Space disasters still have lessons to teach

http://www.msnbc.msn.com/id/41279886/ns/technology_and_science-space/


The Apollo fire, on Jan. 27, 1967 , was made possible by a decision to believe that flammability in pressurized pure oxygen couldn’t be that bad, even if no tests had been run to check out that convenient assumption.

The loss of Challenger, on Jan. 28, 1986 , was made possible by the decision that even though flexible O-ring seals had never been verified to function properly at sub-freezing temperatures, it was convenient to assume they would still do so in the absence of tests proving otherwise.

The loss of Columbia, on Feb. 1, 2003, was made possible by the decision — set down in writing — that even though observers suspected that tank debris might have hit the panel-covered leading edge of the shuttle's left wing, those panels were "probably" just as tough as other tile-covered areas that were previously hit. As was the case for the earlier tragedies, this was a claim that had never been tested.

Appallingly, in hindsight, when tests were made, they showed hazards that should never have been "assumed away." Pure oxygen fires were horrifically violent. Flexible pressure seals at low temperatures did not seat properly in their slots when hit by rocket thrust. And a leading-edge panel hit by a flying piece of insulation didn’t just get scratched — it shattered.