• Thanks for stopping by. Logging in to a registered account will remove all generic ads. Please reach out with any questions or concerns.

What Really Happened Aboard Air France 447

observor 69

Army.ca Veteran
Inactive
Reaction score
3
Points
430
What Really Happened Aboard Air France 447
Two years after the Airbus 330 plunged into the Atlantic Ocean, Air France 447's flight-data recorders finally turned up. The revelations from the pilot transcript paint a surprising picture of chaos in the cockpit, and confusion between the pilots that led to the crash.

For more than two years, the disappearance of Air France Flight 447 over the mid-Atlantic in the early hours of June 1, 2009, remained one of aviation's great mysteries. How could a technologically state-of-the art airliner simply vanish?

With the wreckage and flight-data recorders lost beneath 2 miles of ocean, experts were forced to speculate using the only data available: a cryptic set of communications beamed automatically from the aircraft to the airline's maintenance center in France. As PM found in our cover story about the crash, published two years ago this month, the data implied that the plane had fallen afoul of a technical problem—the icing up of air-speed sensors—which in conjunction with severe weather led to a complex "error chain" that ended in a crash and the loss of 228 lives.

The matter might have rested there, were it not for the remarkable recovery of AF447's black boxes this past April. Upon the analysis of their contents, the French accident investigation authority, the BEA, released a report in July that to a large extent verified the initial suppositions. An even fuller picture emerged with the publication of a book in French entitled Erreurs de Pilotage (volume 5), by pilot and aviation writer Jean-Pierre Otelli, which includes the full transcript of the pilots' conversation.

We now understand that, indeed, AF447 passed into clouds associated with a large system of thunderstorms, its speed sensors became iced over, and the autopilot disengaged. In the ensuing confusion, the pilots lost control of the airplane because they reacted incorrectly to the loss of instrumentation and then seemed unable to comprehend the nature of the problems they had caused. Neither weather nor malfunction doomed AF447, nor a complex chain of error, but a simple but persistent mistake on the part of one of the pilots.

Human judgments, of course, are never made in a vacuum. Pilots are part of a complex system that can either increase or reduce the probability that they will make a mistake. After this accident, the million-dollar question is whether training, instrumentation, and cockpit procedures can be modified all around the world so that no one will ever make this mistake again—or whether the inclusion of the human element will always entail the possibility of a catastrophic outcome. After all, the men who crashed AF447 were three highly trained pilots flying for one of the most prestigious fleets in the world. If they could fly a perfectly good plane into the ocean, then what airline could plausibly say, "Our pilots would never do that"?

Here is a synopsis of what occurred during the course of the doomed airliner's final few minutes.

More at LINK




 
Incredible, and I am truly shocked. Did they investigate the possibility of hypoxia or some other intoxication, maybe carbon monoxide?
 
camouflage said:
Did they investigate the possibility of hypoxia or some other intoxication, maybe carbon monoxide?

They investigated and found that

AF447 passed into clouds associated with a large system of thunderstorms, its speed sensors became iced over, and the autopilot disengaged. In the ensuing confusion, the pilots lost control of the airplane because they reacted incorrectly to the loss of instrumentation and then seemed unable to comprehend the nature of the problems they had caused

 
The article is a good read and an excellent primer on some of the dangers that come with increased cockpit automation.

Camoflage- that a flight crew mismanaged a multi-million dollar aircraft through 41,000 feet of air and into the ocean is entirely consistent with the evidence available. In periods of high stress, it is difficult for most people to analyze properly- which is why we in the military attempt to reduce emergency reactions to rote memorization that (hopefully) forms part of your muscle memory and preserves what little brain power has not shutdown, for thinking.
 
SeaKingTacco said:
The article is a good read and an excellent primer on some of the dangers that come with increased cockpit automation.

Camoflage- that a flight crew mismanaged a multi-million dollar aircraft through 41,000 feet of air and into the ocean is entirely consistent with the evidence available. In periods of high stress, it is difficult for most people to analyze properly- which is why we in the military attempt to reduce emergency reactions to rote memorization that (hopefully) forms part of your muscle memory and preserves what little brain power has not shutdown, for thinking.

Interesting point. Many years ago an Air Canada 767 ran out of fuel in mid flight but the pilot managed to bring it in for a safe landing at an abandoned airfield. The pilot was a seasoned captain and also a recreational glider pilot, which may explain why he was able to pull this off; no other Air Canada pilots were ever able to recreate the feat in simulators, perhaps because they did not have the characteristics of the plane "burned into" their heads, nor were they glider pilots...
 
The point about increased cockpit automation raised by SKT is very germane to 447's situation. 

That three very highly trained pilots were unable to work through the automation's contorted failure mode, and themselves did not revert to (relatively) 'fail-proof' first principles of aircraft control, speaks volumes to the challenges that more and more aircrew will face as automation becomes more deeply embedded within aviation. Clearly, the situation was so unfamiliar and/or made such little sense to the pilots, that it kept them from reverting to the basics of aircraft control; aircraft attitude, power setting and aircraft configuration. 

Training for future aircrew should include not only the basics and the new automation, but how to identify failure of, and deal with the new automation with the basics when the systems do fail.  Without doing so, the current indoctrination to "trust the machine" can badly lead the pilots astray when a failure mode not foreseen by the engineers, nor programmed into the automation, does in fact occur.  That the cockpit indications to 447's crew so confused them that they physically held the aircraft in a deep-stall for over two and a half minutes reinforces the need to treat automation with a respectful, yet critical eye, and to train accordingly.

Regards
G2G
 
I think it's merely further evidence that airplanes are at their best....


.... when being jumped out of  :nod:
 
I'm not sure what to make of this.

Some pilots pull off bone headed stunts (another French pilot flew an Airbus into the trees at the end of the runway during an airshow), while others pull off the most amazing feats of flying imaginable (dead stick landing in the Hudson river, a DC-10 crew who flew a crippled airliner home with no tail controls at all by adjusting the throttles...)

Blaming the cockpit automation might be applicable in a few cases, but perhaps what is really needed is more rigorous basic training in flying so pilots are aware of what their airplanes really can do.
 
Thucydides said:
I'm not sure what to make of this.

Some pilots pull off bone headed stunts (another French pilot flew an Airbus into the trees at the end of the runway during an airshow), while others pull off the most amazing feats of flying imaginable (dead stick landing in the Hudson river, a DC-10 crew who flew a crippled airliner home with no tail controls at all by adjusting the throttles...)

Blaming the cockpit automation might be applicable in a few cases, but perhaps what is really needed is more rigorous basic training in flying so pilots are aware of what their airplanes really can do.

One of the big problems though, is understanding what automation does...I mean deep in the heart of the programming.  The Airbus demo flying (Air France 296) into the trees was a perfect example of the pilots not being aware of some of the programming that significantly affected engine response when the aircraft was flown in a full landing config (gear down, flaps in 'notch 4' [landing setting]).  When the aircraft settled slightly during the fly past, the pilot increased throttles, but the engine response had been deliberately software limited for the landing configuration, and was not as fast as if the aircraft were in a 'cleaner' configuration (gear up and/or flaps in notch 3 or less).  By the time the engines wound up to provide increased power to clear the trees, precious time had passed and a positive rate of climb couldn't be established before contact with the trees.

That said, there was a whole lot of other stuff going on beyond the software-driven reduced engine spool-up response, and there was a rash of flight computer software "patches" that streamed out of Airbus in the months that followed the Habsheim crash...of course many at Airbus and Air France and DGAC noting it had "nothing to do with the AF296 crash"...  :Tin-Foil-Hat:

The road to Hell is paved with good intentions, and when well-meaning engineers put things into the primary flight computer software to make things easier for the pilots, problems occur when the aircraft flight manual doesn't (or can't) capture all the nuances of what that software is going to do when the pilots find one of those situations that wasn't fully addressed within the aircraft type's initial certification test program.  A quick check of what happened to American Airlines flight 587 and you can't help but wonder about the "don't worry, the software is looking after you" attitude by the manufacturers of particular products.  :Tin-Foil-Hat: :Tin-Foil-Hat:

Regards
G2G
 
G2G thanks for that calm and rational response to Thucydides.

As I knew I would be unable to be so polite I thought it wise to avoid comment.
 
The crash of a B-2 in Guam shows you some of the pitfalls of automation........No amount of pilot training was going to prevent that one.
 
I remain impressed with the fact that we are able to find enough good people, aircrew and ground-crew and controllers, to keep so many aircraft, civil and military, in the air, safely. My hat's, always, off to the aviation community.
 
The other thing I am interested in - there is one mention of CRM in that analysis (crew resource management).  One mention explicitly, and other mentions about their communication in general. 

The other time I heard CRM discussed was with the "Miracle on the Hudson" and the article I read about that incident gave Capt Sullenberger huge credit for being highly skilled at CRM, and that those skills, in addition to his skills as a pilot, made the difference.

This is the article I read on CRM and the Miracle on the Hudson. http://www2.macleans.ca/2009/01/26/maclean%E2%80%99s-interview-robert-helmreich/#more-30313

The human factors (over and above their ability to actually fly the plane) are huge.  I hope the industry doesn't focus entirely on the automation issue.  The way people work together in a crisis is huge.
 
Baden, I wasn't trying to rebut what Thucydides was referring to directly, as much as I was trying to address the challenges of dealing with the reality of more and more complexity entering into systems that humans interact with on a near-daily basis.  The issue with aviation-related changes are that the consequences are more often catastrophic than others.  A 'bricked' cell phone is a lot different than a frozen flight computer.  As an aviator, one of my greatest concerns is that those who fly newer systems heavily enhanced with automation (and for the most part, it certainly is enhancement) may retain fewer of the fundamental, but-saving instincts when that automation fails, or at least doesn't work entirely as advertised.  I can't imagine the combination of confusion/frustration/terror amongst the AF447 crew (and passengers) as they plummeted towards the ocean for several minutes.  Not to quarterback their actions, but to learn from them, one can only imagine how a reversion to refer back to old-style control and performance instruments (artificial horizon, VSI, engine performance indicators) would have helped regain control of the aircraft...wings level, dot above the horizon, engines at 85% N1 for cruise power...unfortunately, the lack of sensical information that the air data computer was pumping out to the crew threw them enough that they weren't able to 're-cage' their assessment of what was going on around them.

EGT, yes, cockpit/crew resource management (CRM) has been further refined to what is now commonly referred to in the military as HPMA, or human performance in military aviation -- human factors both designed into the machine and the manner in which humans crewing the aircraft will assess interact with the aircraft systems and within the surrounding environment.  Capt Sullenberger is an excellent example of a very experienced aviator who, aside from knowing his machine inside and out (glider performance, etc...) made an excellent decision to accept the situation as it was and to commit early to a less than desirable but well-assessed by Sully as inevitable outcome, and set up for the best execution of the water landing as possible.  The calmness with which he advised LaGuardia that he wasn't going to make Teterboro, and was going to carry out a landing in the Hudson epitomizes the pinnacle of aviation professionalism and capability.  His first officer will most likely be incredibly appreciative of having been with Sully on that landing, and aviators around the world review Sully's handling the situation (via CVR, personal appearances, etc...) to reinforce their own ability to deal with the most unlikely or undesirable situation.

Regards
G2G 
 
Journeyman said:
I think it's merely further evidence that airplanes are at their best....


.... when being jumped out of  :nod:

Somebody buy this man a beer!
 
exgunnertdo said:
The other thing I am interested in - there is one mention of CRM in that analysis (crew resource management).  One mention explicitly, and other mentions about their communication in general. 

The other time I heard CRM discussed was with the "Miracle on the Hudson" and the article I read about that incident gave Capt Sullenberger huge credit for being highly skilled at CRM, and that those skills, in addition to his skills as a pilot, made the difference.

This is the article I read on CRM and the Miracle on the Hudson. http://www2.macleans.ca/2009/01/26/maclean%E2%80%99s-interview-robert-helmreich/#more-30313

The human factors (over and above their ability to actually fly the plane) are huge.  I hope the industry doesn't focus entirely on the automation issue.  The way people work together in a crisis is huge.

Listening to the tape, they knew what their options were, the tower could not wrap their heads around it. for the Air France jet, you would think a simple bubble system would give them the clues they needed to orient the aircraft.
 
Colin P said:
you would think a simple bubble system would give them the clues they needed to orient the aircraft.

Have you ever been disoriented flying an aircraft ?
 
Does this suggest that fly-by-wire systems are sub-optimal for civil aircraft?  At least to the extent that they are replacing hard mechanical, or even hydraulic systems, with pure electrical systems. 

Even plants that reside on the ground are better with manual over-ride capability.  Ones cruising at 40,000 feet would probably be better served with similar systems.

I can see aircraft the fly at the ragged edge of the flight envelope, and can't stay up  without the aid of a computer, going the fly by wire route.  After all usually there is only one "passenger" and they can follow Journeyman's advice.

But for a civil craft, which IMHO, should be built to stay up in the air on its own, just like a glider, isn't there something to be said for cables and hydraulics?
 
Kirkhill,

The issue is less of cables and push rods (which are heavy) vs fly-by-wire.

The issue in this case seems to be a sub-optimal understanding of the control laws, coupled with  poor CRM, coupled by a refusal to revert to first principles of flying- put the nose on the horizon (or just slightly down), manage your power, watch your airspeed.  In this case, the airspeed was in and out, so they could have gone to the ground speed readout and realized that they were basically free falling, with little forward speed.

We probably aren't ready for Pilotless airliners anytime too soon...
 
SeaKingTacco said:
The article is a good read and an excellent primer on some of the dangers that come with increased cockpit automation.

Camoflage- that a flight crew mismanaged a multi-million dollar aircraft through 41,000 feet of air and into the ocean is entirely consistent with the evidence available. In periods of high stress, it is difficult for most people to analyze properly- which is why we in the military attempt to reduce emergency reactions to rote memorization that (hopefully) forms part of your muscle memory and preserves what little brain power has not shutdown, for thinking.

Interesting, good point and thanks
 
Back
Top