🚀 go-pugleaf

On Mon, 06 Aug 2012 21:26:55 -0400, Walter Banks
<walter@bytecraft.com> wrote:

>Jon Kirwan wrote:
>
>> On Mon, 06 Aug 2012 16:15:32 +1000, Don McKenzie <5V@2.5A>
>> wrote:
>>
>> >Been there, done that :-)
>> >
>> >http://www.abc.net.au/news/2012-08-06/nasa-celebrates-successful-mars-landing/4180454
>>
>> I started watching the NASA channel about 4 hours ago and
>> stayed with it. (Still watching, while they are doing the
>> news conference event and just finished congradulating each
>> member of the EDL team (entry, descent, and landing.)
>>
>> Australia was part of this success, as well -- at least in
>> terms of participating in the very much needed communications
>> portions.
>>
>> What impresses me the most is that this trip was about 350
>> million miles, ending in a tight coordination with two
>> orbiting satellites, the ODY and MRO, and deploying novel
>> technologies to land a 1 ton vehicle (if I heard correct.)
>> Hard to believe that all of this could come together as well
>> as it did.
>
>I like many watch the landing  coverage. There are two
>things that really impressed me.
>
>1) The navigation to get them there on a really small landing
>    footprint
>
>2) With all of the technology in the systems to land this
>successfully, one number still is impressive. There were still
>250 single point possible failures. The math against it working
>was astronomical. If there was anyone involved reading
>this I am all ears and you have my congratulations.
>
>Listing to the systems readout after the landing showed it
>almost everything was nominal.
>
>Congratulations all

Of all those single points of failure, the ones that worried
me the most were the pyros. You can test a lot of things --
like the 65,000 pounds of force the parachute had to bear.
But you can't test the ACTUAL pyros you will use. No matter
what you do, the ones you place in there can't have been
tested. We know how to make explosives of great uniformity,
of course. But all of that has to go into a system and it
must fire exactly correctly, under buffeting circumstances,
without a single point of failure in a single pyro. Just one
and that is it. Not that the rest wasn't also difficult. But
some of the things, since as the novel use of an imbalance in
weight distribution in order to permit direction control
during entry, can have errors in one part of the software be
compensated by the outer control loop in another part. So
even there, there is a backup hope. But the pyros either
work, or don't. That's what I was watching mostly for, though
the rest was also sincerely knuckle-whitening as well.

Jon