Alternate anatomy of an air crash
What started the Pittsburgh USAir accident? That's something that has fascinated Paul Czsyz (pronounced sizz), an aviation expert at St. Louis University who worked for years for Boeing's archrival, McDonnell-Douglas. In common with many experts, Czsyz thinks a movement of the rudder started the crash. But while the National Transportation Safety Board has not completed its investigation, and thus has not offered its explanation, Czsyz is ready to offer his.

After studying the flight data, Czsyz puts forward this scenario for the crash -- a scenario that leads to some explicit recommendations.

1. Something, possibly a signal that the right engine had failed, caused the 737's rudder to go "hardover" all the way to the left. The plane has a big rudder, because it has only two engines, and it needs a large surface area to counterbalance the thrust of a single engine in case one fails. (Right away, Czsyz departs from other investigators in his focus on the possibility that the rudder was automatically commanded by the plane's software to go hardover. He asks: Does the autopilot software contain a "switch" that commands a rudder hardover?)

2. The plane started to "skid to the left" in response to the rudder movement.

3. The autopilot (defined) sensed a turn to the left. But because the autopilot was not linked to the rudder control, it did not "know" that the rudder was all the way to the left -- a highly abnormal position in normal flight. In other words, the autopilot did not "know" that an engine was supposedly out.

4. The autopilot moved the ailerons (control surfaces on the wings) to bank the plane and smooth out a turn it "assumed" was desired. (Airplanes bank, or roll, during turns to create a force that counteracts their momentum. Momentum causes any object to continue moving in the same direction, unless acted upon by another force).

5. This aileron movement accelerated the roll that was already occurring. "The autopilot put it into a roll that would give you a coordinated turn," Czsyz says, "except that it accentuated the roll started by the rudder."

6. Within about 10 seconds, Czsyz says, the plane had flipped over and was diving steeply. At some point, the pilots idled the engines, an action that "blows my mind," Czsyz says, since they needed power to pull out of the dive. He says the idling only makes sense as a desperation measure born of the pilots' realization that something was wrong with the autopilot -- indirect evidence that they thought the rudder hardover was a response to the loss of an engine.

7. But by then the plane was lost since there was not enough room to pull out of the dive. "If the pilots had recognized the trouble immediately, they would have needed at least 5,000 feet of altitude to pull out of the roll," Czsyz says. And since the roll started at 6,000 feet, they had already lost so much altitude that they were doomed.

There are 1 2 3 4 5 6 7 8 9 10 11 12 documents. (Glossary | Bibliography)