PJ2
Thank you for a very detailed answer. I understand the difficulty in explaining these systems in a clear and succinct manner. The paper manuals for the aircraft are hundreds of pages long. I'm going to take a little time to be sure I understand the available aircrew information. At this point, it appears that Airbus did an admirable job in their cockpit design save for the typical airliner aversion for displaying AOA. Can I assume that the manual trim wheel turns as the automatic system trims the aircraft?
That the THS began to trim NU to the point of "deep stall" (once the autopilot disconnected?) as a result of erroneous data signaling lower airspeed (requiring more NU trim) was not observed/noted in the other thirteen pitot/airspeed events which are listed at the end of the BEA Interim Report 2. I don't believe that the THS would be trimmed NU, while on autopilot, to the point where a deep stall condition would be latent with a resulting LOC once the autopilot disengaged. Just a thought....I may be misunderstanding your thought however! PJ2
Actually, the thought runs a bit differently.
If the initial airspeed errors generated were to the high side of the true value and did not yet exceed rejection criterea, the autothrottle system would throttle back to bring the aircraft back on calculated target speed. As the aircraft slowed, the auto flight system would trim nose up to keep the elevator centered and the aircraft on altitude. Suppose the aircraft lost perhaps 30 knots actual IAS before the airspeed rejection criterea activated and thrust lock activated, throwing the aircraft into alternate law. As long as the aircraft was in alternate law the THS would continue to autotrim and the flight control system would maintain aircraft attitude absent an input from the aircrew, however the aircraft would be seriously under-throttled and decelerating. When the aircraft then transitioned to Direct Law, auto trim would stop, but by then THS position could be very nose up. Of course, Airbus may very well have written protecting limits into their software that would stop this sequence before it progressed this far.
(Note: writing this at work so I don't have references at hand)
Thank you again PJ2 for your response. After all the posts on this and related threads, it is probably hard not to say, "Oh No! Here is another one of those."
Machinbird