Air France 447 - On topic only!

[MCM]

I am a Flight Dispatcher, also undergoing ATPL course training, so I have reason to believe I think I know what I'm talking about

There are also people on this thread with thousands of hours as jet transport pilots and engineers, some with A330 experience. They have reason to think they know what they're talking about too.

TeeVee,

The evidence leads us all down one pretty strong path - but its the fine details that everyone wants to get right. When you are dealing with these sorts of delicate issues, it is very important that, as a reporter, you are very accurate, precise and sure of what you are saying.

For the trained eye there is more than enough evidence in the interim report to lead to one of only a couple of conclusions.

Ball lightning, for what its worth, is not high on the list. There are far more plausible reasons for the messages and the loss of control.

[TeeVee]

Evan: typical response for you. Again, your qualifications are????

MCM: I defer to you as a pilot, but ACARS messages are computer generated and anyone that truly understands computers knows that in the event of shorts, component failures, software failures etc, all sorts of weird crap happens.

If the "evidence" was so compellingly strong, I suspect that the report would be far more conclusive.

Again, the French are about as trustworthy as the Taliban when it comes to pointing fingers at themselves. So if you all are happy believing a pack of essentially liars, that's great. I don't believe any of it.

re-read my post: i was not defending the ball lightning theory...

[Evan]

Evan: typical response for you. Again, your qualifications are????

My qualifications are that I've studied the reports and understand what is contained within them. You might take notice of that fact that MCM, who you will agree is highly qualified, is telling you essentially the same thing.

[Gabriel]

Evan: typical response for you. Again, your qualifications are????

MCM: I defer to you as a pilot, but ACARS messages are computer generated and anyone that truly understands computers knows that in the event of shorts, component failures, software failures etc, all sorts of weird crap happens.

If the "evidence" was so compellingly strong, I suspect that the report would be far more conclusive.

Again, the French are about as trustworthy as the Taliban when it comes to pointing fingers at themselves. So if you all are happy believing a pack of essentially liars, that's great. I don't believe any of it.

re-read my post: i was not defending the ball lightning theory...

First, I think that you should make a distinction between the French accident investigation board and the French judicial system.

The French accident investigation board didn't have problems saying that, other than the metal strip left by another plane on the runway, the weakness that downed the Concord (that is, fuel tanks being punctured by debris of tyre) has happened before and was very well known and not addressed as necessary to prevent an accident like the one that happened.

Second, yes, in a "computer gone crazy" even you can get all sorts of random errors. BUT!

What is the chance to get that random sequence exactly as the one that you'd have in the case of blocked pitots, similar to those that have happened in previous incidents of blocked pitots where the crew was able to manage the situation, and exactly as the plane was flying through weather conductive of this type of blockage?

The investigation board said (not that you car: you don't believe a word of what they said anyway) that all the evidence points to a speed disagree event led by pitots being blocked by ice. They also said that that alone doesn't explain the accident by itself, since there are procedures to fly without reference to speed and since other crews managed to fly through this. And hence the flight recorders are needed to have a full picture of the situation and of the actions taken by the crew.

Now, disregard all these facts, and yes, any theory is equally probable.

But you've said:

from what i recall from reading the "interim report" (which will most likely be the final report) i'm not sure that you can say the "evidence" contradicts his theory

And that is plain wrong. The evidence in the interim report clearly contradicts this theory. Whether you believe in the interim report or not is another story.

[VANGHELL]

a few more ideas

I disagree, mainly to the french report because, as people have said before, it is biased, yes! It is very protective, of both AF and Airbus! We all know, without offending anybody, french at heart, or citizenship, or birth, that the french GOVERNMENT is very protective of their own companies. Let me remind you TWO of the accidents in which Airbus has been shielded from harm:

1. Air Inter flight 148, which the BEA believe crashed because the pilots inadvertently left the autopilot set in Vertical Speed mode (instead of Flight Path Angle mode) then entered "33" for "3.3° descent angle", which the autopilot interpreted as a descent rate of 3,300 feet (1,000 m) per minute (PILOT ERROR). Afterwards, of course Airbus modified the display, to show four digit numbers on the display for rate of descent.

2. Air France flight 296, which crashed at the airshow near Mulhouse-Habsheim Airport, near Alsace: the pilot was blamed for not reacting to the decreasing altitude, and not being able to react in time to this situation, even though, third party investigations have reason to conclude anomalous behaviour of the A320. (PILOT ERROR) The month prior to the accident, Airbus posted two Operational engineering bulletins indicating anomalous behaviour noted in the A320 aircraft. These bulletins were received by Air France but not sent out to pilots until after the accident.

So, out of 5 Airbus related crashes, involving Air France, 2 were hijacked (don't count for the conversation), 1 was on French soil, AF 296 (pilot error blamed), 1 was in Canada, and the TSB, was unbiased, and pointed out the correct causes of the accident, without solely blaming the pilot, and it seems to me that AF 447, will be heading to the PILOT ERROR direction, if another serch should give no results. (hopefully, they will find, at least the wreckage).

Now:
1. considering the high level of automation aboard the Airbus (in general).
2. the incredible safety record of the Airbus range of aircraft.
3. the amazing safety features of this family of aircraft, which makes it almost FOOLPROOF, and almost impossible to crash,
I wonder... why in the name of God, were 3 pilots counting the relief pilot as well, not able to save such a piece of impressive machinery.

Let me give you an example: During flight training, I jumped from the VFR Cessna 172N, to the Skylane 182, for IFR, and being slightly different, but not essentialy, I employed the same stick and seat of the pants technique flying it as I did the 172. Now, the first time I flew it the instrutor told me to stay at 1500 ft for my procedure. Now as I was flying along, visually, of course, and within 2 minutes I was at 2200! At which point the instructor told me: "fly the settings, not the airplane" - stab set according to wheight, and 70 to 78% power setting for level flight. Now, I wonder, if the pitot theory is the one that is the cause of the crash, why in the hell didn't they apply stab 4, 5, .... and power accordingly, if the instruments failed? Wouldn't it be logical (and instinctual to all airline pilots) to do so! I mean fly the settings (keep the bird flying), and then see what the heck is going on?

I am not blaiming biased attitudes, and missconceptions, but remeber South African 295 guys. They never said what it was carying, that caused the fire, and it most certainly, wasn't computers in celofane wrapping. The type of thinking that lead to never exposing what caused this flight to crash, could prevent AF 447, from ever being laid to rest.

[Deadstick]

I'm of the opinion (way poorly informed) that the vertical stabilizer separated from the aircraft due to extreme turbulance and possibly excessive rudder control input. I hope one day they find the FDR.

[VANGHELL]

I'm of the opinion (way poorly informed) that the vertical stabilizer separated from the aircraft due to extreme turbulance and possibly excessive rudder control input. I hope one day they find the FDR.

As far as I know, most modern jets, have a rudder movement limiter, which prevents it from travelling too far in flight, even though claims are, that the pitot tubes were obstructed, and thus rendering rudder limiter useless because of no speed readings, or lower speed readings than normal, I still think they would have had no reason to use rudder pedals, knowing that the speed at FL 350 is M0.82 or more, in some cases. More so, because Airbus has had a rudder an vertical stabilizer accident before: AA 587, and important lessons were learnt from it.

[Evan]

and it seems to me that AF 447, will be heading to the PILOT ERROR direction, if another serch should give no results.

While pilot error could one day be determined to be a factor, I doubt this will be reconciled to a primary cause. More paramount is the issue of previously undocumented weather phenomena and the inability for certain air data probes to reliably operate within it. Another large factor is the cascade of vital system failures following an unreliable airspeed event, and the barrage of ECAM messages that ensue. Add darkness and turbulence to this confusion, loss of autoflight, alternate law, loss of instrumentation in IMC, and you have a recipe for disaster. What happened here is most certainly a recipe for pilot error rather than pilot error itself.

As for preventing another occurrence, we must simply avoid traversing such weather systems (in the ITCZ), build more robust pitot probes, and establish better QRH procedures for remaining in a safe flight envelope without airspeed data. It would also be wise for Airbus to reexamine the cascade of interdependent vital systems during a loss of airspeed data. Airbus has already introduced a Back-Up Speed System provision on the PFD, which is in standard use on the A380.

The fin did not separate in flight. That has been established by the evidence.

[Gabriel]

I disagree, mainly to the french report because, as people have said before, it is biased, yes! It is very protective, of both AF and Airbus!

Have you read it thoroughly? It doesn't sound like you did.

So, out of 5 Airbus related crashes, involving Air France, 2 were hijacked (don't count for the conversation), 1 was on French soil, AF 296 (pilot error blamed), 1 was in Canada, and the TSB, was unbiased, and pointed out the correct causes of the accident, without solely blaming the pilot, and it seems to me that AF 447, will be heading to the PILOT ERROR direction, if another serch should give no results. (hopefully, they will find, at least the wreckage).

Now:
1. considering the high level of automation aboard the Airbus (in general).
2. the incredible safety record of the Airbus range of aircraft.
3. the amazing safety features of this family of aircraft, which makes it almost FOOLPROOF, and almost impossible to crash,
I wonder... why in the name of God, were 3 pilots counting the relief pilot as well, not able to save such a piece of impressive machinery.

Let me give you an example: During flight training, I jumped from the VFR Cessna 172N, to the Skylane 182, for IFR, and being slightly different, but not essentialy, I employed the same stick and seat of the pants technique flying it as I did the 172. Now, the first time I flew it the instrutor told me to stay at 1500 ft for my procedure. Now as I was flying along, visually, of course, and within 2 minutes I was at 2200! At which point the instructor told me: "fly the settings, not the airplane" - stab set according to wheight, and 70 to 78% power setting for level flight.

Aha! And HENCE is why the MOST LIKELY theory is a ball lighting?
You sound very much like "I saw something flying that I could not identify and doing maneuvers that I can't explain, HENCE it's an extraterrestrial spacecraft with a crew of intelligent aliens"

There is something very wrong with these syllogisms, you know.

Now, I wonder, if the pitot theory is the one that is the cause of the crash, why in the hell didn't they apply stab 4, 5, .... and power accordingly, if the instruments failed? Wouldn't it be logical (and instinctual to all airline pilots) to do so! I mean fly the settings (keep the bird flying), and then see what the heck is going on?

Well, I'd say that in EVERY accident where pilot error was a necessary factor (NEVER "the" cause), the pilot DID something different as he should have. Or explain Colgan. Or Turkish at Amsterdam. Or AA 587.

[Gabriel]

More so, because Airbus has had a rudder an vertical stabilizer accident before: AA 587, and important lessons were learnt from it.

It's interesting to see how you choose only the information that fits your position, or you choose, from all the possible interpretations of something, the one that fits your theory, or you take one piece of information, cut and paste the part that fits your theory, and scrap the rest.

Examples:

"Airbus had a fin accident before, so lessons were learnt"
vs
"Airbus had several fin accidents and incidents before, so it wouldn't be strange if that was the case again".
(by the way, you might find it interesting that Airbus did nothing on the design of the rudders and its control system after this accident)

"In these accidents the BEA shielded Airbus from harm"
vs
"In the Concord accident the BEA clearly appointed responsibility to EADS due to their lack of action on a known issue that was a key factor in the accident"

"Air Inter flight 148, which the BEA believe crashed because the pilots inadvertently left the autopilot set in Vertical Speed mode (instead of Flight Path Angle mode) then entered "33" for "3.3° descent angle", which the autopilot interpreted as a descent rate of 3,300 feet (1,000 m) per minute (PILOT ERROR). Afterwards, of course Airbus modified the display, to show four digit numbers on the display for rate of descent"
vs
"BEA report of such accident: ... the commission considers that the ergonomic design of the auto-pilot vertical modes controls could have contributed to the creation of the accident situation . It believes the design tends to increase the probability of certain errors in use, particularly during a heavy workload."

There is a name for this "illness": Confirmation bias.

[Curtis Malone]

It's interesting to see how you choose only the information that fits your position, or you choose, from all the possible interpretations of something, the one that fits your theory, or you take one piece of information, cut and paste the part that fits your theory, and scrap the rest.

It's no more interesting than the way most everyone else around these parts does the same thing.

[Gabriel]

It's no more interesting than the way most everyone else around these parts does the same thing.

In my previous post I've said that confirmation bias is an illness.
That was not fair. Confirmation bias is the normal human behaviour. If left "alone", humans do confirmation bias. SO what would be the opposite of "left alone?". Understanding and proactive attitude.

I try to keep myself aware of the confirmation bias and make a honest and reasonable effort to try avoid it. Not that I always succeed.

[Evan]

In my previous post I've said that confirmation bias is an illness.
That was not fair. Confirmation bias is the normal human behaviour. If left "alone", humans do confirmation bias.

Not exactly. Confirmation bias is the normal human behavior when one wants to adhere to a conviction, or to convince others of one, and is not interested in an objective investigation of the evidence for the purpose of finding the truth. The reason we have the ability to learn is because there is also a normal human tendency to want to question one's own convictions and explore new or opposing ideas. So, in the larger sense, human behavior is balanced somewhere between conviction and curiosity, and if left "alone" one might take either approach depending upon the individual and the circumstances.

[Gabriel]

Not exactly. Confirmation bias is the normal human behavior when one wants to adhere to a conviction, or to convince others of one, and is not interested in an objective investigation of the evidence for the purpose of finding the truth.

I don't agree.

I'm not an expert in human factors, but what I've read:

Humans tend to form an opinion early (be it in politics, science, religion or accident investigation). The natural tendency after that is to defend that opinion. This is not an intentional process, it's just how it flows naturally. You simply likes the things that agree with you, and dislike those that doesn't, so it's natural to put more weight in those that you like.

Studies have demonstrated that people faced with exactly the same information in different order reacted differently based on what was their first impression.

As I've said, it's not something intentionally done to defend one's agenda or because one is not interested in the "truth". Think for example the confirmation bias in the context of a pilot error. Do you really think that the pilot wanted to defend his theory and was not really interested in knowing that there was a mountain out there even if that contradicted his theory?

Avoiding confirmation bias takes awareness and effort.

Curiosity, in general terms, is more related to the unknown than with what one considers known. That's why there are so many courses and tools for creativity, thinking out of the box, braking paradigms, etc. It's a powerful tool, not one that flows naturally for the unaware human, that's why creativity is an important value for a job profile. If everybody was creative, it wasn't very valuable (supply and demand they call it)

[EconomyClass]

Has a "final report" been published? I mean, past history suggests that the real information might come out in 15 or 20 years, but I don't expect the BEA to wait that long for a "final report".