Air France Flight 447

[captainslow]

This is one of the best articles I've read in a long time. It's a bit long, but a quick read. It's about the investigation into the Air France flight (447) that went down (mysteriously) en route to Paris from Rio.

Investigators are closer to finding a cause, thanks to an underwater investigation by the same people who found the Titanic , but among other things, the article describes a possible conflict of interest between the French investigators and Air France and Airbus, which are partly owned by France (and which are the companies funding the underwater investigation).

The description of the desolation of Tasil Point and the underwater mountains below it is fascinating, but there were several aspects of the story, particularly the interview with the Brazilian coroner, that I found quite disturbing. Not least was the possibility raised that there could have been a failure of the air speed probes (called pitot probes) which would have resulted in a total autopilot shutdown. According to the article, even though pilots are trained to fly manually, their ability to fly safely depends on their speed at the moment of failure, and dirt and ice are just enough to take out the pitots. Apparently, a bug's nest growing in one pitot probe led to a plane crash off the Dominican Republic in 1996, killing 189. The particular model used on the Air France plane was supposed to have been replaced with a more reliable model due to frequent icing problems which caused the pitots to fail. Little things like this make me nervous. It's not that I lack faith in technology - I just have tremendous faith in human laziness.

http://www.nytimes.com/2011/05/08/magaz ... l?_r=1&hpw

In any event, it's worth reading.

And not to get too depressing, it got me thinking about a plane crash in the 1980s that happened not too far from where I currently live that I read about recently - http://en.wikipedia.org/wiki/Air_Florida_Flight_90 Such a tragedy but the bravery is quite inspiring. Well, I still love flying and I don't mean to scare anyone. Hopefully, we'll have the reasons for the Air France crash soon enough so that any necessary fixes can be made to Airbus fleets around the world.

[Mr Kleen]

I watched a great documentary on Flight 447 last year. Their conclusion was blockage of the pitot tubes as well. With the probes blocked, the autopilot thought the plane was loosing airspeed (stalling) so it put the plane into a dive to get back to the necessary speed. With the tubes blocked the autopilot system never saw an increase in airspeed so the plane hit the water at full throttle. At least it was quick...




So, who want's to fly over for a visit?

[zaxrex]

Oh, me, mememe. Pick me!

Funny, I have the article IRL and was saving it to read.

[Raven]

I watched a great documentary on Flight 447 last year. Their conclusion was blockage of the pitot tubes as well. With the probes blocked, the autopilot thought the plane was loosing airspeed (stalling) so it put the plane into a dive to get back to the necessary speed. With the tubes blocked the autopilot system never saw an increase in airspeed so the plane hit the water at full throttle. At least it was quick...

I'm confused as to why no one noticed the auto pilot going nose down.

[Mr Kleen]

The tube clogged and the autopilot immediately sent the plane into a nose dive to regain airspeed.
The crew noticed.
The passengers noticed.
The Atlantic ocean did not.

[scheherazade]

These things are hard to imagine for 'regular folk', but airplanes in low visibility, without instrumentation, can be a hell of a thing to control.

There's false gravity, and you really don't know which way is down.
http://www.youtube.com/watch?v=H4BpzIY6 ... re=related
http://www.youtube.com/watch?v=ZLg94fFc7k8

Combine that with cloud cover that touches the ground, and you can fly straight into the ground and never even realize that you weren't going straight and level.



If you have a virtual horizon, it gets easier.
Some clues to an approaching stall can be changes in pitch.

If you're nose is dropping, and you need to pull up to stay level, you're too slow.

If your nose is pitching, and you need to push back to stay level, you're too fast - AND you're most likely fine. However this is an issue for trans sonic flight where the airframe is not made for the speed you reach. Air flow is too disrupted if the airfoils aren't made for it, and you lose control authority.
A good way to lose speed is to climb. Can't really climb if you can't control where you're going.

An airliner cruises ballpark 700 knots. Air is thin so it can't slow down too much without losing lift and altitude (which is a non issue if you have the fuel to plod around slowly at lower altitude).
You can't speed up much because you're ballpark 50 knots from supersonic.

Stupid modern planes will self-trim themselves - and you don't even know if the nose is drooping or pitching, because the airplane will auto level itself. You'll only ever see the nose droop/pitch if it's SO friggin bad that the airplane can't handle it - which means that by then YOU can't handle it.

Although, if you have the altitude, it's better to go too slow than too fast. You can always speed up, but you can't slow down a supersonic diving couple hundred thousand lb aerodynamic anvil.
(A slow speed stall [optionally with spin] [with lots of altitude] is actually pretty 'meh'. push the stick, counter rudder, wait for it to pick up some speed, level off. You can rinse and repeat that all day.)

Note that in the article, the coroner mentioned that people's fractures were consistent with a belly flop in the 'crash position'. And the airframe was damaged mostly on the belly - but largely in-tact. Indicating a hard water landing, and indicating that some people could have survived, and floated around [damaged and crippled] till the water took them.

-scheherazade

[zaxrex]

Umm, you are about 300 miles an hour too fast there buddy.
Do not try to recover a spinning A330 like it is a C-172 or a Piper. The wingspan is so large that the inside wingtip and control surfaces can actually be flying backwards, not just stalled, but counter-reactive to inputs.
The crew has to de-couple the throttle balancing and the aileron interlock, pop full upper panels, keep the rpms of the inside engine down to avoid low pressure turbine stall from the uneven inlet air and hope that the descent angle and rate is not great enough for the jet wash to make the tail surface controls twitchy and bilaterally amplified.

But regardless, the article would leave you the believe that the crew had some sort of control rather than an uncontrolled descent.

I wonder how long it will be and how much it will cost to do a root cause failure analysis. Oh wait, they can't call it that...

[scheherazade]

Umm, you are about 300 miles an hour too fast there buddy.
Do not try to recover a spinning A330 like it is a C-172 or a Piper. The wingspan is so large that the inside wingtip and control surfaces can actually be flying backwards, not just stalled, but counter-reactive to inputs.
The crew has to de-couple the throttle balancing and the aileron interlock, pop full upper panels, keep the rpms of the inside engine down to avoid low pressure turbine stall from the uneven inlet air and hope that the descent angle and rate is not great enough for the jet wash to make the tail surface controls twitchy and bilaterally amplified.

But regardless, the article would leave you the believe that the crew had some sort of control rather than an uncontrolled descent.

I wonder how long it will be and how much it will cost to do a root cause failure analysis. Oh wait, they can't call it that...

You're right, I should have said mph not knots. (And even so, I rounded up from the mid 600's I've seen simply while traveling back and forth cross atlantic).

I think you're describing more of a tight/flat spin than a stall.
You can be going 100+ mph forward, stall and drop a wing, without any of that drama.

-scheherazade

[zaxrex]

You described a spin recovery for a light GA platform.
I regurgitated parts of what the deadheading right-seater next to me said was the procedure for spin recovery for a modern passenger jet. Thought it had a little more bearing on the topic at hand.

[scheherazade]

You described a spin recovery for a light GA platform.
I regurgitated parts of what the deadheading right-seater next to me said was the procedure for spin recovery for a modern passenger jet. Thought it had a little more bearing on the topic at hand.

Understood.
I hadn't had anything that dramatic in mind, but I see how I implied it.

-scheherazade

[Mr Kleen]

I love this place.

[Raven]

Ok, so it would be hard to do recover from a serious problem...