Air France plane missing?

[beachfront]

It may not seem like much to us, but that one accident cost 228 lives. If I were a loved one, I would want answers very much, and if that meant grounding the fleet, I would demand it, even if my pleading would not be heeded.

If this accident didn't happen at altitude, I would still be worried, but not as worried. It appears that we will never know exactly what plucked this bird from the sky from the aircraft herself. I think it is worth looking into.

With all due respect, I don't believe in the concept of "acceptable risk" if there is a known fault or issue.

If there was no fault or issue there would be no "acceptable risk" ..

With all due respect, I don't believe in the concept of "acceptable risk" if there is a known fault or issue.

you would have to live in a bubble to avoid this.. it comes down to almost every single thing you do ...

You like to walk in the park.. there is an acceptable risk that you trip on a piece of concrete the city has deemed not bad enough to replace yet ...

You like drive your car ... there is an acceptable risk the tire manufacturers use to make sure your tires do not blow...

Boeing has an acceptable risk on thier planes based on what they believe has an extremely minute chance of happening...

You better believe in the concept of "acceptable risk" because you are accepting that risk every day ... to not beleive in it would be foolish.
Most items, etc. have a no fault or issue based on acceptable risk ... other wise there would be no way to produce them ...

Nothing is perfect ... a fault or issue can be attributed to most things we do, buy, or participate in ... this includes air travel, 401k's and on and on and on...

[SYDCBRWOD]

Another highly unlikely scenario. All "icing at instantly the same time"

Why is that so difficult to believe? 3 identical devices exposed to identical conditions and yet you expect them to act differently? Now, that I find hard to believe.

[seahawk]

It is enough when one of the pitot tube systems ice over to put the flight controls into alternate law, removing some limitations on possible control inputs. Add heavy turbulence and a certain known weakness in the structure of the plane and there you go.

[mfeldt]

I realize nobody wants to listen to me because I'm a novice, and an alleged airhead and idiot, but I have a serious concern that I posted previously and nobody commented on.

Of course people can specuate on the reason for an airliner falling from FL350 without rhyme or reason. All of the obvious suspects should be discussed and debated.

However, until the investigation is complete or until the investigators show an official copy, we don't even know for sure if the aircraft sent any communications at all. Yes, I know, AF said that it did. I can also say that I saw John Lennon having lunch with Jim Morrison at Burger King, but that doesn't make it true. The ACARS messages that we have so heavily relied upon are not factual. Someone made a best-guess copy from a screenshot from the TV news. That's good and well, but my question is, how did the TV news get the copy? Who gave it to them? Surely AF was not making copies and passing them out to the media.

My point is we don't know what the ACARS said or when, if anything. I am sad to know that this is the case.

It may not seem like much to us, but that one accident cost 228 lives. If I were a loved one, I would want answers very much, and if that meant grounding the fleet, I would demand it, even if my pleading would not be heeded.

If this accident didn't happen at altitude, I would still be worried, but not as worried. It appears that we will never know exactly what plucked this bird from the sky from the aircraft herself. I think it is worth looking into.

With all due respect, I don't believe in the concept of "acceptable risk" if there is a known fault or issue.

Isn't there a clear requirement for grounding a fleet? I read somewhere that a fleet would be grounded when the accident provability is greater than 1x10^-6 per hour, which translates to one crash every three months for the A330/A340 fleet. Since there have been only 3 crashes of the type since 1994, I guess it shouldn't be about grounding a particular type of aircraft. However it appears to me that this crash fits well into a series of sensor failures that cause modern, mostly computer controlled aircraft to go down (remember the Turkish B737 that crashed in Amsterdam a few months ago, because the A/P thought the plane had already landed due to erroneous altitude input? The LOT B767 which diverted to Toronto after pitot tube problems? The Qantas A330s with the same problem? The A320 crash of Air NZ in the Mediterranean?). It seems that all the computer control - which I'm sure has already avoided lots of accidents - becomes a nightmare when certain input sensors fail. Apparently such situations put so much stress on the crews, that they are unable to concentrate on recovering the aircraft any more... Maybe this is the point that needs to be addressed next... m.

[Evan]

This is probably going to ruffle a bunch of feathers here but I am going to give you my opinion anyway. I have stayed away from this thread up until now, but I think this is what really happened. I wish ITS were here because I think he would agree with most of what I am about to say.Yes it is possible for the aircraft to have entered a "wall" of super cooled water at that altitude and iced up almost instantaneously,however I doubt that seriously. I think that at the time of time of the encounter with the weather, the Captain had already gone to the bunk. The 2 First Officers are now sitting up front at altitude fat dumb and happy. Probably shooting the shit and not paying enough attention to the weather radar. Now they are in it, the aircraft is being beat to shit by severe to extreme turbulence. The old man tries to get out of the bunk and make his way upfront to save his bird, but in that kind of turbulence it is impossible to move let alone make his way up front. The 2 First Officers over react to the situation, over stress the airframe and pull the aircraft apart. At first the engine pods are pulled off the wing, hence the ACARS message about loss of electrical power, no engines, no generators, no electric. Now the airframe becomes compromised and the loss of pressurization, ACARS still working off battery power only. Just my opinion but the chance of all of the pitot static system being iced over all at once, remember these units are heated,four pitot tubes,two for each pilot, upper left and lower right for the Captain and upper right and lower left for the First Officer. The same with the static ports, so this theory just does not fly with me.


BB

I can see a number of flaws in that theory, but I go over just the most obvious ones.

First of all, 'fat, dumb and happy' pilots cannot overstress the A330 in normal law. Unlike the A300, the A330 has FBW protections that limit pilot input to the operational environment. AF447 dropped into alternate law, but only after whatever set things in motion.

Secondly, (Have you been following this thread?) since the accident, a large number of pilots have come forward with previously unreported cases of total airspeed loss due to icing at altitude. So there is a strong precedent. In comparison, there is no precedent for an A330/340 airframe failure due to pilot inputs.

Thirdly, the ACARS messages received in the final minutes of AF447 nearly match those of an Air Caraibes A330 that experienced unreliable airspeed due to pitot tube icing in 2007. So you have a 'fingerprint" match.

Lastly, there have been no reports denouncing the skills of the F/Os on AF447. and, seeing as they are among the deceased here, referring to them as 'fat. dumb and happy...shooting the shit' without anything to back that up seems to violate a code of respect for the deceased among pilots. I mean, correct me if I'm wrong...

[Evan]

It is enough when one of the pitot tube systems ice over to put the flight controls into alternate law, removing some limitations on possible control inputs. Add heavy turbulence and a certain known weakness in the structure of the plane and there you go.

Very good point.

[SYDCBRWOD]

It is enough when one of the pitot tube systems ice over to put the flight controls into alternate law, removing some limitations on possible control inputs. Add heavy turbulence and a certain known weakness in the structure of the plane and there you go.

Earlier on in this thread (30 pages ago? Dunno, not about to spend an hour plus checking) wasn't it stated that if the info from one pitot disagreed with the other two that the systems would go into alternate law, but the chain of failures seemed to suggest that it wasn't just one that went down but it must have been at least two and probably three to crash both flight computer channels and the final backup flight info system (forgotten what it was called). Why the cascade of electronic failures if not from compromised data from all the pitots? Are you suggesting that the failures cascaded from the ACARS was after the aircraft had started to break up and therefore sensors etc were being disrupted?

[shamrock]

Isn't there a clear requirement for grounding a fleet? I read somewhere that a fleet would be grounded when the accident provability is greater than 1x10^-6 per hour, which translates to one crash every three months for the A330/A340 fleet. Since there have been only 3 crashes of the type since 1994, I guess it shouldn't be about grounding a particular type of aircraft. However it appears to me that this crash fits well into a series of sensor failures that cause modern, mostly computer controlled aircraft to go down (remember the Turkish B737 that crashed in Amsterdam a few months ago, because the A/P thought the plane had already landed due to erroneous altitude input? The LOT B767 which diverted to Toronto after pitot tube problems? The Qantas A330s with the same problem? The A320 crash of Air NZ in the Mediterranean?). It seems that all the computer control - which I'm sure has already avoided lots of accidents - becomes a nightmare when certain input sensors fail. Apparently such situations put so much stress on the crews, that they are unable to concentrate on recovering the aircraft any more... Maybe this is the point that needs to be addressed next... m.

A computer is only as good as the idiot in charge of imputting the information into it....

The AA flight to Cali some years ago proves that when it went into a mountain, or the USAF 737 that was meant to land at Dubrovnik and hit a mountain instead...and there are many other cases where tiny, insignificant mistakes have caused catastrophic accidents.

When computerised gadgets meet humans, mistakes and cock-ups ensue...no matter how good a pilot you are, no matter how well trained you may be. Once day that computerised thingamabob will bite you in the ar$e and all it takes is one wrong letter in the spelling or one wrong digit in the direction.

Computers only make mistakes if the human programmes a mistake into it.

[seahawk]

Earlier on in this thread (30 pages ago? Dunno, not about to spend an hour plus checking) wasn't it stated that if the info from one pitot disagreed with the other two that the systems would go into alternate law, but the chain of failures seemed to suggest that it wasn't just one that went down but it must have been at least two and probably three to crash both flight computer channels and the final backup flight info system (forgotten what it was called). Why the cascade of electronic failures if not from compromised data from all the pitots? Are you suggesting that the failures cascaded from the ACARS was after the aircraft had started to break up and therefore sensors etc were being disrupted?

I can´t answer this. I have not enough knowledge of how the systems interact.

But look at this list:

02:10Z:
Autothrust off
Autopilot off
FBW alternate law
Rudder Travel Limiter Fault
TCAS fault due to antenna fault
Flight Envelope Computation warning
All pitot static ports lost
02:11Z:
Failure of all three ADIRUs
Failure of gyros of ISIS
02:12Z:
ADIRUs Air Data disagree
02:13Z:
Flight Management, Guidance and Envelope Computer fault
PRIM 1 fault
SEC 1 fault
02:14Z:
Cabin Pressure Controller fault

But then again the rudder limiter is supposed to lock itself in place when the speed data is conflicting.

[Evan]

A computer is only as good as the idiot in charge of imputting the information into it....

In this case the idiot was a pitot probe.

In all of the recent fatal crashes involving erroneous system behaviors (Turkish 1951 Colgan 3407, AF447[assumed]), the ultimate cause has been human.

Turkish 1951: Crew was allowed to use the autothrottle with a defective radio altimeter that failed on 8 previous flights / pilot in training was still PF despite a reported instrument failure in flight / pilots not flying the plane, hand on the throttles.

Colgan 3407: Administrative failures / pilots not adequately vetted and trained on transition to Q400 systems / pilot fatigue policies violated.

Air France 447: Regulatory failure / recommendation issued for known issue instead of an AD / AD previously issued for A320 series but not covering A330/340 / piloting into hazardous weather / inadequate training and QRH procedures for the given situation.

[BoeingBobby]

The airbus A330 as with every commercial Airliner flying from the CRJs to the A380 has a RAT that will deploy automatically usually when the main busses fail.

Funny that none of our 27 747's 200's 300' & 400's have RAT's, The 707's and 720's that I flew did not have them. The DC-8 and the CV-880 neither. WTFO did I miss something here? I know that DC-10's and MD-11's do. I seem to remember some of my friends that flew L-1011's did as well. But your statement above is completely erroneous sorry.

BB

FOF, made it to OKA just fine, I am at Yakota AFB now. Thanks for the thought.

[BoeingBobby]

Evan, You are taking my figures of speech to literally. By "Fat dumb and happy" I mean that you are, 2 plus hours into the trip, Auto-Pilot and the FMS are doing most of the work, the Captain is back in the bunk or flirting with the F/A's. Not much to do but talk on the radio every 10 degrees, do a plotting chart, and "Shoot the shit" this is just a way of saying talking to each other. I am sure that you and you friends get together and "Shoot the shit" from time to time. We do it all the time, it keeps us from getting bored or worse falling asleep. The expressions were not meant to be derogatory or disparaging at all. Also I never implied that they over controlled the aircraft by flying manually. If they entered a strong cell at FL 350 unintentionally, or the radar was out and they did, the forces inside a cell like that can exceed almost any airframe limitations in a couple of minutes. You would be shaking around so violently that it would be almost impossible to reach for let alone activate switches. If this is the case, it would not have mattered if the were in a Boeing or anything else less than a NOAA P-3.

BB

[mfeldt]

In this case the idiot was a pitot probe.

In all of the recent fatal crashes involving erroneous system behaviors (Turkish 1951 Colgan 3407, AF447[assumed]), the ultimate cause has been human.

Turkish 1951: Crew was allowed to use the autothrottle with a defective radio altimeter that failed on 8 previous flights / pilot in training was still PF despite a reported instrument failure in flight / pilots not flying the plane, hand on the throttles.

Colgan 3407: Administrative failures / pilots not adequately vetted and trained on transition to Q400 systems / pilot fatigue policies violated.

Air France 447: Regulatory failure / recommendation issued for known issue instead of an AD / AD previously issued for A320 series but not covering A330/340 / piloting into hazardous weather / inadequate training and QRH procedures for the given situation.

This is a rather philosophical matter of interpretation .... of course you can always claim that the ultimate cause is human, the point is that it becomes nearly impossible to avoid human error at the point when the aircraft's computer says: "Oh, the input data is completely confusing, that's too much for me, now you take it! And by the way we have horrible weather outside and only a few hundred metres of altitude left and there's a bunch of other problems as well..."

It is clear that the ever better protection from human error triggers a (false) impression of infallibility, and tends to overstress humans when the systems finally do fail...

And your talking about actual crashes ... so in the LOT and Qantas incidents there were no human errors, because the planes didn't crash in the end? What were the causes for these "incidents" then?

m.

[Evan]

Evan, You are taking my figures of speech to literally. By "Fat dumb and happy" I mean that you are, 2 plus hours into the trip, Auto-Pilot and the FMS are doing most of the work, the Captain is back in the bunk or flirting with the F/A's. Not much to do but talk on the radio every 10 degrees, do a plotting chart, and "Shoot the shit" this is just a way of saying talking to each other. I am sure that you and you friends get together and "Shoot the shit" from time to time. We do it all the time, it keeps us from getting bored or worse falling asleep. The expressions were not meant to be derogatory or disparaging at all. Also I never implied that they over controlled the aircraft by flying manually. If they entered a strong cell at FL 350 unintentionally, or the radar was out and they did, the forces inside a cell like that can exceed almost any airframe limitations in a couple of minutes. You would be shaking around so violently that it would be almost impossible to reach for let alone activate switches. If this is the case, it would not have mattered if the were in a Boeing or anything else less than a NOAA P-3.

BB

Ok, sorry, I get it now.

[Evan]

It is clear that the ever better protection from human error triggers a (false) impression of infallibility, and tends to overstress humans when the systems finally do fail...

No argument there.

And your talking about actual crashes ... so in the LOT and Qantas incidents there were no human errors, because the planes didn't crash in the end? What were the causes for these "incidents" then?

I think the lack of human error in those cases is why they weren't crashes.

The introduction of FBW was supposed to lighten the pilot workload and provide protection against pilot error as well as simplify mechanical design. It was never supposed to fly the plane by itself, or make decisions without pilot supervision. If fact, it is not allowed to make decisions by speculation, only by ratio and programmed values. There is no AI involved. If pilots are doing their job, monitoring their instruments, and really know the systems, they should be able to detect and deal with system failures, i.e. take the plane and handfly it. If not, if systems are overstressing the pilots and confusing their situation awareness at critical moments, the systems have weaknesses that need to be rethought.

But a sensor failure is a bit different from a system failure because it deprives not only the system of vital information, it deprives the pilots as well. Then things get religious.