Comments on: X-Prize cup 2007, a test for reusable rocketry

By: Jonathan Goff

Jonathan Goff — Tue, 06 Nov 2007 20:21:34 +0000

James,
I wonder if there is something in the nature of the problem that makes this pattern different for rocket vehicle/systems.

My guess would be that automobiles have a large and growing market, whereas large rocket systems have almost always been financially losing (or marginal at best) propositions. If we got to a point again where there was significantly growing demand and growing capacity for rocket systems, I think you’d see more university research being pulled in. But as it is right now, the market’s fairly stagnant.

~Jon

By: James Kempf

James Kempf — Tue, 06 Nov 2007 02:27:07 +0000

The analogy with the Darpa Grand Challenge occured to me also. It is interesting comparing the two, in particular, how different the general background of the competing teams is. The Grand Challenge consists mostly of university teams with sponsoring by car companies. The Lunar Lander Challenge consists mainly of small startup companies which, with the exception of Armadillo, are financially somewhat marginal. In one case (Paul Breed’s Unreasonable Rocket team), it is basically a guy and his son. There were something like 20 teams in the Grand Challenge, only one ended up competing in the Lunar Lander Challenge.

This comparison is not to denigate the teams seeking to compete in the Lunar Lander Challenge, they are all dedicated and their ability to make progress with very limited resources is astonishing. I am a great fan of Paul’s blog and the Armadillo updates, and I wish them both the best of success in the next round, and Masten also. I also wish Masten would be more forthcoming about their technical progress, but I assume they have their reasons for not publishing more.

I think the comparison says more about the sad state of research on rocketry than about the competence or ability of the teams, to say nothing of the difficulty of the challenge. Imagine if Lockmart or Boeing were to sponser a university team to compete in the Lunar Lander Challenge the way GM and Volkswagen did for the Grand Challenge, for the win (CMU) and show (Stanford) entries. The teams could benefit enormously from the internal technical knowledge of these companies, to say nothing of the financial resources, such a sponsorship would provide.

I don’t quite understand why this is so, one reason perhaps is that universities in this country don’t seem to do much hands on research in rocketry anymore, though there does seem to be quite a bit of work on computational fluid dynamics for things like pulse detonation engines, hypersonics, and the like, with an occasional subscale demo thrown in if NASA funding becomes available. But research on the system problems of building reliable, reusable rocket vehicles seems completely uninteresting. In other areas of technology, real breakthroughs in research often come from universities, even when small startups ultimately end up commercializing the technology. I wonder if there is something in the nature of the problem that makes this pattern different for rocket vehicle/systems.

By: gravityloss

gravityloss — Mon, 05 Nov 2007 15:50:53 +0000

Thanks for the correction and insight, I edited the original post.
And yeah, I hope you’ll do an update soon. I’m especially interested in the control system stuff as it’s something I plan on working with on my own since without the resources of a full blown rocket garage, propulsion is kinda hard.

By: Jonathan Goff

Jonathan Goff — Mon, 05 Nov 2007 15:18:50 +0000

Gravityloss,
One quick correction. Our first hard start was after about 300-350 runs, not thirty. By that time we had tested at two different test sites with drastically different climates, we had tested in rain, sleet, snow, and dust storms. Our two sites had substantially different altitudes, etc. But the problem wasn’t related to any of that. It turned out to be a marginal igniter interlock design. It worked fine most of the time, but every once in a while the igniter would be strong enough to trip the ignition detection system, without actually being strong enough to light the engine fast enough to avoid an overpressure. It took us another 50 firings to reduplicate the incident and get enough data to really solve the problem (we had originally thought it was a stuck LOX valve). Since then we’ve had 300 more firings (at least), without any repeat of the hard start. But even then we still find new issues and squawks from time to time.

People that test an engine one or two dozen times and think it’s reliable are being a bit naive.

As for control algorithm issues, it’s not exactly what you think. Hopefully we’ll be posting a long update soon. We’re a bit past due for one, aren’t we?