NASA says it has a better jackscrew; others aren't convinced

WASHINGTON -- Engineers at Kennedy Space Center say they have designed a fail-safe jackscrew mechanism that could have prevented the crash of Alaska Airlines Flight 261.

They say it can be easily retrofitted into The Boeing Co.'s jets such as the MD-83 that crashed.

But the National Transportation Safety Board isn't so sure.

And Boeing says a new design isn't needed at all. The current one has worked for 37 years and about 100 million flight hours, and it would not have failed on Flight 261 had it been properly lubricated by Alaska mechanics, according to Boeing.

The design of the jackscrew mechanism, used in about 2,100 Boeing jets in service around the world, was one of the issues raised Tuesday during the NTSB's final meeting here on the crash of Flight 261.

The jackscrew assembly controls the movement of the stabilizer, the winglike structure on top of the vertical fin of the McDonnell Douglas T-tail airplanes. On Flight 261, the mechanism came apart because of a lack of lubrication, sending the Alaska jet into an uncontrollable dive into the Pacific Ocean off the Southern California coast on Jan. 31, 2000. All 88 passengers and crew members died.

Although the safety board turned aside a proposed recommendation from its staff that the jackscrew mechanism be redesigned, it did say the design was a contributing factor in the crash, as was the lack of maintenance by Alaska. It also recommended an engineering review to determine if a fail-safe mechanism is practical.

The jackscrew mechanism in question was designed by Douglas for its DC-9 and was subsequently used on the MD-80 series and still later on the MD-90 and the McDonnell-designed Boeing 717. Only the 717 is in production today, but combined, the models represent the second-most-popular family of jets in the world after Boeing's 737.

"We are satisfied with the (jackscrew) design and certification of that design and how that design has performed in service," said Ron Hinderberger, the director of airplane safety for Boeing, who attended the NTSB meeting. Flight 261 was the first in-flight failure of a jackscrew that led to a crash.

The assembly consists of a 22-inch-long screw made of hard steel that travels up and down through a large fixed nut made of softer copper alloy. Two electric motors, one operated by the pilot, the other by autopilot, turn the 1.5-inch-thick jackscrew. Operating much like a common garage-door opener, the threaded rod can push the horizontal stabilizer up 2.1 degrees from neutral or depress it as much as 12.3 degrees.

On Flight 261, the threads inside the nut were stripped away because of a lack of lubrication, according to the safety board. The jackscrew then tore through a mechanical stop, causing the stabilizer to move well beyond its design limits.

The safety board debated whether the jackscrew mechanism needs to be redesigned so that a single failure -- such as the nut wearing out -- could not cause a catastrophic event.

During the discussion, board member John Hammerschmidt said he had been made aware a day earlier of a fail-safe jackscrew developed by engineers at the Kennedy Space Center. John Clark, the board's director of aviation safety, told Hammerschmidt he was also aware of the space center jackscrew but had not been briefed.

NASA's jackscrew was developed for space shuttle operations, and details about the design are in a Kennedy Space Center technical bulletin issued in September. There, the engineers who designed the jackscrew mentioned Flight 261 and the potential application of their design for commercial aviation.

The Kennedy engineers originally wanted a fail-safe jackscrew to replace one now used on the service arm of the gantry at the space shuttle launch pad. Moments before launch, a jackscrew rotates an oxygen vent hood up and away from the shuttle. If the jackscrew failed, the hood would crush the shuttle's huge external tank as the shuttle climbed away from the pad, causing a catastrophic explosion.

Essentially, the jackscrew designed by the space center engineers uses a follower nut that rotates together with the primary nut. Should the primary nut fail, the load is transferred to the follower nut. A "mechanical or electronic indicator provides a non-intrusive measurement of the wear of the primary nut," the technical bulletin said.

On McDonnell Douglas jets, mechanics must check jackscrews for wear. This is known as an end-play test. But it can give different results depending on certain conditions and how it is performed. Boeing has designed a new end-play tool that gives more uniform testing results. It should be certified soon by the FAA for use by aircraft mechanics.

Although the space center engineers may believe their design could be the solution to the jackscrew safety issue, the NTSB isn't convinced.

"I'm not so sure you can take the concept and apply it (to planes)" said Jeff Guzzetti, a safety board accident investigator who headed the systems group during the Flight 261 investigation.

He pointed out that the single nut on the Douglas-designed jackscrew does not move. The jackscrew travels up and down through the fixed nut. On NASA's jackscrew, the primary load-bearing nut, as well as the following nut, move up and down along the jackscrew.

Boeing said it has not "scoped" out the space center jackscrew design. But the company maintains that a redesign is not needed. "It's not whether it can be done, it's whether you need to," said Boeing's Hinderberger. "It comes back to managing the aviation system. We believe we have done that."

And that system includes the need for maintenance.

Not only were the four board members divided on the jackscrew redesign issue, so was the NTSB staff. Clark, the aviation safety director, urged the board to recommend the jackscrew be redesigned to prevent a single-point failure -- "something to take the issue off the table," he said. "We are not looking for a major redesign, but something that is simple, effective."

Guzzetti, who works for Clark, disagreed. "I understand both sides of the issue," he told the board. "I don't think it's a simple fix. No one has studied it well enough. My intuition is that it isn't as simple as it sounds."

He said he had done a lot of "soul searching" on the redesign question, and decided that it was best to "give maintenance a chance."

"Give some credit to this design that has served airplanes well except for this one accident, when there was clear failure of processes," he said.