Manufacturer and Airline Dispute Cause of 2001 Crash

By MATTHEW L. WALD

Published: September 28, 2004

WASHINGTON, Sept. 27 - During the long investigation into the November 2001 crash of an American Airlines jetliner in Queens that killed 265 people, the airline and the plane's manufacturer have engaged in an unusually public dispute about what happened. Now the National Transportation Safety Board is poised to conclude that actions by a pilot were the main cause.

That conclusion would at least partly vindicate the manufacturer, Airbus, which has questioned the training the pilot received from American Airlines. But in a last-ditch effort, American is campaigning to add another cause: what it asserts was a failure by Airbus to be more forthcoming about earlier problems with a tail control on that model of plane.

The safety board announced on Monday that it would meet on Oct. 26 to approve its final report on the crash of the plane, an Airbus A300, which went down on Nov. 12, 2001, shortly after takeoff from Kennedy International Airport on a flight to the Dominican Republic. All 260 people on board were killed, along with 5 people on the ground in Belle Harbor, Queens, on the Rockaway Peninsula. At the time, New York was still reeling from the 9/11 attacks two months earlier.

The investigation has taken unusually long, partly because investigators first focused their attention on whether there was a flaw in the composite material that made up the vertical fin at the back of the plane, which was ripped off by the aircraft's strong side-to-side motion.

 Although the board does not fix blame, its findings can be influential in court, and American and Airbus have a continuing legal dispute about who will pay the damage claims stemming from the crash. For now they are being divided 50-50, according to one lawyer in the case, but that could change if one side persuades a judge that the other was primarily at fault.

 The board is juggling the relative importance of several factors. The leading one, by all accounts, is that when the plane encountered the wake of a Boeing 747 ahead of it, the co-pilot aggressively worked the rudder pedals back and forth. That threw the plane into an oscillation, with wider and wider swings left and right, until the pressure on the fin exceeded the limit it was designed to handle, and it snapped.

American is arguing, with some support from an independent study commissioned by the safety board, that the control system for the rudder differs from the system used on most other planes, and makes the plane more prone to such oscillations.

 Another potential factor is that for a time, American's pilot training emphasized use of the rudder in recovering from upsets.

American has been arguing to safety board members that an incident in May 1997 involving another American Airlines A300, Flight 903, should have offered clues that the plane was susceptible to oscillation. In that case, pilots slowed the plane too much and it "stalled," meaning that the combination of angle and speed made the wings lose lift. The oscillation occurred during a very rough recovery. But the investigation focused on the stall.

Bruce Hicks, a spokesman for American, said: "Airbus alone had the knowledge about the sensitivity of the system, and most importantly, the grave risk of what happens if you reverse the rudder on this airplane, and the propensity of this airplane for that to happen."

"Had they shared that knowledge with the N.T.S.B. and operators, including American, Flight 587's accident would not have happened, and 265 people would not have died," he said.

But Clay McConnell, a spokesman for Airbus, said the manufacturer had turned over all relevant data. "There were major concerns about structural damage on any transport category airplane that was put through those maneuvers," he said.

Mr. McConnell said the force needed on the pedal to move the rudder was in harmony with the force needed to move every other flight control in the cockpit. "There is nothing that is unique about the rudder system," he said. "It really is a red herring. If the pilot wasn't trained well enough to be familiar with the rudder control forces on that airplane, someone should ask who did that training."

The airline's arguments do not appear to have changed the working hypothesis at the board that the problem originated with the pilot. "There are 20,000 pilots, and only one did this," said one investigator, who asked not to be identified because the case is still open. "There are 700 of these airplanes, and only one had this problem," he said.