This is an old accident, but recently at work I was asked to evaluate the suitability of the type involved for our operations and turned up this account of the accident while looking for evidence I could include in a package to support my recommendation. I didn't get very far into the article before it made me angry.
Two candidates were undergoing flight skills tests. On the day before the accident the first candidate was asked to demonstrate a stall. But the poor weather conditions - turbulence, rain and low cloud - tended to trigger the stick-pusher, and the commander pulled the stall-avoidance system circuit-breaker to prevent nuisance activation.
Now, deactivating stall protection for flight training purposes is not an automatic bad thing. You don't want the trainee to learn to ignore the warnings. A loud stall horn degrades communication in the training environment. The candidate should not become dependent on the a stall warning light to recognize the situation or on a stick pusher to effect the recovery. (A stick pusher is a device that, on detecting an impending stall, automatically applies control force to lower the nose. This simultaneously alerts the pilot to the situation and starts the recommended recovery procedure). Some measure should be taken to make damned sure that CB gets reset before the next flight, but I'm willing to give them the benefit of the doubt here.
But who does stall recovery practice in an actual aircraft in bad weather in IMC? Is this considered normal somewhere?
"The candidate found this exercise frightening as she experienced great difficulties, having to use all her available physical strength to regain normal flight with the engines on full power and in [instrument] conditions," said Norwegian investigation authority SHT.
No shit she was scared. She might have been fired had she refused, or simply marked as someone who didn't have the right stuff. Too bad she or the next candidate didn't refuse, because the next day they died doing it.
For the second candidate's test the following day - also in instrument conditions, and with stronger winds - the examiner instead requested slow flight up to the first indication of stall, and a recovery with minimum loss of altitude. The stall-avoidance circuit-breaker had not been reset.
While the commander added power and retracted flaps at the candidate's request, the pilot "lost control of attitude and airspeed". Altitude increased by 200-400ft (61-122m) and airspeed dropped to just 30kt (56km/h), even though the stall warning activated and full power was applied.
Just 37s after the control loss, and with an eventual sink rate of 10,000ft/min, the turboprop hit the sea in a near-horizontal attitude, killing all three on board.
I didn't recommend the airplane as suitable for our operations, but not because of this accident. Any airplane will stop flying if you get the angle of attack high enough and don't break the resulting stall. I have a theory that the more comfortable it is inside an airplane the easier it is to forget that you have to fly it. That doesn't stop me being happy that we've finally got the heater working well, though.