(1570 + 114x)) = 3200

Now solve for x.   

e3200/3677 = (2150+48.5x)/(1570+114x)
2.39(1570+114x) = 2115+48.5x
3752 + 272x = 2115 + 48.5x
1637 = 224x
x = 7.31

TL;DR = add 7-8 rings to Starship and you reduce conops to one refueling in an HEEO.

That's a stretch too far.

Counting fuel transferred to LEO would be the other way to compare this, as fully amortized fuel will be about $25/kg.

Going all the way to 3.2km/sec also requires weeks of waiting and interference with moon orbits.  so IMHO the orbit needs to have an apogee of 318,000km (just outside Moon's SOI)

I'll save that calculation for another day, my lovely ex-fiancee wants to go to the beach now.


EDIT:  Some bad addition and some dyslexia fixed.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/10/2023 09:19 pm

Here's how to get close to the maximum 3.2km/sec for (easily) recoverable fuel depots using stretched starships, and only one refueling in an HEEO, so far simpler conops, plus less amortization cost on a fueler starship that has to cross van allen belts.

Far simpler conops?   Slightly simpler, barely enough to notice, what do you think is so hard about rendezvous in an elliptical orbit?

Your also over weighting the danger to electronics from the van allen belts.  Lots of satellites orbit in and through them.  Although some sources argue that 20,000 < 13,000 and 60,000  < 36,000 and say everything avoids them because instant death and all.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/10/2023 11:36 pm
...
Now we define a stretch factor x that is 2t of dry mass and 114t of fuel mass.
...

x = 9.88

TL;DR = add 10 rings to Starship...

As an aside, I really like this lazy math trick.  :)

Going all the way to 3.2km/sec also requires weeks of waiting and interference with moon orbits.  so IMHO the orbit needs to have an apogee of 318,000km (just outside Moon's SOI)

I'll save that calculation for another day, my lovely ex-fiancee wants to go to the beach now.

A 200 x 318,000 km orbit has a period of 186 hours, and a delta-v from LEO of 3165 m/s.

Also we normally refer to it as a "wife" and/or "husband."  :D
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/11/2023 12:31 am
...
Now we define a stretch factor x that is 2t of dry mass and 114t of fuel mass.
...

x = 9.88

TL;DR = add 10 rings to Starship...

As an aside, I really like this lazy math trick.  :)

Going all the way to 3.2km/sec also requires weeks of waiting and interference with moon orbits.  so IMHO the orbit needs to have an apogee of 318,000km (just outside Moon's SOI)

I'll save that calculation for another day, my lovely ex-fiancee wants to go to the beach now.

A 200 x 318,000 km orbit has a period of 186 hours, and a delta-v from LEO of 3165 m/s.

Also we normally refer to it as a "wife" and/or "husband."  :D

I had both dyslexia and bad arithmetic, the actual answer is x=7.33 or 7-8 rings (could get away with 7 w/ smaller  cargo section by one ring).

Beach was very nice.  Got what is probably my last big vitamin-D and nitric acid boost of the year.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/11/2023 12:40 am

Here's how to get close to the maximum 3.2km/sec for (easily) recoverable fuel depots using stretched starships, and only one refueling in an HEEO, so far simpler conops, plus less amortization cost on a fueler starship that has to cross van allen belts.

Far simpler conops?   Slightly simpler, barely enough to notice, what do you think is so hard about rendezvous in an elliptical orbit?

Your also over weighting the danger to electronics from the van allen belts.  Lots of satellites orbit in and through them.  Although some sources argue that 20,000 < 13,000 and 60,000  < 36,000 and say everything avoids them because instant death and all.

The conops of the dual boost is really rather difficult, and from an Oberth perspective, possibly not very useful from a C3 standpoint.

You get the initial 2.3km/sec with 3 ships, then your merge to 2 ships.  Then you boost to 3.6km/sec before the final fuel transfer.  You are now on an escape trajectory and leaving the deep gravity well quickly, and you have to transfer the remaining fuel to the final starship and then do the final boost.  I don't think you can do that fast enough, which means you are getting no Oberth boost and are losing all the energy you would have gained by doing the final thrust deepest in the gravity well.

I haven't calculated this from the C3 standpoint, but it's entirely possible your Vinf from this might belower than just the one refuel at apogee and final burn w/ only two ships.   The Oberth boost to Vinf is about 2x or so at 200km altitude from Earth.

If it takes an hour to transfer the fuel (which might be optimistic) you'll be 36,000km above Earth before doing the final boost and will have lost a whole lot of velocity.  That's a lot of Oberth loss.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/11/2023 12:54 am
The conops of the dual boost is really rather difficult, and from an Oberth perspective, possibly not very useful from a C3 standpoint.

You get the initial 2.3km/sec with 3 ships, then your merge to 2 ships.  Then you boost to 3.6km/sec before the final fuel transfer.  You are now on an escape trajectory and [problems]...

Yes. For that example, you'll want to only partially fill one of the tankers. That's why I wrote the spreadsheet to support partial loads of fuel.



New Spreadsheet just dropped. Added better instructions and lots of formatting cleanup. Enjoy!

More improvements coming soon.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/11/2023 02:11 am
The conops of the dual boost is really rather difficult, and from an Oberth perspective, possibly not very useful from a C3 standpoint.

You get the initial 2.3km/sec with 3 ships, then your merge to 2 ships.  Then you boost to 3.6km/sec before the final fuel transfer.  You are now on an escape trajectory and [problems]...

Yes. For that example, you'll want to only partially fill one of the tankers. That's why I wrote the spreadsheet to support partial loads of fuel.



New Spreadsheet just dropped. Added better instructions and lots of formatting cleanup. Enjoy!

More improvements coming soon.

It doesn't matter about partial refill (plus/minus waste arguments) 

You can only get fuel up to +3km/sec in Earth's SOI, above that you either run into Moon interactions problems, multi-week orbits, or being unable to do the final boost at perigee.   Pick your poison.

Now, there may be an orbit involving Moon gravity assist that gets your final burn at > 3km/sec at perigee, but good luck calculating that w/ a spreadsheet. (I've done it with a Jupiter Oberth, and I'm not completely sure I got it right).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/11/2023 02:24 am
The conops of the dual boost is really rather difficult, and from an Oberth perspective, possibly not very useful from a C3 standpoint.

You get the initial 2.3km/sec with 3 ships, then your merge to 2 ships.  Then you boost to 3.6km/sec before the final fuel transfer.  You are now on an escape trajectory and [problems]...

Yes. For that example, you'll want to only partially fill one of the tankers. That's why I wrote the spreadsheet to support partial loads of fuel.



New Spreadsheet just dropped. Added better instructions and lots of formatting cleanup. Enjoy!

More improvements coming soon.

It doesn't matter about partial refill (plus/minus waste arguments) 

You can only get fuel up to +3km/sec in Earth's SOI, above that you either run into Moon interactions problems, multi-week orbits, or being unable to do the final boost at perigee.   Pick your poison.

Yes, effectively that's the situation. As I said a week ago:

To make the math very simple: tandem refilling allows any Starship-compatible vehicle configuration to gain (up to) 3.2 km/s of additional delta-v from LEO. Effectively, your vehicle begins its departure burn at (approaching) Earth escape velocity, instead of a ~circular orbit.

by choosing a more pragmatic (shorter) orbital period, effectively the tradeoff is that you "only" get a boost of 3.0-3.1 km/s, vs 3.2 km/s.




but good luck calculating that w/ a spreadsheet

I don't understand your disdain for spreadsheets, which is evidently difficult for you to contain.

Spreadsheets are great tools for early high-level mission planning like this.

Of course every spreadsheet doesn't perform every calculation. What kind of expectation is that?  ???

Now, there may be an orbit involving Moon gravity assist that gets your final burn at > 3km/sec at perigee... (I've done it with a Jupiter Oberth, and I'm not completely sure I got it right).

Moon Oberth (accelerating toward Earth using a pericynthion burn to accelerate toward Earth beyond escape velocity) won't help nearly as much as simply using that same fuel to have a longer Earth perigee burn, because the Moon's gravity well is much shallower.

An unpowered lunar gravity assist (free-return trajectory) gives a perigee velocity of 10.92 km/s (https://en.wikipedia.org/wiki/Free-return_trajectory), which still doesn't exceed the escape velocity of 11.03 km/s at the same perigee height.


HEEO gets you 3.0-3.165 km/s, and lunar only gives an extra 0.05. If you want more performance beyond that, we're back to a StarPusher stage.

While obviously you can build a smaller one, ideally (for optimal rocket staging) a hypothetical StarPusher would be similar in size to Super Heavy.   :o  It only needs 1/2 to 1/3rd as many engines, however.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/11/2023 03:12 am
A 300/310,000 orbit gives the maximum elliptical orbit that doesn't have to consider the
moon's orbit.  (This is just below Moon's SOI).  It's still an orbital period of ~7 days
so more than one trip around one will have to seriously consider evaporation in the calculations.

That's an orbital velocity of 10812 m/sec at perigee.  That's a deltaV added to LEO of 3.09 km/sec.  This
sets the limit of how fast you can get fuel to go at Earth Perigee, which is where any Oberth boost
needs to happen.

mratio = e3090/3677 = 2.32

Now assume we stretch the standard EDL capable cargo startship by 3 segments (which was discussed
in various places that accompanies 3 extra Raptor-Vacs, so it'll be standard in 10 years)

m1mission = (1532 + 126 + 150)/2.32 = 779

Now we define a stretch factor x that is 2t of dry mass and 114t of fuel mass.  This is
only for the fueler starship.

then burn to until there exactly 1532 tons of fuel left at 3090 m/sec for the final burn:

m'0 = 1400 + 114x + 100 + 779   = 2279 + 114x
m'1 = 1400 + 114x + 100 + 120 + 150 = 1770 + 114x
Δv2 = 3677 * ln((2279 + 114x)/(1770 + 114x)) = 3090

Now solve for x.

e3090/3677 = ((2279 + 114x)/(1770 + 114x))
2.32(1770+114x) = 2279+114x
4106 + 264x = 2279 + 114x
1907 = 150x
x = 13.24

TL;DR = add 13 rings to the fueler Starship and you reduce conops to one refueling in an HEEO,
and get a full Oberth boost at perigee after one apogee refueling.

This is still a stretch too far.  That's 3 rings (which will be std by then) plus 10 rings into the payload
section, which for a fueler, is the payload.  But the payload section is only about 7 rings worth of fuel.

Thus we can define the maximum fuel payload for an EDL capable fueler (EDL lets you return to LEO) as
10 rings on top of 1200t, or 2,340t of fuel.

So with all these constraints, what kind of deltaV above LEO can we get for a perigee burn with
a full load of fuel with payload?  With just one reusable tanker.  Then we can compare to two tankers
and decide whether the extra deltaV is worth the cost.

m'0 = 2340 + 100 + 120 + 150 + 1532  = 4242
m'1 = 1532 + 100 + 120 + 150 = 1902
Δv2 = 3677 * ln((4242/1902 = 2,949

Thus if we can make an EDL capable reusable fuel depot that can hold 2340t of fuel we can
get, with one HEEO refuel, a perigee deltaV of 2,949 m/sec.
  Such a depot doesn't deviate
by the current design except for the addition of 3 rings and the payload section being used for fuel.

That's 19 LEO fuel trips to fuel all this.  That's about $95M of launch costs at $5M/fuel trip.

It requires two refuels and two tankers to add a mere 141m/sec.  That hardly seems worth it.
It'd be better to just reduce the rated payload capability by a few 10s of tons.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/11/2023 03:45 am
A 300/310,000 orbit

...has a perigee that's too high. :)  That's why you're getting (slightly) worse numbers for your Oberth.

A 300/310,000 orbit gives the maximum elliptical orbit that doesn't have to consider the
moon's orbit.

Obviously you can consider higher orbits than this. You just need to use higher resolution numerical calculations, which you were going to do during later stages of mission planning anyway.

So let's keep in mind that this 3.09 km/s number isn't a "real" upper bound.

mratio = e3090/3677 = 2.32

Now assume we stretch the standard EDL capable cargo startship by 3 segments (which was discussed
in various places that accompanies 3 extra Raptor-Vacs, so it'll be standard in 10 years)

m1mission = (1532 + 126 + 150)/2.32 = 779

Now we define a stretch factor x that is 2t of dry mass and 114t of fuel mass.  This is
only for the fueler starship.

then burn to until there exactly 1532 tons of fuel left at 3090 m/sec for the final burn:

m'0 = 1400 + 114x + 100 + 779   = 2279 + 114x
m'1 = 1400 + 114x + 100 + 120 + 150 = 1770 + 114x
Δv2 = 3677 * ln((2279 + 114x)/(1770 + 114x)) = 3090

Now solve for x.

e3090/3677 = ((2279 + 114x)/(1770 + 114x))
2.32(1770+114x) = 2279+114x
4106 + 264x = 2279 + 114x
1907 = 150x
x = 13.24

TL;DR = add 13 rings to the fueler Starship and you reduce conops to one refueling in an HEEO,
and get a full Oberth boost at perigee after one apogee refueling.

Thinking outside the box. I like it!  :)

Probably it's not economical to use a custom vehicle here (vs two standard tankers), but it's an interesting thought experiment nonetheless.

This is still a stretch too far. ... Thus we can define the maximum fuel payload for an EDL capable fueler

I don't see why, and your "maximum" seems totally arbitrary.

You compare it to Starship as if larger aerospace structures (with more rings) are impossible, but it seems a more apt comparison would be Super Heavy.

EDL lets you return to LEO

You don't need full EDL capabilities to return to LEO, just aerobraking.


So with all these constraints

[sic]  8)


what kind of deltaV above LEO can we get for a perigee burn with a full load of fuel with payload with just one reusable tanker? Then we can compare to two tankers and decide whether the extra deltaV is worth the cost.

[snip]

This is the wrong comparison, and therefore the wrong math.

A custom enlarged tanker is the more costly option vs using two (also reusable) standard tankers.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/11/2023 04:02 am
A 300/310,000 orbit gives the maximum elliptical orbit that doesn't have to consider the
moon's orbit.

Why on or off of Earth should that be a constraint?  We know where the Moon is.  We know how gravity works.  We know how to handle weak instabilities.  It's rocket science, equations are a lot tamer than 1000 tonne of LOX.

"Must be able to calculate trajectory" is a valid constraint.  "Must be able to compute trajectory on the back of an envelope" is not. 
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/11/2023 04:16 am
A custom enlarged tanker is the more costly option vs using two (also reusable) standard tankers.

My final proposal has no enlarged custom tanker.

It's merely a standard "+3" ring Starship with the domes moved up such that the entire thing is a fuel tank (i.e. including the cargo section), or alternatively the cargo section as a separate parallel fuel tank suitable for refueling other ships only.  Everything else is bog standard (though you could remove the header tanks they aren't needed, it isn't coming back to Earth, just LEO).

I suspect there are multiple uses, it's basically a "depot" ship.  You could use the same "depot" to refuel 3-4 Starships headed to Mars, thus decoupling launch schedules from MTO windows.

I'm really wondering why anyone cares about +300m/sec, for some non-trivial amount of additional fuel being boosted to LEO.

In the dual tanker setup that gets you the full 3.2km/sec, how many fuel loads up to LEO at 200t/ launch?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/11/2023 04:09 pm
Has anyone worked out the lifetime of a full depot in LEO at 200 km altitude? I'm wondering if that's actually short enough to be a problem.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/11/2023 04:29 pm
Has anyone worked out the lifetime of a full depot in LEO at 200 km altitude? I'm wondering if that's actually short enough to be a problem.

short time from heating/boiling, or from atmospheric friction?

The latter is trivially handled by an occasional burn from one raptor.  I wonder how short a "burp" you can get out of a Raptor.

Boil-off might be enough
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/11/2023 04:34 pm
Has anyone worked out the lifetime of a full depot in LEO at 200 km altitude? I'm wondering if that's actually short enough to be a problem.

short time from heating/boiling, or from atmospheric friction?

The latter is trivially handled by an occasional burn from one raptor.  I wonder how short a "burp" you can get out of a Raptor.

Boil-off might be enough
I'm thinking about atmospheric friction. I figure they can handle boiloff through active cooling if they really have to, although for a "roving depot," such as we've been discussing, you'd obviously want as little extra mass as possible.

Even if you do occasionally do a Raptor burn, how fast will you run through fuel? Also, when the depot first goes up, it'll be pretty empty, so the effect of atmospheric drag will be worse.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 09/11/2023 05:05 pm
After first BECO (post stage-flip-and-boostback), the main tanks are for all intents and purposes empty. Any propellant remaining is dead weight - the main tank lockout valves will close and propellant feed will completely cut over to the stacked header tanks just above the aft dome (possibly with some residual LCH4 within the downcomer) for reignition of the centre engines for the landing burn. In theory switchover could occur during the last few moment before staging once the outer engines have shut down (may be desirable to ensure switchover occurs under positive controlled thrust rather than microgravity or during atmospheric braking) but that trades operational simplicity for oversizing the headers.
Whether it is feasible or desired to dump the residuals in the main tank is another question: dumping reduces dry mass for the landing burn and catch, but dumping also ejects 'free' ullage pressurant for stiffening the main tanks during EDL. Dumping LOX in particular reduces heating of the prop in the headers, as you do not have 'warm' closer-to-boiling LOX sloshing about a thin skin of metal away from the ideally-still-subchilled header props. Then there's feasibility: dumping LOX isn't too bad, but dumping uncombusted LCH4 is a pain both due to flammability and the environmental issues of ejecting it into the upper atmosphere. Dumping LOX and retaining LCH4 may also have a benefit in moving CoG of the landing booster up, which reduces the pendulum effect from TVC on the pose of the forward dome (and thus catch points) at velocities below those when the grid-fins provide useful control authority.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/11/2023 05:27 pm
Has anyone worked out the lifetime of a full depot in LEO at 200 km altitude? I'm wondering if that's actually short enough to be a problem.

The problem as I see it is that LEO is not a single thing.  It's actually a bunch of different "places" separate by various, sometimes large, Δv.  An equatorial LEO solves some of this, but SpaceX is not French, and Panamizing Guiana is Le Terrible idée.

For me a "depot" is simply a tanker that has not yet transferred its fuel, but that is not enough '60s space opera for some.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 09/11/2023 05:56 pm
For me a "depot" is simply a tanker that has not yet transferred its fuel...
We do not yet know that. We know that the HLS mission concept requires transfer from a Tanker to a Depot, and from the Depot to a Starship HLS. There is no known actual requirement for any other type of transfer. All of the specialized transfer equipment, if any, may be aboard the Depot and not part of any other SS.

In particular, universal any-to-any transfer would require an androgynous transfer interface. That is certainly possible to design, but is in general more complex and heavier than a non-androgynous interface. IDSS is an example.

I am not saying that SpaceX will not implement any-to-any transfer. I'm saying that we do not know.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/11/2023 10:19 pm
For me a "depot" is simply a tanker that has not yet transferred its fuel, but that is not enough '60s space opera for some.
That's not NASA or SpaceX's view (https://ntrs.nasa.gov/api/citations/20220013431/downloads/HLS%20IAC_Final.pdf), though. Figure 2 (https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=50157.0;attach=2142897;image) makes a clear distinction between a tanker and a depot. Or that was their thinking as of one year ago, anyway.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/12/2023 05:48 am
Specialization is your friend.

Tankers can be specialized to carry as much fuel as possible, by stripping out all unnecessary mass. Every kilogram of mass reduction is another kilogram of fuel.

Depots can be specialized by enlarging the tanks, adding thermal insulation, and optionally adding a cryocooler. You can also add solar panels and electric thrusters (Hall effect or arcjet) to efficiently counteract drag and change orbits.

Aggregating fuel in depots are also nice for your mission architecture, because it minimizes refueling events for the high-value mission ship. This also minimizes the time for the mission ship to loiter in orbit (vs a plan which uses it as the 'depot' to aggregate fuel), reducing MMOD risk, boiloff, and depreciation.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 09/12/2023 07:53 am
Re depots and boiloff, and considering that a depot can still be useful even if it reaches orbit with nothing spare...

If you use thicker (assume x2) steel for the depot tanks, to withstand higher pressure from higher temperature LCH4/LO2, how much would it reduce the need for passive or active cooling?

Assume this is for months long on-orbit storage. When any maneuvers or transfers happen the depot can spool up / subcool first.
Case 1) LEO with optimal paint & attitude (barbecue roll? light sun side/dark sky side?) and nearly half of its orbit is in shade
Case 2) HEEO with standard TPS, attitude control not active, and a negligible fraction of its orbit in shade

I think vapour pressure scales with T and emission scales with T⁴, but I suspect that might be the difference between nothing and nothing until you have a JWST style sunshade

P.S. for me, the necessary (semantic) defining feature for "depot" is aggregation. It must receive propellant transfer to qualify. It will necessarily (plan to) perform multiple sequential rendezvous in one flight - so effectively a station.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/12/2023 01:46 pm
P.S. for me, the necessary (semantic) defining feature for "depot" is aggregation. It must receive propellant transfer to qualify. It will necessarily (plan to) perform multiple sequential rendezvous in one flight - so effectively a station.
Another key point, which Dan alluded to above, is the gender of the fueling plugs.

A Starship is fueled on the ground through the "Quick Release" (QR) panel, which is a smooth plate with a set of sockets on the side of a Starship. It has big holes for adding the LOX and liquid CH4 and small ones for extracting the gases from the tanks. This is generally called a "female" connector.

The ground support equipment (GSE) has the corresponding male connector on the end of a bundle of hoses. It draws propellants from the big tanks on the ground and pumps them into the Starship.

A tanker, then, is just an ordinary Starship with bigger tanks (and no cargo space). But a depot has to have a GSE connector--possibly in addition to the usual QR panel, although there's a strong desire to minimize cryogenic plumbing and to keep it as simple as possible.

Another distinction is that a tanker wouldn't have pumps to transfer the fuel. Only a depot would have that. As you say, it doesn't matter if a depo reaches orbit with zero leftover fuel, but as Twark says, every kilogram of mass saved on the tanker is another kilogram of fuel per trip. So a tanker is extremely stripped down whereas a depot is loaded with extra equipment.

Our long discussion about "roving depots" has conveniently ignored this distinction, but it's a serious one.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: volker2020 on 09/12/2023 02:58 pm
Has anybody calculated what delta-v / mass you could get to tank one tanker and one starship full in leo, dock them and let the tanker boost the ship as high as possible in a ecliptic orbit? Tanker than returns to leo using aero break.

Seems to me like the simplest solution (no docking in HEO), no tanking in HEO, just one separation and worst case an easy return.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/12/2023 04:21 pm
Has anybody calculated what delta-v / mass you could get to tank one tanker and one starship full in leo, dock them and let the tanker boost the ship as high as possible in a ecliptic orbit? Tanker than returns to leo using aero break.

Seems to me like the simplest solution (no docking in HEO), no tanking in HEO, just one separation and worst case an easy return.
In terms of delta-v, they're absolutely the same, all other things being equal. The two are accelerating in tandem, whether physically attached or not, and they are burning the same amount of fuel (between the two of them) to attain the same ∆v. The only difference is that in the refueling case, the tanker has to (somehow) pump the rest of its fuel into the mission Starship once the burn is done. In the "piggyback" scenario you're describing, the tanker burns all of its fuel before the mission ship does. That nicely handles the problem of the tanker having the wrong gender of connector (not to mention having nothing to pump fuel with), but it requires a really strong connection between the two vehicles, which is going to add a good bit of mass (I suspect).

The other point no one has worked out (as far as I know) is how much ∆v you can get if the tanker reserves some fuel to do an apogee burn to allow it to aerobrake. (Aerobreaking is considered a bad outcome.) :-) It's a cute little problem that actually seems to have an analytic solution, which I'll post once I've had time to finish working it out.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 09/13/2023 12:47 am

Assume this is for months long on-orbit storage.


Why do you assume this?  It only makes sense if your launch cadence sucks.  It's essentially saying assume no rapid reuse. 

P.S. for me, the necessary (semantic) defining feature for "depot" is aggregation. It must receive propellant transfer to qualify. It will necessarily (plan to) perform multiple sequential rendezvous in one flight - so effectively a station.
For me the definition of depot is you leave it in orbit for two or more missions.  Orbital mechanics makes this hard. In the near term unless you plan on replacing GEO satellites wholesale or something there won't be enough missions going to the same place.  If you are going to Mars, you want to work really hard on your launch cadence and hitting your launch windows and refueling schedules.

Also you are all obsessed with gender.  Gender swapping adaptors are ubiquitous and small.  A tanker with a gender adaptor that accumulates from several launches in a week before transferring to the mission ship is still a tanker.   It's role may be depot, but if every ship in the fleet can perform that role with minimal work it's a tanker.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 09/13/2023 12:53 am
I disagree. A depot could last for just one mission. Bu It’s it has to receive fuel from something else and give it to something else, or it’s just a tanker.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 09/13/2023 12:56 am
Has anybody calculated what delta-v / mass you could get to tank one tanker and one starship full in leo, dock them and let the tanker boost the ship as high as possible in a ecliptic orbit? Tanker than returns to leo using aero break.

How do you attach the kick-stage to the mission ship? It has to support the fully fuelled mission ship during the burn, and you want that burn to be fairly high-thrust to maximise Oberth efficiencies. That's fundamentally different to the docking needed for a prop-transfer and much more like stage integration for launch.

If you shed the nose cone on the kick-stage during launch, leaving a ring-mount similar to the new SH hot-staging ring, but with extra prox-ops/docking hardware, then the kickstage can't EDL.

OTOH: If you use a tanker capable of EDL, then how does it support the mission ship. (Which presumably has to be inline with the tanker.)

Presumably, the kickstage would be easier, but it means yet-another-variant, and is either expendable or adding its own complexity for reuse (such as getting into an orbit with the correct ascending-node-angle for its next mission, plus ability to remain on-orbit in a useful/reliable state between interplanetary missions.)

[Plus I feel like discussion of such a kick-stage is more suited to its own thread, if there isn't one already.]
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/13/2023 01:21 am
The other point no one has worked out (as far as I know) is how much ∆v you can get if the tanker reserves some fuel to do an apogee burn to allow it to aerobrake.

My spreadsheet does let you reserve a quantity of fuel for deorbit and landing. Because the ∆v is so small, to avoid circular dependencies it's treated as a constant mass that doesn't change based on apogee height.


Using vis-viva, it's relatively easy to show that

        ∆v2 = 2 GM [ 1 / (ra + rreentry) - 1 / (ra + rp) ]

where rp is the perigee, ra is the apogee, and rreentry is the perigee radius for reentry (or aerobraking).

For a 200 km perigee and reentry at 80 km, the graph of ∆v (m/s) vs apogee (km) looks like this (https://www.wolframalpha.com/input?i=plot+sqrt%28797472061.2+%28+1%2F%28x+%2B+80+%2B+2+*+6571%29+-+1+%2F+%28x+%2B+200+%2B+2+*+6571%29+%29%29+from+x%3D200+to+100000).


EDIT: this equation and graph are wrong; see this post (https://forum.nasaspaceflight.com/index.php?topic=50157.msg2524493#msg2524493) for the correct equation.


I think it's safe to say that it's not a big concern.  :)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 09/13/2023 01:25 am
Re: Tanker vs Depot.

I see three classes:

Launch tanker:
EDL capable Starship (with some allowance for early missions if EDL is not solved but SpaceX needs to launch prop for Artemis demo + III.) Extra rings on tank sections, fewer on nose-section. Probably no pumping ability. Visually, it'll look like any other Starship: nose, heatshield, flapperons.

Normal depot:
Non-EDL-capable. Specialised variant of Starship. Has the extra bits required for propellant accumulation and storage. Perhaps solar arrays, sun-shield, pumps, coolers/radiators, different thrusters, etc. I'm not picking a list now that defines "depot", just "whatever turns out to be required", it will be on this vehicle, with as little as possible on the tanker. On the flip-side, it strips out the equipment necessary for EDL, so it will look quite distinct. Less "shuttle", more "skylab".

Orbital propellant transfer depot/tanker:
Moves propellant from one orbit to significantly different one. Can be any combination of properties of the pure-tanker and the pure-depot. It has to be able to dock with and transfer prop to a mission ship, but other that that it might be very close to a pure depot, or it might be EDL capable and closer to a tanker, or anywhere in between. It might be accumulation-fuelled by tankers, or it might grab a single load from a pre-filled regular depot. It might be called a "depot" or a "tanker", depending on the specifics of its mission, and the naming is likely to be pretty arbitrary.
Using these orbital prop transfer vehicles to try to define what is a depot or tanker is therefore an exercise in futility.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/13/2023 05:21 pm
Also you are all obsessed with gender.  Gender swapping adaptors are ubiquitous and small.  A tanker with a gender adaptor that accumulates from several launches in a week before transferring to the mission ship is still a tanker.   It's role may be depot, but if every ship in the fleet can perform that role with minimal work it's a tanker.
That's a fair point, although when it comes to cryogenic fluids, you shouldn't be too quick to assume that something is trivial. But it's about more than just gender; the real issues, as I see it are, pumps, insulation, and ullage burns.

A depot has to be capable of pumping propellants into/out of other vehicles, and it's going to need something to power those pumps. Even if that's just a modest array of solar panels, it adds weight.

As NASA and SpaceX have said, the first depot will only have passive measures to prevent boiloff, but that likely means a coating of something like Solar White (meaning no reenty tiles) and probably a multilevel insulator like the Webb telescope has.

Finally, it's going to need some engine other than a Raptor that's capable of doing very long, very low-thrust burns for the duration of every fueling operation.

Add all of that together, and a depot ends up sporting quite a bit of hardware that you really don't want on any other Starship variant.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 09/13/2023 05:39 pm
Also you are all obsessed with gender.  Gender swapping adaptors are ubiquitous and small.  A tanker with a gender adaptor that accumulates from several launches in a week before transferring to the mission ship is still a tanker.   It's role may be depot, but if every ship in the fleet can perform that role with minimal work it's a tanker.
That's a fair point, although when it comes to cryogenic fluids, you shouldn't be too quick to assume that something is trivial. But it's about more than just gender; the real issues, as I see it are, pumps, insulation, and ullage burns.

A depot has to be capable of pumping propellants into/out of other vehicles, and it's going to need something to power those pumps. Even if that's just a modest array of solar panels, it adds weight.

Okay, how much weight, and how does it affect various scenarios?  I don't recall seeing any rocket equation solutions.

For example, suppose in the scenario where we want to boost 2400t of fuel to a 3km/sec HEEO, and the dry mass is nominally 130t, and we add 10t of solar panels and cooler to it, what does that do?

The plane jane version leaves you with 989t of fuel remaining.  The solar panel/cooler version leaves you with 983t of fuel remaining.  0.6% difference.

6t is rounding error at these large depot size setups.   I bet the 7 day orbit involved will have enough boil off that solar panels w/ cooler would want to be standard equipment anyways.

The conops scenario that needs to be mass efficient is the launching of fuel from the surface of earth.  Any depot/fueler operating at higher orbits doesn't need to be mass efficient because the mass of fuel is so high it's rounding error.

So there will be eventually two fuelers:   a first for earth-LEO and a second for conops from LEO to higher orbits or injection trajectories.  I think eventually that the second design may have heat shield so if HEEO becomes a standard procedure for getting stuff to the moon, high speed runs to Mars, or high deltaV destinations beyond.





Quote
Finally, it's going to need some engine other than a Raptor that's capable of doing very long, very low-thrust burns for the duration of every fueling operation.

That's an assumption.  We haven't seen how propellant transfer works yet.  We have seen videos of how zeroG fluid transfers work with a type of surface tension siphon.  No burns required.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 09/14/2023 12:03 am
Using vis-viva, it's relatively easy to show that

        ∆v2 = 2 GM [ 1 / (ra + rreentry) - 1 / (ra + rp) ]

where rp is the perigee, ra is the apogee, and rreentry is the perigee radius for reentry (or aerobraking).
Ah, but I'm trying to solve for the case where you don't know ra a priori. You know the total impulse of the two burns (i.e. a fixed amount of fuel), and you know rp and rreentry, but that's all. In particular, you don't know how big the initial burn is going to be--you have to compute it.

This is still solvable (also using via viva) but not quite so straightforward.

However, as you say, the second burn is so tiny that my enthusiasm for the problem has cooled somewhat. :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 09/16/2023 03:12 am
Ah, but I'm trying to solve for the case where you don't know ra a priori. You know the total impulse of the two burns (i.e. a fixed amount of fuel), and you know rp and rreentry, but that's all. In particular, you don't know how big the initial burn is going to be--you have to compute it.

This is still solvable (also using via viva) but not quite so straightforward.

However, as you say, the second burn is so tiny that my enthusiasm for the problem has cooled somewhat. :-)

Good luck!

Sadly, it turns out my earlier equation was a bit too simplistic to be correct. So far, the simplest correct equation I can work out for calculating the deorbit ∆v is:

        ∆v = sqrt(2 GM/ra) (sqrt(rp/(ra + rp)) - sqrt(rreentry/(ra + rreentry)))

...which for a perigee of 200 km and 80 km reentry looks like this (https://www.wolframalpha.com/input?i=plotsqrt%282+*+398736030600+%2F+%286371+%2B+x%29%29+*+%28sqrt%28%286371+%2B+200%29%2F%28x+%2B+200+%2B+2+*+6371%29%29+-+sqrt%28%286371+%2B+80%29%2F%28x+%2B+80+%2B+2+*+6371%29%29%29+from+x%3D200+to+x%3D100000).

Apologies for the error. If anyone wants to build off this result, it would help if it's correct...    :o




If you want to analytically find ra a priori, you "simply" need to solve the following equation for ra:

        ∆vtotal = ∆vdeorbit + ∆vinjection = sqrt(2 GM/ra) (sqrt(rp/(ra + rp)) - sqrt(rreentry/(ra + rreentry))) + sqrt(GM/rp) (sqrt(2 ra / (ra + rp)) - 1)

As I said, good luck!  8)  Honestly, it's probably a lot easier to just use numerical methods here.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/15/2023 10:32 pm
This is a reasonable definition of a depot:

a: a place for storing goods or motor vehicles
b: store, cache

https://www.merriam-webster.com/dictionary/depot (https://www.merriam-webster.com/dictionary/depot)

A depot is for storage. A tanker is a means of transport.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/15/2023 10:46 pm
Non political thoughts on gender identity.

The ground based QD is male. Tankers and  other variants, other than depot, are female to allow mating. To mate with other ships the depot must be male. How to do this?

Simplest solution is to build the depot with male QD and put a female-female adapter on the ground based QD. Put it on and take it off as necessary. Extra mass stays on the ground. Depot launches should be rare enough that this will not be a serious impediment to ops.

The ground QD retraction mechanism and protective dog house may need a redesign to work with and without the adapter, but for SX that should be a minor issue.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: OTV Booster on 10/15/2023 11:13 pm
Current plans for Artemus III is for passive depot cooling. While I fully expect rapid turnaround eventually, expecting it for Artemus III would be (opinion) a loosing hand. The ships may actually turn around faster than the pad but even that might not be all that fast early on.


IIRC SX has been able to turn an F9 pad around at the Cape in four days and that only recently. Using one pad at BC and one at the cape, maybe two launches a week? That may call for more than two tankers if tanker turnaround is slow. We have to wait to find out. So does SX.


The upshot of this is the depot may have to loiter quite a while waiting for tankers and boiloff may be more than a nominal problem. However long the depot and tankers are stretched, it may need more tanker flights than expected and it'll  probably need one last final topoff just before LSS shows up. Not a deal killer but a kink to keep in mind.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/14/2023 10:00 pm
Re: Tanker vs Depot.

I see three classes:

Launch tanker:
EDL capable Starship (with some allowance for early missions if EDL is not solved but SpaceX needs to launch prop for Artemis demo + III.) Extra rings on tank sections, fewer on nose-section. Probably no pumping ability. Visually, it'll look like any other Starship: nose, heatshield, flapperons.

Normal depot:
Non-EDL-capable. Specialised variant of Starship. Has the extra bits required for propellant accumulation and storage. Perhaps solar arrays, sun-shield, pumps, coolers/radiators, different thrusters, etc. I'm not picking a list now that defines "depot", just "whatever turns out to be required", it will be on this vehicle, with as little as possible on the tanker. On the flip-side, it strips out the equipment necessary for EDL, so it will look quite distinct. Less "shuttle", more "skylab".

Orbital propellant transfer depot/tanker:
Moves propellant from one orbit to significantly different one. Can be any combination of properties of the pure-tanker and the pure-depot. It has to be able to dock with and transfer prop to a mission ship, but other that that it might be very close to a pure depot, or it might be EDL capable and closer to a tanker, or anywhere in between. It might be accumulation-fuelled by tankers, or it might grab a single load from a pre-filled regular depot. It might be called a "depot" or a "tanker", depending on the specifics of its mission, and the naming is likely to be pretty arbitrary.
Using these orbital prop transfer vehicles to try to define what is a depot or tanker is therefore an exercise in futility.

Apologies for the tardy response; I kinda lost track of this thread.

I think a hybrid tanker/depot only makes sense if it's EDL-capable.  Otherwise, just use a depot.  And if it's EDL-capable, then the depot-ish parts of it have to be capable of surviving EDL.  That means they have to stow themselves or be positioned on spots on the dorsal side of the vehicle, where they don't mess up the aerodynamics too much.  If all you need is a gender-bender to connect the two QDs and a near-CoM stabilizing dock / grapple / tether / whatever, that might be OK.

If you're going to do HEEO refuelings, then depots need to be capable of >1000m/s orbital maneuvers, for two main reasons:

1) Accumulating prop in a high energy orbit is extremely expensive.  It'll always make more sense to return the depot to VLEO for accumulation, even if the return is propulsive, than it will to have all of the tankers waste a substantial amount of prop getting up to the depot.

2) While it may be OK to deploy a depot operationally to an HEEO, MMOD risk makes it unwise to store it in that orbit.  Note that the depot is simultaneously at risk from MMOD and of becoming MMOD.

If you have a depot in something like a 200 x 9800km operational orbit (200 x 200 + 1500m/s), an explosion or debris-generating MMOD strike is going to strew debris all over a bunch of extremely important orbits.  That's a tolerable risk for the short amount of time needed to do a Starship refueling operation, but then you either need to lower the orbit to VLEO (something like 400 x 400, 1440m/s) or raise the perigee to something like 1000 x 9800 (~150m/s), to get it out of the way of the bulk of MMOD. 

One of these is obviously cheaper than the other, but it doesn't solve the problem of wanting the depot in VLEO for accumulation.  Also, in the event of an explosion, a 1000km perigee means debris forever, while one in VLEO will clear most of itself out in a few years.  But the price for maintaining a depot permanently in VLEO is a non-trivial amount of prop dedicated to stationkeeping.

I continue to think that HEEO refueling is a conops of last resort.  It's a lot easier to make the LSS hold 1500-1600t of prop than to manage shifting depots around in high radiation, with narrow departure windows.

I've been thinking about how SpaceX should manage the SLT flights (post-Option B, reused LSS), and it seems as if the best solution is a second depot in NRHO.  You'd still have a VLEO depot, which handled accumulation, but the last tanker to VLEO, instead of moving its prop into the depot, would extract all prop from it.  It would then boost up to NRHO (whether this is a fast NRHO or a BLT depends on how boiloff gets managed in the tanker) and dump almost all prop into the NRHO depot.  The tanker would then do a direct EDL from TEI (or a reverse BLT).

If you really do have an EDL-capable depot/tanker hybrid, then you don't necessarily need the depot in NRHO.  You just launch the hybrid as the last tanker, fill it up from the depot, send it to NRHO, transfer the prop to the LSS, and return to EDL.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/14/2023 10:13 pm
[...] put a female-female adapter on the ground based QD. Put it on and take it off as necessary. Extra mass stays on the ground. Depot launches should be rare enough that this will not be a serious impediment to ops.

Thanks for sharing this thought. If I'm seeing it correctly, any Starship with a male QD interface acts as a depot, or at least a depot prototype that could be used to demonstrate on-orbit propellant transfer. Count me as "extremely eager" to see a Starship fire up its engines with propellant loaded on-orbit!
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 11/14/2023 10:53 pm
Re: Tanker vs Depot.

I see three classes:

Launch tanker:
[...]
Normal depot:
[...]
Orbital propellant transfer depot/tanker:
Moves propellant from one orbit to significantly different one. Can be any combination of properties of the pure-tanker and the pure-depot. [...] It might be called a "depot" or a "tanker", depending on the specifics of its mission, and the naming is likely to be pretty arbitrary.
Using these orbital prop transfer vehicles to try to define what is a depot or tanker is therefore an exercise in futility.
I think a hybrid tanker/depot only makes sense if it's EDL-capable.  Otherwise, just use a depot. [...]

And it might be called a depot. (Perhaps an "orbital transfer depot".) My point was that trying to define "depot" based on the properties of the hybrid cases is futile. Each example will be its own unique combination of properties.

Better to use the "pure" examples to define what is a "tanker" vs "depot". But also accept that there are a bunch of potential use-cases that will be somewhere in between, and whose naming will be somewhat arbitrary.

If you're going to do HEEO refuelings, then depots need to be [...bunch of stuff about HEEO...]

I wasn't referring to HEEO when talking about the third category. Just that their defining trait will be transferring propellant between orbits. It could be LEO to Lunar. It could be interplanetary. Didn't specify. Just that their properties are going to be fuzzier to define than a "pure" depot or tanker.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Paul451 on 11/14/2023 10:59 pm
The ground based QD is male. Tankers and  other variants, other than depot, are female to allow mating. To mate with other ships the depot must be male. How to do this?
Simplest solution is to build the depot with male QD and put a female-female adapter on the ground based QD. Put it on and take it off as necessary. Extra mass stays on the ground. Depot launches should be rare enough that this will not be a serious impediment to ops.

It's not just the inny and outy bits on the connectors, but the compatibility of all the plumbing and hardware behind those connectors.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 11/15/2023 01:15 am
The ground based QD is male. Tankers and  other variants, other than depot, are female to allow mating. To mate with other ships the depot must be male. How to do this?
Simplest solution is to build the depot with male QD and put a female-female adapter on the ground based QD. Put it on and take it off as necessary. Extra mass stays on the ground. Depot launches should be rare enough that this will not be a serious impediment to ops.

It's not just the inny and outy bits on the connectors, but the compatibility of all the plumbing and hardware behind those connectors.
To be more specific, I think the biggest distinction will be pumps. Neither a tanker nor LSS needs to have any pumps to transfer fuel. (Except to the engines, of course.) A depot can pump fuel out of a tanker or into an LSS (or other Starship variant). Nothing else can do this. (Not the way I'd like to define the word, anyway.) In addition, a depot can have big solar panels, active cooling, and whatever else it needs, but it cannot EDL.

If it doesn't have the pumps, there's no point in putting any of that other stuff on a tanker. This makes a tanker as simple as possible; it's just a version of Starship with big tanks and zero cargo space.

A "hybrid" would be something that capable of EDL which also had pumps to let it exchange fuel with a tanker, an LSS, or even another hybrid. You can do some cool things with hybrids, but they seem to add a lot of engineering and operational complexity.
1) Accumulating prop in a high energy orbit is extremely expensive.  It'll always make more sense to return the depot to VLEO for accumulation, even if the return is propulsive, than it will to have all of the tankers waste a substantial amount of prop getting up to the depot.

2) While it may be OK to deploy a depot operationally to an HEEO, MMOD risk makes it unwise to store it in that orbit.  Note that the depot is simultaneously at risk from MMOD and of becoming MMOD.

Oh, I don't think anyone is talking about doing that! (This is the confusion that comes from calling hybrids "depots.") Let's see if I can describe what I think is meant by "HEEO refueling." I'll over-simplify a bit in the interest of brevity.

1) One hybrid and one LSS are launched into the same circular LEO with just a bit of space between them.

2) Tankers make as many trips as necessary to fuel them both up.

3) When they're both full, they make a single burn at the same time that consumes a bit less than half of their fuel. They are now together in HEEO.

4) Before they reach apogee, the hybrid transfers the rest of its fuel to the LSS, saving just enough for EDL. The LSS is now fully fueled and the hybrid is almost empty.

5) At apogee, the hybrid does a very small burn to lower its perigee enough for EDL at perigee.

6) At its own perigee, the LSS does a second burn that lifts it to the moon.

No one would ever try to refuel an actual depot in HEEO. No one would ever try to put one there in the first place. Or, at least, that the conclusion I drew from the (extensive) discussions upthread. (Pretty much for all the reasons you give.)

It's still pretty complicated, though, and it only works if you have hybrids.

By contrast, a depot in NRHO has a lot going for it. It's a lot easier to keep the propellant cool. (For methalox, passive measures alone may suffice.) And a fully-fueled depot should only have to burn 2/3 of its fuel getting from LEO to NRHO. You can top it up by sending two regular tankers, which fill up at an LEO depot, deliver about 1/3 of their load of fuel to NRHO, and then return to Earth for EDL.

In this scenario, we don't introduce any new vehicle types (beyond the four SpaceX has already talked about), and the genders on the connectors even work out.

If only we could convince NASA to build an NRHO depot instead of Gateway. :-)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: BitterJim on 11/15/2023 01:41 am
Non political thoughts on gender identity.

The ground based QD is male. Tankers and  other variants, other than depot, are female to allow mating. To mate with other ships the depot must be male. How to do this?

Simplest solution is to build the depot with male QD and put a female-female adapter on the ground based QD. Put it on and take it off as necessary. Extra mass stays on the ground. Depot launches should be rare enough that this will not be a serious impediment to ops.

The ground QD retraction mechanism and protective dog house may need a redesign to work with and without the adapter, but for SX that should be a minor issue.

Who not skip the female-female adapter and just give the depot both male and female versions of the QD?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/15/2023 04:33 am
Better to use the "pure" examples to define what is a "tanker" vs "depot". But also accept that there are a bunch of potential use-cases that will be somewhere in between, and whose naming will be somewhat arbitrary.

I think what you're saying is that there will be an a la carte menu of accessories to be added to a common Starship core.  Something like the attached?  (I've called your hybrid "OTV Tanker" below.)

It's not just the inny and outy bits on the connectors, but the compatibility of all the plumbing and hardware behind those connectors.

I've got this subsumed under "Pumping System" in the attachment, but, other than the actual pumps, I'd guess that the existing plumbing, which should handle both fill and drain under gravity or ullage acceleration, should be just about it.  If you can avoid lots of internal plumbing changes, things get a lot easier.

1) One hybrid and one LSS are launched into the same circular LEO with just a bit of space between them.

2) Tankers make as many trips as necessary to fuel them both up.

3) When they're both full, they make a single burn at the same time that consumes a bit less than half of their fuel. They are now together in HEEO.

4) Before they reach apogee, the hybrid transfers the rest of its fuel to the LSS, saving just enough for EDL. The LSS is now fully fueled and the hybrid is almost empty.

This sounds wildly improbable and terrifying to me.  Formation thrusting, even if it weren't dangerous, is unlikely to result in two ships in the same orbit.  There are always unquantifiable residuals.  Those alone make a first-orbit RPOD unlikely, and certainly not something that you'd count on.

A "traveling depot" (one that changes orbits but isn't EDL-capable) is more expensive than an EDL-capable tanker, but not by massive amounts.  Remember that, even if you need 1500-2000m/s of delta-v times two to accomplish the refueling, the second 1500-2000m/s takes a lot less prop than the first one.

Put the depot or tanker in HEEO first and determine its orbital parameters as best as possible.  Then put the target Starship into the HEEO, slightly out of phase, and then phase to do the RPOD.  (I have an HEEO RPOD thread that didn't really catch on somewhere.  I still think HEEO RPOD is hard.)

The other option is to make a "pusher/tanker" depot, which we've discussed elsewhere.  It acts more like a first stage than a formation-flyer, and eliminates all the RPOD scariness.

Quote
By contrast, a depot in NRHO has a lot going for it. It's a lot easier to keep the propellant cool. (For methalox, passive measures alone may suffice.) And a fully-fueled depot should only have to burn 2/3 of its fuel getting from LEO to NRHO. You can top it up by sending two regular tankers, which fill up at an LEO depot, deliver about 1/3 of their load of fuel to NRHO, and then return to Earth for EDL.

Best I can tell, a 1625t tanker (which is what I get if the tanker moves the domes and bulkheads to eat all of the 8m of the cylindrical portion of the payload bay) winds up in NRHO with only a tad more prop than is necessary to transfer to the LSS for NRHO-LS-NRHO, plus what's needed to get back to EDL.  If a tanker can refuel the LSS directly, that's fine.  If not, a depot can mediate the transfer.  But when it's almost exactly one tanker per mission, things seem pretty straightforward.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/15/2023 04:55 am
To be more specific, I think the biggest distinction will be pumps. Neither a tanker nor LSS needs to have any pumps to transfer fuel. (Except to the engines, of course.) A depot can pump fuel out of a tanker or into an LSS (or other Starship variant). Nothing else can do this. (Not the way I'd like to define the word, anyway.) In addition, a depot can have big solar panels, active cooling, and whatever else it needs, but it cannot EDL.
I suspect you are overestimating the complexity of pumps.  They'll need significantly less than 100kg of pump/motors to do the transfer in a couple of hours.*  Putting pumps in the tanker to do the offload will significantly simplify the rest of the plumbing and pushing rather than sucking fluids should make ullage a lot easier.

* They might need more than that in solar cells or batteries, but those can be on the depot.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 11/15/2023 01:40 pm
I suspect you are overestimating the complexity of pumps.
Could be. I'm a software guy, and this is definitely a hardware problem. :-)

They'll need significantly less than 100kg of pump/motors to do the transfer in a couple of hours.*  Putting pumps in the tanker to do the offload will significantly simplify the rest of the plumbing and pushing rather than sucking fluids should make ullage a lot easier.
Are you sure about that? Is pushing that much easier than sucking? (Uh, maybe I should rephrase that . . .) ;-)

They might need more than that in solar cells or batteries, but those can be on the depot.
Ah, but that only works if the other vehicle is actually a depot. If it's some other Starship variant (or even another tanker), then the power has got to come from somewhere. Not a problem if the rule is that tankers (and only tankers) fill depos; depos (and only depos) fill everything else.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 11/15/2023 02:09 pm
Pumps are also an option rather than a necessity: you have a free vacuum-sink all around you, and your propellants boil off with minimal effort (i.e. you have to actively stop your boiloff mitigation for a bit). That means you have both a sink-to-vacuum and source of pressurised gas without any energy input beyond that needed to open vent valves. Since the tanks need vent valves already, that means pressure-based fluid transfer can be performed with effectively no additional hardware or energy input.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/15/2023 02:15 pm
Pumps are also an option rather than a necessity: you have a free vacuum-sink all around you, and your propellants boil off with minimal effort (i.e. you have to actively stop your boiloff mitigation for a bit). That means you have both a sink-to-vacuum and source of pressurised gas without any energy input beyond that needed to open vent valves. Since the tanks need vent valves already, that means pressure-based fluid transfer can be performed with effectively no additional hardware or energy input.
This also means that all SS types are "equipped" to supply pressure or vacuum to move the fluids in either direction. I still think all transfers will either be to the Depot or from the Depot for other reasons, though.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 11/15/2023 04:17 pm
Pumps are also an option rather than a necessity: you have a free vacuum-sink all around you, and your propellants boil off with minimal effort (i.e. you have to actively stop your boiloff mitigation for a bit). That means you have both a sink-to-vacuum and source of pressurized gas without any energy input beyond that needed to open vent valves. Since the tanks need vent valves already, that means pressure-based fluid transfer can be performed with effectively no additional hardware or energy input.
Oh, I like that! Since the ullage acceleration is so small, you might not even need to apply heat to the source side; simply venting the gas on the destination side could be enough to just suck all the propellant through it. I wonder how much propellant you'd lose to evaporation in the process though.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/15/2023 04:49 pm
They'll need significantly less than 100kg of pump/motors to do the transfer in a couple of hours.*  Putting pumps in the tanker to do the offload will significantly simplify the rest of the plumbing and pushing rather than sucking fluids should make ullage a lot easier.
Are you sure about that? Is pushing that much easier than sucking? (Uh, maybe I should rephrase that . . .) ;-)
Blowing is easier than sucking?  Maybe some other rephrase ... .

Look at the physics of sucking.  To suck you lower the pressure at one end and let the pressure at the other end do the work.  In this case it means the ullage is doing the work between the source and the pump.  Controlling the ullage will be a lot more difficult than controlling a pump.  Modern pumps with variable speed controllers are wonderfully controllable, quite unlike every other part the hydraulic engineers throw at a problem.  I think hydraulic engineers hate modern pump because you don't have to sacrifice a virgin to make them work properly.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/15/2023 04:51 pm
Pumps are also an option rather than a necessity: you have a free vacuum-sink all around you, and your propellants boil off with minimal effort (i.e. you have to actively stop your boiloff mitigation for a bit). That means you have both a sink-to-vacuum and source of pressurized gas without any energy input beyond that needed to open vent valves. Since the tanks need vent valves already, that means pressure-based fluid transfer can be performed with effectively no additional hardware or energy input.
Oh, I like that! Since the ullage acceleration is so small, you might not even need to apply heat to the source side; simply venting the gas on the destination side could be enough to just suck all the propellant through it. I wonder how much propellant you'd lose to evaporation in the process though.
At least an order of magnitude more than the mass of a pump.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/15/2023 04:53 pm
They'll need significantly less than 100kg of pump/motors to do the transfer in a couple of hours.*  Putting pumps in the tanker to do the offload will significantly simplify the rest of the plumbing and pushing rather than sucking fluids should make ullage a lot easier.
Are you sure about that? Is pushing that much easier than sucking? (Uh, maybe I should rephrase that . . .) ;-)
Blowing is easier than sucking?  Maybe some other rephrase ... .

Look at the physics of sucking.  To suck you lower the pressure at one end and let the pressure at the other end do the work.  In this case it means the ullage is doing the work between the source and the pump.  Controlling the ullage will be a lot more difficult than controlling a pump.  Modern pumps with variable speed controllers are wonderfully controllable, quite unlike every other part the hydraulic engineers throw at a problem.  I think hydraulic engineers hate modern pump because you don't have to sacrifice a virgin to make them work properly.
You control the flow with high precision by using a variable valve, not by attempting precise control of the ullage pressure.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/15/2023 05:04 pm
They'll need significantly less than 100kg of pump/motors to do the transfer in a couple of hours.*  Putting pumps in the tanker to do the offload will significantly simplify the rest of the plumbing and pushing rather than sucking fluids should make ullage a lot easier.
Are you sure about that? Is pushing that much easier than sucking? (Uh, maybe I should rephrase that . . .) ;-)
Blowing is easier than sucking?  Maybe some other rephrase ... .

Look at the physics of sucking.  To suck you lower the pressure at one end and let the pressure at the other end do the work.  In this case it means the ullage is doing the work between the source and the pump.  Controlling the ullage will be a lot more difficult than controlling a pump.  Modern pumps with variable speed controllers are wonderfully controllable, quite unlike every other part the hydraulic engineers throw at a problem.  I think hydraulic engineers hate modern pump because you don't have to sacrifice a virgin to make them work properly.
You control the flow with high precision by using a variable valve, not by attempting precise control of the ullage pressure.
Which requires headroom in the ullage pressure, so higher ullage pressure than otherwise needed.


Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 11/15/2023 05:16 pm
Pumps are also an option rather than a necessity: you have a free vacuum-sink all around you, and your propellants boil off with minimal effort (i.e. you have to actively stop your boiloff mitigation for a bit). That means you have both a sink-to-vacuum and source of pressurized gas without any energy input beyond that needed to open vent valves. Since the tanks need vent valves already, that means pressure-based fluid transfer can be performed with effectively no additional hardware or energy input.
Oh, I like that! Since the ullage acceleration is so small, you might not even need to apply heat to the source side; simply venting the gas on the destination side could be enough to just suck all the propellant through it. I wonder how much propellant you'd lose to evaporation in the process though.
At least an order of magnitude more than the mass of a pump.
An empty LCH4 (~600 m^3) tank at 1 bar pressure has ~395 kg ullage mass. You could in theory store props for long term coasting at as close as possible to their vapor pressure (~0.12 Bar) and drop that to closer to 40kg. And that's if you vent down to vacuum every time: as long as there is some pressure differential between the sending and receiving ullage, then there will still be fluid flow. You can start by not venting below flight pressure in the tanks of the delivery tanker (6 bar) and will not even need to vent any depot ullage for the first few transfer ops, and even then you can get away with a slow-bleed to keep receiver pressure below sender pressure to minimise losses (or still vent down, you can trade efficiency for speed here).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/15/2023 05:37 pm
as long as there is some pressure differential between the sending and receiving ullage, then there will still be fluid flow.
Some pressure differential means some fluid flow for some value of some.  A small pressure differential means a small flow. The alternative of using a small pump still saves mass and increases flexibility and control.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/15/2023 06:21 pm
They'll need significantly less than 100kg of pump/motors to do the transfer in a couple of hours.*  Putting pumps in the tanker to do the offload will significantly simplify the rest of the plumbing and pushing rather than sucking fluids should make ullage a lot easier.
Are you sure about that? Is pushing that much easier than sucking? (Uh, maybe I should rephrase that . . .) ;-)
Blowing is easier than sucking?  Maybe some other rephrase ... .

Look at the physics of sucking.  To suck you lower the pressure at one end and let the pressure at the other end do the work.  In this case it means the ullage is doing the work between the source and the pump.  Controlling the ullage will be a lot more difficult than controlling a pump.  Modern pumps with variable speed controllers are wonderfully controllable, quite unlike every other part the hydraulic engineers throw at a problem.  I think hydraulic engineers hate modern pump because you don't have to sacrifice a virgin to make them work properly.
You control the flow with high precision by using a variable valve, not by attempting precise control of the ullage pressure.
Which requires headroom in the ullage pressure, so higher ullage pressure than otherwise needed.
Yep. Coarse control by controlling ullage pressure with a bit of overpressure, and precision control using the valve. I'm fairly agnostic on the pump-versus-valve issue. Let the actual engineers make the choice.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/15/2023 07:41 pm
Yep. Coarse control by controlling ullage pressure with a bit of overpressure, and precision control using the valve. I'm fairly agnostic on the pump-versus-valve issue. Let the actual engineers make the choice.

Team Pump and Team Pressure have extensively litigated this issue way, way up-thread (I'm on Team Pump), but one thing that we didn't think about a lot is whether the existing QD and the existing plumbing had everything needed to implement pump-fed transfer.

The pump itself is tiny (less than a couple 100W) for each of the two tanks, but that assumes that it's easy to connect the ullage space between the source and destination tanks.  Without that connection, ullage pressure will fall in the source and build up in the destination, creating back pressure.

There are O2 and CH4 prepress lines on the QD, but they may be one-way valves.  If you can tie the prepress lines for both the source and destination together, then you likely have all the plumbing you need.  But if something prevents that, there's work that needs to be done to get pump-fed transfer working.  That work might be easy, but it still keeps Team Pressure in the game.

Venting is such sweet sorrow.

PS:  I've always assumed that the pumps themselves were part of the depot's gender-bender package, as would be vent lines that fed to any cryocoolers for the depot.  That also kinda seems to require being able to turn the prepress plumbing into a vent.  Here's a block diagram of a "depot kit", which turns any vanilla Starship or variant into a depot-capable vehicle.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 11/15/2023 07:47 pm
At least, it seems you don't need multiple redundant pumps if you're not a depot, given a backup option for pressure control during transfer - albeit a rather inconvenient one
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 11/15/2023 08:16 pm
as long as there is some pressure differential between the sending and receiving ullage, then there will still be fluid flow.
Some pressure differential means some fluid flow for some value of some.  A small pressure differential means a small flow. The alternative of using a small pump still saves mass and increases flexibility and control.
A small pump. Plus the power source for the pump. Plus the extra plumbing the tie the pump into the main prop feed lines. Plus the extra prop used to chill down the mass of the pump hardware (in addition to chilling the feed lines). Etc.
That's a lot of extra components, systems, and mass, to save a less than 1% of tanker's propellant load.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 11/15/2023 08:26 pm
A small pump. Plus the power source for the pump. Plus the extra plumbing the tie the pump into the main prop feed lines. Plus the extra prop used to chill down the mass of the pump hardware (in addition to chilling the feed lines). Etc.
That's a lot of extra components, systems, and mass, to save a less than 1% of tanker's propellant load.
How are you sure it really is less than 1% though?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/16/2023 05:08 pm
Coarse control by controlling ullage pressure with a bit of overpressure, and precision control using the valve.

Any indication such precision is needed?

Obviously the flow rate can't vary by orders-of-magnitude, but ±20% shouldn't make any difference. By eliminating flow rate valves you not only delete mass and complexity, but you also eliminate a flow restriction which reduces pump mass and power (or alternately reduces the total vented ullage mass required to transfer propellant).

"The best part isn't."
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/16/2023 05:52 pm
Coarse control by controlling ullage pressure with a bit of overpressure, and precision control using the valve.

Any indication such precision is needed?

Obviously the flow rate can't vary by orders-of-magnitude, but ±20% shouldn't make any difference. By eliminating flow rate valves you not only delete mass and complexity, but you also eliminate a flow restriction which reduces pump mass and power (or alternately reduces the total vented ullage mass required to transfer propellant).

"The best part isn't."

Often the place you need fine control is during startup or shutdown to avoid things like water hammer and resonances, temperature effects and other things you did not even think of putting into the simulation before it failed on the test stand.

You can view this as navigating a state space with odd walls and landmines. The more control the better.  It's even more fun if you are required deal with failures, so that not only do you need a single safe path, but also safe alternates if some component fails at the wrong time.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/16/2023 06:10 pm
Coarse control by controlling ullage pressure with a bit of overpressure, and precision control using the valve.

Any indication such precision is needed?

Obviously the flow rate can't vary by orders-of-magnitude, but ±20% shouldn't make any difference. By eliminating flow rate valves you not only delete mass and complexity, but you also eliminate a flow restriction which reduces pump mass and power (or alternately reduces the total vented ullage mass required to transfer propellant).

"The best part isn't."

Often the place you need fine control is during startup or shutdown to avoid things like water hammer and resonances

If the startup/shutdown is caused by a pump (gradually) spooling up and down rather than a valve opening and closing, there's no source for water hammer.

For valves they usually solve this with a slow-opening/slow-closing valve.


temperature effects and other things you did not even think of putting into the simulation before it failed on the test stand.

You can view this as navigating a state space with odd walls and landmines. The more control the better.

Avoiding this sort of hyper-preemptive design work is exactly what distinguishes SpaceX from OldSpace.

If your logic is "why not add it?", you're going to add a lot of stuff you don't actually need.

This is why Elon specifically says "if you're not sometimes adding things back, you're not deleting enough." This piece of advice is simultaneously non-obvious (see this exchange) yet important.

It's even more fun if you are required deal with failures, so that not only do you need a single safe path, but also safe alternates if some component fails at the wrong time.

Business as usual, sure. I think we're all assuming this.  That's just the price of admission!
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/16/2023 06:44 pm
Coarse control by controlling ullage pressure with a bit of overpressure, and precision control using the valve.

Any indication such precision is needed?

Obviously the flow rate can't vary by orders-of-magnitude, but ±20% shouldn't make any difference. By eliminating flow rate valves you not only delete mass and complexity, but you also eliminate a flow restriction which reduces pump mass and power (or alternately reduces the total vented ullage mass required to transfer propellant).

"The best part isn't."
As I said, I'm agnostic. I replied to a message about precision by mentioning controllable valves. The idea here is to control a valve that must already exist in the system, so not much added complexity and no additional flow restriction. There must be one since the flow is between two ships, so you get the precision (if needed) by replacing an on/off valve with a controllable valve. Sure, leave that sucker all the way open when you don't need precision and just use the ullage pressure.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 11/16/2023 06:57 pm
Another reason to want to control the flow rate is that the ullage burn is very low-gravity, so there's apt to be a lot of sloshing.

Really clever would be to use the vented ullage gas to fuel the special engines used for the long ullage burn. I'm only half-serious, since, assuming those special engines are only on the depot, it might be too complicated make that work both for filling the depot (when you use the depot's ullage gas) and draining it (when you need to use the Starship's ullage gas).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/16/2023 08:00 pm
Coarse control by controlling ullage pressure with a bit of overpressure, and precision control using the valve.

Any indication such precision is needed?

Obviously the flow rate can't vary by orders-of-magnitude, but ±20% shouldn't make any difference. By eliminating flow rate valves you not only delete mass and complexity, but you also eliminate a flow restriction which reduces pump mass and power (or alternately reduces the total vented ullage mass required to transfer propellant).

"The best part isn't."
As I said, I'm agnostic.

Ditto.

I replied to a message about precision by mentioning controllable valves. The idea here is to control a valve that must already exist in the system, so not much added complexity and no additional flow restriction.

If there's never any additional flow restriction, then it won't effect the precision.  ;)  If there is, we can't claim "no additional flow restriction."

I was talking about flow regulating valves (hence my word choice: "flow rate valves"), not shutoff valves. There is a need for shutoff valves. There is not (AFAICT) a need for flow regulating valves.


There must be one since the flow is between two ships, so you get the precision (if needed) by replacing an on/off valve with a controllable valve. Sure, leave that sucker all the way open when you don't need precision and just use the ullage pressure.

It sounds like we're in agreement, then, that there isn't necessarily a need (let alone desirability) to adding intentional flow restriction to the system.

If included, such hardware would be for emergency measures only. If you're getting any substantial pressure drop across the valve while in operation, something has gone very wrong!  :o

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/16/2023 08:21 pm
Another reason to want to control the flow rate is that the ullage burn is very low-gravity, so there's apt to be a lot of sloshing.

Hence my "±20%."

Ultra-fine precision isn't really needed here, especially when it negatively effects the dry mass and pumping losses

Really clever would be to use the vented ullage gas to fuel the special engines used for the long ullage burn. I'm only half-serious, since, assuming those special engines are only on the depot, it might be too complicated make that work both for filling the depot (when you use the depot's ullage gas) and draining it (when you need to use the Starship's ullage gas).

Yeah, the big advantage to ullage venting over pumping is to eliminate the gaseous return pipe and interconnect. This sacrifices that advantage.

One could always take the L and only burn the ullage mass when transferring from tanker to depot. Half is still better than none.  :)

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 11/16/2023 08:36 pm
If you have a dedicated pump you can probably get away with just letting the ullage regulate itself passively.

Venting the receiver tank below vapor pressure will be detrimental anyway because  it will just boil until it reaches the boiling point which might be wasting more propellant than you really need to.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/16/2023 08:52 pm
What eriblo said. If you want to add a flow restriction ("resistor") somewhere in the flow path to regulate the flow rate, it's far better to restrict it at the gaseous vent.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/16/2023 09:50 pm
What eriblo said. If you want to add a flow restriction ("resistor") somewhere in the flow path to regulate the flow rate, it's far better to restrict it at the gaseous vent.

Between the flow of liquid and the gaseous vent is a large tank partially full of gas.  In effect a hydraulic accumulator.
 This can add a huge integral term to the control loop.   This doesn't necessarily make it uncontrollable, but it does have potential for adding oscillations that make control interesting.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/16/2023 10:49 pm
Thanks, good observation.

I guess I just intuitively expected that the (unavoidable) pipe losses are sufficient to passively damp oscillations, so long as the vent maintains a stable ullage pressure delta (and thus motive force). SpaceX tends to prefer simple bang-bang closed loop ullage pressure control.



Yet another reason why I prefer pumps.  :D
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/16/2023 11:07 pm
On Earth, everyone just uses the pressure differential to move the cryogens.

People use pumps for non-cryogens because non-cryogens are not self-pressurizing. Cryogens make pressure based transfer super easy, actually easier than for non-cryogens which would need a separate pressure system to do that, so non-cryogens just use gravity or pumps.

Cryogenic pumps are also kind of annoying. Cryogenic seals are annoying. Cryogenic lubricants are annoying.

Sure you could use pumps, but I seriously doubt they will.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 11/17/2023 12:37 am
If you fill a sealed container with a single fluid the distribution of liquid and gas and the pressure is only dependent on the temperature (at thermodynamic equilibrium the pressure is the vapor pressure at that temperature).

In microgravity you will not have a pressure gradient - while you can have temperature gradients supported by heat flux the liquid can not be hotter/the gas can not be colder than the boiling point.

If you add liquid it might initially increase the pressure as the gas compresses and heats up (together with the gas/liquid interface) but it will eventually decrease down to the vapor pressure of the new average temperature.

In theory the only thing you need to do to transfer liquid from one tank to the other is to maintain a temperature gradient (or start with one large enough to finish the desired transfer). The reason to add a pump if you already have temperature control would be if you need to speed up the process or have finer control.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/17/2023 12:55 am
Also, as far as venting, you can use venting for ullage thrust.

And ullage thrust can be pro-grade, ideally close to perigee.

Works even as cold gas thrusters (low Isp, like 100s methane and 70s oxygen), but you could also use the vented gas in hot gas thrusters... in which case a pressure boost pump would probably make sense at least eventually (although some hot gas thrusters exist that can work at the ~6 bar ullage pressure of Starship), which is like 300s Isp in vacuum.

So all that venting and ullage thrust could be done in a way that has almost no loss overall, as you'd be using the boiloff/vent/ullage gas as thrust to raise your apogee, nearly as efficient as using Raptor. At least, once you get really good at it. I'm sure at first, you'll be bad at it so there will be more losses than you'd prefer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/17/2023 04:20 am
On Earth, everyone just uses the pressure differential to move the cryogens.

People use pumps for non-cryogens because non-cryogens are not self-pressurizing. Cryogens make pressure based transfer super easy, actually easier than for non-cryogens which would need a separate pressure system to do that, so non-cryogens just use gravity or pumps.

Cryogenic pumps are also kind of annoying. Cryogenic seals are annoying. Cryogenic lubricants are annoying.

Sure you could use pumps, but I seriously doubt they will.

Yet another case where Analogy Over Physics could be more misleading than enlightening.

1. On Earth, you can just inexpensively recondense more cryogen out of the atmosphere. Venting gas is basically free on Earth, not so much in space. In the few cases where venting is not "free" on Earth (eg MRI helium), they often use... cryogenic pumps.  :)

2. If you prefer not to pump cryogenic liquids, you can always pump gas instead.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/17/2023 04:23 am
On Earth, everyone just uses the pressure differential to move the cryogens.

People use pumps for non-cryogens because non-cryogens are not self-pressurizing. Cryogens make pressure based transfer super easy, actually easier than for non-cryogens which would need a separate pressure system to do that, so non-cryogens just use gravity or pumps.

Cryogenic pumps are also kind of annoying. Cryogenic seals are annoying. Cryogenic lubricants are annoying.

Sure you could use pumps, but I seriously doubt they will.

Yet another case where Analogy Over Physics could be more misleading than enlightening.

1. On Earth, you can just inexpensively recondense more cryogenic out of the atmosphere. Venting gas is basically free on Earth, not so much in space. In the few cases where venting is not "free" on Earth (eg MRI helium), they often use... cryogenic pumps.  :)

2. If you prefer not to pump cryogenic liquids, you can always pump gas instead.

And before somebody asks, no, using ullage for reboost doesn't suddenly mean it's free. Actual reboost needs are much smaller than ullage mass would provide, so venting is still mostly a waste of upmass.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/17/2023 04:55 am
On Earth, everyone just uses the pressure differential to move the cryogens.

People use pumps for non-cryogens because non-cryogens are not self-pressurizing. Cryogens make pressure based transfer super easy, actually easier than for non-cryogens which would need a separate pressure system to do that, so non-cryogens just use gravity or pumps.

Cryogenic pumps are also kind of annoying. Cryogenic seals are annoying. Cryogenic lubricants are annoying.

Sure you could use pumps, but I seriously doubt they will.

Yet another case where Analogy Over Physics could be more misleading than enlightening.

1. On Earth, you can just inexpensively recondense more cryogenic out of the atmosphere. Venting gas is basically free on Earth, not so much in space. In the few cases where venting is not "free" on Earth (eg MRI helium), they often use... cryogenic pumps.  :)

2. If you prefer not to pump cryogenic liquids, you can always pump gas instead.

And before somebody asks, no, using ullage for reboost doesn't suddenly mean it's free. Actual reboost needs are much smaller than ullage mass would provide, so venting is still mostly a waste of upmass.
No one asked.

The higher your apogee, the more useful your propellant is.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/17/2023 05:26 am
And they rarely use cryogenic pumps on Earth for liquid helium. I’ve never seen it used, and they use liquid helium at my work.

I’m speaking from experience, not aNaLoGy.

And the physics also says you only need a slight pressure difference to transfer fluid.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/17/2023 05:47 am

The higher your apogee, the more useful your propellant is.

Much of the incidental boost does not go into the propellent.  It goes into the dry mass of the tanker.  Worse it requires all later tankers to attain a higher orbit.

Once the accumulator is in a significantly non-circular orbit it complicates future operations.  Rendezvous and fuel transfers need to be timed to the eccentric orbit.  Making sure the apogee is in the right direction adds further operational constraints.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 11/17/2023 09:07 am
Also, as far as venting, you can use venting for ullage thrust.

And ullage thrust can be pro-grade, ideally close to perigee.

Works even as cold gas thrusters (low Isp, like 100s methane and 70s oxygen), but you could also use the vented gas in hot gas thrusters... in which case a pressure boost pump would probably make sense at least eventually (although some hot gas thrusters exist that can work at the ~6 bar ullage pressure of Starship), which is like 300s Isp in vacuum.

So all that venting and ullage thrust could be done in a way that has almost no loss overall, as you'd be using the boiloff/vent/ullage gas as thrust to raise your apogee, nearly as efficient as using Raptor. At least, once you get really good at it. I'm sure at first, you'll be bad at it so there will be more losses than you'd prefer.
Using ullage vent for settling thrust is attractive in its elegance, but I think is a result of "spend a million to save a gram" thinking that is the norm for spacecraft so far, rather than optimising for cost and reduced complexity.
The problem is that using ullage vent for settling then couples propellant settling with propellant transfer, which introduces system complexity and adds feedback loops (e.g. tackling any slosh by varying transfer rate means varying venting, which then varies settling thrust, which then influences slosh...). If instead the vent is a regular non-propulsive setup (as is the norm for spacecraft vents) and dedicated ullage motors are used, then gas can be vented for ullage management at any time independently of applying settling thrust, and settling thrust can be applies when needed without the need to vent ullage. For long-duration settled coasts (e.g. if you are doing 'rapid' prop transfers to a depot over a few days) - where you want to avoid settling, then returning to microgravity and allowing propellants to slosh, then having to re-settle - more efficient biprop settling thrust also avoids having to boil off propellant just to vent it for cold-gas thrust, and you likely require biprop RCS anyway for on-orbit manoeuvres outside the minimum impulse bit of a Raptor.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/17/2023 10:19 am
Also, as far as venting, you can use venting for ullage thrust.

And ullage thrust can be pro-grade, ideally close to perigee.

Works even as cold gas thrusters (low Isp, like 100s methane and 70s oxygen), but you could also use the vented gas in hot gas thrusters... in which case a pressure boost pump would probably make sense at least eventually (although some hot gas thrusters exist that can work at the ~6 bar ullage pressure of Starship), which is like 300s Isp in vacuum.

So all that venting and ullage thrust could be done in a way that has almost no loss overall, as you'd be using the boiloff/vent/ullage gas as thrust to raise your apogee, nearly as efficient as using Raptor. At least, once you get really good at it. I'm sure at first, you'll be bad at it so there will be more losses than you'd prefer.
Using ullage vent for settling thrust is attractive in its elegance, but I think is a result of "spend a million to save a gram" thinking that is the norm for spacecraft so far, rather than optimising for cost and reduced complexity.
The problem is that using ullage vent for settling then couples propellant settling with propellant transfer, which introduces system complexity and adds feedback loops (e.g. tackling any slosh by varying transfer rate means varying venting, which then varies settling thrust, which then influences slosh...). If instead the vent is a regular non-propulsive setup (as is the norm for spacecraft vents) and dedicated ullage motors are used, then gas can be vented for ullage management at any time independently of applying settling thrust, and settling thrust can be applies when needed without the need to vent ullage. For long-duration settled coasts (e.g. if you are doing 'rapid' prop transfers to a depot over a few days) - where you want to avoid settling, then returning to microgravity and allowing propellants to slosh, then having to re-settle - more efficient biprop settling thrust also avoids having to boil off propellant just to vent it for cold-gas thrust, and you likely require biprop RCS anyway for on-orbit manoeuvres outside the minimum impulse bit of a Raptor.

I think we worked this out way up-thread and determined that cold-gas ullage venting wasn't even close to being able to provide the needed impulse for settling.  You just leave too much enthalpy on the table without a combusting gas thruster.  It doesn't have to be a big thruster (indeed, it needs to be pretty small), but it has to be able to generate more than a few tens of seconds of Isp.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/17/2023 01:03 pm

The higher your apogee, the more useful your propellant is.

Much of the incidental boost does not go into the propellent.  It goes into the dry mass of the tanker.  Worse it requires all later tankers to attain a higher orbit.

Once the accumulator is in a significantly non-circular orbit it complicates future operations.  Rendezvous and fuel transfers need to be timed to the eccentric orbit.  Making sure the apogee is in the right direction adds further operational constraints.


So all that venting and ullage thrust could be done in a way that has almost no loss overall, as you'd be using the boiloff/vent/ullage gas as thrust to raise your apogee, nearly as efficient as using Raptor. At least, once you get really good at it. I'm sure at first, you'll be bad at it so there will be more losses than you'd prefer.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 11/17/2023 01:17 pm
Is everyone talking about venting ullage assuming that the receiving tank always operates right at the upper pressure limit? Because otherwise you can just wait a little to let the ullage cool down/condense like it will naturally want to do.

I.e. venting with active pumping assumes that you either load fast enough that the pressure temporarily exceeds the maximum or that the final average temperature in the tank (also accounting for the work done to pump the fluid) will exceed the boiling point at the maximum pressure (and that there is no other way to cool the contents).

With passive pumping the same applies using the source tank pressure instead of the maximum rated pressure of the tank or the pump.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/17/2023 01:26 pm
Let’s put some harder numbers on here for how much acceleration is needed for settling.

Lockheed/ULA claims about 0.0001m/s^2 acceleration is sufficient. And they’ve demonstrated 0.001m/s^2 (10^-4 gee) already, which they say is already good enough.

https://www.ulalaunch.com/docs/default-source/extended-duration/settled-cryogenic-propellant-transfer-2006-4436.pdf

This other paper showed mass transfer rates for vented transfer was up to 16t/hour of liquid hydrogen with a 10cm diameter pipe.

Methalox has a bulk density of 10 times that of liquid hydrogen, so about 160 tonnes per hour would be doable for 10cm diameter pipes.

To transfer 1200t would require 7.5hours (or just use larger pipes), and with 1600tonnes mass roughly, that would require about 1600kg/hr of settling propellant for thrust at 10^-4 gees. So about 1% the mass of the propellant transferred. Rough figures, of course. There may need to be intermittent filling which might require more settling propellant. And at the same time, the acceleration could be even lower, like 10^-5 gees, requiring even less propellant.

But probably this amount of propellant will already be available due to chill down of the transfer pipes boiling off some of the propellant.

https://ntrs.nasa.gov/api/citations/20080032611/downloads/20080032611.pdf
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 11/17/2023 01:42 pm
I think we worked this out way up-thread and determined that cold-gas ullage venting wasn't even close to being able to provide the needed impulse for settling.

What's the determined requirement?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 11/17/2023 01:47 pm
I think we worked this out way up-thread and determined that cold-gas ullage venting wasn't even close to being able to provide the needed impulse for settling.

What's the determined requirement?
i worked it out in the post immediately above.

Cold gas actually WOULD be enough if you can do the more challenging (yet still demonstrated) 10^-5 gee settling acceleration. 10^-4gee thrust probably needs hot gas to keep settling thrust to just 1% of transferred mass, but the factor of 4-5 lower Isp of cold gas can be compensated by using a factor of 10 lower settling acceleration.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/17/2023 02:24 pm

The higher your apogee, the more useful your propellant is.

Much of the incidental boost does not go into the propellent.  It goes into the dry mass of the tanker.  Worse it requires all later tankers to attain a higher orbit.

Once the accumulator is in a significantly non-circular orbit it complicates future operations.  Rendezvous and fuel transfers need to be timed to the eccentric orbit.  Making sure the apogee is in the right direction adds further operational constraints.


So all that venting and ullage thrust could be done in a way that has almost no loss overall, as you'd be using the boiloff/vent/ullage gas as thrust to raise your apogee, nearly as efficient as using Raptor. At least, once you get really good at it. I'm sure at first, you'll be bad at it so there will be more losses than you'd prefer.
Being good at something does not override orbital mechanics.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 11/17/2023 02:55 pm
I think we worked this out way up-thread and determined that cold-gas ullage venting wasn't even close to being able to provide the needed impulse for settling.

What's the determined requirement?

i worked it out in the post immediately above.

Cold gas actually WOULD be enough if you can do the more challenging (yet still demonstrated) 10^-5 gee settling acceleration.

An electrical boom (https://forum.nasaspaceflight.com/index.php?topic=45674.msg1820996#msg1820996) could deliver that modest acceleration most efficiently, and continuously.

You'd need attitude control under acceleration, but that also can be managed electrically with magnetorquers, as we saw on the Eu:CROPIS mission (https://forum.nasaspaceflight.com/index.php?topic=34036.msg2537511#msg2537511) and elsewhere.

Just for efficiency.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/17/2023 03:03 pm

Cold gas actually WOULD be enough

But why?  Is it really so hard to develop a 1600N rocket?  The hypothetical thing could weigh as much as a Raptor and it would still be a win.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/17/2023 08:58 pm

Cold gas actually WOULD be enough

But why?  Is it really so hard to develop a 1600N rocket?  The hypothetical thing could weigh as much as a Raptor and it would still be a win.

Worst case would be a 110t dry + 1625t prop depot transferring prop to a 120t dry + 150t payload target = 2005t system.  1E-5G = 1E-4m/s² yields an ullage thrust of 200N.  So far, so good.

Next problem:  For how long?  An hour?  A day?  I'd kinda guess that no matter what transfer scheme you chose, you could count on transferring  100kg/s.  Worst case would be to fill a totally dry target with 1200t of prop, which would take 3.3hours (12,000s).

What's the specific impulse for a very low-pressure chamber?  (I'll assume a 1bar injector drop, so the tanks need to be at 4bar.)  For methane at 3bar, I get 90s (many guesses involved here), which is vexit=875m/s.

So: T = 200N = 875m/s * mdot, which makes mdot = .229kg/s.  Over 12,000s, that's  2.75t of ullage.  That's not terrible at first blush.

But wait!  There's more!  Both ships need to provide some of the ullage thrust, in proportion to their current mass, to keep things from rotating.  And we're only using methane, not oxygen.

The depot is much heavier than the target ship to start with, but the target is heavier at the end of the transfer.  However the mass of the ship is inversely proportional to the available ullage mass, so the ship with the highest mass flow requirements is the one with the smallest supply of ullage gas.

A completely full depot effectively has zero ullage gas.  But a nearly empty target with 1200t tanks at 4bar has about 3.4t of ullage GCH4, which is adequate.

If the ullage spaces are tied together (i.e., if you're using a pump-fed instead of pressure-fed transfer), balancing the thrust requirements is easy, although you likely need a bit more gas to keep the ullage spaces at reasonable pressure.  So at the very least, you need plumbing to heat some LCH4 to backfill the gas you're using for thrust.

Note that things are much, much worse for a pressure-fed transfer system.  Now you need to be boiling LCH4 at a pretty hefty clip to backfill not only your cold-gas thrusters at their peak throttle, but also to keep the pressure differential up as the ullage space expands (while the liquid space shrinks).

It's a lot easier just to run some LCH4 into a COPV, seal the COPV, and heat the LCH4 to supercritical pressures.  But that requires more plumbing.

I assert you need that plumbing anyway, for the waist landing thrusters.  Those need to be plumbed for burning methox.  And if you're going to do that, you might as well plumb the whole ship the same way.  I doubt it's more than an extra tonne of dry mass, which is probably less than what you'll save with a ~200ish Isp thruster, which reduces your ullage prop from 2.75t to 1.23t.

This discussion always comes back to the thruster architecture, which is gonna be really freakin' complex for a vehicle the size of Starship.  You need 80-100kN thrusters for the waist landing thruster for the LSS, you need thrusters capable of (at a guess) 5Ns minimum impulse bits for RCS (that's pretty big, for a big vehicle with big moments of inertia), and you need these long-duration, tiny-thrust ullage verniers.  I'm willing to believe that the RCS and vernier thrusters can be the same, especially if they're both methox, but the need for waist thrusters militates strongly for the COPVs needed to heat and manage both LCH4 and LOX.  If you have to develop it for one thing, you might as well use it for everything.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/17/2023 10:03 pm
But wait!  There's more!  Both ships need to provide some of the ullage thrust, in proportion to their current mass, to keep things from rotating. 
The other way to do this is to tie the ships together and have one (probably the depot) thrust through the combined center of mass.   The forces  are small, so the docking structure does not need to be robust.  A dozen rolls of duck tape would be strong enough (though difficult to deploy)

Mind you, the ships still have a lot of intertia.  You have to plan ahead and act gently.  But that should be no more difficult than flying the ships in formation without docking. Doable as long as you don't let humans near the controls.


This discussion always comes back to the thruster architecture, which is gonna be really freakin' complex for a vehicle the size of Starship.  You need 80-100kN thrusters for the waist landing thruster for the LSS, you need thrusters capable of (at a guess) 5Ns minimum impulse bits for RCS (that's pretty big, for a big vehicle with big moments of inertia), and you need these long-duration, tiny-thrust ullage verniers.  I'm willing to believe that the RCS and vernier thrusters can be the same, especially if they're both methox, but the need for waist thrusters militates strongly for the COPVs needed to heat and manage both LCH4 and LOX.  If you have to develop it for one thing, you might as well use it for everything.
AIUI the LSS landing requirements are unique to LSS.  Don't let it poison the architecture.  If need be kludge on some solids or hypergolics, take NASAs money, learn what you can, and on to Mars.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/18/2023 02:59 am
Quote from: Marcia Smith, @SpcPlcyOnline
Hawkins: SpX will use both Starbase and LC39A launch pads for Artemis III HLS. In the "high teens" of number of launches needed to aggregate fuel. Lander is about half way thru that sequence. Must do all in quick succession due to boil off issues.

I'm confused: what rendezvous orbit is accessible from both Starbase and LC-39A? And why is HLS itself launched halfway through the propellant launch campaign?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 11/18/2023 03:57 am
I'm confused: what rendezvous orbit is accessible from both Starbase and LC-39A? And why is HLS itself launched halfway through the propellant launch campaign?

[IANAE, but AIUI...]
Starbase is a bit limited due to having Caribbean needles to thread, but that inclination (26.x⁰ ?) is not far off KSC's latitude (28.6⁰ N). With a dog-leg left out of KSC you could launch into the same plane. Maybe Starbase can also launch on a more northerly bearing and dog-leg right to thread the needle? which would reduce the cost of the KSC dog-leg.

With the current size of Starship, the total amount of fuel you need to accumulate is more than the depot can hold. For the last few tanker launches, you need a 2nd tank in orbit... and presumably it makes more sense to use HLS for that instead of a standard tanker or 2nd depot.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/18/2023 04:54 am
Quote from: Marcia Smith, @SpcPlcyOnline
Hawkins: SpX will use both Starbase and LC39A launch pads for Artemis III HLS. In the "high teens" of number of launches needed to aggregate fuel. Lander is about half way thru that sequence. Must do all in quick succession due to boil off issues.

I'm confused: what rendezvous orbit is accessible from both Starbase and LC-39A? And why is HLS itself launched halfway through the propellant launch campaign?

Any orbit with an inclination greater than 28.5º 28.6º is accessible from both BC and the Cape.  As for why HLS is launched halfway through, beats me.  The only thing I can think of is if they need to top off the LSS while it's in NRHO.  That would be something to do closer to the SLS launch, and would indeed have the LSS launching in the middle of the fueling campaign.

BTW, according to Eric Berger, Lisa Watson-Morgan is citing "high single digits" of tanker launches, which jibes with what I get (nine) if each tanker can take 150t to LEO.

PS:

Starbase is a bit limited due to having Caribbean needles to thread...

If they launched direct to 28.6º, the Starship would fly over the northern suburbs of Tampa, but it would be 1450km downrange by the time it made landfall.  The instantaneous impact point would be moving awfully fast by then, and they might meet the FAA hazard-to-public standard.

They could also launch farther north than that (landfall near Cedar Key, south of Gainesville, north of Ocala, in between St. Augustine and Daytona Beach), which is about as low-population a path as you'll find in central Florida.  That's an inclination of 30.9º.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/18/2023 05:16 am
Any orbit with an inclination greater than 28.5º is accessible from both BC and the Cape.

Absolutely true from an orbital mechanics perspective. Launching to a 90 degree orbit out of BC might involve a flight path that seemed to regulators uncomfortably close to Dallas or Mexico City. Of course it would be silly to suggest a polar rendezvous orbit; launching out of BC to other inclinations might prove to be similarly (if not so extremely) problematic.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/18/2023 05:33 am
But wait!  There's more!  Both ships need to provide some of the ullage thrust, in proportion to their current mass, to keep things from rotating. 
The other way to do this is to tie the ships together and have one (probably the depot) thrust through the combined center of mass.   The forces  are small, so the docking structure does not need to be robust.  A dozen rolls of duck tape would be strong enough (though difficult to deploy)

You'd still need to switch from one ship to another as the ullage space changed, just to get the prop you needed.  Note that thrust vectoring, either via gimbaling or throttling multiple thrusters, will be required either way.

Quote
AIUI the LSS landing requirements are unique to LSS.  Don't let it poison the architecture.  If need be kludge on some solids or hypergolics, take NASAs money, learn what you can, and on to Mars.

I actually left out an important use case:  low delta-v orbital maneuvers (10-200m/s).  It'd be tough to do those accurately with a Raptor, but easy with a methox thruster.

If they didn't have to do the waist landing thruster, I suspect that they'd forgo developing methox thrusters for now, and wait and see.  But, given that they do need something (and they probably need something for Mars as well), once they have the tech and the plumbing for methox, it's cheaper and cleaner to use it everywhere.

Ullage-based thrusters are a fine idea for SuperHeavy, where they have to dump tonnes of ullage gas overboard to lighten the load, but they're dumb for Starship.  Starship is an interplanetary-class spacecraft, which has to do small maneuvers weeks and months apart.  They're gonna need methox thrusters eventually.  Might as well get the development out of the way now, and use them for LSS.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/18/2023 05:56 am
For ease of reference: 14 CFR 450.101(a)(2)(i): "The risk to any individual member of the public, excluding neighboring operations personnel, must not exceed a probability of casualty of 1 × 10-6 per launch."
https://www.ecfr.gov/current/title-14/part-450#p-450.101(a)(2)(i)

And yes, the IIP trace moves really fast as the vehicle approaches orbit. For launches out of the Cape the trace often crosses Africa, but that's just before it disappears completely because the vehicle has made it to orbit.

EDIT to add:
1,509 km (Matamoros  to Tampa)
10,848 km (Cocoa Beach to Luanda)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/18/2023 08:20 am
For ease of reference: 14 CFR 450.101(a)(2)(i): "The risk to any individual member of the public, excluding neighboring operations personnel, must not exceed a probability of casualty of 1 × 10-6 per launch."
https://www.ecfr.gov/current/title-14/part-450#p-450.101(a)(2)(i)

And yes, the IIP trace moves really fast as the vehicle approaches orbit. For launches out of the Cape the trace often crosses Africa, but that's just before it disappears completely because the vehicle has made it to orbit.

EDIT to add:
1,509 km (Matamoros  to Tampa)
10,848 km (Cocoa Beach to Luanda)

Arrrgh, this is going to be one of those horrible expected value (1E-4) vs. probability (1E-6) issues.  I think you want the expected value.  You, as a member of the public, have a 1E-6 chance of dying, and so do I.  But the launch itself has an expected value of killing 1/10,000th of a person.

I didn't know that the IIP didn't disappear until Africa.  So the real question is how fast it's moving when it hits Tampa.  If it crosses the whole Florida peninsula in a couple of seconds, that's no big deal.  If it takes a couple of minutes, then SpaceX probably can't launch anywhere except through the Florida Strait without a dogleg.  And that would be a hell of dogleg, because you basically have to wait long enough that the IIP will stay out to sea and bend around Florida, which means that it'll occur at high velocity, and therefore require a huge amount of delta-v.

Launching from BC (latitude 26.0º) through the Florida Strait yields an inclination of 26.3º.  It would probably be cheaper to launch from LC-39A as southerly as possible, then dogleg left to achieve that inclination.  Apparently when I wrote my dogleg calculating spreadsheet, I only considered cases of increasing inclination, not decreasing it, because it explodes and I'm not doing spherical trigonometry tonight, even with a cheat sheet.  So I can't quantify the delta-v hit.  I suspect it's... not terrible?

The on-orbit plane change is pretty easy to do.  In a 400 x 400 x 28.6º orbit, getting to 400 x 400 x 26.3º costs 308m/s.  That would reduce a tanker's prop load from 150t to 128t.  Not great, but not the end of the world.  The dogleg will cost less than that.

I'm still hoping for a really fast IIP when it hits Tampa.  I'd be well within the corridor, but I'm willing to take my chances.  Take that, FAA!

Update (after the launch):

Looks like crossing the Florida peninsula with a launch is gonna be a no-go, based on Starship's projected debris track today.  So some combination of dogleg and plane change to get to a 26º-ish inclination will be necessary.  I'll SWAG the prop loss at 20t.

I wonder if that's why various folks at NASA have upped the number of launches (although various other folks at NASA still think it's high single-digit number of launches).  If they can only get, on average, 137t of prop to LEO (150t from BC and 130t from LC-39A with dogleg, with 2x as many LC-39A launches as BC launches), I get 11 tanker launches, if depot boil-off can be limited to 180kg/day, with a 60-day accumulation campaign.

LSS assumptions:

Dry Mass: 95t
Crew Module: 15t
Vacuum Payload (deployed on surface): 5t
Samples returned: 1t
Prop capacity: 1500t (eats 6m of cylindrical payload space, leaving 2m for garage)
Isp: 369s (3 RVacs @ 100% + 1 RSL @ 50% throttle)
Boil-off:  150kg/day during a 120-day TLI-BLT-NRHO-loiter period

Note that I'm assuming that the LCH4 dome and LCH4-LOX bulkhead are merely stacked farther forward, and that the outer mould line isn't changed.  If Starship gets stretched (likely, but unknown), then a vanilla propulsion section will probably work, and you can have a full 8m-high garage.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/18/2023 08:58 am
Launching from BC (latitude 26.0º) through the Florida Strait yields an inclination of 26.3º.  It would probably be cheaper to launch from LC-39A as southerly as possible, then dogleg left to achieve that inclination.  Apparently when I wrote my dogleg calculating spreadsheet, I only considered cases of increasing inclination, not decreasing it, because it explodes and I'm not doing spherical trigonometry tonight, even with a cheat sheet.  So I can't quantify the delta-v hit.  I suspect it's... not terrible?

The on-orbit plane change is pretty easy to do.  In a 400 x 400 x 28.6º orbit, getting to 400 x 400 x 26.3º costs 308m/s.  That would reduce a tanker's prop load from 150t to 128t.  Not great, but not the end of the world.  The dogleg will cost less than that.


It's not an either/or (dogleg vs plane change burn). If you want to achieve "true" optimality, you do both.  :D

On F9 GTO flights out of the Cape, we typically see SpaceX do a small inclination-reducing dogleg so the transfer orbit inclination is a few tenths of a degree below the Cape's latitude. Then they still do the standard plane-change over the equator. They even (again optimally) plan for a small portion of the plane-change to occur at apogee, resulting in a GTO that isn't exactly 0°.


Mathematically this all falls out of the small angle approximation (https://en.wikipedia.org/wiki/Small-angle_approximation). During any burn, adding a small amount of perpendicular delta-v is "free" due to the SAA. Therefore the optimum trajectory distributes the plane change across all the burns, with the exact distribution "weighted" by the relative efficiency of doing a plane change during that burn.


This isn't exotic or speculative. This is SOP in industry, as you can verify by checking the coasting orbits for any bog standard F9 GTO launch.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: JayWee on 11/18/2023 09:17 am
It's a lot easier just to run some LCH4 into a COPV, seal the COPV, and heat the LCH4 to supercritical pressures.  But that requires more plumbing.
Aren't you getting heated GCH4 during launch for free thanks to self-pressurization anyway? Would that be enough?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/18/2023 09:42 am
For ease of reference: 14 CFR 450.101(a)(2)(i): "The risk to any individual member of the public, excluding neighboring operations personnel, must not exceed a probability of casualty of 1 × 10-6 per launch."
https://www.ecfr.gov/current/title-14/part-450#p-450.101(a)(2)(i)

And yes, the IIP trace moves really fast as the vehicle approaches orbit. For launches out of the Cape the trace often crosses Africa, but that's just before it disappears completely because the vehicle has made it to orbit.

EDIT to add:
1,509 km (Matamoros  to Tampa)
10,848 km (Cocoa Beach to Luanda)

Arrrgh, this is going to be one of those horrible expected value (1E-4) vs. probability (1E-6) issues.  I think you want the expected value.  You, as a member of the public, have a 1E-6 chance of dying, and so do I.  But the launch itself has an expected value of killing 1/10,000th of a person.

The rule was either written or explained by somebody who is bad at math.*

1E-6 was about the probability that an individual Londoner would be killed by each V2 aimed at the city in WWII.  All it requires is really bad guidance.  One would hope that the actual FAAs standards are a tad higher.

* Or, given the legacy of the space program, somebody who thought the Blitz was a good idea.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/22/2023 11:33 am
For ease of reference: 14 CFR 450.101(a)(2)(i): "The risk to any individual member of the public, excluding neighboring operations personnel, must not exceed a probability of casualty of 1 × 10-6 per launch."
https://www.ecfr.gov/current/title-14/part-450#p-450.101(a)(2)(i)

And yes, the IIP trace moves really fast as the vehicle approaches orbit. For launches out of the Cape the trace often crosses Africa, but that's just before it disappears completely because the vehicle has made it to orbit.

EDIT to add:
1,509 km (Matamoros  to Tampa)
10,848 km (Cocoa Beach to Luanda)

Arrrgh, this is going to be one of those horrible expected value (1E-4) vs. probability (1E-6) issues.  I think you want the expected value.  You, as a member of the public, have a 1E-6 chance of dying, and so do I.  But the launch itself has an expected value of killing 1/10,000th of a person.

The rule was either written or explained by somebody who is bad at math.*

1E-6 was about the probability that an individual Londoner would be killed by each V2 aimed at the city in WWII.  All it requires is really bad guidance.  One would hope that the actual FAAs standards are a tad higher.

* Or, given the legacy of the space program, somebody who thought the Blitz was a good idea.

6,000 launches of the V2 killed about 5,000 people in enemy territory. That's an expected value of 0.87 morts per launch. The FAA requires less than 1E-4 morts per launch.

So yes, the FAA standards are indeed a tad higher. ;)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/26/2023 07:34 am
For ease of reference: 14 CFR 450.101(a)(2)(i): "The risk to any individual member of the public, excluding neighboring operations personnel, must not exceed a probability of casualty of 1 × 10-6 per launch."
https://www.ecfr.gov/current/title-14/part-450#p-450.101(a)(2)(i)

And yes, the IIP trace moves really fast as the vehicle approaches orbit. For launches out of the Cape the trace often crosses Africa, but that's just before it disappears completely because the vehicle has made it to orbit.

EDIT to add:
1,509 km (Matamoros  to Tampa)
10,848 km (Cocoa Beach to Luanda)

Arrrgh, this is going to be one of those horrible expected value (1E-4) vs. probability (1E-6) issues.  I think you want the expected value.  You, as a member of the public, have a 1E-6 chance of dying, and so do I.  But the launch itself has an expected value of killing 1/10,000th of a person.

The rule was either written or explained by somebody who is bad at math.*

1E-6 was about the probability that an individual Londoner would be killed by each V2 aimed at the city in WWII.  All it requires is really bad guidance.  One would hope that the actual FAAs standards are a tad higher.

* Or, given the legacy of the space program, somebody who thought the Blitz was a good idea.

6,000 launches of the V2 killed about 5,000 people in enemy territory. That's an expected value of 0.87 morts per launch. The FAA requires less than 1E-4 morts per launch.

So yes, the FAA standards are indeed a tad higher. ;)
I was referring to the 1E-6 individual risk.  The numbers I have for V-2 aimed at London are about 2 morts per launch out of population in the target area of about 8 million.

With a total limit of 1E-4 morts per launch the individual risk limit is nugatory.  A member of the public at that risk limit is either trespassing or is being targeted.  Members of the public do not have a 1E-6 chance of dying during a launch.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tyrred on 11/26/2023 07:48 am
What is "nugatory"?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: catdlr on 11/26/2023 07:56 am
What is "nugatory"?

of no value or importance.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: tyrred on 11/26/2023 08:13 am
So which parts of on-orbit refueling are nugatory?

Best part is no part, eh?

There has got to be a way to make it nearly as simple as refueling a tanker truck, to take more fuel to another place.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/26/2023 11:54 am
For ease of reference: 14 CFR 450.101(a)(2)(i): "The risk to any individual member of the public, excluding neighboring operations personnel, must not exceed a probability of casualty of 1 × 10-6 per launch."
https://www.ecfr.gov/current/title-14/part-450#p-450.101(a)(2)(i)

And yes, the IIP trace moves really fast as the vehicle approaches orbit. For launches out of the Cape the trace often crosses Africa, but that's just before it disappears completely because the vehicle has made it to orbit.

EDIT to add:
1,509 km (Matamoros  to Tampa)
10,848 km (Cocoa Beach to Luanda)

Arrrgh, this is going to be one of those horrible expected value (1E-4) vs. probability (1E-6) issues.  I think you want the expected value.  You, as a member of the public, have a 1E-6 chance of dying, and so do I.  But the launch itself has an expected value of killing 1/10,000th of a person.

The rule was either written or explained by somebody who is bad at math.*

1E-6 was about the probability that an individual Londoner would be killed by each V2 aimed at the city in WWII.  All it requires is really bad guidance.  One would hope that the actual FAAs standards are a tad higher.

* Or, given the legacy of the space program, somebody who thought the Blitz was a good idea.

6,000 launches of the V2 killed about 5,000 people in enemy territory. That's an expected value of 0.87 morts per launch. The FAA requires less than 1E-4 morts per launch.

So yes, the FAA standards are indeed a tad higher. ;)
I was referring to the 1E-6 individual risk.

And my point is that "the actual FAAs standards"—and the safety they provide—comes from the totality of the rules, not just one rule in isolation.


With a total limit of 1E-4 morts per launch the individual risk limit is nugatory.  A member of the public at that risk limit is either trespassing or is being targeted.  Members of the public do not have a 1E-6 chance of dying during a launch.

Exactly, that's why the rule exists.

You can't have a launch that "targets" any group of people too much, even if unintentionally. This can sometimes impose additional geographic constraints beyond just the 1E-4 total limit.

The fact that "they'd be trespassing"...   only means the rule is working!  Why do you think SpaceX made sure to control (and/or own) that land?  ???
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/26/2023 07:27 pm

The fact that "they'd be trespassing"...   only means the rule is working!  Why do you think SpaceX made sure to control (and/or own) that land?  ???
They have to control the land because a crowd of (more than 100) gawkers would exceed the total morts before any individual risk would exceed 1E-6.  Similarly, overflights of Florida, Cuba or Africa put many thousands or millions at risk so the total risk controls by several orders of magnitude.

Individual risk makes sense testing a small rocket with limited range at White Sands.  But for a large orbital rocket in a range the size of the planet total risk will dominate, barring really contrived situations.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/27/2023 04:17 am

The fact that "they'd be trespassing"...   only means the rule is working!  Why do you think SpaceX made sure to control (and/or own) that land?  ???
They have to control the land because a crowd of (more than 100) gawkers would exceed the total morts before any individual risk would exceed 1E-6.

Yet that doesn't explain why they don't allow crowds smaller than 100 people.  ::)

Face it, individual risk does have a real impact on launch safety planning. Anyone who claims otherwise is the one making the weird positive claim, and the burden of proof is entirely on them (ie you) to prove that re-planning based on individual risk has never occurred.

Individual risk makes sense testing a small rocket with limited range at White Sands.

Thanks for providing a counterexample to prove my point.  :D

I never said that individual risk always changes the planning. That'd be silly. However sometimes it does, and that's a part of the overall safety that the totality of the FAA rules provide.



Anyway this is drifting off-topic. Let's get back to on-orbit refueling shall we...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/27/2023 05:49 am
For orbital rendezvous of vehicles launched from different locations is there an option akin to what's so cleverly done for GTO launches, where the inclination change is performed at the same time as an orbit raising burn? (AIUI for GTO the vehicle enters a temporary lower parking orbit and then coasts until an equator-crossing where it performs a combined apogee-raising and inclination-reducing burn, thereby incurring a considerably lower delta-v cost for the inclination change.)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/27/2023 10:16 pm
For orbital rendezvous of vehicles launched from different locations is there an option akin to what's so cleverly done for GTO launches, where the inclination change is performed at the same time as an orbit raising burn? (AIUI for GTO the vehicle enters a temporary lower parking orbit and then coasts until an equator-crossing where it performs a combined apogee-raising and inclination-reducing burn, thereby incurring a considerably lower delta-v cost for the inclination change.)

I don't know how to the do the full optimization of 28.6º to GTO to GEO, but my understanding is that you take out only a small amount of inclination at LEO-to-GTO injection, and most of it at GTO-to-GEO.  The reason for this is that inclination changes are dependent on the orbital speed at the point of impulse (Δv = 2*v*sin(Δi/2).)  (Update:  Fixed formula.) Orbital speed in 200x200km VLEO is 7784m/s.  Orbital speed at apogee of a 200 x 35,786km GTO is 1597m/s.

The problem with on-orbit plane changes for reaching the depot is the depot will likely be in VLEO.  That's what optimizes tanker prop to orbit.  But that means that plane changes are extremely expensive.  Finding a common inclination where both sites can launch directly is much more economical.  If I did it right, launching out of BC at an azimuth of 108º yields a 30.8º inclination,¹ which can be reached from LC-39A via either azimuth 77º or 103º.

___________
¹There are some issues:  Your IIP can't cross the Yucatan, Cuba, the Caymans, or Jamaica.  To figure this out, you need to know the exact time and velocity for a lot of downrange points, then adjust the current earth-fixed longitude for the rotation of the Earth since the launch.  You also need a (very complicated) model for converting each <downrange, velocity> tuple to an <impact point, velocity> tuple.  (The IIP velocity is important because risk of killing somebody is high when it's moving slowly and low when it's moving quickly.)

I've only arm-waved this, but there's a pretty good chance of overflying heavily populated portions of Venezuela.  However, Venezuela is more than 3800km downrange.  The IIP will be really moving by that time, and the expected value may easily be below the 1E-4 people/launch requirement.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 11/28/2023 05:01 pm
For orbital rendezvous of vehicles launched from different locations is there an option akin to what's so cleverly done for GTO launches, where the inclination change is performed at the same time as an orbit raising burn? (AIUI for GTO the vehicle enters a temporary lower parking orbit and then coasts until an equator-crossing where it performs a combined apogee-raising and inclination-reducing burn, thereby incurring a considerably lower delta-v cost for the inclination change.)
If your launch site is at a latitude less-than or equal-to the inclination of the target orbit, then you can always launch into an orbit with the correct inclination. Only when the launch site is too far north do you have to worry about changing the inclination after launch. That's important because, as TheRadicalModerate illustrated, inclination changes are very expensive. Sometimes they're so expensive that it'd be cheaper to land the vehicle, ship it to a better launch site, and relaunch it than to try to do the inclination change in space! (With some assumptions about EDL and recovery costs being negligible.) :-) The one real exception (as TRM alludes to) is when the orbit is highly elliptical. Then you can do the inclination change at apogee relatively cheaply.

You can also get the ascending node right at launch time, but only once or twice a day (depending on how close to the equator your launch site is). However, given that the inclination is correct, any vehicle in LEO tends to precess fairly rapidly (days or weeks). If the target orbit is higher or lower than your initial orbit, then you can simply wait until the two orbits are aligned (i.e. until the ascending node is at the right longitude) at zero fuel cost. Then you do the usual boost to raise apogee on the opposite side of the orbit, and then boost again at the new apogee to raise perigee. Some fine tuning is also going to be required to actually match up with another vehicle (e.g. enter an orbit just a bit lower than the target and then wait to catch up to it.)

For example, if you've got a depot in VLEO at the inclination of Cape Canaveral, then you could easily launch ships to that orbit (right altitude, inclination, and ascending node) from either Kennedy or Boca Chica spaceports once a day (or from any spaceport closer to the equator). You could do multiple launches a day if you were willing to wait a few days (or weeks) for them to line up.

Given that we're talking VLEO, I suppose it's just barely possible that someone could take advantage of the atmospheric drag to make bigger (or faster) changes to the orbit (e.g. with a "space rudder"), but I've never seen that seriously explored.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/28/2023 07:05 pm
For orbital rendezvous of vehicles launched from different locations is there an option akin to what's so cleverly done for GTO launches, where the inclination change is performed at the same time as an orbit raising burn? (AIUI for GTO the vehicle enters a temporary lower parking orbit and then coasts until an equator-crossing where it performs a combined apogee-raising and inclination-reducing burn, thereby incurring a considerably lower delta-v cost for the inclination change.)
If your launch site is at a latitude less-than or equal-to the inclination of the target orbit, then you can always launch into an orbit with the correct inclination. Only when the launch site is too far north do you have to worry about changing the inclination after launch. That's important because, as TheRadicalModerate illustrated, inclination changes are very expensive. Sometimes they're so expensive that it'd be cheaper to land the vehicle, ship it to a better launch site, and relaunch it than to try to do the inclination change in space! (With some assumptions about EDL and recovery costs being negligible.) :-) The one real exception (as TRM alludes to) is when the orbit is highly elliptical. Then you can do the inclination change at apogee relatively cheaply.

You can also get the ascending node right at launch time, but only once or twice a day (depending on how close to the equator your launch site is). However, given that the inclination is correct, any vehicle in LEO tends to precess fairly rapidly (days or weeks). If the target orbit is higher or lower than your initial orbit, then you can simply wait until the two orbits are aligned (i.e. until the ascending node is at the right longitude) at zero fuel cost. Then you do the usual boost to raise apogee on the opposite side of the orbit, and then boost again at the new apogee to raise perigee. Some fine tuning is also going to be required to actually match up with another vehicle (e.g. enter an orbit just a bit lower than the target and then wait to catch up to it.)

For example, if you've got a depot in VLEO at the inclination of Cape Canaveral, then you could easily launch ships to that orbit (right altitude, inclination, and ascending node) from either Kennedy or Boca Chica spaceports once a day (or from any spaceport closer to the equator). You could do multiple launches a day if you were willing to wait a few days (or weeks) for them to line up.

Given that we're talking VLEO, I suppose it's just barely possible that someone could take advantage of the atmospheric drag to make bigger (or faster) changes to the orbit (e.g. with a "space rudder"), but I've never seen that seriously explored.

RAAN is important for HEEO but, as you say, you can manage it cheaply (although not quickly) with precession.  The harder problem is the argument of perigee.  There's nothing cheap to be done there.  Since that's what governs where the apse line points, it's a fairly big deal.  You can fiddle with it a bit, but changing it dramatically will cost you.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 11/28/2023 07:08 pm
the inclination of Cape Canaveral, then you could easily launch ships to that orbit [...] from [...] Boca Chica

As mentioned up-thread (or in HLS thread?), yes you could easily do that... if there weren't regulations you had to follow about the safety of people in e.g. Florida & Cuba. There is also a discussion about which orbital inclinations you can get regulatory approval to launch into from BC, with or without a dogleg

Normally, launch sites are chosen to avoid being constrained too much by these considerations... and launch sites are normally not built in Boca Chica, but on edges of larger oceans or deserts
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/28/2023 08:52 pm
For orbital rendezvous of vehicles launched from different locations is there an option akin to what's so cleverly done for GTO launches, where the inclination change is performed at the same time as an orbit raising burn?

It's important to realize that real vehicles perform inclination change during every burn in the flight.


     •  During launch they change inclination slightly (0.1-0.2 degrees) by performing a small dogleg.

     •  During the orbit-raising burn they perform the majority of the inclination change.

     •  During the apogee burn they perform a small change in inclination (0.3-0.4 degrees, maybe more).


Doing it like this is actually more efficient than doing 100% of the inclination change during the second burn, because math (https://en.wikipedia.org/wiki/Small-angle_approximation).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/29/2023 12:33 am
For orbital rendezvous of vehicles launched from different locations is there an option akin to what's so cleverly done for GTO launches, where the inclination change is performed at the same time as an orbit raising burn?

It's important to realize that real vehicles perform inclination change during every burn in the flight.
[...]     •  During the orbit-raising burn they perform the majority of the inclination change.

Doing it like this is actually more efficient than doing 100% of the inclination change during the second burn, because math (https://en.wikipedia.org/wiki/Small-angle_approximation).

Thanks for this, and all the prior responses. Is it correct that this efficiency is only because the delta-v of the orbit-raising burn for the GTO flight profile is so high? So for example the gain is much less if the orbit-raising burn were only from 200 km LEO to 400 km LEO?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 11/29/2023 01:33 am
Fact check: What's the inclination of GEO-1800?

26.5⁰ or thereabouts isn't it?

Which would imply 90%+ of the plane change is done in the apogee burn, where the downrange speed (prior to the burn) is much much lower than in LEO
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 11/29/2023 04:05 am
If I did it right, launching out of BC at an azimuth of 108º yields a 30.8º inclination,¹ which can be reached from LC-39A via either azimuth 77º or 103º.
Why 30.8º? Wouldn't optimal be Cape Canaveral's latitude (28.39º) since Boca Chica is slightly south of that (26º). Then you should launch due east (or close to it). (Azimuth 90º, right?) In that case, I'd expect the IIP to clear the tip of Florida, and, by that point, it ought to be moving fast enough not to matter what it crosses after that. From Canaveral (where it really is due east), there's nothing but the ocean all the way to Africa.

Launching into any other orbit from Boca Chica might be a bear, but a (mostly) due-east launch seems to be pretty safe.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/29/2023 04:24 am
     •  During the orbit-raising burn they perform the majority of the inclination change.

     •  During the apogee burn they perform a small change in inclination (0.3-0.4 degrees, maybe more).

This is backwards.  See here (https://www.reddit.com/r/spacex/wiki/launches/gto_performance/).

The main reason to apply any inclination change at LEO-to-GTO injection is because you have some extra performance left in your second stage, and you might as well use it before deploying the payload and losing the ability.  That extra performance can be applied either to some moderate inclination change, to boosting up to supersync, or some combination of both.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/29/2023 05:09 am
If I did it right, launching out of BC at an azimuth of 108º yields a 30.8º inclination,¹ which can be reached from LC-39A via either azimuth 77º or 103º.
Why 30.8º? Wouldn't optimal be Cape Canaveral's latitude (28.39º) since Boca Chica is slightly south of that (26º). Then you should launch due east (or close to it). (Azimuth 90º, right?) In that case, I'd expect the IIP to clear the tip of Florida, and, by that point, it ought to be moving fast enough not to matter what it crosses after that. From Canaveral (where it really is due east), there's nothing but the ocean all the way to Africa.

Launching into any other orbit from Boca Chica might be a bear, but a (mostly) due-east launch seems to be pretty safe.

First, to get to 28.6º inclination (which is LC-39A's latitude--you had an arcminutes to decimal degree confusion) from BC, you need to launch along one of the following two azimuths:¹

77.0º, which pretty much goes over Tampa
103.1º, which might work, but also might go over parts of Cuba and Hispaniola.

The big takeaway for me from the IFT-2 mission was that a lot of debris can land way downrange.  Pretty sure that takes the northerly azimuth off the table.  The southerly one is a possibility, but IFT-2 had a debris trail that showed up 3400km downrange, while the vehicle was well short of even its suborbital target.  Hispaniola is only 2700km downrange.

Note:  You can't just draw the azimuth line and assume that's where the IIP runs.  The Earth will rotate about 300km over the course of a launch, so the IIP, even with no doglegs, will curve northerly for northerly azimuths and southerly for southerly azimuths.  You need a lot of trajectory data for an accurate trace of the IIP--and even then, you don't really know the ground speed of the IIP, which, along with population density data,² is essential to computing the Dead People Expected Value.

That southerly curve might cause the 103.1º azimuth to just barely clear both Cuba and Hispaniola.  If that works, then it's open ocean all the way to the Windward Islands, which are 4000km downrange, and tiny specks of land.  If this one works, it's the winner for the depot.

I chose the 30.6º inclination because it looked like it would sneak between the Yucatan and the Caymans, and then it was open ocean all the way to Venezuela, which is 3700km downrange.  But I'm not even sure that's far enough for the IIP speed to make the expected value small enough.

________
¹For a primer on the math, I recommend this (https://www.orbiterwiki.org/wiki/Launch_Azimuth).  Be sure to have a couple of tablets of Advil Extra-Strength Spherical Trigonometry Formula on hand.

²An obvious question is why the FAA allows the Cuban Dogleg to allow flights to get to  polar orbits from Florida.  I suspect a big chunk of that answer is that the F9 is insanely reliable, which also factors into the expected value.  That, and the IIP looks like it goes over a narrow, not particularly heavily populated part of the island.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/29/2023 05:29 am
Fact check: What's the inclination of GEO-1800?

26.5⁰ or thereabouts isn't it?

This risks going off-topic, and hopefully discussion can return to orbits more likely than GTO to be used by Starship for on-orbit propellant transfer. Atlas V offered a wide range of geo-transfer orbit (TO) inclinations, all the way down to 18°, according to the Mission Planner’s Guide. (See attached.) And F9 launched SES-22 into a remarkable 310 x 35627 km x 16.5° transfer orbit. (On Twitter see planet4589/status/1542337107716444162.)
The math alluded to in a prior post isn't immediately intuitive. It involves combining the orbit raising delta-v "a" and the inclination reducing delta-v "b" not by summing them, but by using sqrt(a^2 + b^2). (Further explanation from the NSF archives is at https://forum.nasaspaceflight.com/index.php?topic=36954.) If a is large enough, the influence of b is markedly reduced.

Personally I'm quite curious about the path from BC that leads between Yucatan and Cuba. The STS-36 launch lends some credence to the notion that overflight of land during ascent can in some circumstances be deemed "okay."
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/29/2023 06:43 am
This risks going off-topic, and hopefully discussion can return to orbits more likely than GTO to be used by Starship for on-orbit propellant transfer. Atlas V offered a wide range of geo-transfer orbit (TO) inclinations, all the way down to 18°, according to the Mission Planner’s Guide. (See attached.) And F9 launched SES-22 into a remarkable 310 x 35627 km x 16.5° transfer orbit. (On Twitter see planet4589/status/1542337107716444162.)

Think of it this way:  Based on the size of the payload, the F9 S2 (or any other launcher's final stage) arrives in its parking orbit with x amount of delta-v left in the tank.  You want to use all of it.  If you can just barely get to GTO, then the GTO inclination is likely to be the same as the launch inclination.  If you have more performance than that, you can decide to take out some inclination, or do some kind of supersync GTO (setting yourself up for a bi-elliptic transfer (https://en.wikipedia.org/wiki/Bi-elliptic_transfer), if it turns out to be more efficient).

Small payload = more delta-v left over = more inclination take-out.

Quote
The math alluded to in a prior post isn't immediately intuitive. It involves combining the orbit raising delta-v "a" and the inclination reducing delta-v "b" not by summing them, but by using sqrt(a^2 + b^2). (Further explanation from the NSF archives is at https://forum.nasaspaceflight.com/index.php?topic=36954.) If a is large enough, the influence of b is markedly reduced.

It's actually the law of cosines, applied to the initial velocity vector (i.e., the orbital speed at apogee), the target velocity vector (i.e., the orbital speed after the perigee's been raised), and the inclination change, which is the angle between the two vectors:

Δvtotal = sqrt( vinit² + vfinal² - 2*vinit*vfinal*cos(Δinclination) )

You can work out vinit and vfinal by applying vis-viva (https://en.wikipedia.org/wiki/Vis-viva_equation) at your GTO's apogee and the GEO orbital speed.  Same thing for any two orbits that share a common apogee or perigee.  (Don't forget!  Altitude and radius are not the same thing!  They differ by the Earth's radius.)

Quote
Personally I'm quite curious about the path from BC that leads between Yucatan and Cuba. The STS-36 launch lends some credence to the notion that overflight of land during ascent can in some circumstances be deemed "okay."

Computing the Dead People Expected Value (not the official term, for obvious reasons) is insanely complicated, but one of its terms is the reliability of the rocket along its flight path, which changes based on what kind of maneuver is being done (e.g. staging is less reliable than exoatmospheric steady boost).  The F9 is insanely reliable, so its DPEV is much lower.  Starship won't be proven to be that reliable for quite a while, likely long after SpaceX needs to start the refueling campaign.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 11/29/2023 07:00 am
Note:  You can't just draw the azimuth line and assume that's where the IIP runs.  The Earth will rotate about 300km over the course of a launch, so the IIP, even with no doglegs, will curve northerly for northerly azimuths and southerly for southerly azimuths.
"curve northerly"... I know what you mean, but just in an attempt to minimize confusion for anyone else watching:

In the rotating frame, you start accelerating from rest due [launch azimuth] and as a result of the Coriolis force the ground track will always curve right (in the northern hemisphere).
(Speculation: The IIP should also curve right, unless [flight path elevation] >> [latitude])

In Earth's inertial frame, you can make some assumptions and draw the great circle of the orbital plane, due [orbital plane azimuth] from the launch site. If 000<azimuth<090, due to earth's rotation (i.e. the map moving east) the ground track immediately heads off to the left by some angle (as opposed to curving left). Namely, it goes due [launch azimuth] instead.

The ground track then "curves" right slightly faster than the [orbital plane azimuth] is increasing (in the northern hemisphere), thus matching the behaviour seen in the rotating Earth frame.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/29/2023 08:29 am

RAAN is important for HEEO but, as you say, you can manage it cheaply (although not quickly) with precession.
There's a point where not quickly becomes not cheaply.  I think you're going to reach that point here.

Differential precession is going to be less than half a degree per day.  If you don't match RAAN quite closely during launch you could be waiting for weeks or months.  Taking weeks or months to rendezvous will cost in boiloff and limited tanker reuse.

The harder problem is the argument of perigee.  There's nothing cheap to be done there.  Since that's what governs where the apse line points, it's a fairly big deal.  You can fiddle with it a bit, but changing it dramatically will cost you.

For circular, or near circular, orbits the argument of perigee doesn't matter.

The vast majority of rendezvous will be between just launched tankers and depots in low or very low circular orbits.  These are the operations you need to optimize and can almost ignore argument of perigee.

If you need additional steps on a refueling ladder then you have to worry about argument of perigee, but it should be easily manageable by gathering one or more depots and the LSS in a low circular orbit and departing either in formation or at appropriate short intervals to match argument of perigee.

If we had an equatorial launch site and could rendezvous in an equatorial orbit we could ignore RAAN as well.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Greg Hullender on 11/29/2023 05:12 pm

RAAN is important for HEEO but, as you say, you can manage it cheaply (although not quickly) with precession.
There's a point where not quickly becomes not cheaply.  I think you're going to reach that point here.

Differential precession is going to be less than half a degree per day.  If you don't match RAAN quite closely during launch you could be waiting for weeks or months.  Taking weeks or months to rendezvous will cost in boiloff and limited tanker reuse.
Well, how bad is it? Actually, it does look pretty bad.

Using the orbital precession formula (https://en.wikipedia.org/wiki/Nodal_precession#Equation) from Wikipedia, if the tanker were at in a circular orbit at 200 km inclined by 28.39° (this is Cape Canaveral; LC-39 is 28.61°, but that only makes a tiny difference), it would precess at 7.90°/day. If you put the depot up near the Van Allen belts, say 700 km up, it would precess at 6.11°/day. Accordingly, they'd line up once in just over 200 days. Not great. Especially when 200 km is probably too low and 700 km is probably too high.

If we do 200 km and 400 km, it's just once every 456 days, but even once ever 200 days is probably unworkable.

It does work much better for HEEO. An orbit with perigee of 200 km (above the surface) and a period of 3.41 days (1/8 sidereal lunar period) has an apogee of 185,000 km and a precession period of 26 years. A vehicle in a 200 km circular orbit aligns with this orbit once every 46 days. But HEEO has it's own problems.

So that means we don't really have a good way to fuel a single VLEO depot from Boca Chica/Cape Canaveral. Not unless there's a clever way to rotate the RAAN using air friction or something. Or have I missed something?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 11/29/2023 07:18 pm
it would precess at 7.90°/day. If you put the depot up near the Van Allen belts, say 700 km up, it would precess at 6.11°/day. Accordingly, they'd line up once in just over 200 days. Not great.
[...]
So that means we don't really have a good way to fuel a single VLEO depot from Boca Chica/Cape Canaveral. Not unless there's a clever way to rotate the RAAN using air friction or something. Or have I missed something?
My 2c:
RAAN can be primarily controlled by (siderial) time of launch. Having that (<)1.79⁰/day of precession to play with just means your launch window can be longer. If you need a 1 hour launch window to find a gap in the weather, that corresponds to a 15⁰ range in RAAN (less any LV cross-range capability) - so you're looking at waiting up to (>)~8 days for the RAAN to align

If the inclination is sufficiently above the launch site latitude so that you have 2 distinct launch azimuths, far enough apart in time to turn the pad/SH around, maybe you can get 2 launches a day from KSC, each with a non-instantaneous launch window. If you need 3 or 4 launches a day... then you're looking at waiting for RAAN to drift by 15⁰/hr * [pad turnaround time], which will take a long time, but it's still not 360⁰

-- some calculations --
For that 30.8⁰ inclination 200km & 400km, precession is ⁻7.72⁰/day and ⁻6.95⁰/day. Say we want 1 BC and 2 KSC launches a day:
I get 2 KSC launch windows a day 2h 46 mins apart (2h 43 factoring in ongoing precession)
If you need 20 min launch windows, and 6 hours for pad/SH turnaround, then the 2nd launch window (az=103⁰) runs until 6h40 - 2h43 = 3h57 after RAAN alignment and the RAAN will be 60.4⁰ too high, which will take 79 days to align.

Well, it's still so high that it defeats the purpose of launching twice a day
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 11/29/2023 07:47 pm
it would precess at 7.90°/day. If you put the depot up near the Van Allen belts, say 700 km up, it would precess at 6.11°/day. Accordingly, they'd line up once in just over 200 days. Not great.
[...]
So that means we don't really have a good way to fuel a single VLEO depot from Boca Chica/Cape Canaveral. Not unless there's a clever way to rotate the RAAN using air friction or something. Or have I missed something?
My 2c:
RAAN can be primarily controlled by (siderial) time of launch. Having that (<)1.79⁰/day of precession to play with just means your launch window can be longer. If you need a 1 hour launch window to find a gap in the weather, that corresponds to a 15⁰ range in RAAN (less any LV cross-range capability) - so you're looking at waiting up to (>)~8 days for the RAAN to align

If the inclination is sufficiently above the launch site latitude so that you have 2 distinct launch azimuths, far enough apart in time to turn the pad/SH around, maybe you can get 2 launches a day from KSC, each with a non-instantaneous launch window. If you need 3 or 4 launches a day... then you're looking at waiting for RAAN to drift by 15⁰/hr * [pad turnaround time], which will take a long time, but it's still not 360⁰
I'm confused. I can launch once a day, or I can launch twice a day if I am free to choose either ascending or descending. That puts me into the correct plane. It's fairly cheap to get to the correct phase after I'm in the correct plane. Missions to the ISS do this routinely. no need to wait 8 days.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Barley on 11/29/2023 08:47 pm
There is a subtle problem with using precession to give wider launch windows.

Precession of lower orbits is to the west, the same direction as a delay in launching.

If you are sending a tanker to rendezvous with a depot in LEO the best point in a launch window is when they lineup, at the end of the launch window.  You'd really prefer for the optimum to be at the beginning of the window, so you can plan on best performance and take a small hit if you miss.  As is you have to plan on less than best performance and take a hit with every launch that is not delayed.

If you need say 8 tankers, it's probably better to have 16 instantaneous launch windows over eight days and miss a few than trying to do hour long windows over four days.

It's different if you want to launch the LSS directly to an elliptical orbit.  You can have the depot wait in a low circular orbit until the LSS launches so short delays can be dealt with by delaying the depot boost while the depot precesses.  If you insist on have the depot in the elliptical orbit first, you have the same problem as above.

You should work really hard on hitting instantaneous launch windows before you start working on ways to deal with not hitting the window.  Get the operational stuff right first.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 11/29/2023 09:27 pm
Precession of lower orbits is to the west, the same direction as a delay in launching.
You're saying the opposite of what I was thinking, so let's clarify this

Using Earth's inertial frame, the target orbit RAAN drifts to lower numbers, and the parking orbit drifts to lower numbers faster. So a parking orbit with a higher RAAN is the one that can be easily corrected.

If a launch is delayed, the Right Ascension of the launch site is quickly increasing, and so the RAAN of the parking orbit will be high... which is the forgiving option

Does that make sense? What part is in doubt?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 11/29/2023 09:30 pm
...

Does that make sense? What part is in doubt?

The part where "just launch on time" is a huge problem...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/30/2023 04:56 am
Note:  You can't just draw the azimuth line and assume that's where the IIP runs.  The Earth will rotate about 300km over the course of a launch, so the IIP, even with no doglegs, will curve northerly for northerly azimuths and southerly for southerly azimuths.
"curve northerly"... I know what you mean, but just in an attempt to minimize confusion for anyone else watching:

In the rotating frame, you start accelerating from rest due [launch azimuth] and as a result of the Coriolis force the ground track will always curve right (in the northern hemisphere).
(Speculation: The IIP should also curve right, unless [flight path elevation] >> [latitude])

In Earth's inertial frame, you can make some assumptions and draw the great circle of the orbital plane, due [orbital plane azimuth] from the launch site. If 000<azimuth<090, due to earth's rotation (i.e. the map moving east) the ground track immediately heads off to the left by some angle (as opposed to curving left). Namely, it goes due [launch azimuth] instead.

The ground track then "curves" right slightly faster than the [orbital plane azimuth] is increasing (in the northern hemisphere), thus matching the behaviour seen in the rotating Earth frame.

Once the launcher is clear of the atmosphere (which largely rotates with the Earth), its orbital plane is in the inertial frame, and isn't rotating with the Earth any more.  So the apparent ground track (and the IIP), is west of the position/velocity track of the orbit itself.  So it'll appear to curve north of the azimuth for northerly azimuths, and south of the southerly azimuths.

Note that this is a simplification, because the inclination-to-azimuth formula I used assumes an impulsive launch, which isn't the case.  So there are some Coriolis forces caused by the acceleration of the launcher as it burns along the trajectory, but I think they're pretty small.

Attached are the two instantaneous azimuths to 28.6º in red, and the approximate southerly ground track in yellow.  The end point should be relatively accurate, and it based on how long it takes the Starship to get 4000km downrange, which is then used to compute how far westerly it should be from the azimuth at that latitude. 

Note that the ground track is not the IIP; that will be some unknown distance downrange of the ground track, but it'll have the same approximate shape.

Still not enough info to figure out if a directly launch would work or not.  And I'm way too lazy to think about doglegs.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/30/2023 05:31 am
Precession of lower orbits is to the west, the same direction as a delay in launching.
You're saying the opposite of what I was thinking, so let's clarify this

Using Earth's inertial frame, the target orbit RAAN drifts to lower numbers, and the parking orbit drifts to lower numbers faster. So a parking orbit with a higher RAAN is the one that can be easily corrected.

If a launch is delayed, the Right Ascension of the launch site is quickly increasing, and so the RAAN of the parking orbit will be high... which is the forgiving option

Does that make sense? What part is in doubt?

I wouldn't worry too much about the RAAN.  For a wide-ish window, you launch when the RAAN of the tanker's parking orbit should be that of the depot.  For minor delays, you:

1) Decrease the azimuth for northerly launches, then dogleg right when your RAAN matches up.

2) Increase the azimuth for southerly launches, then dogleg left when the RAAN matches up.


For LC-39A, both azimuths should be relatively open.  For BC, only the southerly azimuth will work, but the dogleg should be open--at least until it can't thread the needle between Cuba and Jamaica any more.  (This assumes the 28.6º inclination will work from BC.  Your inclination--and geographic impediments--may vary.)

I've kinda lost track of who said what up-thread, but precession is also not going to be much of an issue.  If the orbit is circular (and I think it will be), you launch to the proper inclination/RAAN and a slightly trailing true anomaly, then you haul ass to phase up to the depot.  Taking your time increases boil-off, which decreases prop to the depot.  If you have to wait for RAAN precession to do your job for you, you're gonna have a pretty dry tanker.

I still don't know how you recover a tanker--or any Starship, for that matter--to its launch site.  AFAICT, all of the approaches are overland through high population areas.  It's possible that the hypersonic entry phase is so high, and with such a high deceleration, that the bulk of the IIP stays out to sea, and then the belly flop goes straight down, and can be safely off-shore until the last few moments.  But I think it remains to be seen how SpaceX does this.  (This should be Yet Another Thread Topic (https://forum.nasaspaceflight.com/index.php?topic=59970.new#new).)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: sdsds on 11/30/2023 05:37 am
[...]
It's actually the law of cosines, applied to the initial velocity vector (i.e., the orbital speed at apogee), the target velocity vector (i.e., the orbital speed after the perigee's been raised), and the inclination change, which is the angle between the two vectors:

Δvtotal = sqrt( vinit² + vfinal² - 2*vinit*vfinal*cos(Δinclination) )

Thanks for the pointer! This is indeed the textbook answer.

There seems to be a commonly held belief that inclination changes in LEO are prohibitively expensive. Fortunately the rather messy law of cosines cleans up quite nicely in the simple case of a pure inclination change between two circular low-Earth orbits with the same radius. Of note, Δv is proportional to sin(Δi/2), hinting that for small inclination changes the Δv might not be prohibitive. The spreadsheet seems to indicate a Δv of only 355 m/s to get between orbits "natural" of LC-39 and BC when the orbital altitude is 205 km.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/30/2023 05:41 am

RAAN is important for HEEO but, as you say, you can manage it cheaply (although not quickly) with precession.
There's a point where not quickly becomes not cheaply.  I think you're going to reach that point here.

Differential precession is going to be less than half a degree per day.  If you don't match RAAN quite closely during launch you could be waiting for weeks or months.  Taking weeks or months to rendezvous will cost in boiloff and limited tanker reuse.

Let's be clear:  This is a big deal for HEEOs, where both RAAN and argument of perigee are forcing pretty serious restrictions on TLI or interplanetary injections.

And in an HEEO, RAAN precession is hard to compute, because the quadrupole moment is dependent on altitude which is... highly eccentric.

I'll spare you my usual rant about why HEEO Is Bad for Depots.  But this is one of the big problems.

Quote
The harder problem is the argument of perigee.  There's nothing cheap to be done there.  Since that's what governs where the apse line points, it's a fairly big deal.  You can fiddle with it a bit, but changing it dramatically will cost you.

For circular, or near circular, orbits the argument of perigee doesn't matter.

The vast majority of rendezvous will be between just launched tankers and depots in low or very low circular orbits.  These are the operations you need to optimize and can almost ignore argument of perigee.

If you need additional steps on a refueling ladder then you have to worry about argument of perigee, but it should be easily manageable by gathering one or more depots and the LSS in a low circular orbit and departing either in formation or at appropriate short intervals to match argument of perigee.

If we had an equatorial launch site and could rendezvous in an equatorial orbit we could ignore RAAN as well.

Again, omitting the rant, yes, circular orbits simply don't have this set of problems, and getting the RAANs to match up is a ridiculously well-solved launch and rendezvous problem.

But if depots are in circular VLEOs, then the delta-v budget they can provide for a 1200t Starship is probably too small to be able to do LEO-TLI-BLT-NRHO-LS-NRHO, properly burdened with human-rated flight performance reserves, boil-off, and sump losses, without a refueling somewhere else.  It could be in NRHO (expensive in terms of prop, but fairly forgiving of RPOD oopsies), or in HEEO (Bad!! Insert rant here!).

But if you have 1500t of prop in your LSS, things are pretty good from VLEO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/30/2023 06:12 am
[...]
It's actually the law of cosines, applied to the initial velocity vector (i.e., the orbital speed at apogee), the target velocity vector (i.e., the orbital speed after the perigee's been raised), and the inclination change, which is the angle between the two vectors:

Δvtotal = sqrt( vinit² + vfinal² - 2*vinit*vfinal*cos(Δinclination) )

Thanks for the pointer! This is indeed the textbook answer.

There seems to be a commonly held belief that inclination changes in LEO are prohibitively expensive. Fortunately the rather messy law of cosines cleans up quite nicely in the simple case of a pure inclination change between two circular low-Earth orbits with the same radius. Of note, Δv is proportional to sin(Δi/2), hinting that for small inclination changes the Δv might not be prohibitive. The spreadsheet seems to indicate a Δv of only 355 m/s to get between orbits "natural" of LC-39 and BC when the orbital altitude is 205 km.

355m/s doesn't sound like a lot, but let's look into a delta-v budget to the reference 200x200 LEO with 150t of payload.  (We'll arm-wave average Isp=362s, since all 3 RSLs have to run for quite a ways, and then one RSL has to keep running at reduced throttle to do attitude control.)

Dry mass=120t
Prop mass=1200t.
Isp=362
Payload=150t
EDL Delta-v reserve=150m/s.
7t for ullage and sump losses (computed at 6bar, with no boil-off allowance)

Prop reserved for EDL = (exp( 150/362/9.8 ) - 1)*(120t + 7t) = 5.5t of prop.

Delta-v from staging to LEO = 362*9.8*ln( (120+150+1200) / (120+150+5.5+7)) = 5850m/s.

But now we add 355m/s to that, and we need 6205m/s.  How does that affect our payload x to LEO?

(120+x+1200) / (120+x+5.5+7) = exp( 6205/362/9.8 ) = 5.749
x + 1320 = 5.749x + 761.7
x = 97.1t 117.6t of payload.

Lopping 53t 32t off of each prop delivery to the depot is non-trivial.  It's not something you want to do if you don't have to.

And you don't have to.  There are common inclinations reachable from both BC and LC-39A, and they're all close enough to azimuth 90º for both sites that the rotational delta-v losses are in the low double digits.  You get a lot more prop to the depot that way.

Update:  Removed stupid math error.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 11/30/2023 06:40 am


Prop reserved for EDL = (exp( 150/362/9.8 ) - 1)*(120t + 7t) = 5.5t of prop.

Delta-v from staging to LEO = 362*9.8*ln( (120+150+1200) / (120+150+5.5+7)) = 5850m/s.

But now we add 355m/s to that, and we need 6205m/s.  How does that affect our payload x to LEO?

(120+x+1200) / (120+x+5.5+7) = exp( 6205/362/9.8 ) = 5.749
x + 1320 = 5.749x + 761.7
x = 97.1t of payload.

Lopping 53t off of each prop delivery to the depot is non-trivial.  It's not something you want to do if you don't have to.

And you don't have to.  There are common inclinations reachable from both BC and LC-39A, and they're all close enough to azimuth 90º for both sites that the rotational delta-v losses are in the low double digits.  You get a lot more prop to the depot that way.

The math doesn't seem right, better double check.  I"m getting 117.6t.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 11/30/2023 06:59 am


Prop reserved for EDL = (exp( 150/362/9.8 ) - 1)*(120t + 7t) = 5.5t of prop.

Delta-v from staging to LEO = 362*9.8*ln( (120+150+1200) / (120+150+5.5+7)) = 5850m/s.

But now we add 355m/s to that, and we need 6205m/s.  How does that affect our payload x to LEO?

(120+x+1200) / (120+x+5.5+7) = exp( 6205/362/9.8 ) = 5.749
x + 1320 = 5.749x + 761.7
x = 97.1t 117.6t of payload.

Lopping 53t 32t off of each prop delivery to the depot is non-trivial.  It's not something you want to do if you don't have to.

And you don't have to.  There are common inclinations reachable from both BC and LC-39A, and they're all close enough to azimuth 90º for both sites that the rotational delta-v losses are in the low double digits.  You get a lot more prop to the depot that way.

The math doesn't seem right, better double check.  I"m getting 117.6t.

Yup, you're right.  Forgot to subtract 1 from the mass ratio before dividing by it.  The conclusion still holds, though.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 12/01/2023 02:20 am
approximate southerly ground track in yellow.

Tried to put some math on this. No coriolis, just a great circle calculator (https://edwilliams.org/gccalc.htm).

Skirting just south of Cuba (21.7050° N, 85.1200° W) means launching from Boca Chica at an azimuth of 108.67° (22 m/s loss), reaching an inclination of 31.56°. A modest dogleg might reduce this inclination by ~1.0° or so (eg AsiaSat-8), but let's be conservative and assume no dogleg.

Assuming a final inclination of 31.56°, launches from the Cape would target 104.21° azimuth, incurring a 12.5 m/s loss.


Obviously, a lot smaller than 355 m/s!   8)

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/01/2023 04:56 am
approximate southerly ground track in yellow.

Tried to put some math on this. No coriolis, just a great circle calculator (https://edwilliams.org/gccalc.htm).

Skirting just south of Cuba (21.7050° N, 85.1200° W) means launching from Boca Chica at an azimuth of 108.67° (22 m/s loss), reaching an inclination of 31.56°. A modest dogleg might reduce this inclination by ~1.0° or so (eg AsiaSat-8), but let's be conservative and assume no dogleg.

Assuming a final inclination of 31.56°, launches from the Cape would target 104.21° azimuth, incurring a 12.5 m/s loss.


Obviously, a lot smaller than 355 m/s!   8)

Google Maps or Earth will turn the azimuth into a great circle.  That's not the problem. 

The problem is that you have to compute not only the IIP for each point on the trajectory, but also how long it takes from launch to impact for that point.  Based on that, you know how far east the Earth has rotated at that latitude, and therefore how far west of the azimuth line impact will occur.

I suspect that you're right that there's a dogleg that'll clean this up a bit, but that just takes the already impossible IIP computation problem and makes it trans-impossible.

I'm mostly convinced there's a pony in there somewhere, even though we can't compute it.  And as long as there is, then prop loss to an inclination not very far above 28.6º is tiny.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 12/01/2023 05:20 am
approximate southerly ground track in yellow.

Tried to put some math on this. No coriolis, just a great circle calculator (https://edwilliams.org/gccalc.htm).

Skirting just south of Cuba (21.7050° N, 85.1200° W) means launching from Boca Chica at an azimuth of 108.67° (22 m/s loss), reaching an inclination of 31.56°. A modest dogleg might reduce this inclination by ~1.0° or so (eg AsiaSat-8), but let's be conservative and assume no dogleg.

Assuming a final inclination of 31.56°, launches from the Cape would target 104.21° azimuth, incurring a 12.5 m/s loss.


Obviously, a lot smaller than 355 m/s!   8)

Google Maps or Earth will turn the azimuth into a great circle.  That's not the problem.

Mad overkill, but sure.  ;)

The problem is that you have to compute not only the IIP for each point on the trajectory, but also how long it takes from launch to impact for that point.  Based on that, you know how far east the Earth has rotated at that latitude, and therefore how far west of the azimuth line impact will occur.

I suspect that you're right that there's a dogleg that'll clean this up a bit, but that just takes the already impossible IIP computation problem and makes it trans-impossible.

Eh, that's why I used a simplification.  What's the maximum error this causes, 100 km? So maybe 1/2 a degree when "threading the needle?"

Feel free to run the calc for 108.17 degrees instead. Should be easy, you can even use your fancy Google Earth.  ;)


Until then, it seems like these numbers are close enough to use as baseline assumptions for on-orbit refueling.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/01/2023 05:31 am
What's the maximum error this causes, 100 km? So maybe 1/2 a degree when "threading the needle?"

In the picture I had up-thread, I used sorta-kinda IFT-2 data, plus a bit of arm-wavery near the end for a full trajectory to orbit, and the difference is about 400km at that point in the trajectory.  The IIP for that point will be thousands of km downrange, and the IIP could be >1000km west.

The place that's more likely to cause problems is early in the launch, where the IIP is in the process of racing ahead of the trajectory as downrange velocity climbs.  It's very hard to compute the westerly drift of the IIP then, and it can make that needle-threading exercise impossible without SpaceX-grade data.

BTW, doglegs give the IIP strange, curvy tracks.  I have myself half-convinced that there's a way to have the IIP track curve around the Yucatan and into southern Central America for a Starlink launch with a dogleg that'll take the payload to 54º.  But again, it's too hard to calculate with any confidence without access to the secret sauce.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Twark_Main on 12/01/2023 05:45 am
Until then, it seems like these numbers are close enough to use as baseline assumptions for on-orbit refueling.

Bad News Everyone.

Looks like the Vera Rubin Observatory (https://en.wikipedia.org/wiki/Vera_C._Rubin_Observatory) is at 30 degrees south, so the Starship, Depot, and Tanker orbits will have a high density in that location and will pass directly overhead.

If you thought Starlink was bad, just wait for the astronomical apocalypse Starship depots will cause...  :o
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: yg1968 on 12/04/2023 05:26 pm
https://twitter.com/SpcPlcyOnline/status/1731731958571429944
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: FutureSpaceTourist on 12/04/2023 05:45 pm
https://twitter.com/nasaspaceflight/status/1731741102909038657

Quote
This is cool. It's not the refilling demo HLS has on the books; it'll be Ship 28 transferring prop internally (i.e between header tank and main tank).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/04/2023 06:24 pm
https://twitter.com/nasaspaceflight/status/1731741102909038657

Quote
This is cool. It's not the refilling demo HLS has on the books; it'll be Ship 28 transferring prop internally (i.e between header tank and main tank).
It says "third flight". It does not say "ship 28".    SHIP 26 FOREVER!!!

Whichever ship it is, I do not think this can be done without actually going into a stable orbit. Not enough time.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: wannamoonbase on 12/04/2023 06:58 pm
It says "third flight". It does not say "ship 28".    SHIP 26 FOREVER!!!

Whichever ship it is, I do not think this can be done without actually going into a stable orbit. Not enough time.

This is exciting, I think they will have the time to complete this in less than 1 orbit. 

Depending on how much you're going to transfer this maybe a 10 or 20 min start to finish test.

May as well stack up the test objectives on these vehicles, while figuring out getting to orbit and trouble shooting that heat shield.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/04/2023 07:09 pm
It says "third flight". It does not say "ship 28".    SHIP 26 FOREVER!!!

Whichever ship it is, I do not think this can be done without actually going into a stable orbit. Not enough time.

This is exciting, I think they will have the time to complete this in less than 1 orbit. 

Depending on how much you're going to transfer this maybe a 10 or 20 min start to finish test.

May as well stack up the test objectives on these vehicles, while figuring out getting to orbit and trouble shooting that heat shield.
Perhaps, but I think they should start with the ship in a fairly quiescent state, to emulate a Depot. I'm not sure the ship and the tanks can reach equilibrium that fast. I don't think you get valid test until all those messy interactions between gas phase and liquid phase have stabilized. Then, the ship must do whatever it's designed to do to settle the liquid, generate appropriate ullage, etc, and this may require more time. Finally, they can perform the actual transfer. If they stay in orbit, they can continue to perform more transfers under different conditions to gather more data.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: wannamoonbase on 12/04/2023 07:20 pm
It says "third flight". It does not say "ship 28".    SHIP 26 FOREVER!!!

Whichever ship it is, I do not think this can be done without actually going into a stable orbit. Not enough time.

This is exciting, I think they will have the time to complete this in less than 1 orbit. 

Depending on how much you're going to transfer this maybe a 10 or 20 min start to finish test.

May as well stack up the test objectives on these vehicles, while figuring out getting to orbit and trouble shooting that heat shield.
Perhaps, but I think they should start with the ship in a fairly quiescent state, to emulate a Depot. I'm not sure the ship and the tanks can reach equilibrium that fast. I don't think you get valid test until all those messy interactions between gas phase and liquid phase have stabilized. Then, the ship must do whatever it's designed to do to settle the liquid, generate appropriate ullage, etc, and this may require more time. Finally, they can perform the actual transfer. If they stay in orbit, they can continue to perform more transfers under different conditions to gather more data.

All good points, depends on what the test conditions are.  I was thinking if they just want to transfer a volume.

I agree if in a stable orbit they could do multiple tests over time and characterize behavior.  MY thought was to just get started, then add more complexity once they have a more reliable launch vehicle.  (which I don't think will be too far away)
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: catdlr on 12/04/2023 07:27 pm
It says "third flight". It does not say "ship 28".    SHIP 26 FOREVER!!!

Whichever ship it is, I do not think this can be done without actually going into a stable orbit. Not enough time.

This is exciting, I think they will have the time to complete this in less than 1 orbit. 

Depending on how much you're going to transfer this maybe a 10 or 20 min start to finish test.

May as well stack up the test objectives on these vehicles, while figuring out getting to orbit and trouble shooting that heat shield.
Perhaps, but I think they should start with the ship in a fairly quiescent state, to emulate a Depot. I'm not sure the ship and the tanks can reach equilibrium that fast. I don't think you get valid test until all those messy interactions between gas phase and liquid phase have stabilized. Then, the ship must do whatever it's designed to do to settle the liquid, generate appropriate ullage, etc, and this may require more time. Finally, they can perform the actual transfer. If they stay in orbit, they can continue to perform more transfers under different conditions to gather more data.

All good points, depends on what the test conditions are.  I was thinking if they just want to transfer a volume.

I agree if in a stable orbit they could do multiple tests over time and characterize behavior.  MY thought was to just get started, then add more complexity once they have a more reliable launch vehicle.  (which I don't think will be too far away)

I agree also. seems like they might need more than one orbit and perhaps why they plan to ditch the ship in the Indian Ocean instead.  Unless they have installed empty tanks for transferring liquid oxygen (Lox) and methane (Meth), do you think the first test could be as simple as drawing fuel from the Lox header tank, reversing the process, and refilling it? Ship 28 has been sitting at the rocket installation stand for a long time, and now we understand the reason why. However, we didn't see any additional tank installed in the payload area. That's why I conjecture that the fuel transfer demonstration could test the proposed fuel connection equipment inside the ship and utilize the header tank as the fuel transfer destination, at least for this initial round. They can then make it more robust in a later flight once they have analyzed the captured data to determine what worked and what didn't.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: FutureSpaceTourist on 12/04/2023 07:32 pm
https://twitter.com/thesheetztweetz/status/1731771334135291946

Quote
Video replay of today's NASEM meeting about NASA, with Hawkins' remarks beginning ~1:21:00

https://vimeo.com/event/3924129
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 12/04/2023 07:44 pm
I personally have kind of been expecting a desire to demonstrate a Raptor ignition (or at least compatible conditions) in space before actually going fully orbital. Especially since the RCS might not be able to do a sufficiently targeted deorbit burn.

If they were to go orbital already on the next flight there would have to be a fair bit of confidence in their models and performance.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/04/2023 07:55 pm

All good points, depends on what the test conditions are.  I was thinking if they just want to transfer a volume.

My guess is that they intend to fulfill their non-Artemis, non-HLS contract to do a fuel transfer demo for NASA, so they will meet the contractual requirements. It's not a huge amount of money, but revenue is revenue.
    https://spaceflightnow.com/2020/10/16/nasa-selects-companies-to-demonstrate-in-space-refueling-and-propellant-depot-tech/
Quote
An award to SpaceX worth $53.2 million will go toward a “large-scale flight demonstration to transfer 10 metric tons of cryogenic propellant, specifically liquid oxygen, between tanks on a Starship vehicle,” NASA said.

SpaceX’s Starship is designed to carry more than 100 metric tons of cargo to low Earth orbit. With the docking of a refueling tanker in Earth orbit, SpaceX could refill a Starship with methane and liquid oxygen propellants to fire off to more distant destinations, such as the moon or Mars.

SpaceX will collaborate with Glenn and Marshall on the Starship propellant transfer demo, NASA said.

“When we think about companies like SpaceX and the Starship, their architecture is heavily reliant on the ability to transfer cryogenics in low Earth orbit for the purpose of taking a system all the way to the moon,” Bridenstine said Wednesday. “Their system, in fact, doesn’t appear to require a fuel depot around the moon. Their system would require a fuel depot in orbit around the Earth.”
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: catdlr on 12/04/2023 09:13 pm
https://twitter.com/thesheetztweetz/status/1731771334135291946

Quote
Video replay of today's NASEM meeting about NASA, with Hawkins' remarks beginning ~1:21:00

https://vimeo.com/event/3924129

To save us time, the Cryogenic Fluid Transfer Slide (presentation) comes at 2:51 by John DanKanich.  There are some milestones and dates on that slide (just in year date).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Yggdrasill on 12/05/2023 11:21 am
I personally have kind of been expecting a desire to demonstrate a Raptor ignition (or at least compatible conditions) in space before actually going fully orbital. Especially since the RCS might not be able to do a sufficiently targeted deorbit burn.

If they were to go orbital already on the next flight there would have to be a fair bit of confidence in their models and performance.
The Raptor ignitions on IFT-2 Starship and Super Heavy at ~75 km altitude should derisk the models somewhat.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 12/05/2023 12:05 pm
I personally have kind of been expecting a desire to demonstrate a Raptor ignition (or at least compatible conditions) in space before actually going fully orbital. Especially since the RCS might not be able to do a sufficiently targeted deorbit burn.

If they were to go orbital already on the next flight there would have to be a fair bit of confidence in their models and performance.
The Raptor ignitions on IFT-2 Starship and Super Heavy at ~75 km altitude should derisk the models somewhat.
I was thinking more about managing temperatures and pressures and propellant settling with mostly empty tanks during a prolonged coast in microgravity. Something that the initial ascent does not have a lot of commonality with (by design).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: martiantime on 12/05/2023 01:03 pm
They might use header tanks for deorbit burn. Isn't this simpler than starting the engines from the main tanks, which requires settling the fuel and maintaining pressurization in a large volume?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: rsdavis9 on 12/05/2023 02:02 pm
They might use header tanks for deorbit burn. Isn't this simpler than starting the engines from the main tanks, which requires settling the fuel and maintaining pressurization in a large volume?

Still have to settle the header tank. But much smaller volume requires a lot less pressurization.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: wannamoonbase on 12/05/2023 02:20 pm

All good points, depends on what the test conditions are.  I was thinking if they just want to transfer a volume.

My guess is that they intend to fulfill their non-Artemis, non-HLS contract to do a fuel transfer demo for NASA, so they will meet the contractual requirements. It's not a huge amount of money, but revenue is revenue.
    https://spaceflightnow.com/2020/10/16/nasa-selects-companies-to-demonstrate-in-space-refueling-and-propellant-depot-tech/
Quote
An award to SpaceX worth $53.2 million will go toward a “large-scale flight demonstration to transfer 10 metric tons of cryogenic propellant, specifically liquid oxygen, between tanks on a Starship vehicle,” NASA said.

SpaceX’s Starship is designed to carry more than 100 metric tons of cargo to low Earth orbit. With the docking of a refueling tanker in Earth orbit, SpaceX could refill a Starship with methane and liquid oxygen propellants to fire off to more distant destinations, such as the moon or Mars.

SpaceX will collaborate with Glenn and Marshall on the Starship propellant transfer demo, NASA said.

“When we think about companies like SpaceX and the Starship, their architecture is heavily reliant on the ability to transfer cryogenics in low Earth orbit for the purpose of taking a system all the way to the moon,” Bridenstine said Wednesday. “Their system, in fact, doesn’t appear to require a fuel depot around the moon. Their system would require a fuel depot in orbit around the Earth.”


Reading more about this last night, I agree with you.  Hard to do in one pass.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: StraumliBlight on 12/05/2023 02:29 pm
Reading more about this last night, I agree with you.  Hard to do in one pass.

This possibly outdated 2015 overview (https://www.nasa.gov/wp-content/uploads/2015/03/kortes_perseverancecfm_tagged.pdf) states there will be an orbit insertion.

Quote
• Demonstration conducted as part of a Starship flight
• Orbit insertion, header tank venting
• Coast phase (propellant settling, heating, boil-off rate and propellant stratification data collection)
• Active settling maneuver
• Autogenous pressurization and propellant transfer
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 12/05/2023 02:41 pm
Reading more about this last night, I agree with you.  Hard to do in one pass.

This possibly outdated 2015 overview (https://www.nasa.gov/wp-content/uploads/2015/03/kortes_perseverancecfm_tagged.pdf) states there will be an orbit insertion.

Quote
• Demonstration conducted as part of a Starship flight
• Orbit insertion, header tank venting
• Coast phase (propellant settling, heating, boil-off rate and propellant stratification data collection)
• Active settling maneuver
• Autogenous pressurization and propellant transfer
Thanks!
When I click on the link, it indeed is a PDF with a name that implied it's from 2015. However, the document itself appears to be from Q1 FY 2021 2022, and it uses nomenclature (e.g., "Starship") that was not used in 2015. Look at page 11, and specifically at the timeline at the bottom.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: eriblo on 12/05/2023 02:54 pm
They might use header tanks for deorbit burn. Isn't this simpler than starting the engines from the main tanks, which requires settling the fuel and maintaining pressurization in a large volume?
Still have to settle the header tank. But much smaller volume requires a lot less pressurization.
With a well designed propellant sump and helium (or possibly nitrogen) pressurization in the headers they can avoid a lot of risk with regards to a deorbit burn. The problem is that the benefits of this approach would mostly be negated if you use the header for propellant transfer tests during flight.

It is worth noting that the S26 deorbit burn static fire test had plenty of LOX in and did a lot of interesting venting of the main tank (not conclusive as the headers were also loaded).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 12/05/2023 03:15 pm
https://twitter.com/thesheetztweetz/status/1731771334135291946

Quote
Video replay of today's NASEM meeting about NASA, with Hawkins' remarks beginning ~1:21:00

https://vimeo.com/event/3924129

To save us time, the Cryogenic Fluid Transfer Slide (presentation) comes at 2:51 by John DanKanich.  There are some milestones and dates on that slide (just in year date).
BTW, Lakiesha Hawkins really seems to “get” SpaceX’s strategy to development. At 2:01:17, Lester Lyles kind of pushes back on her characterization of IFT-2 test as “completed” (given it was a /launch/ failure), and she responded explaining how SpaceX conducts development with a hardware-rich, iterative, test-then-fix approach and NASA HLS is embedded, lending NASA’s expertise for the investigation (to help SpaceX make continual progress on each test).
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: martiantime on 12/05/2023 04:45 pm
Quote
• Demonstration conducted as part of a Starship flight
• Orbit insertion, header tank venting
• Coast phase (propellant settling, heating, boil-off rate and propellant stratification data collection)
• Active settling maneuver
• Autogenous pressurization and propellant transfer
Thank you, very interesting.
Does "header tank venting" mean that it will be depleted, and then there will be propellant transfer from the main tank to the header tank? As far as I understand, the engine will be fired, which allows to replenish an autogenous pressurization in the main tank, and then propellant will be squeezed into the header tank by this pressure.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/05/2023 08:53 pm
I heard Dankanich mention "no-vent fill" into a warm tank on the call.  That would seem to indicate that they either think that the flash to gas during the chilldown phase won't overwhelm the pressure differential between the two tanks, or that they're planning on equalizing pressures  between the tanks and using a pump.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: alugobi on 12/06/2023 01:24 am
Ars Technica article.

https://arstechnica.com/space/2023/12/nasa-wants-to-see-gas-stations-in-space-but-so-far-its-tanks-are-empty/
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 12/07/2023 02:40 am
Quote
• Demonstration conducted as part of a Starship flight
• Orbit insertion, header tank venting
• Coast phase (propellant settling, heating, boil-off rate and propellant stratification data collection)
• Active settling maneuver
• Autogenous pressurization and propellant transfer
Thank you, very interesting.
Does "header tank venting" mean that it will be depleted, and then there will be propellant transfer from the main tank to the header tank? As far as I understand, the engine will be fired, which allows to replenish an autogenous pressurization in the main tank, and then propellant will be squeezed into the header tank by this pressure.

A good reason to vent the header tank is to let it warm up.  They seem to be quite concerned--and rightly so--about flowing prop through warm plumbing, into warm tankage.  That's probably the primary test goal.  But watching large-scale venting behavior on both sides of the vent is probably pretty interesting as well.  Otherwise, they could just launch with an empty header tank and have done with it.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: LMT on 01/18/2024 04:59 pm
Potential cross-pollination?

Quote from: Joint Press Release
Historic Cryogenic Refueling Mission in Space! (https://spaceium.com/historic-cryogenic-refueling-mission-in-space/)

January 15, 2024

Space Machines Company and Spaceium are thrilled to announce a groundbreaking partnership, set to embark on the first-ever cryogenic refueling mission in space in 2025. This collaboration marks a significant milestone for the future of the space industry.

Under this agreement, Spaceium will showcase its cutting-edge cryogenic storage capabilities on the Space Machines Company’s platform. Following this demonstration, Spaceium will refuel Space Machines Company spacecraft’s tank with cryogenic fuel using the stored reserves to highlight Spaceium’s refueling capability.

The implications of cryogenic refueling in space are monumental for the future of the space industry and exploration, as exemplified by the Artemis program. This pioneering mission sets the stage for a future of increased efficiency and innovation in space exploration, showcasing the immense benefits this partnership brings to the forefront of the space industry.

“We are thrilled to embark on this groundbreaking journey with Space Machines Company, setting the stage for a new era in space exploration. Together, we are not only demonstrating advanced cryogenic technology but also fueling the future possibilities of space exploration and technology.” Said Ashi Dissanayake, CEO of Spaceium.

“In-space refueling and servicing will change the economics of space infrastructure, with cryogenic refueling essential for earth missions and beyond. We are excited to be partnering with Spaceium to demonstrate these cutting edge capabilities on orbit.” said Mark Ramsey, COO of Space Machines Company.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DistantTemple on 01/19/2024 01:44 am
Potential cross-pollination?

Quote from: Joint Press Release
Historic Cryogenic Refueling Mission in Space! (https://spaceium.com/historic-cryogenic-refueling-mission-in-space/)

January 15, 2024

Space Machines Company and Spaceium are thrilled to announce a groundbreaking partnership, set to embark on the first-ever cryogenic refueling mission in space in 2025. This collaboration marks a significant milestone for the future of the space industry.

Under this agreement, Spaceium will showcase its cutting-edge cryogenic storage capabilities on the Space Machines Company’s platform. Following this demonstration, Spaceium will refuel Space Machines Company spacecraft’s tank with cryogenic fuel using the stored reserves to highlight Spaceium’s refueling capability.

The implications of cryogenic refueling in space are monumental for the future of the space industry and exploration, as exemplified by the Artemis program. This pioneering mission sets the stage for a future of increased efficiency and innovation in space exploration, showcasing the immense benefits this partnership brings to the forefront of the space industry.

“We are thrilled to embark on this groundbreaking journey with Space Machines Company, setting the stage for a new era in space exploration. Together, we are not only demonstrating advanced cryogenic technology but also fueling the future possibilities of space exploration and technology.” Said Ashi Dissanayake, CEO of Spaceium.

“In-space refueling and servicing will change the economics of space infrastructure, with cryogenic refueling essential for earth missions and beyond. We are excited to be partnering with Spaceium to demonstrate these cutting edge capabilities on orbit.” said Mark Ramsey, COO of Space Machines Company.
They just need Blue Origin to come on board to provide the Lawyers to Patent every possible way of in space cryogenic transfer refueling (whether there is prior art or not!). Ideally before SpaceX have actually demonstrated it - the race is on...
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Solarsail on 01/22/2024 06:40 pm
I'm reminded of an idea from a comment on Ycombinator forums ("hacker news")...  could propellant settling and ullage acceleration be provided using small propellers on the sides of the tanks, to spin the entire contents around in a circle?  Essentially a centrifuge...  Any drag induced by friction with tank parts would end up heating the propellants...  But you could get away with a fairly slow rotation, I think.  1 mm/s^2 acceleration on the outer surface only requires 1 rotation per 7 minutes, for a fluid flow speed of 0.24 km/h.  Would that workable, and/or preferable to adding ullage thrusters?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Vettedrmr on 01/23/2024 02:40 am
I think you just end up "stirring the tanks".  Conservation of energy says that any rotational energy you can impart into the contents of the tank will be opposed by rotation of the tank itself, and the friction of the fluid to the tank walls would just end up with a chaotic volume.  Not certain about this, but it's the only thing my brain could resolve to.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: KilroySmith on 01/23/2024 06:47 pm
From a fundamentals point of view, what does LOX/LCH4 look like in a tank in orbit?

Does it coalesce into a giant ball in the middle of the tank?
Does it coalesce into one or more balls attached to a tank surface?
Is it a chaotic mess of droplets floating freely?
Is it a combination of the above?

From a surface tension point of view, it seems like blobs should coalesce when they run into each other - which they'll do as a result of ship accelerations, from engine cut-off to thrusters.  From this point of view, it also seems like they should coalesce on one or more surfaces (but, obviously, not on the surface you'd like them to collect on). 

Anyone seen video of the inside of a partial tank in orbit?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Brigantine on 01/23/2024 08:04 pm
From a fundamentals point of view, what does LOX/LCH4 look like in a tank in orbit?

Does it coalesce into a giant ball in the middle of the tank?
Does it coalesce into one or more balls attached to a tank surface?

Alternative conjecture: Does it spread to cover all the tank walls, bit a big bubble in the middle of the tank?

Also just a note that as things scale up, inertial forces dominate over surface tension / viscous forces, so it will take a very long time to settle into a steady state, and any little RCS correction might be enough to make it turbulent again - What effect does fluid behaviour in half-full tanks have on RCS control? Is a steady state even possible?

In LEO, what is the order of magnitude of tidal forces within the tank? Small for sure, but enough for methane to preferentially coalesce at zenith and nadir? (or cause observable tides in the tank if the Ship has any other spin?)

∆a = μ/(r+∆r)² - μ/r² ≈ ⁻2*∆r*μ/r³ = g₀*(6,371/r)² * ⁻2*∆r/r ≈ -2.8*10⁻⁶ * ∆r N/kg [where ∆r is from Ship CoM]
(For a 200 km orbit. 50% more than that i.e. -4.2*10⁻⁶ * ∆r if the Ship rotates to keep the nose at zenith, from centrifugal forces)

That's... smaller than settling thrust, but not by much. Consider a vertical ship, the sump of the downcomer will get near 10⁻⁵ g

Maybe that's where the "10⁻⁵ g settling thrust" comes from, just enough to overcome tidal forces so fuel doesn't stick to the top of your tank?

I remember far upthread we were discussing tidal settling by tethering 2 ships over a distance

[EDIT: for 500 kg/m³ liquid methane, this can be converted to -2.1 mN/m²
for 1,200 kg/m³ LOX, this can be converted to -5.0 mN/m²
Not sure yet what bearing this has on predicting what methane looks like.
There will also be converging tidal forces in the port-starboard axis, +0.7 mN/m² for methane and +1.7 mN/m² for LOX]
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: InterestedEngineer on 01/23/2024 08:10 pm
From a fundamentals point of view, what does LOX/LCH4 look like in a tank in orbit?

Does it coalesce into a giant ball in the middle of the tank?
Does it coalesce into one or more balls attached to a tank surface?

Alternative conjecture: Does it spread to cover all the tank walls, bit a big bubble in the middle of the tank?

Also just a note that as things scale up, inertial forces dominate over surface tension / viscous forces, so it will take a very long time to settle into a steady state, and any little RCS correction might be enough to make it turbulent again - What effect does fluid behaviour in half-full tanks have on RCS control? Is a steady state even possible?

In LEO, what is the order of magnitude of tidal forces within the tank? Small for sure, but enough for methane to preferentially coalesce at zenith and nadir? (or cause observable tides in the tank if the Ship has any other spin?)

∆a = μ/(r+∆r)² - μ/r² ≈ ⁻2*∆r*μ/r³ = g₀*(6,371/r)² * ⁻2*∆r/r ≈ -2.8*10⁻⁶ * ∆r N/kg [where ∆r is height within the tank]
(For a 200 km orbit. 50% more than that i.e. -4.2*10⁻⁶ * ∆r if the Ship rotates to keep the nose at zenith, from centrifugal forces)

That's... smaller than settling thrust, but not by much. Consider a vertical ship, the sump of the downcomer will get near 10⁻⁵ g

Maybe that's where the "10⁻⁵ g settling thrust" comes from, just enough to overcome tidal forces so fuel doesn't stick to the top of your tank?

I remember far upthread we were discussing tidal settling by tethering 2 ships over a distance


The surface tension of LOX is 0.0132 N/m.

that means along the 9m diameter the force is 119mN.   That sounds like more force than tidal forces
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: CMac on 01/23/2024 08:48 pm
I had been thinking about the centrifugal settling idea.

Have a trough around the circumference: bottom of trough is against the wall and sides are perpendicular to wall. It can be a wide trough with shallower sides. Paddles in the trough would make the fluid stay somewhat stationary.

You spin the ship slowly while shutting down thrust. Some propellant settles into the troughs, some doesn't. A feeder line goes to manifold, fed from a few points around circumference.

Maybe a header tanks is easier though.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: QuantumG on 01/25/2024 06:48 am
Stages that thrust together, settle together.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Eer on 02/24/2024 02:21 pm
Wondering about IM-1 cryo-fuel boil-off lessons learned for SpaceX ... sounds like their approach (manage temperatures, allow pressure buildup in tanks) worked for them ... applicability for SpaceX on-orbit refueling tests and operations? Did SpaceX get affirmation of their general approach?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Emmettvonbrown on 03/16/2024 12:05 pm
I just realized that a tanker launcher could be a reusable Super Heavy topped with an expendable, weight trimmed Starship.

Starting from a 1200 mt propellants Starship to be refueled in orbit, I wanted to check the bare minimum number of tanking flights.

With 250 mt it would need five tanking flights, but 300 mt would cut that to four.

And since Super Heavy turnaround may be as low as three hours, those four or five tanking flights could be squeezed over a single day.

Random basic calculations. Pushing mass fractions to 0.95 and 0.96, and specific impulses to 335 and 370 / 385.

I wanted raw payload numbers. To see whether 250 mt was possible, and even 300 mt.

9.81*335*ln((3600+1400+250)/(180+1400+250))+9.81*375*ln((1400+250)/(70+250))
= 9497 m/s with 250 mt.

9.81*335*ln((3600+1400+300)/(144+1400+300))+9.81*385*ln((1400+300)/(56+300))
= 9374 m/s with 300 mt

To be honest I don't think 300 mt is possible. Pushing isp and mass fraction too far.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: steveleach on 03/16/2024 12:58 pm
I just realized that a tanker launcher could be a reusable Super Heavy topped with an expendable, weight trimmed Starship.

Starting from a 1200 mt propellants Starship to be refueled in orbit, I wanted to check the bare minimum number of tanking flights.

With 250 mt it would need five tanking flights, but 300 mt would cut that to four.

And since Super Heavy turnaround may be as low as three hours, those four or five tanking flights could be squeezed over a single day.

Random basic calculations. Pushing mass fractions to 0.95 and 0.96, and specific impulses to 335 and 370 / 385.

I wanted raw payload numbers. To see whether 250 mt was possible, and even 300 mt.

9.81*335*ln((3600+1400+250)/(180+1400+250))+9.81*375*ln((1400+250)/(70+250))
= 9497 m/s with 250 mt.

9.81*335*ln((3600+1400+300)/(144+1400+300))+9.81*385*ln((1400+300)/(56+300))
= 9374 m/s with 300 mt

To be honest I don't think 300 mt is possible. Pushing isp and mass fraction too far.
Why would this be worth discarding all those ships (and engines)?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 03/16/2024 01:25 pm
I just realized that a tanker launcher could be a reusable Super Heavy topped with an expendable, weight trimmed Starship.
Starting from a 1200 mt propellants Starship to be refueled in orbit, I wanted to check the bare minimum number of tanking flights.
With 250 mt it would need five tanking flights, but 300 mt would cut that to four.
Why would this be worth discarding all those ships (and engines)?
It's a contingency plan. If SS reuse is harder than expected, a really cheap expendable tanker allows Starship HLS to stay on schedule and still remain profitable, in parallel with SpaceX getting reusability working. This scenario assumes they recover the SH (33 engines) and expend only the SS. That's only six Raptors per tanker. Each expended tanker might cost $20M, so $100M for the demo flight and $100M for Artemis III.

It also solves the turnaround problem, if any. You can launch 5 expendable tankers in five days. You would certainly need at least two reusables, and possibly as many as five, to do 5 launches in five days. Fast launch is possibly needed to minimize boil-off at Depot. If you need five reusables, the expendables will be a cheaper alternative for the demo. The reusables don't save any money until they are actually reused, and it's unclear if there s a customer for them after demo until Artemis III.

The expendables may also actually serve a useful long-term purpose, if SpaceX can figure out a way to keep them in formation as a sun shield and Earth sheld for Depot.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: steveleach on 03/16/2024 02:02 pm
I just realized that a tanker launcher could be a reusable Super Heavy topped with an expendable, weight trimmed Starship.
Starting from a 1200 mt propellants Starship to be refueled in orbit, I wanted to check the bare minimum number of tanking flights.
With 250 mt it would need five tanking flights, but 300 mt would cut that to four.
Why would this be worth discarding all those ships (and engines)?
It's a contingency plan. If SS reuse is harder than expected, a really cheap expendable tanker allows Starship HLS to stay on schedule and still remain profitable, in parallel with SpaceX getting reusability working. This scenario assumes they recover the SH (33 engines) and expend only the SS. That's only six Raptors per tanker. Each expended tanker might cost $20M, so $100M for the demo flight and $100M for Artemis III.

It also solves the turnaround problem, if any. You can launch 5 expendable tankers in five days. You would certainly need at least two reusables, and possibly as many as five, to do 5 launches in five days. Fast launch is possibly needed to minimize boil-off at Depot. If you need five reusables, the expendables will be a cheaper alternative for the demo. The reusables don't save any money until they are actually reused, and it's unclear if there s a customer for them after demo until Artemis III.

The expendables may also actually serve a useful long-term purpose, if SpaceX can figure out a way to keep them in formation as a sun shield and Earth sheld for Depot.
Oh, right, I totally agree then. In fact, it still makes sense even if they are not reusing the boosters, as long as they are also using the flights to progress towards reuse.

I'm not convinced they'll deliberately expend ships after they've got reuse working though, apart from a few special cases and end-of-life vehicles.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Emmettvonbrown on 03/16/2024 04:15 pm
DanClemmensen: you nailed it !

Well here is my point of view (for what's its worth). Yes - of course the Super Heavy is reused. And its faster turnaround than Starship (barely 3 hours) really shines here.

That "tanker launcher" looks like a giant Falcon 9 or... New Glenn. First stage reusable, second stage is not but it doesn't matter: it's just a tanker and actually, Starship dirt-cheap steel rings tankage with nothing else around it.

It is a matter of massed-produced Starships and Raptors - which drops the unit cost to nothingness - or close. Once reusable Starship is cheap enough, an expendable tanker strictly derived from it cannot be that expansive.

By "strictly derived" I mean: Starship steel, tankage and Raptors and... nothing else. Very much a giant steel can filled to the brim with methalox propellants. Starship tankage trimmed to the bone to get the mass fraction down: from 0.92 to 0.95 (and yes, 3% makes a difference, since the rocket equation is exponential in nature).
Starting from a 1400 tons Starship, a 0.92 mass fraction brings an empty mass of 112 tons. But 0.95 is 70 tons. And 0.96 : 56 tons.
Also cutting into the numbers of orbital refuelings can't be bad. If SpaceX ever wants to fire a fleet of Starships to Mars within one launch window...

It could be called "tankship" LMAO.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: chariotoffire on 03/17/2024 01:10 am
Another point in favor of expendable tankers is that reuse is really only valuable if you can land at your launch site. If you land at Vandenburg, there's not an economical way to transport the ship. You can pull the Raptors off and truck those over to the Cape or Boca Chica, but it's probably not worth it.

The FAA will need to be very confident Starship will not break up before it allows overland reentries, and even then it may limit the number due to noise concerns.

There is some value to attempting reentry--you can refine the process and hardware, demonstrate reliability to the FAA, and analyze the returned hardware. However, it makes more sense to do this with Starlink launches because 1) Tankers will always be mass constrained; Starship may be volume constrained so some of the reuse mass is free and 2) Boiloff means you want to refill the depot as quickly as possible, so fewer flights is good.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edkyle99 on 03/17/2024 01:37 am
Another point in favor of expendable tankers is that reuse is really only valuable if you can land at your launch site. If you land at Vandenburg, there's not an economical way to transport the ship. You can pull the Raptors off and truck those over to the Cape or Boca Chica, but it's probably not worth it.

The FAA will need to be very confident Starship will not break up before it allows overland reentries, and even then it may limit the number due to noise concerns.
Shuttle did overland reentries.  It weighed about the same as Starship.  I heard a subtle sonic boom from one flying over Chicago on its way to land at KSC one morning (pre-Columbia loss).

 - Ed Kyle
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Robotbeat on 03/17/2024 01:57 am
Another point in favor of expendable tankers is that reuse is really only valuable if you can land at your launch site. If you land at Vandenburg, there's not an economical way to transport the ship. You can pull the Raptors off and truck those over to the Cape or Boca Chica, but it's probably not worth it.
...
It's worth barging the ship. Barging costs less than building a new ship. You can also just launch it again from that site on a different mission and have it come back to the original launch site on another pass.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Negan on 03/17/2024 02:44 am
A tanker is going to include unique docking and fueling systems. It will be interesting to see how its cost compares to other Starship variants even if it's made to be expendable.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 03/17/2024 03:49 am
A tanker is going to include unique docking and fueling systems. It will be interesting to see how its cost compares to other Starship variants even if it's made to be expendable.
Tankers must dock to the Depot. Any SS variant that needs to be refueled must also dock to the Depot. SpaceX may choose to put all of the complexity in the Depot, not the tanker.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Negan on 03/17/2024 04:38 pm
A tanker is going to include unique docking and fueling systems. It will be interesting to see how its cost compares to other Starship variants even if it's made to be expendable.
Tankers must dock to the Depot. Any SS variant that needs to be refueled must also dock to the Depot. SpaceX may choose to put all of the complexity in the Depot, not the tanker.

I see all if not most active systems being on the tankers and the other SS variants that are reusable due to the fact that those systems can be much easier to repair or replace on the ground.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 03/17/2024 05:22 pm
A tanker is going to include unique docking and fueling systems. It will be interesting to see how its cost compares to other Starship variants even if it's made to be expendable.
Tankers must dock to the Depot. Any SS variant that needs to be refueled must also dock to the Depot. SpaceX may choose to put all of the complexity in the Depot, not the tanker.

I see all if not most active systems being on the tankers and the other SS variants that are reusable due to the fact that those systems can be much easier to repair or replace on the ground.
Your speculation may end up being correct. I don't work for SpaceX, so I'm just guessing. Do you work for SpaceX? My Depot-only model is based on the idea that in-orbit refuelling will be very rare for a long time to come. The only instances that are currently funded are the three HLS missions. I'm guessing that the refuelling hardware will occupy volume and will have mass, this eats into the volume and mass budgets of all variants that have it, specifically including Tanker and HLS.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: steveleach on 03/17/2024 05:39 pm
Another point in favor of expendable tankers is that reuse is really only valuable if you can land at your launch site. If you land at Vandenburg, there's not an economical way to transport the ship. You can pull the Raptors off and truck those over to the Cape or Boca Chica, but it's probably not worth it.

The FAA will need to be very confident Starship will not break up before it allows overland reentries, and even then it may limit the number due to noise concerns.

There is some value to attempting reentry--you can refine the process and hardware, demonstrate reliability to the FAA, and analyze the returned hardware. However, it makes more sense to do this with Starlink launches because 1) Tankers will always be mass constrained; Starship may be volume constrained so some of the reuse mass is free and 2) Boiloff means you want to refill the depot as quickly as possible, so fewer flights is good.
You're trying to optimise for a completely different variable than SpaceX are.

The cost of the programme is cost of launch (A) times number of launches (B). You want to reduce that by minimising B. SpaceX want to do it by minimising A.

That's not to say that SpaceX won't try to reduce the number of launches, just that we shouldn't expect them to do that in ways that increase the cost of the launches.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Negan on 03/17/2024 07:38 pm
A tanker is going to include unique docking and fueling systems. It will be interesting to see how its cost compares to other Starship variants even if it's made to be expendable.
Tankers must dock to the Depot. Any SS variant that needs to be refueled must also dock to the Depot. SpaceX may choose to put all of the complexity in the Depot, not the tanker.

I see all if not most active systems being on the tankers and the other SS variants that are reusable due to the fact that those systems can be much easier to repair or replace on the ground.
Your speculation may end up being correct. I don't work for SpaceX, so I'm just guessing. Do you work for SpaceX? My Depot-only model is based on the idea that in-orbit refuelling will be very rare for a long time to come. The only instances that are currently funded are the three HLS missions. I'm guessing that the refuelling hardware will occupy volume and will have mass, this eats into the volume and mass budgets of all variants that have it, specifically including Tanker and HLS.

I'm pretty sure dearMoon will require some in-orbit refueling. Maybe that will just be a one and done.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: chariotoffire on 03/17/2024 08:44 pm
Another point in favor of expendable tankers is that reuse is really only valuable if you can land at your launch site. If you land at Vandenburg, there's not an economical way to transport the ship. You can pull the Raptors off and truck those over to the Cape or Boca Chica, but it's probably not worth it.

The FAA will need to be very confident Starship will not break up before it allows overland reentries, and even then it may limit the number due to noise concerns.

There is some value to attempting reentry--you can refine the process and hardware, demonstrate reliability to the FAA, and analyze the returned hardware. However, it makes more sense to do this with Starlink launches because 1) Tankers will always be mass constrained; Starship may be volume constrained so some of the reuse mass is free and 2) Boiloff means you want to refill the depot as quickly as possible, so fewer flights is good.
You're trying to optimise for a completely different variable than SpaceX are.

The cost of the programme is cost of launch (A) times number of launches (B). You want to reduce that by minimising B. SpaceX want to do it by minimising A.

That's not to say that SpaceX won't try to reduce the number of launches, just that we shouldn't expect them to do that in ways that increase the cost of the launches.

I think SpaceX wants to minimize the cost of the program; that doesn't necessarily mean minimizing A. I agree that reusing tankers is the solution long-term, but I'm not sure that's the case for Artemis 3. If there are constraints on the number of resuable flights, that will tend to favor expendable tankers.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: edzieba on 03/18/2024 11:05 am
Another point in favor of expendable tankers is that reuse is really only valuable if you can land at your launch site. If you land at Vandenburg, there's not an economical way to transport the ship. You can pull the Raptors off and truck those over to the Cape or Boca Chica, but it's probably not worth it.

The FAA will need to be very confident Starship will not break up before it allows overland reentries, and even then it may limit the number due to noise concerns.

There is some value to attempting reentry--you can refine the process and hardware, demonstrate reliability to the FAA, and analyze the returned hardware. However, it makes more sense to do this with Starlink launches because 1) Tankers will always be mass constrained; Starship may be volume constrained so some of the reuse mass is free and 2) Boiloff means you want to refill the depot as quickly as possible, so fewer flights is good.
You're trying to optimise for a completely different variable than SpaceX are.

The cost of the programme is cost of launch (A) times number of launches (B). You want to reduce that by minimising B. SpaceX want to do it by minimising A.

That's not to say that SpaceX won't try to reduce the number of launches, just that we shouldn't expect them to do that in ways that increase the cost of the launches.
That doesn't necessarily minimise total cost, and it also lacks consideration for time (which has both cost in days-per-year, but also cost in ongoing employment). A plan that gets you very cheap per-launch Starship launches but only a handful a year is not going to be preferred over a plan that has slightly more expensive per-launch Starships at several tens per year.

On top of that, you have R&D concerns. For the near term (few years) SpaceX are going to be continuously making changes and upgrades to Starship - just as they did with Falcon 9 (8 years and 53 flights between v1.0 and the first Block 5, plus more years and launches to the current stable-ish Block 5).
Every launch is an opportunity to iterate and test and improve. The more launches, the more tests you can conduct, and the more improvements you can incorporate, sooner. For tankers specifically: if you can strip down a tanker to the bare minimum to get a 5-tanker-prop-load for HLS and expend all the tankers or use close to 'stock' Starships and need to go for a 6-tanker or 7-tanker prop-load but make recovery attempts for each one, you get a lot more bang for your buck.
And potentially some Starships back, but I'd discount this in the near term: for the first few years, I'd expect that by the time a Starship has been built, tested, launched, and (maybe) recovered, it is a hopelessly obsolete design, and may even be incompatible with any boosters other than the sole one it flew on. I expect we'll see quite a lot of recover-then-scrap before we see vehicle reflight.

tl;dr initial optimisation will be for "fly the maximum number of vehicles" over "fly vehicles at the minimum possible cost", 'only' needing to be cheap enough for the company not to fold in the meantime.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 03/18/2024 01:33 pm
Another point in favor of expendable tankers is that reuse is really only valuable if you can land at your launch site. If you land at Vandenburg, there's not an economical way to transport the ship. You can pull the Raptors off and truck those over to the Cape or Boca Chica, but it's probably not worth it.
...
It's worth barging the ship. Barging costs less than building a new ship. You can also just launch it again from that site on a different mission and have it come back to the original launch site on another pass.
If you can only launch eastward from an east coast and you cannot land from the west on an east coast, then you have a problem. I really hope landing from the west becomes acceptable, but it not, there are two possible solutions: operate from an ocean platform or island,  or move the returning SS from a different landing site.

An island or ocean platform will add transport costs for everything. Not just the SS, but also payloads, consumables, workers, etc.   Moving the SS from an acceptable landing site (e.g., a west coast) is a hassle. For rapid reuse, an SS end up more time in seaborne or land transport than in any other phase of the turnaround. The elegant alternative would be to launch the SS into a polar orbit from the west coast and land to back on the east coast. This doubles the launch cost of an east coast mission unless you can find enough paying customers for those polar launches. It also adds some turnaround time, (measured from one east coast launch to the next east coast launch), but not nearly as much as land or sea transport.

A nearly empty SS has a TW greater than 1, so another possibility may be a P2P ferry hop. Launch as usual from the east coast, land from the west on a island or ocean platform in easy reach of the east coast, and then use an SS-only P2P flight to get back home. for KSC, Key West or the Bahamas might work, depending on exactly how strict the prohibitions against overland re-entry are.

If short P2P hops over land are acceptable (i.e., less risk to life), then things get easier. You can hop to KSC from Tampa, and hop to Boca Chica from the Mexican west coast.

Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Vettedrmr on 03/18/2024 03:12 pm
If you can only launch eastward from an east coast and you cannot land from the west on an east coast, then you have a problem. I really hope landing from the west becomes acceptable, but it not, there are two possible solutions: operate from an ocean platform or island,  or move the returning SS from a different landing site.

Quote
A nearly empty SS has a TW greater than 1, so another possibility may be a P2P ferry hop. Launch as usual from the east coast, land from the west on a island or ocean platform in easy reach of the east coast, and then use an SS-only P2P flight to get back home. for KSC, Key West or the Bahamas might work, depending on exactly how strict the prohibitions against overland re-entry are.

I'm having problems reconciling your solution to the problem you stated.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 03/18/2024 03:31 pm
If you can only launch eastward from an east coast and you cannot land from the west on an east coast, then you have a problem. I really hope landing from the west becomes acceptable, but it not, there are two possible solutions: operate from an ocean platform or island,  or move the returning SS from a different landing site.

Quote
A nearly empty SS has a TW greater than 1, so another possibility may be a P2P ferry hop. Launch as usual from the east coast, land from the west on a island or ocean platform in easy reach of the east coast, and then use an SS-only P2P flight to get back home. for KSC, Key West or the Bahamas might work, depending on exactly how strict the prohibitions against overland re-entry are.
The problem: EDL over the US mainland may be restricted or prohibited.
Just looking at a map, it looks like you can EDL to Key West or the Bahamas without passing over much land at dangerous altitude. The SS-only P2P may be viewed as safer than an EDL.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Vettedrmr on 03/18/2024 03:35 pm
I admit I don't know much about the P2P concept, but ISTM that it's essentially a short suborbital flight, with an attendant EDL, although at a lower energy state.  Am I missing something?
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: DanClemmensen on 03/18/2024 04:19 pm
I admit I don't know much about the P2P concept, but ISTM that it's essentially a short suborbital flight, with an attendant EDL, although at a lower energy state.  Am I missing something?
Maybe I am completely wrong, but I think a short SS-Only P2P has a much lower descent speed than an EDL from orbit.
For a P2P from the Bahamas to KSC, the path is not over the mainland. From Key West, you would need a dogleg.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: Vettedrmr on 03/18/2024 04:31 pm
I know this doesn't really have anything to do with refueling, but I wonder if they could land on an ASDS off the west coast to demonstrate landings, and gain confidence in crossing N. American landmasses.
Title: Re: Starship On-orbit refueling - Options and Discussion
Post by: TheRadicalModerate on 03/27/2024 06:46 pm
I admit I don't know much about the P2P concept, but ISTM that it's essentially a short suborbital flight, with an attendant EDL, although at a lower energy state.  Am I missing something?
Maybe I am completely wrong, but I think a short SS-Only P2P has a much lower descent speed than an EDL from orbit.
For a P2P from the Bahamas to KSC, the path is not over the mainland. From Key West, you would need a dogleg.

Two things:

1) I spun up a Safe IIPs for Starship Recovery (https://forum.nasaspaceflight.com/index.php?topic=59970.0) thread a while back.   It's been dormant for a couple of months, but nobody really talked about offshore options.  That would be worthwhile to add.

2) P2P is... weird.  If you're doing purely ballistic P2P, the entry angle for most ranges gets steep enough that it's not good for much other than warheads.  That may be OK for extremely short hops, but I'm not sure it gives you much info about lifting reentries.

I think it's much more likely that P2P gets most of its range from hypersonic skips (i.e., dipping into the atmosphere, intentionally generating too much lift for a full reentry, popping back into space, then repeat until you've lost so much energy that you reenter).  That might be an interesting way to gather a lot of data on the hypersonic part of the EDL.  But if you're going to do it at very low entry speeds, you need to depress the trajectory a lot.