Transcript – Columbia Accident Investigation Board Press Conference Tuesday, April 1, 2003 (Part 3)
ADM. GEHMAN: Keep in mind that Revision No. 15 is dated like six weeks after the accident. A lot of work.
A REPORTER: NBC. For the chairman. I would like to ask about the air search issue. Those of us who are covering the story were shocked and horrified by the crash. Our condolences to the team and the people involved in the crash. I’m interested in the other air search that was work going on with the Army compass sensor that has apparently now been withdrawn. Can you describe to me how helpful that sensor was to your location and whether you intend to try and get it back to do any searches out west?
ADM. GEHMAN: We are receiving a brief tomorrow from the chief of the debris collection team. So I will have to defer any answer specifically to platform by platform just because I just don’t happen to currently know it. I will say, however, that the debris search continues with the same vigor as it has been. We’ve got between 4500 and 5500 people every single day still picking up debris, and it continues to be enormously important. Obviously this OEX recorder is a piece of debris. I mean, that’s how it was found; and it was found by somebody walking along, 6 feet away from the person beside them, and it was found in an area which had previously been gone over. So it just emphasizes how important this is.
So we are going to receive a brief tomorrow on the status, and the purpose of the brief is the beginning of the process of trying to determine how to terminate the debris search. We still have several weeks to go before we do that, but we are starting to talk about it. Air searches have been very important, particularly for the larger pieces in the wide-open area. Also, as we complete the debris searches in the areas where most of the debris is, which is essentially in East Texas, we now have to ask ourselves what do we want to do in the areas where there is less probability of finding debris but the debris that you’re going to find is more important. Obviously walking side by side, 5 feet apart through West Texas is a different story when there’s not much debris out there. So we don’t know the answer.
A REPORTER: News 24 Houston. Admiral, at the media bridge yesterday on the information you’re giving on the OEX recorder, a question was asked of you whether or not the debris that you collected represents any part of RCC Panel No. 9. Did you find that out?
ADM. GEHMAN: Roger.
MR. TETRAULT: The question was does any part —
ADM. GEHMAN: In other words, do we have any debris from RCC No. 9 or the attachment points?
MR. TETRAULT: RCC No. 9. Let me take just a second.
ADM. GEHMAN: I took that down and we passed it on to our folks and we’re digging. I also by the way —
MR. TETRAULT: There is nothing on RCC No. 9.
ADM. GEHMAN: We have nothing on RCC No. 9.
MR. TETRAULT: Let me just give you one other piece of information if you would have asked this — which if I were sitting out there, I would. What carrier panels do we have and what carrier panels are missing from the lower edge? We have pieces from carrier panels on Area 5, 6 — I’m sorry, this is the missing ones — from 5, 6, 8, 9, and 12.
ADM. GEHMAN: Missing.
MR. TETRAULT: Missing.
We have pieces from 1, 2, 3, 4, 7, 10, 11, and 13 and 14.
A REPORTER: NBC News. General Barry, could you provide just a little bit more detail on the process through which the pinholes were created on the RCC? It’s a chemical reaction created by this interface with the paint?
GEN. BARRY: It is; and let me just give you a couple of other data points, if I may. The root cause of the pinholes in the RCC, of course, is not generally agreed upon; but the leading candidate we’ve got right now, as I said, is zinc oxide. The other possible contenders were salt spray from the ocean and a TEOS application providing oxygen sourcing.
TEOS is an application that’s applied to the RCC in its development. What you do is you have this oxidation element that goes on that makes these pinholes and then, of course, if they get too large, they get refurbished. I do want to give you this detail, if I may; and it was highlighted last week a little bit. There have been 11 panels and 12 seals on the left side that have been refurbished, and then there have been repairs on 7 panels and 6 seals on the left side. Let me just give you an idea of what happens when we do the refurbishment, and then I’ll talk about the repair real quickly.
If you’re going to do a refurbishment on the RCC, first of all, you inspect the part, you disassemble it, you put a Type A sealant on it, you clean it, you repair the pinholes, you heat-treat the part to 1800 degrees, you apply the TEOS, and then you apply a second Type A sealant. That’s refurbishment. And by the way, refurbishment has to be sent back to the vendor in Dallas. If you repair it, as I have listed the numbers here, what you do is you clean down the damaged area, you only apply a Type A sealant, you heat-cure it, and then you verify the contour is flush and send it on its way. So what we’ve been able to do is track the numbers of them; and as I said, per RCC, there were anywhere from 20 to 40 in them as they have aged over the years. And then we either refurbish it or repair it, depending on how large the pinhole is.
LT. COL. WOODYARD: Any other questions from here? If not, we’ll move to the phone bridge.
A REPORTER: NBC. We’d like to go back to the carrier panel. It is now a likely suspect for the mystery piece you discovered the day after the Columbia reached its place in space. When it was going up and the impact took place, was the Columbia locked in its cradle right after that wind shear? Did that cause something there possibly to spring loose that carrier panel, or could it have come from the Palmdale modifications in ’99 or even from wear of age? What’s your best guess?
GEN. BARRY: Well, there’s a series of incidents here that we’re looking into. Let me start with the Palmdale OMM. They did do the normal depo level work on the orbiter in 1999. As you recall, it flew one more time at STS 113 and then, of course, STS 107. After the OMM, they, of course, removed all the RCC panels, put them back on, and then they found with a step-and-gap test, you know, to see if they were all flush and the right distances between them, that it was not lining up. They had to take them all off again. Then they got some of the old heads that had done this before and reinstituted and put it back in the right order. So there was some concern on our part about whether that was done right. All indications are that it has been done right.
The second thing is when the RCC follow-on flow to see what was done with any kind of changes. We know it was up near the leading edge of the No. 1 RCC that some work was done in the flow between STS 113 and 107. So what we’re seeing is a series of things. We’re trying to examine if any of them had factors. It’s a long story, and it’s still going to take a lot of investigation to be able to go through each and every one of them, from the paper review to the quality assurance review as well as the interviews that we’ve been able to do with some of the folks.
ADM. GEHMAN: The second part of the question was whether or not Q-loading or wind shear or anything like that on ascent could have contributed and what was the timing of it.
GEN. BARRY: As you remember, I brought up an issue where we had some what they call I-load preparation at 62 seconds. There was an out-of-experience movement of the solid rocket motor that put the orbiter in position where it was supposed to be countering a wind shear. We’ve looked into that pretty thoroughly right now. I’m not fully convinced yet. We’ve got another briefing this Friday coming in from NASA integration here at JSC, but all indications are that this was well within tolerance, certainly for the solid rocket motor to move, but also this being at 62 seconds, 61 seconds, that’s about max dynamic pressure, pretty close to it. So we’ve still got some concerns about it, but we do have a briefing coming in Friday and we should be able to nail that one shut as either a contributor or maybe one more piece of the story.
A REPORTER: Newsday. There was a mention of 11 flights previously where there were bit flips prior to entry interface, and I’m just wondering what conclusions you draw from that.
MR. TETRAULT: When I mentioned that, I was referring to left main landing gear brake Temperature D, which is the first one that went off or went off nominal in the prior time line. The conclusion that NASA had was that that temperature sensor may not be off nominal and that the time line may need to be corrected as a result of that.
ADM. GEHMAN: And that we had seen this before.
MR. TETRAULT: I say may. They have not made the decision that that’s correct, but they had found that 11 flights prior to that flight had had that same kind of impacts prior to the same time that this flight did.
A REPORTER: Aviation Week. For Admiral Gehman. You mentioned that underneath the context factors. I wonder if you could talk a little bit more about the methodology you’re going to use as you get at those and how you’re going to try to drive out such activity and things like the psychology of continued success.
ADM. GEHMAN: We’re going to approach it from two angles. Group 4 has put together a panel of people who have written on, commented on, and studied manned space flight for ever since the beginning of it. They are going to put down some macro-level thoughts — that is, thoughts of previous studies. NASA’s always being studied by somebody, and manned space flight is always under review. So there’s a rich treasure trove of studies and data and views. They are going to collect those, as well as their own. They’re going to take source material like not only budgets — I’ll just give you a little example here.
Take budgets, for example. There are budgets and then there budgets. There is the budget which is appropriated by Congress that NASA uses to spend on the space shuttle, but you can go to another level of degree on budgets. You can compare what the space shuttle program asked for with what NASA submitted with what the administration submitted after OMB marked it up with what the appropriation committees finally gave them.
And there’s actually a trend in there or there may be a trend in there. For example, you can find trends where the space shuttle program office asked for a lot of money and was not supported by somebody, any one of those categories. You can find trends where the administration asked for money and Congress didn’t give to it them or you can find trends where Congress felt that NASA was underfunding the space program and Congress gave them more money than they asked for. There’s all kinds of trends in there.
So Group 4 is going to take a kind of top-down look to see what history teaches us, to see what the previous studies teach us, to see what NASA and the administration have been told or were warned about. Meanwhile, my colleagues here are approaching the same question from the bottom up. As they do their inquiries, they are coming across habits and traits and events that they have looked at in the STS 107 investigation; and then we will intersect the two. Either their work will suggest to Group No. 4 an area that they ought to look at — like I think the normalization of deviation was one of them that we mentioned — or else, as the Group 4 reports out to the board as a whole on what they’ve found, we will find specific evidences that either confirm or deny their theories.
For example, if they say that there’s been a migration of some work force trend — this is popularly written on about changes in the NASA work force — and they therefore find that the work force has done this, that, and the other thing, the work that people, my colleagues on the left will do, will either confirm or deny that. Then we weave that into our report. I hope that answers your question.
A REPORTER: I would like you to clarify a couple of things for me. You said that on the RCC panels there’s a tolerance level of 40 thousands of an inch for a pinhole. Does that mean below 40 thousandths of an inch, it will not be repaired or refurbished and, above, it will? Second, you said that there were 20 to 40 pinholes per panel. Is that over time or is that at any given moment you could pull an RCC panel and see that many holes and would those all be repaired or would those be passed on if they’re below a certain threshold? And I was wondering, finally, do you have to resolve the issue of whether the breach occurred in a carrier panel or an RCC panel. Is that a major development or conclusion that’s still facing the investigation? Thank you.
GEN. BARRY: First answer to the question. If it’s less than .04, that’s correct. It’s a visual test and kind of a press on the area to see if there’s any voids underneath it, the way NASA’s been operating to date. We do and are encouraging them to take some hard looks at NDE on RCC panels so you can get a better look at what exactly that pinhole has underneath it; but right now if it’s less than .04, yes, it does past the test.
The 20 to 40 is over time. This is not per panel per sortie or per launch. So over the period of time, as I said, the flight the first time it was noticed it was in ’92.
ADM. GEHMAN: But it does mean that on average over a long period of time there have been 20 to 40 pinholes either refurbished or repaired to each RCC panel, which tends to accumulate. Obviously what we’re studying is whether or not — we’re studying the effects of ageing is what we’re studying. We’re not suggesting that RCC panel disintegrated and fell off the orbiter.
GEN. BARRY: But we are concerned about oxidation. As the Admiral explained a couple of press conferences ago, it’s like termites, you know, underneath, digging holes. If we want to know what’s underneath those pinholes, then the only way you’re probably going to do is that certainly getting a little better discipline on the RCC non-destructive evaluations or examinations.
The final comment you had was the breach in the carrier panel. Yes, it is definitely one of the scenarios that is being looked at. We have about ten that we’re looking at for the main landing gear and as Roger kind of commented on a little bit in his analysis, but also from the RCC or the carrier panel but certainly that is one of the theories that we’re trying to go down and track down and nail shut.
MR. TETRAULT: Let me comment on the last one, if I could. We seem to have narrowed this down pretty well to the leading edge of the wing. In that leading edge, there are a number of items that could fail that could cause the accident. My list of those items includes the RCC panel, the T seal, the stainless steel structure which supports the RCC panels, the bolts that support all of those, and the carrier panel. So we have to sort our way through all of those and make a determination of which one is the real failure mode. If we fail to do that properly and we get the wrong one, then we could have a future accident. So we’ve got to bring it at least that far.
ADM. GEHMAN: Or else our recommendations would be so broad and general and not specific enough that they won’t help the return-to-flight decision. And we would not be happy with that.
A REPORTER: CBS News. Just to follow up on that, for either Admiral Gehman or Roger Tetrault. Do your failure scenarios, if you have a missing carrier panel either with or without the horse collar, is the missing carrier panel by itself enough to generate the heat that you see in that sensor on the front of the wing spar behind RCC 9? Do you need any associated damage to the RCC to make that happen, I guess is what I’m wondering. Finally just as an interpretive question, if you guys are seeing something that looks pretty good right now that’s an RCC panel, perhaps it came off on Day 2, why shouldn’t I consider that the point where this breach occurred? I mean, that seems pretty straightforward to me, assuming that your continued testing pins that down.
MR. TETRAULT: The last item, I’d say your logic could lead to someone’s death; but, you know, you can’t rule out all these without specifically knowing and make this leap of faith that something that you know happened between Panels 5 and 12 actually happened in this particular area. You just can’t make that leap of faith.
With regard to the size of the hole, without running a thermal analysis, I can’t answer it specifically, but I would say I would tend to think that it would be more than plausible that that size of the RCC panel could cause all of these issues. We do know that the thermal analysis when it was done down at Panel No. 6, which started with a 6-inch hole, tended to indicate that the heat could get back into the wheel well and so on and actually support an accident. I did a quick check on the square area of an RCC panel at Position No. 6 before I came here, and it was about 92 inches. So given the heat, the thermal stuff that I saw based on 6 inches, I would say it’s pretty well assured that 92 square inches could give us the kinds of conditions that we’re seeing here. I’m saying that with some degree of confidence, not having run a thermal analysis.
ADM. GEHMAN: That’s right. We have to caveat that by reminding everybody of my introduction in which the thermal story has to line up with the aerodynamic story which has to line up with the time line reconstruction story. So if you assume that you have a breach of some sort in the vicinity of RCC No. 6, either a carrier panel or a broken piece of RCC or something like that and then your closest temperature sensor is in the vicinity of RCC Panel No. 9, you have to calculate whether or not the heat can get over there and how does the heat get over there and how would the sensor react. Then you have to find out what the aerodynamic forces are. So you’re way ahead of us. We would in no way conclude that we found the initiating event here.
A REPORTER: Earth News. For General Barry. With what you showed with your hand motion of the maneuver of the shuttle on Flight Day 2, was that at 23 hours after launch, the inertial measurement maneuver, or was it some other time and, if so, do you have that time? All I’ve heard for re-entry was four days later over the Pacific Ocean. The Pacific Ocean’s fairly large. Can you narrow that down a bit or at least give me a time for when it re-entered and whatever the mystery object is? Any idea of what the orientation was, especially if it was a carrier panel? Was it edge on, slip stream, or sideways or what?
GEN. BARRY: We’ll have to get back to you on the exact time. It was on the second day. I just don’t know the exact location. My understanding was the velocity vector was going backwards. You know, they did about 250 maneuvers on these 16 days as part of their experiments. So the orbiter was changing many, many times. On this particular case, it was going on a rearward velocity vector. It did turn into a right wing going downstream and then came back to having a rearward velocity vector. In the course of that time — and I don’t know exactly where it was located — we can get that for you — that is when we saw this piece come off, and we’re relatively confident now that it did come off the orbiter. It wasn’t something that showed up somewhere as a meteorite or something like that. The velocity vector that it had coming off there did lead us to believe that it did come off the orbiter.
Things have come off in the past, as you know. We’ve seen blankets when the doors have opened, and astronauts have told us that screws and different things that have been debris in the inside of the payload bay have come out. So this is not unusual. But the size of this — and certainly to be tracked. Now, remember this was not picked up by any DOD sensors until six days — really we started looking right after the mishap, but it was a combination of all of the looks from all those radars that I mentioned to you and by the sixth day we concluded that we did see something there and then we tracked it through to when it burned up in the atmosphere.
ADM. GEHMAN: Let me follow on John’s answer there. John said we saw it come off the orbiter. We did not see it come off the orbiter. The first radar observation of something on orbit, accompanying the orbiter, as I recollect, was an hour after the last of these yaw maneuvers. So the orbiter was back flying stern first and the DOD system just happened to look at the orbiter about an hour after that and there it was. So it would be wrong to characterize that we saw it come off the orbiter.
The second — part of the question was the object is tumbling. Is that correct? The object is turning in space? It doesn’t have an edge attitude or anything like that?
GEN. BARRY: Initially it doesn’t show a lot of tumbling. Of course, as it re-enters the atmosphere, it starts going into more tumbling. Sheila Widnall has been looking into that as something else that we can combine with Lewis Labs and be able to get some more information. So there’s still some more work to do.
ADM. GEHMAN: We can calculate the mass and shape of it.
A REPORTER: New York Times. Could you tell me, please, do you have a candidate mechanism in mind for how on earth you get a debris strike near max Q on liftoff, you damage the carrier panel either on the upper side of the wing or the lower side of it — I don’t know if you can say which — and it survives all the way into orbit and is gently shaken loose by the motion of the shuttle yawing?
ADM. GEHMAN: Well, the answer is, no, we don’t. But part of our testing at Southwest Research will be to characterize debris strikes different than the body of knowledge about debris strikes in the past, to see whether or not these debris strikes can create the kind of debris that the panel here has been talking about, including a carrier panel.
As far as the second part of your question, all of us have spoken to astronauts who on their mission have looked out in space and seen some part of the shuttle, which has been attached by nuts and bolts and screws, go drifting by. So, you know, these thermal blankets in the payload bay, they’re all held on with clips and they’ve got wires. They’re not supposed to come off either. One of the astronauts told me he looked out the window on Day 2 and saw a washer go by.
So, no, I don’t think we’re going to be able to prove that something came off, but I know we’ll be able to get plenty of testimony of instances in which things which were firmly attached to the orbiter at one time came unattached later on. So that’s the best I can do for answering that question.
GEN. BARRY: The only thing I might add is, you know, we did have a negative angle on attack on launch for a brief period of time. Then it goes to positive; but the higher you go, the less dynamic pressure you have. Of course, the less dynamic pressure you have, you know, maybe there is some issues there where there can be less chance for things to come off. We haven’t quite figured that out, but certainly with the testimony we’ve gotten, as the Admiral has said, and the indications that we’ve had from certainly the radar cross-section, you can’t rule out the aspect of something coming off the orbiter. We’ve still got some more work to do.
ADM. GEHMAN: One of the more common events are blocks of ice coming off the orbiter. Occasionally big blocks of ice which are attached to dump nozzles or exhaust nozzles — and they’re really attached pretty hard, I guess — will just detach themselves and float away.
You might mention ice as a radar reflection candidate.
GEN. BARRY: We did. We looked at it, and it didn’t meet the test.
ADM. GEHMAN: Ice is out.
A REPORTER: Discovery Channel. I’d like to know if you could please go back through three different points in time, to the time of the Boeing analysis during the shuttle flight to the data from last week and from the new analysis now and give us the dimensions of the debris, of the foam, and the probable strike zone on the orbiter and how that has changed as time has progressed.
MR. TETRAULT: I can give you what it is now. I’m not sure what was used in the analysis, so I can’t make the comparison for you. Maybe one of my other friends up here knows that data off the top of their head. What we’re looking at now, based on the photo analysis, is 25 by 15. These are 25, plus or minus 3, by 15, plus or minus 3, by 5. And I think that is probably larger than what they were using in the crater analysis, although I must say that what we’re looking at is a strike which is not in the area where they used crater analysis to analyze.
GEN. BARRY: I’d just mention a couple of things on the crater. It is 20 years old plus, and it’s a simpler algorithm than certainly we have today with hydro codes that are available. So I know NASA is looking into that as another thing that can be added to improve things. The model doesn’t really account for lift because, you know, it just takes the impact velocities, the angles, and the density. Stiffness is not really in the crater program to speak of. The crater can do 3-D orientation, ballistic characteristics, and the crater analysis did not really look at the worst case with carrier panel impact or RCC as a case in point.
The one that is interesting also is the debris assessment team judged that the densified layer will remain attached to the crater even though it predicted damage exceeding tile thickness. So we’re trying to not only look into the analysis that was presented for crater but also what follow-on improvements could be made based on the current technologies of today.
MR. WALLACE: I would just add two quick points, having gone through a lot of communication on crater analysis but not into the analysis in technical detail itself, is that there were several sets of dimensions. So I think there were different assumptions made. If it comes back to one specific number, we’ll get that and get back to you.
LT. COL. WOODYARD: This concludes the press briefing today. Again, we thank you for coming out.
