This was extremely discouraging. It’s hard to understand now, because NASA has been taking at least two years to put the shuttle back on track. But at the time, I thought it would be a matter of days.
I went over to Mr. Rogers and said, “We’re going to Florida next Thursday. That means we’ve got nothing to do for five days: what’ll I do for five days?”
“Well, what would you have done if you hadn’t been on the commission?”
“I was going to go to Boston to consult, but I canceled it in order to work 100 percent.”
“Well, why don’t you go to Boston for the five days?”
I couldn’t take that. I thought, “I’m dead already! The goddamn thing isn’t working right.” I went back to my hotel, devastated.
Then I thought of Bill Graham, and called him up. “Listen, Bill,” I said. “You got me into this; now you’ve gotta save me: I’m completely depressed; I can’t stand it.”
He says, “What’s the matter?”
“I want to do something! I want to go around and talk to some engineers!”
He says, “Sure! Why not? I’ll arrange a trip for you. You can go wherever you want: you could go to Johnson, you could go to Marshall, or you could go to Kennedy…”
I thought I wouldn’t go to Kennedy, because it would look like I’m rushing to find out everything ahead of the others. Sally Ride worked at Johnson, and had offered to work with me, so I said, “I’ll go to Johnson.”
“Fine,” he says. “I’ll tell David Acheson. He’s a personal friend of Rogers, and he’s a friend of mine. I’m sure everything will be okay.”
Half an hour later, Acheson calls me: “I think it’s a great idea,” he says, “and I told Mr. Rogers so, but he says no. I just don’t know why I can’t convince him.”
Meanwhile, Graham thought of a compromise: I would stay in Washington, and he would get people to come to his office at NASA, right across the street from my hotel. I would get the kind of briefing I wanted, but I wouldn’t be running around.
Then Mr. Rogers calls me: he’s against Graham’s compromise. “We’re all going to Florida next Thursday,” he says.
I say, “If the idea is that we sit and listen to briefings, it won’t work with me. I can work much more efficiently if I talk to engineers directly.”
“We have to proceed in an orderly manner.”
“We’ve had several meetings by now, but we still haven’t been assigned anything to do!”
Rogers says. “Well, do you want me to bother all the other commissioners and call a special meeting for Monday, so we can make such assignments?”
“Well, yes!” I figured our job was to work, and we should be bothered—you know what I mean?
So he changes the subject, naturally. He says, “I understand you don’t like the hotel you’re in. Let me put you in a good hotel.”
“No, thank you; everything is fine with my hotel.”
Pretty soon he tries again, so I say, “Mr. Rogers, my personal comfort is not what I’m concerned with. I’m trying to get to work. I want to do something!”
Finally, Rogers says it’s okay to go across the street to talk to people at NASA.
I was obviously quite a pain in the ass for Mr. Rogers. Later, Graham tried to explain it to me. “Suppose you, as a technical person, were given the job as chairman of a committee to look into some legal question. Your commission is mostly lawyers, and one of them keeps saying, ‘I can work more effectively if I talk directly to other lawyers.’ I assume you’d want to get your bearings first, before letting anybody rush off investigating on his own.”
Much later, I appreciated that there were lots of problems which Mr. Rogers had to address. For example, any piece of information any of us received had to be entered into the record and made available to the other commissioners, so a central library had to be set up. Things like that took time.
On Saturday morning I went to NASA. Graham brought in guys to tell me all about the shuttle. Although they were pretty high up in NASA, the guys were technical.
The first guy told me all about the solid rocket boosters—the propellant, the motor, the whole thing except the seals. He said, “The seals expert will be here this afternoon.”
The next guy told me all about the engine. The basic operation was more or less straightforward, but then there were all kinds of controls, with backing and hauling from pipes, heating from this and that, with high-pressure hydrogen pushing a little propeller which turns something else, which pumps oxygen through a vent valve—that kind of stuff.
It was interesting, and I did my best to understand it, but after a while I told the fella, “That’s as much as I’m going to take, now, on the engine.”
“But there are many problems with the engines that you should hear about,” he says.
I was hot on the trail of the booster rocket, so I said, “I’ll have to put off the main engines till later, when I have more time.”
Then a guy came in to tell me about the orbiter. I felt terrible, because he had come in on a Saturday to see me, and it didn’t look like the orbiter had anything to do with the accident. I was having enough trouble understanding the rest of the shuttle—there’s only a certain amount of information per cubic inch a brain can hold—so I let him tell me some of the stuff, but soon I had to tell him that it was getting too detailed, so we just had a pleasant conversation.
In the afternoon, the seals expert came in—his name was Mr. Weeks—and gave me what amounted to a continuation of my JPL briefing, with still more details.
There’s putty and other things, but the ultimate seal is supposed to be two rubber rings, called O-rings, which are approximately a quarter of an inch thick and lie on a circle 12 feet in diameter—that’s something like 37 feet long.
When the seals were originally designed by the Morton Thiokol Company, it was expected that pressure from the burning propellant would squash the O-rings. But because the joint is stronger than the wall (it’s three times thicker), the wall bows outward, causing the joint to bend a little—enough to lift the rubber O-rings off the seal area. Mr. Weeks told me this phenomenon is called “joint rotation,” and it was discovered very early, before they ever flew the shuttle.
The pieces of rubber in the joints are called O-rings, but they’re not used like normal O-rings are. In ordinary circumstances, such as sealing oil in the motor of an automobile, there are sliding parts and rotating shafts, but the gaps are always the same. An O-ring just sits there, in a fixed position.
But in the case of the shuttle, the gap expands as the pressure builds up in the rocket. And to maintain the seal, the rubber has to expand fast enough to close the gap—and during a launch, the gap opens in a fraction of a second.
FIGURE 7. Joint rotation is caused by pressure from inside the rocket pushing the walls out farther than the joints. A gap opens, and hot gas flows past one or both of the O-rings.
Thus the resilience of the rubber became a very essential part of the design.
When the Thiokol engineers were discovering these problems, they went to the Parker Seal Company, which manufactures the rubber, to ask for advice. The Parker Seal Company told Thiokol that O-rings are not meant to be used that way, so they could give no advice.
Although it was known from nearly the beginning that the joint was not working as it was designed to, Thiokol kept struggling with the device. They made a number of makeshift improvements. One was to put shims in to keep the joint tight, but the joint still leaked. Mr. Weeks showed me pictures of leaks on previous flights—what the engineers called “blowby,” a blackening behind an O-ring where hot gas leaked through, and what they called “erosion,” where an O-ring had burned a little bit. There was a chart showing all the flights, and how serious the blowby and erosion were on each one. We went through the whole history up to the flight, 51-L.
I said, “Where does it say they were ever discussing the problem—how it’s going along, or whether there’s some progress?”
The only place was in the “flight readiness reviews”—between flights there was no discussion of the seals problem!
We looked at the summary of the report. Everything was behind little bullets, as usual. The top line says:
•
The lack of a good secondary seal in the field joint is most critical and ways to reduce joint rotation should be incorporated as soon as possible to reduce criticality.
And then, near the bottom, it says:
•
Analysis of existing data indicates that it is safe to continue flying existing design as long as all joints are leak checked* with a 200 psig stabilization…
I was struck by the contradiction: “If it’s ‘most critical,’ how could it be ‘safe to continue flying’? What’s the logic of this?”
Mr. Weeks says, “Yes, I see what you mean! Well, let’s see: it says here, ‘Analysis of existing data…’“
We went back through the report and found the analysis. It was some kind of computer model with various as-
Figure 8 Thiokol attempted to cure the joint-rotation problem with shims.
Figure 9 Two examples O-ring erosion. Such erosion would occur unpredictably along 2 or 3 inches of the 37-foot O-ring.
Figure 10 The correlation between temperature and O-ring incidents.
Figure 11 The self-contradictory recommendations of the seals report are underlined.
sumptions that were not necessarily right. You know the danger of computers, it’s called GIGO: garbage in, garbage out! The analysis concluded that a little unpredictable leakage here and there could be tolerated, even though it wasn’t part of the original design.
If all the seals had leaked, it would have been obvious even to NASA that the problem was serious. But only a few of the seals leaked on only some of the flights. So NASA had developed a peculiar kind of attitude: if one of the seals leaks a little and the flight is successful, the problem isn’t so serious. Try playing Russian roulette that way: you pull the trigger and the gun doesn’t go off, so it must be safe to pull the trigger again…
Mr. Weeks said there was a rumor that the history of the seals problem was being leaked to the newspapers. That bothered him a little bit, because it made NASA look like it was trying to keep things secret.
I told him I was entirely satisfied with the people Graham had brought in to talk to me, and that since I had already heard about the seals problem at JPL, it wasn’t any big deal.
The next day, Sunday, Bill Graham took me with his family to the National Air and Space Museum. We had an early breakfast together, and then we went across the street to the museum.
I was expecting to see big crowds there, but I had forgotten that Graham was such a big shot. We had the whole place to ourselves for a while.
We did see Sally Ride there. She was in a display case, in an astronaut’s suit, holding a helmet and everything. The wax model looked exactly like her.
At the museum there was a special theater with a movie about NASA and its achievements. The movie was wonderful. I had not fully appreciated the enormous number of people who were working on the shuttle, and all the effort that had gone into making it. And you know how a movie is: they can make it dramatic. It was so dramatic that I almost began to cry. I could see that the accident was a terrible blow. To think that so many people were working so hard to make it go—and then it busts—made me even more determined to help straighten out the problems of the shuttle as quickly as possible, to get all those people back on track. After seeing this movie I was very changed, from my semi anti-NASA attitude to a very strong pro-NASA attitude.
That afternoon, I got a telephone call from General Kutyna.
“Professor Feynman?” he says. “I have some urgent news for you. Uh, just a minute.”
I hear some military-type band music in the background.
The music stops, and General Kutyna says, “Excuse me, Professor; I’m at an Air Force Band concert, and they just played the national anthem.”
I could picture him in his uniform, standing at attention while the band is playing the “Star Spangled Banner,” saluting with one hand and holding the telephone with the other. “What’s the news, General?”
“Well, the first thing is, Rogers told me to tell you not to go over to NASA.”
I didn’t pay any attention to that, because I had already gone over to NASA the day before.
He continued, “The other thing is, we’re going to have a special meeting tomorrow afternoon to hear from a guy whose story came out in the New York Times today.”
I laughed inside: so we’re going to have a special meeting on Monday, anyway!
Then he says, “I was working on my carburetor this morning, and I was thinking: the shuttle took off when the temperature was 28 or 29 degrees. The coldest temperature previous to that was 53 degrees. You’re a professor; what, sir, is the effect of cold on the O-rings?”
“Oh!” I said. “It makes them stiff. Yes, of course!” That’s all he had to tell me. It was a clue for which I got a lot of credit later, but it was his observation. A professor of theoretical physics always has to be told what to look for. He just uses his knowledge to explain the observations of the experimenters!
On Monday morning General Kutyna and I went over to Graham’s office and asked him if he had any information on the effects of temperature on the O-rings. He didn’t have it on hand, but said he would get it to us as soon as possible.
Graham did, however, have some interesting photographs to show us. They showed a flame growing from the right-hand solid rocket booster a few seconds before the explosion. It was hard to tell exactly where the flame was coming out, but there was a model of the shuttle right there in the office. I put the model on the floor and walked around it until it looked exactly like the picture—in size, and in orientation.
I noticed that on each booster rocket there’s a little hole—called the leak test port—where you can put pressure in to test the seals. It’s between the two O-rings, so if it’s not closed right and if the first O-ring fails, the gas would go out through the hole, and it would be a catastrophe. It was just about where the flame was. Of course, it was still a question whether the flame was coming out of the leak test port or a larger flame was coming out farther around, and we were seeing only the tip of it.
That afternoon we had our emergency closed meeting to hear from the guy whose story was in the New York Times. His name was Mr. Cook. He was in the budget department of NASA when he was asked to look into a possible seals problem and to estimate the costs needed to rectify it.
By talking to the engineers, he found out that the seals had been a big problem for a long time. So he reported that it would cost so-and-so much to fix it—a lot of money. From the point of view of the press and some of the commissioners, Mr. Cook’s story sounded like a big exposé, as if NASA was hiding the seals problem from us.
I had to sit through this big, unnecessary excitement, wondering if every time there was an article in the newspaper, would we have to have a special meeting? We would never get anywhere that way!
But later, during that same meeting, some very interesting things happened. First, we saw some pictures which showed puffs of smoke coming out of a field joint just after ignition, before the shuttle even got off the pad. The smoke was coming out of the same place—possibly the leak test port—where the flame appeared later. There wasn’t much question, now. It was all fitting together.
Then something happened that was completely unexpected. An engineer from the Thiokol Company, a Mr. McDonald, wanted to tell us something. He had come to our meeting on his own, uninvited. Mr. McDonald reported that the Thiokol engineers had come to the conclusion that low temperatures had something to do with the seals problem, and they were very, very worried about it. On the night before the launch, during the flight readiness review, they told NASA the shuttle shouldn’t fly if the temperature was below 53 degrees—the previous lowest temperature—and on that morning it was 29.
Figure 12a Progression of a flame, possibly from the leak test port area(© NASA.)
Thiokol reversed itself, but McDonald refused to go along, saying, “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.”
That was so astonishing that Mr. Rogers had to ask, “Did I understand you correctly, that you said…,” and he repeated the story. And McDonald says, “Yes, sir.”
The whole commission was shocked, because this was the first time any of us had heard this story: not only was there a failure in the seals, but there may have been a failure in management, too.
Figure 12a An incorrectly sealed leak test port could provide an escape route for a flame which burns past the primary O-ring.
Mr. Rogers decided that we should look carefully into Mr. McDonald’s story, and get more details before we made it public. But to keep the public informed, we would have an open meeting the following day, Tuesday, in which Mr. Cook would testify.
I thought, “This is going to be like an act: we’re going to say the same things tomorrow as we did today, and we won’t learn anything new.”
As we were leaving, Bill Graham came over with a stack of papers for me.
“Geez! That’s fast!” I said. “I only asked you for the information this morning!” Graham was always very cooperative.
Figure 14 Puffs of black “smoke” (fine unburned particles) were seen escaping from the same place where the flame was observed.(© NASA.)
The paper on top says, “Professor Feynman of the Presidential Commission wants to know about the effects over time of temperature on the resiliency of the O-rings…”—it’s a memorandum addressed to a subordinate.
'What Do You Care What Other People Think?' Page 11
