TM,
It should be made perfectly clear that many, if not most multi-engine accidents occur during training exercises. The first statement made by my ATO when I went for my META was,
"This is a whole new way for students to try and kill you."
I know it is probably not the light in which you were portraying the examples, but
when the instructor (30,000 hours) decided to cut the right hand engine mixture control
airspeed fell quickly well below best single engine rate of climb before the instructor could re-start the engine and re-introduce power.
Why cut mixture when you can just close a throttle and have the engine instantly available? Piper is one manufacturer who specifically warns against using
either mixture or fuel cuts to simulate engine failure. I will admit I was taught to conduct simulated failures using mixture, but after researching the topic with several engineers and senior colleagues, I changed to throttle cuts. If something goes wrong and it was found you acted contrary to manufacturer's recommendations, guess who gets the blame?
The accident report in this case also makes mention of the ATO simulting the failure at a point where safe continuation of flight (the path ahead) could not be assured. The impact speed was determined to be about 55 Kts, WELL below blueline and only just above the stalling speed. Additionally this flight was being conducted at night, the dangers of which were referred to with reference to the Metro that went in at Tamorth during night asymmetric training. I think there was more to this accident (as is often the case) than just a windmilling prop.
for some reason the pilot (10,000 hours plus) failed to feather the propeller.
I think this just illustrates the importance of getting the drill right. The pilot indicated he was feathering the engine, but the investigation found this had not been done. Maybe the rpm had decayed below min rpm for feathering??
The pilot (13,000 hours) selected gear and flap up as per his training but never got around to feathering the engine due to his pre-occupation with selection of the gear and flap handles.
I know for a fact that even a Duke (380 BHP/side which would
seem to be more than enough on one engine) will not perform with gear out.
In any case,
Analysis of DH-104 performance indicated that, at the time of the right engine failure, it was possible for the aircraft to achieve a positive rate of climb, assuming that the engine failure drills were performed promptly and correctly, and proceeded without interruption. However, when the landing gear failed to retract on the first attempt, any possibility of the pilot being able to attain the required aircraft performance was lost. As a result, he was probably forced to abandon completing the emergency procedures in order to maintain control of the aircraft.
It is interesting to note that while the pilot became preoccupied with the gear, he still has the presence of mind to reduce power on the operating engine as Vmca was approaching to maintain control.
All being said, there is no one correct answer to the questions that have risen here. Every situation is different. I'm not saying my way is the perfect one. I am open to and have have made changes to behaviour patters (ie. LEARNING) in the past. All we can do is look at the available facts and make the best decision according to the circumstances. The trick is making an INFORMED decision, not just acting because that's how your instructor said to do it. Understand the reasons behind the actions.
Not saying these are the only interpretation of these incidents, but again students cannot be babied along thinking they can handle the situation.
This stuff will kill you if you don't get it right (and that goes for instructors too). The stress imposed by a descending (non-performing) aircraft due to the incorrect configuration (be it gear out, flaps extended or windmilling prop) can be a very good motivator for the student to get it right.
I'm going to drift a little here onto decision speed vs decision point. I have heard several variations on the continuing portion of the engine failure brief.
One of these variants is "when there is no runway left over the nose, we are going to continue."
If you are below Vmca, we lose control.
If below Vyse, the only way to regain speed is to lower the nose = descent.
Why not put the aircraft back on the runway if below Vyse?
We are talking about FAR23 aircraft here, so forget V1 speeds. FAR23 guarantees
control with an engine failed, whereas FAR25 (>5700kg) = performance.
Even if it means you go through the fence at 40 kts, surely this is better than trying to reach blueline while avoiding the terrain?
Guarantee, most people will keep raising the nose as they see the ground approaching, neglecting the impending loss of control.
Marathon completed.