Apogee Tugs and Cryo Transfer

About that “light blogging”…little did I know that I’d catch the bug that Jonny had, and be stuck at home with a nasty stomach flu…

Anyhow, one of the ideas that I’ve given a lot of thought to lately has been that of orbital cryogenic propellant transfer, and how to actually make it happen. The problem is that while there are a lot of markets that could eventually use on-orbit cryo propellant transfer eventually, when it is on the shelf, not very many of them are near-term, or low risk. The main markets I had been looking at before were for sending satellites to MEO and GEO, sending crew/cargo for NASA to Lunar orbit and beyond, and cislunar space tourism (with eventual private lunar settlement and exploitation). All of these markets are kind of long-term or uncertain though.

For reusable or expendable tugs to fly commercial satellites to MEO or GEO, you need to build up a fairly hefty track record, due to the overly conservative nature of those markets–why save $10M when you’re risking a $500M bird? The DoD seems to be willing to take a lot more risks (stuff like flying TacSat on the first Falcon I flight), so they might serve as an intermediate market, but that’s not too clear.

For NASA the biggest risk is political. On-orbit propellant transfer really gets rid of most of the need for Ares I and V, and makes the existing shuttle workforce obsolete. In fact, in order to gain the biggest benefit from orbital propellant transfer, they would have to scrap the Ares I and V developments, as the development cost of those systems is about $15B, and the yearly fixed operating costs is in the $2-4B range. Without killing those jobs, there’s no way NASA will be able to provide enough demand to really do anything constructive on the moon. However, killing those jobs is a very politically difficult thing to do. Seeing how little cajones either side of the House or Senate have these days, it’s a longshot indeed, even if it would save billions of dollars and make the VSE actually worth doing.

Cislunar space tourism (round the moon flights or visits to a Sundancer in L-1 or L-2) is rather interesting, but won’t become a reality until costs come down enough that a ticket costs about the same as a ticket to ISS does today. Such tourism will rapidly drive the demand for propellant deliveries up high enough to justify a true RLV, but the market suffers from a real chicken-and-egg issue. Until costs come down far enough, there won’t be enough demand. Settlement even more so.

So, I was kind of wondering if there was any realistic, near-term market that could use on-orbit cryogenic propellant transfer, and that could give the experience base to leverage those longer term markets off of. While I was thinking about this, I stumbled across an idea that Henry Spencer had discussed a few times on usenet over the years–Phil Chapman’s “Apogee Tug” idea.

The Apogee Tug
The basic idea consists of a reusable tug and a dumb cargo module. The tug “lives” at a space station (Sundancer, Nautilus, ISS, doesn’t matter). When a cargo launcher (or passenger/crew launcher for that matter) goes up, it places it’s load into a low parking orbit (something between the 165x185km orbit used in Apollo and the 200km circular orbit often used as the baseline for performance to LEO), and provides attitude control for the short period while waiting for the tug to arrive. The tug then leaves the station, does a burn to rendezvous with the target, and then docks or berths with it. Once mated, small fuel containers launched with the cargo module pump their contents over to the tug, which then carries the cargo module (minus the now detached upper stage) up to the station. The tug transfers enough propellant to both deliver the payload to the station, as well as return the tug to the parking orbit the next time a payload comes up. If needed, the tug could be built with oversized tanks allowing it to do both up and downmass maneuvers to and from the station and parking orbit.

The interesting result you get from this idea is that most of the time, there is an actual performance benefit (in extra cargo delivered to the station) compared to not using the tug, even if you assume fairly reasonable propellant ratios and propulsion Isps. While crunching numbers, I investigated several existing US and Russian orbital boosters, including Proton, Soyuz, K-1, Falcon IX, Delta-IVM, Delta-IVH, and Atlas V 401. While all of them showed a small to moderate payload bonus over having the launch vehicle itself deliver the payload directly to station orbit, this did require a few caveats. For many of the higher performance systems (Atlas and Delta in particular), the tug propulsion system had to have as good of Isp as the upper stage they were working with, and the tug dry mass also had to be fairly low. While this is reasonable for the LOX/Kero upper stages, it might be a bit of a stretch for the LOX/LH2 upper stages (RL-10s are pretty darned efficient, and making a smaller version of them might be tough–not to mention quite expensive).

Here were some interesting observations I made based on my calculations:

  • The apogee tug is more of a benefit if the upper stage has to provide a large amount of the Earth-to-Orbit Delta-V, with the biggest benefits going to fully reusable almost SSTO stages like the K-1 OV. Russian boosters also showed more of a benefit, mostly because they tend to be a bit beefier and have a higher upper stage drymass to payload ratio. After K-1 (which had a 9-11% payload benefit), the next two biggest benefactors would be the Soyuz and the Proton, which both had something like a 1-2% direct benefit.
  • The delivered payload to station isn’t tightly coupled with the drymass. For example, the theoretical numbers I used for a tug sized for the K-1 had about 200lb drymass, and about 360lb of propellant, for a delivered payload of 9855lb (compared to 8800lb without the tug). Upping the drymass of the tug to 500lb only costs about 21lb of payload. Upping it to 1000lb only drops the payload by about 62lb compared to a 200lb tug, and a ginormous 2000lb tug (at this point a gold-plated one with manned compartment, robot arm, solar panels, berthing interface, and lots of other goodies) only drops the on orbit payload by 145lb vs the 200lb drymass version. That’s still over 9700lb to the station, which is still almost a 10% increase.

    This appears to be due to the fact that since the Delta-V’s are so low, since the Tug is only taking the cargo one way, and since the tug hardware is left in orbit after each shot (and you only need to provide enough propellants with their transfer mechanisms to fuel the tug). Think of the tug as a reusable third stage that you don’t have to haul with you. Even for a lot of the higher performance US ELVs, beefing up the tug substantially doesn’t cut very far into one-way station payload

  • A corollary of the previous point is that a given tug could probably be sized for the bigger payloads, while still being able to provide a decent payload bonus for less capable boosters. This also means that it might be possible to add substantial Delta-V reserves just in case, or make the tug capable of both one-way and two-way operations. And adding enough mass to allow the thing to be manned doesn’t cost you very much payload-wise.
  • Even if the tug didn’t give a direct payload benefit, it saves the requirement of carrying a lot of docking specific “smarts” on the actual cargo module. The payloads can be much dumber, and easier to design, and hence much cheaper. These rendezvous and docking specific pieces of hardware cost mass and money, and when you combine that with the small payload bonus that exists for most launchers using the tug, it becomes a fairly sizeable benefit. Not having to have each cargo module or space station chunk have its own avionics suite, its own star tracker, its own RCS system and plumbing, its own AR&D hardware makes them a lot cheaper and easier to make. The actual interfaces for these dumb cargo containers could be standardized and open source.

    [Note: Pete Zaitcev points out that the term “open source” might not be the most accurate for what I meant to say. He suggests “open specification”, or a couple others. Maybe “open architecture”? Freely available for download online? I guess I’m just not that savvy when it comes to IT jargon.]

  • This last point is particularly noticable for light payload systems, like many potential first generation RLVs will be. Being able to have a tug carry hundreds of pounds of good rendezvous, docking, power, and communications hardware for you could make a huge difference for say a 1-2ton to orbit RLV. Smaller RLVs are potentially cheaper to develop and operate, but tend to suffer disproportionately from systems that only scale down so far. A tug might very well make a total payload benefit of as much as 50% for such a vehicle.
  • Making a light manned version of the tug that is still capable of providing enough of a payload benefit for all potential orbital vehicles is fairly reasonable, and would eliminate the need for autonomous rendezvous and docking. Manned docking is like state of 1960s technology. We’ve been doing this since my dad was in elementary school, and we have a lot more experience with it (in the US) than we do for autonomous rendezvous and docking.
  • This would provide a great way to gain experience with “orbital assembly”, rendezvous and docking, and cryogenic propellant transfer. The experience gained from this system would probably be 90% of the work needed to make the first orbital propellant depot possible. And once such a system is available, things get really interesting fast…
  • If you have some flights carry extra propellants which get stored at the station, you can do occasional tug flights where the payload itself doesn’t provide propellants, allowing a much bigger maximum payload to be taken to the space station than would be otherwise possible. This also tends to become a larger effect the smaller the initial payload size.

Other Tug Ideas
This idea isn’t particularly new or revolutionary. It was proposed as part of the original Space Transportation System (that ended up getting axed when Shuttle costs overran). The Russians have been talking about their Parom tug concept recently that they are trying to develop. It’d be good for 60 flights, and would be a great asset, even though it would probably use hypergolics (which are a lot easier to transfer than cryogenic propellants). Constellation Services International was also thinking about using a similar system, except based around using the Progress module. All of them are good ideas. Using cryo propellants would improve the efficiency a bit, and get us a lot of experience with that technology before we start trying to run our first orbital propellant depot, or trying to fuel on-orbit our first lunar transfer stage. There are plenty of ways of skinning that cat, and I may be doing some R&D on one of the more clever methods in the near future.

Especially with something like Bigelow’s Sundancer module planning to go up before the end of this decade, and with NASA’s need to transition to commercial ISS resupply there’s a real potential near-term market that benefits from on-orbit cryogenic propellant transfer.

Just a thought.

The following two tabs change content below.
Jonathan Goff

Jonathan Goff

President/CEO at Altius Space Machines
Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and is the founder and CEO of Altius Space Machines, a space robotics startup in Broomfield, CO. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew. Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
This entry was posted in Uncategorized. Bookmark the permalink.

12 Responses to Apogee Tugs and Cryo Transfer

  1. Anonymous says:

    Flying TacSat on Falcon 1 should not be taken as evidence of risk-taking behavior by the DoD. It was a student bird with no budget allocated to launch it. The alternative was letting stay on the shelf.

  2. Anonymous says:

    One of the benefits of being old is you remember things like when NASA first trotted out the space station idea it was one of three things that were supposed to work in concert. The Space Tug, the Space Station, and the Space Shuttle were SUPPOSED to be built at the same time. That was the follow on to Apollo. Now somebody pared that down to just the Shuttle during the Nixon Administration. It would be interesting, if you Rand, could take a look back at what they planned to do with that Space Tug and compare it with the present notions of the Space Tug.

  3. Jon Goff says:

    Yeah, while some parts of the DoD are willing to take some risks and be innovative, a large part of the DoD is just as stodgy conservative as the comsat operators. There *might* be some possibilities there that don’t exsist with the comsat guys, but it’s a longshot–which is why I was focusing on commercial station flights using a tug as a more likely market.


  4. Jon Goff says:

    Yup, it was definitely intended to be a combo deal. I’m not sure that they wouldn’t have botched the STS anyway had they gotten everything they asked for for all three components though. It’s amazing how many good ideas were abandoned by NASA way back in the day. The nice thing about a small tug like that is that I think it could be developed by the private sector for far less than NASA would budget for it, which means its more likely to get done.


  5. tankmodeler says:

    The Orbital Express program demonstrates a lot of the technologies needed to validate a tug (and a few more as well). All we need now is a couple of companies to step in and create the partnerships needed to benefit both the COTS programs and themselves…

  6. Anonymous says:

    (in response to anonymous at 7:10 AM) TacSat 1 was originally scheduled to fly on the first flight of Falcon 1 out of Vandenberg AFB. Due to launch pad conflicts, the launch was moved to Kwaj and FalconSat (the Air Force Academy payload with no launch budget) was placed on top for the first flight. We all know what happened next.

  7. jake says:

    As you partially observed, the benefit from the apogee tug is almost entirely in reducing the non-payload mass taken all the way to the final destination.

    In other words, it’s staging. It’s cheater staging, along the lines of the L-1011 for the Pegasus, or the original Kistler Launch Assist Platform, but it’s staging nonetheless.

    If your stages are operating on the flattish part of the rocket equation curve like good classical rocket stages do (the Atlas 401 is archetypical here, the Kistler K-1 somewhat less so), there’s no benefit. If your stage is operating out in the steep part of the curve (Shuttle, Ares I, any hypothetical SSTO), then you can save big.

    You can get a very good feel for this by considering launches to GTO. Many launchers can place payloads directly into GSO, and using something like the IUS it works out OK. But if you use the satellite itself as an apogee tug, you can simplify your launch vehicle to the Atlas V 401, with all of the benefits that come with it.

  8. jak says:


    You didn’t mention this in your posting, but are you assuming the RLV is going to orbit? If not, then I have another question. In the past, you’ve expressed some skepticism about tethers due to the redevouz problem. Why would that be different in this case? Is it because the tug is a larger target, or something to do with the orbital dynamics of a tether platform v.s. a tug which is going between orbits? Thanx.

  9. ザイツェヴ says:

    It’s all right to use “open source”, but an “open standard” would probably be better. It’s just weird to see it getting mainstreamed by you outside of computers.

    I have no idea about numbers, but even if Jake is right about mass advantage not being there, there’s still an advantage of not throwing away a bunch of engines and other equipment.

  10. Jon Goff says:

    In other words, it’s staging. It’s cheater staging, along the lines of the L-1011 for the Pegasus, or the original Kistler Launch Assist Platform, but it’s staging nonetheless.

    Yeah. It’s a really nice form of “cheater staging” because you have an upper stage with effectively zero drymass (from the perspective of the launch vehicle). This still isn’t enough *directly* to always come out as a huge win (except in the cases you mentioned about Shuttle, Shaft, and SSTOs), but between the indirect mass savings of being able to not have to have RCS systems on the payload, not needing a docking system (a berthing interface is enough) etc, you actually do end up getting an effective payload boost.

    More importantly you have direct cost savings from not having to pay for those RCS systems, advanced avionics, rendezvous and docking hardware, etc for every single flight. Even if the payload benefit were zero, what really matters (at least in a commercial market) is the cost. It doesn’t matter if you save money by getting more payload for a given launch buck, or if you get it by dropping per delivery cost from all the AR&D hardware–either way you’re saving money, which is what really matters.


  11. Jon Goff says:

    You didn’t mention this in your posting, but are you assuming the RLV is going to orbit?

    For the post in question, I was assuming that the RLV is going to at least a low parking orbit (something that’s stable for at least a day or two, if not more). Now, Phil Chapman had suggested the idea that you could have a tug rendezvous with a suborbital payload….but like you I think that’s a little too gutsy for my taste.

    If not, then I have another question. In the past, you’ve expressed some skepticism about tethers due to the redevouz problem. Why would that be different in this case?

    It wouldn’t be different in this case, which is why I don’t like the idea of a suborbital rendezvous like that. You have a rendezvous and docking procedure that has to happen right, in a very short time frame (single digit minutes), every time…It’s just nowhere near as tolerant to, oh say reality, as I would like. By slowing things down enough that you days to solve the problem, you make a lot less likely that you’ll end up losing payloads.


  12. Anonymous says:

    Another potential advantage of using tugs: you can afford some uncertainty in the final orbit of the cargo and run your engine until tanks are completely empty and vented through the engined – no reserve residuals. The tug will take care of the rest.

    At this point in the curve every little bit of fuel makes a big difference in delta-V but usually you can’t use it because of the uncertaintly.

    This is assuming, of course, that the upper stage is capable of running dry without damage. A pressure-fed upper stage like Falcon 1 may be able to do it, for example.


Leave a Reply

Your email address will not be published. Required fields are marked *