As a new member in this forum it has been very interesting to follow this thread. As a former Q400 pilot this accident has really caught my mind.
I share the view with a lot of you guys that the icing does not appear to be the obvious reason. I would like to share the following thoughts:
The configuration before selecting gear down and flaps 15 is normally flaps 5, at 25t this gives a Vsr (stall reference) of 110 kts in non-icing cond and 102 kts at flaps 10 which I understand is where the flaps got before selecting flaps 10 again.
Given the additional 20 kts margin added by the SPMs when ref speed is on it makes sense that the shaker activated in the 130 kts region.
A normal reaction at when this happens would be to initiate a go-around which according to SOP is full-power/flaps 10 and at positive climb rate gear-up. A two-engine Q400 go-around requires quite a lot of pitch up to avoid flap overspeed and to stay close to V2.
An abruptly performed go-around would increase the AoA and could in this case trigger the pusher.
The Operators Manual which I assume follows bombardiers input has a little bit different wordings on stall-recovery and windshear recovery. Stall recovery is basically a go-around proc with level pitch while windshear has the wordings:
• Increase the pitch attitude by rotating the aircraft smoothly and steadily to achieve V2.
• Simultaneously, select go around, disengaging the autopilot and if required, advance power
levers to the mechanical stops to arrest descent.
• Trade speed for improved climb capability in order to prevent further altitude loss.
• Do not accelerate back to final approach speed.
• The stick shaker may activate at pitch attitudes lower than normal.
• Pitch attitude must be reduced just enough to silence the shaker.
Note that this escape procedure promotes pitch attitudes limited by stick-shaker instead of re-gaining airspeed.
Another thing that comes to my mind is that even though 134kts in flaps 5 configuration is a anormally low operating speed it is still equal to Vref flaps 5 non-icing. Personally I agree that some kind of distraction caused this, I still think it is of interest to note that the speed bugs are set manually and that even though I put increase ref-speed on I can faulty set the bugs for non-icing. The bugs for 25t non-icing would be set at Vref = 119kts and Vfto = 146.