[LINK] Finally up to date on the Shuttle
Mon, 03 Feb 2003 13:26:55 +1100
I recall from Richard Feynman's account that the whole of NASA was under
pressure to pretend that the risk of failure was 1/1000 or less, because
they had all been convinced that if they admitted it was any riskier,
that Congress would cancel the entire project. So I guess they figured
a lie was better than no Space Shuttle at all.
But this caused denial of real problems.
The engineers at Morton Thiokol, who made the solid fuel boosters,
looked at the various partial O-ring failures and found them to be
strongly correlated with low temperatures at launch. The planned launch
was in much colder conditions, with ice on the launch pad. I recall
that they tried to get word upwards in their own company and in NASA, in
the day or so before the launch, but were not successful in halting it.
There was intense political pressure on the launch - an election due, a
popular woman teacher flying into space etc.
The O-ring is a 37 food circumference rubber ring, as thick as your
little finger and is one of two which seals the re-usable steel sections
of the solid fuel booster. When the booster is ignited, the gap expands
and the O-ring must expand instantly and maintain pressure in the larger
gap, otherwise putty flows past it, and the flame follows the displaced
putty and burns through the gap between the cylinders, burning a hole in
the side of the rocket, which is what happened in the Challenger launch.
But rubber's ability to rebound instantly when the gap is increased is
severely curtailed by cold temperatures, as Richard Feynman demonstrated
with some samples of an O-ring, some iced water and some C (we call then
G) clamps in the Inquiry's Committee room.
Richard Feynman also notes how the Mercury, Gemini and Apollo projects
were built from the ground up by engineers - each component optimised
and robustly engineered and then used as the basis for the next biggest
part of the design.
But the Space Shuttle was specified by committee, with assumptions in
many key areas that in the process of designing it, that engineers would
be able to exceed current engineering limits by 10, 20% or so.
Consequently, they built things which were too highly stressed, such as
turbo pump for the rocket fuel, which was smaller than anyone would
normally have designed it. But it fractured after a few runs, so they
had to build a bigger one. But then, that meant they had to redesign
everything else, because the rest of the design assumed the too-small
size of the committee-generated turbo-pump specification.
There's a lot to learn from all this. It is in Richard Feynman's "What
Do YOU Care What Other People Think? - Further Adventures of a Curious
Character". Along with many other fascinating things, there is his
account of the Challenger enquiry, with detailed photos and diagrams of
the O-ring and all the stresses on the solid fuel booster. Likewise, I
really enjoy "Surely You're Joking, Mr Feynman". The erudite physicist
whose lectures are highly renowned - and which I want to read and listen
to soon - who plays bongos, looks at an atom bomb without covering his
eyes, gets to the bottom of the mechanical and organisational failures
which caused the Challenger destruction . . . a great man!