CAIB Press Briefing Transcript 29 Apr 2003 (part 2)

ADM. GEHMAN: Thank you very much.

MR. TETRAULT: Good afternoon. I’m representing Group 3, which is the technology group, if you will. My teammates this week are busy. Sheila Widnall is at MIT. Scott Hubbard is leaving to go to Southwest Research, where the foam tests will begin on Thursday. Dr. Doug Osheroff is here from Stanford, and he’s working on chemical analysis of debris. Dr. Jim Hallock is also here, and he’s working on fault tree closeouts and sensors.

What I would like to do today is go ahead and do a quick review of the critical evidence that indicates the location that we are seeing in the breach of the left lower wing. I’m going to very quickly note two items and then come to this slide. All of you have seen the photographic analysis that showed the foam coming off the external tank and striking the RCC panel area, which is measured to be in the 6, 7, and 8 lower-wing area. You’ve also heard General Barry talk about the Day 2 radar event. The importance of that particular event, as far as I’m concerned, is that the OEX sensors show that very early on in this particular incident we have heating in the left wing; and that, in turn, indicates that the breach was probably there and that it existed at interface. Certainly if that’s a fact, then it lends credence to the potential that a piece left the left wing, departing sometime during flight, and is the piece that we’re looking for on the Day 2 radar event.

Now, let’s look at this debris because that begins to point us into specific area in the left wing where we have some concerns. This debris shows some debris from Panel 9, 8; and this is 8, as well. What we see is there are some very unique events happening at Panels 9 and No. 8. These are the ridges, if you will, that go to the RCC panels and are normally attached to the panel on the sides. What is unique is that they have knife edges on the sides, which are shown here; and that is obviously a sign of high-temperature erosion during re-entry. Again, we don’t see anything like this anyplace else in the debris field. We also have a very heavy element of slag that is molten metal which has been thrown forward on to the back side of the reinforced carbon-carbon panel here at Panel No. 8. Then in addition to that, there was a third element that was unique in that the carrier panels which would be located out in this area, we do have the tiles from the RCC No. 9 area carrier panel and those tiles are significantly slumped and molten compared to any of the others that we have seen. So those are the unique features which are pointing us kind of in that area, or at least were up until fairly recently.

In fact, probably the favorite hypothesis which we were using at that particular time was, since this particular T-seal was missing, that it might very well be the T-seal between these Carrier Panels No. 8 and No. 9. And there’s reason why we favored the T-seal.

If I could have the next slide. It’s because it made a nice gap through which we could see the potential for entry of hot gases creating the kind of incident we had, and it offered a protected area. This is the gap, if you look at it here. What I’ve shown is a make-believe T-seal covering the upper part of the wing, if you will, leaving a gap without the T-seal if the bottom half of the T-seal had been extracted. If that gap is there without any of the T-seal structure, including this rib face here, then you would have a gap of 1.15 inches. If the rib face remained in, such as it is here, and the gap was the size of this red piece, then it would be .775 inches.

Let me have the next slide, please. Now, one of the reasons that we favored the T-seal was because we saw a lot of T-seal failures in the debris down at Kennedy; and there were three different failure modes that we observed while we were at Kennedy. The first of those failure modes is a cross-sectional fracture, which is shown here in Photograph A, where we just cut across both the rib and the face plate and it separates. The second failure mode was a face plate fracture, which is shown here in B. That would be this plate, which is normally the sliding plate against the RCC, breaks off at the intersection with the rib. Then finally there is a flange fracture where you can slap this face — and this is not the best photograph of it — but you can slap this face and the rib actually pops off. So you may have the entire face or some piece of the face, as we have here, without attachment to the rib. So those are three different failure modes that we’ve seen in the debris; and recognizing that they do, in fact, fail or could, in fact, fail, we tended to have a preference for a T-seal as a potential culprit and where the breach occurred.

Next chart. Of course, as many of are you aware, last week NASA put a piece down on the floor which was labeled as being T-seal No. 8. This is in the gap between Panels 8 and 9. This is the area that we previously had thought we didn’t have a T-seal for. They put this down as being certain that it was in the right spot. A subsequent audit of that proved that it was not a certainty, that, in fact, it was probably something less than that and maybe in the 70 to 80 percent range.

What brought them together on this issue is, in particular, this erosion mark on both sides of these pieces; and these two pieces do appear, because of that erosion mark, to be coupled together. When you couple them together, there are very few positions that you can put this piece down at. The preferred position was, as I said, at Position No. 8. However, since that time they have been doing additional work; and as of this morning, we have found similar erosion to this pattern here, in Position No. 10. So we’re uncertain at this time whether this piece belongs in Position 8 or Position 10. It’s probably a 50/50 at this particular point, so the potential of having no T-seal between Panels 8 and between Panels 9 is still a possibility.

Having said that, I’m now going to move away from that T-seal. If I could have the next slide, I’ll begin to show you why we are actually moving more inboard on the wing in terms of where we believe the event actually occurred.

This is a shot of an X-ray taken of RCC Panel No. 8. There’s actually three pieces of it. If you’re familiar with what’s been going on, these three pieces are all top half. So we have no bottom half of RCC Panel No. 8. What you see on the X-ray, you do not see visually when you look at these pieces of the RCC; and what you begin to see is deposits of heavy metals in certain patterns that we don’t see when we look at it visually.

Let me show you what the patterns are. If you look in this area, you can see that there’s definitely a pattern to how this area is being laid down. What you can’t see from the photograph that I’m showing you is that, in fact, when you get down into this area, there are small nodules of metal which are being deposited in a very uniform pattern, going straight vertically up and down; and this would be where the RCC attaches to the T-seal. So it gives you an idea of how that orientation is. This orientation is on the side.

So we are going, if you will, if you stood behind the RCC panel and took a paint brush and sprayed it at the back of the RCC panel, you would get deposits that were very similar to this; and they’re very uniform and very straight up and down. That begins to indicate that we’re not getting a lot of side flow from, say, a T-seal position or even a panel further back.

Now, let me go to the next chart. This is a closeout photograph of the spar area, looking from inside the wing at the wing spar in the front. What I’m drawing on here are six out of the 15 wire runs of OEX sensors that run from basically left to right toward the fuselage. We haven’t plotted all 15 of these yet because they’re extremely difficult to plot. You have to virtually do them by eyeballs. The drawings aren’t necessarily correct. For instance, the drawings actually show only three bundles; and as you can see from here, there are five bundles.

What I’ve also plotted is where we are directionally behind the spar. If you look at these four bolt heads that go through here, that’s where T-seal No. 8 would be. That’s the T-seal between RCC Panel No. 8 and RCC Panel No. 9. So that’s the area of the T-seal. RCC Panel No. 9 extends from here to here; No. 8, from here to here where this spar is.

I’ve drawn the wires in here and shown you where they run, and in some cases I put them front of obstacles so you can follow them, instead of behind the obstacles. For instance, here obviously it wouldn’t go across the front of this truss bar, it would go behind it, but I wanted you to be able to see and follow how they go.

The red one is a very famous sensor. That’s the temperature sensor that goes out on the front end of the spar in the clevis area between 9 and 10, and that’s coming in as one of the furthest out and running right along the center. Now, what’s interesting is on the far left. As I’ve shown you, the order of failure in seconds from entry interface — and these are the failures of those wires which we presume are being cut.

If you look at this, it tells us a few things. One, they’re being cut from the center outward. They go red, then yellow, then blue, and then deep blue. So it’s going outwards. It’s coming from the center. Two, it includes failure of these two wires, which obviously tells you that the breach is not anywhere outboard of this area. If it was, it wouldn’t be cutting these two wires. So it’s beginning to localize this; and what you see then is it’s localizing it further into 8 and away from T-seal No. 8.

Okay. In addition, what we see is this wire, the green one, which is this one right here, is the last one to go off, by some 25 seconds, which is a substantial period of time. So this breach, if you will, is progressing down the wall and hits this one. As you turn this corner — I haven’t shown this picture, but this wire actually comes back up in a short span of time and connects back to these bundles. So what this differential in time also suggests is that the breach and the cutting of these wires occurs somewhere in this area and not likely to be very far to the right of the picture that I’m showing you because otherwise this wire bundle would have been cut much sooner.

So the bottom line to the presentation I’m giving to you is that we don’t know yet whether that is a T-seal No. 8 or whether it belongs in Position No. 10. It’s probably 50/50. It is not likely, based on what we’re seeing, to be that significant that we are moving slightly to the right and, as the General said, we’ve had this oscillating pattern of where it is but it’s getting closer and closer and I feel that we’re probably within 30 inches of where the actual breach occurred now.

That’s all I have.

ADM. GEHMAN: Thank you very much.

LT. COL. WOODYARD: Okay. He’ll begin again. Let’s go over the rules of engagement one more time. We ask you that limit your questions to one question, not one two-part question or two one-part questions. Please give your name and publication, as well.

A REPORTER: Houston Chronicle. I think I want to follow up on Mr. Tetrault’s last comments. I’m wondering if you can sort of tie together what the radar cross-section work needs to do to help you move to the right place, because it sort of seems like you’re saying there may have been a breach either in the RCC panel itself or another fitting. I guess I’m not sort of clear where this is all pointing or pushing you now.

MR. TETRAULT: I think what’s still open is a piece of RCC Panel No. 8 and/or some portion of T-seal No. 7, the one that’s between Panel 7 and Panel No. 8. Those are the locations that these wires would tend to indicate that you should be looking at most closely.

In order for it to pass the two tests that there are on this ballistics, it appears, based on what we know, that it is likely to have to have an angle, it is likely to have to have a size that’s more than 100 square inches. It requires that size in order to get the ballistics and it requires the angle to meet the radar signature items.

So if it has to have an angle, it probably has to have a rib. Because of the pristine nature of the carrier panel’s tiles on the bottom of RCC 8 — they look like they’re new — we doubt very seriously that, if there is a rib that it went out in space, that it would include the bottom part of the bottom panel called the spar rib. It would not include that because that spar rib probably had to be there to protect those tiles. So it would be more likely to have a side rib as opposed to a bottom rib.

A REPORTER: Florida Today. For Dr. Ride. The Rogers Commission wrote a chapter in their report about NASA’s silent safety system, and I was wondering in your interviews to date after the Columbia accident how you would characterize NASA’s safety system today and what role it may have played in the accident.

DR. RIDE: You’re right, we did spend a lot of time during the Rogers Commission assessing NASA’s safety program at the time; and it’s our intent to, along with John Logsdon’s group, assess the role of NASA’s safety program today and see how that’s evolved. We’re at a relatively early stage in assessing that, I’d say. We’re spending some time understanding safety’s role, for example, in the disposition of foam leading up to the events and in the decision-making process during STS 107. We know that there were safety people involved in all the meetings during the decision-making process. You know, as organizationally they should have been, they were in attendance at the meetings. They didn’t make significant contributions one way or another in most of these meetings.

We’re not in a position yet to draw any conclusions from that because we really want to be able to take a step back and look at the larger picture; but we’re trying to lay the groundwork in almost the case study, if you like, that we’re putting together, that is, the on-orbit decision-making during 107 and the history of the foam events because we think that that will lay the basis for understanding how the safety program functions.

A REPORTER: Dallas Morning News. General Barry, I know you may not be ready to be specific; but I’m interested in what generally you think the interesting potential events are on ascent and this time line.

GEN. BARRY: Let me go over a couple of things. As the Admiral said, we want to get as much detail as we did on the re-entry. So here’s a couple of things we’re taking a hard look at.

First of all, the bipod design. The bipod itself has been designed three different times. The one that flew on Columbia was the third version. So we’re taking a look at those.

The application procedures, environmental charts when the bipod was actually built, to make sure it was actually within spec.

The bipod debris. We’re looking at the four previous debris problems where we have concrete evidence that it came off. There’s one in question, but it looks like we’ve got a couple of hard to look at.

Looking at the specs for the foam to come off and hit the orbiter. The spec actually called for a .006-foot-pound impact limit on the orbiter, which is absolutely minuscule. So obviously the orbiter was not designed by spec to take hits. So we’re taking a look at that.

Sally Ride’s group is going to take a look at what decisions were made, as she mentioned, in the certification of flight. Also the FRR, the PRCB, and the PAR are all parts of looking to see what management decisions were made.

We’re looking at the RCC aging, for the pin holes, the substrate loss. We know there was a mate, de-mate, and re-mate that was done on this.

There was crushed foamed that was identified prior to launch in the bipod region. So we’re taking a look at the PR, the Problem Report that was done on that.

We’ve looked at the time-on-pad statistics for how much moisture was evident and any infrastructure issues. The launch out of experience the Admiral mentioned. We’re looking at any and all of those, up through and including when the ET separated from the orbiter. And then, of course, the 81 seconds when the bipod actually allegedly came off.

We’re looking at the MADs data, the cryopumping, the crater use, the Huntington Beach experience that was made, and the move that contributed to the crater analysis and then the spectral analysis. Finally, we’re looking at the SRB bolt-catcher to make sure there’s no issues there, and then finally the second-day orbital debris.

That will include a time line that will cover quite a bit. We want to have the detailed analysis just like we did for the re-entry last week and we are certainly going to include all of those and many more.

A REPORTER: And the temperature rise in the left wing?

ADM. GEHMAN: Whether there was one or not. It’s not clear there was a temperature rise in the left wing.

A REPORTER: USA Today. For Dr. Ride. Since you’ve been doing more interviews about the decision-making during 107, can you shed any more light on what happened to the request for the imaging? You said, when you addressed us most recently, that there was kind of a tangle of relationships and discussions. Can you clarify on that for us, please?

DR. RIDE: A little bit. Since some of the information comes from interviews that were done under privilege, you know, we have to be a little bit careful about what we describe and how we describe it, as you can appreciate; but we know that there were several or a couple of different avenues where requests for photos or initiations of questions about photos arose. There was one from the tiger team that was working on this, as we discussed a couple of weeks ago. There was also questions about the possibility of photos that were initiated from people at KSC.

It looks to us, just to shed a little more light on the tangle that was this process, it looks as though the management chain heard about the requests that were coming from Florida, the questions about images that were coming from Florida, and at least in part were responding to those questions coming from, you know, not through a channel from engineers but from channels of just people who had seen the videos and were inquiring about whether photos should be taken and how to initiate the process, including making a couple of calls to DOD to ask that question.

Management appears to have been responding to those requests and saying, "Well, who’s got the requirement for the photo?" Those responses were occurring at about the same time that the request from the tiger team was making its way up through another channel. So at least in some instances, the, "Well, who’s got the requirement," kind of collided with the request coming up through another channel.

As a result, at least in part, the, "No, we’re not ready to take photos," going back to the people asking from the KSC side came down to the engineers as a no to their request. Now, this is not the whole part of the story, but it sheds maybe a little more light on the many, many people that were involved and the way that the different roles in the organizational structure kind of played into all this. That’s what we really hope to be able to tease out as we start looking at the organizational models to see what sort of useful things we can see from that.

A REPORTER: Did the MMT ever hear about either request?

DR. RIDE: It was not briefed in the MMT itself, no. The request for photos was not briefed in the MMT. Program management did hear about the desire to take photos, were unable to identify precisely where that request was coming from.

A REPORTER: New York Times. General Barry. You’ve been talking for weeks about doing foam impact tests at Southwest Research Institute. Have you started and, if you’re holding off, are you waiting for Mr. Tetrault to tell you where to aim the foam or where is the holdup?

GEN. BARRY: Yes and no, I guess, is the answer. Bottom line is we’re going to be monitoring that. It’s mainly Group 3. I don’t know if you want to talk to that.

MR. TETRAULT: The first testing of the foam is actually going to occur on Thursday of this week, and I think there’s a second shoot that will be on Friday. That will be against tiles from a presumed lower wheel well door. We really won’t get to the RCC panels until probably early June. That’s because they’re fabricating those particular panels. It will take a while to get them in the shape we need to get them in.

We’ve had great debate as to how that test should be run, as you indicated, in terms of how it should be shot, what types of instrumentation should be in there, should we only be looking at RCC panels or T-seals or both, what about the LESS structure underneath it, how do we try to get the best information about what’s happening to that structure as we make these shoots with very limited pieces that we have available to us. So there’s been an awful lot of scientific debate, if you will, about how best to run those tests, but they are being constructed right now and the expectation is that they will run on schedule, in June. For the RCC, the tiles will run —

ADM. GEHMAN: The testing starts this week but the testing is limited by the target, not by the bullets. We have plenty of foam. When we decided to do foam testing 30 days ago, we were talking about wheel well breaches, and so the first test rig that we built was a wheel well. That’s what we’re ready to shoot at. Now we’re talking about the leading edge system, and it will take us a couple of weeks to build a representative leading edge.

A REPORTER: This new leading edge you’re fabricating, that’s from the Enterprise or what?

MR. TETRAULT: It’s both.

ADM. GEHMAN: Part of it’s going to be new and part of it — we’re going to do both.

MR. TETRAULT: We’re shooting at a No. 6 panel which is an older panel which has had a number of flights on it. I don’t recall exactly how many, but I thought it was 20.

ADM. GEHMAN: We’re building up a leading edge section that’s going to be essentially from Panel No. 6 to Panel No. 9 so that we can shoot at anything we want. By the time we get ready to actually shoot at it, we’ll know more about what angle till we cover — we may want to shoot at an angle that covers two T-seals and one panel or two panels and one T-seal, and we may shoot a couple of times.

MR. TETRAULT: There are significant angular differences between a panel at No. 6 position and one that’s out at 8 and 9.

ADM. GEHMAN: So the limitation is the target. It just takes a long time to build that target up and get it instrumented right.

A REPORTER: Associated Press. For Mr. Tetrault. From what I gathered from your presentation, it no longer looks like T-seal 8 is in the running, so to speak, even if those pieces are not identified to be from No. 8, and that it’s more now No. 7. If that’s the case, do you have any pieces of T-seal No. 7 lying around and do you think that could have been the missing object in its entirety?

MR. TETRAULT: Let me answer the multiple questions here. With regard to T-seal No. 7. We do have piece of the upper half of T-seal No. 7, but we do not have the bottom half of T-seal 7. So that could be a candidate. I wouldn’t say T-seal No. 8 is totally out of the picture in this game. I’ve found that every time you think you have an answer, you find that you’ve closed but you don’t have an answer and something shows up that takes the wind out of your sails and you have to backtrack a little bit.

I would say it is getting more probable that the breach is a little bit further to the inboard side of the wing than where T-seal No. 8 is and we’re beginning to understand a little bit of the unique features that occurred at No. 8 that led us to look at T-seal No. 8. For instance, I showed you the slag; and that may, in fact, be from a breach in No. 8, not from a T-seal. The slumping that is on No. 9, I mentioned that the bottom flange, the spar flange protected the No. 8 tiles. There is also a gap from that bottom flange at the corners of panel No. 8. That gap becomes a potential funnel that funnels hot air right at the carrier panel tiles, and that may be a very good explanation for why we have this extra slumping at Panel No. 9 is that the heat is coming out of 8, down at the bottom flange, through that gap, and is directed directly on those tile faces. So I think we’re getting a better handle on why some of these pieces are unique, which allows to some degree to move a little bit to the right, as well.

ADM. GEHMAN: But I think, Roger, that it would be correct to say that a large fragment of an RCC panel is also — as long as it contains an edge, as long as it contains a flange, that a fragment of lower RCC is also —

MR. TETRAULT: Right, a fragment of the lower RCC, the T-seal at 7, a little less likely a T-seal at 8.

A REPORTER: And the RCC fragment at No. 8.

MR. TETRAULT: That would be at Position No. 8.

GEN. BARRY: And that’s one of the remaining five that we’re going to test at the radar cross-section.

A REPORTER: NBC. Back off on the debris. Now, that we’ve been moving back and forth with the RCC panels, we’re out a little further outboard. We’re almost far enough outboard, aren’t we, that the Maui AMOS pictures with viewing above the orbiter with the open payload bay doors, the doors eclipsed a good part of the inboard RCC panels, but at a certain point on the way out you begin to see RCC panels, the tops of them. I’m not sure exactly where —

MR. TETRAULT: I’ve seen that picture, and it starts at 9. We’ve seen from the wiring that we really don’t believe we had a breach anywhere from 9 outboard.