New Clues Deepen Flight 587 Crash Mystery

As work continued into the night on reading the plane's flight data recorder, investigators late Tuesday revealed several new clues about the crash of American Airlines Flight 587, but the information merely deepened the mystery of what caused the accident.

With both of the Airbus A300-600's engines and its vertical stabilizer coming down before the rest of the airplane, early accident-related speculation was on a catastrophic engine failure that triggered collateral structural damage. But investigators have found nothing that backs such a scenario, and the near-pristine condition of the tail pieces indicates that something besides debris caused them to separate from the plane.

Visual inspection of the plane's two engines showed "no physical evidence" of an engine or fan failure, U.S. National Transportation Safety Board (NTSB) Investigator-In-Charge George Black told reporters in the first of two briefings held late Tuesday. Black also confirmed what AviationNow.com reported several hours before the briefing: neither of the General Electric CF6-80C2 engines showed evidence of a bird strike.

"We will have to await tear-down to completely analyze the condition of the engines and any part they might have played in the accident," Black cautioned, adding that both engines will go to American's Tulsa, Okla., maintenance base for disassembly.

While an engine failure seems less remote based on the lack of evidence, other facts clearly indicate that something went suddenly and terribly wrong shortly after Flight 587 left Runway 31 Left at John F. Kennedy Airport Monday morning. All 251 passengers and nine crewmembers were killed when the jet plunged into a residential waterside area of Queens, New York.

In perhaps the most perplexing development Tuesday, the A300-600's vertical stabilizer and rudder were pulled from Jamaica Bay -- about a half-mile from the main debris site -- and both pieces "appear to be complete," Black said. Television images of the tail showed no marks, holes, or other structural damage that would indicate that the tail was knocked off by debris -- such as from a disintegrating engine.

Relying on what Black described as a "second quick look" at the cockpit voice recorder (CVR) tape recovered hours after the accident, investigators have determined the first part of the flight was "normal."

About three minutes into the flight -- or 107 seconds after power was applied to start the A300-600's takeoff roll -- an "airframe rattling noise" was captured on the tape, Black said.

Seven seconds later, one of the pilots mentioned a "wake encounter," but there was no further discussion, Black said.

Four seconds later, the first officer called for maximum power to the plane's engines. This could indicate several things, such as setting up engine thrust to help steer the plane -- an option to compensate for a lack of rudder control -- or an effort to climb in preparation for an emergency return to JFK.

At 127 seconds, one of the pilots indicated that control of the plane has been lost.

The vertical stabilizer and rudder were pulled from Jamaica Bay Tuesday, and both of them "appear to be complete," Black said. Television images of the tail showed no marks, holes, or other structural damage that would indicate that the tail was knocked off by debris -- such as from a disintegrating engine.

The CVR information alone doesn't reveal when Flight 587 took off or how long it stayed airborne, Black said. The CVR information will be merged with data from the FDR, air traffic control radar tapes, and other sources to develop a detailed look at the plane's flight profile.

The rattling noises captured on the CVR will surely get significant scrutiny from investigators. Black, who was relaying information from other investigators and has not listed to the tape, could not provide details on the noises.

Aside from the two rattling sounds, Black did not indicate that other unusual noises were captured by the CVR. In the past, onboard explosions -- such as what would occur due to an uncontained engine failure, fuel tank ignition, or bomb detonation -- have often been picked up on recorders.

While no sounds pointing to explosions have been picked up from the CVR analysis sessions, Black said investigators noted "noises" on an air traffic control tower tape of communications between controllers and Flight 587. The noises -- which Black did not describe in detail -- were heard at about the time crew lost control. He did not say what the sounds indicated.

Detailed analysis of the tapes and coordinated timeline of the events will help shed more light on how the clues fit together, Black said. Transcription of the CVR is slated to begin Wednesday, with participation of representatives from France's Bureau Enquêtes-Accidents as well as Airbus.

Flight 587 took off about two minutes and twenty seconds behind a Japan Airlines 747-400, Black said. While investigators will examine that fact in connection with the Flight 587 crew's reference to wake turbulence, Black noted that the separation between the two jets exceeded the two-minute requirement.

Evidence from eyewitnesses has provided some insight on Flight 587's final moments, Black said. Members of two airline crews that witnessed at least part of the American Airlines jet's final flight have been interviewed, and each pilot tells roughly the same story, Black said. Each pilot recounted that Flight 587 "wobbled," spiraled out of control, and went down at a severe nose-first angle, Black said.

Black said a construction worker shot video of Flight 587's takeoff roll, but turned his camera away as the A300-600's gear was retracting. Black did not indicate that the tape showed anything amiss with the plane during the departure, but investigators are hoping to talk with witnesses -- including airline crews -- that were on the ground as Flight 587 taxied out and took off.

Investigators have not determined if any debris was on the runway when Flight 587 took off, but Black said the board hopes to have more definitive information in the coming days.

All major parts of the plane have been recovered, Black said. The vertical stabilizer and rudder were pulled from Jamaica Bay. The rest of the plane, including both engines, came to rest on land.

Flight 587 took off to the northwest and investigators believe it completed about half a turn and was heading southeast when it began to break up. Black said the Airbus stopped sending transponder signals at an altitude of about 3,000 feet, indicating that something -- such as a major structural failure -- cut the unit's power. "Primary" target returns -- indicating aircraft or large pieces of debris without working transponders -- began at that point, Black said.

Tracing the plane's assumed flight path, the vertical stabilizer was the first piece of debris found, followed by the rudder about 200 yards further down the flight path. Both pieces were in the water, about half a mile from the main crash crater.

The plane's No. 1 engine ended up in a gas station parking lot about 700 feet from the crater, while the No. 2 engine came down in about 800 feet from the main crater. All of the debris is "more or less in a line," Black said.

A sweep of a 16-block area around the crash site turned up one additional piece of wreckage, and a helicopter flight over the area indicated that no major pieces remained in the shallow bay waters. Divers made 15 trips into the bay Tuesday and found nothing substantial, Black said.

Examination of the wreckage showed that the plane's flaps were retracted, which would be expected during a climb-out. The horizontal stabilizer jackscrew was in the neutral position, while the rudder trim actuator indicated a 10-degrees left displacement.

Black also released information on Flight 587's two pilots. The captain was a 16-year veteran of American and had 8,050 total flight hours, including 1,723 as the left-seat occupant on an Airbus A300-600. The first officer joined American in 1991 and had 4,400 total flight hours, including 1,835 as an A300-600 first officer.

Tuesday's progress meetings, attended by more than 100 investigators, included updates from most of the subgroups tasked with probing specific aspects of the accident. Progress reports from the maintenance, human factors, flight data recorder, and cockpit voice recorder groups are expected in the next day or two.

The maintenance group has been sent to Tulsa to review the A300-600's records, while the two recorder groups were working to extract detailed information from the plane's CVR and FDR.

could the pilot have mistaken vibration caused by the unusual movement of a loosened vertical stab or other component to be a "wake encounter"?

Although I'm not an air transport pilot (private with single and multi ratings), I would think the answer would be yes.

Black said the Airbus stopped sending transponder signals at an altitude of about 3,000 feet, indicating that something -- such as a major structural failure -- cut the unit's power. "Primary" target returns -- indicating aircraft or large pieces of debris without working transponders -- began at that point, Black said.

Has there ever been anything other than a explosion or on-board fire that caused enough damage to an airframe that power (and/or the audio input) to the cockpit voice recorder is lost before impact? Oh yeah, I forgot: an exploding center fuel tank. I wonder if anyone has the titanium balls to try that theory.

Bulkhead failure has caused loss of hydraulics and transponder previously. That failure can be caused by corrosion as well as the ways you discussed.

It has to do with the official start time of the flight...ergo...when it leaves the gate.

The takeoff roll begins when the airplane is lined up on the active runway, is cleared for takeoff, and power is applied. Has nothing to do with the gate.

Wake turbulance from a 747 can be very violent. Essentially a level 5 tornado is generated by each wingtip. The winds were not too strong that day, so the vorticies would have persisted.

I think you have it backwards. The v-stab is more like a kleenex, and the rudder is more like a pen. Throw both laterally, and guess which lands farther downrange.

So then the article was wrong. 3 minutes does not equal 107 seconds.

Wake turbulence is very strong, but its effect on an A300 shouldn't have had this impact. A G4 or Citation sure, but an A300?

Perhaps the heavier piece fell more quickly which is why it was the first to hit the water.

Huh?.

---max

Perhaps the rudder actuator stayed with the fuselage after the stab separated. The rudder actuator might have held the rudder with the rest of the fuselage for a short time after the stab was gone.

Yes. This noise thing has me troubled. On another thread, I was questioning why the persistent use of the term "rattle" in all the press accounts - like it's being used as a euphemism for something else more sinister.

Do the pilots out there know if "rattle" normally has a specific meaning when used by pilots?

I think "rattle" is being repeated by everybody since that is how the noise was described by the NTSB. Nobody else (reporting anyway) has heard the CVR to make an alternate description.

Doesn't necessarily need to be progressive failure because I don't think it would have been detected. the non-destructive inspection of airframes, especially with composite materials, is not done in a fashion to uncover these potential failures.

I might be wrong but I'm not a maintenance expert.

Now, my question is this...why no reports of eyewitnesses who spoke quite definitively about flames and smoke near the wings. Or did I miss this in the article. Again...only airline pilots and ground crews are credible eyewitnesses? Check out any news footage of interviewing the people who actually saw the thing go down. There's plenty of 'em. Everyone says flames and smoke.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.