US Airways accuses Boeing: Says data withheld, may have prevented crash

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

US Airways accuses Boeing: Says data withheld, may have prevented crash

Friday, February 18, 2000 By JAMES WALLACE mailto:jameswallace@seattle-pi.commailto:jameswallace@seattle-pi.com

SEATTLE POST-INTELLIGENCER REPORTER

In documents unsealed this week, US Airways officials accuse Boeing of withholding vital information that could have prevented a disastrous air crash near Pittsburgh in 1994.

The accusations say the data, which were from Boeing's 737 flight test program and regarded a control problem, could have averted the crash of Flight 427, which killed 132 people.

Boeing strongly denied the US Airways allegations and said the information in question was made available to the Federal Aviation Administration when the airplane was being certified in 1984 for commercial operation.

The documents were part of a petition for punitive damages filed by attorneys representing families of two of the 132 passengers and crew who died when the 737 rolled over and dived into a gully near Aliquippa, Pa.

Though the two cases were recently settled out of court, the documents show the depth of the falling out between Boeing and US Airways, which had been a good Boeing customer before the Pittsburgh crash.

In 1998, the two companies settled a legal dispute involving the airline's cancellation of Boeing airplane orders in favor of planes made by Airbus Industrie.

An FAA pilot was in the cockpit of a 737-300 during a test flight when it experienced the aerodynamic phenomenon that US Airways now claims Boeing tried to hide, Craig Martin, Boeing's chief 737 spokesman, said yesterday.

"We have always been upfront and honest with the FAA," Mike Denton, former chief engineer for the 737, said in an interview with the Seattle Post-Intelligencer.

He said he was disturbed by the allegations.

"They paint a picture of The Boeing Co. that is so different than the company I work for," he said. "If that (allegation) were true, I would not work here."

After the longest crash investigation in U.S. history, the National Transportation Safety Board ruled last March that the probable cause of the crash of Flight 427 was a malfunction of the plane's rudder, the movable part on a plane's tail that controls the side-to-side movement.

A pilot banks or turns a plane by using ailerons that raise or lower the wings, not by moving the rudder.

A hydraulic valve that moves the 737 rudder apparently jammed, causing it to fully deflect in the opposite direction commanded by the flight crew, the board said.

But the board acknowledged that even after five years of investigation and testing there is no actual physical proof that that is what happened.

For Boeing, the 737 rudder issue will not go away.

The previously confidential documents about the crash were unsealed this week in a Cook County circuit court in Chicago. They included depositions from US Airways officials as well as confidential internal Boeing documents. Boeing said it decided to lift previous objections to the documents being unsealed "so that we can set the record straight."

Although the court record includes Boeing's legal response to the petitions, the company prepared for reporters a 16-page page paper titled "USAir 427 litigation: The Boeing Perspective." It contests, point by point, various allegations made by the lawyers for the two plaintiffs.

The two Chicago plaintiffs who had been seeking punitive damages recently settled out of court, and the judge never ruled on the cases.

One of the petitions was filed by the widow of Timothy McCoy of Chicago, a passenger on USAir Flight 427. The estate received a record $62 million in damages in the out-of-court settlement.

James Gibbs, who was flight manager for USAir at the time of the Pittsburgh crash, and Gordon Kemp, the airline's 737 flight manager, both gave depositions that are included in the now-public court file. (Since the crash USAir has changed its name to US Airways.)

If Boeing had not withheld 737 flight test data, Gibbs said in his deposition, the crew of Flight 427 would have successfully been able to fly out of the situation.

The Gibbs deposition was taken 10 months ago.

The flight test data in question involve the 737's flying characteristic at a certain speed and flap settings. Flaps, one of the moveable parts of a plane's wings, are extended upon landing to provide more lift at slower speeds.

Flight 427 was preparing to land at the Pittsburgh airport when it crashed. The jet was moving at 190 knots and the crew had just extended flaps to what's known as "flaps one," the initial deployment position.

During its flight test program on the 737 in 1984, Boeing found that at a speed of 190 knots and the flap-one setting, the plane could not overcome a full rudder deflection by using the ailerons on the wing, as would normally be the case.

"There is no FAA requirement that says this phenomenon is not acceptable," Denton said. In the wake of the Pittsburgh crash and subsequent investigation, airlines have instructed their pilots to fly the 737 at 10 knots faster when first deploying the flaps.

This keeps the plane above what's known as the "crossover speed" so the ailerons can overcome a full rudder deflection. The higher speed was officially adopted into Boeing's 737 operating manual last March.

Gibbs and Kemp said in their depositions that no one with Boeing ever told them the 737's crossover speed was higher than that being flown by USAir pilots at the time of the Pittsburgh crash.

A spokesman for US Airways said yesterday the airline had no comment on the depositions, but noted that the final report on the crash by the safety board supported the actions taken by the crew.

Denton, who is now an engineering executive with Boeing's product development team, said the information about the 737's crossover speed was not hidden.

"The information was out there," he said. "I don't know if the airlines saw the certification data or had access to it. But it was an observed phenomenon, one that was not viewed as unacceptable by the FAA."

He said there was a discussion about the 737 flight test data in the safety board's report on the 1991 crash of a United Airlines 737 as it prepared to land in Colorado Springs.

Initially, the board said no cause for that crash could be determined, though a rudder malfunction was suspected.

Last March, when the board issued its final report on the Pittsburgh crash, it said the probable cause of the United crash was a similar uncommanded deflection of the rudder.

But, according to Boeing, even if the USAir 737 was flying 10 knots faster, it probably would not have made any difference.

That's because the pilots had not been trained in the proper procedures to recover from the kind of upset that sent Flight 427 rolling over and into a fatal dive.

Denton said Boeing conducted simulator tests and found that pilots who waited three seconds to take appropriate corrective action in event of a full rudder deflection had a much more difficult time recovering the airplane than if they waited only one second.

"The added 10 knots didn't buy you a lot if you waited three seconds," Denton said. Boeing says officially that it accepts the safety's board findings in the Pittsburgh crash. But Boeing engineers such as Denton are not convinced.

"As an engineer it drives me nuts that we don't know what caused that full rudder displacement for a fact," he said. "We tried to look at all the data. We truly don't know."

Link

http://www.postintelligencer.com/business/boe18.shtml

-- Carl Jenkins (Somewherepress@aol.com), February 18, 2000

Answers

Again, probably not y2k-related...but an excellent description of how complex an airline accident investigation can become, especially when so much information is missing. Long ago I read an excellent series of articles written about what an investigator went through in trying to determine the cause of this crash; the articles are long, but well worth the read. Unfortunately, I no longer have the link.

Keep 'em coming, Carl. :)

-- Rachel Gibson (rgibson@hotmail.com), February 18, 2000.


Moderation questions? read the FAQ