Stall recovery technique

[foxmoth]

Also, if you read the transcript of the AF, that was not the case, they really did not know WIHWH

[AirRabbit]

Of course, it is true that it would be “nice” if crew members were to be able to “identify” the fact that they are in a stall, if, indeed, that happens to be the case. However, it seems to me that there is at least a possibility that these “pilots” (Colgan and AF447) knew they were in a stall. Why do I say that? Well, both of these pilots responded to something – and what they did, was not only deliberate and continuing, demonstrating to me, at least, that they were committed to do what they were doing. Additionally, the actions they each took seem to have been driven by the fact that what they were doing was just exactly what they wanted to do. That's pretty deliberate. What drives that kind of "deliberateness?"

They each pulled back on the controls: in the Colgan accident, the pilot even went to the point of attempting to over-control the stick pusher; in the AF447 accident the pilot also pulled back on the controls, allowing the airplane to climb (at one point at a rate of 7000 fpm) AND, something, or someone, managed to move the trimmable horizontal stabilizer from a nominal 3 degrees nose up to a whopping 13 degrees nose up position, doing so in about 1 minute and that extreme nose up position was maintained for the duration of the flight; also both pilots managed to advance the power to, or almost to the maximum for go around setting. There is that attitude of being deliberate ... again!

In a previous post of mine, on this thread (at #14), I described an accident that occurred on December 22, 1996, involving ABX Air (Airborne Express), flying a Douglas DC-8-63, in Narrows, Virginia, USA. Both pilots were very experienced instructors with lots of experience in the subject airplane, and were doing a post maintenance flight on an airplane having just recently completed a routine maintenance check and routine servicing. One of the tasks their procedures called for during such a functional check out was to perform a recovery from an approach to stall.

These pilots happened to have been trained on a simulator that was not programmed correctly – and had been trained to recover with an absolute minimum altitude loss, using only an increase in power setting/thrust. Attempting to comply with these 2 “standards,” they attempted to keep the nose of the airplane on or above the horizon (unsuccessfully), while attempting to add power to recover (also, unsuccessfully). What they managed to achieve was a fully developed aerodynamic stall. But, because they apparently understood from their training that the procedures to follow to recover from a fully developed aerodynamic stall were exactly the same as a recovery from an approach to stall. Therefore, they stubbornly attempted to maintain altitude and recover by simply adding power. Even though they could accomplish neither, they continued to attempt to do this until they crashed. Why? Could it be that they wanted to trust their training?

OK, I certainly am aware that there may be some here who can, or would want to, point to at least some facts that may not line up between the ABX accident and these other 2 accidents. But, the fact still remains that these critical actions – attempting to maintain attitude/altitude and adding power – are clearly present in all 3 accidents. This leads me to conclude that they were very likely attempting to do what they were trained to do – but the training let them down. They all were very likely either incompletely or incompetently trained, and very likely left the training school fully confident that they could handle whatever “mother nature” or airplane malfunctions could throw at them … and unfortunately we now know that wasn’t the case.

It is for these kinds of reasons that I continue to work, even fight, for a regular, professional review of training standards and training methods for pilots, instructors, and evaluators.

[Zaphod Beblebrox]

I will pollute this thread with some musings.

It has been ordained in the US that carriers are strongly encouraged to create training modules for AQP on revised stall recovery.

The last one I did in training involved allowing a auto thrust failure while operating at 370 in B737-400 classic. This was to allow the aircraft to get well into the stall at altitude before effecting recovery.

The surprise is in the effort it takes to significantly reduce the pitch and angle of attack on an out of trim nose up aircraft at FL370. There is a real push required and additionally when you push up the power there is not a lot there and it takes a while to get spooled up.

You will end up giving up quite a bit of altitude to get things right. Try to recover too soon and the buffet starts right in. The aircraft does not respond as quickly as it does down low. Maintain altitude? forget it!

[AirRabbit]

Originally Posted by Zaphod Beblebrox

I will pollute this thread with some musings.

It has been ordained in the US that carriers are strongly encouraged to create training modules for AQP on revised stall recovery.

Actually, the “strong encouragement” came from the Flight Standards Offices in Washington, through the AQP Office, when various circumstances illuminated during the investigation of the Colgan Air accident. During this investigation, it came to light that airlines training under AQP were authorized to substitute some required tasks with another required task - where the tasks involved were thought to use or require sufficiently similar control applications. With this substitution/reduction in the tasks involved, it was almost immediately recognized that the newly available time could then be used to address other tasks or other training applications, like line scenario training. Unfortunately, in at least one case, a major airline was found to have completely eliminated the required training/testing task of “Recovery from Stalls/Approaches to Stall” and authorized the substitution of “Windshear Encounter on Approach” or “Windshear Encounter on Takeoff” to take its place, and this substitution had been in place for an undisclosed number of years.

When senior FAA managers learned that all AQP authorized airlines were able to make such substitutions, and without further, detailed, analysis of all other AQP authorized airlines, it would not be possible to say what other airlines, if any, may be authorized to make the same or similar substitutions. Without waiting for such analysis and because of the increased focus on airline stall training in the wake of the Colgan crash, senior FAA management officials immediately took steps to ensure that each AQP authorized airline was, in fact, training/checking on the appropriate “Recovery from Stalls/Approaches to Stall” task.

It would appear that instead of returning to the traditional task descriptions of training/testing in such “stall” tasks, either the airline or AQP managers suggested or encouraged the development of a task that was different from what was contained in the existing regulations and practical testing standards. As far as I know, whether or not there is one or more “newly developed” tasks and whether or not this/these task(s) are to be trained only, or both trained and tested, is unknown to anyone outside of the airline and the AQP management officials.

[lalitya]

Clarification

Thank you for your reply.
What I meant by repeated training is that, most check rides, proficiency checks, line checks etc, involve stall recovery as one of the tested elements.

[lalitya]

Thank you all!

Thank you all for your replies. It is a great discussion.

@AirRabbit: Great post! Thank you for your wonderful thoughts on this.

The douglas stall recovery (at least takeoff stalls) also included firewall power, flaps fifteen. Configuring you for flying in case you had forgotten the flaps (and slats) on takeoff!

@glendalegoon: Can you please explain what firewall power is?

@eckhard: Great post! Thank you!

@+TSRA: Agreed with your views!