First, with regard to Pool’s comment about groundings: we did effectively ground the ATR during the winter of 94 to 95 by prohibiting its operation in icing conditions until a redesigned set of boots had been retrofitted. I was there, and it was not a happy time, but it was the right thing to do.
This thread is so long that my next comments are statistically almost sure to be repetitive, so please forgive me if that is the case.
There may be some confusion about the nature of a stick pusher. FAA Advisory Circular 25-7A, Flight Test Guide For Certification Of Transport Category Airplanes, never refers to a stick pusher as a stall “avoidance” device. The promulgation of this myth in training may be part of the problem. Rather, the AC identifies the stick pusher as one of several types of stall identification devices. The role of the pusher as a stall identification device is thoroughly explored by Dave Davies in Handling the Big Jets.
A conventional marriage between airfoil section and configuration results in an early flow separation, consequent aerodynamic buffet (felt through the control column in the case of unboosted elevators), and a pronounced nose down pitch at the stall. More contemporary combinations of airfoil section and configuration, as well as wing sweep, may seriously compromise the traditional stall warning, identification and characteristics. There may be no early aerodynamic buffet, or due to powered elevators and/or T-tail, the turbulent flow may not be felt as clearly as need be. The actual stall may not yield a nose down pitch. Indeed, it may result in a nose-up pitch in some swept wing cases. Hence the stick shaker, which is intended to artificially replace the early aerodynamic buffet, and the stick pusher, which is likewise intended to replace the traditional pitch down of a conventional airfoil section. The general idea was to replicate the conventional stall behavior.
For a good review of this, take a look at FAR 25.201 and 25.203, AC 25-7A and, if you can get one, Davies wonderful book.
Davies makes one point that is very powerful and, in the Colgan case, poignant. On page 135, he says, “Get to know what the push feels like during training.” As I sat in the audience at the NTSB Public Hearing for the Colgan case, I listened to testimony to the effect that Captain Renslow had been exposed to the Saab 340 stick pusher, but not the Q400 pusher. I thought this to be very sad. In an earlier part of my career, I was a check airman on the Metroliner. The early Metro II’s would periodically require an inflight calibration of the stall warning horn, AoA indicator and stick pusher. This allowed me to develop some expertise in performing a sort of falling leaf at around 85 knots while the technician in the right seat rapidly twirled his little jeweler’s screwdriver to make the adjustments. As you might imagine, this system was soon replaced through an STC allowing bench calibration. But the experience with it led me to always demonstrate a stall right into the pusher during while training new captains and first officers. I like to think that everyone I trained would have recognized the pusher for what it was, and been comfortable managing it.
I cannot understand why the industry finds this so difficult. If we did it in the airplane, it can surely be done in the simulator.