It’s been a little while since I last posted anything to the blog. Michael Mealling suggested that after the RTM articles, it might be a good idea to take a break for a week or so, in order to not burn myself out. Between that piece of advice, the fact that the ideas I wanted to write about last week are only about half-way baked, and how many trips we needed to make to our remote test site for my day job last week, blogging has been pretty light. We’re still going to be pretty busy over the next few weeks, but I’ll try to find some time to post occasionally.
Anyhow, I’ve noticed that a lot of my posts recently have been rather negative about NASA, and not particularly constructive. While I probably am not going to stop bringing up what I see as legitimate concerns with NASA’s policy choices, I figured it was about time to post something a wee bit more constructive.
I wanted to take a few minutes to briefly describe one possible way that NASA could better organize it’s return to the moon. It probably is still a far cry from perfect, probably isn’t as purely libertarian as I would prefer, and may not have a chance in heck of ever happening, but I figured it would be worthwhile to put a few thoughts on the table.
A Modest Proposal
The basic idea is to have NASA change the lunar transportation architecture from a closed, shuttle-derived system, to an open, commercially launched architecture.
Instead of developing two new heavy lift launch vehicles, NASA should go with an architecture that dry-launches the various components, and then refuels them on-orbit using fuel modules boosted by existing or future launch vehicles. The lunar lander module would be a single stage lunar orbit to surface and back vehicle instead of the two stage, Apollo-esque system they have now.
By using on-orbit refueling, and by having the lunar lander return to lunar orbit, the vehicles could be reused multiple times, greatly reducing the costs per mission. This style of architecture is also more able to take advantage of lunar derived propellants if they become available.
By putting the propellant launch needs out on the open market, the price of launching it will be driven down substantially, allowing for more ambitious projects. There is a current glut in the 10-20 ton to LEO launch vehicle market, and prices would drop considerably with higher utilization. The high launch prices currently bandied about by SDV supporters are high precisely because these vehicles are used far less frequently than the infrastructure was designed to handle. Higher flight rates would drive the per launch prices down substantially compared to their current place, and with competition from companies like SpaceX, launch prices might actually start getting close to where they need to be for a sustainable architecture.
This architecture has the benefit that it can rely on existing launchers, while still being able to take advantage of future lower-cost launch vehicles if they become available. Between Atlas V and Delta IV, there is plenty of currently idle US launch capacity that could be used. SpaceX with their Falcon V and IX could eventually be tapped, as could Zenit, Proton, Soyuz, or Arianne V if the government were willing to purchase launch on an international market. This could also encourage future players with RLVs or low cost ELVs to enter the market, without putting them on the critical path like Griffin is afraid of. In contrast, if someone came up with a magical RLV that cost $50/lb to orbit next year, NASA would have to completely change its whole architecture to benefit from it.
There is an additional benefit in that by going with multiple suppliers, the program has less programatic risk of any one supplier going out of business or having to stand-down their launcher due to a launch accident. Not to mention avoiding the programatic risk of having the sole launch vehicle canceled due to cost, like the Saturn V.
If NASA doesn’t trust the commercial sector to provide on-orbit refueling capacity, they could develop a docking and propellant transfer module that could be integrated into otherwise dumb propellant tanks that could then be launched by commercial providers. The commercial provider would launch the tank with module into LEO, then the module itself would perform the docking and propellant transfer maneuvers, followed by a deorbit burn after the propellant has been transfered. A propellant depot in LEO isn’t required, but might make logistics easier in the long term. While using a NASA designed docking and propellant transfer module is probably not as cost effective as having an all commercial system with NASA only defining the docking interface, it at least gets rid of the “last mile problem” and the difficulties of dealing with NASA proximity ops bureaucracy.
Anyhow, this is just a few initial thoughts. I may flesh out various parts of this as time permits, but I wanted to put the basic idea out in the open for some discussion.