Slashdot Mirror
Bob Ebeling, Challenger Engineer Who Forewarned of Shuttle Disaster, Dead At 89 (huffingtonpost.com)
An anonymous reader quotes a report from HuffingtonPost: For three decades, retired NASA engineer Bob Ebeling blamed himself for being unable to stop the 1986 launch of space shuttle Challenger. He had warned that the shuttle might explode, and it did shortly after liftoff, killing seven crew members. Ebeling was one of five engineers at a NASA contractor then called Morton Thiokol who warned the space agency that cold temperatures predicated at the time of the launch could prove disastrous. The warning was ignored. The night before the launch, Ebeling reportedly told his wife, Darlene, "It's going to blow up." He told another daughter, Kathy Ebeling, that he had toyed with the idea of bringing his hunting rifle to work to threaten NASA not to launch, according to an article last month in The Washington Post. In the final weeks of his life, however, thanks to an outpouring of support following a National Public Radio story in January on the 30th anniversary of the disaster, Ebeling, 89, finally found peace. Ebeling died Monday in his home in Brigham City, Utah, after a prolonged illness with prostate cancer, NPR reported.
There should have been, at NASA, a launch validation team composed entirely of top-notch mid-to-senior level engineers and scientists.
They should carefully consider each known risk prior to each launch.
They should debate it only in terms of risk level = probability of occurrence x probability distribution of consequence severity.
That team should make the go/no-go call, fully documenting their reasons.
Any divergence from this sort of technical review with final authority is a gross violation of responsible process for something as complex as this.
Where are we going and why are we in a handbasket?
Not only did they ignore the warnings, in some cases they directly ignored the protections in place to protect against unsafe launches. A few years back, I had the privilege of being seated next to Roger Boisjoly, another of the Morton Thiokol whistleblowers, who was to be a guest lecturer for 650 engineering ethics students at Texas A&M University the following day. It was fascinating to hear him describe his firsthand account of the conference calls and back-and-forth taking place the night before the disaster.
From what I recall of what he said, prior to every launch, NASA required that Morton Thiokol engineers sign off on their systems, and one of those sign-offs fell to him, but he refused to sign anything due to the concerns he had about the O-ring in cold temperatures. While Morton Thiokol management tried to convince him to change his mind, they were on a conference call with NASA, who was asking what the delay was about. Morton Thiokol management played it off as a minor issue on their end that was being worked out (i.e. "He's driving into the office right now...just give us a minute" sort of stuff). When they were unable to convince him to sign it, his non-engineer manager relieved him of duty and signed-off on the launch himself, completely contrary to protocol.
NASA accepted it regardless of that fact, and the rest is history.
You only learned that 'lesson', because Bob Ebeling and the other Morton Thiokol engineers only tell half the story - the half that makes them look good.
They don't tell you their initial design was flawed. The don't tell you that they 'fixed' it by putting a band aid over it. They don't tell you that the flaw resurfaced when the Shuttle began flying. They don't tell you that the flaw was bad enough that a Challenger type loss was possible even well within the temperature spec. They don't tell you they papered over the problems because the design "hadn't actually completely failed". When they tell the tale of their too little, too late attempts to reverse themselves, they leave out all these things.
Bob Ebeling and the other Morton Thiokol engineers should feel responsible for the death of the Challenger's crew - because they are. The loss of Challenger is every bit as much a tale of engineers fucking up as it is of managers failing to adequately oversee them.
Posted Anon mostly because I don't want to deal with the clueless morons who only know the urban legend version of Challenger's loss (that is, they haven't actually read and understood the Rogers Commission Report - they only repeat what they've heard from other, equally clueless, morons), or with the idiots who worship engineers as a religion.
Mating the segments of the SRBs was a difficult task
It was a needlessly difficult task. The fundamental problem was that the SRB sections would deform while being shipped long distance by train, making both the O-rings and the alignment of the sections critical. They were shipped long distance by train so that they could be manufactured in the district of someone important to funding. Earmarking was the root cause. Make the SRBs on-site and avoid the need for O-rings entirely.
The alignment problem was aggravated by really poor markings on the sections, because the "usability" of the alignment process was ignored, leaving the techs stuck trying to line up these small and cryptic markings.
Feynman's book on all of this was a great read.
Socialism: a lie told by totalitarians and believed by fools.