Friday, February 11, 2000

ALASKA AIRLINES FLIGHT 261 - A Preliminary Analysis

Eighty-eight people needlessly gave their lives in the crash of Alaska Airlines Flight 261. Another eighty-eight who will be missed by their families, mourned by their loved ones, remembered in stunned disbelief. Why them? It just doesn’t seem fair that having enjoyed what must have been a joyous vacation in Mexico these marvelous people in the prime of their lives have been silenced forever.

Sound familiar? We uttered these laments just last year for two hundred seventeen in the crash of Egyptair 990, two hundred thirty-two in the crash of Swissair 111 the year before, and so on and so on, year after year. While the causes of these crashes are different, the myths surrounding them are the same.

The official investigation into this crash has just begun. A lot more needs to be learned, but enough information has been released so a preliminary analysis can be made.

Alaska Airlines Flight 261 was most likely caused by a stabilizer trim that got so far nose down that there was no longer sufficient elevator authority to counteract the nose down pitch. In simple terms that means that the leading edge of the horizontal stabilizer (that’s the little horizontal wing on top of the tail) moved upwards on its motorized jackscrew (it’s hinged at the back) so far that the movement of the elevators (they’re the little flaps that attach to the rear of the horizontal stabilizer on the tail) couldn’t pitch the nose of the airplane back up. The reason the airplane was seen to move so wildly in the air before impact may have been due to the crew trying to regain control even to the point of rolling inverted so they could use the down pitch in reverse (the pitch would tend to cause the aircraft to climb if the airplane were upside down). It may also have resulted from loss of a control surface or flap departing the aircraft on one side.

Pundits on television and in the press have all lauded the MD-83’s safety record. Indeed, like all big airplanes, the MD-80’s accident record is pretty good, but that’s not the true test of an airplane’s reliability. Each time there is a problem reported with the stabilizer trim it is an accident that didn’t but could have happened. There have been hundreds of reports of stabilizer trim problems and several Airworthiness Directives issued by the FAA to correct stabilizer trim problems that could result in loss of aircraft control. One of these AD’s was not yet complied with by Alaska Airlines, the one that requires an extensive inspection for corrosion. That’s because the FAA gave the airlines eighteen months to complete the inspections!

Just as it does after every airline accident, the FAA ordered inadequate inspections of a suspect part, this time the stabilizer trim jackscrew. Had the FAA ordered immediate inspections for corrosion instead of allowing eighteen months for compliance, damaged jackscrews and other failed or nearly failed components would have been immediately discovered.

The stabilizer trim system on the MD-83, and all DC-9 based aircraft like it, is a very complex yet redundant control system. As with every complex aircraft system, it requires constant maintenance and strict adherence to a principal that in aviation allows no exceptions. That principle is simple. Every airplane will telegraph its intention to fail long before the failure takes lives. Here the MD-83 and its predecessors for years warned the FAA, the aircraft manufacturer and the airlines that its stabilizer trim system needed attention, immediate attention, otherwise there would be a crash. Neither of the three entities who should have heeded these warnings did enough, in a timely way, to prevent the loss of eighty-eight lives.

The engineers at McDonnell Douglas built into the stabilizer trim system many redundancies so if a trim problem occurred, it could be stopped before the loss of aircraft control. There were switches on the pilots’ control yokes that could stop the trim motor. Suitcase handles on the control pedestal to apply a motor brake, a guarded switch to turn the system off, and three circuit breakers to cut off the power. There was also a standby trim motor to substitute if the primary motor failed. Either all of these systems failed (extremely unlikely) or the flight crew was troubleshooting a failing system in-flight and the stabilizer trim was allowed to progress beyond the point where aircraft controllability could be maintained.

Communications between the cockpit crew of Alaska Airlines Flight 261 and ground based mechanics in Seattle and Los Angeles reveal an effort to analyze the problems being experienced in the air with the stabilizer trim system while the aircraft was still under control. It had successfully taken off and climbed to thirty-one thousand feet. Control trouble was experienced and it descended to twenty-six thousand before experiencing uncontrollability out of seventeen thousand. Cockpit communications revealed thus far suggest continued troubleshooting, even when the aircraft was still under control. The system could have been shut off electrically before loss of control occurred and an emergency landing made even with the stabilizer trim inoperative, but after the stabilizer was allowed to run beyond the limit of elevator effectiveness, a safe landing is much more difficult. Since the accident, the aircraft manufacturer has warned all MD-80 series pilots not to troubleshoot stabilizer trim problems in-flight, but instead land at the nearest suitable airport.
Information now disclosed by the investigation reveals that the autopilot was shut off and the aircraft hand flown during much of the two-hour flight. That likely means that the stabilizer trim was not working properly, since MD-83s are otherwise flown on autopilot most of the time, especially at higher altitudes.

Airline pilots are retrained every six months. That’s why they are so good. We look at them in awe and respect them (and pay them pretty well, too) because they are , as we see it, all there is between us passengers and our worst nightmare come true, a crash. Unfortunately, recurrent training in an MD-83 simulator does not include troubleshooting a stabilizer system failure, except to recognize a runaway trim and deal with it before it gets to aircraft uncontrollability. No training is conducted to demonstrate loss of aircraft control or how to regain it if the stabilizer limits are exceeded. Thus, troubleshooting in the air a malfunctioning stabilizer trim system is like taking passengers on an experimental test flight! Aircraft maintenance is designed to be performed on the ground, not in the air. Malfunctioning flight controls are an emergency that requires an immediate landing at the closest airport, period! Unfortunately, without training to the limit, the pilots can’t know how dangerous it is to test a failed stabilizer system in-flight. Alaska Airlines failed to train, and the FAA failed to require it. This made victims of the flight deck crew, as well as the passengers of Flight 261.

What can be expected from this point forward? Alaska Airlines will engage in damage control, putting a sympathetic face on a very, very sad and preventable accident that took the lives not only of innocent passengers, but also innocent co-workers. The NTSB will raise portions of the tail and determine whether corrosion, a failed trim motor or failed safety devices allowed the stabilizer to run beyond controllable limits. The FAA, the agency in charge of safety, will fall all over itself defending the indefensible, which is its continuing failure to order correction of known aircraft problems in a time frame sufficient to prevent loss of life. The NTSB will criticize the FAA and make safety recommendations, which the FAA will either ignore or delay. The families of the victims will cry a lot, suffer disbelief and anguish, become angry and tormented that the Federal Government has let them down again, and file lawsuits to help assuage in the smallest ways their indignation over what has occurred again, needless loss of life.

God bless the memories of all aboard Alaska 261, God bless their families, their loved ones and friends. God’s speed to all those who investigate. We hope that others will be spared for the sacrifice of all that were not.

February 11, 2000

This preliminary analysis is offered so those interested can participate in a dialogue of information gathering in the hope that this tragedy will not fuel unwarranted fear of flying, one of the safest of all human endeavors.