Air France jet crashed nose-up after 4 minute ordeal

Reuters/Yahoo ^

[nuconvert]

PARIS (Reuters) – Pilots wrestled with the controls of an Air France airliner for more than four minutes before it plunged into the Atlantic with its nose up, killing all 228 people on board, French investigators said Friday.

The 2009 emergency began with a stall warning two and a half hours into the Rio-Paris flight and nine minutes after the captain had left the cockpit for a routine rest period.

The Airbus A330 jet climbed to 38,000 feet and then began a dramatic three and a half minute descent, rolling from left to right, with the youngest of three pilots handing control to the second most senior pilot one minute before the crash.

[freebird5850]

I’ve always been wary of fly-by-wire controls. No “feel” as to what the plane is trying to tell you.

[Blueflag]

Here’s another take on the same story.

Reveals that at one point both pilots had their hands on their (left and right seat) ‘joystick’. In the old days, two people might have been needed to wrestle a cable- or hydraulically-controlled aircraft back to controlled flight. TWO people trying to fly a fly-by-wire system doesn’t sound good.

Read on - another perspective:

http://www.cnn.com/2011/WORLD/americas/05/27/air.france.447.crash/index.html?eref=igoogledmn_topstories

[Tzfat]

The computer locked the pilots out of the controls and had a 10+ degree positive nose pitch while flying down the runway and crashed into the trees at the end.

The "computer" did exactly what it was designed to do in that case. Disinformation about the "computers" in Airbus aircraft is almost comical. I have flown the Airbus family aircraft for over 10 years. They are well designed and well built.

[Tzfat]

Reveals that at one point both pilots had their hands on their (left and right seat) ‘joystick’. In the old days, two people might have been needed to wrestle a cable- or hydraulically-controlled aircraft back to controlled flight. TWO people trying to fly a fly-by-wire system doesn’t sound good.

The issue is not fly-by-wire, but the fact that the sidestick controllers are not linked. Some fly-by-wire aircraft have interconnects.

You are correct about it being bad, however. The two inputs are algebraically summed. That means they can either fight against each other, or worse, dramatically over control in the same direction(s).

Bright lights and aural warnings tell pilots when both are providing input, as well as when one pilot overrides the other. Basic Airbus pilot stuff. Sounds like inexperience at least on the part of one pilot.

[Blueflag]

Not sure I’d call it a hatchet job, but ... here’s the WSJ/Fox version of the same story. DOES make the junior crew look like they reacted incorrectly.

*IF* the artificial horizon was FUBAR I can understand how they might get (AND KEEP) the aircraft in a 35 degree AOA **after*** a stall warning, but damn ...

http://www.foxnews.com/world/2011/05/27/air-france-captain-absent-descent-began/

“Throughout the descent, according to the report, “inputs made by the [pilot flying] were mainly nose-up” and the “angle of attack,” or the position of the longitudinal axis of the plane in relation to the airflow “remained above 35 degrees.”

Serious stall.

” ... apparently confused by repeated stall warnings, pilots of an Air France jetliner in 2009 continued to pull the nose up sharply—contrary to standard procedure—even as the Airbus A330 plummeted toward the Atlantic Ocean”

and ...

“... paints a somewhat unflattering picture of a seemingly confused cockpit, with the crew making extreme inputs to their flight controls and the engines spooling up to full power and later the thrust levers being pulled back to idle. At one point, according to the report, both pilots sitting in front of the controls tried to put in simultaneous commands.”

Not good to try to fight the controls in a fly by wire tandem.

Read more: http://www.foxnews.com/world/2011/05/27/air-france-captain-absent-descent-began/#ixzz1NZ2OpLs7

[DManA]

I believe Airbus has quietly gotten all airlines to replace their original pitot tubs with an improved version since the crash. The original was susceptible to freezing up.

[SeeSac]

Honestly, ground speed & wind direction are all that I need

Baloney.

[Blueflag]

Indeed. I didn’t know that the inputs were ‘summed’. Yikers. (I fly slow movers, VFR, below 5000 AGL generally)

The various articles read like the aircraft Q-cornered, and basically performed a “falling leaf” maneuver, descending at roughly 10,000 FPM.

Several articles had suggested this as a pilot input induced “deep stall”, but how does a non-T-Tail jet get and STAY in a deep stall?

[DManA]

I saw a tv show on this. They were able to recreate the situation in the cockpit. Older pilots did the right thing (Add power, lower then nose?) Younger pilots all did what the pilots on that plane did. They concluded that pilots were not being trained for this situation (which should never have occurred in the first place). All pilots of this plane were given remedial training after the crash.

[CodeToad]

“that is largely driven by GPS input”

Indicated Airspeed in never derived from GPS position data. Indicated Airspeed must be driven by measured air over the wings.

[John S Mosby]

Controls on Airbus are single hand, like a computer game. Why Airbus’s are routinely referred to as “PacMan”. The so-called brain locking out pilot control is recurring event in many crashes. Airbus should be sued out of existence.

[Tzfat]

I’ve always been wary of fly-by-wire controls. No “feel” as to what the plane is trying to tell you.

Any aircraft with significant control loading does not provide "feel" anyway. Hydraulic controls provide artificial feedback. Of all the jet airliners flying today only, the DC-9 provides actual flight control feedback.

As for fly-by-wire, a number of modern fly-by-wire airliners provide artificial feedback, so the issue is not fly-by-wire. I have flown fly-by-wire with feedback and others without. Most pilots will tell you that the difference is negligible in "feel" because after a while in an aircraft, your brain provides the feedback. It is weird the way it works.

[r9etb]

You are correct about it being bad, however. The two inputs are algebraically summed. That means they can either fight against each other, or worse, dramatically over control in the same direction(s).

Summing the inputs would have been a conscious design choice ... but it seems like a really odd thing to do, for precisely the reasons you state.

Is there some advantage to summing the inputs?

[CodeToad]

“Air France uses “pilot error” to explain a lot. “

Yes, they do. Typical euroweenie mentality. I had a BMW motorcycle a few years back that would leave third gear for neutral while under power. I was told by the BMW tech that I didn’t know how to ride motorcycles. It seems this transmission issue is common with BMW but it is pilot error anyway.

[stubernx98]

Imagine that as you are driving in traffic, and your GPS map goes to blue screen and you have to re boot it, while texting a message to the software company for a solution to the problem. then people start honking at you as you are waving all over the road, and a cop with flashing lights is behind you. Oh yes you are on cruise control and it won’t disengage.

[Blueflag]

It dismays me that any pilot would react to a stall warning by pulling the nose up, TV show or not.

That is akin to stepping on the GAS when a child starts to run into the street.

NOSE DOWN is so BASIC to pilot training that I am incredulous that BOTH jet jockeys would input and maintain a nose up attitude.

Unless they somehow thought they were in an overspeed (I never have to worry about that in the Champ, really) condition and were trying to lose KIAS, I dunno.

But again, when you get a stall warning (horn or stick shaker of just the seat of your pants) you just NEVER pull back on the ‘stick’. A stalling plane is crying out for airspeed over the wings and the return of lift to the wings and control surfaces. WHY react with nose up is beyond me.

REM: I don’t fly jets. These guys LAND faster than our plane will fly. So i speak from a low and slow perspective.

[John S Mosby]

Intl. overseas flight. He went back to sleep in the qtrs, so he’d be pilot on landing in France. Routine. What’s not routine is why did the computer take over, and why can’t they override it. Either way, if the airspeed wasn’t right, they couldn’t very well fix it, and the computer uh, would insist it was right.

[Tzfat]

Controls on Airbus are single hand, like a computer game. Why Airbus’s are routinely referred to as “PacMan”. The so-called brain locking out pilot control is recurring event in many crashes. Airbus should be sued out of existence.

You are seriously misinformed. Holiday Express, not withstanding. As an airline pilot for 30 years, and over 15,000 hours, I can tell you that the Airbus is well designed and well built. "Boeing snobs" are people who have never flown anything else. I have 8,000 hours in Boeings as well.

[DManA]

Not a pilot so some of what they said went over my head. The way they explained it, the thing the pilots who screwed up in the simulator did was not completely irrational but did seem to make sense if you hadn’t been trained.

The key to the whole situation was the pitot failure. Generated a cascade of errors. Or so said this tv show.

[Blueflag]

Folks who might criticize the Airbus design might also want to equally harp on the F-16 — one of the world’s finest machines to ever surround a pilot. It too flies via a “PACMAN” controller ;-)