Stalls in the Pattern

[steveinindy]

You're correct in that you can't make anything idiot proof ... we'll just make better idiots. The AF447 stall is a great example of the highest technology not saving the airplane and all the people on-board. Personally, I think that if we continue to make systems smarter and smarter, we, as a people, will get dumber and dumber.

Panic and fear are more powerful than almost any other force on the human mind. As someone else pointed out, training is a partial answer but even then you are still going to get folks who will freeze no matter despite the best training and screening in the world. The answer is probably a combination of technology AND training. I think all too often folks tend to get in the mindset that technology means you can neglect training and others have this mindset that technology is a crutch that "real pilots" don't need. After helping pick up pieces of shattered pilots who crashed for want of a GPWS or instrument training, I can say that (at least in my book) real pilots use whatever they have at their disposal to try to avoid getting hurt or killed.

[Ron Blum]

Along those lines, GPS has shown to reduce the number of fatal accidents of VFR pilots into IMC. In other words, they now know where the mountain is. GPS doesn't address the special disorientation in IMC, though.

[Bill Greenwood]

I am not sure what a GPWS is, but I am pretty sure I don't have one. I have been lucky enough to fly for 34 years. I have some specific training, some probably better than average skills in some areas like stall recognition and recovery. But I don't get to train as much as I'd like, I don't have regular access to a T-6 like I used to.

As much or more than being a skilled pilot, my approach to danger is to plan and fly so as to avoid trouble. The simple old motto of "Using your superior judgement to avoid having to use superior skill."

I am fortunate to live in an area where I don't do much bad weather flying in actual IMC. Here in the mountains we often have very nice days where the visibility is 50 miles or more, and if it if really bad it is much too bad to fly.

[martymayes]

I think that crew catches a lot more shit than they should honestly. They were in the mindset of icing (references to transcripts about "the worst icing they had ever seen" etc) and in an aircraft where it is drilled into your head (thanks to the Comair crash on approach to Detroit Metro) that tailplane icing stall is a major concern.

That's interesting, because "tailplane icing stall" is not mentioned in the Comair 3272 probable cause published by the NTSB. "Tailplane icing stall" was never part of Colgan's stall training curriculum. In fact, no US certificated aircraft has been identified as being succeptible to "tailplane icing stall." Perhaps you can enlighten us as to who exactly is drilling this into the heads of pilots? Also, I believe the "worst icing" comment from the FO was it was the worst she had seen in a long time.

What happen is the crew, while going though the motions, was totally disengaged from the tasks at hand. The speed decayed below profile and stayed there for most of the arrival up to the loss of control point. Not once is a speed callout recorded on voice recorder. The light was on but nobody was home.

The onset of that is an abrupt and unexpected pitchdown possibly proceeded by control vibrations and a change in configuration. The upset came after they extended the flaps. The response they gave (flaps up and stick back) is exactly what you would expect if someone were responding to the tailplane stall.

Yes, when the stick pusher activates, there is a very abrupt pitchdown. That's what it's designed to do.

If pilots of a 121 operator are inventing their own stall recovery methods, contrary to published company manuals, and contrary to how they are trained, then the training problem would be even more serious. But pilots are not doing that. The SPS activated on this aircraft, exactly as designed, startled the pilot back into real time where panic replaced a trained response that could have saved the aircraft. There was adequate room to recover, as evidenced by subsequent simulation events.

The unfortunate thing is that they were too low to have time to figure the problem out. I am not excusing their culpability in causing the crash. I am simply pointing out that instead of painting them as a pair of clueless stupid idiots who didn't deserve their licenses, let's look at the bigger problems here and find real solutions to them.

The bigger problem there was a marginal pilot was trained to minimum standards and placed on the line where skills atrophied to a dangerous level. Trying to "figure the problem out" after the aircraft departs controlled flight is usually not a good strategy. Didn't help being paired with a non-assertive crewmember who was still mostly disengaged (quite possibly from illness/fatigue/lack of rest).

[Flyfalcons]

Let's not discount that in training, stalls only go to the first indication of stall and at the time, recovery procedures focused on minimum altitude loss while accelerating out. I don't know how the Dash handles large power changes but in a couple jets I've flown, there is a large pitch-down moment when advancing thrust to full. Because of those two factors, stall recovery muscle memory for me was increase thrust to full power and pull back on the stick hard to avoid the nose-down pitch moment to avoid altitude loss (the Excel is particularly bad with this). In my opinion, stall training in the simulator should go to fully stalled or stick pusher before recovering.

[PA11]

It's more about being trained that the aircraft can power out of the stall, not about pitch changes with the addition of power. After Colgan, all that changed. Now we lower the nose just like we did in Cessnas.

[martymayes]

recovery procedures focused on minimum altitude loss while accelerating out.

Not anymore. There is no reference to altitude. What PA11 said....lower angle of attack for recovery.

[Ron Blum]

GPWS is Ground Proximity Warning System (larger airplane (high $$$) equipment). But this information (through GPS data and terrain mapping) is something that we can get in a "poor man's" version. We need to use technology to bring costs down (and safety up).

[Flyfalcons]

Yes I'm aware of that, I just finished recurrent a couple weeks ago, but I was referring to issues at play at the time of the accident. Stall demonstrations are still performed to "first sign" though.

[wacoc8]

During recurrent training last month on the Hawker 900xp, we let the airplane stall, in the landing configuration with the auto-pilot on. Let the airplane fly through the shaker until we got the pusher, then recovered using full power with a nose down pitch attitude (about 7 to 10 degrees). We only lost 400 ft. Granted, we knew it was coming and were not discussing what was for dinner or what is our favorite Abita beer. Practice is good and learn from others mistakes.