stewart@JPL-VLSI.ARPA (02/11/86)
I was just reading through all the bulletins that have collected in the last 2 weeks, when I came upon an article that said that the landing and not the launch was the phase of the mission most likely to have a cata- strophic failure. I had thought the opposite to be true. Can someone set the record straight. I've managed to delete the message so cannot address the poster directly.
elt@astrovax.UUCP (Ed Turner) (02/17/86)
The claim that landings are more dangerous than launches may have been based on several risk analyses carried out before the loss of Challenger. I believe that some or perhaps all of these concluded that the landing procedure was quite hazardous. One particularly controversial study (Rand Corp. maybe?) concluded that the chance of destroying the Orbiter on landing is of order 1% per mission. Also there is the empirical fact that there have been a couple of "close calls" during shuttle landings. In any case it is not hard to see why a no power landing in a "plane" as massive as the Shuttle with such poor low speed glide and handling characteristics could be quite tricky. One unhappy possibility that must be considered is that the Shuttle has several critical failure modes, all of roughly the same small probability but adding up to something in the few percent range. Since identification and evaluation of such failure modes and probabilities is always a difficult and uncertain business, one might be left with finding them in the same way that the O-ring/SRB problem (if that's what it was) was uncovered. This sometimes happens with experimental aircraft. Ed Turner astrovax!elt
dietz@SLB-DOLL.CSNET (Paul Dietz) (02/24/86)
Did the RAND study say the dangerous part of landing was the reentry or the touchdown? I wonder what the chances of disaster are if a tile comes off the bottom of the orbiter. The shuttle's problem may be that recognized flaws are difficult to fix, not just that there are many low probability unknown flaws. For example, it's known that there's no abort mode for the first two minutes of flight, but this is a fundamental feature of the shuttle design and cannot be easily changed. One flaw I worry about during liftoffs is catastrophic failure of the SSME turbopumps. These pumps have suffered from cavitation on previous flights; if one of the pumps were to fly apart high velocity metal fragments could go flying through the engine compartment. This was one early speculation about the Challenger accident (since discarded). Continuing upgrades of the shuttle make extrapolations of reliability suspect. For example, if and when the SSME's are upgraded to 109% of rated thrust we won't be able to use previous flights as indicators of their reliability.