Additional Thoughts on the SA’08 Propellant Depot Panel

It’s been long enough since I put up the presentations from the panel, that I figured it would be worth starting a new post to mention some of my thoughts on what we discussed on the panel. Some of these ideas are thoughts that were briefly discussed during the panel, and others are ideas I just didn’t have time to bring up.

While the panel went much better than expected, I also noticed several things I could do next time to make it even better.

One thing is that I probably should have mixed audience participation, discussion, and presentation a little more thoroughly. Expecting people to stay awake through an hour of presentations right after dinner was a bit much–surprisingly enough most of the attendees did, but I think mixing it up would’ve been better.

Another thing that in hind-sight would’ve been useful is to set aside more time in the panel for actual discussion, particularly on points where one or more of us had a different preferred approach than the others. For the most part we all agreed on the importance of propellant depots, the benefits to commercial and governmental programs, and the fact that a lot of the technology either exists, or is on the cusp of existing. But we disagreed on some of the details, and much like statistics, it’s the disagreement or outliers (and the reasoning behind them) that hold the most information.

Tugs or Reusable Tankers or Both

One point of disagreement was between myself and Dallas Bienhoff regarding the best way for handling prox-ops for propellant deliveries. I am a big proponent of using space tugs to offload as much of the weight and complexity of a prox-ops system to an asset that stays on-orbit and gets used a bunch of times. Dallas on the other hand disagreed with me on that point and suggested that a reusable tanker with autonomous rendezvous and docking capabilities was a better way to handle complexity and reduce cost. Both of us were advocating for reusing as much of the complicated prox-ops hardware as possible, but going about it in different ways.

To me, there are several benefit of having the tanks dumb as possible, with all of the complicated subsystems on the tug:

  • Performance: the tug stays in orbit, and is very weight insensitive. By not having to relaunch the tug every time, you get a lot more propellant per pound of delivered mass on orbit. When you look at unmanned delivery craft right now, they typically have a really poor payload to drymass ratio. For a tug you want as high of a payload to drymass ratio as possible.
  • Easier Open Interface Standards: It is probably a lot easier to get ITAR approval for openly publishing a very simple common interface specification if that interface spec is just a few handholds and a commercial, off-the-shelf quick-disconnect receptacle. That way, anyone who can launch stuff into your station’s orbit can deliver propellants to your station, even if they’re durned furiners. An Autonomous Rendezvous & Docking (AR&D) system with automatic fluid couplings would likely be a lot harder to get export approval for.
  • Flexibility: The simpler the propellant module, and the less smarts it has built-in, the easier it is for people to just stretch it to whatever size best suits their launch vehicle. Unlike a reusable tanker which might require significant redesign to be used on a different launch vehicle.
  • RLV Simplicity: Making a high-flight rate reusable launch vehicle is going to be hard enough already without trying to also make it into a satellite as well. For RLVs, you want a relatively dumb tank that never leaves the payload bay, but you don’t want to have to have all the prox-ops stuff cut into your already very limited payload budget.
  • Other Tug Uses–On-Orbit Assembly: Tugs are useful for lots of other things too, especially if they have arms. They can help in assembly of stations and large in-space vehicles. No need to make each and every space station or space vehicle component be its own independently operable mini-station complete with its own GN&C, power, etc.
  • Other Tug Uses–Satellite Recovery: They can also perform satellite recovery missions. Imagine if there had been a tug already developed, and set aside on standby in case of a botched launch like the recent Proton upper stage failure. If done properly, the tug could’ve been launched on short notice on an otherwise empty upper stage. The tug could’ve then transfered the satellite from the malfunctioning upper stage to the still functional, and mostly full upper stage that delivered the tug. There are some very tricky technical details I’m glossing over, but its a capability that could become rather standard once you have tugs available. In case anyone from the DoD is reading this, yes, I’m saying that tugs are an important part of ORS.
  • Other Tug Uses–Rescue Missions: They can perform rescue missions. Right now, one of the most hazardous parts of a lunar mission is the ascent, rendezvous, and earth return legs. Imagine if there was a staging point in L1, L2, or LUNO, instead of basing all lunar missions from earth’s surface. You could store one or two of these tugs at the small staging/refueling base. If something went wrong with the LSAM US or CEV, you could send a tug in to help out. If you were using lunar ejector seats, and had to abort to orbit, this would give you a quick way of getting a rescuer to a stranded astronaut. This would greatly reduce your odds of losing a crew due to a LSAM/CEV rendezvous failure, or CEV propulsion failure prior to (or during) TEI.

And there are probably other ideas I’m overlooking.

On the other hand, Dallas a good point in favor of reusable propellant tankers, and I can think of some others as well:

  • The more expensive propellant handling hardware your tanker needs, the better it would be to reuse it. For instance, say you don’t think you can get first-orbit or even first-day rendezvous with your propellant depot. You might want to invest more heavily in insulation, zero boil-off systems, and other cryo handling hardware. You don’t want to be tossing that away after every flight.
  • You’re eventually going to want to have smaller depots located on the other ends of your transportation system (ie in the lunar vacinity, around Mars, around Venus, etc). Some of these locations, especially at first, will need to be fueled from Earth. That means tanker modules are necessary. Once again, once the flight duration gets longer than a couple of hours, you’re going to start wanting to add other bells and whistles. And those bells and whistles are expensive enough that not throwing them away after every flight is a good idea.

Dallas may have had some other points that I’m not remembering right now, but I think that both sides have valid points, and that the best option may be to do a little bit of both.

Full-Service vs. LOX-Only?

I had been somewhat surprised when Dallas (who works for the company that was the lead on developing and flying Orbital Express) suggested against using a tug. I was even more surprised when Frank Zegler suggested a LOX-only depot. Before I had met Frank over the internet, I considered LH2 to be an unmitigated evil, almost on the level of Nitrogen Tetroxide and UDMH. But he was one of the main people to talk me into thinking that Hydrogen isn’t always evil, and sometimes can be tamed, and can make a lot of sense. So, when he sided with the sentiment that meiza and several other regulars here at Selenianboondocks have expressed–namely that your first depot should probably be LOX only–I was very surprised.

I didn’t have time to bring up this point of disagreement at all during the panel, but here were some of Frank’s points in favor of LOX-only depots:

  • LOX is much easier to store and handle cryogenically due to its much higher boiling point.
  • LOX is much denser, and thus you can store a lot more of it in a given size tank.
  • LOX makes up the majority of the propellant mass than for any fuel combo you would likely use.
  • Storing only one liquid is much easier than two, because you can eliminate the heat transfer from the warmer propellant (LOX) into the colder one (LH2). Even with a sunshield or a ton of MLI, you still have a significant heat source in the fact that your LOX is way hotter than the boiling temperature of your LH2. You probably never thought of LOX as a heat source, did you?

If these arguments sound familiar, its because they’re the same ones that many of you have made over the years. I can see Frank’s point, especially if you think that your main (or only) market is going to be “topping up” EDSes and LSAMs for NASA. I’ve never disputed these facts. But I still think that going all the way and providing at least one fuel to go with that LOX is a good step. These arguments probably aren’t new, and probably aren’t going to change anyone’s minds, but in case you haven’t heard my spiel before here’s my case:

  • Without modifying existing or future stages, they only have so much hydrogen capacity. Unless you launch a complete stage as your payload, topped-off to the brim, you’re going to use some of that capacity getting to orbit in the first place. Which quickly cuts into your maximum payload you can deliver to your final destination (and also how much LOX you can actually use).
  • For many payloads, prior to the time when reusable LEO-GEO or LEO-Luna ferries are available, the best way to use a propellant depot is to launch the payload on a refuelable upper stage, top that upper stage up in LEO, and then immediately go to your destination. If you reuse your upper stage as your transfer stage, the inability to top of the hydrogen as well effectively halves your payload you can deliver to other destinations. If you fly a separate refuelable stage that has a full load of LH2, you’re greatly cutting into how big of a payload you can put into LEO in the first place. For instance, a Centaur stage with a full load of LH2 weighs about half of the payload capacity of an Atlas V 401 to LEO.
  • If you can’t provide both oxidizer and fuel, you can’t reuse interorbital transfer stages/ferries.
  • If you can’t provide both oxidizer and fuel, you can’t reuse lunar landers.
  • If you can’t provide fuel on-orbit, you can’t make up for boil-off caused by unexpected delays, variance in the thermal properties and boil-off rate of your stages, equipment malfunctions etc.
  • On a psychological level, going LOX-only allows people to continue to disbelieve in the feasibility and utility of propellant depots. Look at the mindset two years ago. It said that propellant transfer of any sort on orbit was deep, black magic, and that it should be avoided at all cost. Now that Orbital Express has shown that it is doable and not that hard for storable propellants, critics say “well, that’s all good and fine, but cryogenic propellants are a whole different beast entirely.” If we went to LOX-only depots, those same critics would likely say “well we knew LOX was doable all along, it’s the hydrogen that’s the unrealistic part–there’s no way you could store that long enough to be practical.” At some point I want to stop giving skeptics ammunition.
  • More importantly, both Dallas and Frank agree that LH2 storage on-orbit is completely feasible. Dallas going so far as to say that for 1-2kW and 50-60kg, you could install a ZBO system that could completely eliminate boil-off.

I guess I’m still convinced that in spite of the added extra difficulty, that the real markets that I think there will be for propellants on-orbit will be much better served if you can provide fuel as well as oxidizer. But you can draw your own conclusions.

To ZBO or Not to ZBO?

Another disagreement (this time between Dallas and Frank) was on whether or not to go with a Zero Boil-Off system for long-duration cryo storage. Dallas seemed to think that not only would it not be that hard to implement, but that it would be very desirable, while Frank seemed to prefer passive storage techniques, and in fact considered ZBO to actually be a detriment! I think both sides have points, but that in a way they’re somewhat talking past each other. And in the end, I think I side more with Dallas on this one.

Frank is right that ZBO systems, done the typical way, (without doing a proper passive-storage design and without settling propellants) is likely going to be an expensive development project, and a complicated system in operations. Frank also made the point that at least some of the boiled-off hydrogen is actually useful. That warm GH2 can be propulsively vented to cause the other propellants to stay in a settled orientation. It is still pretty cold, so it can be used to pull heat from the avionics away from the propellants. It can be used for prechilling lines and valves. It can be used to provide propellants for GOX/GH2 RCS engines. It can be combined with oxygen in a fuel cell to provide water. The single most important benefit for Frank and ULA is that first one–settling the propellants makes everything easier, and propulsive settling is by far the highest maturity and easiest way of settling propellants.

As Dallas pointed out, a properly designed ZBO system when added on-top of a good passive insulation system doesn’t need to be that big and complex. 1-2kW isn’t that much power. And especially if the propellant is settled in some fashion, running a cryocooler becomes even easier, because you can avoid two-phase flow. If your cryocooler works better taking gas in and spitting out liquid (or chilled gas), settling allows you to guarantee you’re only pulling in gas. If pulling in liquid, subcooling it, and spraying it through the gas is more effective, settling allows you to pull liquid from a part of the tank where you know liquid will be, and to inject it into a part of the tank where you know it will be gas.

Lastly, there are other ways to settle propellants, and if you use them, you no longer need boil-off gases to provide the settling. You might still intentionally allow some LH2 to boil-off, for use in RCS engines for stationkeeping or to run fuel cells. But with a ZBO system, you have a choice.

A more convincing argument against the complexity of a ZBO system, that can be derived from Frank’s presentation, is the fact that a good passive system can get boiloff rates low enough that you just don’t care about them anymore. In those cases, a ZBO system might not buy you that much extra performance. Now, there’s an argument against ZBO that I’m more willing to buy.

So, I guess the real answer may be–it depends. In situations where your propellant is expensive enough to deliver to, where deliveries are somewhat infrequent, and storage times are long, a ZBO system might make a lot of sense. But in situations where the propellant can be readily topped off from tankers on a regular basis, even though ZBO is doable and not that hard, it still might not be worth it.

Once again, draw your own conclusions.

NASA vs Other Markets

This last topic isn’t so much a disagreement, as an area that I thought deserved a little more commentary. I think that all of the panelists would agree that NASA is unlikely to have a change of heart tomorrow, and completely overhaul Constellation to take more advantage of propellant depots. However, in spite of this recognition that NASA isn’t likely to become a good customer anytime soon, most of the panel was still very NASA- and Constellation-centric. While there were mentions made by all the panelists about performance benefits that normal ELVs could get for delivering payloads to GEO and beyond, most of the discussion of benefits was focused on augmenting NASA’s return to the Moon.

Admittedly, if NASA ever gets its act together and actually makes it back to the Moon, it will be annually consuming more propellant mass in orbit than the combined launch mass of all other launches combined, and if they were actually buying that from propellant depots, it would be a truly transformational event. But lets face it guys–it makes too much sense for NASA to willingly go along with it. Much like Zero-G demonstrated, while NASA might eventually be willing to abide by the law and purchase commercial services that they used to provide for themselves in-house, it will take many years to get them to change. As it is, it’ll take NASA a decent amount of time before they can even take advantage of depots, even if they recognized the potential right away.

As I said in my presentation, it doesn’t matter how critical propellant depots are for creating a spacefaring society, or how much better lunar exploration would be with propellant depots involved. If you can’t find a way to get enough real customers to wrap a business case around, it will never happen.

That said, I also wanted to note that NASA and the DoD are actually doing some good things regarding propellant depots. First, they’ve regularly put out SBIR solicitations for technologies that could be relevant to propellant depots. Second, on the larger scale, they’ve funded actual technology demonstration missions like Orbital Express, DART, and XSS-11 to demonstrate useful related technologies. Third, even though Michael Griffin’s NASA hasn’t been doing much action-wise to enable propellant depots, Griffin has at least been a vocal proponent of the capability in many public forums. Fourth, NASA was at least interested in offering a propellant depot related Centennial Challenge–if they had actually been given any new Centennial Challenges funding in the past three years.

Even though I don’t think either the DoD or NASA is likely to outright fund a propellant depot anytime soon (and personally I wouldn’t want them to!), there are lots of things that can be done within the system to help move things closer to reality. Better, clearer ties can be made between the technologies needed for propellant depots, and the needs and desires of NASA and the DoD. Tugs and depots, for instance, are an important part of a truly Operationally Responsive space transportation infrastructure. By making that connection more and more in public, additional funding for research, development, and demonstration might become available. NASA also desperately needs good long-duration cryo storage and handling technologies in order to make ESAS work, and at least some of those technologies will also be useful for propellant depots. Propellant depots (especially commercial ones) might allow NASA to launch larger interplanetary missions than would otherwise be possible, etc. So while I think the key to propellant depots lies in markets outside of NASA, I think there’s a lot of good NASA and the DoD can still due, even in spite of the political environment and constraints that both of them operate in.

Anyhow, those were some of my thoughts I wanted to discuss from the panel. Once I get the video from Henry, I might find a couple changes or additional comments to bring up, but for now those are my thoughts.


[Update: 4/4 8:30AM PDT]
Here is the link to a thread on where I have been discussing propellant depots with some of the other regulars.

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.
Jonathan Goff

About Jonathan Goff

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 Commercial Space, NASA, Propellant Depots, Space Development, Space Policy. Bookmark the permalink.

20 Responses to Additional Thoughts on the SA’08 Propellant Depot Panel

  1. Bill White says:

    Use LOX & kerosene for the LSAM and the LEO to LLO shuttle and you can store the kerosene easily enough in big Kevlar bags far away from the LOX.

    Could you simply fill a small Bigelow hab (Genesis sized?) with RP-1? Even if it freezes (-73 C?) just use sunlight to heat it up before fueling a customer.

    Trade lower ISP for extended shelf life, lower handling costs and cheaper tanks.

  2. Bill White says:

    PPS – The lower performance of a kerosene single stage reusable lunar lander might well be offset by the ability to send bags of RP-1 to an L point via solar ion propulsion and Belbruno trajectories.

    What you lose in LSAM performance you might win back with lower fuel delivery costs to EML-1 and EML-2.

    Obviously, the numbers need to be run to validate or disprove this idea.

  3. Brock says:

    Considering that propellant depots are at least a few years out, how does the analysis change if you assume that a good chunk of your business will be VASIMR driven?

    I understand that VASIMR uses hydrogen only, so would it make sense to provide hydrogen to the chemical rockets too to reduce the number of types of fuels you have to stock?

    Of course, if you decide to stock multiple types of fuels, Argon and Neon are easier to store than Hydrogen.

  4. Jon Goff says:

    Converting a Bigelow module into a storage tank for storable propellants should not be very hard. The technologies for flexible room-temperature fuel bladders are more or less off-the-shelf. You’d need to add a few features either to the propellant or to the tankage to allow for settling (for easier handling), but those are relatively straightforward. Honestly, you could probably turn a Bigelow module into a tank for *cryogenic* propellants. That’s actually been one of my pet ideas for a while. There’s some non-obvious trickiness needed to pull that off.

    The thing is that with the current high costs of launch to orbit, the extra cost of handling LH2 is probably more than offset by the lower total launch cost. Sure you may have more initial launch mass for your station, and sure, there are going to be some extra headaches involved. But if it halves your IMLEO for a mission, it’ll probably pay for itself on the first flight.

    The main reason to include kerosene is if there’s a customer (say an existing LV provider) who wants to modify a LOX/Kero upper stage for refueling. Think Falcon, Soyuz, or Zenit. Otherwise, if there isn’t actually a specific customer need for Kerosene, while it’s easier, it’s not clear how much it really buys you until launch costs start coming down substantially.

    Just my opinion though.


  5. Jon Goff says:

    VASIMR and other non-chemical propulsion systems such as NTRs, Microwave or Laser Thermal, etc. all look pretty nice with LH2. Which is another reason why having the ability to store LH2 is nice.

    That said, most of those are a ways off yet. The nearterm demand is lots of LOX and a smaller amount of LH2 or kerosene. Well, there’s also a market for N2O4 and Hydrazine as well.


  6. Anonymous says:

    While one does eventually want to have both the oxidizer and the fuel at a depot, or colocated separate depots, one has to get from here to there. The depots that Dallas showed in his presentation look as complex as the ISS, composed of 6 or more Delta upper stage tanks, a central truss, power, radiators, long complex plumbing that has to be chilled to enable any form of transfer, etc. This will cost more than $1B just in launch costs, not to mention development. ISS costs are over $100B. I’m sure that a commercial depot, even of Dallas’s complexity, could be developed for a fraction of that, but you are still talking many billions of dollars. That initial investment is a hurdle that is likely never going to be crossed for the first depot!

    Frank’s concept provides an entry into providing propellant on orbit at a fraction of the initial price. A single launch of a fully integrated depot, passive, settled cryo storage, short plumbing routes, maximum use of technologies already flying or currently in development. I can easily see such a simple depot being developed and launched for ~$500m.

    As has been mentioned, LO2 Composes most of the mass required in LEO for any mission beyond LEO, GEO or exploration. The fact that LO2 is also simpler to work with makes it an ideal choice to start with. If proven successful, a nearly duplicate depot could be launched to support the LH2 required.

    I think too many people place depots just out of reach by thinking of ideals vs. what is good enough.

  7. Jon Goff says:

    I think you may be presenting somewhat of a false dilemma. I definitely agree that the depot Dallas presented is a lot more complicated than I think a first-generation depot will end up being. However, I don’t think that having both LOX and LH2 necessarily implies a bit ISS-scale endeavor. Especially if the demand at first is relatively light (ie Centaur-loads worth of propellant or smaller), it might be possible to do a pilot depot that is basically just a single modified Centaur stage. There’s nothing really that precludes you from doing a simple station with both propellants. And at least so far as I’ve seen in my looking, most applications outside of NASA’s needs really want both propellants in order for a propellant depot to really provide enough benefit to be worth their while.


  8. Mike Puckett says:

    “ISS costs are over $100B”

    Only if you count the cost of the Shuttle program over the past 15 years. That number is seldom justified. I suspect minus the shuttle, which would have liekly flown in circles anyway, 30-35BN would have been a better estimate.

    Still, the prop depot needs to be less than this, much less.

    The nice thing about it is alot of the pieces already exist albiet in need of some mods.

    I was thinking co-locate too.

    Build your O2 depot and then put an H2 depot next door.

    Mabey you could eventually join them with a truss.

  9. Joe Silva says:

    Has anyone looked into launching and storing water? Later you could break it up using solar power as needed, maybe have smaller holding tanks for O and H. For sure we need water for manned missions anyway.

  10. Joe Silva says:

    Has anyone looked into launching with just water and beaming power too it to make high pressure steam as exhaust?

  11. Jon Goff says:

    Splitting water back into H2 and O2 is a *very* energy intensive process. I can’t remember the exact numbers, but they were pretty impressively challenging. Much more energy efficient (although not necessarily space efficient) to store the LOX and LH2 as LOX and LH2.


  12. Anonymous says:

    Likely where we disagree a bit is with respect to what market is there. I don’t see a commercial satellite or large scale science market driving the development of a depot. The cost increment to go from a Delta IV medium to a Delta IV (5,4) and similarly on the Atlas is so low that one would be very hard pressed to justify spending the better fraction of a billion dollars on a depot, not to mention the upgrades to the upper stages to enable use of such a depot. Yes, there are maybe 1 or 2 science mission per decade that could benefit, but they typically don’t have much money for launch. Witness no science mission is riding on a Delta IV HLV, they are all stoping at the “affordable” Atlas 541 or 551 (Pluto, Juno, MSL). The Delta HLV could allow them ~30% growth.

    NASA’s exploration is the sole market that I see as being sufficiently large to justify depots. For this market you are talking 100 mT of LO2 per lunar mission and on the order of 15 mT of LH2.

    The transportation can go forward as NASA currently defines it and still benefit from a depot by being able to top off the EDS tanks, likely requiring both LH2 and LO2. This would close the performance gap.

    Alternatively, one could off load half of the LO2 leaving strictly what was needed to get to orbit. This would hugely increase the rest of the payload that could be launched on ARES V, allowing crew launch on this mission while still closing the performance gap. The LO2 would be topped off at the depot. With the single launch, short time durations, there would be no need for LH2 at the depot.

    There are many other archetectures that take advantage of of depots, some LO2 or LH2 only, some requireing both. If you really need both LO2 and LH2 you can send up a similar, co-located LH2 depot. There really isn’t anything magic about having the 2 propellants joined in a common depot, but it likely mandates multiple launches and orbital assembly to provide the volumes required for exploration.

  13. Anonymous says:

    “That warm GH2 can be … combined with oxygen in a fuel cell to provide water”

    and power. It just occured to me that if both O2 and H2 are stored, you may have all the power you need just from boil-off.

    – renclod

  14. Anonymous says:

    It is quite easy to achieve ZBO for O2, thus to combine it with H2 to feed a fuel cell, one would have to purposly “vent” 6 lb of O2 for every 1 lb of H2 boiled-off. For any length of time beyond a few days this doesn’t trade very well compared to solar.

  15. Lampyridae says:

    The propellant settling could be achieved with a tether and using tidal forces. LOX at one end, LH at the other, perhaps, or mixed tanks with a counterweight (like an Ares V core booster) at the other.

    Probably the best bet is to simply dock at one of the nodes (only a few milliG’s to worry about) and let propellant settle into your departure stage’s tanks. Trickier, perhaps, and maybe not worth simply using thrusters.

    Anyway, I agree with you on ZBO and using LH on-orbit. These are critical techs if you want to go to Mars and are worth taking the time to develop now. There are probably orbits you could design and maintain with minimal propellant where the depot spends more than 50% of its time in the Earth’s shadow but I doubt that trades well against the extra propellant to reach that high.

  16. Lampyridae says:


    With regard to the market for a propellant depot, this is where a lunar COTS-type contract could shine. This is also something the Europeans might try. I doubt they will field a big booster, cautious lot that they are, but a propellant depot would save them Euros as well as enable bigger robotic science missions. There are no indications yet of any willingness to head in that direction, but they never mention such grand things to avoid scaring the member states. (You want more money for WHAT?)

  17. nick says:

    Jon: there’s also a market for N2O4 and Hydrazine as well.

    You’ve hit on the real opportunity in orbital depots. This is orders of magnitude more straightforward than the probably >$10b LOX/LH2 depots, and there is a substantial non-NASA market. Both the DoD and commercial comsats make heavy use of N2O4 and hydrazine for maneuvering and stationkeeping.

    Either the tug or tanker approaches could be implemented with a single launch, for an overall cost of about $100m (two orders of magnitude less than the LH2/LO2 depot) and the technology has already been developed by Orbital Express. No fancy active cryogenics are needed to store these storable propellants for long periods. The biggest problem, and this is the kind of problem for which an entrepreneur not beholden to any one bureaucracy can shine, is convincing satellite makers to make spysats (which need to be highly maneuverable), comsats (which often prematurely end their lives by running out of stationkeeping propellant), and so on to design their craft to be refuelable.

    An example of how an entrepreneur might reorganize how business gets done in order to take advantage of propellant depots: form a new company that sells comsat operators (i.e. communications companies, not space companies) directly on longer-lived comsats, does the overall architecture design for the comsat system, but subcontracts out to the usual satellite builders and launch providers. This turns the problem around — instead of engineers selling only an untested architecture trying to convince the satellite builders to change their ways and add refueling capability, incentivize them to do so if they want to get juicy next-generation comsat contracts. It’s far easier to sell a resulting feature (longer life) to the person who directly benefits from the feature (comsat operators) than an implementation method (depots) to somebody who does not (satellite and spacecraft makers).

    Once we’ve proven the economics of depots with storable propellants, i.e. once we’ve learned to crawl, then we can learn to walk with fancy active cryogenics.

  18. Brock says:

    Nick, that sounds like a nice market, but I think you’ve got your boot-straps backwards.

    If all of the comsats and spysats up there now are not refuelable, then the first sattelite you put up that is refuelable would be a market of one customer. You’d have to put up a gas station to refuel just that one guy. I think that would be unprofitable.

    Until a substantial chunk of sats are refuelable I don’t think there’d be a market for Hydrazine. A rocket refueler in LEO though could make money pretty quickly as I understand it.

    Once you’ve got a refueling station for LO2/LH2 you can announce that you’ll upgrade it to include hydrazine later once there’s a market for it.

  19. nick says:

    brock: If all of the comsats and spysats up there now are not refuelable, then the first sattelite you put up that is refuelable would be a market of one customer. You’d have to put up a gas station to refuel just that one guy. I think that would be unprofitable.

    This is just as true for LO/LH. Indeed, it’s more true, because the very industry proposed as the main market for the depot (trips to the moon) doesn’t exist, and probably won’t exist even if NASA gets the funding to go back to the moon, since NASA doesn’t plan to take advantage of depots. It’s a speculation built on top of a speculation built on top of a technology that would take billions to develop. It all depends on both changing the mind of NASA and big funding for a NASA manned lunar base, neither of which seem to be in evidence. And it’s utterly unecessary to demonstrate the economic viability of propellant depots — this can far more easily be shown using storable propellants.

    In the case of comsats and spysats, they are both already large industries, and they both already use storable propellants in large quantities.

    The chicken-and-egg question you highlight is why entrepreneurial skill, rather than the traditional space industry skill of lobbying for government funding, is needed to demonstrate the economic viability of orbital depots. NASA throwing billions of dollars at something usually demonstrates quite the opposite of the economic viability of that something. Even if by magic you get billions of NASA funding for the most fancy active cryogenic depots, that will demonstrate the economic viability of propellant depots to nobody. NASA funding would do for perceptions of the economic viability of depots what the Shuttle did for perceptions of the economic viability of reusable launch vehicles.

    An entrepreneur can, back to the example strategy I explained above, get contracts for a fleet of refuelable satellites, not just one. The entrepreneur gets this, not by selling satellite and spacecraft makers or government agencies, who are quite happy doing what they already know how to do, but by selling end customers, such as communications companies that operate comsats, on the benefits to the end user that refueling brings, such as longer life and the ability to relocate comsats to cover different parts of the globe as markets and customers change. Once the prime contractor has juicy contracts from end customers, spacecraft makers have a strong incentive to fall in line to get the subcontracts. If you want to sell the means, sell the benefits the means provides, and sell it directly to the people who benefit.

    Furthermore, if we insist on having a large dose of government money up-front, we’re far more likely to get it from the DoD, which has a far higher budget for R&D than NASA and needs highly maneuverable spysats with highly uncertain lifetime propellant use. NASA has already rejected space depots where they are most needed and doesn’t have the funding to do them even if it wanted to. That’s why it was primarily the DoD that funded Orbital Express in the first place, and why it will continue to lead in the technology of on-orbit refueling.

    Given the vast differences in both costs and markets, the relative business prospects for storable propellant depots vs. LO/LH depots are not even a close call. If you insist on LH/LO you will be limiting yourself to government-funded studies (no on-orbit hardware) that will be long since obsolete by the time the business case for them might have become viable.

    In the long term, there are a ton of alternatives to LH/LO propellants. As other posters have pointed out, we may be using VASIMIR, advanced arcjets, solar or nuclear thermal, or any number of other technologies using other propellants, which may or may not be cryogenic.

    The main long-term benefit of depots is facilating use of ISRU propellant. When we use sufficiently cheap ISRU propellant, whether from the moon or asteroids, the economics of rockets change radically. Where power is more expensive than reaction mass (for example where solar mirrors or panels are launched from earth to power ISRU-propelled thermal or VASIMIR rocket engines), higher specific impulse is bad: the power needed increases as the square of exhaust velocity. That means when we are using cheap ISRU propellant hydrogen is the worst, not the best, propellant for thermal or VASIMIR engines. Instead the most important criteria for a propellant becomes its storability, to minimize tankage factor which along with power source is the remaining dominant source of cost in the cheap ISRU propellant regime. LO/LH and LH2 have high tankage factors and are thus extremely poor choices for cheap ISRU propellant.

    To decide today that LO/LH will be the sole possibility we should all focus on, at the expense of much less costly and nearer-term ways to demonstrate the economics of refueling and depots, and at the expense of a wide variety of long-term alternatives, is a classical case of premature optimization. It’s nice that Beinhoff and Zielger got government funding to do these two studies, but two studies do not an industry make and are a horrible excuse for foreclosing exploration of the many other possibilities.

  20. nick says:

    More on what our entrepreneur will be discovering and selling here: gas stations grant real options . In contrast to vague references to “flexibility”, real options, like financial options, can be analyzed mathematically and often have substantial financial value. Refuelable spacecraft and propellant depots will be part of a broad change from fixed plans to the discovery and enablement of real options in space operations.

Leave a Reply

Your email address will not be published.