Air France 447 - On topic only!

[Evan]

I'm still waiting for a reference that would suggest that a crash is likely if a pilot used a power setting, attitude and VS that he was familiar with for normal operations- but didn't have airspeed indications...Please.

A reference... let's see.... Oh, I know:

AF447. The pilot did what he thought was normal. He did not intend to slow, stall, exceed crit AoA and die. He did not intend to do any stunt flying.

Because reality includes 'surprise' factor, coordinative overcontrol, confused situational awareness and pilot error, we have procedures, checklists and CRM. Follow them and you live. And so do all the lives you are entrusted with. It really is that simple.

Done.

[Gabriel]

EASA states that it has been determined that, when there are significant differences between all airspeed sources, the flight controls of an Airbus A330 or A340 aeroplane will revert to alternate law, the autopilot (AP) and the auto-thrust (A/THR) automatically disconnect, and the Flight Directors (FD) bars are automatically removed.
Further analyses have shown that, after such an event, if two airspeed sources become similar while still erroneous, the flight guidance computers will display the FD bars again, and enable the re-engagement of AP and A/THR. However, in some cases, the AP orders may be inappropriate, such as possible abrupt pitch command.

Seems pretty straightforward, not too much to read?

Yes. It's clear. That is something that can happen. Not samething that happened. Not at least in this AF accident.

[Gabriel]

Add to that, ADIRU's popping in and out saying "dead, no wait, I'm up, no wait, I'm dead" (mach part of vid) and giving the pilots the bait of being able to go back on AP (given the first quote above) is likely exactly why they decided to give HAL another try (..."the AP orders may be inappropriate, such as possible abrupt pitch command.")

That was after more than 90 seconds of the pilots doing wrong thing, free of HAL dubitations. And they never decided to give HAL another try. They never re-engaged the AP.

Lets focus on what the pilots had available to them, shall we?

Fair enough.

They had three perfectly working attitude indicators and a working sidestick. That's all they needed to apply 5deg nose up as required by the memory item.

They also had two perfectly working thrust levers. That's all they needed to apply CLB thrust. The only other memory item. In addition, they also had perfectly working engine instruments (all of them) to verify, if they wanted to, that the engines were correctly responding to their thrust lever command, had there benn one.
Other than the above, they had three perfectly working attitude indicators to see that they were going to a nose-up attitude that was unsustainable (and I' not talking about a sudden reaction on a rubber-now-ferrari sidestick, I'm talking about a lot of seconds). They also had three perfectly working vertical speed indicators that shoed that they were climbing at an unsustainable vertical speed with a light airplane low at TOGA, let alone a heavy one at its ceiling and reduced thrust. They also had three perfectly working altimeters that showed that they were climbing like crazy past an altitude that, seconds before the event, they had stated was not achievable do to higher-than-standard temps.

Finally, after more than 30 seconds of the above, they had all that plus a perfectly working stall warning that, for almost one full additional minute, was clearly horning and shouting "stall".

Wouldn't an AoA display on the AB glass cockpit the pilots look at every day, have helped avoid any confusion about what we now have the luxury of seeing on the trace?

The pilots? I agree
These pilots? I doub't it. Why would they have used it, if they decided to ignore (or could not understand) all the other correct indications they had available?

As far as I know, if the AP/AT can re-appear/dis-appear/ad nauseam causing multiple false stall warnings,

Wrong. The flight director bars can re-appear/dis-appear.
The AP remaind off since the start of the event to the impact.
There were no false stall warnings. There were false lacks of stall warnings, only after morwe than one minute and a half of the start of the event, when the pilots had already lost control of the plane and, more significantly, of themselves.

[3WE]

A reference... let's see.... Oh, I know:

AF447. The pilot did what he thought was normal. He did not intend to slow, stall, exceed crit AoA and die. He did not intend to do any stunt flying.

Because reality includes 'surprise' factor, coordinative overcontrol, confused situational awareness and pilot error, we have procedures, checklists and CRM. Follow them and you live. And so do all the lives you are entrusted with. It really is that simple.

Done.

Wrong Evan, instead of power, attitude and VS for normal flight, he chose a low power, nose up, high-altitude, very rapid climb, which, someone with fundamental aerodynamic knowledge would know is anything but normal flight, but instead a very good recipe for a stall.

Try again:

I want a reference that shows that power settings, attitudes and VS that are used for normal flight are likely to result in a crash...never mind, there's some reading comprehension problem here.

[Evan]

Wrong Evan, instead of power, attitude and VS for normal flight, he chose a low power...

No. Wrong! He didn't chose low power. He had only 83% thrust because thrust lock was engaged when the autothrottle disconnected during a selected speed change. He didn't move the thrust levers because he wasn't aware of this. He had a flawed situational awareness.

Meanwhile, the ECAM is saying THRUST LEVERS MOVE and the QRH is saying MOVE THRUST LEVERS and the memory item is MOVE THRUST LEVERS to CLB. But CLB power wasn't restored until well beyond the initial pull-up, because he was flying on instinct instead of adhering to the established procedures and CRM.

Are you starting to get my point? The pilot's instinct was telling him he had CLB thrust and his instinct was wrong. In a surprise moment with flawed situational awareness, a pilot must be expected do the wrong thing (or not do the right thing) instinctively. On the other hand the procedures will account for possible confusion and stabilize the flight to give the pilots time to sort out the situation methodically.

If they had gone directly to procedures and checklists they would have had around 100% thrust instead of 83%. (They also would have been at 5° pitch, and would have landed in Paris).

In the wake of this, it is painfully obvious to see why these things must be respected. This is an object lesson in humility. Yet bravado persists. I really don't understand what your problem with these procedures is.

[KurtMc]

Peace

[Black Ram]

KurtMc As far as I know, if the AP/AT can re-appear/dis-appear/ad nauseam causing multiple false stall warnings, even though Alt Law has been triggered, and the pilots are now in manual flight..... well, as far as I know, perhaps it did so without leaving a trace.

False stall warnings? Really? So they actually weren't stalled. Must have been a 100mph+ downdraft, coming right after the 100mph updraft...

As Evan once said, it's also possible for the crew to have lost the tailfin, with ball lightning welding it back together, without leaving a trace. Confirmation bias?

I still don't quite understand your theory. The sneaky little HAL attacking from the back door by engaging the AP/AT, without leaving a trace on the FDR, which caused false stall warnings......which the crew "respected for awhile, but [threw] on the junk heap like Chicken Little"....and crashed. Which brings us to this:

All the while, our pilots are trying to follow those memory items. Trying real hard to understand the illogical, in a pressure situation with their backs to the corner, interpreting as best they could.

What memory items were they trying to follow? Climb over ceiling when you don't know your speed? Only if they used their altimeters, the bizarre would have instantly looked a lot more logical (I mean, if they had used them prior to reaching FL100).

If the AP's kicked on with that erroneous data and says 'shit, boys, we're fallin' and goes full tilt boogie with up elevator, pushing that THS along with it... please watch again the mach tracings in comparison to stall warnings, in comparison to the position of the aircraft pre-stall/incipient stall.

Now I see why they didn't take you to aircrash investigation school. Before you call this a "personal attack", I'll explain my point - there are trained professionals who have already analyzed the data for months.
But you are also lacking in your logic - if the AP/AT did re-engage (which didn't happen), with the IAS dropped so low due to 1) speed exchanged for altitude to reach FL38000 (I assume we are talking about after the climb, as you say HAL is apparently thinking they are falling), and 2) ice blockage - then the last thing that would have happened is thrust to IDLE. Also, there is no reason for the AP to climb from FL350 to FL380, in alternate law, with speed erroneously low. But this will happen if the AP was off (as it was), with the aircraft in alt law (as it was), and with the crew pulling up (which they did). And as long as the aircraft was not in abnormal alternate law (which it wasn't), autotrim would ensure the THS obeys the pilots' commands, which were mostly nose-up.

Lastly, I keep wondering what's your next masterpiece. With the theory of flat spin and subsequent tailfin separation, and the theory of updraft causing a stall thrown out, and with the theory of AP/AT engaging itself without leaving a trace about to be thrown out too, I can't help but think that aliens are up next.

[Evan]

As Evan once said, it's also possible for the crew to have lost the tailfin, with ball lightning welding it back together, without leaving a trace.

That's my theory and I'm sticking to it, although I'm leaning more towards cube lightning after reading the report (well, after looking at the pictures anyway).

[KurtMc]

Peace

[Gabriel]

Kurt, check your PMs.

[Evan]

Square lightning?

That video is amazing. Not once, not even once, is the problem of inadequate pilot training addressed. Even when the Prof. David Woods (FAA Advisor on automation) testifies to the pilots saying things like "what's it doing now?" and "How do I stop it from doing this?", The issue of training is not raised. Amazing. Then we have Lufthansa's chief safety pilot telling us that the problem with sidestick automation is that, instead of being able to fly by 'feel' without paying attention to the instruments, the pilot now has to constantly look at his instruments. He describes traditional yoke flying as...

"He doesn't have to look at his artificial horizon. He doesn't have to look at his instruments." "he just places his hand on the control column, and he looks outside or he talks to the stewardess (that's right, stewardess), or talks with somebody else, or something like that, and he knows, just by placing his hand on the column, exactly what his aircraft is doing.

THAT'S THE CHIEF SAFETY PILOT SPEAKING! Amazing! I mean, what the hell is going on? We have Air France now completely exposed for lack of proper training (i.e. failed safety culture) and now we see Lufthansa giving evidence that inattention to instruments is SOP for their pilots (i.e. failed safety culture). We already had AA guilty of not instructing pilots on use of rudder. Basics. And these are supposedly the world's leading airlines.

They show the Air Inter Flight 148 incident. Pure pilot error. The pilot was in the wrong descent mode. How could he not know that? That mode is clearly announced on both the FCU and the FMA. Meanwhile the rate of descent is clearly obvious on the altimeter and VSI. If the pilot was WATCHING THE INSTRUMENTS he would have seen the error. This is not a problem of automation. It is a problem in training and pilot screening. But obviously it is also a problem in operator protocol.

Safe piloting requires these two things:
Thing #1: Know the plane COMPLETELY. Know how it works.
Thing #2: ALWAYS fly to plane. "Fly the plane" means monitor the instruments. If you have a hand on the column while you flirt with the 'stewardess', you are not flying the plane.

What led to AF447 is becoming clearer to me every day.

I'm thinking of opening a commercial piloting school. It doesn't seem to require much knowledge or experience. You just churn them out. It's like printing money!

[mawheatley]

But where do we go from here?

What led to AF447 is becoming clearer to me every day.

I'm going to throw a bullet into the campfire here and see what shoots out: I'm a little concerned that this whole incident will be put down to pilot error (which it clearly was) and the book will be closed. But I ask, what can be learned? What improvements CAN be made? What will prevent this from happening again? We've read the reports, we've sat, jaws dropping, as we've digested the chain of events and the pilots' actions that put the aeroplane into the ocean. We've all said words to the effect of "how could they let that happen?" My point is this: if we could go back in time and put the three AF447 pilots in a room 24hrs before the crash, and tell them that by this time tomorrow you will all be dead, because you will, collectively, allow an A330 to stall, at 35,000 ft, and be unable to recover, they'd look at us like we were stupid.

They were all (presumably) intelligent men. They'd all gone through proper training, all had experience on modern jet aircraft, all had experience on the A330. They'd passed their exams, ridden their check rides and proved competency to their examiners. They were not stupid people. Each one of us is convinced that, put in a similar situation the outcome would be different. Thing is, they would have said the same thing! What makes us different from them? I'm sure they all knew WHAT to do. The question is why, with their backs against the wall, did they not do it?

And more to the point, what can be done to ensure that the next group who find themselves in this situation DO take the correct course of action and bring about a much happier ending?

[Fear_of_Flying]

I'm going to throw a bullet into the campfire here and see what shoots out: I'm a little concerned that this whole incident will be put down to pilot error (which it clearly was) and the book will be closed. But I ask, what can be learned? What improvements CAN be made? What will prevent this from happening again? We've read the reports, we've sat, jaws dropping, as we've digested the chain of events and the pilots' actions that put the aeroplane into the ocean. We've all said words to the effect of "how could they let that happen?" My point is this: if we could go back in time and put the three AF447 pilots in a room 24hrs before the crash, and tell them that by this time tomorrow you will all be dead, because you will, collectively, allow an A330 to stall, at 35,000 ft, and be unable to recover, they'd look at us like we were stupid.

They were all (presumably) intelligent men. They'd all gone through proper training, all had experience on modern jet aircraft, all had experience on the A330. They'd passed their exams, ridden their check rides and proved competency to their examiners. They were not stupid people. Each one of us is convinced that, put in a similar situation the outcome would be different. Thing is, they would have said the same thing! What makes us different from them? I'm sure they all knew WHAT to do. The question is why, with their backs against the wall, did they not do it?

And more to the point, what can be done to ensure that the next group who find themselves in this situation DO take the correct course of action and bring about a much happier ending?

The part of your premise I'm not sure is true is that "they all knew WHAT to do." I think, if you caught one of the pilots by surprise at the grocery store, and asked him - "Your airbus has just experienced UAS, what do you do?", he would have been just as mystified then as he was at 35,000 feet.

After Colgan, we were all fairly comfortable with the notion that regional airline pilots, on grueling schedules, who routinely failed checkrides, could slip through the cracks. The difference here is that we have pilots of a major airline acting in a similar manner (albeit under different circumstances). I agree: these pilots were of course reasonably intelligent, well-intentioned people who met the standards for their profession.

So the obvious conclusion is that the standards for the profession are not high enough. If there are memory items for UAS, every pilot needs to know them without thinking and without blinking.

Yet air travel seems to be safer than ever, even though the population of pilots, more and more, consists of people who do not have decades of Air Force experience, for example, learning the trade through hard experience. They come up through a system that provides them with minimum x number of flying hours, written tests, and flight simulations. They don't have the real experience of facing crises. How can it be, then, that planes aren't falling out of the sky left and right?

This apparent contradiction exists because even as the quality of training and experience decreases, the automation keeps ahead of the game, and planes become less and less vulnerable to human error.

And that to me is the answer here. There will of course be recommendations to improve pilot training, and they might have some short term benefit, but we aren't going to change the world and how people enter aviation as a career, receive their training, and ascend through the ranks. What we are going to continue to change, is automation. I think Airbus will modify things here, including how the stall warning functions, and eventually, how the aircraft's computers deal with UAS, carrying out the memory items automatically. The automation, so much maligned by some on this board, will be our salvation.

[Evan]

They were all (presumably) intelligent men. They'd all gone through proper training, all had experience on modern jet aircraft, all had experience on the A330. They'd passed their exams, ridden their check rides and proved competency to their examiners... I'm sure they all knew WHAT to do. The question is why, with their backs against the wall, did they not do it?

This is exactly the issue. They had NOT gone through proper training. By proper training I don't mean what currently passes as proper training. I mean training for all abnormal operations listed in the FCOM and FCTM and QRH (as well as stall recovery, which is not listed there) and regular proficiency checks for these procedures. Read the report. The did not have this training for either UAS or stall recovery. This is why, with their backs to the wall, they didn't do what was required.

The more I think about that Lufthansa Chief Safety Pilot, the more I am convinced this is an industrywide issue of bureaucratic protocol. It's too bad hundreds of people have to be murdered before anything can be done about it. If a few hundred executives lost their jobs instead, that would be a fairer outcome.

[Black Ram]

Square lightning?

Kurt, this outdated TLC show is really of no help to your searching aimlessly for a reason to blame the bus. Do you know that there is an entire group of people out there believing that the air show A320 crash was of an unmanned aircraft? All because of this program. Also, they show the A310 incidents where the crew was unfamiliar with the controls and messed up their trim, resulting in the aerobatics you see....and they present it as if it happened on the computerized A320. What do you have to say to that? It's just terrible reporting, I'm sorry.
As others have mentioned, there is no word on the crew's actions in the air show and Air Inter A320 crashes. "The plane wanted to land and it won"? BS, have you ever heard of stall protection, the thing that may have saved 133 lives in Habsheim? This incident has been covered in this very thread at least once. But apparently you are a writer, not a reader.