(A guest post from Leila Johnston.)
It’s almost exactly 25 years since one of the most tragic incidents in the history of space flight. Mission STS-51-L was to be the 25th Space Shuttle mission, an event surrounded by tremendous excitement and optimism. Children around the world knew that a schoolteacher was on board – a thrilling prospect that made the whole thing so much more human. But Sharon Christa McAuliffe never made it into space. With the world watching, the craft disintegrated just 73 seconds after launch, and all seven crew members perished in a twist of smoke. The story of Challenger is so tragic that it still feels very difficult to even look at the pre-flight footage or publicity shots. It was also avoidable.
To any casual observer, things were looking good. Challenger was a workhorse. It flew the majority of NASA’s manned missions to space between 1983 and 85, safely transmitting the first American woman, the first African-American, and the first Canadian into space. No one expected it complete its list of records this way, and everyone was looking for answers.
The Rogers Commission was assembled to determine the fault. This curious panel of experts featuring celebrity physicist Richard P Feynman, Neil Armstrong and Chuck Yeager, concluded it was all down to the failure of an O-ring seal on Challenger’s right Solid Rocket Booster. As the seal failed due to a design flaw, an enormous flare burned through the SRB’s attaching struts, freeing the booster to pivot and knocking down the disasterous domino stack that lead to the propellants igniting, and the craft falling apart.
NASA came under scrutiny. Feynman, in particular, was vocal about their inaccurate safety ratings and famously demonstrated that O-rings are weakened by the cold during a televised hearing. And it was cold on the night of January 28th, 1986. Very cold. At -3°C, in fact, it was below the qualifying limit for many of the major shuttle components. So how could this have been allowed to happen?
I asked a rocket scientist to explain it to me. Ian researched the Challenger analysis for his degree project in the 80s, and now works on propulsion design at Roxel. Over to him.
The technical reason for the failure was simple. It was a very cold night at Cape Canaveral, an elastomeric seal stiffened up and didn’t seal properly, allowing high pressure gas at 3500°C to escape from one of the booster combustion chambers and impinge on the liquid hydrogen tank. It survived for 2 minutes, then the tank wall burned through, nearly 100 tonnes of liquid hydrogen was released and ignited to create a massive fireball.
The real question is why were the engineers overruled. The day before a launch NASA would hold individual meetings called Flight Readiness Reviews (FFR) with the various subcontractors. These meetings were actually telephone or video conferences, and at the end of the FFR each subcontractors had to issue a recommendation to launch – or not to launch.
Morton Thiocol (MTI) manufactured the rocket boosters. Six months earlier, a memo was sent by Roger Boisjoly to MTI Vice President, Enginering, including the comment: “It is my honest and very real fear that if we do not take immediate action […] we stand in jeopardy of losing a flight along with all the launch pad facilities.”
The manager was Joe Kilminster Vice President for Space Bosters at MTI and his statement from the FRR was pretty clear: “MTI recommends STS-51L launch proceeds.”
There were lots of unspoken, but still real commercial and political pressures on the MTI management to launch, particularly as STS-51L had been delayed several times already and further delays would cause problems for the next shuttle which was waiting to take its place on its launch pad. There was a “race” with the Europeans to intercept Halley’s Comet – further delays would mean the rendezvous would be missed. President Reagan was addressing the nation that night and wanted to speak to Christa McAuliffe – the teacher in space – live on TV.
The engineers knew that this seal had failed before, there were four on each booster – and two boosters. Ten times seals had failed, but each time the back-up seal worked. There were another nineteen times when the seals were damaged but held. The engineers knew that this shuttle had stood in the rain on the pad and water may have penetrated the joints. They also knew that it would be colder that night than the seals were ever designed to work at, and colder than any other mission.
At the FRR Boisjoly produced a vu-foil stating that: “O-ring temperature must be ? 53°F (11.7°C) at launch Development motors […] SRM 15 worked at 53°F.”
NASA was reported to be “appalled” at the suggestion to delay the launch and the MTI managers then adjourned the FFR for private discussions within MTI. The techies tried to argue their case. In fact, they argued so forcefully that Boisjoly noted afterwards that it may have cost him his job. Kilminster and his boss Jerald mason were not receptive to the technical arguments. Joe Kilminster prepared and signed a one-page note stating that even if the primary seal failed, the secondary seal would hold. They went back on line to NASA, and transmitted the note ending with:
“MTI recommends STS-51L launch proceed on 28 January 1986. SRM-25 will not be significantly different from SRM 15.”
Needless to say, it was. The temperature that night was 14 degrees lower than any previous firing. The elastomeric seals were too stiff, the gas leaked the shuttle blew up.
Roger Boisjoly blew the whistle and gave evidence at the Presidential Commission. He was awarded the Prize for Scientific Freedom and Responsibility by the American Association for the Advancement of Science. Perhaps not surprisingly, he could see no future working for Morton Thiokol and resigned. He has since become a speaker on workplace ethics and a folk hero for us downtrodden techies.
Leila Johnston is the editor of Hackers!, the newspaper for ‘Makers, Players and Explorers’. Issue 2 featuring rocket science curios, and all kinds of technical and cultural interestingness (including ideas from the wonderful Mr Reid) will be out in March: Hackerspaper.com.