CHAIRMAN ROGERS: This morning, the first item on the agenda is the Thiokol discussion, and I thought that in order to lay a foundation for the discussion, I might have Dr. Keel, who is our Executive Director, refer to comments that were made in our executive session, whenever it was - I've lost track of the date, I guess last week - by Mr. McDonald.
DR. KEEL: Yes.
CHAIRMAN ROGERS: I think this will give you a good starting point. We didn't ask the questions at that time. We thought we should have the principals here so we could ask them. So with your permission, we will do that, and you can summarize it, A1.
DR. KEEL: Yes. I will just, Mr. Chairman, at your direction just - we have selected excerpts from this, and let the record show that this is excerpts from the closed session of Monday, [600] February 10th, before the Commission. I will begin - this is after Mr. McDonald first rose and explained the fact that there was a discussion prior to launch expressing concern about the temperature at the seal and the performance of the O-rings at that temperature.
CHAIRMAN ROGERS: You really said "discussions," so there were more than one.
DR. KEEL: That is correct.
"Chairman Rogers: Am I hearing you say that you recommended against launch and you never changed your mind?
"Mr. McDonald: No, I did not say that. We did change our mind afterwards.
"Chairman Rogers: What brought you to that decision?
"Mr. McDonald: Well, the data that was reviewed. NASA concluded the temperature data we had was inconclusive, and indeed a lot of the data was inconclusive, because the next worst blow-by we had ever seen in a primary seal, in a case to case seal joint, was about the highest temperature we had launched at,
and that was true."
And I'm skipping ahead now in the testimony:
"Chairman Rogers: Did you change your mind?
"Mr. McDonald: Yes. The assessment of the data was that the data was not totally conclusive that the temperature could affect everything relative to the seal, but there was data that indicated that there were things going in the wrong direction, and this was far from our experience base, the conclusion being that Thiokol was directed to reassess all the data because the recommendation was not considered acceptable at the time, at the 53 degrees.
"NASA asked us for a reassessment and some more data to show the temperature in itself could cause this to be a more serious concern than we had said it would be. At that time, Thiokol in Utah said that they would like to go off-line and caucus for about five minutes and reassess what data they had there or any other additional data.
"That caucus lasted for I think a half hour before they were ready to go back on, and when they came back on they said they had reassessed all the data. They had come to the conclusion that the temperature influence, based on the data they had available to them, was inconclusive, and therefore they recommended a
launch.
"Chairman Rogers: When you say 'inconclusive,' what does that mean?
"Mr. McDonald: Well, the fact is -
"Chairman Rogers: You told them the day before not to do it, and then now you've got some more data and you say it is inconclusive and you changed your mind?
"Mr. McDonald: I was not back at Wasatch when that discussion was being held. I was at Kennedy, and I do not know what other data they were looking at, other than the charts that I had in front of me and others that they had in front of them at both KSC and Marshall. I do not know.
"I do know they came back on and said they had reassessed it and concluded that it was okay to launch, and at that point in time Thiokol was requested to put that in writing.
"Chairman Rogers: Well, I think in view of the very serious nature of this and the fact that it will be scrutinized for years, that we should have precisely what the data was before we present it."
CHAIRMAN ROGERS: Well, this meeting today is in the nature of an investigation and not really a hearing, so it is not necessary to swear anybody in.
When we have our regular formal meetings, we swear all our witnesses, but this really is for us to assess the facts and to see how they will be presented later on.
So, although it is going to be recorded, it is not going to be sworn testimony. That doesn't mean we don't want the truth, but it means that this is not formal, this is informal and in the nature of an investigation, and we want to have you feel free to have a discussion about the facts.
And what I have in mind particularly, and I think the rest of the Commission does, is to be sure that we know precisely what the facts were, who was involved, and as much as you can remember what was said, and if there were any documents relating to those discussions we would like to take a look at the documents.
And if there was any impounding of documents, critical documents, we would like to know that, when it happened, who has them if there are documents of that kind - all of the facts that relate to this very critical period in the launch.
MR. REINARTZ: Mr. Chairman, I'm Stan Reinartz from the Marshall Space Flight Center. I'm the project manager for our propulsion elements.
What we would propose, subject to your
concurrence, is that Mr. Mulloy, our SRB project manager from Marshall, give you the background that led up to the discussion that we had with Morton Thiokol, list the participants, including myself, in that telecon, and run through what we had, a very brief summary, and then we would provide Morton Thiokol the opportunity to then discuss on their side of the situation how they viewed it, if that is acceptable.
CHAIRMAN ROGERS: Sure.
MR. REINARTZ: All right, Sir, Larry Mulloy.
MR. MULLOY: Mr. Chairman and members of the Commission:
Good morning.
(Viewgraph.) [Ref. 2/14-1 1 of 4]
The decision to scrub the 51-L mission on January the 27th was made at approximately 1:00 p.m. by my recollection. Now, I had put approximate symbols in front of these times but I don't have the precise times.
But of course during the launch process, the project managers at Marshall and Dr. Lucas are on a communication net with personnel back at our Huntsville operations support center and also personnel that we have in the backup firing room here at Kennedy. We communicate with those people - they are technical specialists - relative to any problems or during the launch count any anomalies that come up, any potential launch commit criteria violations that come up during the course of the count.
And then subsequent to the launch or subsequent to a launch scrub, we stay on the net and communicate with those people relative to our ability to turn around within 24 or 48 hours on any concerns that might be related to that.
So what happened on the 27th, after the
decision to scrub, we polled our support personnel who were on the net, and that includes USBI, Booster Production Company, and Morton Thiokol personnel, regarding any constraints to a 24 hour recycle, in other words of going at 9:38 on the 28th of January.
This was about 30 minutes after the launch scrub. Some input did come back, and what they had looked at was all elements of the system. Our request then went into the system, which Morton Thiokol will discuss as to how they reacted to that request, any constraints for a 24-hour turn-around.
The input that I got back about 30 minutes after launch scrub on the net was that the concerns that had been worked were related to the recovery battery temperatures on the solid rocket boosters. We have a launch commit criteria red lined for those, and the fuel service module, hydrozene modules in the aft skirt that work the thrust vector control system on the solid rocket booster.
At that time, at about 1:30, there was no concern or no input back relative to any concerns for the solid rocket motor, considering the predicted temperatures.
The next event that I am aware of is about 5:15 that afternoon. In response to that request to
work, Al had - well, I will let him cover what he did, but he called our MSSE resident office, Cecil Houston, to inform him that Thiokol engineering did have some concerns regarding the function of the O-rings at the predicted temperatures.
CHAIRMAN ROGERS: Could I interrupt. Was there a discussion about the weather prior to the scrub?
MR. MULLOY: No, sir.
CHAIRMAN ROGERS: In other words, the first discussions you remember about the weather occurred at 5:15?
MR. MULLOY: Yes, sir. It was in the assessment of, do we have any problems with going at 9:38 on the 28th of January. And then it was pointed out that the temperatures were predicted to get below freezing, and there were a lot of aspects of that consideration for the work. Of course, there was water on the pad and potential bursting of pipes, ice on the external tank, et cetera.
MR. ALDRICH: Excuse me, Larry. To be consistent with the discussion we had yesterday, between the 1:00 p.m. meeting and the 5:15 p.m. meeting there was a mission management team that met, too, which Jess Moore and myself conducted in the launch control center, where we met with all of the projects and discussed whether we should pick up for launch the next day.
And at that meeting, we elected to pick up, and we were most concerned about the temps on the facility. But we asked the facility and the other elements to review the temperatures and call if we developed any complications downstream which would make that decision.
CHAIRMAN ROGERS: So at 2:00 o'clock -
MR. ALDRICH: So at 2:00 o'clock we called the level two and the level one of the projects, which fits between these two times.
CHAIRMAN ROGERS: Excuse me. I think I've changed my mind. If you don't mind, I think all the people from Thiokol that are going to talk today, let's swear them all in at one time, because if we don't and we have sworn the other witnesses it may in retrospect look a little odd.
So if you will all stand up and be sworn.
[603] (The following witnesses were sworn: Jerry Mason, Cal Wiggins, Joe Kilminster, Robert Lund, Don Ketner, Roger Boisjoly, Arnie Thompson, Al McDonald, and Boyd Brinton.)
MR. MULLOY: Sir, I will show on the chart the participants who are here.
CHAIRMAN ROGERS: When you are asked to comment, if you would identify yourself so the reporter will be able to record this.
MR. MULLOY: And following up on Arnie's comment -
CHAIRMAN ROGERS: Now, as much as possible, when you say we called the facility, if you remember who you called or who somebody called, let's use names, so if we trace what happened we have people instead of buildings.
MR. MULLOY: Yes, sir, will do.
Following up on Arnie's comment, what I just stated relative to our concerns, I stated in that mission management team meeting that we had worked those two concerns. And I cannot recall any discussion in that mission management team meeting relative to SRM temperatures.
MR. ALDRICH: I do not recall such discussion.
MR. MULLOY: Now, Al called the resident office. He will testify to that. Cecil Houston about ten minutes later then was able to get in contact with Jud Lovingood, who is here today - he's the Deputy Manager of the Shuttle Projects Office, of Mr. Reinartz - and relaying this concern that Al had pointed out to establish a telecon with Reinartz, myself, George Hardy at the Marshall Space Flight Center, and engineering and program people who were at the Thiokol Wasatch division
in Utah.
That telecon did occur. I did not participate in that telecon. They were unable to get in touch with me at that time, but that telecon occurred between Lovingood, Reinartz, and Thiokol personnel, and other personnel at Marshall Space Flight Center. And the result of that was to listen to the concerns as expressed here just in oral transmission. They had no data at that time.
The result of that was to set up a telecon where we could get all of the data transmitted to all of the parties and have more personnel participating in that.
CHAIRMAN ROGERS: And we will be able to be told by somebody what the comments were at that time?
MR. MULLOY: Yes, sir. At the 8:15?
CHAIRMAN ROGERS: Yes.
MR. MULLOY: Yes, sir, and I'm going into that.
DR. WHEELON: Is it your practice to record these telecons on magnetic tape?
MR. MULLOY: No, sir.
DR. WHEELON: There is no record of that -
MR. MULLOY: To my knowledge, this telecon was not recorded.
DR. WHEELON: Are others, and this not? Or is it your practice not to do so?
MR. MULLOY: It is our practice not to do so.?
DR. COVERT: Do you write up notes?
MR. MULLOY: Yes, sir.
DR. COVERT: Does everybody write up notes? Or if you don't know -
MR. MULLOY: I don't know, sir, that everybody writes up notes. I am aware that I wrote up notes in this case. [Ref. 2/14-1 2 of 4]
[604] DR. WHEELON: May we have copies of those notes?
MR. MULLOY: Yes, sir.
DR. COVERT: We might want to collect these different notes and see if they all attended the same meeting.
MR. MULLOY: I don't think, when you see the number of participants - there were some 30 participants - I cannot say that everyone had notes.
DR. COVERT: I understand. Press on, Larry.
CHAIRMAN ROGERS: Well, we should, if there are notes, we should have them, because there may be a difference of opinion on what was said, and it would be helpful to us to have them all. And if there are some that exist and we don't have them, that would be a mistake. We should get them all.
MR. MULLOY: Yes, sir. We will collect those.
That telecon was a little late starting. It was intended to be set up at 8:15. The 30 minute delay was due to the transmission of the engineering material from Wasatch in Utah to the Kennedy Space Center and to the Marshall Space Flight Center. That did arrive, though, in Huntsville and here at Marshall, and the telecon was begun at 8:45.
And Thiokol will then present to you today the data that they presented to us in that telecon. I will not do that. The bottom line of that, though, initially was that Thiokol engineering, Bob Lund, who is the Vice President and Director of Engineering, who is here today, recommended that 51-L not be launched if the O-ring temperatures predicted at launch time would be lower than any previous launch, and that was 53 degrees. Yes, sir.
MR. WALKER: May I ask a question. I wish you would distinguish between the predicted bulk temperatures and the O-ring temperatures. In fact, as I understand it, you really don't have any official O-ring temperature prediction in your models, and it seems that the assumption has been that the O-ring temperature is the same as the bulk temperature, which we know is not the case.
MR. MULLOY: You will see, sir, in the Thiokol
presentation today that that is not the case. This was a specific calculation of what the O-ring temperature was on the day of the January 1985 launch. It is not the bulk temperature of the propellant, nor is it the ambient temperature of the air.
It was Thiokol's calculation of what the lowest temperature an O-ring had seen in previous flights, and the engineering recommendation was that we should not move outside of that experience base.
I asked Joe Kilminster, who is the program manager for the booster program at Thiokol, what his recommendation was, because he is the gentleman that I get my recommendations from in the program office. He stated that, based on that engineering recommendation, that he could not recommend launch.
At that point I restated, as I have testified to, the rationale that was essentially documented in the 1982 critical items list, that stated that the rationale had been that we were flying with a simplex joint seal. And you will see in the Thiokol presentation that the context of their presentation is that the primary ring, with the reduced temperatures and reduced resiliency, may not function as a primary seal and we would be relying on secondary.
And without getting into their rationale and [605] getting ahead, the point, the bottom line, is that we were continuing - the assessment was, my assessment at that time was, that we would have an effective simplex seal, based upon the engineering data that Thiokol had presented, and that none of those engineering data seemed to change that basic rationale.
Stan Reinartz then asked George Hardy, the Deputy Director of Science and Engineering at Marshall, what his opinion was. George stated that he agreed that the engineering data did not seem to change this basic rationale, but also stated on the telecon that he certainly would not recommend launching if Thiokol did not.
At that time Joe Kilminster requested a five minute off-net caucus, and that caucus lasted approximately 30 minutes.
DR. COVERT: Larry, I'm confused, and could I go back about four meatballs here. What is a simplex joint seal?
MR. MULLOY: No redundancy.
DR. COVERT: That's a single O-ring?
MR. MULLOY: Yes, sir.
DR. COVERT: Is this common notation, or is it named after Charlie Simplex or something?
MR. MULLOY: Perhaps it is an unfortunate
phrase. I guess I consider "simplex" singular and "duplex" dual, and I was using it in that context.
MR. WALKER: So you're talking about the secondary seal.
MR. MULLOY: Yes.
MR. WALKER: In other words, if the first one didn't work the second one would?
MR. MULLOY: Yes, that is correct. And the engineering rationale you will see for accepting the situation was counting on the secondary seal, and that will be developed. Yes, sir.
MR. SUTTER: Is this in line with - I thought the philosophy was at least dual redundancy on there being critical.
MR. MULLOY: Yes, sir, that - -
MR. SUTTER: Where is the rationale that says a single ring is okay and anything single is okay?
MR. MULLOY: The design goal is certainly to have redundancy in critical systems, and in late 1982 it was recognized that this joint design did not provide that redundancy and the criticality on the critical items list, the criticality for this joint was changed from criticality 1-R, which is redundant, to a criticality 1.
And there are numerous systems on the shuttle system that are criticality one. They are not redundant. They are indeed criticality one.
This was changed into that criticality, recognized, and the rationale for the acceptance of that
condition based on analysis and comparison to a similar joint design which flies with a single O-ring, namely the Titan, which at that time had had some 70 successful launches.
So since December of 1982, we have been operating the system with the recognition that under worst case conditions, worst case tolerance buildups, and certain other conditions, that you could indeed not have redundancy in the joint. And what I am stating here is that my [606] conclusion, based upon engineering data presented, was that that situation here was not any different.
MR. HOTZ: But doesn't the Titan use a different joint?
MR. MULLOY: It is a similar joint, sir. It's a tang and clevis with a single O-ring that they don't leak check. The primary - -
MR. HOTZ: But isn't it reversed, that the tang and the clevis go the opposite way?
MR. MULLOY: Yes, sir, I believe it is, yes, sir. It is the tang is down - or, excuse me, the tang is up in flight direction and the clevis is down.
MR. HOTZ: So you're not basing it on exactly the same type of a joint.
MR. MULLOY: No, sir; similar.
MR. WALKER: There's also a quasi-seal of the insulation, and that may be a very critical difference.
MR. MULLOY: There is an interference fit in the insulation, which is not related to the physical joint design. But it is related to the insulation layup.
CHAIRMAN ROGERS: Larry, in criticality one problems in the past, do you remember any other examples where there was such a late decision on a criticality one and whether there were disagreements among the participants?
MR. MULLOY: No, sir, I do not. We are researching those records now for the total shuttle system, and I'm certainly doing that for the SRM. I am not aware since December of 1982. I took over the project in November of 1982.
I do not remember any changes from a criticality 1-R to a criticality 1 in the solid rocket booster. What we're looking at now is what are all of the criticality ones and getting a listing of those for the solid rocket booster. The total system is doing that, I believe, are you not, Arnie?
MR. ALDRICH: Yes, sir, Mr. Chairman. We are preparing a summary for you on the critical items and the nature of some of them and the total set, and that
will be available and we can discuss it later today if you wish.
MR. SUTTER: Well, I think that that discussion shouldn't be discussed today, because I think it is a discussion that should take all day some time. This is not exactly the subject we're discussing now, but I would like to some time have a discussion on the design criteria, and is redundancy an objective or a requirement and how do you go from 1 to 1-R, how do you decide these things, and how do you say that at one place you've got redundancy and then you don't? I think that is a whole day's discussion.
CHAIRMAN ROGERS: I agree. Dr. Keel, would you make a note, and we will schedule one as soon as we can. I think you're absolutely right.
DR. RIDE: Larry, in your discussion relative to the CIL or your decision relative to the CIL, does that mean that you were assuming what the effects of temperature were on a secondary seal?
MR. MULLOY: No, I don't believe that was considered at that time. I believe the total rationale for retention on the CIL from 1-R to 1 is attached there, and all of the considerations are there, and I do not believe temperature is included in that consideration.
DR. RIDE: I guess I meant, when you were making this decision at the meeting were you assuming that the effects of the cold weather - -
DR. MULLOY: Yes.
[607] DR. RIDE: that you were assessing were going to affect the secondary seal and not the primary seal?
MR. MULLOY: Yes, and that will come out, I think, in Thiokol's - -
MR. REINARTZ: Wait a minute. Excuse me. Repeat your question, Sally, if you would, please, for Larry.
DR. RIDE: Okay. I guess I was just looking at the CIL, and it says that - it looks to me like it says that the primary O-ring is I heard now a single point failure, because you can't count on the secondary O-ring. Is that a fair assessment of the CIL?
MR. MULLOY: That is correct.
DR. RIDE: And then I guess my question was, in your discussion the day before launch and the evaluation of the effects of the cold temperatures on the O-ring, if you were going to base your decision on the CIL it seems that you would have to assume that the cold temperature affected the secondary O-ring, but not the primary O-ring, since the primary O-ring is the criticality one.
MR. MULLOY: Let me restate. I did not base
my decision on the CIL. The CIL states that we have a simplex - the rationale for a simplex seal. We do not have a redundant seal. My assessment at the time that I made this comment right here, that it did not seem to change that basic rationale, was related to the engineering data that Thiokol will show, that shows that even with the effects of the cold temperature that we expected to have a simplex seal under the worst case conditions, the basic rationale being simplex and not redundancy, a single functioning O-ring.
DR. RIDE: A simplex seal where the one O-ring was the primary O-ring?
MR. MULLOY: No. The rationale, you will see, says we were counting on the secondary O-ring to be the sealing O-ring under worst case conditions and the worst case analysis that is presented here.
DR. RIDE: But doesn't the CIL say you can't count on the secondary O-ring?
MR. MULLOY: Yes, and you have to see the engineering development for the rationale that states that if the cold effect on the primary O-ring - and I'm getting into Thiokol's engineering data, but basically, and I guess I should say what my understanding was, the effect of the cold on the O-ring is it reduces the diameter of the O-ring. It also reduces the resiliency of the
O-ring.
The primary seal under the leak check pressure is seated in the wrong direction. Therefore, there is a time required before the primary O-ring can become effective. The consideration was that if the primary O-ring under the worst case conditions, which was not predicted, incidentally, as you see the engineering data - it was predicted that the primary O-ring would have sufficient compression and sufficient resiliency to extrude into the gap and serve as a functioning seal.
But we went a step further and said, suppose under worst case conditions the cold effect caused the primary O-ring to be totally ineffective. If the primary O-ring is ineffective, the secondary O-ring, which is in a position to seal, will be pressure actuated in the time before the joint rotates.
DR. FEYNMAN: Excuse me. I'm getting very confused, and I would prefer a more logical description and order. I think your original plan was to explain the order of things quickly and then we could go back and listen to all of the arguments in the order in which they were presented. This way things are being pulled out of order and it is hard to follow. Is that okay?
[608] CHAIRMAN ROGERS: Yes, but if anybody else has questions they should ask them.
DR. FEYNMAN: Yes, but we have the opportunity for some orderliness - -
CHAIRMAN ROGERS: We haven't followed the procedure of not interrupting so far, so let's go ahead.
[Laughter.]
CHAIRMAN ROGERS: If anybody has a question he wants to ask, he should ask it.
MR. REINARTZ: Mr. Chairman, let me do one thing. When we considered the question of the joint and the simplex, that that was the conclusion of a worst case possibility. What we were looking at from the data presented was that there was an increased probability that you may not have a primary seal.
But we did not accept the fact that the basic starting condition was assume that you had no primary seal to start with. There was an increased probability of not sealing with the primary, and if that did occur then we had a simplex with a secondary seal, and that was the final point that Larry was making there.
So there was some increased probability, but it was not a foregone conclusion, that the primary seal would not operate, and I think that will come out today.
CHAIRMAN ROGERS: It's a little difficult to compare what you just said with what the document says, though.
Let's go ahead.
MR. MULLOY: Yes, sir. This is the list of the total participants, the total number of participants in the telecon. At Morton Thiokol in Utah were these. We brought the principal technical disciplines. We did not bring all of these people here today, Mr. Chairman, and of course they are available.
But in Florida there was myself, Stan Reinartz, Cecil Houston, and Jack Buchanan. Did he get in? Okay, Jack Buchanan is here also. He's the manager of the launch support services here, and Al McDonald; at Marshall, the following individuals with the disciplines that they represent indicated. There are some 30 people then involved in that 8:45 telecon.
The next chart, please.
(Viewgraph.) [Ref. 2/14-1 3 of 4]
Moving on, then. That telecon lasted, then, after Thiokol requested their caucus and it took about 30 minutes, at that point Mr. Kilminster came back on the net on the telecon and read the rationale for recommending launch. And that was rapid-faxed to me here at Kennedy about 36 minutes later, and I will show you what that says.
I guess it is an assumption on my part, Joe, that you were reading from that. It sounded the same.
MR. KILMINSTER: I was reading from some other
notes.
GENERAL KUTYNA: During that 30 minute off-the-net conference, were there telephone calls between the centers and Thiokol to discuss this thing?
MR. MULLOY: No, sir. We were off the net and we were not carrying on any discussion on this. Those of us who were here at Kennedy were discussing among ourselves and relooking at some of the engineering data, and I'm sure that perhaps George Hardy can testify to the fact of what they were doing back there. I don't know.
[609] We did not carry on any continuing dialogue during that 30 minute period.
What Joe Kilminster then essentially stated at the end of that telecon and what he subsequently faxed down 35 minutes later, which I then had, was:
"Calculations show that the SRM-25 O-rings will be 25 degrees-20 degrees colder than SRM15" - which was that one that was at 53 degrees.
MR. WALKER: 33 degrees, in other words, was the predicted temperature?
MR. MULLOY: I believe, yes, sir.
DR. COVERT: What's the difference between 25 and 15?
MR. MULLOY: That should be 25 degrees? That's an error? Well, this is the chart that I worked from.
MR. KILMINSTER: That is the way it read, but it should have said 25.
DR. COVERT: What's the difference between 25 and 15? Is that just a serial, or sequence?
MR. MULLOY: Yes, these are the SRM numbers. It's ten SRM builds later. It happens to be the difference between January and January.
DR. COVERT: Okay, fine. Thank you. Press on.
DR. MULLOY: "Temperature data is not conclusive on predicting primary O-ring blow-by. The engineering assessment is that colder O-rings will have increased effective durometer. They will be harder. The harder O-rings will take longer to seat, that more gas may pass the primary O-ring before the primary seal seats as it translates from its upstream position to its downstream position, that the demonstrated sealing threshold, however, is three times greater in terms of erosion than we've experienced on SAM-15."
And I will let Thiokol engineering explain more about the analysis that goes into that.
"If the primary seal does not seat, the secondary seal will seat. Pressure will get to the secondary seal before the metal parts rotate. The O-ring pressure leak check places the secondary seal in outboard position, which minimizes sealing time, and MTI recommends 51-L launch proceed on 28 January 1986 and
that SAM-25" - which are the motors that were on 51-L - "will not be significantly different from SAM-15."
And with that, sir, I will turn this over to Thiokol, who will discuss their deliberations during this period of time. Mr. Jerry Mason is the Vice President and General Manager of the Wasatch Division of Thiokol.
I might mention, Mr. Chairman, we also have from Thiokol here today Mr. Ed Garrison, who is the President of the Aerospace Group in Chicago, Mr. Mason's supervisor.
CHAIRMAN ROGERS: Mr. Mason, might I suggest in your discussion with us today that, please disclose anything that you know about that may turn up. If you have documents that we don't know about that would be embarrassing to you, tell us about them now.
We don't want to have to pry information out of you. You know what's there. Tell us the whole story, if you will.
[Please note that some of the titles to the references listed below
do not appear in the original text. Titles are included to identify
and clarify the linked references - Chris Gamble, html editor]
[610] [Ref.2/14-1 1 of 4] [Chronological events recalled by Larry Mulloy-
January 27-28, 1986] [Ref.2/14-1 2 of 4] [Chronological events recalled by Larry Mulloy-
January 27-28, 1986]
[611] [Ref.2/14-1 3 of 4] [Chronological events recalled by Larry Mulloy- January 27-28, 1986] [Ref.2/14-1 4 of 4] [Chronological events recalled by Larry Mulloy- January 27-28, 1986]
[612] [Ref.2/14-2 1 of 7] [Larry Mulloy's handwritten note regarding 51-L assessment of cold o-rings].
[613] [Ref.2/14-2 2 of 7] [a written testimony/deposition, presumably from Larry Mulloy] [Ref.2/14-2 3 of 7] [a written testimony/deposition, presumably from Larry Mulloy]- continued.
[614] [Ref.2/14-2 4 of 7] [a written testimony/deposition, presumably from Larry Mulloy] - continued. [Ref.2/14-2 5 of 7] [a written testimony/deposition, presumably from Larry Mulloy] - continued.
[615] [Ref.2/14-2 6 of 7] [a written testimony/deposition, presumably from
Larry Mulloy] - continued. [Ref.2/14-2 7 of 7] [a written testimony/deposition,
presumably from Larry Mulloy] - continued.
MR. MASON: We will do that. First off, what we had in mind was that I would give you the kind of overview that Larry has, except to give you more specifics of what occurred at the Wasatch Division in Utah. And then after that, our Vice President of Engineering will go through the specific technical charts that were reviewed that evening, and then Mr. Kilminster is prepared to provide the additional rationale that led us to the final conclusions.
But one thing I would like to explain. Larry's introduction was a little bit in error. I am actually the senior vice president of the Wasatch operations, and I need to explain that. We have three divisions there, the space division, strategic, and the tactical division. And of course, the space division is dedicated to the shuttle program, and we have a general manager for each division.
Mr. Wiggins, who is with us today, is the general manager of the space division, and reporting to him is Mr. Kilminster, who is the program manager over manufacturing and quality operations. And then Mr. McDonald reports to Mr. Kilminster as the program
manager on space and the solid rocket motor itself.
On the other side, also reporting to me, is the vice president of engineering, Mr. Lund. So I hope that clarifies the relationship a little bit.
The chronology is essentially as Larry had identified there.
MR. SUTTER: Can I ask one question now? Between Kilminster and Lund, who makes the "technical" or engineering decisions?
MR. MASON: Lund makes the technical recommendation to Mr. Kilminster.
MR. SUTTER: He can act on it as he sees fit?
MR. MASON: The final decision comes from the program manager to the general manager, and if necessary to me. The engineering, since it is a separate function, does provide a bit of an oversight on the program manager's functions. In other words, he doesn't report to him on a hard line; he reports to him in an advisory role, if you will.
The sequence was essentially as Larry had identified it. We were notified both from Marshall and Kennedy that the temperatures were going to be quite low the next morning for the launch if it were to occur at 9:38.
As a result of that, our engineering people
looked at all of the aspects of the motor, particularly at the propellant mean bulk temperature, which is normally something we are concerned about, and that was not of any concern in this case, and identified that the only area of concern was the seal.
And they wanted to get together more data that existed in various places to take a further look at the seal, which they did. They started gathering that information, and the preliminary telecon occurred and arranged for the subsequent formal telecon.
In the formal telecon then, we reviewed the technical data, which Mr. Lund will cover with you, and the sequence was that we were gathering this data and attempting to reach our firm recommendation at the same time. So we were in a position that we got the charts together and [617] were trying to make the final chart on our recommendations, and at that point had not had any, I would say, complete discussions.
There were differing opinions within the group, and the only opinion that survived everyone was to launch at 53 degree O-ring temperature or higher.
CHAIRMAN ROGERS: What was the nature of the discussions? I mean, you say there were a lot of different views. What were they, some of them?
MR. MASON: Well, they centered around the
effect of the cold on the ability of the O-rings to respond, and it was primarily the primary O-ring because, as has been mentioned, when you pressurize, when you run the leak test, you put the primary O-ring on the wrong side of the roof, whereas the secondary goes onto the side that you want it on.
So the question was, will the cold change the response time enough to keep that primary from sealing? And we did not have hard data that would give us that answer.
DR. RIDE: What was the lowest temperature that your data went down to?
MR. MASON: We had - the only meaningful, I guess the only conclusive data was the flight data, which was the 53 degrees. In other words, we had information on the hardness of the O-ring and various other things like that, which Bob will cover with you in the chart. But we did not have anything that said specifically how long does it take for the O-ring to move across.
Now, the reason, of course, that is the primary issue is that it needs to seal before you get a high enough pressure in the case to open that gap, or the so-called joint rotation. So it needed to seal in the 160 milliseconds.
GENERAL KUTYNA: If I may, sir, this has been a problem since about 1980 off and on, and it was looked at as a criticality one failure mode for a few years and it has been continuing. It seems strange that you had to gather the data on the environments of this thing at this very moment; that if I had a problem like this in an airplane, I would have been working that environment data from the year 1980 or 1981, and not 1985 or 1986.
Why was the environment not considered and why hadn't you gone through an exhaustive review of the environmental effects on the failures prior to this time?
MR. MASON: Well, some of the data that we were gathering is data that was currently being created at that time as a result of the effort we were making to improve the reliability. For example, there is a blow-by test in there. That one had been run with argon and we were going to run it with freon, and we were getting the latest, very latest information we had, because we were giving consideration to that.
DR. RIDE: Let me ask Don's question in maybe a slightly different way. You had at some point to sign up for the SRB's working at 31 degrees ambient temperature, because that is the launch commit criteria on the entire system.
What data did you use to certify the boosters to work at 31 degrees before you allowed that to be taken into account?
MR. MASON: Well, let's see. I think the way it is set up is, we signed up to work - to have it work with the propellant mean bulk temperature from the 40 to 90 degrees. I believe that is the performance requirement.
MR. WALKER: Why was there no separate requirement on the O-rings? That is the issue.
[618] DR. RIDE: I mean, surely that requirement was imposed on you by NASA. NASA must have said: Your solid rocket booster has to work at 31 degrees; do the analysis of all the parts to determine that it does.
MR. MASON: Well, I have to say what my current understanding of the requirements is, which is that it has to perform, the propellant, at mean bulk temperature from 40 to 99, and that says it has to withstand temperatures of 31 to 99. And the attention has all been focused on getting the performance when the propellant mean bulk temperature was 40 to 99.
DR. WHEELON: A question. It is normal to establish specifications on a unit and then to complete a qualification program for a unit, prior to flight. Did you have a
temperature specification on the O-ring or the joint assembly, either one? And can you describe what your qualification program of testing to verify that that specification was being met?
And if you personally don't know the answer to that question, I would like that question then to revolve to whoever in your organization picks it up. But I think we need a clean, crisp, clear answer to the question.
MR. MASON: Okay. I think it would be better that I not - I think I know the answer, but I have not enough confidence to try it.
DR. WHEELON: Who are you going to lateral that football to?
CHAIRMAN ROGERS: Who can answer that?
MR. MASON: Joe, can you answer that?
MR. KILMINSTER: I believe, as Jerry mentioned, the major focus of emphasis was on the 40 to 90 degree requirement for the mean bulk temperature. When it comes to the seal, we have a procurement spec for that material, a Mil R specification, that calls out that material being capable over the temperature range of minus 30 to 500 Fahrenheit. And it was on that basis that we qualified the use of that material as far as the seal is concerned.
GENERAL KUTYNA: Capable of what?
MR. WALKER: Isn't that specification for a captured O-ring inside of a groove with a flat mating surface, not this kind of configuration?
MR. KILMINSTER: I don't believe the specification specifies what type of construction or what kind of design. It is a material capability.
MR. WALKER: Could we have that information? That information must be in Parker's specifications.
MR. KILMINSTER: It's in the Mil R specification.
DR. WHEELON: What did you do to assure your material was meeting that specification? What was the qualification program?
MR. KILMINSTER: Early in the program, it was determined that we would not have a program, a development program, and a test program that would qualify over that full temperature range.
DR. WHEELON: What range were you going to qualify over?
MR. KILMINSTER: Again, the emphasis was based on the solid propellant bulk temperature, and that qualification was done by analysis. We did not conduct a test.
DR. WHEELON: Let's set aside the question of
bulk temperature on the propellant. The question is what specification did you have on the joint and the O-ring, and how did you test to verify that in fact you were meeting that specification, or did you not have a specification and not test?
MR. KILMINSTER: The specification we have is a Mil R specification.
DR. WHEELON: Which is a generic spec to cover a whole range of military equipment, right?
MR. KILMINSTER: In this case, it is this specific material.
DR. WHEELON: Okay. So it is minus 30 to plus 500, is that correct?
MR. KILMINSTER: That's correct.
DR. WHEELON: And how did you shade that requirement and how did you test to make sure that you had met that shaded requirement?
MR. KILMINSTER: We did not test specifically to identify that requirement or test against that requirement.
DR. WHEELON: Don't you find that a little surprising?
DR. KILMINSTER: There are many areas, as I mentioned, based upon the original intention that we would not conduct full-scale firings, full-scale tests, using
a full range of temperatures.
DR. WHEELON: Did you use any subset of that full range of temperatures in your tests?
MR. KILMINSTER: Yes, we did.
DR. WHEELON: What range was that?
MR. KILMINSTER: That will be discussed when Bob discusses his charts. I believe we had a static firing as low as 47 Fahrenheit.
DR. WHEELON: Ambient?
MR. KILMINSTER: No, that was the predicted O-ring temperature, using ambient calculations lower than that.
CHAIRMAN ROGERS: Could I make a suggestion about procedure? Let's not worry about - if there is a question that is asked and somebody can answer it, have them answer it. I mean, you can still continue your presentation that you have organized, but you are all here now and it's reasonably informal.
So when Bud asks a question like that and somebody can answer it, just have them stand up and answer it.
DR. WHEELON: So you think went down to 47 degrees, in terms of a spec for ground testing of the seal?
MR. KILMINSTER: Yes.
DR. WHEELON: But no lower than that?
MR. MASON: That was the seal temperature.
DR. WHEELON: I understand, not the ambient. But you qualified the seal at 47 degrees Fahrenheit?
MR. KILMINSTER: We verified it in a static test at 47 degrees.
DR. WHEELON: A static ground test?
MR. KILMINSTER: Yes, sir.
DR. RIDE: And you did no tests on the joint below 47 degrees?
MR. KILMINSTER: That is correct.
[620] GENERAL KUTYNA: Bob Crippen, I thought somebody said yesterday, maybe it was you, that the shuttle was cleared to fly with a shuttle temperature of 31 degrees or thereabouts.
VICE CHAIRMAN ARMSTRONG: Arnie said that.
GENERAL KUTYNA: Now, how do you correlate the fact that the shuttle is cleared to fly at 31 degrees and yet you have only tested down to 47 degrees, and by analysis only to 40 degrees? How do you explain that?
MR. KILMINSTER: The only explanation I have is that we felt that we had a margin because of the material being capable down to minus 30 as identified in the specifications.
DR. WHEELON: Capable of what?
MR. KILMINSTER: Capable of functioning.
MR. WALKER: I really think we need to understand exactly what that specification implies in terms of the use of O-rings in various kinds of configurations. We would really like to have that information as soon as possible on exactly what that specification implies.
DR. WHEELON: Or better yet, can you give us the specification from which you were working?
MR. BOISJOLY: The specification in question is Mil R 83248A.
DR. WHEELON: A further question. Did you do any further testing of the O-rings or the O-ring material on your own, independent of the motor?
MR. KILMINSTER: Not that I can recall.
DR. WHEELON: So you were just working to the specification of the material as provided in the Mil Spec?
MR. KILMINSTER: I'm talking in the original qualification program. We subsequently have done testing.
DR. WHEELON: How subsequently? The last couple of days?
MR. KILMINSTER: No. Over the last probably year, year and a half.
DR. WHEELON: And what did those tests over the last year tell you?
MR. KILMINSTER: There is one test that is included in Bob's presentation that goes down to 30 degrees and identifies no blow-by at that point in a sub-scale joint configuration.
DR. WHEELON: Is this a scale test?
MR. KILMINSTER: It is a scale model.
MR. MASON: It's a full scale O-ring and full scale joint. It's just short, a small diameter.
DR. WHEELON: How many such tests did you run?
MR. KILMINSTER: I can't recall.
DR. WHEELON: Can you get us that data? I don't mean to tax your memory.
MR. KILMINSTER: Roger, do you recall the number of tests that were run at the cold temperatures?
MR. BOISJOLY: No, I don't.
MR. THOMPSON: Arnie Thompson here. There were two tests run at 30 degrees and two tests run at room temperature with the blow-by device, using argon as a tracer gas. The results were that we could see no blow-by with the apparatus.
We recognized that we needed to have a better sensitivity, so we went to freon 14, which shows better
on the mass spectrometer. And those tests are proceeding and we have some data from that.
[621] DR. WHEELON: In view of that apparently satisfactory validation, why were you concerned on the day before launch? It seems to me these tests should have set you at ease, and yet you were uneasy. Why?
MR. BOISJOLY: It should be emphasized that that test simulates the gap on a subscale basis, but does not simulate the dynamics of the joint and O-ring. It is not a joint. It is a gap, a physical fixed gap. And it was intended to measure if blow-by could occur past an O-ring at a very low temperature - not the temperature, but at a very low pressure, at the beginning of the ignition transient.
That test was specifically set up after the SRM-15 blow-by, which occurred a year ago last January, because at that presentation we made a point of saying that there is a small, miniscule portion of air that will go past any seal, any O-ring seal, in its attempt to do its initial sealing at low pressures. And then rapidly, as the pressure increases, the seal is functioning better and better at higher pressures, and that was the purpose of that test. That is the climate in which that test was developed.
GENERAL KUTYNA: But yesterday we were shown that that joint does in fact compress. The gap moves, opens, closes.
MR. BOISJOLY: That's correct.
GENERAL KUTYNA: at the light of the SRB's. and possibly even before that, when the SSME's are lit and you have the twang effect.
And what you're saying is you did not take that into consideration during this test?
MR. BOISJOLY: On that particular test, that is correct.
GENERAL KUTYNA: It's a totally static joint with no compression?
MR. BOISJOLY: Yes.
DR. COVERT: I would like to ask a procedural question, please. Is that spec you gave us and that number, is that the one that was in force at the time the decision was made, or is that the one that is in force now? Or what's the deal?
MR. BOISJOLY: That is the mil spec that controls the flight-on? O-ring material.
DR. COVERT: What is the date of the spec?
MR. BOISJOLY: I don't think I have that information.
DR. COVERT: I think that would be helpful to know.
DR. WHEELON: Did these tests figure in your
decision to proceed, or were they not a part of your thinking when you re-made the decision to proceed?
MR. BOISJOLY: To proceed with flying?
DR. WHEELON: With flying?
And I'm trying to say, there apparently was a caucus out in Utah at which you were asked to consider your decision not to go and in effect you wound up going. Were these test results a part of that reconsideration?
MR. BOISJOLY: Yes, sir.
DR. WHEELON: What role did they play?
MR. MASON: Well, sir, if I could, maybe I could pull that, the next steps together, and explain what things were involved in the reconsideration. We got to the point where we had faxed these copies of these charts, and at that time our best conclusion was to stay within our experience base.
[622] We had not been able to determine whether it was rational to extrapolate, and so at that point we recommended the 53 degree or higher temperature, because that was our experience base.
DR. KEEL: Mr. Chairman, could we have it clarified for the record exactly what "that point" means in the time sequence?
MR. MASON: That point was at the time when we faxed the charts to start the formal group telecon.
DR. KEEL: And in addition, can we have those charts? Do you have them with you today? MR. MASON: Yes, sir. Mr. Lund is going to review those charts for you.
CHAIRMAN ROGERS: When you say "formal telecon" and "informal," how do you distinguish?
MR. MASON: Well, it really was whether you have the whole group on or there is an individual telecon. Earlier there had been a telecon that Larry mentioned with Mr. Lovingood, in which there was just talk between Marshall and Morton Thiokol. And then there had been the call from Kennedy to us saying it was going to be cold, and so I call that the informal telecon. It was just communicating generally.
And then it culminated in what I call the formal telecon, at which we had everyone on the net - Marshall, Kennedy, and ourself.
MR. LOVINGOOD: To clarify that, Kennedy was on, Jerry, during the 4:45 telecon, but we didn't have all the people in that 4:45 meeting that we had in the later meeting.
CHAIRMAN ROGERS: Let's discard the idea of "formal" and just call it telephone conversation, and who was talking and what was said.
MR. MASON: I think now I probably ought to
address it as the final telecon in which we had everyone that Larry had listed to on the line. And it was interrupted by our caucus, and some people would call that two meetings or two telecons in one, but in my mind it was a continuing one, just interrupted by the caucus.
Now, we reviewed the charts in there and drew our conclusion that we ought to fly at 53 degrees or warmer. And at that point we received a number of comments on the net. The ones that I looked upon as being of consequence were that we were reminded that there really wasn't good correlation or valid correlation between temperature and blow-by, and that was pointed out because we had two cases of blow-by and one had been at 75 degrees and one had been at 50 degrees, and we had a lot of cases at varying temperatures where there had been no blow-by, and so we had very limited basis for saying that blow-by correlated with temperature.
[Ref. 2/14-3]
MR. WALKER: Now you're talking about seal temperatures?
MR. MASON: We're talking about seal temperatures, right.
DR. RIDE: Where was the lowest temperature that you had no blow-by?
MR. MASON: Well, in the static test motors it was 47, I believe was the lowest, but
it's on Bob's charts. We have cases where we had blow-by listed and we have - what we did was pick the coldest static test and the flight test. And of course, the static test didn't have any blowby.
The static tests are slightly different and Bob will explain those. We tended to not put a of weight on those.
[623] GENERAL KUTYNA: But the worst blow-by you had was at the lowest temperature, which was 51 C. did that not give you a strong correlation?
MR. MASON: That was the exact discussion that we had, was whether the fact that that one was somewhat worse than the one at 75 degrees, was that a correlation with temperature or was the fact that they blew by at both the 75 and 50 indicating that it was relatively independent of temperature.
DR. COVERT: Did you have any data on the stacking or the clearances on the 75 degree one as compared to the other?
MR. MASON: Yes, we did.
DR. COVERT: And were those clearances different significantly?
MR. MASON: They were relatively nominal, and we have that on the chart also.
MR. WAITE: Is last year's experience what led
to this increased level of testing? In other words, the decision to go back and test?
MR. MASON: Well, you might say cranked up the gain in August when we had summarized everything and concluded that we needed to get a more aggressive effort on improving things.
MR. WAITE: So it was your experience on this one occasion that led to this need for more testing?
MR. MASON: I would say it was collective experience, sitting down and looking at it totally.
In any event, with the one comment about the lack of hard evidence that there was really a correlation, although instinctively we felt that the cold would make it somewhat worse. The other point that was made was that the concern about the primary having to move and it being colder, it might move slower. We said, don't overlook the fact that the secondary is in position and therefore it doesn't have to move, so the time element is not as of great a consequence on the secondary as it is on the primary.
So with those two comments and others, but those were the ones that we considered of consequence, we decided that we ought to have a caucus and assess whether it would be reasonable to extrapolate below our experience.
DR. COVERT: How much more rigid is the grease at 30 degrees than it is at 50 degrees? MR. MASON: I can't give you a number. The grease is somewhat stiffer, but it is just a film on the outside of the O-ring.
DR. COVERT: At least in principle it is.
MR. MASON: Yes. And we made that observation, that both the grease and the O-ring would be stiffer. Now, we had durometer readings on the O-rings and we had the general knowledge that the grease gets somewhat stiffer.
DR. FEYNMAN: Could I interrupt, because there's some physical thing that I don't understand quite clearly, and that is why the need to move the O-ring makes it so much less likely to seal than if it is in the right place.
MR. MASON: Because if it doesn't move fast enough and the joint opens up before it is seated, then it won't seat.
DR. FEYNMAN: Why?
MR. MASON: Because it doesn't have enough compression of the O-ring to hold the gas that forces down into the crack.
MR. WALKER: In other words, it has to be deformed initially before the gap opens up?
MR. MASON: It has to initially so that when the gas hits it it will flow down into the crack and if you lose too much of that compression then it will flow by instead of pushing it in.
So that was our exact concern, that that very small distance that it had to move, if it didn't move that within the 160 milliseconds before the gap opened.
DR. FEYNMAN: That theory of how it works would account for the very much larger fraction of nozzle joints which failed compared to field joints, in which the nozzle joints have to move about seven times as far as a field joint, and that is consistent with your view.
MR. MASON: Exactly.
So we then had our caucus, in which we revisited all of the things we had talked about before. And we recognized two primary things that are covered in Mr. Kilminster's chart, and that was that the worst experience we had had in erosion was 38 thousandths, and we know from tests that the O-ring would seal with over 120 thousandths of erosion.
So our first thought was that if we had more erosion on the primary because it took longer, it would still seal, even if it were eroded three times as much. So we said we still had a reasonable expectation that
the primary would seal, but we didn't have absolute data that said how long it would take to move.
So we then said, what happens if it doesn't? And we took the second point, which was that the secondary was in position and did not have to move. So we felt that the primary probably would seal, but if it didn't the secondary would because it was already in position.
DR. FEYNMAN: During this discussion, nobody noticed the possibility, I presume, that even though the secondary is in position, if the thing opens up and the resiliency of the material is zero because it is too cold, it won't close, it won't fill into the seal? Was that considered or not?
MR. MASON: That was considered, and the key issue became that what had to happen is either the primary or the secondary had to seal in the 160 milliseconds, before resiliency came into play, because in both cases if it seated before it opened up then it would extrude down into the joint.
But as you pointed out, if it started to open, then the resiliency question came into play.
GENERAL KUTYNA: But that's 160 milliseconds after lightoff of the solids. How about the resiliency effects in the seven seconds that the SSME's are
compressing and bending this particular joint? Is there not an opening and closing of the gap during that time, that could have compressed the seals?
MR. MASON: Well, I can't answer that with certainty. I believe that those effects are relatively nominal on that joint, because the stiffener rings take out the bulk of that load on that joint.
DR. RIDE: Has that analysis been done?
MR. MASON: They have been looking at that effect again since the incident, and I can't tell you what the outcome is right now.
DR. RIDE: Where is that done? Is that done at Thiokol?
MR. MASON: Both. We're doing it both at Thiokol and at Marshall. We cross-check each other.
But I know there were some rough numbers done that said that that is quite stiff and it is not likely to be, but it needed to be done in great detail.
[625] DR. RIDE: But it hadn't been done before 51-L?
MR. MASON: Yes, the analysis, the load analysis in the joint had been done. I can't tell you whether we had looked at whether it had any effect on the O-ring. I can't answer that question.
DR. FEYNMAN: I've got the timing problem
mixed up with the 160 and 600 milliseconds and so on. Let me understand. Supposing the secondary seal is in place at some time and the primary seal has failed or something, and so the secondary seal is supposed to be pushed in by compression.
Now, it opens? Or is that not the way it works?
MR. MASON: Now it opens.
DR. FEYNMAN: How does it stay against the metal?
MR. MASON: Because the pressure is high enough. You see, it's now got 800, 900 psi, so now the pressure overrides the lack of resiliency.
DR. COVERT: Have you calculated how far this thing opens during this twanging motion because of the - it is a two-part system and it is a pin-connected thing, with a clamp there and the rest of it acts like it's free, in effect.
MR. MASON: We've calculated how fast it opens and how much due to pressurization, and I believe that the data will show that that is the primary driver, that that overrides any impact from the loads.
DR. COVERT: From the bending.
MR. MASON: But that I can't say has been verified yet.
DR. COVERT: But I agree with you that the pressurization deformation may be large compared to the
bending deformation. But if we are at a marginal situation, it seems to me you can't throw away the small part, because that might be the thing that tips it across the edge.
Is that a reasonable way to look at it?
MR. MASON: I think we have to look at every small potential contributor, yes.
MR. HARDY: Mr. Chairman, I might make one comment to Sally's question earlier about calculating the effect of the transverse loads on the joint. We did in the early part of the program run static structural tests, and I talked about the dynamics of the joint, but we did run static structural tests where the test article with a flight representation of the joint under pressure, under a number of pressure cycles I think it was said yesterday. I don't know how many, but a large number of pressure cycles, where we did put in the transverse load and the maximum design transverse loads on the case, both at the aft joint and at the forward joint while we were pressure cycling.
GENERAL KUTYNA: But did you do it prior to pressure cycling? Because that's when it happens.
MR. HARDY: We did it in the precise sequence of the pressurization: the twang load, the pressurization, and then the flight dynamics.
GENERAL KUTYNA: And how much opening or closing of that gap did you get?
MR. HARDY: Well, we measured the opening and closing. We measured it for the to effect, that is both the pressure effect and -
GENERAL KUTYNA: But now the pressure happens afterwards. How much did you get?
[626] MR. HARDY: I can't recall that. I don't remember that. I don't remember the precise effect of the pre-ignition load on the joint.
GENERAL KUTYNA: But would that be a factor if we have in fact a seal that has lost its resiliency, that you would be concerned about how much that gap opened and closed and squished that seal prior to the pressurization?
MR. HARDY: Depending upon the effect and how much that load might open the gap, it could be a factor. And we are after that answer right now.
MR. SUTTER: On these tests, was it just on one set of O-rings, or did you change your gaps and the dimensions of the O-rings? Did you run a variation?
MR. HARDY: I can't remember how many sets of O-rings were involved. I do remember that repeated pressure cycles were done on the same O-ring, but I can't remember exactly how many sets of O-rings were
involved.
CHAIRMAN ROGERS: You took part in all of these conversations yourself?
MR. MASON: I took part in the final conversation from the beginning, the whole telecon, from when we faxed the charts down until we drew the final conclusion. I was there.
CHAIRMAN ROGERS: Mr. McDonald testified, at least I thought he did, that the recommendation was not to launch, Thiokol's recommendation was not to launch. Who - at one point. Who was involved in that discussion leading up to the decision not to launch?
MR. MASON: All of us that were in the meeting, we were gathered and we were reviewing the data and attempting to reach our recommendation simultaneously.
CHAIRMAN ROGERS: Now, could you give us, without being too precise, the nature of the arguments against launch? Tell us what was being said. There must have been a consensus against launch if that is what you conveyed to NASA.
MR. MASON: Well, at that point we were trying to meet the deadline of getting the data together and getting the recommendation in, and so the engineering people were generating that and put together the
charts.
CHAIRMAN ROGERS: What were they saying, though?
MR. MASON: They were saying, we're going to be outside of our data base to go colder than 53 degrees, and we're concerned about whether the O-ring will move fast enough to seat and seal before the joint opens up. And that was the thrust of the issue, is not knowing exactly how long it would take for the O-ring to move into position.
DR. COVERT: Jerry, if the O-ring has been eroded a little bit or ablated or charred or whatever the damn thing does, it is now oval rather than round, and the action - is this going to tend to rotate vertically? Is that the stable motion, or does it tend to, say, slide in, skinny in the vertical direction as opposed to rotating like this?
MR. MASON: I don't know. Perhaps. There is probably some theory there, and Roger is probably the best one to answer that. I'm not sure we know exactly.
MR. BOISJOLY: That was one of my major concerns, and I addressed that as a timing function to seal. And I believed and I still believe and I believed that night that there isn't anybody on the face of this earth that can tell you exactly the mechanism that happened in that joint.
And even before the fact, you don't understand if it's going to rotate and walk up and delay or either slide because of its stiffness and delay. But the timing function that I spoke of that night had to do with the fact that I was afraid that that timing function could throw us in from an ignition transient at the start to somewhere after that start time, and that is what my major concern was about.
DR. COVERT: I think Roger and I are thinking along the same line. If it rolls in, then it is not a problem because you have the full exposure. But if it slides in and it is not skinny, it may in fact continue to blow by.
MR. BOISJOLY: But my concern even went beyond that point, because as it is performing this function at the beginning of the transient cycle, it still is being attacked by hot gas.
DR. COVERT: Precisely.
MR. BOISJOLY: And it is eroding at the same time it is trying to seal, and it is a race between, will it erode more than the time allowed to have it seal.
And that was my major concern, because SRM-15 showed erosion and hot gas blow-by at a low temperature, and that was the major issue on the table at that time.
DR. COVERT: Thank you very much.
CHAIRMAN ROGERS: Did you change your mind?
MR. BOISJOLY: No, sir, never.
MR. SUTTER: Can I ask a question, Mr. Chairman?
In these conversations, this discussion about this simplex seal right there, was that discussed with you?
MR. MASON: Well, on the telecon Mr. Mulloy did go through his rationale and talked about the simplex seal.
MR. SUTTER: Did Thiokol agree with the fact that you could work with the simplex seal? Did that influence your decision? What if there was no statement that the criteria said the simplex seal is okay? Would you have still made the decision you did?
MR. MASON: I'm not sure I understand that question.
MR. SUTTER: Well, I guess my question is, were you designing the same criteria that NASA was designing, and who established the criteria, and what were your ground rules for design?
MR. MASON: Well, let's see. If we get back to that fundamental design criteria, I really can't answer that. I'm not sure whether Joe or Bob can.
I know what we were endeavoring to do was to be confident that it was safe to fly, and we weren't
going back to that kind of a fundamental issue.
MR. WAITE: I have the same question in regards to the initial conversation about the single seal versus the two seals. You didn't really address the simplex seal until the latest conversation.
MR. MASON: Our thought has always been that it is essential to have a seal in that first 100 and - well, 160, 180, 200 milliseconds, before the joint opens up. If we get a seal at that point, then it doesn't really make any difference whether it's the primary or secondary. Once it is sealed, then it is reliable.
[628] And so from a practical standpoint, we have been addressing getting the high confidence of getting that seal at ignition, and in that sense - -
MR. SUTTER: But you're saying that when that seal - when one seals, then there's no assumption that any other failure might wipe out that seal?
MR. MASON: Well, everything we have known and everything demonstrated that once you had that seal, that there wasn't another cause that we could identify that would cause it to fail, because once it's sealed there is no more gas flow and then it is simply a benign environment.
DR. COVERT: But if it seals on the secondary flow before the joint enlarges or rotates, whatever you
call it, and now it is a benign environment because there is no gas flow, and if the primary seal is not yet seated and now you get joint rotation and you feel the pressure, is what you said before, the pressure is sufficient to deform the secondary seal so that the primary will never be called into play again, is that right?
MR. MASON: That is correct. And there is - on that point, there is a unanimous position. There isn't any difference in that, that once it is pressurized it will handle the joint rotation. The issue is whether it gets pressurized rapidly enough.
VICE CHAIRMAN ARMSTRONG: I am unclear as to why you considered 53 degrees was the limit of your experience base, rather than the 47 that you had had in the static test.
MR. MASON: Okay, I will explain that. In the static test, we static test it in a horizontal position, and in the process of mating the segments the putty is exposed differently. It flows differently than it does in the vertical mode.
You don't have to touch it in the vertical mode, but in the horizontal mode we have to go inside the motor and tamp the putty to get it in the right position. So our feeling was that that probably made -
well, it made a putty job that could be classed as better than the flight condition, and so it may have masked something.
And so we were hesitant to use it absolutely as evidence that it was okay.
MR. WALKER: Why didn't you ask for inspection of the putty seals in the flight, or didn't you think that was important, the vertical mating?
MR. MASON: Well, we had finally reached the conclusion that you can have - you can trap air in the mating operation. You can cause an air void when you leak test, when it blows back into the putty, and you can't be sure. You can go look and if it hasn't erupted through the putty you may have a path that is there but it hasn't exposed itself.
So even if you inspect it, you have the possibility of that path. So we felt that inspection wouldn't eliminate that possibility, and so we had to have a design approach that would tolerate that. That was our thought.
MR. WALKER: Can you respond to the question as to why the shuttle SRB design is different than the Titan design in terms of these seals? Were both of these designs done by your company?
MR. MASON: No, no. United Technologies.
MR. WALKER: That is probably the answer.
[629] MR. MASON: And I think maybe there are some difference in burning surface requirements. I'm not sure. But anyway, I know the Titan was considered at the time the joint design was done, but I think there were reasons.
But the point I would like to get to, however, is that we had the caucus and had these discussions, and it was clear that we were not going to reach a unanimous position there, and so we were then faced with the issue: Shall we stay with 53 degrees or is it reasonable to extrapolate?
And we had the reasons that I've already identified that made us feel that it was reasonable to extrapolate. So our final conclusion was reached by me having a review with the vice president of engineering, the program manager Lund, Kilminster, and the general manager Wiggins, and my own opinion.
We collectively agreed that it was reasonable to extrapolate, with the rationale that was listed on Kilminster's chart that said that we had a substantial margin, that we could erode more in the primary by a factor of three than what we had seen before; and that even if that didn't happen, that didn't seal, we had the secondary in a position where it did not have to move in
order to seat, and therefore it would seal.
And that was the basis for our conclusion. And we have brought with us today the people that had objections. In fact, Roger, as you just noted, is one who says he didn't change his mind. But it was one of those where it becomes the responsibility of management to make we think a rational decision, and that is what we did.
MR. WALKER: How many people in the meeting were considered seal experts? That is, their primary interest or their knowledge would qualify them as particularly knowledgeable on seals?
MR. MASON: Well, to varying degrees I guess there's a half a dozen people there. Bob, you would be best to respond to that.
MR. LUND: It is a matter of degree. It's a judgment factor. Certainly Roger Boisjoly is a seal expert, and Arnie Thompson.
MR. WALKER: Did you seal experts agree, or is it kind of a confusing issue?
MR. LUND: There was much data, and I will present that data to you in a little bit, that is both ways. Some say it's okay, some say it's not. So each engineer that was there reached his own conclusion from the data that was presented, and so there is a diversity
of opinion.
MR. MASON: Mr. Chairman, I did make one point while you were gone that was consistent with your earlier request, and that is that when we came down here we did review - and first, I made the point that we did not have a unanimous opinion, and because of the fact - we had a unanimous opinion among the key people reporting to me, but because of that we did identify and talk to the people who had concerns, and we brought the people that had concerns, so that if you wanted to you could talk to them, or if they wanted to they would have an opportunity to speak.
CHAIRMAN ROGERS: Very good. Thank you.
MR. MASON: Now, that is as far as I was intending to proceed. I thought Mr. Lund could go through the charts that we looked at that night.
CHAIRMAN ROGERS: Any other questions?
[630] MR. WAITE: I have one more. I still haven't got the sequence of events in terms of this single versus dual seal concept. Did the dual seal discussion come up after you had been contacted or after your caucus, or before the caucus?
MR. MASON: Well, I think that it became a thought, a primary thought, after we had said that we ought to fly to 53 and we had that ensuing
conversation. And it was pointed out at that point that the secondary seal was in position. And we had been focused so much on the primary seal that we said: Wait a minute, let's take a look at that, because we really hadn't - -
MR. WAITE: Who pointed it out?
MR. MASON: Mr. McDonald. He said, don't overlook that. His comments were that the cold is in the direction of badness, but that - and that is the concern with the primary. The secondary, however, is in a better position. Don't overlook that.
And we said, well - -
DR. COVERT: Could I ask you one more.
MR. MASON: Well, I guess we ought to have Al say just exactly what he said.
MR. McDONALD: I commented that lower temperature is in the direction of badness for both O-rings because it slows down the timing function, but that the effect was much worse for the primary O-ring compared to the secondary O-ring because the leak check port puts the primary O-ring on the wrong side of the groove, while the secondary O-ring is in the right direction, and that this condition should be evaluated in making the final decision for recommending the lowest acceptable launch temperature.
Based on the data presented in chart 2-1, I considered this very important because, depending upon how much delay one has with getting a good, reliable primary seal affects the capability for the secondary O-ring to seal.
MR. CRIPPEN: Excuse me. And we made that in light of even though we had a CIL saying the secondary might not be there during separation?
MR. McDONALD: Well, the charts that were presented that night, one of the charts, and I had presented that chart earlier in August in a meeting at NASA headquarters, and it was presented that night and we looked at that and it showed the - -
DR. KEEL: Do we have that chart, so all the Commissioners can see it?
MR. McDONALD: Yes, in Mr. Lund's presentation.
DR. KEEL: Can you pull it up now if you're going to talk about it, though? Can you put chart 3 up and let Mr. McDonald make his point.
(Viewgraph.) [Ref. 2/14-3]
MR. MASON: Do you want to do it right now?
DR. KEEL: Yes, it is up there.
MR. McDONALD: That was one of the charts, and what I was looking at was that we could conclude that this lower temperature doesn't affect the timing functions,
which I said I really feel it does - so that we really don't change the time for the primary seal to really seal, what is being eroded, up to the 170 milliseconds. That is, we haven't changed this launch from any other.
[631] However, if the colder temperature takes longer to not only move the seal, but also to extrude it in the gap because it's harder, then that might throw us into those longer times where, as you can see, it reduces the probability of the reliable secondary seal.
That was my concern, and if you go through it far enough it says that we had already stated that we have a high probability of no secondary seal because of the rotation problem, and I said that is a very important consideration.
MR. MASON: Let me address then the CIL question, because the basis for - -
DR. COVERT: What's a "CIL"?
MR. MASON: Critical Items List.
The reason that that was changed from the 1-R to 1 was this very rationale here, which said that after early on, after the 170 or 330 milliseconds, you didn't have a redundant seal, and so it was changed because it was not redundant all the time.
But it didn't really remove the redundancy at ignition.
CHAIRMAN ROGERS: Could I put the case and oversimplify it just a bit, because this is a good opportunity to be argumentative, if you will, on why you changed your mind. First, it was criticality item one, so that meant that if it was a failure, the mission was a catastrophe, and there had been discussion about that, a lot of it.
Then you made a decision, you and Thiokol made a decision, to recommend against launch for this very reason, that you were concerned. And then Mr. McDonald, in describing what happened, said that Thiokol had notified NASA that they should in effect not launch or conditions were not appropriate for launch, however you want to say it.
And then there was a phone call and he said, the conclusion being that Thiokol was directed to reassess all the data because the recommendation was not considered acceptable, in that - he was a little unclear about what he said, but anyway the recommendation was not considered acceptable.
And then you were supposed to have a five minute phone call caucus, which lasted 30 minutes, and you changed your mind.
Now that, you know, it is hard to explain it to an average outsider. You would think that that was
one of the most critical things you can imagine in this program, and it had been so considered by NASA for a long time, and you obviously knew it.
What caused you to change your mind? Try to explain it. The impression is that you were directed to do it, that there was so much pressure to get this launch off that you were directed to do it, and you did it.
Now, if that is not the case, try to explain it in language that the public will understand: Why you changed your mind and how you did it so quickly?
MR. MASON: Well, I think I would start by saying that when we picked the 53 degree temperature it was clearly the most conservative approach, and it was done because we had only gotten to a position in time to get to the telecon discussion. We had not established it as an unequivocal position, but rather one that was conservative, that stayed within our experience base.
That was what we were thinking at that time.
When we had the telecon, then the other factors that I mentioned came out, and I would not characterize that NASA said, directed us to reassess. What NASA said was: Here is our feeling [632] and our rationale, and we wonder whether you feel strongly that that 53 degrees is as low as it is rational to fly.
And I think we all recognized that it certainly was good at 50 degrees and 45, and so the question wasn't one of do you go below 53 degrees.
It was a question of how much below 53 can you go.
DR. RIDE: What data did you have to make that extrapolation? Did you have any data at all?
MR. MASON: Well, we had the effect of the temperature on the durometer of the O-ring. We had the one blow-by test that showed that it didn't blow by at 30 degrees. And other things that don't come to my mind that are in Mr. Lund's charts.
We had the higher margin of erosion. I mean, we had the ability to tolerate considerably more erosion than we had ever had and still seal.
But the reason for the debate was the fact that we didn't have hard evidence of how quickly that O-ring would move. And it became a matter of judgment rather than a