I had a brief email exchange with Jeff Greason, CEO of XCOR Aerospace but also (and more importantly in the context of this post) member of the Augustine panel, to get his thoughts on last week’s SLS announcement.
RS: To what degree do you think that Congress’s and (by extension, because they have to, as numerous congresspeople have demanded, “follow the law”) NASA’s current plans conform to what the Augustine panel suggested? And do you think that they are the most effective ways of opening up space?
JG: As far as I can discern NASA’s SLS/MPCV architecture, it seems to fall within the parameters we studied on the Committee. An architecture like that needed something like $12B/year for human spaceflight in order to actually do exploration missions within NASA’s traditional cost structure. Right now it looks like the budget available will be more like $9B or even $8B/year. That suggests that NASA is repeating the Constellation mistake of beginning a development path that cannot be carried to a useful point without an increase in budget. The difference is that the last time we did that, there were serious policymakers who thought the budget might actually be increased, and this time, I don’t think there are.
I’m not saying that SLS can’t be built — only that if it is built, the cost of keeping it operating will be so high that NASA’s budget won’t support developing, for example, planetary landers that would be needed to make it useful.
Therefore, I don’t think the SLS approach is a very effective way to open up space, nor to show national leadership. Part of what it takes to do something effective is to match the program approach to the budget available. If NASA had a top-line budget of $25B/year and that could be sustained, this might be an executable approach. At $18B/year or less, I don’t think it is.
RS: Ignoring the issue of whether or not it is executable, is it the best use of NASA’s limited resources, assuming (and I understand this is a generous assumption) that the goal is to actually develop and settle space?
JG: Whether, as I think, opening a space frontier should be NASA’s goal, or whether NASA’s goal is simply human exploration of space, it is hard to see the SLS approach as the best use of limited resources.
Looking simply at exploration, the obvious destinations are low-gravity bodies such as NEOs or Phobos/Deimos, or the Moon, or Mars. Depending on the pace of missions and how the architecture is done, it takes at least 150-200 tons/year for some human exploration, 300-400 tons/year for a fairly robust program, and Mars exploration requires quite a bit more, unless you inject new technologies to reduce it.
The Committee found that while our existing 25-ton launch vehicles were too small or fly too infrequently for a robust human exploration program, 70 tons was ample. We did not thoroughly explore sizes in between but from what I learned there, my opinion is that 35- to 50-ton vehicles, if low cost and scalable to a reasonable flight rate, are probably sufficient. Doing an Apollo-class mission takes two or three launches of 50- to 70-ton vehicles, and with launchers that size propellant depots aren’t mandatory but it really does help to simply transfer propellant between spacecraft. It is tractable in the near term to do missions that way.
The U.S. can have 50- 70-ton vehicles that come from the same industrial base as existing launch vehicles — Atlas 5 Phase II and Falcon Heavy. NASA could fund both of those vehicles for a total of about $3-$4B, comfortably — a small fraction of SLS development costs. And the annual cost to keep the vehicles flying would probably be under $0.5B/year because they come from the same production lines as other, smaller launchers that have other customers. It is hard to say what SLS will cost to keep operating but based on what I saw on the Committee I would expect more than $2B/year. It would be far cheaper to launch two of the alternative vehicles than one SLS, so I cannot see how SLS offers good value compared to the alternatives.
The capsule picture is a lot more complicated and it is hard to say what makes a good capsule strategy without clear mission requirements. I can envision some architectures under which MPCV might be a good tool in the toolbox.
RS: Regardless of the tendency of those who make decisions on the Hill to favor pork over progress, how do you see things playing out over the next few years in terms of both the development of private space, and space development in general?
JG: Guessing what the future will bring depends a bit on policy choices by government that aren’t very predictable. I think suborbital providers will enter into service in the next few years. Orbital cargo, of course, has been done privately for all customers except NASA since 1986 and more recently all NASA science missions have been launched privately. One of the last bastions is ISS cargo resupply, and I think one or both of the private companies working on that will get there.
Commercial crew service for NASA is an arena where most of the programmatic risk is driven by NASA itself. There’s no doubt that some of the U.S. providers, which includes firms like Boeing and ULA, have the technical capability to provide commercial crew transportation to orbit. Whether NASA will choose to purchase those services on commercially reasonable terms remains to be seen. Commercial crew transportation to orbit is a classic situation where the demand, such as commercial habitats, won’t grow without the supply, and the supply won’t be built without the demand. A well-executed commercial crew program could speed up a commercial market considerably — a badly executed one, or the absence of one, could slow it down.
I am a little surprised that the importance of commercial crew for exploration missions hasn’t been more widely appreciated. Once there is a commercial crew transportation service, it is no longer necessary to “man rate” exploration boosters — you can ferry the crew up to the spacecraft on a different launch. So even if something like SLS were used to launch exploration payloads, NASA’s released information suggests that it would be used for cargo well before would be available for crew. Using commercial crew to orbit as part of an exploration strategy could therefore be a very valuable part of accelerating human exploration, but it is seldom considered in mission architectures that I’ve seen.