I think I addressed some of your comments in my general message to both lists (ERPS & aRocket) but I'll respond to a couple of points below:
On Fri, 26 Sep 2003, Randall Clague wrote: > On Fri, 26 Sep 2003 10:48:50 -0700 (PDT), "David J. McCue" > <[EMAIL PROTECTED]> wrote: > > >I find the mean-spirited nature of this comment diffficult to understand. > > Dave, I'm sorry I sounded mean-spirited to you. I had and have no ill > intentions toward you, RRS, or CSULB. Come on, you know me better > than that. > > My comment was meant to express my surprise and dismay that someone > would fly an expensive rocket on an engine that had only been > successfully tested once. One success shouldn't give you enough > confidence in a new design that you are willing to risk very much on > its repeated success under new and more stringent conditions. To us, time and limited opportunity to fly are our biggest enemies. Students have few chances to participate before they move on. That changes our risk/benefit assessment. Rockets (airframes) are not much more dear to us than our test stand. Recovery of a vehicle is a pleasant novelty that allows us to be more adventurous. We are actually pretty damn good at recovery, by the way. > I see two alternative histories: 1) do more static tests before > committing to flight; 2) fly the student payload on a conventional HPR > with a conventional recovery system, fly the experimental recovery > system on an HPR with nothing else aboard, and fly the aerospike > engine on as plain vanilla vehicle as you can build, using a > conventional recovery system. AIUI, the student payload was destroyed > because the engine malfed (to be expected on a first flight, and no > big by itself) _and_ the recovery system failed. The success or failure of an engine has little to do with recovery. A cato on the pad usually results in a satisfyingly big fireball, but most of the rocket is probably just scorched and nothing more. Again, whether an engine works perfectly or fails in flight is irrelevant to successful recovery. If the 'chute comes out, you're fine - if it doesn't, it don't matter how well the engine worked, you're gonna make a hole in the sand... Finally, we don't do HPR. If you want to fly a payload with us, it flies on our EX rocket. As I mentioned in my longer message, I do wish that the high school kids had a better experience. I'd be happy to give their payload a ride in one of my HPR rockets, but that's separate from what CSULB is doing. > My advice going forward is simply to put your eggs in separate > baskets, at least until you're 95% confident that your baskets are 50% > reliable or better. Once all three systems are working, integrate > them into a single vehicle. It's more work, and it costs more and > takes longer. But IMHO it's your only realistic shot at success. > > -R I disagree with your analysis. In my experience, the reliability of the engine has little to do with recovery of the vehicle or payload. As I stated above, if it leaves the pad, getting the rocket back is really all a function of the recovery system working. I also have trouble with your overall approach of dividing up problems and trying to solve them one at a time. I would argue that it takes more time, money and *does not* increase our chances of success, because if we test two things on one flight, we can learn from both of those tests at once. One must consider the interaction of two or more simultaneous tests, but we seem to be conducting multiple experiments every time we fly, planned or not. ;-) Case in point, we learned about a critical tolerance in the engine assembly that must be monitored during manufacture. We also learned that there was a component in the recovery system that could be assembled in either of two ways, a right way and a wrong way. The designer failed to make this clear to the person responsible for integration on flight day, and we crashed. Both problems appeared during this flight, and we learned about both at a cost of one flight. Yes, the rocket was damaged and a payload lost, but that was the result of one factor alone: recovery. We didn't expect a problem with the recovery system, the only thing that could jepordize the payload. The malfunction was not a result of the change in design, but an error in assembly. The pyro design might have overcome the problem where the pneumatic system did not, but the main reason we're going back to pyro recovery is that we can easily incorporate two independent initiator systems controlled by redundant altimeters. I'd like to hear why you think I'm wrong, because many of us are trying to figure out the best way to get to a combination of things that work well - in fact, I'm gonna guess that folks over on aRocket would like to bat this one around as well, so I'll crosspost this -- -Dave Mc _______________________________________________ ERPS-list mailing list [EMAIL PROTECTED] http://lists.erps.org/mailman/listinfo/erps-list
