Report Details Last Minutes of Doomed Air France Jet

[alp227]

Source: The New York Times

PARIS — The commander of an Air France jet that crashed two years ago was in the cockpit, though not at the controls, as the crew struggled to recover from a high-altitude stall that occurred as the plane entered a zone of moderate turbulence over the Atlantic, according to the initial chronology of the flight’s final moments published Friday by French air accident investigators.

The report confirmed investigators’ initial hypothesis that a loss of consistent speed readings — likely due to icing of airspeed sensors on the plane’s wings and fuselage — set off the chain of events that ultimately led to the crash of Flight 447 on its way from Rio de Janeiro to Paris. But they said it was far too early to say whether the crash was the result of errors made by the pilots or a technical malfunction, as the plane plummeted from cruising altitude into the ocean at 11,000 feet a minute, a rate of descent several times faster than even the steepest descents in normal flight.

“We are still trying to interpret this information in order to have a better understanding of what happened,” said Jean-Paul Troadec, the director of France’s Bureau of Investigations and Analyses, or BEA. “That work has just begun.”

“Our role is not to assess blame, merely to determine the facts,” he added.

Read more: http://www.nytimes.com/2011/05/28/world/europe/28flight.html

[swishyfeet]

1. Very interesting comments

Much more informative than the article itself.

Edit: ("highlighted" comments that is)

[cosmicone]

2. Hate to be a Monday moning quarterback but

in addition to the Pitot tubes, it was pilot error.

Instead of trying to reboot the autopilot, they should have disabled it and maintained the pitch/throttle combination required to prevent a stall. They should know this combination for their particular aircraft by heart.

In a simulation of this tragedy in simulators, 97% of the pilots successfully recovered control of the aircraft and prevented the loss. (Saw it on a Nova or a Discovery documentary, don't remember which)

Having said all that, it must have been a horrible 4 minutes for those poor souls in the cockpit. May they rest in peace.

[Peace Patriot]

3. I think I understand what you're saying, but I still don't understand how the plane

could take such a steep dive, in "moderate turbulence," in the few minutes they took to try to reboot the autopilot. Seems like it just nosedived for the ocean. I thought pilots of most aircraft have more time than that, even if they've lost all power, to glide. Otherwise, rebooting the autopilot would be forbidden, right? (--if rebooting the autopilot would permit the plane to just fall out of the sky?) WHILE they're rebooting the autopilot, surely they have control of the plane. And if they have no power, they still have some control--attitude control and wind cushion, don't they? They can keep the plane level and maybe even glide to a landing. (Hard, destructive landing in those conditions, but still a landing of sorts--some might survive it.) What would cause all attitude (level flying) to be lost, power to fail (presumably), all control lost with the plane just turning off and steeply diving to destruction? Is it possible that "moderate turbulence" was just an average, and that some stray hurricane force wind hit the plane just at the wrong moment?

Thanks for bringing your expertise to this matter!

[cosmicone]

4. Sorry, I am not a pilot and have no flying experience

however, the plane did NOT nosedive ... it landed flat on its belly and remained horizontal with the nose slightly higher throughout its descent.

I was only addressing the physics ... the engines stall and lose thrust if the throttle is too high or too low. (Just as moving your car from a slippery surface) So there is a correct pitch and throttle combination that prevents a stall under all conditions and this combination is unique to each model of the aircraft. (I learned this from a documentary)

If the pilots had employed that immediately, the plane would have continued flying and out of the storm.

[Peace Patriot]

5. I see. So it did glide a bit and achieved a sort of "landing" (didn't nosedive).

It then probably broke apart and sank, and if anyone got out, they got drowned in the storm. But I still don't understand how rebooting the autopilot could have caused loss of control. Maybe the answer is simply panic--failure, as you say, to maintain the correct pitch and throttle combination while they tried to get the autopilot back up--in the very unfortunate circumstance that the lead pilot had gotten up to go pee or something.

I'm wondering now about the backup systems on commercial airliners for when the engine stalls and the plane starts descending. That's another puzzle in this terrible event. Why couldn't they restart the engine and gain altitude? A "stall" is not necessarily fatal. (You also pointed that out--most pilots can deal with it.) And rebooting the autopilot shouldn't interfere with the plane's safety systems. Say, panic in the first minute or so, co-pilot failure to keep his head, does a few wrong things, but there is still time to come out of a stall. Why couldn't they?

I'm thinking electronics failure (software) and/or electricity failure (hit by lightning?)-perhaps not loss of electricity but a surge hitting and zapping software or some critical aspect of software. Conditions were really not that bad. So "rare event" comes to mind--a particularly bad lightning strike that somehow hit home, a hurricane force gust or swirl, a rare magnetic event that blew parts of the safety software all hooey?

Well, I guess we just have to wait for further reports from the investigators.

Thanks for your in-put, even if you aren't an expert. Was that Nova (or whatever it was) about THIS accident--or just flight safety in general?

[cosmicone]

6. It did not have a soft landing.

It landed horizontally, but at full terminal velocity. The impact created forces >36G and there was no chance of survival for anyone. This was not like "miracle on the Hudson" situation.

Secondly, Airbus planes are run by computers. It is not simply an autopilot that failed, the entire system failed due to critical error. It is a safety issue -- it means the system can no longer be trusted because its sensors are unreliable.

It is almost like the computer saying, "hey humans, I have done whatever I could. I can't handle it anymore. You humans take over."

In the simulator study that they showed in the documentary, almost every pilot let the computer be disabled, took over controls and manually set the pitch and the throttle combination to escape the storm and not cause a stall.

4 minutes at 35,000 feet is not very much time to take control of the aircraft but is adequate if the pilots handle what they are taught in a steely cold efficiency. Wasting it on rebooting the computer, when the computer had signed off, and not setting the proper throttle/pitch combo was human error.

Perhaps there are some pilots on DU who could shed light on this. My knowledge gleaned from a documentary can only go so far.

I'll post a link to this thread in GD to attract pilots.

[Peace Patriot]

7. Thanks! Yeah, I just re-read a news article about it and "got" how fast they fell

and how hard they hit the water. And I see what you're saying. Sure looks like they waited too long to take over and correct.

It said that the main pilot was in the cockpit but not in his seat. First report I read said he was somewhere else and entered the cockpit after the crisis began. Maybe both are true--at different moments. For what it's worth to his survivors, it looks like it wasn't his error, but, sadly, the co-pilot's family may have to bear this burden. The whole thing is very sad and scary, too. We forget how dangerous our technology can be--whether it's cars, planes, having electricity and gas lines in houses, or a thousand other dangers we face every day from our otherwise awesome engineering abilities and progress. One slip-up, and we and everyone we love could be dead. It's probably a lot better than having to deal with kimodo dragons (just saw that episode of David Attenborough's "Life"--Wow! They are some awesomely nightmarish beasts!), or other natural dangers in the pre-industrial, and, before that, the primitive world, but still--we seem to create our own kind of dragons, that sometimes eat us alive.

May all those poor, terrified people rest in peace!

:grouphug:

[cosmicone]

8. An Airbus pilot responded and posted this

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=439x1177963#1178405

[Trajan]

11. Excuse me ... It DID nose dive ...

There is evidence of a sink rate of 11,000 feet per minute, which is an awful nose down attitude of more than 120 miles per hour STRAIGHT DOWN .... I am sure there was a horizontal component that formed a 'slope' ... but when we speak of 11,000 feet per minute, we are speaking 'nose dive' ....

You are right about this 'pitch/throttle' relationship as being the proper procedure for recovery .... but they failed to recognize their problem until they had already stalled, and they failed to recovery ...

The fact that the plane landed 'flat on it's belly' is a fluke of the end game here .... The plane, by that time, was in 'high speed stall', with the plane's flight path most likely at a steep slope, even though the aircraft had actually rotated at a steep angle from it's actual flight path angle .... This is a definition of a high speed stall ....

They needed to recover from the stall at 30,000 feet, which would have required a 'NOSE DOWN' configuration at the stick .... it is my understanding they did NOT move the stick downward/forward ....

[Peace Patriot]

13. Just want to say my comments may be kind of dumb because I didn't realize it was an Airbus

and what that means, as to lack of pilot control including having no "feel" in the throttle. I wasn't aware of the problems in these "fly by wire" (software controlled) airplanes. In short--from reading some comments by pilots at the other thread--in this largely automatic airbus system, the co-pilots may have been getting bad info from all their (computer controlled) indicators and may also have had trouble wresting control of the airplane from the computer. It may not have been pilot error, but rather a badly designed control system overly weighted toward automation and against pilot judgement which not only made it hard for the pilots to take control but ALSO gave them wrong information, with them having to act against all indicators, to save the plane. And they were especially handicapped by having a "joystick" instead of real throttle. I don't know if this is what happened but it adds a new dimension to this terrible accident--one that I hadn't considered. (Sloppy, inattentive reading on my part.) (My inattentive understanding was that it was a commercial airliner with the standard pilot control/autopilot system, where the autopilot is either on or off, but not confusingly in between and unable to be shut off.)

We only have to consider the 'TRADE SECRET'-programmed voting systems we now have, all over the U.S., to know just how badly awry human activity can go, at high speeds, in computer systems over which we have no hand and visual control.*

--------------------------

*(Granted, the airbus was not likely programmed by malevolent far rightwing billionaires, as our voting machines are, but the principle is similar. With hand-counted paper ballots, seen by human eyes and counted by human hands, at human speed, you may end up with a few scumbags in Congress, by miscount or by ballot box stuffing, but not a whole Congress full of scumbags (s)elected secretly, in invisible systems, operating at the speed of light, across a whole country.)

[Blue_Tires]

9. The report said when the aircraft first stalled

the pilot flying did nose-up inputs? isn't that the one thing you're NEVER supposed to do?

[RebelOne]

10. True. I was flying small planes many years ago.

And if your plane is beginning to stall, you point the nose down.

[Trajan]

12. Yes ... recovery from a stall

Requires that you place the elevator controls to move the nose of the plane DOWN until the airflow over the wing surfaces "normalizes" and smooths, after which the nose of the plane is then 'eased' upward to slowly achieve a straight and level attitude without breaking the airframe ....

According to what I have read, they did NOT perform the basic stall recovery procedure ...