Some here have offered the notion that the THS should stop trimming at some point but what are the all the effects of such a design change? Should it be stopped at a stall indication? In these two accidents that would have been too late as the THS was already at the NU stop. Ironically, for AF447, stopping the autotrim function at that point would remove the ability of the autotrim to follow a ND stick order and reduce from -13.6 back to normal as the trim did in our little sim exercise, as it would have for AF447.
Mhmm, are you sure?
I agree for Perpignan's A320, but not for AF447: In the latter case, the trim was -3° when the stall warning went on, and the motion toward -13° was done with the stall warning shouting all the long.
Still, as stated previously, I'm no more advocating that inhibiting the NU autotrim when S/W in ON would be such a good idea:
1/ Someone quoted the "black swan" theory, and that's worth to think about (plus the fact that under the current implementation, the elevators may try to compensate for the "unavailability" of the autotrim).
2/ As you said, that would be of no use at all in a Perpignan-like scenario.
KISS principle. If autotrim available ("normal" scenario, the crew must be aware of that, and of how it works). If autotrim not available, the crew must be notified of that (USE MAN PITCH TRIM must be displayed, which was not always the case in the system logic, IIRC that has to be (has been?) corrected).