Columbia Accident Investigation Board
Home Board Charter Board Members News Photos Events Contact The CAIB
Columbia Accident Investigation Board Press Briefing
Monday, June 7, 2003

San Antonio, Texas

MR. SCOTT HUBBARD: We have found the smoking gun. We're here at Southwest Research Institute in San Antonio, Texas, where we have just conducted a test designed to show whether foam that fell off the external tank of the Columbia Orbiter could create a hole big enough to destroy the vehicle. The test that we conducted at 1:30 PM today demonstrates that this is, in fact, the most probable cause creating the breach that lead to the accident of the Columbia, the loss of crew and vehicle.

In order to do this test, we used facilities that allowed us to fire a block of foam that we believe approximates the size of the foam that fell off of the external tank at 82 seconds into the Columbia mission. We aimed for a velocity of about 500 miles an hour, around 775 feet per second. The test we conducted was at 777 feet per second. So this was, in fact, a valid test. It came extremely close to what we believe were the accident conditions.

We also fired the projectile, the foam block, to within about a quarter of an inch of the target that we established before the test. We had sixteen cameras looking at this, and when you look at a few of those, what you will see is really very interesting.

It shows the foam crashing into this re-enforced carbon panel, shows the foam moving across the panel, a crack appears at the top, and then at the bottom of the panel, as the foam is crossing, a rip begins to appear. This tears all the way across the panel and produces the hole. The hole is a – very ragged, about sixteen inches by sixteen inches, or about 256 square inches. There are a number of pieces that are inside the wing leading edge, as well as pieces that fell outside. We're going to show you one of those pieces now.

This is one of the first pieces that have been picked up. It's approximately about eleven inches long and something like three to four inches wide.

QUESTION: Scott, could you hold that up for the camera, please? Thank you.

MR. HUBBARD: This is a piece that was torn out by the foam traveling at approximately 500 miles an hour. There are many more pieces that are inside. We've yet to examine those. One of the mysteries of this investigation has been what did the radar see when it saw an object leaving the Orbiter on the second day of its mission.

One of the conjectures was that that could be a part of what's known as a P-seal (?) or it could be a piece of the re-enforced carbon panel of about 100 square inches. I would say that between the size of this hole, which is 256 square inches and this piece, the first one we picked up, which is about 40 square inches, we've also established that that piece that came off in the second day, could very likely be part of the carbon panel itself.

Who's got my notes here?

The angle of the impact was about 22 degrees that approximates what we believe was the angle at which the foam block in the accident actually impacted the Orbiter.

I want to re-emphasize that what we tested today was not the worst case. The first test that we did on the re-enforced carbon panel six showed a five and a half inch crack. This test, which was done under what we believe was the best set of average conditions approximating the accident, produced a hole, sixteen inches by sixteen inches. As a result of getting the data at – from both of these tests, I believe that we have found the smoking gun. I believe that we have established that the foam block that fell off of the external tank was, in fact, the most probable cause, the direct cause of the Columbia accident.

So, now I'd be happy to take some of your questions.

QUESTION: Did they not know –.

QUESTION: Didn't the test fail because, had you had a hole that size on liftoff, wouldn't you have known it before you reached orbit?

QUESTION: Yes, yes –.

QUESTION: – Won't you –.

QUESTION: – Shouldn't you have known it?

QUESTION: – 2.1 mach, air still coming in, further destruction, drag, other things that we didn't see in the real flight?

MR. HUBBARD: I don't know that – the onboard data that was available to the crew did not monitor things like the condition of the wing leading edge. This –.

QUESTION: – Any the tracking cameras?

MR. HUBBARD: The tracking cameras – the one –.

QUESTION: – (Inaudible) from the tracking camera. Isn't there supposed to be the tracking camera images? It wouldn't see that?

MR. HUBBARD: This hole was just at the limit of the resolution of the camera that had the view of the underside. The video camera, the videotape camera, that was working properly, had a resolution, after a lot of enhancement, of about two feet by one foot so this, which is sixteen inches by sixteen inches, is just at the limit of that resolution.

Although, I point – I want to point out that you're looking at black on black. You're looking at the underside of the dark panel with a dark hole in it. So it wasn't clear that there wasn't enough contrast.

QUESTION: Scott –.

QUESTION: So this could have happened on liftoff? A hole this big?

MR. HUBBARD: A hole this big could have happened on liftoff.

QUESTION: Wouldn't it have come down before Texas? If – you know, wouldn't the shuttle have broken apart, perhaps over California or New Mexico? I mean, how could it have made it all the way to Texas with a hole that big?

MR. HUBBARD: The range of calculations that have been done have showed that a hole that is about ten inches across would correspond to the sorts of measurements that were made of the increases in temperature and so forth that we learned after the fact.

I want to point out that the variability in this re-enforced carbon material from panel to panel is huge. Two pieces, straight from the factory, will vary in their breaking strength by 70 percent. So producing a hole sixteen by sixteen inches, my view, with that kind of variability is very close to a hole ten inches in diameter. In other words, I'm saying that the errors involved in making these kinds of measurements with a material that varies so substantially means that a ten inch hole, a one foot hole, a one and a half foot hole are all roughly the same in terms of this test we just conducted –.

QUESTION: – You might not get the same results if you did the exact same thing again –.

MR. HUBBARD: – Exactly same test again, we might have gotten a different –. Gina?

QUESTION: Scott, could DOD assets have seen a hole like this in space? I mean, you guys have the inside information on this, we don't. Is this something that would have been visible, do you know?

QUESTION: Or picked it up when it happened?

MR. HUBBARD: I don't think I should comment on what our national assets are capable of doing. I just don't think –.

QUESTION: (Inaudible). Can you say that they should have taken a look in space, without commenting on what the resolution of the assets?

MR. HUBBARD: It probably would have been very useful to take a look at something and see if you could see this. Again, I'll point out the contrast is black on black, and so you'll have to be careful, be sure you've got the right illumination angle and all of that.

QUESTION: Any other broken pieces?

QUESTION: Would this have any impact on the aerodynamics of the shuttle going up and could that be part of the data that showed the max – through thermal data?

MR. HUBBARD: This hole, although it's very dramatic, if it occurred like this, and remember, we're just trying to simulate the conditions, it's probably not exact. But a hole like this at 82 seconds would still be at a time when you've got millions pounds of thrust. And so relative to the weight of the Orbiter, the weight of the whole system, you haven't even released the external tank yet. When you come back in that's when the aerodynamics play a much bigger role.

QUESTION: What if – if, if the crew was not able to detect it, were they back at NASA able to detect anything, again, their telemetry, anything that was telling them that something had hit the wing like – of this magnitude?

MR. HUBBARD: The sensor data that we were able to read out after the fact, the so-called MDS data that somebody had talked about, was stored on board. This was not read out either on the ground or by the crew. The best available data they had at the time of the accident were the two cameras, and as I said, this hole is at about the limit of one of the cameras.

QUESTION: (Inaudible) that you saw?

MR. HUBBARD: In a quick look, I haven't seen anything. But the full report on this of what happened to the attachment fittings, the hardware, all that sort of stuff, will take a day or two because the panel needs to be removed and everything very carefully inspected.

QUESTION: At a meeting, here a month ago, the Accident Investigation Board had made several recommendations including: inspections and repairs, the ability to do that, also improvements in the thermal protection system, and then foam shedding. Does the lack of those, you think, contribute to – to what happened to Columbia?

MR. HUBBARD: That's a complicated question. Let me characterize my answer this way. The board had a briefing from a retired astronaut, who was a specialist in space walks. And he said that he could imagine devising a rig to go out and take a look, and he could put the face of his buddy this close. Well, obviously, you can see this hole from very far away. I think a space walk - there's no doubt you would have seen this kind of damage.

QUESTION: (Inaudible). Could they have fixed this? Or patched it then? (Inaudible).

MR. HUBBARD: I don't know personally. The experts in thermal protection systems are working hard to try to see what type of damage can be repaired.

As you recall, the board commissioned a study to see if a rescue mission was feasible. We included that with a lot of ÒifsÓ it was feasible, which of those two options you would use, that's just speculation.

QUESTION: (Inaudible).

MR. HUBBARD: I was surprised. I was very surprised.

QUESTION: Did you inform the Admiral? Did you talk to him?

MR. HUBBARD: I talked to Admiral Gehman. I've told him we've – he said you must be grinning from ear to ear. And I said that I felt that the last six months of investment by the board, all the hundreds of people supporting the board, and all the thousands of other people who've collected debris and so forth, are now, I feel, really have come to fruition in the sense of being able to show in a much more direct way what brought down the Orbiter.

QUESTION: I think it's very sobering, very stabbing – very stunning to see at the same time.

MR. HUBBARD: It is. You know, as a physicist conducting a test, I feel gratified that, after months of work, we're able to demonstrate this connection between the foam and the damage, but I know that it was the source of tragedy. And so that makes me feel very sad. This whole six months, we've constantly been reminded by pictures of the seven lost astronauts - what this all means when you stand back and look at the human dimension.

QUESTION: What can or should we learn from this?

MR. HUBBARD: I think our recommendations, the first ones that we've released, say that we need to have better imagery, that the shuttle program should have the capability to do on orbit inspection and repair, that we need to utilize national assets to see if there's been any damage. We need to realize that this is a vehicle that needs to be looked at very carefully because each flight, since there's only been 113 flights, is a special flight. Aircraft, even experimental aircraft, often go through thousands of flights before they are determined to be operational. This vehicle has only had 113 flights.

QUESTION: Why were you surprised?

MR. HUBBARD: I was surprised because the data and the models weren't coming together. As a physicist, you like to see the model and the prediction, you like to see the experiment match that.

And before this test today, I was willing to see anything from a small crack to something completely unexpected, and what we got was something completely unexpected. They –.

QUESTION: – (Inaudible) is there any possibility that we could do some sort of a patch on orbit or are you basically saying that we would have to do a rescue mission?

MR. HUBBARD: As I said, the experts are working to try to determine whether you can do on orbit repairs of the re-enforced carbon. It is the highest temperature area of the Orbiter, so finding materials that would patch that are very difficult. I don't know personally of any way you could patch some hole this big, but that doesn't mean that some expert might not come up with –.

QUESTION: – Given that does that –?

QUESTION: – Does this – does this open the door to the possibility of re-designing the RCC?

MR. HUBBARD: The RCC has proven to be a very tough material. I think the first step is understanding the RCC panels much better and understanding what kind of damage threshold there are. Where is the dividing point between an impact that creates this kind of damage and an impact that you might be able to return with? That database doesn't exist, and I think it's one of the first things that the shuttle program is going to go to work on.

QUESTION: This may be so severe, right, that it could actually delay return to space flight.

MR. HUBBARD: I don't know. That's a good question for the – for the shuttle program. I think what our goal was, the board's goal was, was to connect the dots between the foam shedding event and the proximate or the direct cause of the accident. And that's what this whole test program has been about. And I think today, we've made that connection.

QUESTION: Did the repair – did the repair recommendation though take into account something of this size and dramatics?

MR. HUBBARD: Well, we know that almost surely there was a breach on the order of ten inches in diameter. Here we've got one sixteen, so that's in the same ballpark in my book. So – oh, sorry, we have, in essence, recommended that large breaches be considered. Now, whether the thermal protection experts can do anything with that or not is up to them.

QUESTION: You guys all out of questions?

QUESTION: No, give me a minute.

QUESTION: What was – we heard a lot of our own grasps over here on the press side. What was being said under the tent on your side? What was – did you utter – find yourself uttering anything at the moment that thing –?

QUESTION: – Can you answer that question?

MR. HUBBARD: This is going out live, you know.


MR. HUBBARD: No, no. I – I was – it was in here. It was, ÒUh!Ó like that. You know, it was a visceral reaction. You know, it was shortly followed by, ÒOh my God!Ó It was – it is so dramatic, but yet if you stand back a few moments and say this – this is in the same ballpark as the kind of hole that we believe must have been there when the Orbiter entered the atmosphere. So, you know, I felt surprised at how it appeared, such a dramatic punch through. But it is the kind of damage, the type of damage that must have occurred to bring down the Orbiter.

QUESTION: Scott, you said that the data and the experiment were not coming together. What was lacking in the data that you now think that could have predicted this?

MR. HUBBARD: No one has ever produced a model of how re-enforced carbon would respond to these types of impacts. Some work, some good work, was done in the modeling the impact of the – of various things, including foam on the tiles. But the re-enforced carbon panels, except for testing the impact of things that might be orbital debris, ice and the like, no one had ever tested extensively the – or modeled the panel. There was, many years ago, a set of tests done to see if a static load, this is a dynamic load – this is, you know, a moving slug, but somebody – it had been tested to see if you just put pressure, more and more pressure could you crack this material. But that, to the best of my knowledge, was never followed up on by a modeling effort.

QUESTION: Did this move – in your mind, did this move beyond the most probable cause to the cause? Do you really need that probable any more when you look at this?

MR. HUBBARD: I think it's the direct cause. I think foam hitting the wing leading edge of the Orbiter at 500 miles an hour is the direct cause.

Now the board and Admiral Gehman have been clear that much of the report is going to deal with the other types of causes, contributing causes and other elements of the Orbiter program over the last twenty years. But in terms of the direct technical cause, we think this is it.

QUESTION: This is the real cause –.

QUESTION: – You can't rule out –.

QUESTION: – The direct cause –.

QUESTION: – You can't rule out orbital debris and whatnot though, right?

MR. HUBBARD: You can never rule that out. It remains an unknown. And what I'm saying is I've got a – and I think the board has a direct connection between foam shedding and creating a hole that's the same order of magnitude as what must have been there when Columbia came home on February 1st.

QUESTION: Do we still care about that float-away object on day two any more now from that? I mean, now that we've got a hole. Does the float away day two object matter any more?

MR. HUBBARD: I think in terms of telling a complete story - the second day float away object needs to be compared with what we've learned here. But they'd already done, the Department of Defense and other labs, had done a pretty exhaustive study of the candidates. And one of the candidates was a hundred square inches, that's about this big, piece of re-enforced carbon. You can see here a hole, and we have a piece over here that's in that range. And so, I think that this begins to tie that part of the story in a more direct way.

QUESTION: (Inaudible) side?

MR. HUBBARD: I'm sorry.

QUESTION: Float away object –?


QUESTION: – Was a piece?

MR. HUBBARD: Based on this test, the float away object would have been a piece of re-enforced carbon, you know, on the order of this size of this piece of paper. And we got a piece here that's half that size.

QUESTION: It would have been stuck somehow inside the hull.

MR. HUBBARD: My staff tells me there's two bigger pieces inside. So, if you can imagine a piece that's been knocked into this cavity, and then you begin to do some orbital maneuvers, and let's say that rather than being blasted through it's hanging, just hanging by a thread, and then it drifts away. So I think this data gives us a better understanding of what that object may have been.

QUESTION: Well, then what do you think that that day two object was a result of this or was it something else?

MR. HUBBARD: Let me answer her question, and then I'll get to you, Gina.

QUESTION: What do you think the difference was today? Do you think it was that angle at which it was fired, so that more of that edge or the leading edge of the foam hit?

MR. HUBBARD: I think that the adjustments that we made to try to localize the force, the way we think it happened in the accident, probably made a big difference.

But I also have to add that these panels vary, and we've used aged panels all through this. Panels that have a flight history and this had twenty-seven flights. Columbia was on its 28th flight, and we know that these panels age at different rates. They lose as the result of being heated up to 3,000 degrees and cooled off and heated up and cooled off. They lose mass and so, this panel may have lost it slightly differently. But in the – in the big picture, what we've established is the connection in the foam shedding and breaking a hole that's the size that almost surely had to be there.

QUESTION: Do you have any estimates on the most – the amount of force that it was hit with?

MR. HUBBARD: We don't have an estimate. We have to get the read outs from the string gauges. Previous measurements show impact forces of about 2,000 – 2,200 pounds. So a ton of force was imparted to this panel.


Back to Event
Privacy Policy Disclaimer Accessibility