Transcript: Columbia Accident Investigation Board Press Briefing June 12, 2003 (part 1)

National Transportation Safety Board Conference Center
429 L’ Enfant Plaza, SW

Washington, D.C.

MS. LAURA BROWN: Welcome to our first Washington press briefing. I will turn you over to Admiral Hal Gehman.

ADMIRAL HAROLD GEHMAN: Good afternoon. I’ll make a few preliminary statements and then ask each of my colleagues to make a short statement about what’s going on, and then we’ll go around and do questions.

Obviously, the Board is toward the end of its transition from Houston to Washington, D.C. We still have a foot in each camp. We still have a small office down in Houston with some people working down there. We have board members going back and forth, still have board members traveling. But, the focus of our efforts is now up here in Washington, D.C.

We are focusing on writing – drafting I should say. But, that does not mean that investigating has ceased. As a matter of fact, I suspect that by the time we get finished here today, you’ll see that my colleagues here have some interesting things to report, and we’ll demonstrate very clearly that the investigation goes on.

But, what we’re trying to do is we’re trying to ramp down investigating and ramp up drafting. The purpose of that, of course, is to meet our goal of having this report done prior to the August congressional recess. I will repeat what I’ve said before, that that’s a goal. We will try to make it. It is my intention to make it, but we would rather get this report right than get it in a hurry, and I know Congress and the White House share that view.

That’s all I’m going to say at the front end. And I’m gonna start by calling on my colleagues here, and I’ll ask Dr. Ride to lead off.

DR. SALLY RIDE: I’ll be very brief because we’ve got some interesting things coming up. But, our group, Mr. Wallace, General Hess and I are all in Washington, D.C. We’re focusing on reading and wrapping up the work that we’ve been doing on decision making and also into the safety process.

I will say that one of the things that we spent just a little bit of time on recently, as we’ve been reviewing the MMT – the Mission Management Team – and the process that it used, we will probably be recommending increased training for the Mission Management Team as it prepares for shuttle flights in the future.

And I think, with that, I’ll turn it over to -.

ADMIRAL GEHMAN: – All right. Thank you very much.

Dr. Osheroff, are you ready to be next?

DR. DOUGLAS OSHEROFF: Well, I – first, let me say, this is the first one of these things where I’ve been representing group three, or anyone else I suppose. I suppose it’s because I’m naturally modest and quiet.

However, I really wanted – well first, let me say that this is – they’re having troubles with my computer -.

ADMIRAL GEHMAN: – That you’re not stable.

DR. OSHEROFF: It’s a good Mac G4 and I’m sure it’ll work eventually. It worked for me. Maybe I’ll have to go back and stroke it for awhile. Should we go on and – to someone else and then come back? No?

ADMIRAL GEHMAN: No, they’re – no, we shouldn’t do that.

DR. OSHEROFF: They’re not – I can’t do anything.

I am the foamologist!

Yes, it’s great, but why is it flickering?

Well, let me – I’ll start out – look, it’s – there it goes. I’ll talk anyway.

I’m actually an experimental physicist, low temperature physicist, in fact. And when Hal Gehman asked me to join this Board, I said well, what can I bring to this because I knew nothing about NASA or the workings of – I have nothing – never done anything that had to do with space before, except I guess, as Department Chairman, I assign space in the Physics Department at Stanford.

However, I became rather rapidly interested in the properties of this foam, which is really fascinating stuff. It’s highly anisotropic; that is to say, its mechanical properties depend upon which direction you squeeze it, for instance.

They’re still having problems with my computer, aren’t they. Why? It’s not -.

And so I decided that I would do some – some experiments on the foam. And I’ll simply describe the experiments. The idea was to glue a piece of BX-250 – this is the foam which is sprayed onto the bipod ramp, and this is the foam which is believed to be responsible for the breakup of Columbia on re-entry. Glued it to a metal plate with a little tube that came through the plate so that I could apply either a liquid or gas pressure. And the idea was to try to understand, as I increased the pressure, how ultimately this resulted in some sort of a fault that propagated through to the surface.

Now, this is important because, for many years, people at NASA have assumed, in fact, that one of the main mechanisms for foam shedding by the – from the external tank was that liquid cryogens, liquid nitrogen in particular, would somehow condense at the – you know, in a void or something inside the foam near the metal surface, near the external tank, and that as you started getting aeroheating, in fact, this liquid nitrogen would warm up, pressure would build up, and it would throw foam off of the external tank.

And so the question was, in my mind, do I understand the process by which a fault propagates through the foam? And the answer was no – and these guys don’t understand my computer either, it looks like.

And so we started doing these experiments and, if we ever got to – we’re not getting there at all. But the bottom – I think we’re not – I don’t know if these guys are help. I think that was the word.

What I found was, that in fact the mechanism by which the liquid expands is not consistent in any way with the ejection of foam from the surface. What it does is it tends to make a two-dimensional, rather flat crack or fault, which propagates up to the surface, and it meets the surface normal to the surface in almost every case. This has to do with the anisotropic properties of the foam.

So, you tend to split the foam this way, but that doesn’t throw anything off in this direction. And now people at Marshall Space Flight Center are doing very similar experiments and finding very similar results.

Now, it’s interesting that – I’m not going to be able to show you all of the beautiful results we have, but they really are quite nice. There’s another side to this question, though, and that is, you know, for STS-107, it was 81 seconds into launch when it – when the foam fell off the bipod ramp. And that’s really less than 30 seconds into aeroheating.

Is it possible for the heat to propagate through the foam and actually boil off the liquid nitrogen that might have condensed in that brief period of time? And there is someone named Marshall Joy at the Marshall Space Flight Center that has been doing very careful thermal analyses of the foam, and finds, in fact, that the thermal relaxation times are much too long for that.

So, the conclusion that I’ve reached, and that independently the people at Marshall Space Flight Center have reached, is that the process by which foam is ejected is undoubtedly a very complex one, involving more than just cryocondensation and ejection. And I dare say, in fact, that the sorts of experiments which I’ve done – which were actually done in my kitchen at home for about $100 – are the sort of things I think that we need to see more of done. Specifically, experiments to try to understand the physical mechanisms why the foam behaves the way it does.

And the last thing I had to show you, which is really the reason you’re not seeing anything, is that there are two –
Roger Tetrault was going to give this talk, but he very graciously gave up his spot so that I could make a fool out of myself this afternoon. And he had – he had a very nice set – these are being done at -.

Okay, give me number two. Okay, so this shows the heating. Now my time is all screwed up, but you guys are – you have nothing to do anyway.

So, here we have the bipod ramp, and they have three temperature sensors on here. This is on the outside of the foam. This is next to the super lightweight ablator – SLA, as it’s called – and then there’s one that’s against the aluminum.

And this is after launch, and you can see that, at launch, there’s a little tiny bit of heating on the surface, but it’s really only about 65 seconds into launch that you start seeing very dramatic heating. And the temperatures here are, of course, in degrees Fahrenheit, and it gets very hot.

But in fact, if you look at the temperatures that are down here, they don’t do anything for a long time. This is actually liquid hydrogen temperature and this starts rising because, in fact, the hydrogen level is dropping in the tank. But in fact, you have to go for really quite long periods of time before the SLA, which would suck up the liquid nitrogen if it condensed. In fact, in this experiment, it never got very hot.

Now, I’d seen this a long time ago, probably back in March, and I always thought it was real data, and it looked like it was here. You can see it plateaus out like that. It’s not real data at all. This is a simulation. And in fact, these measurements have actually never been made by NASA in the 20 some years of the program.

Can I have the next slide?

Here are some real measurements that were made by Marshall Joy at Marshall Space Flight Center. This is a piece of BX-250 glued to a metal plate, which is sitting at a temperature of 20 Kelvin. That’s 20 Celsius degrees above absolute zero. And this is in fact the temperature distribution of the eight sensors that he’s imbedded inside this foam. You can see, in fact, only when you get very close to the plate, you actually see the temperature dropping to the point 77 Kelvin, where you can actually condense nitrogen into a liquid. So, if you’re going to form liquid nitrogen, it can only be very, very close to the plate. Keep that in mind.

And now let’s see the next view graph.

Now he’s warmed the temperature. He didn’t warm it up to 600 degrees, but he warmed the temperature of the surface from roughly 300 up to 350 degrees. And here, this is the one that’s closest to the surface. It shows a very fast response.

But, if you look at the one – the thermal sensors which are very far down – this is launch here, and I don’t like his time scale very well – but in fact, you see essentially no heating until you’re well over 200 seconds beyond launch. So, again, it does not look like simply having heat propagating through the foam will result in the boiling of any liquid cryogen, which has been condensed down inside there.

Now, that doesn’t mean it can’t happen. In fact, they’re doing much more complex tests at Marshall Space Flight Center. I mean, one has to worry about vibrations, but they’re not worrying about vibrations in the tests that are going on right now.

I mean, it’s a very, very complicated environment that’s going on. You have vibrations from the surface of the foam because of aerovibrations, and you have vibrations coming – particularly because of the solid rocket boost – you know, I guess they’re gone by this time, but – it’s – there’s a lot of vibration.

ADMIRAL GEHMAN: One hundred twenty seconds.

DR. OSHEROFF: One hundred twenty seconds, so it’s still there.

If I could have the next view graph.

So, how does the foam fracture under hydrostatic pressure? And my answer as an experimental physicist is to do an experiment. So, if I could have the next slide.

This was actually done by my – I should give a plug to my graduate student, Jim Bumgardner (sp), who was actually working on NSF grant when I asked him to do this. But this took less than an afternoon.

So, he took a piece of foam, glued a metal plate to it, which I’ve cut off here, and we were very unsophisticated. This is a green food coloring dye in water, and we pressurized it up and it came through the surface as a very narrow slit. And so I said this is fascinating, and then I basically broke this thing in two, and you can see, in fact, the fracture itself was essentially a one dimensional thing. If you cut this in the other direction, you see essentially nothing, and I’ll show you that in a minute.

Could I have the next?

So then I got very sophisticated and I got the people in the machine shop to – on a Saturday afternoon – to make a very nice plate for me. The advantage here is in fact that this is a quarter-inch diameter hole, and I could glue on a piece of foam without actually disturbing the foam in any way. Next.

Okay, this is a piece of BX-250, and I put some epoxy on here. In the next slide you’ll see that it’s glued to that plate. I just wanted to point out, these are called the “knit lines” (sp). When they make the foam, they spray back and forth, waiting 40 seconds nominally between passes. And every time they do that, they get one of these lines. So the growth direction is always perpendicular to the knit lines. Next.

Okay, this is this thing all glued together and it’s all ready to go. And then I put it in the kitchen sink. I used a battery operated tire pump to generate the pressure. I don’t have pictures of that, but if I could have the next view graph.

Okay, here – here in fact – now I’m – we’re much more sophisticated. We are using red ink, actually Mont Blanc ink, if you must. Anyway, so it starts here and it comes up to the top. And so now we’re gonna section this thing. Next.

So this is what it looked like on the bottom. I actually cut down a little bit because it was kind of disturbed. Now, if we see the next one, you’ll see all the sections. So this is the bottom and then we’re starting to come up. You notice it gets larger and then eventually it gets smaller and, in fact, looks very similar to what we saw with the green food coloring dye, except now in fact, you know, we’re actually seeing it in a different plane. So if I could have the next one.

Now, here it is again, but I’ve actually used PhotoShop to increase the contrast. And if you look, you can see in fact that, while the knit lines are parallel in this direction, in fact, they’re curved in the other direction. I don’t know –
this almost looks like it came from a bipod ramp.

So, you notice in fact that it started right in the center of this place – of the plate, but in fact it broke through the surface way over here, and that’s entirely consistent with saying that it was propagating in the direction which is perpendicular to the knit lines. That is, parallel to the growth of the foam. And that is entirely understandable in terms of the mechanical properties of the foam. Next, please.

Okay, so – it’s interesting, by the way. It takes very little pressure for the foam to fracture, two to three atmospheres. And in fact, the strength of the foam itself is frequently, you know, not much more than this. The foam fractures in the plane and it fractures in the plane normal to the knit lines; thus, the fracture exists normal to its surface as it exits the foam and this alone cannot cause the ejection of foam.

And I think there – this is a very simple experiment. A very elegant answer with a really nice conclusion. And I think that it would be nice to see more of these sorts of things done by – I can imagine every high school student playing with foam. I think we have to send it off to all of the high schools.

Now, let me say that there’s a lot of caveats. The pressure that’s built up in the foam can operate in conjunction with something else. And for instance, if you get delamination of the foam from the coating which coats the aluminum surface, that in fact would give you perhaps a very large area of pressure, and that could definitely get – if that area, dimension of that area is large compared to the thickness of the foam, it would undoubtedly eject foam. So, in fact, it’s more complicated, but again, I think that it’s really time for us to develop – to do experiments that allow us to constrain models that we may develop as to how this foam is ejected.

Now I should say, and the last thing, of course, that the bipod ramps will be eliminated in future flights. However, there are other places on the external tank where there are very large pieces of BX-250 foam. And I think that even though the only place where we know that it’s ever come off is in fact the left-hand bipod ramp, in fact I think it behooves NASA to understand these processes better. And the reason is, in fact, that as the foam falls off and hits lets say the RCC panels, you have the danger that you will get a hole develop. And if that happens, you’ll see what happens in the next view graph. This is the movie, thanks to Roger Tetrault. Could I have the first movie please?

MS. LAURA BROWN: I don’t believe we’re gonna be able to show them.

DR. OSHEROFF: We don’t get to see the movies? That’s the best part.

MS. LAURA BROWN: I know it is, but they had technical problems.

DR. OSHEROFF: I’ve been deep-sixed. I protest.

Well then let me tell you what – these are experiments that are being done by NASA now. They take what’s called an arc jet. They use an arc, which is to say, you know, a lot of electricity, to heat up air to a temperature of 8,000 degrees Fahrenheit and they direct that at an aluminum plate that’s a tenth of an inch thick. That’s much thicker than the spar in the inside of the wing of the orbiter.

It takes less than 20 seconds for a hole to grow from one inch to six inches. It’s really very impressive. You cannot imagine the destructive power of the gasses that would flow in through that hole. It’s really pretty scary to see, and I’m sorry I can’t show it to you.

Thank you.

ADMIRAL GEHMAN: Thank you very much.

Okay, Dr. Logsdon.

DR. JOHN LOGSDON: Well, I’m told that there’s about a minute-and-a-half needed to switch from a Mac to a PC, so let me make some general remarks.

A number – oh, there they are. Anyway, I’ll make the general remarks anyway.

A number of your papers reported recently on the outline of our report. And it’s always dangerous to correct the press, but what you did was take a lot of questions we’re asking and turned them into conclusions. We don’t have conclusions yet. The reports were accurate enough that that version of the outline identified a whole range of topics that we’re looking at, but the outline’s very much a living document. It changed yesterday a bit, so -. That’s just a caveat that, don’t report our conclusions.

There was a column in one local newspaper three weeks ago that complimented us for our conscientious and thorough report. You know, we’re just starting to write. We’ll find out what we think when we try to write it down.

I though I had a slide before that. Is that really the first one? Let me run – somebody run the slides real quickly because that doesn’t look like the right order of things. Try that one first instead of the one in the middle.

Group four – which is not just me; Sally Ride and Scott Hubbard are making contributions. We have board members, and we have a group of consultants working with us – are looking at the broader management budget and organizational issues. I think I’ve said this at a press conference before. We have a whole long list of factors that we’re looking at that have provided the context within which the shuttle program has operated. And as we work our way through these factors, we can begin to report them at these various press briefings.

The one I chose to talk about today was budget because I knew that Marcia Smith was gonna talk about budget, and I think I have a little bit to add to what she had to say. So, if we go to the next one.

I think one of the things to realize is that NASA’s priority in the national scheme of things, compared to other places that federal government spends money, has not been very favorable. That, while the defense budget has increased, the discretionary budget has increased, and the non-defense budget has increased even more over the past decade, NASA’s budget, if you measure it in purchasing power and constant dollars, for most of the ‘90s actually shrunk, and just got a little bit above where it was at the start of the decade in the most recent fiscal year.

So, the environment in which NASA has operated has been one of virtually no absolute budget growth. And when you correct for inflation, as this does, for most of the time, shrinking budgets. And yet NASA hasn’t cut any programs and hasn’t closed any field centers. It’s been a very, very constrained budget environment.

Most of those decisions, and I’ll say it again, are decisions made in the process of formulating the President’s budget each year. And so it’s interactions between NASA and OMB, and then the policy and political people in the White House. That’s a process that none of us have very much visibility into, as compared to the congressional process, which is quite transparent.

And we on the Board have not insisted on getting the information on the NASA/White House interaction, because that’s an Executive Privilege area. It’s a lot broader issue than just NASA and the shuttle budget. And so we have not been using any OMB pay us back or similar data in our analysis. Let’s go to the next slide.

All right, when you’ve not – don’t have much budget growth in real terms, and you want to continue lots of programs and you have one big program called Space Station that keeps growing in cost, where do you cut the budget? And I would argue that the space shuttle budget, which is the largest single item in the NASA budget, as you saw in Marcia’s charts this morning, has taken a disproportionate share of the cuts in NASA’s budget, allowing NASA to do other things, some initiatives plus fund the station.

So you see the difference in this chart between the growth in the NASA budget overall, and the basically lack of growth in the shuttle budget. The shuttle has been, if you want, the cash cow to finance other parts of the agency. Next slide.

Here are the numbers. You see there are two components to the shuttle budget. One to actually operate the system, and the other to invest in the upgrades. And again, you heard Tom Young say this morning that the upgrades – several people said – Allen Li said the same thing. This ambivalence about how long you’re gonna need the shuttle, and therefore how much you invest in keeping it and improving it, I think shows that -.

In the ’94, ’95. Look at the difference between ’94, where there was a fairly significant upgrade budget, and then the next several years, when NASA said well, we’re probably gonna replace it with X-33 – with VentureStar – relatively quickly, so let’s not invest in it. And then a shift to saying, well, maybe not. Maybe it’s gonna fly to 2012 and we better start investing in it. And then in 1999, in terms of the policy decision, in fiscal 2000 in terms of budget, beginning another round of upgrades. Where the operations budget has been pretty constant over this time period, the variation has been really a willingness to invest in the system’s future. Let’s go to the next one.

And this is a NASA chart, which again shows you the shuttle budget over the past decade and projected for the future. The big cuts came before EDSPA (sp). They came from, first of all, taking advanced solid rocket motor out. But then the contractor work force went down from 21,000 to 17,500 between fiscal ’91 and fiscal ’94.

So that – the budget has been more or less level, certainly not growing very much, and is not projected to grow in the future. Just to give you a sense of the financial context within which the program is operating.

And I think that’s the last slide, right? Yep.