Author Topic: Starship On-orbit refueling - Options and Discussion  (Read 596749 times)

Offline rsdavis9

Re: Starship In-orbit refueling - Options and Discussion
« Reply #1200 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? 
With ELV best efficiency was the paradigm. The new paradigm is reusable, good enough, and commonality of design.
Same engines. Design once. Same vehicle. Design once. Reusable. Build once.

Offline OTV Booster

  • Senior Member
  • *****
  • Posts: 5105
  • Terra is my nation; currently Kansas
  • Liked: 3553
  • Likes Given: 6006
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1201 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.
We are on the cusp of revolutionary access to space. One hallmark of a revolution is that there is a disjuncture through which projections do not work. The thread must be picked up anew and the tapestry of history woven with a fresh pattern.

Offline Greg Hullender

  • Full Member
  • ****
  • Posts: 608
  • Seattle
    • Rocket Stack Rank
  • Liked: 443
  • Likes Given: 338
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1202 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, 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.

Offline sebk

  • Full Member
  • ****
  • Posts: 774
  • Europe
  • Liked: 969
  • Likes Given: 27121
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1203 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.

Offline Greg Hullender

  • Full Member
  • ****
  • Posts: 608
  • Seattle
    • Rocket Stack Rank
  • Liked: 443
  • Likes Given: 338
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1204 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?

Offline sebk

  • Full Member
  • ****
  • Posts: 774
  • Europe
  • Liked: 969
  • Likes Given: 27121
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1205 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.

Offline Greg Hullender

  • Full Member
  • ****
  • Posts: 608
  • Seattle
    • Rocket Stack Rank
  • Liked: 443
  • Likes Given: 338
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1206 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.

Offline daavery

  • Full Member
  • **
  • Posts: 204
  • denver CO
  • Liked: 268
  • Likes Given: 103
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1207 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)

Offline TheRadicalModerate

  • Senior Member
  • *****
  • Posts: 4395
  • Tampa, FL
  • Liked: 3313
  • Likes Given: 639
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1208 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?

Offline daavery

  • Full Member
  • **
  • Posts: 204
  • denver CO
  • Liked: 268
  • Likes Given: 103
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1209 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

Offline TheRadicalModerate

  • Senior Member
  • *****
  • Posts: 4395
  • Tampa, FL
  • Liked: 3313
  • Likes Given: 639
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1210 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.

Offline BT52

  • Full Member
  • *
  • Posts: 139
  • Liked: 59
  • Likes Given: 189
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1211 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.
« Last Edit: 07/27/2022 11:56 pm by BT52 »

Offline OTV Booster

  • Senior Member
  • *****
  • Posts: 5105
  • Terra is my nation; currently Kansas
  • Liked: 3553
  • Likes Given: 6006
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1212 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.
We are on the cusp of revolutionary access to space. One hallmark of a revolution is that there is a disjuncture through which projections do not work. The thread must be picked up anew and the tapestry of history woven with a fresh pattern.

Offline OTV Booster

  • Senior Member
  • *****
  • Posts: 5105
  • Terra is my nation; currently Kansas
  • Liked: 3553
  • Likes Given: 6006
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1213 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.
We are on the cusp of revolutionary access to space. One hallmark of a revolution is that there is a disjuncture through which projections do not work. The thread must be picked up anew and the tapestry of history woven with a fresh pattern.

Offline daavery

  • Full Member
  • **
  • Posts: 204
  • denver CO
  • Liked: 268
  • Likes Given: 103
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1214 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

Offline Paul451

  • Senior Member
  • *****
  • Posts: 3553
  • Australia
  • Liked: 2518
  • Likes Given: 2180
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1215 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.

Offline sebk

  • Full Member
  • ****
  • Posts: 774
  • Europe
  • Liked: 969
  • Likes Given: 27121
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1216 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.

Offline Greg Hullender

  • Full Member
  • ****
  • Posts: 608
  • Seattle
    • Rocket Stack Rank
  • Liked: 443
  • Likes Given: 338
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1217 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? :-)

Offline TheRadicalModerate

  • Senior Member
  • *****
  • Posts: 4395
  • Tampa, FL
  • Liked: 3313
  • Likes Given: 639
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1218 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.
« Last Edit: 07/28/2022 03:50 pm by TheRadicalModerate »

Offline OTV Booster

  • Senior Member
  • *****
  • Posts: 5105
  • Terra is my nation; currently Kansas
  • Liked: 3553
  • Likes Given: 6006
Re: Starship In-orbit refueling - Options and Discussion
« Reply #1219 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.



We are on the cusp of revolutionary access to space. One hallmark of a revolution is that there is a disjuncture through which projections do not work. The thread must be picked up anew and the tapestry of history woven with a fresh pattern.

Tags: HLS 
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1