by John Gribbin
The Challenger disaster occurred shortly before noon, Eastern Standard Time, on Tuesday, 28 January 1986, when the space shuttle exploded, a little over a minute after lift off, killing all seven crew members. Feynman hadn’t taken much interest in the shuttle programme, having noticed that none of the results from the supposedly important scientific missions that it carried into Earth orbit were ever published in the main scientific journals, and suspecting that the whole exercise was a bit of a boondoggle.6 But like millions of other Americans he saw the lift off and explosion of the Challenger on the TV news.
What Feynman didn’t know was that the Acting Head of NASA was William Graham, who had been a Caltech undergraduate 30 years before and had attended Feynman’s famous Physics X course. Graham had gone on to work at Hughes Aircraft, where he often attended Feynman’s Wednesday lectures, sometimes accompanied by his wife. Graham had the unenviable task of drawing up a shortlist of candidates to participate in the inevitable Presidential Commission looking into the causes of the disaster. Most of the people who ended up on the Commission had some sort of expertise involving the space programme – which, unfortunately, meant that they could not truly be regarded as disinterested investigators, no matter how hard they might try to be impartial. They included Air Force General Donald Kutyna, who was responsible for shuttle operations for the Department of Defense, Sally Ride, the first American woman in space, Neil Armstrong, the first person on the Moon, and other people associated either with NASA or the space programme. The Chairman of the Commission would be William Rogers, a former Secretary of State and Attorney General. When Graham’s wife suggested asking Feynman, a truly independent and original thinker, to join the team, Graham leaped at the idea.7
Graham phoned Feynman to ask if he would be available, not knowing that the request was particularly ill-timed. By now, Feynman had undergone two operations for abdominal cancer, was suffering from heart trouble and had been found to have another rare form of cancer, involving his bone marrow and affecting his blood, making it sticky and prone to clotting. Health concerns aside, he had spent much of his life avoiding responsibility, following his independent path and in particular steering clear of anything to do with Washington. His immediate reaction was to say no. But first he checked with his closest friends, including Gweneth. They all told him that he had to do it, because he could make a unique contribution. As Gweneth put it:
If you don’t do it, there will be twelve people, all in a group, going around from place to place together. But if you join the commission, there will be eleven people – all in a group, going around from place to place together – while the twelfth one runs round all over the place, checking all kinds of unusual things. There probably won’t be anything, but if there is, you’ll find it. There isn’t anyone else who can do that like you can.
She was right.
So Feynman agreed to join the Rogers Commission, only to find that its remit extended far beyond finding the immediate cause of the disaster, to address questions such as ‘What should be our future goals in space?’ He foresaw the possibility that the Commission’s work might be never-ending, and gave himself a deadline he would serve for a maximum of six months, then quit, no matter what. But he would give Washington an honest six months, and wouldn’t do anything else in that time – no teaching, no consultancy with the Thinking Machines Corporation, no physics. As he told Gweneth, ‘I’m gonna commit suicide for six months.’
The call from Graham confirming that Feynman was a member of the Commission came at 4pm on Monday, 3 February. He would be expected in Washington for the first meeting on Wednesday morning, which gave him a whole day to prepare, Feynman fashion, for the task ahead. He arranged with Al Hibbs, one of the friends who had urged him to serve on the Commission, to visit the Jet Propulsion Laboratory for an intense briefing on the shuttle, so that he would be up to speed. He learned a lot that day, but one of the most important things he learned was just about the first. On the second line of his handwritten notes from the briefing, he made the comment ‘O rings show scorching’.
The O rings were part of the two solid fuel booster rockets that helped to launch the shuttle into orbit. The booster rockets are made in cylindrical sections and joined together. The O rings are like huge rubber bands, 37 feet in circumference, that fit into the joint between two sections of rocket and are supposed to seal the joint tight so that hot gas cannot escape through the crack as the fuel burns. After they have done their work, the spent booster rockets separate from the shuttle and fall into the sea, from where they are recovered and refurbished for future use. If the O rings on some of the recovered spent boosters were scorched, that meant that hot gas was escaping from the joint. If the seal failed entirely during a launch, it could cause just the kind of disaster that had engulfed Challenger. But why should the O rings have failed on the Challenger launch on 28 January 1986, and not on any of the previous shuttle launches?
Feynman ‘sucked up information like a sponge’ at JPL, without finding any answer to that question. Then he caught the overnight flight to Washington, and made it to the first meeting of the Commission, in Rogers’ office, on Wednesday, 5 February. Wound up by the intense cramming of the previous day, and by lack of sleep, he was bothered to find that the first meeting was just an informal get-together, and nobody seemed to share his sense of urgency about getting to grips with the real work. On the other hand, he was relieved to learn that the investigations were expected to take no more than 120 days, less than his promised six months.
Although Feynman didn’t know any of the other members of the Commission, he couldn’t help noticing General Kutyna, resplendent in his uniform among the group of civilians and who Feynman happened to sit next to at the first meeting. For once, though, it turned out that there was a human being inside the uniform – Feynman was delighted to find that while many of the commissioners were met by fancy limousines after the meeting, Kutyna was heading off for the Metro.
I thought, ‘This guy, I’m gonna get along with him fine; he’s dressed so fancy, but inside he’s straight. He’s not the kind of general who’s looking for his driver and his special car; he goes back to the Pentagon by the Metro.’ Right away, I liked him.
The feeling was mutual, and Kutyna took Feynman under his wing, showing him the way things worked in the Washington bureaucracy:
Feynman had three things going for him. Number one, tremendous intellect, and that was well known around the world. Second, integrity, and this really came out in the commission. Third, he brought this driving desire to get to the bottom of any mystery. No matter where it took him, he was going to get there, and he was not deterred by any roadblocks in the way. He was a courageous guy, and he wasn’t afraid to say what he meant.8
Kutyna was lucky to find out the kind of person Feynman was, and to establish such a good relationship with him so quickly, because by the end of the week the General would have a problem. He would be given a strong hint about what had caused the Challenger to explode, but the information would come from a sensitive source, a NASA astronaut who could get fired for telling tales out of school. The fact that this was a real possibility is itself, of course, an indictment of the way NASA was being run at the time, but Kutyna knew it wasn’t just paranoia. There had been a previous occasion when one of the astronauts, an old friend of Kutyna, had passed him a document describing how safety procedures had been violated during manufacture of the shuttle segments. The astronaut was seen passing Kutyna the document, and promptly demoted.
Shortly after the Commission started its work, another astronaut told Kutyna about some sensitive information. The contractor involved had been testing O rings under conditions of extreme cold for at least six months prior to the accident. Clearly, there was some concern about what happened to the O rings when they got cold. This was potentially a key piece of information, since the fatal Challenger launch had been the first shuttle lift off to take place when the temperature was below freezing. If the cold was impli
cated in the disaster, perhaps causing a failure of the O rings, then those data ought to be available to the Commission – but they had never been mentioned in the material that was to go before the commissioners. Kutyna desperately needed to get the possibility out in the open, without damaging the career prospects of his astronaut friend. The best way to do this would be if he could steer Feynman, the only truly independent member of the Commission, into thinking about the problem of cold affecting the O rings.
But he would have to be subtle. Right at the beginning of their work on the Commission, Kutyna had given Feynman a personal briefing, at the Pentagon, on the whole space programme, to put their deliberations on the shuttle in perspective. He had offered to get Feynman clearance to have classified information, but Dick had refused, saying, ‘I don’t want to clog my mind with secrets that I can’t talk about. I want to be able to talk about anything that you tell me. So don’t give me anything classified.’9 So there was Kutyna’s dilemma. The only person on the Commission he could trust to put the puzzle of how cold affected the O rings on the agenda was also the only person on the Commission who adamantly refused to be involved in secrecy.
Feynman was also the only person on the Commission who was unaccustomed to the slow pace of their work. After the informal meeting on Wednesday, lasting just a couple of hours, they were free for the rest of the day. On Thursday, in the first public meeting, the commissioners had the opportunity to question senior representatives from NASA. It turned out that all but a couple of the members of the Commission had degrees in science or engineering, and they fired off technical questions that the administrative ‘big cheeses’, as Feynman called them, were not equipped to answer. ‘We’ll get back to you on that’ became the litany of the day. Friday wasn’t much better. Although Kutyna gave the commissioners a rundown on the way an earlier investigation he had worked on, into the causes of a failure of an unmanned Titan rocket, had been carried out, Rogers (one of the few people on the Commission with no technical background) dismissed this practical experience as inappropriate to the shuttle investigation. ‘We won’t be able to use your methods here’, he told Kutyna, ‘because we can’t get as much information as you had.’
To Feynman, this was patently false. Because the Titan was unmanned, it didn’t have anywhere near as much monitoring equipment as the shuttle, and nor had the launch been filmed in close-up for national TV, whereas the pictures from the Challenger launch were good enough to show a flicker of flame coming from the side of one of the booster rockets just before the explosion. It was another frustrating day. ‘Although it looked like we were doing something every day in Washington, we were, in reality, sitting around doing nothing most of the time.’
Then came the weekend. As it turned out, the Commission faced a long weekend break. They were scheduled to go to Florida the following Thursday, to get a briefing from NASA officials and tour the Kennedy Space Center. Such a formal, guided tour had no prospect of providing any real insight into what went on; and even that cosmetic exercise was five days away! Devastated, and on the point of pulling out of the investigation, Feynman called Bill Graham, who had got him into it in the first place, and asked if there were any way he could get to do some real work, talking to engineers, trying to find out what had gone wrong. Graham thought this was a great idea, and offered to arrange for Feynman to visit the Johnson Space Center, as soon as he liked. But Rogers vetoed the proposal. Graham suggested a compromise – Feynman would stay in Washington, but Graham would arrange for NASA experts to give him a briefing at NASA headquarters, right across the street from Feynman’s hotel. At first, Rogers objected to this proposal, too; but eventually gave it a reluctant OK.
So on Saturday Feynman began to get down to some real work on the problem, picking up where he had left off at JPL. When he talked to the expert who knew all about the seals in the joints where the pieces of booster rocket fitted together, it soon became clear that there was a known problem with the O rings, that had been brushed aside, largely (it seemed) through wishful thinking. There had been minor leaks on previous flights, and sometimes parts of the O rings on the recovered boosters had been burnt away. But only a few of the seals had failed, on only some of the flights. NASA’s attitude, as Feynman described it, had been ‘if one of the seals leaks a little and the flight is successful, the problem isn’t so serious’. He likened it to playing Russian roulette. The first time you pull the trigger, the gun doesn’t go off, so you assume it is safe to pull the trigger again, and again, and again …
He actually found a report which began ‘the lack of a good secondary seal in the field joint is most critical’ and concluded ‘analysis of existing data indicates that it is safe to continue flying’. How could it be safe, if the situation were ‘most critical’?
By now, the press were on to the story of the problems with the seals, and the following day, Sunday, a story appeared in the New York Times. One result of this was that Rogers called an emergency meeting of the Commission for Monday, 10 February. Kutyna phoned Feynman at his hotel on the Sunday afternoon, to inform him of the special meeting, and to invite him over to dinner that evening. He had heard the news about cold affecting the O rings almost a week before, and he was still looking for a way to put Feynman on the scent. He found it after dinner, when he was showing Feynman his pride and joy, an Opel GT 1974 that he was working on in the garage. There were a couple of carburettors on the workbench. One of the important components of such a carburettor is a seal formed by a rubber O ring, a miniature version of the shuttle O rings, to stop leaks where two of the subcomponents are joined.
‘You know, Professor Feynman,’ said Kutyna, ‘these damn things leak when it gets cold. Do you suppose cold has any effect on the rubber O rings in the carburettor?’10 It was enough to set Feynman on the trail to what would become his most famous, and public, experiment. Thanks to Kutyna’s hint, he was already thinking about the effect of cold on the O rings when he went along to the special meeting of the Commission on Monday. The first part of the meeting was a waste of time. The ‘exposé’ in the newspapers hadn’t contained any information that Feynman didn’t know already. But then things got interesting. First, they were shown pictures they hadn’t seen before, revealing puffs of smoke coming from one of the joints on the booster rocket before the shuttle had cleared the launch pad. The smoke seemed to be coming from the same place as the flame that had appeared just before the explosion, indicating strongly that the seals had been faulty and leaking from the very beginning of the launch.
Then came a real surprise. An engineer from the Thiokol company, responsible for the seals, spoke to the Commission. He had come on his own initiative, without being invited – if it hadn’t been for the special meeting called because of the newspaper stories, he wouldn’t have found the commissioners there. He told the meeting that the Thiokol engineers were so concerned about the possible effects of cold on the seals that the night before the launch they had advised NASA not to fly the shuttle if the temperature fell below 53° Fahrenheit, the lowest temperature at which the shuttle had flown before. But, said the engineer (Feynman only gives his name as ‘Mr MacDonald’), NASA bullied Thiokol into reconsidering its opposition to the flight, which was launched, fatally, when the temperature was 29° Fahrenheit. Only MacDonald refused to go along. He told the Commission that he had said to his colleagues, ‘If something goes wrong with this flight, I wouldn’t want to stand up in front of a board of inquiry and say that I went ahead and told them to go ahead and fly this thing outside what it was qualified to do.’ MacDonald’s testimony was so stunning that Rogers asked him to repeat the whole story.
There were two aspects to MacDonald’s story. First, it pinpointed cold as the immediate cause of the O ring failure. Kutyna’s hint to Feynman had given him a bit less than 24 hours’ start on the rest of the commissioners, but even without that hint, after MacDonald’s testimony he would have been hot on the trail. Secondly, it showed, as Feynman had begun to suspect afte
r his briefings on Saturday, that there had been two failures. One was a technical failure; the other was a human failure, a failure of management. Even when the engineers had expressed concern, the managers had pressed ahead.
The news was so important that Feynman wanted to get to grips right away with finding out how the properties of the rubber in the O rings were affected by cold. But Rogers decided to call another public meeting for the following day, Tuesday – not to air MacDonald’s news, which he considered too sensitive to go public with just yet, but to go over the old material that had been in the New York Times. The idea was to go over much the same ground that they had covered in closed session on Monday afternoon (and which had been old news to Feynman even then!), but to do it in front of reporters and TV cameras. Feynman hated the thought of wasting more time, when he was now in a position to get some real information about what happened to the O rings under freezing conditions. But he was stuck in a hotel in Washington, distanced from the labs where the necessary experiments could be carried out. It was while eating dinner alone that night that he noticed the glass of ice water on the table, and said to himself, ‘Damn it, I can find out about that rubber … I just have to try it! All I have to do is get a sample of the rubber.’
He knew that there was always ice water available at the Commission meetings, and thought about doing the experiment, for real, while they were all sitting around listening to the same old stuff they had heard already. The idea was irresistible to the showman side of his personality. But first, he needed a sample of the rubber used in the O rings. Once again, he called Graham, who came to the rescue. There was a model of one of the joints at NASA headquarters, which was going to be shown during the open meeting the next day. It contained two strips of the rubber (the O rings were only about as thick as an ordinary pencil, in spite of the importance of their job; what mattered was their flexibility, their ability to squeeze into the tiny gaps that opened up between the joints in the rocket under the stress of take-off, and block the exit of any hot gases). But Feynman would have to get the sample of rubber out of the joint himself.
