Sep 08, 2004 17:32
Well, that didn't quite go as planned. I'm surprised neither chute opened though - I would have thought that even if the drogue failed to deploy that the main chute would have at least made an attempt to deploy, even if it couldn't sustain the shock or the tumbling. But there was no sign of either.
Which means that either something failed on board in a way that prevented either chute deploying (power failure to the pyros?) or the main chute can't deploy if the drogue hasn't deployed (which could be software - the program never gets into the "main chute" loop) or hardware (perhaps the chute can't deploy if the cover didn't come off with the drogue).
Anyway, whatever, it's a bit of a bummer. Hopefully it will turn out to be a "couldn't be helped" random problem rather than the kind of "they should have spotted it" problems that occured with Mars Polar Lander... that one crashed because (if I recall rightly) a variable wasn't cleared going into a subroutine after the landing legs were swung out. If the legs "bounced" then a variable got set at that point, and not reset before the program got into the "looking for touchdown" routine. The set variable was the "legs have touched something" variable... so it thought it had touched down while still quite a long way up. It wasn't noticed in testing because it only happened on fairly rare occasions. Unfortunately, the descent to Mars was one of them.