It is fairly obvious that there are many benefits to having refueling capabilities in space. Perhaps not to the monster rocket fanatics with blinders and a few others, but to most of us it approaches no-brainer territory. The problem is getting to the first egg. The first bird we would accept as a chicken came from an egg, but that egg did not have to be laid by a bird we would accept as a chicken. Mules do not have mules for parents. The first refueling capability doesn’t have to have depots in the business plan, perhaps a buddy tank for starters.
The problem with getting refueling capability in space is often called a classic chicken and egg problem. Without a depot there is no refueling capability or reason to build vehicles to use it, and without the vehicles to use it there is no reason to build a depot. Dual launch architectures try to get around this by designing missions around a propellant launch and a mission launch. So far, this hasn’t worked well either as the mission planners try really hard to avoid boxing themselves into a corner where if either launch  fails, the mission fails.
One possibility hinted at in comments a while back is to have the primary mission fully capable of success if the primary launcher works properly. Then have enhanced secondary capability if the propellant launch succeeds and can top off the tanks in the primary. It took me a while to wrap my thoughts around the concept, after which I started working it over to see what I could think of. I can’t give due credit to the original thinker because I didn’t really grasp it until days later, just that it was somewhere in the 40 depot posts by Jon Goff and hundreds of comments associated with them. The originator can chastise me in comments  for not digging back for is name.
So this is my take on the idea. A vehicle arrives in LEO with payload for LEO. It hits the refueling vehicle either before or after placing its’ primary in the correct orbit. It off loads any excess propellant it managed to bring up if it is going to deorbit, or if it simply has excess for the secondary planned mission(s). Alternately it accepts enough more propellant to accomplish other missions that are non-critical to the primary mission.
The secondary missions suggested were along the lines of deorbiting various dead satellites. Surveying satellites with minor issues with the intent to design possible repair missions. Rendezvous with satellites in useless orbits and boosting them to the proper orbits. Creating the capability of the upper stage herding dead GEO sats after a GTO burn for the primary customer. Testing servicing options that would be too expensive for a dedicated launch.
The whole point would be to have some capabilities in place and in use for the cost of one launch that did not have to satisfy any customers other than the company that launched it. It would be replenished with excess propellant in vehicles that didn’t need all they hauled up. It would refuel various company vehicles that were scheduled to do extra work for revenue. The refueling wouldn’t have to be a maximum top off, just a calculated amount for individual missions. By using the capability in house, the problem of selling to and satisfying outside customers goes away. By not putting the capability in the critical path, the primary customers should have very few concerns about endangering their ROI.
It would seem that a capability could be started in the two digit millions. It could be very crude with perhaps LOX only to start, or LOX/Kero, or hypergolics, just as long as it fit that particular companies’ business plan.
Later on it might be possible to put this in the critical path of missions because a solid experience base and track record had been built up. Then after that it should be much simpler to get true multi-propellant depots up and running with whatever the market expresses a need for.
johnhare
Latest posts by johnhare (see all)
- Regenerative cooled turbopump - March 16, 2023
- Second Guessing Starship - September 19, 2022
- Projectile Fusion - May 19, 2022
Two observations:
“secondary missions suggested were along the lines of deorbiting various dead satellites. Surveying satellites with minor issues with the intent to design possible repair missions. Rendezvous with satellites in useless orbits and boosting them to the proper orbits. […]
It would be replenished with excess propellant in vehicles that didn’t need all they hauled up. It would refuel various company vehicles that were scheduled to do extra work for revenue.”
As described, you seem to require everything to be in the same orbit.
“A vehicle arrives in LEO with payload for LEO. It hits the refueling vehicle either before or after placing its’ primary in the correct orbit.”
If you are launching a docking-capable upper-stage and a refuelling vehicle on its own launch, why not just launch the “refuelling” vehicle as a docking-capable upper-stage directly to the second target?
What do you gain, in the early days, by having the “refueller” dock with that first upper-stage?
For me, the stepping stone would be:
Develop a docking-capable booster-stage. This can be used to deliver generic payloads to dockable/berthable platforms like ISS. It can also be used for double-launch-boosting payloads (launch heavy payload into LEO on a cheap launcher, launch the booster separately to kick it into GEO/BEO.) The latter function also serves the role of a satellite rescue vehicle (without requiring everything to be conveniently in the same orbit.)
Note the absence of refuelling. All it needs is the ability to do prox-ops and docking. (“All” it needs. Heh.) Each target is planned, and therefore can be pre-fitted with appropriate attachments for the docking-booster. There’s no need to dock with an uncooperative payload. (Obviously, it needs the ability to restart.)
Note the absence of long stays in orbit. The docking-booster is launched directly to its target. Therefore boil-off doesn’t rear its vaporous head.
On the negative side, note that it’s initially expendable. Which means the docking-booster has to be only fractionally more expensive than a baseline upper-stage, which means it is a modified upper-stage.
Things like reusability, refuelling, docking with uncooperative payloads, depots, modular stacking the booster for heavy BEO burns, etc etc, can be added later. But you start with a docking system and you have a product that’s immediately useful.
A depot proposal requires docking and refuelling and long-duration prop storage, and enough stuff going on in a single orbital inclination and a narrow range of orbits to justify all the infrastructure in that spot.
A refuellable upper-stage (LM-Jupiter) needs to have a launch of the new payload and propellant each time on yet another upper-stage. Why not launch that stage directly to the target?
Depots do only work with things in the same orbit. This was a post about working up to having depots. With your input, I would say that the tanker would simply be a modified upper stage that could be retasked as a little depot after delivering its’ payload. In this way, the depot development is piggybacked on revenue launches similar to the attempts at Falcon9 stage recovery. Your addition could get the initial costs into the single digit millions.
SpaceX would be more profitable now if they had never bothered trying to recover stages, but later on that could reverse. I see the same thing happening on depots, except that it could actually be a lower initial investment.
A long time ago I mentioned that with fuel transfer capability all reusable spaceships WERE essentially fuel depots. The idea never got traction.
Reusability gets a whole lot easier if you have ships that never deorbit.
There is the flight test of the Falcon Heavy soon. Did it find a payload or a dummy payload?
A starter depot containing a single tank would make a dummy payload that has the ability to recover some of the cost by selling the propellant. Use a cheap propellant like water, LOX or Argon – alternatively a mono propellant.
Being a satellite the depot would need an attitude control system, roll control, solar panels and a homing beacon. To save time and money the connectors and pumps could be the same as those used on the 2011 Robotic Refuelling Mission or the Orbital Express mission in 2007.
“Reusability gets a whole lot easier if you have ships that never deorbit.”
Don’t see the logic. You still have to launch an entirely separate vehicle with extra fuel. What are you saving?
Is you fuel delivery tug reusable? If so, then you have to engineer that and a long-duration orbital vehicle. If not, then you are throwing away the delivery tug.
And with any long-duration orbital asset, you are back the problem of needing everything in the same orbit.
If it never deorbits the transfer vehicle does not have to be shaped like a capsule. For example its living quarters could be shaped like a Bigelow B330 habitat with airlocks, radio antennas, solar panels and docking ports sticking out of the sides.
It may be possible to design a combined launch faring with reentry ‘top’ heat shield to protect the fuel delivery vehicle’s tanks, allowing reusability.
Andrew,
Nothing earlier assumed a manned facility, AFAICT. Certainly I didn’t.
Yes, it might be better to leave a space-station in orbit and just launch fuel, rather than launch all the systems contained in that space-station every single launch. But I was just talking about depots.
Not seeing the advantage of a reusable upper-stage, unless your target is always in the same orbital plane. (Eg, if your reusable system only services the ISS.)
In space fuelling becomes worth while when the mass to LEO exceeds the maximum payload of the launch vehicle. The payload is launched (nearly) dry. 30+ tonne payloads are likely to be manned.
Smaller launch vehicles, say 2 tonne to LEO, could use inspace refuelling to get 2 tonne satellites to GEO or a lunar Lagrange point.
As a safety measure I hope fuel tanks and oxidiser tanks have connectors that are a different shape. Uncontrolled mixing of the two can have nasty effects. This does not have to be heavy just 2 or 3 pins sticking out the side of the pipe.
Andrew,
Or having the connectors be different sizes–which probably would happen automatically since you need more LH2 volume than LOX volume.
~Jon
In Britain the car companies tried having different sized nozzles for diesel, petroleum and lead free petrol. This has a high failure rate size as according to the RAC car breakdown organisation there are “roughly 150,000 motorists filling their tank with the wrong fuel every year.”
http://www.rac.co.uk/breakdown-cover/wrong-fuel-recovery/petrol-in-a-diesel-car
Spacecraft may be better since an air tight seal is needed. Just ensure that a small connector can not insert liquid into a large connector and vis versa.
“monster rocket fanatics with blinders”
I resemble that remark! No blinders though. I can see liquids being handled on the ground in ever larger LVs, with liquids handled in space at a minimum.
I really don’t like the ideas of depots. If we ever get a scenario anywhere close to the movie GRAVITY–it won’t be from a missile test–but some rupturing depot freaking out all over LEO.
Fill HLLVs on the ground–exchange liquids for thrust/momentum–and dare that to leak.
Depots come into their own for the return trip and reusable lunar landers.
As a safety measure I hope fuel tanks and oxidiser tanks have connectors that are a different shape.
Or you put them at fixed locations on the docking adaptor so that there is only one way to get hooked to both.
Ech wrote > Or you put them at fixed locations on the docking adaptor so that there is only one way to get hooked to both.
Works if you are refuelling via the docking adaptor, this is how power and data connections work. However the propellant connectors could be on the side of the stage.
Jeff Wright: “I really don’t like the ideas of depots. If we ever get a scenario anywhere close to the movie GRAVITY–it won’t be from a missile test–but some rupturing depot freaking out all over LEO.”
Kessler Syndrome is more likely to occur due to the number of dead satellites currently in orbit. For most of them, they’re only dead because they’re out of fuel and can’t maneuver anymore – so they can’t avoid collisions.
Also, once you have depots in orbit you are no longer limited to refueling them from Earth. We;re going to need that ISRU capability eventually, and so we’re going to have to figure out orbital propellant storage and transfer anyhow.
Jeff,
Fuel vapour isn’t a Kessler risk. Therefore for the purposes of debris risks, the depot contains no more material than the dry-weight of the tankage. Being actively managed, the tanks on a depot are at less risk of rupture than the tanks on a dead satellite.