Inquiry Into Flight 587 Crash May Take 6 More Months

After more than 11 months of investigation, some 30 hours of testimony and nearly 3,500 pages of documentation, the biggest question remains unanswered: Why did an American Airlines Airbus jetliner crash into a New York neighborhood a year ago?

It’s not that a list of suspects hasn’t been identified.

The National Transportation Safety Board hearings here last week focused on the actions of the co-pilot, the sensitivity of the controls and the training and advance warning given to pilots about aggressive flight maneuvers. But the board adjourned Friday with a caution that it could take an additional six months of investigation before the probable cause is officially determined.

The board usually concludes that multiple factors caused or contributed to an airplane accident, and the findings on the Nov. 12 crash of American Flight 587 probably will be no exception.

The Airbus A300 jetliner took off without incident from John F. Kennedy International Airport on a scheduled flight to the Dominican Republic. About two minutes after that, as it climbed out over Jamaica Bay, the big jet wobbled slightly as it flew through the wake generated by one of the wings of a bigger jet ahead of it, a Japan Air Lines Boeing 747.

Seven seconds later, the Airbus plane wobbled more noticeably as it encountered the wake from the 747’s other wing.

In what may have begun as an effort to compensate for this wobble, the A300 co-pilot, 1st Officer Sten Molin, apparently pushed hard on his control pedals, moving the big rudder on the tail sharply to the right and causing the plane to abruptly jolt sideways.

The rudder is a large movable panel on the tail. It is normally used only to keep the plane aligned properly on takeoffs and landings and in extremely rare instances of engine failure.

Perhaps trying to compensate for an initial overreaction, Molin appears to have moved the rudder back to the left. He only succeeded in making matters worse. He then apparently went right again, then left, then right one last time, generating such forces that the plane’s tail snapped off. It all happened in 7 1/2 seconds.

Flight 587 slammed into a residential neighborhood in Queens, killing all 260 aboard the plane and five on the ground.

Last week’s four-day hearing by the NTSB highlighted the design of the rudder system and the training American gave its pilots on using that system.

It boiled down to a battle between Airbus, the company that designed and built the A300, and American, the company that employed and trained the pilots who flew it. Caught in the middle was Molin, the man at the controls, and, to a lesser extent, the U.S. and French officials who had certified the A300 as safe for passenger flight.

The Airbus rudder controls are designed so that it takes less force on the pedals to apply full rudder at higher speeds than at lower speeds. Thus the pilot has little indication of how far the rudder has swung.

Did this design feature confuse Molin, leading him to overreact on the controls? Should the Federal Aviation Administration and French air officials have certified this design? Should manufacturers be expected to anticipate inappropriate control movements? Can any airplane be expected to withstand the forces generated by such movements?

A senior American Airlines pilot, John Lavelle, told the NTSB last summer that five years ago, while flying with Molin, he noticed that Molin used the rudder too aggressively during a takeoff.

Lavelle said Molin told him that American had trained him to use the rudder that way, and Airbus contends that American trained its pilots to rely excessively on the rudder.

American denies this. However, in 1997, four years before the crash, an American Airlines pilot, David Tribout, expressed concerns about the way his company’s pilots were being trained to use rudders, saying what they were being taught was “potentially dangerous.”

In 1998, reacting to a turbulence incident involving another A300 flown by American, Airbus warned the NTSB in a memo that rapid back-and-forth movements of the rudder “can lead to structural loads that exceed the design load of the [tail] fin and other associated airframe components.”

A copy of the memo was sent to American.

Officials for the NTSB and the carrier say privately that because there had been no crash owing to tail failure at the time, the memo was probably filed without attracting much notice. In any event, the warning about the rudder was not the main subject of the memo. Pilots were not told about it.

Since the crash of Flight 587, the NTSB has warned all airline pilots to go easy on the rudder.

The assumption thus far is that the rudder on Flight 587 moved in response to Molin’s pressing on the pedals. There is no evidence thus far that the control system malfunctioned, causing the rudder to move on its own.

Some concerns raised early in the investigation now seem to have been laid largely to rest.

Testimony during the hearing indicated that the composite material used to build the tail fin was as strong as the metal used on other planes. It failed only when subjected to stresses far above those experienced during normal flight. There is no evidence the tail on Flight 587 had been damaged before the crash.

The outcome of the NTSB investigation is of more than academic interest, especially to Airbus and American. Both companies have reputations to protect, and public perceptions could have an effect on the financial futures of each.

In addition, there is the question of liability.

A number of lawsuits have been filed on behalf of the families of the crash victims, and although the total monetary liability has yet to be decided in the courts, it could well exceed $100 million. Flight 587 was an international flight, and under international agreement, punitive damages are not considered.

Airbus and American -- or, more exactly, their insurers -- will probably pick up most, if not all, of the tab. Despite their sometimes acrimonious disputes at the hearing, they probably will eventually sit down quietly together and figure out how to split it.

In most such cases, the award to each litigant is to be determined through a formula designed to gauge the “value” of each victim. The computation typically depends on the victim’s remaining life expectancy, earning power, number of dependents and other related factors.

The NTSB isn’t interested in any of these monetary concerns, some of which are resolved before the board issues its final ruling on proper cause.

The agency’s objective is to prevent another similar crash. And to that end, its investigation will continue, as long as it takes, probably until sometime next spring.

“We will pursue every lead in our search for answers,” board Chairwoman Carol Carmody said Friday as she closed the hearing.