Transcript

COLUMBIA ACCIDENT INVESTIGATION BOARD
NEWS CONFERENCE
JOHNSON SPACE CENTER
FEBRUARY 25, 2003

Adm. Harold Gehman, USN
Brig. Gen. Duane W. Deal, Commander, 21st Space Wing, USAF
Maj. Gen. Kenneth W. Hess, Commander, HQ USAF Chief of Safety
Mr. G. Scott Hubbard, Director, NASA Ames Research Center

STAFF: Hi. I'd like to welcome you all, with the members of the Columbia Accident Investigation Board. We'd like to let you know that we have a new website up and running today. It's www.caib.us. And with that, I'd like to turn it over to Admiral Hal Gehman.

GEHMAN: Good afternoon. As usual, I'm joined by three of my fellow board members, who I will introduce to you. I'll make a few opening comments. They will make a few opening comments.

And then we'll be delighted to take your questions and questions from any of the other centers. To my right is Brigadier General Duane Deal, U.S. Air Force, who is on the group that's looking at material and management issues. To his right is Major General Ken Hess, U.S. Air Force, the group that's looking at operations and flight issues. And on his right is Mr. Scott Hubbard, who is on the group looking at technical and engineering evaluation issues.

This is really your opportunity to dialogue with the board. As you know from my previous statements, we don't save up news to be released on Tuesdays. But I do have some announcements to make as to what we've been doing this week and some things like that, as will my colleagues.

Each one of them will make a statement as to what activities they have been doing and where they are. But in general, we don't save up news to be released at this conference.

Our commitment to finding out what happened to the Columbia and her crew remains just as firm. The energy that we are pursuing has increased. We are now more active. We've got more things going on than we did before, far more activities going on.

We have now got investigators out in the field. And as I will tell you a little bit later about our activities, I think you'll see that the pace has picked up considerably.

But our commitment and our resolve to get to the bottom of this is undiminished. I'd like to give public thanks to the many agencies that are assisting us. In particular, I'd like to mention the NTSB and the Department of Defense, who have been very helpful to us.

I would also like to mention state agencies, particularly in the states of Texas, Louisiana, New Mexico, Utah, Nevada, California. It's these state agencies and local police, firefighters, sheriffs that have really borne the burden of all the debris pickup and are really carrying a great deal of the load. And we're very much grateful to them.

Debris continues to come in. We'll get some numbers here for you. We have some numbers to pass out.

Debris continues to be important to us. We are now beginning to learn some things. That's probably too strong. But we are now beginning to see some interesting trends and evidence in the debris. So debris continues to be very important to us.

The board is now working simultaneously in a number of areas as we speak. We are working simultaneously at JSC, KSC--Kennedy Space Center--Marshall, Palmdale--the plant where we used to overhaul shuttles and make them, and the Thiokol plant in Utah.

We now have our own budget, of which I am the budgeting officer. It's controlled by me. As Laura Brown indicated, we now have our email address. Our 1-800 number and websites are up and running. And I believe, at the end of our broadcast here, those will be put up on the screen.

Our first public hearing will take place on Thursday, March the 6th at the University of Houston, Clear Lake. Details can be found on our website.

We are, as I indicated, we are finding some interesting things from debris. Let me show you one here. And I think that it may appear on our television screens as we do that.

This is the same--these two are two pictures of the same piece of debris. This is the underside of a piece of tile that was found about 30 miles west of Fort Worth in the town of Powell, Texas in Navarro County.

This is still being analyzed. But what I would point out to you, this is the underside of the tile. This is the tile which is near the metal surface. What appears to be at the ends here--here and here--are essentially worn away, probably by hot gases.

I'm now going to show you the top of the tile. This is the surface of a tile which faces the heat. And while it's being analyzed right now, including what those little orange specks are, I am told that this is not typical of a re-entry tile. This is very unusual.

One of the riddles that we have to sort out is whether or not this damage was done while the tile was still attached to the orbiter or whether this damage was done after the breakup and that's what it looks like when you try and re-enter the atmosphere at 15,000 miles an hour in a non-aerodynamic state.

So this is now what I like to call--we used to call this the westernmost piece of debris. We now have another piece of debris, which has been found farther west of this, near the town of Littlefield, Texas. It is a fragment of a piece of tile. And I don't have a picture of that here.

The tile at Littlefield, Texas that we don't have in the system yet appears to be a tile from the upper surface of the wing, near what we call the glove area, which is near where the wing attaches to the fuselage. So it appears to be from that section on the top part of the wing, forward, near the body of the orbiter itself.

So the fact that that tile was found significantly farther west is interesting. We'll have to get the tile into analysis to see what it is.

So debris is important to us. And we thank the public for their energy in going out and picking up debris. We reaffirm the instructions to the public to call the number, get some help. Some of this stuff may be hazardous. It's very important to us.

For my own plans, I'm traveling to Washington, D.C. tonight. And I will spend all day tomorrow in meetings with senators and congressmen who are in the oversight committees that exercise oversight over NASA. And I will be explaining to them what we're doing and how it goes.

So those are my opening comments. I would now like to turn the microphone over to Brigadier General Duane Deal. And he has some opening comments.

DEAL: Good afternoon. I'd like to echo the admiral. Thank you for the opportunity to get with you today and explain some of the things that we are up to.

To reiterate what our panel does, we call ourselves kind of the "3-M Panel"--the Materiel, Maintenance and Management Panel. I'll give you a little bit more granularity into that. In the materiel and maintenance aspect, we're looking at the things that you would probably expect us to: design modifications of the shuttle; the entire shuttle system refurbishment procedures; maintenance processing; the overhaul; quality assurance; inspections; the vendor issue; aging platforms; and also, the entire shuttle integration.

Then under the management aspect of it, which all the groups will be contributing to, we're looking at: the boards and the committees that work the various aspects of the shuttle; the internal processes that lead up to launch and also recover from launches; safety programs; the decisions that have been made and the decision levels at which those decisions were made; the oversight and, what some of you are familiar with, the end sight programs; budgetary programs; contract management; and things such as that on the management side.

And on our team, we have Major General John Barry from Air Force Materiel Command, Rear Admiral Steve Turcotte, from the Naval Safety Center, plus a great staff team. And as the admiral said, we're bringing vast experience in these past investigations. And we're committed to do our part to get to the bottom of this terrible accident.

As the admiral alluded to, we're going from being simply wide now to being a deep and wide type of analysis. We're in what we affectionately call the "show me the numbers mode."

We're entering this with no assumptions. We're not assuming anything is broken. And we're not assuming everything is solid, 100 percent. And our duty is to examine everything objectively.

What we have been up to is we're gathering and reviewing a number of past reports for the actions that were taken and also for the follow-ups to those actions, some things that you are familiar with and many of you have reported on, things such as: the foam incidents on Flights 732, 50 and 112; other past reports, such as the Aerospace Safety Advisory Panel reports, the Shuttle Independent Assessment Team reports; maintenance actions that have occurred, not just on Columbia, but also throughout the fleet; the non-destructive inspection type of processes we have; and something some of you are familiar with, the government mandatory inspection points. Those are the items that the government is still required to inspect in quality check of what the contractor has done.

We have started the process of conducting interviews. And we kind of lump that into three different categories.

We bring in outside speakers, some of which are provocative and some of which have appeared in your media. We bring in factual background experts that are independent experts. An example of one of this that we just had last week is a Dr. Jean Gebman of the RAND Corporation, who brought some perspectives on his extensive study of aging aircraft.

We're also starting the process of interviewing witnesses that were involved with the Columbia program.

So in addition to finding the cause, another key aspect that we're all going to be looking at is not just what the cause was, but also what we have to eliminate from what happened, what we do not believe was an impact. And that's going to include the entire program again.

And I could give you an example of looking at the payload itself of the Spacehab modules. There was experimental equipment that we have to rule out that included a huge range that many of you are familiar with, from an ergometer to student experiments that were on board, to crystal growth experiments. Not that we assume that that's a part of it, but we have to eliminate them.

We're on the road this week, as the admiral mentioned. The other two members of our team are on the road. One was at Thiokol in Utah yesterday and today. Another was at Palmdale yesterday and today.

I'll leave and join them tonight at Kennedy Space Center. And then we'll split again, where one of us will go off to Marshall Space Flight Center and another of us to the Michoud assembly facility in New Orleans.

And again, the bottom line is that we are looking at everything, deliberate and thorough, and looking to do it right, for the causes and to prevent the recurrence.

And I'll leave you with two final perspectives. The first is to reiterate what you all already know. The shuttle is a research and development vehicle. Columbia was on its 28th flight and the shuttle system, on its 113th flight.

It's not an airliner or a production operations aircraft; otherwise, it would have had thousands of flights before it went operational. And it's not a Mercury or a Gemini or Apollo that were one-use type of systems.

And so it's very much an R&D environment, which leads us to why we brought in someone like Dr. Gebman, because we are examining it as an aging spacecraft in an R&D environment.

And the second and final point I'll leave you with is the complexity of what we all face. Admiral Gehman is leading a huge task force, doing all this. NASA has what I would term the mother of all fault trees, its tremendous fault trees that we're examining and looking at, in addition to our own scenarios.

I took 30 seconds and measured on of the fault trees, which was three feet by 15 feet long. And that only went down 15 different levels. And it can go another 50-plus levels down below it.

So you can imagine the steps taken to design, produce and maintain just one component that's a part of that fault tree, which also includes the designing, the producing and maintaining the equipment with which it's manufactured. So you can appreciate the task that we all have ahead of us.

So we're pressing forward, as smartly as possible. No longer just receiving data but, as the admiral mentioned, analyzing it with our experts and generating our own evaluations.

And I thank you for the opportunity to convey where we're headed. We have much to do. But we wanted you to have a flavor of what we're up to.

HESS: Sir. Good afternoon. I'd like to echo the comments not only of the admiral, but also of General Deal and the fact that, in order for the Accident Board to complete its work, we're going to have to look at a lot of processes that will take us into places that are not currently being covered by the news media at all.

And when we do that, like General Deal says, it's not because we actually think that there is going to be a cause there. But to do a complete investigation, we have to look at those programs because there may be ideas and concepts that would be worthy of note back to NASA that have nothing to do with approximate cause of the mishap.

Now our portion of the board is going to take a look at things like training, mission operations and life sciences. And in the training aspect, as General Deal said, there is a decision tree that is very, very detailed that we have to work our way through to be able to get there.

Not only do we have to look at the process for how the crew was selected for this particular mission, but also how the crew was trained, not only to do their individual jobs with the orbiter itself on orbit and during ascent and re-entry, but also how they were trained to be able to conduct the experiments that were in the Spacehab that General Deal talked about. We have to look at how the crew was trained that ran the flight from the ground, both the ascent and re-entry crews, as well as the crews who monitored the performance on orbit.

Each of those was trained by an expert team that we have to look at, to look at their qualifications and look at the certifications of the individual members who were on, so we can assess the overall training for the mission itself.

And so, in the first part of our business here, we're going to take a look at the training operations and do some of the initial medical clearance stuff that has to do with the crews and their health and their readiness to do their mission. And once we have finished that background look, we'll transition into more of an operational look, starting with the flight readiness review, and take the flight readiness review into the lock sequence and then look at performance on orbit to measure it against the plan and see how that went.

In all, we think that the background processes will probably take us another 2.5 weeks to get through. We have just hired an expert in the training processes that will help us get through and understand the different words and vernacular that are being used here.

We also have a life sciences specialist that was in the public release this morning, Dr. Jim Bagian, who is helping us in the life sciences area. And so we're making good progress. And our interviews are starting this week and taking a look at those people who did such things as the payload integration, looking at the safety aspects of the initial operations inside the shuttle before launch.

Now our team travel is expanding a little bit. Steve Wallace, whom you met last week, is on travel to Kennedy this week. And he's going to endeavor to tease out the information involved with the launch sequence, the ice control or ice teams that take a look for ice as a potential debris problem on launch, as well as the last waiver processes at the go-for-launch decision.

And to augment our staff, we have added two people. We have added an NTSB investigator this week. And we have another FAA accident investigator coming in.

Both who have tremendous background in investigating some of the major aircraft mishaps that have happened over the last several years, but also the FAA investigator is an expert in air traffic control and radar. Because we'll have to get into the range and range safety issues to ensure that they were in accordance to the plan as well.

So kind of to summarize, we're just now getting to the point where we have enough understanding to begin digging very, very deeply into a lot of the key processes. And we're adding more process each day, as our understanding grows.

And the key members interviews with NASA and members of the USA contractor are just beginning toward the end of this week for us.

HUBBARD: So I can fit with the rest of the brass up here. The engineering and technical analysis group--which includes Jim Hallock, Roger Tetrault, Sheila Widnall and myself--is applying a systems approach to our work.

It's well known that accidents in complex systems often involve a chain of events. It's not often one single thing that caused the whole accident.

So this complex series of events needs to be evaluated together. For example, eventually, we would like to see if you can couple the external tank shedding event with TPS sensitivities with the calculations that are being done by the aerodynamics folks.

So what we're doing right now is trying to bound the event in the orbiter. We're trying to go from the town to the ballpark to the seat. Let me give you three examples of data that the NASA Mishap Response Team has put out there and point out the challenges of being able to bound this event.

They did a hypothetical study of a 20-square inch breach in the wing, four by five. Near the main landing gear door of the seal, it seems to account for the temperature rise. But this is a first cut, with an initial set of assumptions.

Figuring out what might happen to a plume going inside of a wing is a very complex task. It's a good first step. But we don't have an answer there yet.

Secondly, the sensor data that we have all looked at--the rise in temperature, the sensors going off line--seem to be consistent with wires going through the wheel well being severed. But there is a sensor near the front of the spacecraft that also had an anomaly.

How do you account for that? Again, it's a little early to conclude that we have a complete story.

And finally, some initial aerodynamic analysis seems to indicate there was some disturbance going on in the vicinity of the left main landing gear wheel well. But eight different calculations all gave somewhat different answers.

So the story I'm trying to communicate to you is that it's early yet to draw conclusions that we have boxed this story in, that we really have bounded the event. We're making good progress. But we need to validate the data interpretations with analysis of debris, experimental tests and, in selected cases, with independent calculations and perhaps some independent tests.

A couple of current data findings that are relevant. First of all, the downlink telemetry during ascent and on orbit--this is the telemetry now, not the visual observations that were made--all appear to be in family nominal. As far as we can tell, everything in ascent and on orbit, from the telemetry, looked like the mission was fine.

Subsequently, we have looked at the 32 seconds of data. And we have been able to analyze the MRT, as provided to us data, in which we look at the first five seconds and the last two seconds. The first five seconds, I think you know about. It shows that the orbiter was running normally, except for the temperature rises and the other sensor outages that have been noted.

The control jets were firing. The attitude of the vehicle was stabilized. The vehicle was under control. And the auxiliary power units--the so-called APUs--were running.

The last two seconds show that the APUs were still running, but the hydraulic fluids are now at zero. And the hydraulic fluid reservoir is at zero percent for all three units.

The conclusion seems to be--and again, we have to be very careful as we move on this--is that somewhere in the 25 seconds of unrecovered data, there was some sort of a failure that affected all three hydraulic units. Another piece of data in this puzzle that we're working on.

Now our group has plans for some next steps. We are initiating independent expert input. We had a presentation yesterday by Professor Elisabeth Pate-Cornell, a Stanford colleague of mine, who was the principal scientist doing studies on shuttle tiles, on the risk analysis on what was known through the first 50 flights or so. She gave us some excellent ideas for things we might do as a follow-up analysis.

We're developing themes in our group, which will help to test the NASA fault tree. We're going to follow the sensor data, follow the external tank insulation strike, follow the thermal protection system--its history, its sensitivity. And we're going to follow the debris.

Roger Tetrault will be down at KSC starting tomorrow with two experts--one in thermal protection systems, one in entry dynamics. And also, beginning tomorrow, the board will have an on-site resident there to help us review the debris and begin to look at what we might want to do or what might make sense, how we interpret things.

So to conclude with my part, the data and the twisted metal are speaking to us. But we're just developing the ears to hear.

GEHMAN: Thank you very much. And we will be pleased to take questions, whomever Ms. Brown wants to recognize.

STAFF: I'll take one question from each person here. And then we're going to move on to the other centers and take questions from them. And then we can come back here and do some follow-up.

QUESTION: So on the photographs of the tiles that have been found so far, on the first side of it, the inside--and you say it shows wear around the edges, from heat, you said. And I wonder if that means that whatever it is that makes this tile come off, that that indicates an inside source of the heat that makes the tile separate.

And second part of the question is on the flip side of the tile, you see those orange specks. And what are the various things that might cause orange specks?

GEHMAN: The answer to your question is it's premature to assign a cause to that eroding channeling on the underside. We're going to look very hard, through all kinds of different kinds of analysis, at what might have caused that.

Keeping in mind that that may have caused post-breakup. In other words, that just may have been what it looked like when it was going 12,000 miles an hour, no longer on the shuttle.

On the other hand, it might be a clue as to what caused the breakup. That's why we were anxious to find things that came off the shuttle early.

We also--the serial number on that tile was obliterated, regretfully. So we can't tell exactly where it came from. But we can tell, from its shape, thickness, makeup, dimensions, approximately where it came from. And NASA is working on that right now.

Scott, do you want to follow up with any of the details of what we know about that tile?

HUBBARD: I think the fairest thing to say is that we're going to get a first good look by some thermal protection system experts later this week. They will be down there with Roger Tetrault. And at least one of the fellows--Howard Goldstein, I believe--has significant--he helped develop some of these tiles initially. And I think he can look at it and begin to give us some ideas of what we ought to go further and do.

If the tile was hit laterally and sheared off down to what's called the densified layer--this is the layer at the bottom of the tile that helps it stick to the materials underneath--then what you see in some of these places might be like reheating styrofoam, if you will. It tends to slump.

But it is too early to say that this is an explanation. I think, stay tuned for the expert analysis.

HUBBARD: The orange specks, I don't want to speculate on. It could be a result of landing out in a field somewhere.

QUESTION: A real quickie and then my question. Which wing is the second tile from? You didn't really say right or left.

And then my other question is, in the 32 seconds of analysis, how do you account for that big gap of information in between? What's your theory on why we've got five seconds, nothing and then two seconds?

GEHMAN: We believe that the tile found in Littlefield--the fragment of tile found in Littlefield, Texas, which is the westernmost piece of debris, is from the left wing. That's preliminary, because it's a fragment of tile.

HUBBARD: I think you know that the way the data comes in the control room, there are voting procedures. There are multiple lines of signal. And if you don't get a good signal from the signal path, then it's not recorded. It's not sent on to the control room.

During the period of atmospheric entry, it is not uncommon to have signal dropout. This has to do with the positioning of the antenna, as well as the receiving assets, like TDRS, you know, the tracking and relay satellite and ground assets.

So the fact that there was some dropout was not unexpected. It is fortunate that we have been able to recover what we have. The actual signals themselves in that period, at least what I've looked at, it's barely visible at all. It may be that that never gets recovered.

QUESTION: How much of the left wing have you actually recovered? And what does that tell you, what you have recovered from the left wing?

GEHMAN: I believe that the number of debris that we have recovered is 8,110 pieces total, of which 5,297 have been identified. That's total.

Once again, as I mentioned last week, probably weight is more significant than numbers because some of them are little chips. We are over 10 percent of the weight of the vehicle.

And if you were to look at the floor and look at the left wing, it's not impressive. What I mean is, not a lot of the area is covered. So a small part of the left wing has been recovered. A small fraction of the left wing has been recovered.

GEHMAN: Not yet. Not yet. But debris still comes in. And as you can see from this tile, we'll learn something from this tile. And we'll learn something from the next one too.

QUESTION: What do you know about the object, the mysterious object that was seen trailing the shuttle some 24 hours after it came into orbit? And also, what information do you get on this and other matters from DOD assets, please?

DEAL: Well, it's important to note that it wasn't seen 24 hours later. It was seen on a delog, an extensive delog of the tapes after the mishap. So . . .

DEAL: That's where you go back and you look at all the tapes of all the observations throughout the shuttle's 200-some odd, 50-some odd revolutions around the Earth. There were nearly 3,200 observations.

Now an observation does not mean you are tracking it very closely. That means that, okay, you know the shuttle is coming overhead. You get a radar track on it. And then it passes. So that's an observation.

What happened after the mishap is the people in the Space Command went back and they started doing what is called a delog. That's where you look at every single observation as fine tooth as you possibly can. And I can tell you, it's been the most laborious examination that's ever taken place in the history of Space Command, looking at every single one of those observations.

And then on the--I believe it was the 6th of February is when they had enough analysis to say, "Okay, something is there." And then further analysis, which has gone further now, is it appears to be a piece about .3 meters by .4 meters of undetermined composition. As a matter of fact, the Air Force Research Lab at Wright Patterson Air Force Base is going to be working with NASA to do some tests, based on radar reflectivity, of what type of material it could be.

So they followed this piece. And it was characteristic of something accompanying the shuttle. And then it started to separate from the shuttle. It started--it was initially semi-stable, in a slow rotation.

Then as the shuttle went away, it had no propellant and it didn't stay with the shuttle, making its maneuvers. And it did a characteristic re-entry of something of that size.

And on the 20th, it re-entered over the South Pacific. So that's what we know right now. More tests are ongoing to determine what it was. It does not have the characteristics of being anything dense.

It has the characteristics of being some type of lightweight piece of that size that I told you. So again, that was not seen by anyone on the second day. It was seen afterwards, as they went back and did the delog?

DEAL: DOD has provided information that has helped on the timeline that NASA has reconstructed. And they have been helping in numbers of ways, from providing additional analysis to having National Guard troops on the ground looking for debris.

I mean, I could go into a long list of different things DOD has provided. But as far as the timeline itself, they have provided whatever data that they have to NASA. And they have used it in building their timeline.

QUESTION: General Deal, you referred to a series of interviews that you have now begun. I'm wondering whether the order in which you are doing those interviews, the areas in which you're examining, is that essentially arbitrary, based on the availability of people? Or does it reflect the strength of your interest in one area of inquiry, such as aging aircraft?

DEAL: Initially, we're starting out, as we mentioned, wide. And we're starting to go deeper. We're starting at higher levels, with those that are easily available targets right here. I hesitate to use the word target--but interviewees.

And we'll work with them. And they are all giving us names of people. And we're building our interview list based upon that.

And it's also something we're sharing with the rest of the groups because we also have some common things that we need to ask of certain people so that, as we're asking one individual, they may have another member of their group in our interview at the same time.

We are starting to drill down, if that answers your question. It's going to take us to get a little bit smarter, and that's what we're doing on the road this week, with some of the systems themselves, so that we can then pull in some people and say, "We have been through this maintenance data. This is our conclusion." Can you comment on that?

QUESTION: You said that the debris was starting to tell you some interesting stories. I wonder if you can elaborate on that.

GEHMAN: It has not been reported to me that we have any evidence of space debris. But it's way too early.

The tiles that we have are pretty beat up. And it's not until we get enough of them to put them together in some kind of order that we're able to develop a picture of what happened.

So it's a bit too early. But my comments were meant to encourage the continued pickup of debris because they are beginning to talk to us now. Some things are beginning to emerge. But no answers are beginning to emerge.

You must understand that you get a piece of tile here and a piece of tile here, it might not be tiles. It might be metal. And then, pretty soon, we'll be able to connect the dots. But not yet.

Could you discuss what you know, where you're at in the whole concept of the Columbia's left wing being rougher than the other wing? And the boundary layer transition, laminar flow, turbulent flow--all of that, as it fits into your investigation?

HUBBARD: I'll start and we'll follow up. The aerodynamic analysis has just begun. And very fortunate that we've added to the board Sheila Widnall. Aerodynamic analysis is a specialty of hers. And she has taken this on as something she wants to look at personally.

As I mentioned in my opening remarks, there have been some calculations about something called "center of pressure." This has to do with whether or not the wing was responding in the manner in which you would expect. There is a whole series of discussions and calculations about this.

About eight different timelines and data entries were looked at. And they don't quite agree with each other. And this is not surprising at all because we're dealing with a regime of flow and turbulence here, which is very complex.

So I think on this one again, stay tuned. And we definitely, I think based on some of the comments by Professor Cornell, need to look at things like the roughness. Was Columbia any different than the other orbiters? What happened to it during the major orbital maintenance it went through?

HESS: Scott has pretty much echoed a good strategy for us to follow here. We think when we start looking at the operational period of the shuttle past the launch, we'll want to do some comparisons against other Columbia re-entries. And key to it is to take a look at when the shuttle actually went through its major maintenance overhauls to see if there are actually any changes in its entry performance coming back through there.

And we'll probably, at some time, have an opportunity to talk with the astronaut that had indicated that he felt that there was more roughness in Columbia. And we haven't done that yet. But it's coming.

GEHMAN: I might add that a review of many post-flight reports, many shuttle commanders report that each shuttle flies uniquely differently. The light ones, they say, are quite discernible from the heavy ones.

And several pilots, several commanders have reported this left wing roughness on Columbia. And others have said they never noticed it.

QUESTION: I wanted to know what the crew might have known during that last 32 seconds. Would they have gotten some master alarms in the cockpit so they would have been aware that the hydraulic system had failed, for instance?

And what kind of G forces would they have been feeling normally at that time? And might they have--the people who were in the mid-deck, would they have felt anything?

GEHMAN: The G forces, I'm told, as you may be aware, we have all flown simulator missions, including the re-entry. The G forces, I'm told by our astronaut instructors, who have multiple missions, are relatively light. Compared to ascent, the descent is not a vigorous ride.

The data that we have on such things as control surface operations, attitude of the craft, all appear to be operating normally. As far as we can tell, the computer was still flying the orbiter. And the yaw jets and elevons were acting to keep the orbiter in its proper attitude up until loss of signal, including the 32 seconds.

Therefore, we have no reason to believe that there were any unusual G forces involved. There is no sign that the pilot or aircraft commander took control. There is no sign of anything like that.

And of course, there is no--as of this time--there is no more voice communication in the 32 seconds. So the best we can tell is that, at loss of signal, the craft was operating properly. When I say operating properly, with the exception of these anomalies that have been reported.

But the craft attitude was right. It was on course. It was in its proper entry for a normal landing at Kennedy. And beyond that, I wouldn't want to speculate on what the crew may have been able to see or sense.

HESS: No, sir. I think that the G forces are going to be very, very early in the onset measured intense at this point.

HESS: Perhaps. Yeah. I mean, if they have gotten to the G-onset piece there, it would be.

STAFF: We're going to take two more questions and then move to the other centers. And we will come back and do follow up here after that.

QUESTION: A follow up for Doctor or General Hubbard there, please? On this data--and we're getting a lot of the data being dribbled out--when do we expect that this last 32 seconds, you'll have a report on what it is you have found, when you are satisfied you can't find any more?

And can you comment specifically on the report of the four RCS jets firing? Was that in the first five seconds? And also, the reports of RCS leak alarms in the last seconds?

GEHMAN: Let me address the question of the report. And then I'll let Scott talk about individual jets firing and things like that.

Since we don't know where we're going, I can't tell you where we are. At this time in the Challenger investigation, 20-some days into the Challenger investigation, they already knew what did it. And they were busy going around finding out whether or not NASA people had done all the things they were supposed to do to prevent it.

Since we don't know what happened, we are obliged to do a full court press on every one of these avenues--safety, flight crew performance, materiel, aerodynamics, thermodynamics, imagery reconstruction, debris reconstruction--because we don't know where the golden nugget is going to be. So I can't tell you when we will submit a report or anything like that, except to say that the board has talked about that, in general terms, and that we probably would follow the FAA model in which if we find a specific fault or flaw or cause, we will advise NASA of that so they can get started fixing it, at the same time, advising NASA that that's not the end of the story, where there's a lot more to go.

There may be multiple causes. And then, whether there is one or more direct causes, there are certainly going to be many, many contributing causes--safety, quality assurance, contractor performance, you know, et cetera, boards and committees, oversight, interdepartmental communications. I mean, I'm just making all this up, you understand?

So it's pretty hard to--it's hard for us to say how this is going to come out at the end. I wouldn't want to get into any kind of a speculation on that, except to say that, in order that we can make sure that NASA gets started on fixing things, when we have a preliminary finding, we'll say that. Of course, NASA will know it at the same time and they will already be halfway down the road.

But at least for the Congress and the president, in order to make the return to flight decision, they will know our views.

HUBBARD: Yes. The first two came on earlier. And then in the reconstructed five seconds, we see evidence of the third and the fourth RCS jets firing to maintain the attitude of the orbiter.

QUESTION: Were RCS leaks detected--leak alarms detected in the last two seconds?

QUESTION: Okay. I'd like you to clarify a little bit for me the last few seconds of data regarding the APU. You said that the fluid is at zero. And I'm wondering if that's pressure, temperature, quantity, what we're talking about?

You indicated that, at some time, all three APUs failed. Then at another point, I think you said the APUs were running. So I need a little clarification on that.

And then finally, have you looked at whether--at the point at which the orbiter broke up, whether that was going at a sufficient speed to actually vaporize a fair amount of the debris, such that you will never find it?

HUBBARD: Right. The APUs were all running throughout the 32 seconds. They were running in the last two seconds, the auxiliary power units.

The telemetry shows that the hydraulic fluids were now at zero pressure; that is to say no pressure in the line. And the reservoirs that measure the hydraulic fluid had come down to zero percent. So it looks like the hydraulic fluid was dumped somewhere.

So the thought here is that even though the power was still there and things were turning, that there had been some failure in that period of time that affected the hydraulic lines and therefore, the fluids and the reservoirs.

GEHMAN: As far as the weather, the weather at re-entry is being looked at very, very carefully by a number of people, including independent agencies, not including NASA, because we are using the weather in our trajectory prediction of where the debris is. Our analysis of the videos and pictures of the return is getting more sophisticated.

And as you may be aware, we have asked a sheriff in Nevada to go look in a certain spot. That's because we are now down to actually tracking piece by piece.

Now don't get me wrong. We're not tracking thousands. But we've got a couple of pieces that we have been able to track. In order to do that, we have to be aware of the upper air patterns.

And it turns out that, in general, in the south central and southwest part of the United States, on that day, it was a particularly calm and benign environment. Jet stream was relatively low. Upper altitude winds were relatively low.

QUESTION: Was the heat of re-entry sufficient to vaporize some large portion . . .?

STAFF: Thank you. We're going to take some questions from Kennedy Space Center now. If the reporters could identify themselves and give their affiliation, it would help.

QUESTION: This is for General Deal. You mentioned looking at the NASA chart with the various fault trees and fish bones. Can you give us an estimate of how many hundreds, thousands or whatever scenarios that the board is looking at?

And we've all seen quite a few of the scenarios heavily in the press, I suppose from us. But can you give us some examples of scenarios you're looking at which have not appeared in the press?

DEAL: Well, I'm not sure that I can give you any examples of anything that probably hasn't appeared in the press because I think just about everything in the world has appeared in the press.

Again, we are not ruling anything out. And I hate to continue to give that generic statement. That's why we have gotten a list of everything that was on the Spacehab and we're looking at things like that--not that we think that's part of it, but we're looking at it.

If I were to try to tell you that we're focusing in on a few, that would be improper for me to do, because we're trying to eliminate everything that is not a part of it, as we still focus in on these key indicators that Scott Hubbard has alluded to, as well as the admiral and General Hess.

There are a number of key things that you, through your questioning, have keyed on. Of course, we're keying in on those. But that doesn't mean that we're eliminating anything else at this point.

HUBBARD: I think it's fair to say that we are going to track that. And as NASA works through the various elements of the fault tree and makes a determination that some element of it just couldn't have contributed to this tragedy and they close it out, we will be looking at that and see if we agree with that or if we--and say, "Go ahead, that's closed," or if there is additional things that we feel need to be done.

QUESTION: Okay. But can you give us an estimate of the numbers, whether it's hundreds or thousands of different scenarios you're looking at? I certainly hope it isn't . . .

QUESTION: I've been watching launch replays the last several days. And I have noticed that there is one piece of debris that seems to be bouncing off the left wing of the shuttle.

And I'm not talking about the pulverized foam image that we all have seen, but I'm talking about a solid, white object that seems to be just bounce off the left side. And I was wondering if you have any idea of what that might be?

HUBBARD: Sure. I know the video--at least I believe I do--the one you're referring to. And we have looked at that piece that was taken from a camera observing the orbiter on ascent at 82 second period. And there are several things in there that are quite interesting.

The visual analysis sub-team working group that's over at Johnson Space Center is going through this meticulously. We have stepped through it. In fact, I asked for a special viewing of that, stepping through it frame by frame.

The initial piece of debris, which appears to be a single element, appears. There is what we think is an optical illusion that it appears to actually go up. This may be due to the angle at which you're looking at it.

It's actually coming out and away. But because of the angle, it looks like it's rising, which would be a little counterintuitive.

Then the three pieces or so appear. And at that point, even if you step through it frame by frame, it's a little difficult to determine exactly what happens next. Those elements disappear. The objects disappear.

And then you see a plume. You see an object, apparently at the bottom of the wing. And then you see more of the debris crossing across the bottom of the solid rocket motor.

I don't draw any conclusions from this yet, except to say that this is a fruitful area for further enhancement by the video experts. And they're working on it and are going to get back to us when they have reached the next stage of sophistication.

QUESTION: This seems to be moving to the right on the screen. I mean, it just bounces and then moves to the right.

HUBBARD: There is an element that appears to do that. What I'm trying to emphasize is that I think we've already seen one optical illusion because of the angle at which we view this. So I am reserving judgment on which direction it went and whether it hit something or not until we go a little further.

QUESTION: For Brigadier General Deal, how close has Space Command placed the object to the shuttle on flight day two?

DEAL: Well, I can't give you an exact number of feet. It was floating near the space shuttle. And again, I'll go back to the laborious effort that's taken place, trying to detect what happened to this particular object.

It was floating nearby is when they detected it. And I don't have the exact number of meters there, but far enough away so they could tell it was a separate object.

But again, this is in what we call the scan mode. And it's not unlike a ground-based radar that's looking at a scan. And then it may lock in, like a fighter aircraft radar, lock in on a specific target.

They weren't doing that. They were doing a regular scan of everything crossing the horizon in their normal, everyday, day-to-day mission, 24-hour-a-day mission, looking for missile warning and for low altitude space tracking. So they were observing it come overhead.

And then, as they went back and did the delog I was talking about, again in a scan--not what we call a pencil beam where they'll send a pencil beam-sized burst up to look and identify what it is--they identified it slowly starting to fade away. And it was from four different worldwide sites that were able to see this as it floated away. Had sites in Eglin Air Force Base in Florida, Beale Air Force Base in California, Cape Cod Air Force Station in Massachusetts. And there is a system called the Navy Fence that's at several locations across the United States that looks.

So they all saw this small piece of debris that appeared to be accompanying the shuttle and separate from it. So it went from very close to a standard degrade of an orbit as it separated from the shuttle and did what most low-Earth orbit objects will do. And it degraded three days later and came back into the atmosphere.

QUESTION: I'd like to explain to my editors what's a fault tree and what's a decision tree. Can you gentlemen help me, please?

GEHMAN: A fault tree is a tool for making sure that you have a visual representation of all the possible causes of an anomaly. And essentially what you do is you draw a diagram that allows you to account for every possible thing that could go wrong. And you keep breaking it down into more levels of detail and more levels of detail until it looks like either the roots of a tree or the branches of a tree, depending on which way you draw it.

And it's an analysis tool. It's normally used as an analysis tool for fixing something which breaks. It's not specifically designed for a catastrophe like this.

But there is a fault tree for every part of the STS system, more or less sophisticated and more or less complete. And so it's useful if you think that some part of the orbiter or some part of the system broke to go to the fault tree and start using it as a road map so you don't overlook anything.

QUESTION: A question for Scott Hubbard on the--the assumptions you were telling us about in your opening remarks about an area that you assume is damaged and then you can model and get plume that explains what you might have been seeing in the wheel well.

And then you mentioned again some sensor data up on top. My questions are: where exactly was your presumed area of damage? I'm trying to get a sense. Did that actually cross a landing gear door hinge line or the seal area? Was it in front of it? And you're assuming plume got in.

Can you give me some specifics? And also, when you mentioned above the wing, I've been fascinated from day one about the data from the longeron sill area, which is well above the wing, and other side wall sensors. Is there any sense of what's going on there to explain that stuff?

HUBBARD: Okay, let's take the first one first, Bill. Let me go back and make sure the point I was trying to make is clear. What the NASA Mishap Response Team engineers have done is an initial set of thermal calculations.

What they were trying to explain was the very rapid 60 degrees in five minutes temperature rise in the wheel well area. And to do that, they started off with a fairly straightforward and simple set of assumptions. This is what you always want to do if you're trying to box in and go from the town to the ballpark to the seat.

They said, "Let's just put heat on the outside of this box, this wheel well." And they raised the temperature and increased the heat to what is known about the environment. And they could not reach the kind of temperatures inside the wheel well that were measured.

So they said, "Well, let's imagine then that we have a breach somewhere. And that breach was not specified. It just was a breach of 20 square inches--you know, a hole four inch by five inch--somewhere and you just dump this heat into the box." If you do that, you can, in a simple calculation, show the temperature would rise to the observed level.

That's about as far as they had gone a day or two ago. What they need to do next is exactly what Bill Harwood is asking about, which is a much more sophisticated analysis of moving this breach around, dealing with the structure on the inside, which is very complicated.

And once this superheated air would get in there and try to move around, you're faced with a much more complicated area, which is the area of computational fluid dynamics. It's not impossible to calculate this. In fact, there are people that have done that for a living for decades. But it's going to take a while to get to the next level of sophistication that we would all like to see.

The second thing has to do with where the sensors are and what kind of data we've seen. And the team is in the process of providing the board with data that is in three dimensions, not just two, so that we can stand off and see where the sensors went up in temperature, not just looking down at it from a top-eye view, but looking from the side and then stepping through that in time sequence.

And I think when they finish that analysis and we can make it available, we'll be able to address some of the questions about upper versus lower.

I covered the Challenger accident. And I covered the Apollo fire. So I've been around during several accident investigations. And there has never been complete agreement among the panel. I'm thinking specifically about the Rogers Commission.

Have you considered what, if you do not agree with each other 100 percent, what you are going to do then?

GEHMAN: Yes, we have. And it hasn't come up yet. But if someone feels strongly that they need to in some way express their opinions different from where the report reads, then we will have a minority report attached to it.

QUESTION: For General Deal, you talked about the approximately one foot by one foot piece of debris that was tracked after the second day of the flight. You said it was lightweight.

Is it consistent then with what would be a space shuttle tile material? And is it consistent with something that could have been knocked loose by a hit of orbital debris?

DEAL: Well, again, it's too early to speculate on that. They will be doing some tests at Wright Patterson Air Force Base, based upon the radar signature that they were able to get off of it. You could conjecture any type of scenario, anywhere from tile to--if you think about where this was in the flight profile of the Columbia, they had already been on orbit and they had opened their payload doors.

Could it have been something separated from there? Sure. You could build any kind of scenario.

So that's why the continued analysis, based upon what they got from the signature, from the size, from the number of hits and from its orientation, will be critical for them to help eliminate that or determinate that that is a factor in this mishap.

QUESTION: Admiral Gehman, can you tell us more about your trip to Washington tomorrow? Who invited you to come? Who will you be meeting with?

And is this a fact-finding trip for you, wanting to interview members of Congress? Or is this is a briefing trip, where you're giving them the information?

GEHMAN: I'm going to go see half a dozen members of Congress--half a dozen members of the House and half a dozen senators who are on the oversight committees that provide oversight for NASA. And I'm going to give them an update on where we are, what we're doing and how we're organized and let them meet me and see me so they can put a face with a name.

And I'm also going to listen to them, to make sure that our report is comprehensive enough to satisfy their needs and their requirements. So there will be a little bit of me talking and a little bit of me listening. It certainly is not part of the investigation of the accident.

QUESTION: For Scott, what is the status of new tile foam impact physical tests? When and where and if they're going to be done independently?

And is it fair to suggest that nothing so far has contradicted the theory that the foam impact may have done something to the tiles that may have contributed to this breach?

HUBBARD: Okay, let me address your two questions there. The first one: what is the status of testing? There have been, as you know, tests in the past by Southwest Research Institute of firing various types of shuttle foam materials at targets.

What we're interested in is eventually some testing that may include not only, say, the tile or the reinforced carbon/carbon leading edge, but also the structure underneath it. This is part of our systems approach.

Those tests are not scheduled yet. NASA is moving out smartly in doing a whole lot of different tests. They are preparing to look at things like external tank insulation that's been produced for, let's say, a cousin tank, cousin external tank.

So there are a lot of things underway. Can't give you a schedule right now. But it is our expectation that we will need this kind of test data in order to begin to conclude whether or not the external tank foam, in fact, played a major role in this tragedy.

Now that leads into the second question. And I would, I think, turn it around and say: What is there that proves that external tank foam insulation by itself can cause enough damage for something like this mishap to occur? Remember, I said that there is almost always a chain of events in accidents and complex systems.

So in looking back at the previous situations where it was observed that external tank insulation--and I'll come back to insulation in a minute, what does that really mean?--fell off and then when the orbiter returned, there was observed damage to the tiles. Now that's a correlation, but it may not be a cause and effect. The routine number of dings that the tiles get is somewhere over 100 from a typical flight.

In pressing down through levels of detail in observations of external tank insulation falling off and actual damage to the tiles, there is only a few cases where I think you could actually document for sure that a tile damage was correlated with a external tank insulation falling off. So I think we need to be careful about connecting every single event.

One of the things that our group is very interested in: is it only the lightweight foam? Is it the lightweight foam plus the material underneath, which goes by the name of SLA-561. It's an ablative material. It's a protecting material.

It is the same ablative material that was used for the Viking heat shield and is being used for the Mars Entry Vehicles that will be there next year. It's been around a lot time.

And third: could there be other things in there? Could, perhaps, some piece of a cap of a heater element or something come off?

So in the spirit of looking across the board, I want to emphasize that this external tank insulation event needs to include more than just the lightweight white stuff that turns orange in the sun.

QUESTION: Two questions. One: I was wondering if you have learned anything from the piece of the left wing that was found earlier?

And two: I have heard and read that the reason there was the burst of--the loss of contact between the five seconds and the two seconds in that 32-second period is because the antenna on the shuttle wasn't properly lined up to send signals back to Earth because it was tumbling so much. And I was wondering what you think of that idea.

GEHMAN: In the case of the first question, all of this tile debris is still being put together in kind of a jigsaw puzzle. We're not ready to announce any findings or anything at all from what we've learned or from what we found.

We found some interesting things, which we have relayed to you. But we're far too far away from coming to any conclusions or any findings.

As far as the missing 25 seconds in between the five seconds and the two seconds, I wouldn't characterize it the way you did. As the shuttle turns and pivots, as it travels around the world, as it shifts, it has multiple antennas, radio antennas, it always tries to keep a direct line of sight between one of the TDRS satellites and its antennas.

Sometimes there are a few seconds where it's interrupted. Sometimes, the re-entry hot gases and flow around the shuttle temporarily, for a few seconds, interrupt communications. That's been normal.

But I wouldn't characterize it as some antenna was pointed in the wrong direction or something like that. This is just a normal interruption of communications for a few seconds that happens every once in a while.

STAFF: Thank you. We're going to move to questions from Langley Research Center in Virginia.

QUESTION: I have a two-part question relating to the NASA-Langley emails released last week.

What evidence do you have that supports or contests a Langley engineer's theory that tires bursting in the left wheel well could have caused the catastrophe? And secondly, another Langley engineer speculated that the launch debris could have been ice from the external tank dump line. Have you studied that scenario?

GEHMAN: The answer to both questions is "yes." We are entertaining all scenarios very seriously.

There are some on-orbit telemetry to suggest that the tire exploding is not very plausible. But we're going to run it down since we have good solid telemetry of tire pressure.

But the ice business is being taken very seriously. Scott clearly alluded to that, that whatever hit the shuttle isn't necessarily just foam. It could be a lot of things.

And oh, by the way, it's possible that that left bipod ramp is not the only source of the foam. That has shed foam in the past. But there are other places that foam could have come from.

That foam could have been saturated with ice, which would make it very dense. All these are things we're looking at very seriously.

STAFF: Okay. We are going to move back to questions here, probably take about four or five questions here max.

And then, what I'd like to let the TV folks know is that we're going to have a nine-minute feed uplink of debris recovery efforts around Lufkin and Nacogdoches. Just want to let you know it's available.

QUESTION: Can any of you gentlemen tell us whether the areas of closeouts--that is, patches of foam applied either at the tank assembly facility or at Kennedy--is an area of special concern or study?

GEHMAN: I'll speak in general. And this is being followed up on by the gentleman over here to the right, particularly by General Deal and his guys.

We have studied very carefully every time anyone touched the Columbia between flights. We have the records of every piece of tile that's been repaired, replaced, minor repairs, big repairs. We have all those records.

And we are looking at them very, very carefully to see whether or not any of them are of any interest. And then, of course, if we get a piece of debris that is one of those repairs, we can put some things together.

But the answer to your question is "yes." That's an area of great concern to us. And the Materiel and Refurbishment Group, specifically, that's on their list.

DEAL: Not much I can add to that, other than echo what the admiral said. We will be looking in-depth at it. That's one of the reasons we are going to the Michoud Assembly Facility this week.

We'll be looking at the quality assurance processes, what the inspections were, the government inspection points, in addition to the contract inspection points, what the processes are, how they inspected it through non-destructive inspection as well. And, as has already been mentioned, we have a kind of little bonus here that the sister tank to the one that was on board of Columbia, which was number 93, 94 is still in the plant and impounded. So we'll be able to group with group three and do types of testing on that because it is in family.

QUESTION: My question is for Scott Hubbard. And it, too, has to do with your opening remarks about the thermal testing that's planned on the wheel well.

Can you elaborate a little bit on to what extent this testing is intended to tell you whether this breach occurred around the seal of the wheel well door, on the leading edge where you have the carbon composite, or somewhere on the top of the wing and how that answer, when you get it, would fit into the larger story of what happened here? Could you sort of connect those dots, please?

HUBBARD: My first point is this is not testing. This is analysis in a computer. For decades, there has been a trade called computational fluid dynamics. And we have people out there that are very adept at taking wind tunnel data, coupling it with computer methods and then being able to use that to extrapolate what might happen in some future design.

That's the way most modern aircraft have been done. And there was tremendous amount of this calculation done in developing the shuttle.

So what's going to happen is they will take these computer models that they've got of flow of this rarefied atmosphere--I mean, we're up at 200,000-plus feet. The atmosphere up there is 100,000 times less dense than it is here on the ground.

And they're going to introduce into those models--those computer models--holes of breaches of various sizes and shapes in different parts of the wing. And then they're going to look and see how the heat of that very rare, sparse atmosphere could get inside the wing and then where it might go from there.

And if you have access to the website and go in--I believe it's available--and you look at a cross section of the wing, you will see there is all kinds of structure in there. So as that flow goes in, it gets very complicated.

And that's why the first cut analysis was a good thing to do, which is just assume that you can get somehow--through the seal, through the door or somehow--this amount of heat in there and see if that correlates with the measured temperature rise. The first cut, it does.

Now though, you have to ask yourself the question: can you box this in? Put a breach at the front leading edge, put a breach at the wheel well door, put a breach at the seal of different sizes and shapes and then see where that takes us and see if that begins to match with the data that we've got.

So this analysis is ongoing. But again, it's stuff done in a computer. It's not testing.

GEHMAN: I'd like to follow up on that because there have been several questions on this. We actually have six of these analytical detective works going on. And the idea is that if you come up with a thermodynamic scenario, it also has to match the aerodynamic scenario, which also has to match the timeline that we have constructed as to which sensors went off.

That then has to match the debris shedding scenario. That has to match the imagery, the photographs of the debris shedding, which also has to match the evidence that we pick up from the debris on the ground.

And then we hope that we'll be able to go back into the documentation of the repairs of whoever touched the shuttle. So there are six separate detective stories, one of which is the thermodynamic detective story.

And once we get a convergence of what appeared in the telemetry, what appeared in the photographs, a theory of how it could have happened thermodynamically, a theory of how it could have happened aerodynamically, we'll be getting somewhere now. It won't be proof. But it will be a convergence of possibilities.

And we have been beating up on the thermodynamic one here because it's very interesting. But even when we get a thermodynamic scenario which matches, then it also has to match these other five that I mentioned in order for us to have any confidence.

QUESTION: Admiral, you were saying you can't be sure exactly where the tile we saw came from. How specific can you be? Any more specific than simply saying it came from the glove area of the wing?

And you also said it did not show typical wear for tiles heat facing surface. Can you explain what's untypical about it?

GEHMAN: First of all, we got our tiles mixed up. The tiles which came from the glove area was the tile which I don't have a picture of here, the farther westernmost one that was found near Littlefield, Texas.

This tile, which was found in Powell, Texas, just 30 miles west of Fort Worth, the one I showed the picture of, it's the one that has the very extreme heat damage to it. And whether that damage was caused while it was still on the shuttle or after it broke up or something like that, we have a lot of investigating to do.

But this heat damage on this tile is not re-entry damage of a shuttle. It's not what they look like when they come back.

GEHMAN: It should be smooth. And it should be slightly gray, very, very slightly gray, smoky and smooth. Maybe a couple little tiny pocket marks in it from the 100 hits that Mr. Hubbard referred to.

QUESTION: I had a question about the final two seconds of data, telemetry data. Is it possible to say anything more specific about this? And by that, I mean we know the APUs were working. Can you say what other systems the telemetry data said were working on board the system, such as computers and other things?

And also, we know that they hydraulic fluid had drained out. Were there any other sensors, particularly sensors in the left wing, that had been working prior to that two-second period that were off scale low or not working in the final two seconds?

GEHMAN: Correct me if I've got this wrong, Scott. As far as we can tell, we have told you everything that we know that wasn't working with one exception; and that is that in the final two seconds, there were no signals from the left wing. Is that right?

HUBBARD: We've said everything that wasn't working. There is a whole long list of things that were working.

GEHMAN: That's right. As far as we know, in those last two seconds, the guidance system was talking to the jets, was talking to the elevons. The nose of the craft was talking to the tail. All those systems were intact.

And the big anomaly that we noticed was that the reservoirs of the hydraulics were empty. That's the big anomaly that seems to stand out to us.

QUESTION: Gentlemen, I wonder if you have looked at the three analyses produced by Boeing during the flight. There is now a contention out there that Boeing's analysis of the possibility of damage on liftoff was deficient because Boeing has restructured, moved personnel here, lost some of its knowledge base.

And in the fault tree, failure to make proper evaluation then led to failure to take preventive steps later, whether you have followed that or reviewed any other aspect of that analysis and whether your supervision of the NASA investigation also extends to whether the people doing that investigation have any conflict of interest, whether they are investigating what they formerly directly supervised?

GEHMAN: The answer to all your questions is "yes." We are looking very carefully at the foam analysis from a half a dozen points of view. You've heard three or four of them here today.

And then all the other things that you mentioned about conflicts of interest and who did it and were they qualified and all that, all get looked at. It will all get looked at.

It's in progress. Though I don't have anything to tell you about it. It may have been done wonderfully. It may have been done expertly.

There has been some speculation in the press by some experts about the statistical methods, that they weren't happy with some of the statistical methods and things like that. We're going to look at all of that. We're going to look at all of that.

I do apologize that I am going to have to cut it off here after an hour-and-a-half. As I say, we will be available every week so you can dialogue with us. And the public will be a week from Thursday. It will be open to the press.

STAFF: Thank you everyone. And there will be a screen up for a few minutes that will have information about the board's website and how to contact us.