Starship On-orbit refueling - Options and Discussion

[Greg Hullender]

1) Use some kind of tanker that can act as a temporary depot (which would require being able to deploy and stow a QD gender-bender at various times in the conops), and then return to a direct EDL.

I tend to rule this one out on the grounds that it's a cryogenic-plumbing complication. I was traumatized some years ago by a discussion in this thread about cryogenic plumbing. :-) Accordingly, I think the only scenarios that are realistic are ones where fuel is either transferred to a depo or from a depot, but no other vehicle combinations (i.e. without a depot involved) can do it at all.

2) Boost the depot itself from its VLEO to the higher orbit, fuel the target, and then return propulsively back to the VLEO.

I don't think this makes sense unless the target is already fully fueled, in which case, I think this is the same as your case 3. I guess the difference is that the depot might hold a lot more fuel than the target can hold, so after fueling up the target, they both boost into a little bit higher orbit where the depot transfers the rest of the fuel. Or have I misunderstood what you're getting at here?

3) Place a second depot in the target orbit, then have a tanker move the prop between them, followed by an EDL.  This probably makes sense for refueling in stable lunar orbits (e.g. NRHO or a frozen LLO), but HEEOs are likely to be defined by specific mission requirements, and will move around, in terms of both altitude/energy and inclination.

I wouldn't expect any full depot to be in HEEO alone. A mission vehicle would always rendezvous with a depot in VLEO and then they'd boost together to HEEO. I don't see any scenario where a depot is fueled and then placed in HEEO to wait for a mission craft to rendezvous with it--much less where a depot is actually fueled up in HEEO! In other words, I think depots would "live" in LEO and only make brief excursions to HEEO in the company of a mission vehicle--never alone.

Depots in lunar orbits make sense if NASA allows refueling up there. If they've signed off on this strategy of refueling the mission vehicle in LEO and then topping it off again in HEEO, then I guess they've got over their objection to refueling the mission vehicle more than once, so putting a depot in NRHO or LLO could simplify a lot of things.

1) Do you agree that small delta-v brakings will be fine with a depot?  How large do you think they could get without inflicting any damage or risking aerodynamic instability?

I would think this is already a well-studied topic. But my searches for "aerobraking maximum aerodynamic load" didn't give me what I wanted . . .

5) The depot needs a little bit of prop to lift its perigee at the end of the aerobraking, and it needs some for attitude control and small maneuvers to hit the proper aerobraking entry window each pass.  Does that amount of prop (especially for the window maneuvers) exceed what would be required just to do the propulsive return? 

I wonder if the ullage engine would be enough for minor course corrections. Then you'd only need to fire the main engines to raise perigee at the end.

6) Does the extra heat from the aerobraking increase boil-off enough to make this not worthwhile?

Maybe it's not that bad. This article suggests putting a shuttlecock-shaped "sail" at the front, so you could probably confine most of the heating to that structure. The bigger problem may be boiloff over two months in close orbit. Aerobraking may not work without some sort of cryocooler, at which point you really have to ask if it saves enough fuel to pay for itself vs. the simpler propulsive return. (Although I expect the early depots will all be expended after one use.)

[Greg Hullender]

SpaceX may well be planning a network of depots in a variety of orbits with different energies around different bodies, with tankers constantly moving between them keeping them all supplied.  Ships for cargo and crew would then utilise that depot network, accepting that they won't necessarily have the optimum refuelling orbit as a price to pay for operational efficiency.

At this point, I think the only sensible long-term Earth orbits for depots are in circular VLEO and inclined 28.4°. That puts them in easy reach from Kennedy and Boca Chica. I could see eventually having a depot in NRHO or LLO. A Martian depot in VLMO might make sense eventually--fueled with Martian LOX and LCH4. What other bodies and orbits did you have in mind?

[Greg Hullender]

However, assuming the depot must transfer more than a full tank to the LSS, it would have to refuel several times between the first transfer and the Final Tanking Orbit. That adds loiter time to the LSS. Some version of option 3 solves this problem by having all necessary propellant in orbit before the LSS launches. Regarding Dan's note of depot maneuvers' greater efficiency, perhaps the VLEO depot fills the FTO depot, returns to VLEO, and gets topped up before the LSS launches.

I don't think it makes sense to try to fill up a depot in HEEO nor for a mission vehicle to try to rendezvous with a depot in HEEO. You want the depot in the lowest orbit possible so the tanker flights burn no more fuel than necessary. (Apologies if I've misunderstood you and am telling you stuff you already know!) :-)

I think the plan of record is to fuel up two depots, so if one of them boils off a bit while you're fueling the second one, you send a special tanker flight to top it up. By the time the mission craft rendezvous with the pair of depots, they're both full to the brim. The mission craft fills up from depot #1 and then it and depot #2 take off into HEEO. Between the two of them (the mission craft and the depot) there should be just enough fuel to fill up the mission craft, with a bit left over so the depot can return to VLEO.

That seems to add up to enough fuel for the mission craft to get to NRHO, loiter for a few months, descend to the moon, and return to NRHO. But it also seems to enable all sorts of other interesting scenarios.

[Narnianknight]

I guess the difference is that the depot might hold a lot more fuel than the target can hold, so after fueling up the target, they both boost into a little bit higher orbit where the depot transfers the rest of the fuel.

I think the plan of record is to fuel up two depots, so if one of them boils off a bit while you're fueling the second one, you send a special tanker flight to top it up. By the time the mission craft rendezvous with the pair of depots, they're both full to the brim. The mission craft fills up from depot #1 and then it and depot #2 take off into HEEO. Between the two of them (the mission craft and the depot) there should be just enough fuel to fill up the mission craft, with a bit left over so the depot can return to VLEO.

Both of these ideas do sound better than the profile I suggested. There seems to have been at least some idea of a depot significantly larger than a standard ship, so the former theory may hold water. I would see that as a desirable option.

[OTV Booster]

Depots would be optimised to stay in roughly the same place for a long time, with tankers visiting them to move propellants around. If they need to change their orbits they will do it entirely with engines. I can imagine challenging mission requirements utilising single-use depots, as the cost of moving them is more than the cost of the depot. Maybe they'll get solar-electric propulsion later on?

Tankers would be optimised for moving propellant to, and between, depots, and for returning to the launch site. They would be able to aerobrake aggressively if returning to a LEO depot is required, as well as EDL to launch site from any orbit.

SpaceX may well be planning a network of depots in a variety of orbits with different energies around different bodies, with tankers constantly moving between them keeping them all supplied.  Ships for cargo and crew would then utilise that depot network, accepting that they won't necessarily have the optimum refuelling orbit as a price to pay for operational efficiency.

I think we have been misleading ourselves based on the names "Tanker" and "Depot".  Tanker is optimized for EDL. This capability adds a lot of dry mass.  Depot is optimized to contain as much propellant as is feasible and stay in space for a long time. But when you are in space, you can move from place to place easily and efficiently, while the name "Depot" makes you think of a fixed location. To efficiently move a large quantity of fuel, you should use Tankers to fill Depots in VLEO, and then move the entire Depot to its destination. If needed, use additional Depots to fill the first Depot after it arrives. Using Tankers is less efficient because they must move more dry mass. Except possibly for edge cases, sending a Tanker beyond VLEO is a waste of fuel. You can mentally rename the ships "fuel lifter" and "fuel transporter" to help clarify the situation.

When designing missions, you can pretty much always send a Depot together with the mission Ship. Think of the Depot as a tank extender for the Ship. Instead of thinking about where the ship will rendezvous with Depot, think of where the extended Ship/Depot must go. You can also think of Depot as a third stage. Note that you are still flying the same missions, but you are thinking about them differently. You leave Depot with enough fuel to recover itself, just as you leave fuel in Booster for the boost-back or leave Tanker with enough fuel to EDL.

Unless the QD becomes androgynous, depot to depot transfer won't happen. I think you're on the right track with "fuel lifter" and "fuel transporter". Yet another variant. A tanker that moves propellant from one depot, presumably LEO, to another depot at higher energy. HEO or lunar.


The "orbital tanker" would have normal tanker QD, no EDL hardware and little if any specialized depot hardware. It stays up there, the lightest of the bunch. Once it's loaded it would head out. It doesn't hurt to start positioning early and it definitely helps to get away from Earths heat.

[TheRadicalModerate]

I think we have been misleading ourselves based on the names "Tanker" and "Depot".  Tanker is optimized for EDL. This capability adds a lot of dry mass.  Depot is optimized to contain as much propellant as is feasible and stay in space for a long time. But when you are in space, you can move from place to place easily and efficiently, while the name "Depot" makes you think of a fixed location. To efficiently move a large quantity of fuel, you should use Tankers to fill Depots in VLEO, and then move the entire Depot to its destination. If needed, use additional Depots to fill the first Depot after it arrives. Using Tankers is less efficient because they must move more dry mass. Except possibly for edge cases, sending a Tanker beyond VLEO is a waste of fuel. You can mentally rename the ships "fuel lifter" and "fuel transporter" to help clarify the situation.

FWIW, for a depot that's v3-based,¹ here are some prop consumptions to get back to VLEO (200x200) propulsively:

NRHO-VLEO (via reverse BLT): 230t
NRHO-VLEO (fast): 270t
LLO-VLEO (via reverse BLT):  290t
LLO-VELO (fast): 330t
HEEO(LEO+2400)-VLEO: 150t
HEEO(LEO+2400) to 10º plane change: 20t
HEEO(LEO+2400) with 15º apse line rotation: 10t

Some of those numbers are big enough to warrant playing around with aerobraking.

Note that "fuel transporter" must be a synonym for "depot", because otherwise you have a QD-gendering issue.  Depots need to be able to shuttle prop directly, or you need to go through an annoying depot-shuttle-depot conversion to make the prop available to a target Starship.  So I'd be tempted to ensure that depots are aerobrake-capable and have done with it.

When designing missions, you can pretty much always send a Depot together with the mission Ship.

Not quite, because you really don't want to commit a payload mission until you know that the depot (or shuttle) supporting it is happy and healthy in the staging orbit.  That may be trivial if you're staging from HEEO, but if you need a lot of prop in lunar orbit, you may want to go through a BLT to save prop (120t saved, for a fully-fueled prop shuttle).  That means that the payload mission and the prop transport may be separated by as much as 2-3 months.

__________
¹I've been assuming that a v3-based tanker will have the forward dome pushed forward 3 ring segments, which makes it carry 2612t of subcooled prop for the same dry mass, and the depot is pushed forward 4 ring segments, which makes it carry 2534t of boiling prop (2716t of subcooled, which is probably irrelevant).  Note that tankers want to launch full, while depots don't have to.

[TheRadicalModerate]

At this point, I think the only sensible long-term Earth orbits for depots are in circular VLEO and inclined 28.4°. That puts them in easy reach from Kennedy and Boca Chica. I could see eventually having a depot in NRHO or LLO. A Martian depot in VLMO might make sense eventually--fueled with Martian LOX and LCH4. What other bodies and orbits did you have in mind?

Mostly agree, but with a nitpick:  28.5º isn't accessible from BC without a dogleg.  But I think that 30.5º is directly accessible by flying in between Cuba and Jamaica.  At the very least, any required dogleg is tiny.

[TheRadicalModerate]

However, assuming the depot must transfer more than a full tank to the LSS, it would have to refuel several times between the first transfer and the Final Tanking Orbit. That adds loiter time to the LSS. Some version of option 3 solves this problem by having all necessary propellant in orbit before the LSS launches. Regarding Dan's note of depot maneuvers' greater efficiency, perhaps the VLEO depot fills the FTO depot, returns to VLEO, and gets topped up before the LSS launches.

The kinds of missions that need complete refueling in lunar orbits or lunar-distance HEEOs are likely StarKicker missions.  If you want to send a 50t probe to Titan in five years, this is awesome.  Otherwise, I don't think it's something you optimize for.

If you want to go fully down the rabbit-hole on things like this, search for Twark's user ID and the term "laddering".  Don't say I didn't warn you.

[steveleach]

Depots would be optimised to stay in roughly the same place for a long time, with tankers visiting them to move propellants around. If they need to change their orbits they will do it entirely with engines. I can imagine challenging mission requirements utilising single-use depots, as the cost of moving them is more than the cost of the depot. Maybe they'll get solar-electric propulsion later on?

Tankers would be optimised for moving propellant to, and between, depots, and for returning to the launch site. They would be able to aerobrake aggressively if returning to a LEO depot is required, as well as EDL to launch site from any orbit.

SpaceX may well be planning a network of depots in a variety of orbits with different energies around different bodies, with tankers constantly moving between them keeping them all supplied.  Ships for cargo and crew would then utilise that depot network, accepting that they won't necessarily have the optimum refuelling orbit as a price to pay for operational efficiency.

I think we have been misleading ourselves based on the names "Tanker" and "Depot".  Tanker is optimized for EDL. This capability adds a lot of dry mass.  Depot is optimized to contain as much propellant as is feasible and stay in space for a long time. But when you are in space, you can move from place to place easily and efficiently, while the name "Depot" makes you think of a fixed location. To efficiently move a large quantity of fuel, you should use Tankers to fill Depots in VLEO, and then move the entire Depot to its destination. If needed, use additional Depots to fill the first Depot after it arrives. Using Tankers is less efficient because they must move more dry mass. Except possibly for edge cases, sending a Tanker beyond VLEO is a waste of fuel. You can mentally rename the ships "fuel lifter" and "fuel transporter" to help clarify the situation.

When designing missions, you can pretty much always send a Depot together with the mission Ship. Think of the Depot as a tank extender for the Ship. Instead of thinking about where the ship will rendezvous with Depot, think of where the extended Ship/Depot must go. You can also think of Depot as a third stage. Note that you are still flying the same missions, but you are thinking about them differently. You leave Depot with enough fuel to recover itself, just as you leave fuel in Booster for the boost-back or leave Tanker with enough fuel to EDL.

That's not really the thinking behind my distinction, tbh. Depots happen to stay in roughly the same place, but that isn't what makes them a depot.

To me, the distinction is that tankers come home, depots don't. Coming home means EDL, which means heat shield and flaps. The advantage of aerobraking is just so large that it is rarely worth doing anything apart from return to Earth after dropping the propellants off at whatever depot needs it.

If you move propellant to a higher energy orbit in a depot, you either leave that depot there once it is empty, or you bring it back to the lower energy orbit. It is the bringing it back that is challenging. My argument is that it isn't worth the challenge; just pay the mass penalty for heatshield and flaps, and bring the ship all the way home to the surface.

[DanClemmensen]

Depots would be optimised to stay in roughly the same place for a long time, with tankers visiting them to move propellants around. If they need to change their orbits they will do it entirely with engines. I can imagine challenging mission requirements utilising single-use depots, as the cost of moving them is more than the cost of the depot. Maybe they'll get solar-electric propulsion later on?

Tankers would be optimised for moving propellant to, and between, depots, and for returning to the launch site. They would be able to aerobrake aggressively if returning to a LEO depot is required, as well as EDL to launch site from any orbit.

SpaceX may well be planning a network of depots in a variety of orbits with different energies around different bodies, with tankers constantly moving between them keeping them all supplied.  Ships for cargo and crew would then utilise that depot network, accepting that they won't necessarily have the optimum refuelling orbit as a price to pay for operational efficiency.

I think we have been misleading ourselves based on the names "Tanker" and "Depot".  Tanker is optimized for EDL. This capability adds a lot of dry mass.  Depot is optimized to contain as much propellant as is feasible and stay in space for a long time. But when you are in space, you can move from place to place easily and efficiently, while the name "Depot" makes you think of a fixed location. To efficiently move a large quantity of fuel, you should use Tankers to fill Depots in VLEO, and then move the entire Depot to its destination. If needed, use additional Depots to fill the first Depot after it arrives. Using Tankers is less efficient because they must move more dry mass. Except possibly for edge cases, sending a Tanker beyond VLEO is a waste of fuel. You can mentally rename the ships "fuel lifter" and "fuel transporter" to help clarify the situation.

When designing missions, you can pretty much always send a Depot together with the mission Ship. Think of the Depot as a tank extender for the Ship. Instead of thinking about where the ship will rendezvous with Depot, think of where the extended Ship/Depot must go. You can also think of Depot as a third stage. Note that you are still flying the same missions, but you are thinking about them differently. You leave Depot with enough fuel to recover itself, just as you leave fuel in Booster for the boost-back or leave Tanker with enough fuel to EDL.

Unless the QD becomes androgynous, depot to depot transfer won't happen. I think you're on the right track with "fuel lifter" and "fuel transporter". Yet another variant. A tanker that moves propellant from one depot, presumably LEO, to another depot at higher energy. HEO or lunar.


The "orbital tanker" would have normal tanker QD, no EDL hardware and little if any specialized depot hardware. It stays up there, the lightest of the bunch. Once it's loaded it would head out. It doesn't hurt to start positioning early and it definitely helps to get away from Earths heat.

Since we have not yet seen the Depot, we do not know what its plumbing and docking hardware looks like.  The simplest approach may be to leave the QD plate as-is, and add a separate SQD arm at the front. If so, the aft end is like any other ship and can act as a target for another Depot.

I'm convinced by arguments here that "gentle" aerobraking makes sense, and that a Depot (or almost any spacecraft) can use it with at most minor modifications. Yes, it will take awhile. This just means the Depot fleet must be larger.

[OTV Booster]

Depots would be optimised to stay in roughly the same place for a long time, with tankers visiting them to move propellants around. If they need to change their orbits they will do it entirely with engines. I can imagine challenging mission requirements utilising single-use depots, as the cost of moving them is more than the cost of the depot. Maybe they'll get solar-electric propulsion later on?

Tankers would be optimised for moving propellant to, and between, depots, and for returning to the launch site. They would be able to aerobrake aggressively if returning to a LEO depot is required, as well as EDL to launch site from any orbit.

SpaceX may well be planning a network of depots in a variety of orbits with different energies around different bodies, with tankers constantly moving between them keeping them all supplied.  Ships for cargo and crew would then utilise that depot network, accepting that they won't necessarily have the optimum refuelling orbit as a price to pay for operational efficiency.

I think we have been misleading ourselves based on the names "Tanker" and "Depot".  Tanker is optimized for EDL. This capability adds a lot of dry mass.  Depot is optimized to contain as much propellant as is feasible and stay in space for a long time. But when you are in space, you can move from place to place easily and efficiently, while the name "Depot" makes you think of a fixed location. To efficiently move a large quantity of fuel, you should use Tankers to fill Depots in VLEO, and then move the entire Depot to its destination. If needed, use additional Depots to fill the first Depot after it arrives. Using Tankers is less efficient because they must move more dry mass. Except possibly for edge cases, sending a Tanker beyond VLEO is a waste of fuel. You can mentally rename the ships "fuel lifter" and "fuel transporter" to help clarify the situation.

When designing missions, you can pretty much always send a Depot together with the mission Ship. Think of the Depot as a tank extender for the Ship. Instead of thinking about where the ship will rendezvous with Depot, think of where the extended Ship/Depot must go. You can also think of Depot as a third stage. Note that you are still flying the same missions, but you are thinking about them differently. You leave Depot with enough fuel to recover itself, just as you leave fuel in Booster for the boost-back or leave Tanker with enough fuel to EDL.

That's not really the thinking behind my distinction, tbh. Depots happen to stay in roughly the same place, but that isn't what makes them a depot.

To me, the distinction is that tankers come home, depots don't. Coming home means EDL, which means heat shield and flaps. The advantage of aerobraking is just so large that it is rarely worth doing anything apart from return to Earth after dropping the propellants off at whatever depot needs it.

If you move propellant to a higher energy orbit in a depot, you either leave that depot there once it is empty, or you bring it back to the lower energy orbit. It is the bringing it back that is challenging. My argument is that it isn't worth the challenge; just pay the mass penalty for heatshield and flaps, and bring the ship all the way home to the surface.

Let me suggest a different way to split this hair. A depot accumulates and stores propellant. A tanker delivers propellant to a depot. This focuses on core function rather than physical properties or ancillary function that furthers the core function.


To restate what I think we mostly agree on, VLEO is good for filling a depot but bad for maintaining a depot. It will take less propellant to raise a depot from VLEO to LEO than to raise multiple tankers past VLEO to LEO. OTOH, thermal input from earth and unpredictable atmospheric drag make VLEO a bad place for a depot to loiter longer than necessary.


Another point we mostly agree on is a lunar mission will need to retank at least one more time somewhere above LEO. There are arguments for HEO, NRHO and LLO. Assuming no major changes to the QD arrangement (a weak assumption but consistent with all current evidence) these higher depots will most likely need servicing by tankers. On this hinge the current discussion about what to do with these tankers. Ditch em or retrieve em?


A historical digression. In 1919 it took Dwight Eisenhower 8 weeks to move a truck convoy from New York to San Francisco. The roads sucked, there wasn't a network of fuel stops, repair facilities or restaurants. Today a reasonably caffeinated person can easily do this in six days. Five if they push a bit, four if they push hard and less if chemically augmented (not recommended).


The major difference is infrastructure. We got it and they didn't. I suspect that Musk is aware of this lesson.


Early depots will be throwaway but building infrastructure is always a good investment if the amount of use justifies it. Musk is expecting heavy use in the long run so we should ultimately expect to see depots as long term investments and not thrown away if they're still physically capable of performing their task.


Short term, a depot at one of the high energy locations will probably be ditched, but what about the tankers that service them? It's easiest to strip off the tankers EDL hardware to maximize propellant payload, then discard after use. But if we think about infrastructure, that's the loss of a valuable resource. As long as they work, why not use them?


HEO seems a no brainer. A stripped down tanker servicing depots here could lower perigee for aerobraking to a lower apogee targeting LEO, then a small burn to raise perigee to LEO and hook up with a LEO depot for another load to a high orbit depot. It's target needn't be HEO.


NRHO and LLO depot servicing tankers would probably be throwaways until there's lunar propellant ISRU. That is infrastructure and infrastructure always makes it easier.

[DanClemmensen]

Let me suggest a different way to split this hair. A depot accumulates and stores propellant. A tanker delivers propellant to a depot. This focuses on core function rather than physical properties or ancillary function that furthers the core function.


To restate what I think we mostly agree on, VLEO is good for filling a depot but bad for maintaining a depot. It will take less propellant to raise a depot from VLEO to LEO than to raise multiple tankers past VLEO to LEO. OTOH, thermal input from earth and unpredictable atmospheric drag make VLEO a bad place for a depot to loiter longer than necessary.

Another point we mostly agree on is a lunar mission will need to retank at least one more time somewhere above LEO. There are arguments for HEO, NRHO and LLO. Assuming no major changes to the QD arrangement (a weak assumption but consistent with all current evidence) these higher depots will most likely need servicing by tankers.

If one Depot of fuel is not enough, don't send more tankers. Send another Depot. HLS can suck fuel from two Depots.

[OTV Booster]

Let me suggest a different way to split this hair. A depot accumulates and stores propellant. A tanker delivers propellant to a depot. This focuses on core function rather than physical properties or ancillary function that furthers the core function.


To restate what I think we mostly agree on, VLEO is good for filling a depot but bad for maintaining a depot. It will take less propellant to raise a depot from VLEO to LEO than to raise multiple tankers past VLEO to LEO. OTOH, thermal input from earth and unpredictable atmospheric drag make VLEO a bad place for a depot to loiter longer than necessary.

Another point we mostly agree on is a lunar mission will need to retank at least one more time somewhere above LEO. There are arguments for HEO, NRHO and LLO. Assuming no major changes to the QD arrangement (a weak assumption but consistent with all current evidence) these higher depots will most likely need servicing by tankers.

If one Depot of fuel is not enough, don't send more tankers. Send another Depot. HLS can suck fuel from two Depots.

Nothings cast in stone but my impression is that a depot will hold more than the ship can take on. Ok so the second depot, filled at VLEO can most likely make it to HEO with a useful amount of propellant.


Ahh, now I get it. From there it drops back down and aerobrakes to LEO OR VLEO for reuse - and potentially stays a part of the infrastructure.


Something to look at: in the long term would it be more propellant or operationally efficient for the depot to stay at HEO and be serviced by light weight stripped down non EDL tankers that tank up at LEO?

[DanClemmensen]

Let me suggest a different way to split this hair. A depot accumulates and stores propellant. A tanker delivers propellant to a depot. This focuses on core function rather than physical properties or ancillary function that furthers the core function.


To restate what I think we mostly agree on, VLEO is good for filling a depot but bad for maintaining a depot. It will take less propellant to raise a depot from VLEO to LEO than to raise multiple tankers past VLEO to LEO. OTOH, thermal input from earth and unpredictable atmospheric drag make VLEO a bad place for a depot to loiter longer than necessary.

Another point we mostly agree on is a lunar mission will need to retank at least one more time somewhere above LEO. There are arguments for HEO, NRHO and LLO. Assuming no major changes to the QD arrangement (a weak assumption but consistent with all current evidence) these higher depots will most likely need servicing by tankers.

If one Depot of fuel is not enough, don't send more tankers. Send another Depot. HLS can suck fuel from two Depots.

Nothings cast in stone but my impression is that a depot will hold more than the ship can take on. Ok so the second depot, filled at VLEO can most likely make it to HEO with a useful amount of propellant.


Ahh, now I get it. From there it drops back down and aerobrakes to LEO OR VLEO for reuse - and potentially stays a part of the infrastructure.


Something to look at: in the long term would it be more propellant or operationally efficient for the depot to stay at HEO and be serviced by light weight stripped down non EDL tankers that tank up at LEO?

No. Just send up a full one and send back the empty. A Depot ought to have a fairly low dry mass even with its refuelling gear. A Depot is already a "stripped-down non-EDL tanker", except it has really big tanks. To my mind, it's all about the ratio of propellant mass to dry mass.

[Narnianknight]

However, assuming the depot must transfer more than a full tank to the LSS, it would have to refuel several times between the first transfer and the Final Tanking Orbit. That adds loiter time to the LSS. Some version of option 3 solves this problem by having all necessary propellant in orbit before the LSS launches. Regarding Dan's note of depot maneuvers' greater efficiency, perhaps the VLEO depot fills the FTO depot, returns to VLEO, and gets topped up before the LSS launches.

The kinds of missions that need complete refueling in lunar orbits or lunar-distance HEEOs are likely StarKicker missions.  If you want to send a 50t probe to Titan in five years, this is awesome.  Otherwise, I don't think it's something you optimize for.

If you want to go fully down the rabbit-hole on things like this, search for Twark's user ID and the term "laddering".  Don't say I didn't warn you.

I'm not sure what you're responding to. I didn't mention lunar orbits or lunar-distance HEEOs. I'm talking about the FTO mentioned in the FCC filing. If you just mean depots having enough prop even after a transfer burn from VLEO to FTO rather than being completely filled, I agree.

[steveleach]

Depots would be optimised to stay in roughly the same place for a long time, with tankers visiting them to move propellants around. If they need to change their orbits they will do it entirely with engines. I can imagine challenging mission requirements utilising single-use depots, as the cost of moving them is more than the cost of the depot. Maybe they'll get solar-electric propulsion later on?

Tankers would be optimised for moving propellant to, and between, depots, and for returning to the launch site. They would be able to aerobrake aggressively if returning to a LEO depot is required, as well as EDL to launch site from any orbit.

SpaceX may well be planning a network of depots in a variety of orbits with different energies around different bodies, with tankers constantly moving between them keeping them all supplied.  Ships for cargo and crew would then utilise that depot network, accepting that they won't necessarily have the optimum refuelling orbit as a price to pay for operational efficiency.

I think we have been misleading ourselves based on the names "Tanker" and "Depot".  Tanker is optimized for EDL. This capability adds a lot of dry mass.  Depot is optimized to contain as much propellant as is feasible and stay in space for a long time. But when you are in space, you can move from place to place easily and efficiently, while the name "Depot" makes you think of a fixed location. To efficiently move a large quantity of fuel, you should use Tankers to fill Depots in VLEO, and then move the entire Depot to its destination. If needed, use additional Depots to fill the first Depot after it arrives. Using Tankers is less efficient because they must move more dry mass. Except possibly for edge cases, sending a Tanker beyond VLEO is a waste of fuel. You can mentally rename the ships "fuel lifter" and "fuel transporter" to help clarify the situation.

When designing missions, you can pretty much always send a Depot together with the mission Ship. Think of the Depot as a tank extender for the Ship. Instead of thinking about where the ship will rendezvous with Depot, think of where the extended Ship/Depot must go. You can also think of Depot as a third stage. Note that you are still flying the same missions, but you are thinking about them differently. You leave Depot with enough fuel to recover itself, just as you leave fuel in Booster for the boost-back or leave Tanker with enough fuel to EDL.

Unless the QD becomes androgynous, depot to depot transfer won't happen. I think you're on the right track with "fuel lifter" and "fuel transporter". Yet another variant. A tanker that moves propellant from one depot, presumably LEO, to another depot at higher energy. HEO or lunar.


The "orbital tanker" would have normal tanker QD, no EDL hardware and little if any specialized depot hardware. It stays up there, the lightest of the bunch. Once it's loaded it would head out. It doesn't hurt to start positioning early and it definitely helps to get away from Earths heat.

Yeah, but is another variant actually needed? Feels like we're trying to optimise the outbound leg rather than the full round-trip. Without EDL, orbital tankers have to use fuel to return to the lower orbit, spend a long time (unutilised) aerobraking, or just be discarded.

Just have tankers move prop to LEO depot until it has almost enough to fill a tanker. Then another tanker fills up from LEO depot, then moves it all to HEO (or lunar) depot, then returns to Earth. Repeat this until the HEO depot is full.

 

[Greg Hullender]

HEO seems a no brainer. A stripped down tanker servicing depots here could lower perigee for aerobraking to a lower apogee targeting LEO, then a small burn to raise perigee to LEO and hook up with a LEO depot for another load to a high orbit depot. It's target needn't be HEO.

Just have tankers move prop to LEO depot until it has almost enough to fill a tanker. Then another tanker fills up from LEO depot, then moves it all to HEO (or lunar) depot, then returns to Earth. Repeat this until the HEO depot is full.

The trouble with having a pair of depots, one in VLEO the other in HEO, with a dedicated tanker shuttling between them is that they'll precess at different rates, so windows when you can lift fuel from the lower one to the higher one are going to be rare. If the HEO is at 900 km, the orbits will only align twice a year. If HEO is at geosynchronous orbit, they'll align once every six weeks.

I like the idea of a depot in NRHO (or other lunar orbit) that gets serviced by dedicated tankers from VLEO, but the problem with that is that we've got no way to service those tankers unless they EDL. That removes most of the value of having dedicated tankers. The last time we did this calculation, if I remember correctly, we figured that a standard V1 tanker could carry 1/3 of the propellant from a VLEO depot to LRHO and still return to Earth for EDL, so maybe it's not so bad.

A further implication of the "there is no infrastructure for orbital maintenance" problem is that any depot can only be counted on to boost itself out of VLEO one time. Maybe the Raptors are more robust than I'm giving them credit for, but I think it's a good question just how many orbital restarts can be counted on without any servicing whatsoever.

Once there's enough traffic to justify an "orbital drydock," a lot more things become possible.

[Twark_Main]

Yeah, but is another variant actually needed? Feels like we're trying to optimise the outbound leg rather than the full round-trip. Without EDL, orbital tankers have to use fuel to return to the lower orbit, spend a long time (unutilised) aerobraking, or just be discarded.

This. A tanker's extra dry mass is only "a waste of fuel" if you don't care about reusability.

Just have tankers move prop to LEO depot until it has almost enough to fill a tanker. Then another tanker fills up from LEO depot, then moves it all to HEO (or lunar) depot, then returns to Earth. Repeat this until the HEO depot is full.

This, except no HEO depot. You just directly transfer propellant into whatever ship you were going to fill up from the depot. The tanker raises its orbit "just in time" so there's no precession problem, and it immediately returns to Earth afterward.

If the QD needs to be androgynous, make it androgynous. Or use a gender adapter. Personally I prefer the former ("the best part is no part").

[yg1968]

Low Earth Orbit. SpaceX will conduct a range of Starship operations in low-Earth orbit (“LEO”). Each fully reusable Starship spacecraft is capable of carrying up to 150 metric tons to Earth orbit, and this authorization will enable SpaceX to reliably launch and deploy satellites to support broadband, mobile connectivity, earth observation, science, and other use cases that will benefit humanity. Missions beyond LEO will also require a tanker version of Starship for propellant aggregation. During these missions, SpaceX will launch one or more propellant tanker versions of Starship. Some of these tanker variants will remain in LEO as “depots,” and will be filled with propellant by subsequent tanker launches. LEO operations will occur in a circular orbit at 281 km altitude (+/- 100 km) and an inclination ranging from equatorial (0 degrees) to polar.

Medium-Earth Orbit/High-Earth Orbit/Final Tanking Orbit. Missions beyond LEO will also require space station operations in medium-Earth orbit (“MEO”) to high-Earth orbit (“HEO”). For example, crewed lunar missions will include a secondary propellant transfer in MEO/HEO, the Final Tanking Orbit (“FTO”). Operations in MEO/HEO will occur in an elliptical orbit of 281km x 34,534 km and an altitude tolerance of +116,000/-24,000 km apogee and +/- 100 km perigee, with inclination between 28 and 33 degrees (+/- 2 degrees).

Attachments to the Application (see the Technical Annex attachment):
https://licensing.fcc.gov/cgi-bin/ws.exe/prod/ib/forms/attachment_menu.hts?id_app_num=154189&acct=599269&id_form_num=12&filing_key=-519848

Application:
https://licensing.fcc.gov/cgi-bin/ws.exe/prod/ib/forms/reports/swr031b.hts?q_set=V_SITE_ANTENNA_FREQ.file_numberC/File+Number/%3D/SATLOA2024121800288&prepare=&column=V_SITE_ANTENNA_FREQ.file_numberC/File+Number

[TheRadicalModerate]

I think a single shuttling depot is going to always beat out a shuttling tanker between two depots, as long as the depot can aerobrake¹ back into VLEO.  The aerobraking maneuver takes anywhere from a few days to a few months, depending on how aggressively you reduce delta-v per pass, but lunar missions are only going to happen every few months anyway.  Note the following:

1) Shuttling tankers (probably) have higher dry masses than shuttling depots.

2) Shuttling tankers lack whatever boiloff-reduction tech is on the depots.  This is especially important when the tanker needs to use a BLT to get to a lunar orbit, since transit will then take 2-3 months.

3) A tanker's direct EDL likely requires more delta-v for the landing burn than a depot's aerobraking campaign requires to manage attitude, course corrections to line up the next pass, and perigee raise at the end of the aerobraking.

4) Shuttling depots are completely standalone once they go to high orbit.  Their QD system will work directly with the target ship.  Shuttling tankers always need a depot on each end to get the QD gendering to work out, or they need the development of a completely new and androgynous QD for all Starships.

As long as aerobraking is possible, it seems like a no-brainer over tanker shuttles with EDL.

___________
¹When I say "aerobrake" here, I mean a gentle set of braking maneuvers, with somewhere between 20m/s and 200m/s of delta-v removed per pass, rather than a single-pass aerocapture.  All such maneuvers require about 70m/s to raise the perigee at the end of the aerobraking, plus whatever maneuvering and attitude control is required between passes.  At 20m/s removed per pass, it takes 75 days to return to 300x300.  At 200m/s, it takes 5 days.