The problems we all know about were because minimising height loss was put first above good stall recovery practice, not because it was considered.
While the subject is centred around light training aircraft, it is instructive to compare stall recovery technique in a transport jet type such as a 737 (ie not FBW) at high altitude and at low altitude. I am talking about what should be taught in a full flight simulator.
At high altitude cruise (say 37,000 ft) the approach to the stall is preceded by heavy buffet and I mean really strong buffet. You cannot miss it. The recovery should be started then and not wait until the stick-shaker.
Assuming however for the purpose of the training exercise the recovery is started at the stick shaker.
From the Boeing 737 FCTM: Quote:
"High Altitude Recovery. At higher altitudes above 20,000 feet, the airplane becomes increasingly thrust limited. If an approach to stall indication is experienced, nose down elevator and stabiliser trim is required to initiate a descent. This is because when the airplane is thrust limited, altitude needs to be traded for airspeed. Therefore a recovery at high altitude results in a greater altitude loss than a recovery at low altitudes". Unquote.
The nose should be kept just below the horizon while accelerating to a safe recovery speed with the aim to recover to a safe speed before levelling out. As thrust is increased forward elevator and stabiliser trim is needed to keep the nose on or just below the horizon. A typical example of a safe speed to attain before stopping the descent is the published high altitude holding speed which approximates full flap landing speed plus 100 knots. Typically around 230 knots IAS at high altitude. Expect an altitude loss of 3-4000 feet before this speed is attained.
It is a different recovery technique altogether when practicing a low altitude stall since it very much depends on the aircraft configuration at the time the stall occurs. Readers may recall the Turkish Airlines Boeing 737 accident at Amsterdam where a defective radio altimeter caused the autothrottles to retard prematurely during a ILS coupled approach.
When this happened, the autopilot attempted to maintain the iLS glide slope by raising the nose. With both thrust levers at idle, the speed reduced until the stick shaker operated at around Vref minus 30 knots (roughly 105 knots IAS). By then the altitude was about 1000 feet agl. The autopilot stabiliser trim wound steadily back while all this was happening as the autopilot attempted to maintain the glide slope. At the point of stick shaker the stabiliser trim was nearly fully aft. A successful go-around could have been made but the pilot was too slow to react and the aircraft stalled wings level into a field.
The recovery technique used in the simulator for this type of event is to disconnect the autopilot and autothrottles and apply full manual power. At the same time, reduce the angle of attack by lowering the nose sufficiently to unstall the wings. Lowering the nose too much at that altitude will cause significant height loss with no room to recover. Adjusting the nose attitude to around 5-8 degrees nose up once the stall is broken, permits a slight rate of climb while accelerating towards first flap retraction which occurs after reaching Vref.
The pitch up moment is very strong with full thrust. At full thrust settings and very low airspeeds, the elevator, working in opposition to the stabiliser, has limited control to reduce the pitch attitude. This must be countered by forward elevator and immediate forward stabiliser trim otherwise there will be insufficient elevator control to prevent the nose from pitching up under the influence of high thrust. In the simulator example described above, where the autopilot steadily applied almost full back stabiliser in its attempt to maintain the ILS glide slope, it was found that up to eight seconds of continuous forward stabiliser movement may be needed to retain normal elevator control. If full forward elevator combined with continuous forward stabiliser trim does not prevent further pitch up, and loss of control is imminent, consideration should be given to reducing thrust to help lower the nose. If that has only limited effect, the following extract is applicable from the Boeing 737 Flight Crew Training Manual:
Quote:
" If normal pitch controls do not stop an increasing pitch rate, rolling the aircraft to a bank angle that starts the nose down should work. Bank angles of about 45 degrees, up to a maximum of 60 degrees, could be needed. Unloading the wing by maintaining continuous nose-down elevator pressure keeps the wings angle of attack as low as possible, making normal roll controls - up to full deflection of ailerons and spoilers - may be used. The rolling manoeuvre changes the pitch rate into a turning manoeuvre, allowing the pitch to decrease. The reduced pitch attitude allows airspeed to increase, thereby improving elevator and aileron effectiveness . After the pitch attitude and airspeed return to a desired range, the pilot can reduce angle of attack with normal lateral flight controls and return the aircraft to normal flight". Unquote.
In fact, this technique is applicable to most light training aircraft, where, with full flap extended on final approach, application of full throttle in a low level, low airspeed go-around may cause a marked pitch up that could progress to a stalled condition unless immediate action is taken prevent the pitch up.
Obviously in the case of the Boeing 737 under discussion, the aim is not only to recover from the stalled condition with minimum loss of height but to accelerate from Vref minus 30 knots at the point of stick shaker, to a safe airspeed where the first flap retraction sequence is started. Normally that would be at Vref speed, typically around 140 knots in the 737 Classics where a normal go-around procedure then follows. Flown correctly from the point of stick shaker to a safe recovery results in a height loss of around 300 feet.