What really happened aboard Air France 447

[z0mb13]

This is such a chilling read..

This is a sequence of catastrophic mistakes: first why didnt they reroute to AVOID the storm!!?!? is it Airfrance policy to go thru a friggin storm??

Second: WTF was the captain sleeping when the plane is going thru a FRIGGIN STORM??

ANd yes Bonin is an idiot.. but anyone couldve made the same mistake as him in dire conditions.

 

[iamwiz82]

Moral of the story:
Bonin was a fucking moron, the senior officers didn't remove him from control, and fly Boeing.

No one recognized the problem until way way way too late. This isn't a tale of idiocy, it's a tale of way in which the human mind panics. Not everyone can handle stressful situations in stride.

 

[angminas]

cliffs?

1. If you're not sure whether your instruments are right, at least find out what the other guy is and has been doing.

2. Teach your pilots the meaning of the word "stall".

3. Don't fly directly into the worst weather in the region without even exploring the possibility of going around.

4. Don't ignore alarms. Even incorrect alarms went off for a reason.

And probably half a dozen more.

 

[KK]

If you read the article, it explains the entire thing. Basically, the senior captain was absent from the cockpit for most of the crisis, and one of the copilots screwed up pretty badly.

Still does that mean the copilots are totally incompetent? I would doubt it, I mean they would have to have many a many hours flight time. Sure not as much as the pilot had, but still enough to handle the situation.

As for the air speed and ground speed, yes they can vary greatly, but what is the cruising speed of that plane vs. what the possible max tail/head wind in that storm could have been? I dunno, I would think stall warning the first thing anyone that has done 10 hours in a general aviation plane would be able to tell you the first thing you do is push the yoke forward.

 

[911paramedic]

As the stall warning continues to blare, the three pilots discuss the situation with no hint of understanding the nature of their problem. No one mentions the word "stall."

Even I know that you lower the nose if you begin to stall, especially if the alarm has been going off for nearly, what, fifteen minutes or so. Wow.

 

[LTC8K6]

Stall warning is inhibited at IAS of less than 60 knots, I believe. This is to silence it when you are landing, I think. It would also be inhibited with clogged pitot tubes, since the IAS might read below 60 knots. Or in this case, it might actually have been less than that at times.

What was happening, in part, is that when the pilot made a correct move, he gained some air speed, making the stall warning active again.

So, he would bring the nose down a little, get the IAS above 60kts, and he'd hear the stall warning, and get confused.

He was too inexperienced or confused to realize the stall warning would go away again if he just kept going in the right direction with the nose.

The IAS was only incorrect for less than one minute, actually. The tubes were clogged by heavy icing conditions, but the pitot heaters got them clear again in about 50 seconds.

If these guys had simply sat on their hands for 60 seconds, everything would have returned to normal, and they could have re-activated the auto-pilot.

 

[z0mb13]

The problem with this is because pilots are never put in such real life dire conditions when their lives (and countless many other lives) are at stake. Yes they do simulators but if they fail they still live.

IMO pilots should be retested periodically (once every year) to make sure that they remember their training. Make them go through flight simulator test in dire conditions, and if they fail they should be punished severely (like losing their pilot license for a certain period).

ANd yes AIrfrance should be sued due to their poor decision making.

 

[Bignate603]

With all the instruments they have nowadays, I don't see how thats possible. If you got a GPS in there, it'll give you the ground speed and altitude. I just don't know how they could go down with only the pitot tube getting plugged, just doesn't add up.

It wasn't the instruments that was the problem, the pilots didn't correctly interpret the information that they had and that confusion caused the plane to crash.

 

[alfa147x]

I'm guessing someone should have listened to the alarms... wow.

Good read. Thanks OP

 

[skyking]

This is what I posted in an earlier thread:

I hope to shed some light into the kinds of situations that lead to an accident of this nature.
The Pitot system compares the pressure coming into the tube due to forward motion to the static air pressure. If the Pitot freezes over or is otherwise blocked, the airspeed will not change unless the static pressure changes, usually due to a change in altitude.
A decrease in altitude will result in higher static pressure.
As the static pressure increases the indicated airspeed will go down. This leads to a perceived slow speed condition which can be very hard to interpret if the aircraft is in clouds. The attitude instrumentation may indicate a level flight, but the airspeed and stall warnings may not.
In the opposite scenario, the pitot is not blocked but the static ports are. as the plane descends the lower pressure on the static side will cause the airspeed instruments to indicate an ever-increasing airspeed. I am willing to speculate it was a failure on the static side.
In this scenario the auto throttles would disengage to prevent an overspeed. Check.

The stall warning system may be triggered by an AOA vane. as the pilot flying takes over, he lost airpeed indication, and then it came back. If the plane has descended with a blocked static system, the airspeed is rising rapidly towards the Maximum Mach Operating speed. He will cut throttle and raise the nose, even though his actions are causing stall warnings. Exceeding MMO is simply that dangerous, and it can lead to fixation.
I speculate that both pilots locked into the pegged airspeed indication and did not interpret the conditions properly.
In flight training with the old conventional systems, we learn to trust that a certain aircraft configuration (flaps, gear, spoilers, etc) and a certain power setting (RPM, Manifold pressure, etc) will result in level flight at a given airspeed. The only variables that will effect that are airframe ice or faulty power indications. It is the fallback method to cope with all the possible flight instrument failures. Set the engine(s) to this, you get this.

I had a 50-50 chance, I blew it.

 

[BladeVenom]

hmm so no force feedback for the same function stick on airbus craft? would have helped too if the guy said 'i was pulling back the whole time'

That has to be the most awful control scheme possible.

I'm never getting on an Airbus.

 

[Gothgar]

I find it weird that the alarms were supposedly blaring about the stall for so long, yet not one person said anything about the alarms or the possibility of the plane actually stalling till it was too late. I wonder if there was a short or a bug that caused the alarm to not go off, but that still made it seem like it was registered as going off to the black box.

Seems conspiracy-ey

 

[foghorn67]

I find it weird that the alarms were supposedly blaring about the stall for so long, yet not one person said anything about the alarms or the possibility of the plane actually stalling till it was too late. I wonder if there was a short or a bug that caused the alarm to not go off, but that still made it seem like it was registered as going off to the black box.

Seems conspiracy-ey

Jeez, it explains that. The rookie thought he could ascend above the clouds.
The FC computers went from normal law to alternate law, allowing them to continue climbing.
By the time anyone got their bearings straight, it was too late. IAS warnings shut off at really low speeds, assuming you are landing.
A catastrophe is not one single event. It's a chain reaction of events, and removing one will likely avert it.

I am not going to say I told you so. But Boeing engineers always preferred the two yoke system being tied together.
There are plenty of advantages to the Airbus flight stick system, but it clearly didn't help much here. Not sure if anyone imagined all these events happening. And as much as a Boeing fanboy I am, I will never think AB's flight control system is to blame.

 

[alfa147x]

Why did it take the captain so long to come up to the cockpit?

 

[z0mb13]

Why did it take the captain so long to come up to the cockpit?

He was getting a frenchie?

 

[RavenSEAL]

Damn, that's a tough read. I might not know jack about airplanes, but it sounds to me like the pilots were just blatantly stupid.

 

[Born2bwire]

Why did it take the captain so long to come up to the cockpit?

If you read the time stamps it took the captain only a minute to get there. In fact, the whole thing happened over the course of around four and a half minutes.

 

[alfa147x]

If you read the time stamps it took the captain only a minute to get there. In fact, the whole thing happened over the course of around four and a half minutes.

Wow. I missed that. Reading it made it seem like a good 15 - 20 mins

 

[SKORPI0]

Video recreations, animations...

Air France Flight 447's harrowing end

Inside AirFrance Flight 447 cockpit

Air France Flight 447 Recreation *UPDATED*

TV documentary - The Crash of Air France Flight AF 447 - part 1 of 4... look for other parts.

 

[KillerCharlie]

hmm so no force feedback for the same function stick on airbus craft? would have helped too if the guy said 'i was pulling back the whole time'

Airbus controls are different in 2 respects. First, the pilot and copilot sticks move completely independently. One can be pushing down while the other pushes up. Boeing's yokes move together. Second, they are not back driven. Back in the day (and still today with some airliners), you had physical feedback from the mechanical control systems. You lose that with fly-by-wire, but Boeing puts in motors to artificially provide feedback while Airbus provides no feedback.

Another thing, noted in the second page of the article, is that Airbus's fly-by-wire control systems typically prevent you from doing bad things like stalling - this may be why he's pulling back on the stick. However, if something bad happens or for whatever reason you revert to direct mode, you can stall. Boeing control systems typically still let you stall the airplane with fly-by-wire.

Personally I am not a fan of Airbus's control systems (based on what I know about them), but since I'm an aerodynamicist for Boeing I'm pretty biased. Airbus has had a few crashes

 

[Mike Gayner]

Half of the accident reports I read about in commercial airlines seem to go back to faulty/iced pitot tubes. Why are they still using this technology? It obviously doesn't work.

 

[KillerCharlie]

Half of the accident reports I read about in commercial airlines seem to go back to faulty/iced pitot tubes. Why are they still using this technology? It obviously doesn't work.

What alternatives are there? Pitot probes have worked forever, you just have to keep them heated. The original Goodrich pitot probes on the A330 worked fine, but they replaced them with European pitot probes made by Thales that had known icing problems. If this popular mechanics article is correct, ultimately the pilots botched it.

There is really no other way to measure airspeed. The only other alternative I'm aware of is laser measurement sometimes used in wind tunnels, but they probably have a ways to go to before being used on airplanes.

That brings me to another thought. I think on Boeing airplanes there is logic with the multiple pitot tubes and static ports that try to determine if one of the pitots is bad. If 2 of the probes are reading 200 knots and one is reading 10 knots, I think it throws out the bad reading. From the popular mechanics article, the airspeed from one of the probes gets displayed on pilot's side and the other on the copilot's side. I'm not sure if that's actually the case with this airplane or how Boeing and Airbus actually differ with their air data systems. I don't think the airspeed indicator failure was really the problem here.

 

[Scouzer]

Even I know that you lower the nose if you begin to stall, especially if the alarm has been going off for nearly, what, fifteen minutes or so. Wow.

It was 5 minutes from beginning of the problem to death.

Bonin was a relatively inexperienced pilot, perhaps he shouldn't have been flying with Air France quite yet. That being said, Robert made the critical error of not taking clear, organized control of the lesser experienced pilot.

The captain then makes the mistake of taking his break right before flying into shit weather, and all mistakes combine for disaster.

 

[OVerLoRDI]

Good read. Incredible the mistakes that were made.

The thing that confuses me is pulling back on the stick while your altitude is dropping. From the pitch and your rate of ascent it should have been obvious you are actually in danger of stalling or are stalling. Also wouldn't they have felt the acceleration as the plane began falling and realized what is happening? It just sounds all too weird.

 

[sportage]

It doesn’t take long for a chunk of metal to fall to earth. That’s for sure...
I feel the whole issue here was flying into that storm.
As was said, all other flights were routed around the storm.
These guys did not have the experience to fly through violent storms.
And that is/was why the other flights did not attempt to do so, or even risk it.
Maybe I missed the reasoning, but why they chose to fly into the storm was their first and last fatal mistake.
And some of the pilots being clueless as to the St. Elmo's fire effect was a bit un nerving.
Where did they go to flight school? Wal-Mart? Or out of some "flight school for dummies" guide?
I took many small craft flight lessons back in the 1970's. And the first thing drilled into your head is about that little and common fatal thing called "stalling".
Spacing out during any confusion or distraction, and putting the plane into a stall, is one of the most common pilot errors. The cause of a huge number of small craft crashes.

Another common pilot error is flying at night with no sense of the horizon. Getting so disorientated, they fail to trust their instruments as to whether they are flying up, down or level. Then end up actually pulling up into a stall, or flying pitched down, and hitting the ground.
Even though every instrument plainly shows the craft "is" flying level, the human senses insists the plane is not flying level, and the pilot ends up stalling or flying into the ground.

That was exactly what happened to JFK Jr when he crashed his small plane, killing all aboard.
Flying at night, not trained for night flight, and became disoriented because there was no visible horizon.
Pretty cocky and fatally stupid of the rich, late, famous Kennedy clan member.