Another Airbus loss of control incident

Posted on 15. Apr, 2010 by in Airline Safety, Blog

An A330 on Cathay Pacific (CX) Flight 780 experienced un-commanded power changes causing the crew to land at an excessive speed — nearly double the normal approach speed. The normal speed at their weight was reported to be 130 knots, the aircraft touched down at 239. The Rolls Royce Trent 700 engines are controlled by Full Authority Digital Engine Controllers (fly by wire). Throttles that do what they want despite the position selected by the operator; starting to sound familiar? Toyota maybe? Toyota announced today they are going to do an expansive test on their Lexus version of an SUV, I suspect they got a nudge. Quick action; I wonder if an extensive test will be done on fly-by-wire applications?

Unconnected to A330 problems of late? Maybe, maybe not; I think not. The FADECs are integrated into the ADIRU’s (aircraft computers) that control every aspect of flight. In the recent control excursions the pitot system has been blamed (AF, NWA, etc) and in the case of the QANTAS excursions the angle of attack system (AOA). Now we get FADEC (duel, very unlikely) failures. There is not a single point of failure; unless you dig a bit deeper and analyze where the information is processed and acted upon. The ADIRU #1. How is engine thrust related? If the ADIRU senses the aircraft approaching stall; taking input from speed (pitot) or angle of attack the engines will be directed to over-ride the pilot and run power up and if the AOA spikes the flight controls will be directed to push negative g. Ah Ha!

Remember the engines on Sully’s A320 would have still produced thrust, probably allowing him to reach an airfield. However the FADECs over-rode the crew and shut them down. Humans are smarter than computers, they understand the unquantifiable; for example it is better to burn engines up then go for a swim.

Back in the day, when we fought an F-16 slow we didn’t fight the aircraft (we’d loose) we fought the computer. It would over-ride the pilot exposing him to a guns shot. The FA-18 was different (also fly-by-wire); if you wanted to fly zero airspeed it would let you, the pilot had the over-all control. Not that I haven’t witnessed a fight for control between an ex A-7 pilot and his brand new F-18 (Hornet does a hands off catapult launch) off the cat. However, in general the Navy wanted the pilot to be the final arbiter. Personally I agree.

Tags: , , , , , , , , , ,

3 Responses to “Another Airbus loss of control incident”

  1. Rick 12 May 2010 at 08:31 #

    12 May 10 Another A330 down in Tripoli, and apparently in reasonably clear weather! Aaargh! It will interesting to see initial reports out of there.

  2. Simon 16 February 2012 at 06:25 #

    I kind of agree with you Chip, there are so many instances where errors in system designs have caused fatal results but we are in an evolving industry and there are I am sure countless unpublicised areas where automated soulutions have over-ridden inorrect pilot decisions.

    I remember vividly the Mulhouse A320 crash in which the causes have been disputed but what was notable was that the aircraft maintained a controlled attitude into the terrain that resulted in just 3 of the 130 passengers and crew surviving. If the pilot had been able to override the flight control system the aircraft would have stalled and probably resulted in the complete loss of all life.

    We have to accept that design errors will occur and the challenge for us all in the industry is to work as hard as we can to mitigate these during the development process.

  3. Chip 20 February 2012 at 13:44 #

    I think that the fly-by-wire system maintaining a level attitude is part of the problem in pilot recognition of an extremis situation. When I was in flight test I remember a near miss over the hill at Edwards. A C-17 was on a flight test when they radioed to the control room a total loss of lateral control and a rate of decent they could not stop with maximum power. All the engineers monitoring and the crew could not figure out a fix; they could not find a system problem. A T-38 from the USAF Test Pilot School was vectored over to give a visual inspection and see if the crew could identify any damage. After swooping by they radioed: “Dump the nose you are in a full stall.” Rumor has it a few descriptive nouns were included.

    It sounds ridiculous; a room full of Engineers monitoring every parameter, not to mention an augmented crew of Test Pilots on board. And yet they missed the very basics; the number one priority: keep the wing flying. Look at the Air France crash on the equator; eerily similar.

    Basic airmanship has become a casualty of automation. “Fly the damned airplane!” seems obvious; however some of the new systems simply won’t allow it. Often when flying the line, a runway change or some other situation, over tasks the crew in close. Click, click; auto pilot-auto throttles off, is normally the best solution.

    When I was flying old F-4 Phantoms as an Adversary against the Fleet, we had a saying: “We can’t beat the airplane, but we can defeat the computer.”

Leave a Reply