Open main menu
TWA Flight 800 investigation
Portal:TWA Flight 800 investigation

NTSB Board Meeting
August 23, 2000

Day 2, part 4of 4
part1, part2, part3

Subject to revision

JIM HALL: Member Carmody?


JIM HALL: Any other questions from the members on the Witness Report?

(No response.)

JIM HALL: If not, we may get back to that in a minute. But that concludes the presentations. I would like now to get into any areas that the Board feels has not been covered.

Dr. Loeb, I do think that there has been some questions about the radar data, and I would like to have the individual that worked on the radar data make a presentation on what he has done, so that the individuals, the family members, will know this is the person that has worked on this area.

And also, if Mr. Grossi is in the office here, in the building, Mr. Grossi, if you would come up and take a seat. Some people have said that you have -- that there has been a problem with the flight data recorder information and the handling of the flight data recorder information. And if people are going to make accusations, I want to be sure my Board staff, have an opportunity to respond.

Mr. Dennis Grossi, he is the individual that works on the flight data recorder. And, Mr. Pereira, you might give us an overview of the radar data.

MR. PEREIRA: Yes, sir, Mr. Chairman. Early on in this investigation, we received radar data from the FAA and also from Sikorsky. They have a field in Connecticut. There were a total of nine radar that we reviewed data from. Five of these were FAA approach control radars: at the Islip Airport; at JFK Airport; at Newark, New Jersey Airport; White Plains, New York; and Stewart Field. Each of those had a 60 nautical mile range and each of those recorded what is referred to as Primary and Secondary radar data.

There were also three long-range radars that had coverage of this area: FAA radars in Trevose, Pennsylvania; Riverhead, New York; and North Truro, Massachusetts. Each of those had a 250 nautical mile range. Each of those also recorded primary and secondary radar data.

The Sikorsky Field got a feed from the Riverhead, New York FAA radar that was basically a duplicate data set. And they also had their own flight test radar that had a 60 nautical mile range. However, that flight test radar only recorded the Secondary returns; no Primaries.

What do all these radar data show?

In general summary, they show a normal takeoff and climb-out from Runway 22-Right at JFK. In the area were numerous commercial airlines, general aviation and the surface vessel tracks. There were also several military aircraft over the coast and in open water. And after review with the DOD, none of these were reported to have weapons on board.

The TWA 800 Secondary radar returns end at 13,700 feet altitude, followed by a large number of Primaries, consistent with the breakup sequenced discussed so far in this Board meeting.

There were quite a few early discussions about whether or not there was any radar evidence of a midair collision or a missile or projectile. After a thorough group review, we came to the conclusion that there were no sequence of radar returns that intersected TWA 800's position at any time. There were also no radar returns consistent with a missile or a projectile moving towards TWA 800 at any time.

That's my basic summary of the radar data. I do have some graphics if you're interested.

JIM HALL: Well, show us what you have there, Mr. Pereira.

Let me ask you, first, Mr. Pereira, and I don't like asking this question but I want to be sure it's on the record: Has anybody directed you or the Radar Group to come up to any conclusions in regard to what your work is?

MR. PEREIRA: No, sir. It's totally free and objective.

JIM HALL: Are you a professional in this area? Do you consider yourself one?

MR. PEREIRA: Yes, sir, I try to be.

JIM HALL: So the Group has not received any particular direction? Or is any information excluded? Is all the information there?

MR. PEREIRA: No, sir, we were free to talk to whoever we wanted to and form our own conclusions.

JIM HALL: Very well. Please, proceed.

MR. PEREIRA: Okay, the image here shows the Long Island coastline in the upper left. It's a 10-minute time period we're looking at here for the Islip Primary and Secondary radar data. Now, Primary radar data is essentially just where you send out energy and, if you receive a signal back, the radar system identifies the range and azimuth at which that energy came back from. Those data, referred to as Primary radar data, are represented on this image as the white triangles.

And, John, if you could point those out with the mouse, please.

The blue dots, some of which are accompanied by arrows and nomenclature such as the "P-3" and the "USAir 217," those are Secondary radar returns, which is essentially where you send out a coded piece of electrical energy out into the area and, if an airplane receives it and replies, you get information on the identity and the altitude to go along with the Primary radar data regarding range and azimuth. And so we have Primaries and Secondaries on here which show a variety of airplanes, some of which we have labeled here, and Primary radar data.

Now, climbing out to the northeast, in green, we have TWA 800. And there's a yellow square, sort of in the center of the image there, which is where the last secondary occurs.

You can see, to the north of TWA 800, is TWA 900, climbing out to the northeast, as well.

USAir 217, which Mr. Mayer mentioned in his witness discussions, is coming up from the south, above TWA 800 in terms of altitude.

Moving from the northeast to the southwest is a series of primaries, which is the Navy P-3. And the reason we had primaries on him instead of secondaries was because his transponder was, for the most part, broken and not functioning. We did at one point, well on past this, get one secondary return from his broken transponder. We know, from his accounts, that he was at 20,000 feet. And the one transponder return that we did get was at 20,000 feet.

And I would like to point out, at the end of TWA 800's secondary returns there, you'll see kind of a cluster of primaries that extend from the last secondary, to the east-northeast. And then, if you saw it in time sequence, it would seem to drift off to the southeast. And that happens to be the direction of the wind, the prevailing winds in this case. So, once the airplane broke apart, these primaries move in the direction of the wind.

When we first received this radar data, we plotted this on a map for the search and rescue people and sent it up there to them so that they could search in the appropriate areas. And the promptly notified us that they had found the wreckage on the ocean floor in that area.

And I just want to carefully point out that this is, again, 10 minutes' worth of radar data. Then I'm going to show you some images later, in a more short time frame, right around the accident. And you'll see that, actually, at the time of the accident, there wasn't really all this traffic in the area.

The next image, please. These are the same image with the secondaries removed so that you can see just the primaries for that 10-minute time period. Next image.

Now, this is the shorter time frame that I was referring to. This is the Islip Primary and secondary radar data for a period of about 42 seconds prior to the accident. And you can see that we have TWA 800's last secondary, the yellow dot in the center.

The position of the P-3 at the time of the last secondary is at the end of the arrow there, for the P-3. So you can see he was well past TWA 800, and headed to the southwest. USAir 217 was physically above TWA 800, in terms of altitude, and coming up from the south. TWA 900 was off to the west and climbing to the northeast. And that's it for our secondaries in the general vicinity of TWA 800.

As far as primaries are concerned in this last 42 seconds, you'll note that there are not any sequence, as I mentioned, of primary returned headed at TWA 800, as though representing a missile.

Now, we're not sure that we would even be able to see a missile on primary radar data if it were moving towards TWA 800. Some of the tests that we did down at Florida demonstrated that we probably would not see a small-cross-section projectile such as that.

As far as the primaries that we do have here, you'll note that one of them is labeled "30 knot target." And that's headed in the slightly southwest direction there. And it is several miles past TWA 800's position at the time of the last secondary. That track is consistent with a boat moving on the surface of the water.

As well as off to the left of TWA 800 there, there are several white primaries that you see there. Those are also 12 to 15 knot, what appear to be surface vessels; constant heading, constant speed, very consistent with a surface vessel.

Next slide, please. These are just the primary radar data alone; the same basic image, just with the other secondaries removed. Next slide, please.

Member Hammerschmidt asked me a question, and I showed this slide yesterday, regarding some of what we termed false primaries from the Islip Radar. This is, again, 10 minutes' worth of radar data for primaries and secondaries. The whole purpose of this one is to point out false primaries, down in the lower left.

In addition, this group does extend further off the page. It's showing four of the returns in question here. There were several more that followed after that in the same direction. There were also three additional groups -- or two additional groups -- that were identified over the course of the next 20 minutes in this same general area.

As I mentioned yesterday, we came to the conclusion, after discussions with the FAA radar specialist, that these were false primaries that occurred as a result of reflections off of some structure around the Islip Radar.

And just to get an idea of whether or not it's typical to have a bunch of boats and Secondary traffic in this area, we took a completely random day from approximately three years later -- if you'll go to the next slide, please.

About a little more than three years later, in the same area, from the same radar. And you can see similar Secondary high-speed, apparently airliner, traffic, low-speed general aviation traffic, and also numerous Primary tracks, low speed, consistent with boats, the same as the day of the accident, which is to be expected for this type of area this time of year.

There are numerous recreational opportunities out there, which I take part in almost every weekend myself, in a similar type of area. And when I'm out there, there are always numerous large, anywhere from 5 knots if they're trawling, to 15 knots if they're cruising, commercial maritime vessels, be they trawlers, scallop trawlers, longliners, commercial ships. So this is not unusual. And this was a good backup for us in that regard.

JIM HALL: Okay. Thank you. Now, Mr. Grossi -- you have questions? Okay, we'll take questions on Charlie's presentation now. Since we've completed the work, I'll just ask the recognized members that have questions. Member Goglia.

JOHN GOGLIA: Charlie, the secondary response that you gave, it's through a transponder. And I don't think we adequately explained that out for the people who don't understand what that is. Could you explain to them what the transponder is and also what information is sent back? Because it's not just solely a radar return, as you know.

MR. PEREIRA: Yes. A secondary return, if you'll note, most of the airports that -- like if you go over here to National or quite a few other large airports, you'll see an orange disk spinning around on top of a large metal tower. And there's actually two radar disks there, if you'll notice, one Primary and one Secondary. And the Secondary one sends out a coded signal. And the purpose is to try to elicit a response from an airplane so that you'll know the altitude that it's at and its identity for air traffic control purposes.

JOHN GOGLIA: It's not an echo return like Primary radar. It sends a signal out and asks the receiving airplane to send me a signal back.

MR. PEREIRA: That's correct. The transponder receives the signal from the Secondary radar transmitter, and transmits back out into the atmosphere, which the radar then receives, and says, okay, I will combine this identity and altitude information with the Primary radar data that I received from my other radar, and I will feed this to the controller, who will get a dot on his screen, to say this is TWA 800, climbing out at this altitude and this position.

JOHN GOGLIA: So it sends altitude?

MR. PEREIRA: Altitude as well as beacon code, is what we call it. The pilot dials in, in the cockpit, as part of his flight planning, his assigned beacon code. For example, TWA 800 was beacon 2633. So when the Islip Radar was sending out energy in this Secondary signal, it was contacting TWA 800's transponder antenna, and it was returning the information, "I am 2633 and I am at this altitude." And the Islip Radar is using that data, in conjunction with its Primary radar data, to feed to the controllers to help them track the airplane.

JOHN GOGLIA: Now, that transponder, which it's called, located in the airplane, must be powered in order to send that signal back; is that correct?

MR. PEREIRA: Yes, sir, that is correct.

JOHN GOGLIA: And it's powered by an AC power supply?

MR. PEREIRA: I would have to refer to the systems people for that answer. Bob is nodding "yes."

JOHN GOGLIA: Well, I sort of led you, because I know the answer to that.

MR. PEREIRA: Thank you.

JOHN GOGLIA: But what I'm trying to do is make sure that everybody in this room and everybody that's listening to us understands the difference between Primary and Secondary returns. And that wasn't clear in your presentation. And it is important to us, because it gives us so much data. Simply, those few data points really help us a lot. And especially if we lose them, it's important to know that we've lost power on the airplane.

MR. PEREIRA: That's correct. And we've seen that on numerous other accidents we've had recently. And it gives us a lot of clues as to when there are electrical problems on an airplane.

And to give you another analogy to maybe help the audience a little bit, the Primary radar is basically like what a bat does when it's trying to catch insects. It's sending out energy, and it knows there's something there, and it goes and grabs it. And then we have refined that a little bit, with our engineering skills on the human side. We use that technology now to not only say where something is but who it is and what it is. Essentially, it was developed for wartime purposes, but has been used for air traffic control since.

JOHN GOGLIA: Okay, thank you. That was the only questions I had.

JIM HALL: Any other questions for Mr. Pereira?

(No response.)

JIM HALL: If not, Mr. Grossi, if you could touch on this issue of the last four seconds of the flight data recorder tape, anything else that you want to address in terms of the recorders? I think most citizens in our country are aware that the first thing, when we do have an aviation tragedy, that the Board tries to recover is the so-called black box. We have a number of people on staff -- Mr. Grossi, how many years have you been with the Board?

MR. GROSSI: Since 1972.

JIM HALL: There's not too much for me to add but, anyway, a long time. And we have individuals that are trained to read these out. We have a group both for the flight data recorder and then we have another group for the cockpit voice recorder.

Now, in terms of this particular accident, Mr. Grossi was the individual that was in charge of the group. And, Mr. Grossi, as you are aware, there have been some accusations about the last four seconds of the tape. And if you could clarify that for us, I would appreciate it.

MR. GROSSI: I will attempt to do that, sir.

I'm glad you used the word "group." This is a flight data recorder group and, as with all the other investigative functions, it is conducted as a group. And I might add at this time that all of the members of the group, who had complete access to the data and were present during the recovery process and on subsequent examination of the recording, are all in agreement with the findings as we've presented them in the factual reports that were produced.

As an explanation for some of the confusion that's surrounding the interpretation of that factual information that was presented, originally, the information that was recorded right up until the end of Flight 800, we looked at that information. In fact, we worked it through the night once we received the recorders. Essentially, that information has not changed since day one.

And possibly the areas where confusion has set in is, in my report, I included data from Flight 803, which is the flight from the previous day. The way flight data recorders work, they produce a 25-hour record. But, in doing that, they write over the oldest information on the tape. And in this case, the information that was being written over by data from Flight 800 was the data from Flight 803 from the previous day.

Because these are not continuous records, as you can imagine, through transitioning from one flight to another flight that happened on the previous day, there is some discontinuity in the data. Digital data must be continuous. Once you interrupt the data, then there is a period of time where you have discontinuity and the data are no longer correct. You cannot directly interpret that data straight away using normal playback features.

This is something that is well-known by people who deal with flight data recorders. And if I were to look at a report and somebody said that they got all of the data right up to the end of a recording, and I did not see this discontinuity in the data or this break in the data, where the data were not correct, I would be suspicious immediately. And just knowing how computer software works when it deals with data of this type, where it's out of synchronization, quite often software will just ignore that data and then pick up on data that is in sync.

Therefore, you have the opportunity to miss critical data. We know this. We've done this for years. And so our first effort is to go to the end of the tape to determine where that end is. That's the very first thing. And then we match up the cockpit voice recorder with the flight data recorder.

That's the second thing we do, is to try to see if both recorders were ending at the same time. This doesn't always occur. There are times when the CVR will run longer or the flight data recorder will run longer. So that's a critical piece of information. So that was another step we did. And what we found was that they did in fact, keeping within the bounds of the fidelity of both recordings, they did in fact end at the same time.

Now, at this point I would like to acknowledge that I did make a mistake in my original report. I did not make a mistake in that I didn't determine the last bit of data that was recorded. I did make a mistake in interpreting those bits, what those sequence of bits were. I found that error. I subsequently corrected that error.

What are the consequences of that error? Nothing really. Actually, it allowed us to bring the two, the CVR recording and the DFDR recording, the end points, to within closer tolerance, or closer ending time. So it gave us that possibility that both recorders came even closer to ending at the same time.

I probably should get into a little discussion about how this recorder works. It does not record the information directly to the tape. It first fills up what's called a buffer. So this buffer, you can think of it as a bucket or a glass, once it's full, it then dumps the information onto the tape. So there's a trickle of information going in and then it gets dumped onto the tape.

That dumping of the information onto the tape only takes .07 seconds to occur. For the rest of the time the tape is either backing up or repositioning itself so it's not being written to. So for the recording to work properly to know where this data is being put on and where the break points are, where the data are actually being dumped onto the tape, the recorder puts on additional codes into the data stream.

And that's where I made my mistake. I misinterpreted the code, this postamble code, as being part of the data word. And I recognized that it was a postamble, but I thought I only had part of the postamble. But under reexamination I realized it was the whole postamble. So once I determined it was the whole postamble, I couldn't determine how much within that buffer, or that glass, whether it was this full, this full, or this full.

Being all the way full was a little over a second. So now we're able to go a little bit further, at least have the potential of going a little bit further in time, which brought us even into closer correlation with the end of the CVR recording.

Another match that we did to ensure that we had the end of the data to match up with the other evidence was the correlation with the radar data, the altitude information recorded by the radars. And that was closely correlated by Mr. Pereira and we got the same correlation. We got everything ending at the same time. So it was another correlation point, another thing that told us we had the end of the data.

It's pretty conspicuous where the data ends on this particular type of recorder. There is a gap in the recording. It's about three inches in length. And we were able to find that very precisely, both visually on the tape and electronically.

There has been a lot of confusion surrounding this. And I think a lot of it has stemmed from the fact that people do not really understand completely how we process the data, and were inferring a lot into that process and coming up with their own conclusions that supported their own prejudices on the case but were not founded in fact.

With that, I hope I've helped you out in trying to answer some of those questions.

JIM HALL: Well, I think that's clarified the matter, and I want the family members to be able to know and see the individual who does the work. And obviously there is a whole group of other individuals that participated in the work and signed off on the work, but you were the person from the Safety Board that was there to guarantee the independence of the work.

MR. GROSSI: Yes, sir. That's a very important point. I think when you consider that the people that had the in-depth knowledge and were there firsthand and observed the process all signed off on it -- and these are not just anybody, these are experts, and even the recorder manufacturer was involved -- and they all signed off on the work and are in complete agreement with our findings.

JIM HALL: Very well. Any other questions for Mr. Grossi on this matter? I wanted to try to get all these matters that the family members have been given that raise questions and concerns about the independence and, most importantly, the integrity of this Board responded to. Member Goglia?

JOHN GOGLIA: Just one question. Somebody here earlier had handed me a piece of paper, which I put in my pocket, that actually challenges eight seconds. Now, is it possible for you to be off by eight seconds?


JOHN GOGLIA: I didn't think so, either, because I have seen your work. And, again, I don't think that there's anybody out there that's really -- at least not the ones I've talked to -- are lying to us. But not everyone has the complete package of data. And virtually everybody in this room knows that both the flight data recorder and the voice recorder are very, very privileged in the course of an accident. We go to great lengths to secure the data on both of those. And only those people that really have a role to play, the experts that work in the groups, have access to all the data.

MR. GROSSI: Yes. It's a very privileged group that listens to the CVR. We do share the DFDR data as much as possible with the groups. And that is wide open and we electronically transfer files to people like the Airline Pilots Association, the aircraft manufacturer; they are always given full access to the data.

The problem always comes when people aren't familiar with the format in which the data are recorded and then try to read things into any inconsistencies -- at least inconsistencies that may be in their mind -- in the data.

JOHN GOGLIA: Thank you.

JIM HALL: Other questions for Mr. Grossi?

(No response.)

JIM HALL: If not, do any of the Board members have any other areas that they want to explore?

(No response.)

JIM HALL: I only have one more, but I will defer since I have had some time here with these two presentations.

Mr. Sweedler, would you give us the status of any of the outstanding recommendations that you have not previously presented so that the Board knows exactly, on previous recommendations on this accident, so everyone knows what the current status is? And you might explain how you determined whether something is Acceptable, whether it's Unacceptable, or the various other categories.

BARRY SWEEDLER: Yes, Mr. Chairman. In the course of this investigation, the Safety Board made 11 specific recommendations on three separate occasions. The first set of recommendations was made actually only five months after the accident. These were the four recommendations that dealt with changes, both design and operational changes, to preclude explosive fuel-air mixtures in the fuel tank. These were the four recommendations that dealt with that issue.

Until recently, two of the recommendations -- actually three of the recommendations -- had been held in the Unacceptable classification. And the way the Board determines how that recommendation is classified as Acceptable or Unacceptable, the recommendation is made and, under the law, the FAA has 90 days to respond to our recommendations and tell us precisely what they plan to do or not do in answer to our recommendations.

JIM HALL: That's under statute; that's the law of the land?

BARRY SWEEDLER: That's under statute; they have 90 days to respond to us. When we get their answer -- and, in this case, the answer came in well within the 90 days -- but when we get their answer, we then very carefully analyze what they're planning to do to implement the recommendations. And this is a process where my office works with the Office of Aviation Safety, the Office of Research and Engineering, and the staff develops a position as to what we would recommend to the Board that the classification should be.

And in three of the four recommendations, the classification had been Unacceptable. Now, since that time -- and then we would go respond back to the FAA and explain why we think what they're planning does not meet what our intent was. And then we would have additional correspondence. We've had a series of lengthy meetings with the FAA so they can further understand what our concerns are and how the recommendations may or may not be being met.

In this case, we now have additional information, correspondence and from information we've gathered at meetings, from the FAA that will allow the staff to propose to the Board that the recommendations be held in an Acceptable status because of the different actions that the FAA has taken. And we have discussed that through the last few days. And I could go through that again, but basically the FAA is now acting more in concert with that we had proposed originally.

One of the recommendations had been closed and is no longer applicable. It dealt with changes to the handbook that TWA would use for 747's. TWA no longer operates 747's. And even Boeing had looked at the handbooks of other carriers, and that particular recommendation just is not applicable any longer.

The only one from that first set of recommendations that still remains open-Unacceptable deals with the temperature probes in the center wing tank that would give the cockpit crew an indication of what the temperatures would be in that tank. And the FAA has responded that some of the steps that they are planning to take -- they haven't taken yet, but planning to take -- on directed ventilation and fuel tank inerting systems, these are all very promising but they are not in operation yet.

So we felt that, until these things are in operation, it would still be useful for the crew to have the information about the temperatures in the center wing tank. So that is the only recommendation of the 11 total that the staff was proposing be kept in an Unacceptable status.

There was one recommendation that was issued in February of 1997. And that dealt with an issue that had been brought up about the explosive training aids that had been used on this airplane and had left, apparently left traces of explosives in the fabric. We recommended that the FAA develop and implement procedures for handling and placement of the explosive training aids by K-9 explosive detection teams to prevent contamination of aircraft and airport facilities.

The FAA has accepted our recommendation and they have developed and put in place a system that meets the intent of our recommendations. And this recommendation, back in February 1999, was closed by the Board as Acceptable action.

The last five recommendations that I'd like to discuss deal with the fuel quantity indication system wiring and the energy sources that could be found in the tank through the wiring. These recommendations were issued in April of 1998. And Mr. Swaim has discussed in great detail the various aspects of these recommendations.

Two of them have been closed Acceptable action, because of the action by the FAA issuing AD's that require inspection and testing to verify wiring and tubing components inside the center fuel tank.

We have a number of others. There are four others. One that deals with the survey of fuel quantity indication systems on 747's equipped with systems other than Honeywell Series 1 through 3 probes, that is still in the open-Acceptable status. The FAA has issued an NPRM, and this NPRM would also require the development of specific fuel tank system maintenance and inspection instructions. That is still ongoing, and we're waiting for the final rule.

In addition, the recommendation asked the FAA to require research into copper sulfide deposits. And Mr. Swaim has gotten into that pretty heavily. And what staff is proposing is that this recommendation be superseded as Acceptable-closed and that a new recommendation, that we now have some more information on it and we'd like the FAA to do a little bit more, so we're saying let's close this one and get into additional work through a new recommendation that will be discussed shortly.

There are two others. One was open-Acceptable and one was open-Unacceptable. And they dealt with Boeing 747 airplanes and other airplanes with fuel quantity indication system wire installation that are co-routed with wires that may be powered. We asked for a physical separation of these wires. And the other recommendation asked that the FAA require surge protection systems.

While there have been a number of AD's that have been issued that deal with both the 747 and some of the 737's, the NPRM that led to the AD included a requirement for transit suppression systems as an optional, or additional, requirement to the shielding and separation. But the requirement at that time that the AD was issued, this particular technology was not fully developed for application to in-service airplanes.

The FAA has since determined that transit suppression technology, if it's executed properly, will adequately ensure that high-energy signals are not introduced on wiring inside of fuel tanks. And the FAA is continuing to evaluate this technology. And because of that new work that they're doing, the staff is proposing that this recommendation be reclassified from open-Unacceptable to open-Acceptable.

So, in summary, 10 of the 11 recommendations are in an Acceptable status, either open or closed. And it's only that one dealing with the temperature probes that may be overcome by events if and when the FAA requires certain operational changes. So more than 90 percent of the recommendations that we made in the course of this investigation are in an Acceptable status, Mr. Chairman.

JIM HALL: Now, Mr. Sweedler, since we will probably be taking action shortly on some additional recommendations, I would like to ask your successor to try to -- I will contact Administrator Garvey and see if, within a 90-day period of time, I can set up a meeting with her and we can get a status on all of these outstanding recommendations in a window of 90 days from now to provide that information to the families and to the public.

BARRY SWEEDLER: Yes, Mr. Chairman.

JIM HALL: Any other questions for Mr. Sweedler?

(No response.)

JIM HALL: Any other questions that any of the Board members have on the report?

Dr.. Loeb, do any of the staff have any other comments on the report?

DR. LOEB: No, sir, I do not think so.

JIM HALL: I think we have covered all the items in contention in some areas as well as the full report of the various agenda items that you laid out, and we have a number of conclusions. There is quite some work to do here. We've got conclusions. We've got a probable cause. And we've got recommendations to consider. But let's take a 15-minute break, and then we'll come back and conclude this meeting.


JIM HALL: [Starts already in progress on videotape] -- presents an avoidable risk of an explosion.

Number 22, the placement of heat-generating equipment under a fuel tank containing Jet A fuel can unnecessarily increase the amount of time that the airplane is operating with a flammable fuel-air mixture unless measures are in place to either, one, prevent the heat from entering the center wing tank or, two, eliminate the flammable vapors inside the center wing tank.

Number 23, the condition of the wiring system in the accident airplane was not atypical for an airplane of its age, and it was maintained in accordance with prevailing accepted industry practices.

Number 24, until recently, insufficient attention has been paid to the condition of aircraft electrical wiring, resulting in potential safety hazards.

And, number 25, the issues defined in the Federal Aviation Administration's Aging Transport Nonstructural Systems Plan are important safety issues that need to be addressed through appropriate changes, including rulemaking.

Those are the 25 findings. The probable cause is as follows -- the proposed probable cause:

The National Transportation Safety Board determines that the probable cause of the TWA Flight 800 accident was an explosion of the center wing tank, resulting from ignition of the flammable fuel-air mixture in the tank. The source of ignition energy for the explosion could not be determined with certainty. But, of the sources evaluated by the investigation, the most likely was a short-circuit outside of the center wing tank that entered the center wing tank through electrical wiring associated with the fuel quantity indication system.

Contributing factors to the accident were:

One, the design and certification concept that fuel tank explosions could be prevented solely by precluding all ignition sources; and, two, the design and certification of the Boeing 747 with heat sources located beneath the center wing tank with no means to reduce the heat transferred into the center wing tank or render the fuel vapors in the tank nonflammable.

You have heard the 25 findings and the proposed probable cause. Is there discussion by the Board?

(No response.)

If not, do I hear a motion that the findings and probable cause be adopted?


JIM HALL: Moved by Member Hammerschmidt.


JIM HALL: Second by Carol Carmody. All in favor, please signify by saying aye.

(A chorus of ayes.) The Board unanimously adopts the proposed probable cause and findings.

The new recommendations:

As a result of the investigation of TWA Flight 800's accident, the National Transportation Safety Board makes the following recommendations to the Federal Aviation Administration:

Number one, examine manufacturers' design practices with regard to bonding of components inside fuel tanks, and require changes in those practices as necessary to eliminate potential ignition hazards.

Number two, review the design specifications for aircraft wiring systems of all U.S.-certified aircraft, and identify which systems are critical to safety, and require revisions as necessary to ensure that adequate separation is provided for the wiring related to those critical systems.

Number three, require the corrective actions to eliminate the ignition risk posed by silver sulfide deposits on fuel quantity indication system components inside fuel tanks.

Number four, regardless of the scope of the Aging Transport Systems Rulemaking Advisory Committee's eventual recommendations, address all of the issues identified in the Aging Transport Nonstructural Systems Plan, including the need for improved training to ensure adequate recognition and repair of potentially unsafe wiring conditions, the need for improved documentation and reporting of potentially unsafe electrical wiring conditions, and the need to incorporate the use of new technology, such as arc-fault circuit breakers and automated wire test equipment.

To determine whether adequate progress is being made in these areas, within 60 days, the FAA should brief the Safety Board on the status of its efforts to address all of the issues identified in the Aging Transport Nonstructural Systems Plan. Member Goglia.

JOHN GOGLIA: Dr. Loeb, I have a question on number four, the need for improved training to ensure adequate recognition and repair of potentially unsafe wiring conditions. Do you think it's advisable that we specify who that training should be provided to? The way it's written, I could provide that to my head of engineering, and that would satisfy the requirement.

BERNARD. LOEB: I understand. Do you want to suggest that we add who the training is to?

JOHN GOGLIA: For the persons accomplishing the work.

BERNARD. LOEB: Persons accomplishing -- how do you want to word that? I see your point.

JOHN GOGLIA: To make sure we reach down and get the people that are in the tank that should be able to recognize the problems, or in any area.

BERNARD. LOEB: Yes. Should we say to inspection and repair personnel?

JOHN GOGLIA: The maintenance personnel.

BERNARD LOEB: Training of maintenance personnel. Okay.

JIM HALL: Members of heard the revision proposed by Member Goglia. Is there any discussion?

(No response.)

If not, we will incorporate that additional language in recommendation number four.

CAROL CARMODY: Mr. Chairman.

JIM HALL: Carol Carmody.

CAROL CARMODY: I have a suggestion on recommendation number three. I know it's been modified, and I was suggesting a further modification. Because yesterday, when I asked about corrective actions for eliminating the ignition risk posed by silver sulfide deposits, I got the response there really are no -- we don't know what those corrective actions are.

So it seems to me it might be more equitable to ask FAA to require the development and implementation of corrective action rather than implementing something we haven't identified yet. It's not a large change, but I think it might make more sense.

BERNARD LOEB: I think that does make sense, Carol Carmody. I think that's a good change.

JIM HALL: Any comment on the revision proposed by Carol Carmody?

(No response.)

Has staff gotten the revision language as proposed?

BERNARD LOEB: Yes, we have.

JIM HALL: Any objection by the Board members, or more discussion?

(No response.)

If not, that will be incorporated in recommendation number three.

BERNARD LOEB: Just for the record, in four, the word "addresses" is plural. It should be singular. We're going to take care of that.

IM HALL: Additional discussion on the four recommendations proposed by staff?

CAROL CARMODY: Mr. Chairman.


CAROL CARMODY: One additional thought. I note that, in four, we're asking that the FAA should brief the Board in 60 days. And you mentioned earlier that you were going to ask the Administrator to brief you in 90 days on other -- I wonder if we might combine that and perhaps make it 90 for all?

JIM HALL: Yes. Let's say 90 days, and ask them to brief the entire Board on all of the efforts on the status of both these efforts as well as the outstanding recommendations.


J IM HALL: Any other comments on the four proposed recommendations by staff? Do any of the Board members have additional recommendations to propose? Member Goglia.

JOHN GOGLIA: Well, I asked staff's opinion on an additional recommendation. Does staff have an opinion?

BERNARD LOEB: We do not have the backup in the report right now to support this recommendation -- the recommendation that we're looking at, Member Goglia. That's not to say we can't develop it. We could develop it. But it does mean we don't have a conclusion that goes with it, and it would delay.

JOHN GOGLIA: I will withdraw that.

BERNARD LOEB: But what we will do is look at this and see if there is a possibility of developing an independent letter on this or incorporating it in something else. Because I think it is a good idea.

JOHN GOGLIA: Okay, I withdraw it. I don't have anything additional, Mr. Chairman.

JIM HALL: Very well. Any others?

(No response.)

If not, I think we have adequately reviewed, Mr. Sweedler, have we not, the outstanding -- the previously issued recommendations? And therefore I will, in the interest of time, not read all of those recommendations again, but they will be included in the report as part of the record with the report. Any other comments before we move to a vote?

(No response.)

If not, I would entertain a motion that the recommendations and the report be approved.

JOHN BLACK: So moved.

JIM HALL: Moved by Member Black.


JIM HALL: Seconded by Member Goglia. All in favor, please signify by saying aye.

(A chorus of ayes.)

And therefore this TWA four-year report is approved.

I would ask the members if they have any -- well, first, I would go to the staff, if they have any closing comments. Dr. Loeb?

BERNARD LOEB: Having unfortunately not heard what the Board members think in terms of the written product, I'm going to just assume that you think that this report was well written and edited. And the credit to that goes to Karen Bury, Jodi Moffett and Kristen Sears. Kristen has her work cut out, because she has to do a final edit before it can go out. There are some typos and things that need to be taken care of, but I would just like to acknowledge their work in putting together what I personally think is a report that this Agency could be proud of.

JIM HALL: Thank you, Dr. Loeb. I have a closing statement in which I intend to acknowledge a number of people, but first I would like to give the floor to the members for any closing comments they might have, beginning with Member Hammerschmidt.

JOHN HAMMERSCHMIDT: I would just like to close my thoughts on this investigation by saying that, yet again, we have an event that was so very tragic yet from which we can extract a lot of safety improvements in the years ahead. And improvements have already begun concerning what we've learned from this accident investigation.

My congratulations go out to staff for a very excellent report and for presentations at this Board meeting that were superb. I would just like to say, in that regard, that this will be Barry Sweedler's last Board meeting. And I certainly wish to congratulate him on a long tenure here at the Safety Board. And I wish to thank him for all his service and support through the years. I wish him all the best in the future. Thank you.

JIM HALL: Member Goglia.

JOHN GOGLIA: Well, this certainly was a tragic accident that cost 230 people their lives and caused so many other people considerable pain and suffering. And I don't think that any of us sitting in this room can really appreciate the feelings that those people have experienced over the last four years.

Over the last two days, we have discussed a number of issues that we believe caused the accident. And these same factors existed on many other airplanes today. This wasn't an accident of one manufacturer, because it could have been any manufacturer. This wasn't an accident of one airline, because it could have been any airline operating in this environment with these same conditions existing.

The staff -- Bernie, your staff -- has done an outstanding job in delivering to the Board a very thorough report that I believe all of us can be proud of.

As a former mechanic, I can recall the feelings that I experienced after my first accident. I can still feel it. That pain doesn't go away. Many in this room feel that over and over and over. It does take its toll on us.

To that effort, towards the end, with the reconstruction of the airplane in New York, we have available to us a tool that's unequaled in the world. I have brought hundreds upon hundreds of people, as you know, to that facility so that they could feel the pain.

It affected me my whole life. It never changed. I have heard from some of those people that I brought that it has affected them as well. It changes the way they think. It changes the way they're doing their job today. That's a major plus. We need to share that with everybody, everybody, in the aviation community.

We will have failed in our efforts if we don't take that airplane and put it to good use. And I know the Chairman is going to talk about that. And we must not let this airplane go away. We must use this as a teaching tool for everybody in the aviation community -- everybody.


Much of the work that you have done has already seen some progress, as we've said here. The FAA, the industry, they've all moved forward with a number of issues. Even some of the airlines, on their own, have moved forward. I have a letter that was sent to me, a personal note, sent from one major airline. Just the changes that they have instituted on this action are considerable, considerable.

The real challenge for all of us, everybody in this room, the real challenge is to not let this work go away, to not forget the lessons learned. We must continue, with our diligence -- and I know our staff has the diligence -- to continue with the diligence to make sure that every single item that was uncovered is brought to a very satisfactory conclusion.

I thank every one of you for the work that you've done.

JIM HALL: Thank you. Member Black.

JOHN BLACK: That's sort of a tough act to follow. I started by making a list of people that I wanted to express my appreciation to for their parts in this. And then it got so long that I decided that I wouldn't do that. But I would share with you visions that I had during this investigation.

I tried to spend a few days a month during the most active part of this investigation up at Calverton. And what I have are slide shots in my mind of Jim Wildey lying or crunched up in his 6'7" frame under something, trying to trace fracture faces for the sequencing; David Mayer struggling initially with reams of paper and computer printouts on the documentation system, and then later the witness issue; and Al worrying about everything; and Bob Benzon and Merritt Birkey; and of course Swaim climbing all over and into and out of everything, looking for wiring pieces. I'll just mention them because those are the slides that come back in my mind.

This is a Herculean effort on the part of the Safety Board's technical staff, and it's probably the largest forensic investigation in history, and certainly the best. And I would like to thank them for it.

JIM HALL: Carol Carmody.

CAROL CARMODY: Thank you. I was not on the Board when TWA 800 occurred. I was in Canada. And I remember watching with keen and terrible interest all the developments. Because of my contacts with the FAA and the Board, I was able to follow, at a distance, the investigation. I'm glad to be a part of this now, on the Board.

I've been so impressed with the work that's been done, both previously as well as today. I've been impressed with the commitment of the Board members, with the commitment of everybody to do the best and the most thorough and the fairest job possible. And I think we've done that.

I think this accident was indeed a signal that something needs to attention that perhaps had not gotten attention. It has gotten a great deal of attention since the accident from the FAA. I give them full credit for all their work. I think we still want them to do more, but they've made good progress, and my congratulations to them.

Once again, I think all of us who are here today will remember this and will commit to ensuring that this does not get forgotten and that the work continues. Thank you.

JIM HALL: Thank you. Over the past two days, you have heard the NTSB staff presented a great deal of complex technological information on the investigation into the crash of TWA Flight 800. In the final analysis, what it tells us is that the 230 men, women and children on board TWA 800 lost their lives not as a result of a bomb or a missile or some other nefarious act but as a result of a tragic accident, an accident triggered by an ill-fated sequence of events in the center wing tank of an aging aircraft that was set in motion years before by the manufacturer's and the FAA's design and certification philosophy.

What is also evident is that we have learned a great deal over this past four years, especially about fuel flammability, possible ignition sources, and the wiring in aging aircraft. And in the four years since the accident, much has been done to improve the safety not only of the 747 fleet but of other aircraft as well.

In December of 1996, the Board recommended that the Federal Aviation Administration study design changes to deal with heated flammable vapors in aircraft fuel tanks and that they require operational changes to enhance the safety of those tanks. In April 1998, the Board issued another set of recommendations, focused on aircraft wiring and the fuel quantity indication system. Today, we made four more recommendations that we believe will enhance the safety of commercial aviation.

The FAA has taken action on a number of our recommendations. They have issued almost 40 airworthiness directives on fuel quantity indication systems wiring separation and inspections and modifications of fuel pumps, fuel quantity indication system wiring, the wiring in fuel tanks and fuel pumps. They have also proposed a special aviation airworthiness directive to require a review of existing airplane fuel tank system designs and mandatory fuel tank maintenance and inspections. And they are also looking at the issue of aging wiring.

I am gratified to see that they are doing also a lot of work in the area of flammability reduction, including looking at minimizing flammable vapors and ground-based inerting. I trust that the FAA will be equally responsive to our new recommendations.

But what I hope most is once the spotlight has moved away from this accident, that the FAA's leadership will ensure that the promises that have been made to the American people, to the traveling public, and to the TWA 800 families are fulfilled.


I recently came across the following quotation:

"It is recommended that every effort be expended to arrive at a practical means by which flammable air vapor mixtures are eliminated from fuel tanks."

This was presented to the FAA by our predecessor agency, the Civil Aeronautics Board, on December 17, 1963, following the crash of Pan American Flight 214, near Elkton, Maryland. Almost 40 years later:

"It is imperative that at long last the aviation community move with dispatch to remove flammable fuel-air mixtures from the fuel tanks of transport category aircraft."

But these safety improvements are not the only improvements to come out of this tragedy. In the months following the crash of Flight 800, the President and Congress gave the National Transportation Safety Board the important mission of being the Federal coordinator for services to the families of the victims of aviation disasters. Because of that action, all airplanes flying in the United States must have a family assistance plan in place that must be activated following an accident.

Taking care of families is now at the forefront of airlines' planning, not an afterthought. And as a result, families are now treated with the compassion and care they deserve, treated the way any of us would want to be treated were our families to experience a similar disaster.

Families in other transportation-related accidents have also benefited from this change. The Board now launches family assistance personnel to all accident scenes, including, most recently, the pipeline explosion that occurred in New Mexico.

Before I continue, I want to recognize Peter Goetz and Betty Scott for stepping forward at my request long before the legislation was passed to provide much-needed assistance to the TWA 800 families.


In addition, I would like to also recognize two individuals who are no longer with the Board, as Peter is. Betty is the only with us, regrettably. Matt Furman, Public Affairs Officer, and Liz Cotham, a former Family Affairs Specialist. I think all of those individuals received a special mention to me from the families.

This Board meeting, of course, does not mark the end of the TWA Flight 800 lessons to be learned and implemented, because we believe the reconstructed wreckage can provide invaluable insights for aviation professionals. The Safety Board has, over the years, allowed maintenance professionals to visit that hangar at Calverton to view the wreckage and discuss information gained during the accident investigation. In fact, Member Goglia has personally briefed more than 400 maintenance personnel at Calverton.

In the future, I hope that the Safety Board can inaugurate its own training academy. That proposal is on the street now. The centerpiece of which will be the Flight 800 reconstruction, so that future generations of aviation professionals and accident investigators from around the world can learn the lessons that it has to teach.

Before I close, I also want to express my appreciation and admiration to the Safety Board staff who worked so hard for so long on this investigation. Your dedication and tenacity not only produced an exhaustive, well-researched report, but they have also helped create a safer aviation system.

I know that everyone involved with this investigation, the investigators, the writers and editors, the Public Affairs staff, the Families Affairs staff, the Office of Research and Engineering, under the leadership of Dr. Vern Ellingstad, and the management and administrative staff, all sacrificed untold hours away from your families, often missing weekends, holidays and special family occasions so that we can complete our mission to ensure that the traveling public in our country is safer.

Every office and every person at the Safety Board was touched by this accident, and contributed in ways, untold ways, to this investigation. So I want to thank all of them for your selflessness, energy, integrity, and determination to find the cause of the accident and to prevent similar accidents in the future. I truly believe these individuals are outstanding public servants. They are paid by you and they work hard for you.

But I want to especially thank Dr. Bernard Loeb for his leadership in this investigation. Bernie directed the activities of this staff as they pursued every theory under investigation in this accident during a period of unprecedented activity for the Board. As many of you may remember, just two months before Flight 800 crashed, ValuJet Flight 592 crashed in the Everglades. And since then, we have had a number of other accidents, all of them tragedies, including Korean Airline Flight 801, American Airlines Flight 1420, Egypt Air Flight 990, Alaska Air Flight 261, two different cargo plane crashes -- one in Miami, one on the West Coast, and the JFK Junior and Payne Stewart accidents.

The NTSB and the American people are very fortunate to have someone of Dr. Loeb's caliber, his integrity and his ability to be in charge of the Office of Aviation Safety.

And I also want to join the other members in acknowledged Barry Sweedler's last Board meeting. Barry is retiring, as has been mentioned many times now, after 31 years of dedicated service. And, Barry, I want you to know that we wish you and your family godspeed as you move to the West Coast.

On a personal note, as you know, you may have seen the Chairman leaning back in his chair or turning around occasionally. And that is because most of the brains in my office sit behind me at these meetings. And that is my Special Assistant, Lieutenant Colonel (Retired) Deb Smith. She has been invaluable, and I am grateful for her efforts and assistance throughout this investigation.

I want to pause just a minute and let you see the names of the other individuals here at the Safety Board who have been an integral part of the investigation. I also want to thank the parties to this investigation for their persistence and their assistance. The expertise and advice provided by the parties have been invaluable over the years. And I believe that this investigation has shown once again that this party system works and that the participants in our process have much to contribute in developing the factual record of an accident and enhancing the analytical process.

I would like to thank the many individuals from the various industrial, academic, government, and military organizations that assisted us both on scene and throughout the investigation for their contributions to this effort.

I want to acknowledge and thank the members of the media who covered this accident and cover the work of the National Transportation Safety Board. We are certainly fortunate to live in a country where we have a free media and a free press.

And, finally, of course, I want to thank this audience for being attentive and respectful during these proceedings. And, of course, in closing, I thank the Flight 800 families for your patience and your understanding as the Safety Board worked over the past four years to discover the cause of this accident.

I know that this Board may not bring you the closure that you seek, but we certainly hope that you will be able to move forward. Nothing we can do here, obviously, can compensate for the tremendous loss you have suffered. However, I hope that the knowledge that we know what happened and are working to ensure that it will not happen again will provide you some comfort.

Again, this then completes the four-year investigation into the tragedy of TWA 800, and I adjourn this Board meeting.

(Whereupon, the meeting was adjourned.)