We are a commune of inquiring, skeptical, politically centrist, capitalist, anglophile, traditionalist New England Yankee humans, humanoids, and animals with many interests beyond and above politics. Each of us has had a high-school education (or GED), but all had ADD so didn't pay attention very well, especially the dogs. Each one of us does "try my best to be just like I am," and none of us enjoys working for others, including for Maggie, from whom we receive neither a nickel nor a dime. Freedom from nags, cranks, government, do-gooders, control-freaks and idiots is all that we ask for.
Our Recent Essays Behind the Front Page
Wednesday, June 5. 2013
Flight 447: The final leg
The most puzzling airline disaster in history has the most befuddling answer in the history of airline disaster solutions.
In brief: In 2009, in one of the most modern airliners ever built, Air France 447 took off from Rio in Brazil and was safely on its way to Paris, last contacting land while partway over the Atlantic.
And then it disappeared.
A few months later, the PBS show 'Nova' put together a team of experts to figure out what most likely happened. They had some serious heavyweights on board, like a guy who took part in the investigation of the tragic, final flight of the SST, and a guy who worked on what was considered the 'most puzzling crash in history' of its day, Flight 800 out of New York, which inexplicably blew up just off the coast, later traced to faulty wiring in a fuel tank.
But at least the Flight 800 team had little clues to work, like "where it went down".
As referred to, say, if it went down.
Because of the curvature of the earth, cross-Atlantic flights always go off radar at some point, and it was only hours later when it didn't show up on African radar that the alarm was even sounded. If a small electronic monitor box hadn't automatically sent a quick 'update' to headquarters via satellite link somewhere between 1 and 15 minutes before the crash, they would have been totally clueless where it ended up, and even then they just had its last known location and were forced to consider where the wreckage could be spread over hundreds of square miles of open ocean. And in 15,000 feet of water.
And they eventually did find it.
Black boxes, befuddlement, and all.
1. The air speed indicators are small tubes that project outside the aircraft. There are three of them for redundancy and they're electrically heated so they won't ice up. Yet that last electronic burst indicated that, indeed, the uncloggable tubes had clogged, which automatically turns off the autopilot.
2. But this is no cause for alarm because all airplanes have a standard configuration that automatically puts them in level flight at the correct speed, in this case the engines at 85% thrust and the nose up 5 degrees. Keep the wings level, hope your compass still works, and you're good to go.
But the whole concept of not paying enough attention to the power and attitude settings made no sense, whatsoever, because the captain was a seasoned old veteran with two able crewmen to back him up.
Adding to the mystery is that while very few pieces were recovered, there were enough that they could determine the plane was in one piece and fairly level when it hit the water, which rules out any kind of mid-air catastrophe. From the evidence, it appeared the crew was in some semblance of control until the very end, which would also explain why no call for help was issued.
So there was zero reason for the crash.
A recap from the end of the Nova documentary follows, which mentions the two main points, above:
1. Ever heard of 'super-cooled water'? It's a bizarre physical phenomenon that occurs when 100% purified water is cooled far past the freezing point, yet it remains a liquid because ice, on a molecular level, actually requires an initial bit of 'dirt' to form around. In a brief shot in the clip below, he's removed the cap from the super-cooled, still-liquid bottle, screwed it right back on and bonked the bottle. The micro-dust that gathered on the inside of the cap in the second it was off starts the freezing process. Running into a freak event like this could explain the clogged air speed indicators.
2. Okay, the uncloggable tubes got clogged, but so what? Here you have three experienced pilots flying one of the most modernized, digitized, electronicized, fly-by-wire-icized airplanes in history, so it's a little hard to come up with the old 'pilot error' dodge. In the following recap, they'll theorize that perhaps they were a little slow in applying the thrusters, thereby possibly initiating a stall, but they acknowledge that's kind of a long shot earlier in the piece, simply because they had loads of time to pull out of a stall, and it's only, like, the first thing they teach you at flight school.
But our aviation super sleuths gave it their best.
And, as I said on the documentary's page, it looked like they nailed it. Air speed and autopilot quit working, the crew runs into an unexpected ultra-violent storm, the plane's bouncing all over hell and gone, the computer's spitting out error message after error message, and somehow in all that chaos the thrusters and attitude were never set correctly and the plane slowly lost altitude and eventually went into the drink.
What other explanation could there be?
As for this quote from the above clip:
Aviation does not do well with unsolved mysteries.
I'd say this was well borne out when Air France [read: the French taxpayers] kept paying some high-tech, deep-water search rig to keep looking for the wreckage and, amazingly, two arduous years later, they found both it and the two black boxes.
And it certainly wouldn't be any surprise if the black boxes verified Nova's findings. Ultra-bad weather, three clogged air speed indicators, error messages left and right, chaos in the dark, bouncing cockpit, with the resultant inattention to the power and attitude. Result? A steady downward path to oblivion.
What other explanation could there be?
Well, there was one tiny little thing.
One possible contributing factor that the entire investigative team never even considered.
One lone factor that, had one of the team members dared to suggest it, he would have been laughed from the room, dropped from the team, and, ironically, driven from the agency in disgrace for 'total mental idiocy'.
Ironic, because that's exactly the factor he raised.
Also known as brain lock, brain freeze and brain fry in psychiatric circles. The fact that he just held it and held it and held it and never said a word until the very end simply staggers the mind.
You would certainly say it's befuddling.
And the fact that Airbus had both eliminated the yoke and disconnected the controls from each other made the biggest point of all. As was vividly shown, the range of the control handle is fairly short, and would certainly be hard to spot across a dark cabin in a bouncing airplane, but at least you'd have a chance of spotting its unnatural position glancing at your own (connected) handle.
Never mind that the brain freeze would have been impossible to miss if Airbus hadn't removed the conventional yokes for a more (delicate cough) 'modern' look.
And there you have it. The thousands of articles, the millions of words, the endless hours of speculation and conjecture, the hordes of money spent on research and investigation and documentaries and TV specials, the air time, the radio time, the newsprint and news site space, the pundits and writers and columnists and bloggers, all of them guessing away at the mystery...
And not one of them got it right.
That's why I thought the word 'befuddling' in the second paragraph of this article deserved a most in front of it.
The full transcript from the flight recorder with running commentary is here.
So, to sum up the last and final leg of ill-fated Flight 447, the reason it crashed is because the captain was out of the cockpit, the computer spit out some error messages, and one of the two junior crewmen panicked and suffered what's known as "brain lock" in medical circles, where his brain told his right hand that the last place it wanted the plane to go was down, so it reflexively pulled back on the handle to go up — whereupon the brain promptly forgot all about it and continued acting normally until the very end, when, surprised, it noticed the locked position of the right hand for the first time in minutes and called attention to it. Because the range of the handle is so short, being in the 'up' position wasn't noticed by the other co-pilot in the dark, bouncing cabin, and even the captain's return did no good because he, too, couldn't see the position of the still-clutched handle.
Brain fry killed Flight 447.
And no one guessed that.
Posted by Dr. Mercury in Hot News & Misc. Short Subjects at 10:00 | Comments (29) | Trackbacks (0)
Trackback specific URI for this entry
Display comments as (Linear | Threaded)
Superb post, Doc, and all I can say is "Wow" to that list clip. I had read the report after the black boxes were analyzed, and I recall it mentioning the wayward pilot's actions, but I don't recall it specifically noting that he just held the damn thing back in one continuous "brain lock" moment. In an effort to spread the blame around (and divert it from Air France's cockpit redesign and the role it played), they probably referred to the co-pilot's "unconventional" actions or some such, but nowhere did it say he was basically 100% to blame. Or use the term "brain fry," for that matter.
The NTSB probably has a big book of "causes," with "Human Error" heading up a big chapter, with the various types of human error broken down into types.
I'd love to know what this one's filed under.
Bob's right, that's a great question. I'd suppose they'll have to fire up a whole new column. But, rather than using some un-PC term like "brain lock", they'll probably use some big piece of psychiatric jargon to make it more palatable.
And I'm pretty sure "design flaw" won't be mentioned.
Damn good question, Dave.
Doc, cool post, perfectly handled. When I read the lead-in to the last vid, I knew you didn't literally, actually mean "total mental idiocy." Wrong again. The mind just reels at the thought of someone actually doing such a thing, but the black boxes don't lie.
Fun post, much thanks.
Just caught this before I headed out the door. I'll do a "real read" when I have more time.
The air speed indicators are small tubes that project outside the aircraft.
Not to be pedantic but the Pitot/Static System operates a number of flight deck systems:
That will help 'splain it a little. I aalso read where the pitot tubes had a "Safety Alert" issued which required them to be replaced but I don't recall what was wrong with them. It was an "Alert" as opposed to a manditory "Do It Now" bulletin. That a/c was one, in the AF fleet, that had yet to be upgraded.
As well, the Airbus cockpit/flight control systems are a "little problematic for some pilots re Human Factors issues. Capt Scully (the Hudson River Pilot) mentioned that factor in an interview I watched on a news channel some time ago. This has been a concern for many years. The pilots are trained to fly the computer not the aircraft.
"Not to be pedantic but the Pitot/Static System operates a number of flight deck systems:"
Not to be pedantic, but I didn't say it didn't. The tubes outside the aircraft measure air speed. The 'static port' the article mentions measures air pressure and thus altitude. The computer acts as the third piece, hence 'system'.
As for the pitot tubes, yes, Air France was in the process of replacing them, but this plane hadn't received the upgrade yet. This might not have prevented the problem, though, as the previous problems had been simple failure, not freezing over, so the super-cooled water might still have knocked out the upgraded version.
Cap'n Sully is in the last vid and touches on the point you mentioned. While we grant quickly that "brain fry" isn't usually the culprit, it still makes a point that in such a situation, and given the redesign, we've run the whole gamut from "impossible to miss" with a yoke system to "almost guaranteed to crash" in a dark, bouncing cockpit with the aforementioned brain-fried hand feverishly clutching the control. Given that, you could almost say it was inevitable. All it took was the usual 'multitude of factors' to line up correctly.
Getting "touchy" in your 'old' age.
last contacting Rio tower while partway over the Atlantic.
While I don't have any Jeppesen Terminal and HE Charts handy I'd imagine there'd be an Oceanic Sector in an ARTCC that would have been its last ATC contact.
Airbus had difficulty selling the aircraft when it first came out due to the "fly-by-wire" configuration of the cockpit. Pilots didn't like it. The company commissioned a study, costing many thousands of dollars, to determine what sort of individual would acclimatize to the cockpit with the least amount of difficulty. The result?
The individual would:
1/ be able to complete Super Mario 3 in less than an hour,
2/ be able to type 60 wpm and
3/ not possess a drivers license.
"Getting "touchy" in your 'old' age."
I think you need to work on your definition of the word "touchy". All I did was expound on the issue you raised.
"While I don't have any Jeppesen Terminal and HE Charts handy I'd imagine there'd be an Oceanic Sector in an ARTCC that would have been its last ATC contact."
While writing the piece, I thought of actually re-watching the video to decipher this key piece of evidence, but then I figured, "Oh, it's just a blog site. We're just here to have fun. What ultra-pedantic (fill in blank) would even mention such a trivial detail?"
It's now changed to "land". I was going to just change it to "Brazil", but with ultra-pedantic fill-in-blanks around, you can never be too cautious.
"3/ not possess a drivers license."
I still can't figure that out. Any thoughts? What person in the civilized world in the 21st century doesn't have a driver's license? They might have an official ID card, instead, but there's no real dif there except for the car aspect. And if it DID make a difference and they accepted official IDs, why would you want a person who doesn't know that if you turn the steering wheel to the left, the car goes to the left? None of #3 makes any sense.
While I never read the thought behind it (only the 3 stated findings my speculation is this: if you have a drivers license you've already started to program your system to the analogue senses of "feeling the vehicle". Ever drive into a fog bank and your quickly start to have a bad feeling? Pilots learn to fly in visual conditions and develop a "feel" for the aircraft. The transition to IFR flight isn't easy for everyone but you are flying the aircraft via the instrument readings. In an Airbus you don't fly the aircraft you fly the computer which flies the aircraft. If you have a less developed "feel" I would imagine its felt you'll learn how to fly the aircraft (oops ) the computer in less time and with less difficulty.
I'm sure there's a "Farmer" out there with much more knowledge/experience in planes than myself. I just told them where to go for 29+ years!
#22.214.171.124.1 Garry on 2013-06-05 14:06 (Reply)
You mean, you sent them someplace --and they had to stay there for 29+ yeaRS ???
#126.96.36.199.1.1 buddy larsen on 2013-06-06 04:23 (Reply)
"touchy" = sensitive???
#188.8.131.52.2 Garry on 2013-06-05 21:44 (Reply)
I've read about this flight before, in "Extreme Fear," I think. It's amazing, but brain-fry is exactly it. We underestimate how prone people are to it. This is why people who expect to be thrown into emergencies drill, drill, drill--because what's obvious under normal circumstances is nothing like what's obvious when the lizard brain takes over. Some people have a knack for clear, high-level, wide-angle, creative cognition in the face of panic, but most of us get tunnel-vision. Training can help with the problem to some degree.
I've re-watched a handful of 'Mayday' episodes over the last few days, and two of them touch upon the point you made in that last sentence.
In one, I think The Gimli Glider, one of the pilots said afterward that as much as you train for it, you never really know how you'll react until the going gets tough. Like you said, you either have it or you don't.
In regards to the 'tunnel vision', the main pilot on Japanese Flight 123, which blew out the huge big seal at the back of the fuselage, suffered from some kind of brain fry and refused to take certain orders from the tower, and the junior pilots couldn't say a word because the Japanese culture didn't allow it. While it probably, ultimately didn't make any difference, it still took away that 1-in-a-million chance they might have had. Even ditching in the drink would have been better than flying straight into a mountain.
That's one reason I revere Capt. Sullenberger. He treats his landing on the Hudson as no big deal, and perhaps it would be no big deal if the ability to function even at a mundane level of competence in an emergency weren't as rare as it is. It's not as though he had to execute a triple-gainer backflip, he just had to keep track of some ordinary considerations and exercise some ordinary skill--but he did it in the clutch, when many, many people would have brain-fried.
Although very few airliners have actually scuttled in recent years, it seems like the few that have dipped one wing into the water first, which, understandably, disintegrated the aircraft. Sully kept his wings dead level and that's mainly what made the difference.
I thought he had a great delivery in the video clip.
(question is asked)
(question is asked)
No sense in overstating the obvious.
He was certainly "Capt Kool" on that day and it was quite a feat IMO. For Capt "Sully" it was almost just another day at the office. I haven't heard whether he's "back on the line", yet, but I don't believe he is.
It's not brain fry, it's autonomic muscle tensioning. A natural tendency in stressful or even loud situations. The alarms most likely made the problem worse by startling the pilots system.
The same tensioning is why untrained shooters pull their gun, why drivers in a car crash brace their arms into the steering wheel sustaining serious shoulder injuries. It even has a common name, "the grip of fear". I can be trained to minimize but it happens outside the conscious brain when alarms and crisis is about.
Now, what you don't do is design control systems that go full range on such natural muscle tensioning in the hand and forearm. The arm doesn't move much but the fingers gripping do move several inches. Which the engineers decided was the range of motion needed for their fancy controls. I suppose they never say a movie or read a book where someone had to pry another's hand off a gun or control to save the day.
You can get around this where there is a group as on this plane by putting up a large, very visible even in the dark, yoke position indicator. They do that on ships, the rudder angle indicator must be high in front, lit at night and visible from several feet away. The captain upon entering should have had that right in his face.
Question is: We know all this, why is humanity getting stupider? Is it a love of technology and engineers being unable to visualize what happens when it all goes to sh...? Or are to many coming to trust in the technology and it will save us.
The "autonomic muscle tensioning" might have caused the hand to grip, but it was the way he kept responding 'normally' to the other co-pilot that was so sobering. As I said, it was like he suddenly noticed his hand at the very end and figured he might as well mention it. To be so oblivious to the "autonomic muscle tensioning" -- that your own actions are going to cause your death -- is where the 'brain fry' part comes in.
As for your question, what makes it a toughie is that there's no one big reason why airliners crash. It's different almost every time. Not enough gas put in because of a metric/imperial conversion glitch. Bad riveting job. Duct tape left over the pitot tubes after a cleaning. Lack of understanding on a tower's part that a plane was ultra-low on fuel. Newly-installed entertainment system shorts out the system and the plane catches fire. Hose clip 2mm too short which allows the hose to rub against a fuel line and eventually rupture it. Volcanic dust. Terrorist bomb.
And that's just the first season of 'Mayday'.
And even flat-out "pilot error" is different almost every time. Forgot to lower the flaps. Didn't obey the tower. Didn't obey the collision-avoidance unit. Flew at the wrong altitude. Made a mistake with the radio settings and got the wrong instructions. Dialed the wrong number into a cockpit instrument. On and on.
A tough question, indeed.
While I watch very few shows on aviation (they just aren't factual enough for me (just like everyone thinks the "Tower" controls ALL the aircraft ;-}} ) I must say b"Mayday" is very good.
Did you ever see the one on Air Transat 236 that ran out of fuel and "deadsticked it" into the Azores?
The "Gimli Glider" is also a favourite.
I watched the Azores one just the other night. I think I have four or five airline disaster vids on my site, and three of them are glider scenarios. Although I think my all-time fave is the British flight that went through the volcanic ash, the Gimli one, where he friggin' sideslipped the big girl, was maybe the coolest moment in aviation history. Like a few similar episodes in the 'Mayday' series, when they later put multitudes of teams through the same event in the simulator, every one of them put the bird in the ground.
The Transat/Azores had a conclusion (that I heard about) but am still looking for the source. Yes, a "wrong" part was installed off a RR engine manufactured for the L1011. The mounting bolts lined up but the hydraulic ports were different thus the chaffing against the fuel line. It was meant for a "quick fix" to get the plane into the air and save the lost revenue of chartering another airline to haul their load. The problem was when that a/c came home the proper pump was not installed. It made at least 3 or 4 more flights before the fuel line wore through on AT236.
#184.108.40.206.1 Garry on 2013-06-05 14:53 (Reply)
Well, maybe we need some ballroom dance teachers to consult. When I first started and apparently according to my instructor it wasn't unique to me, as once concentrates on the steps, music, etc., some have a tendency to slowly grip their hand. An experienced teacher knows how to place the hands together to avoid the crush...at least for a while.
The pilot's brain was not fried or froze up, it was occupied with other processes. It didn't occur to his brain to check the tension nor to check the flight level indicator, at least to the point of actual awareness.
They can train to stay out of "tunnel vision" but really one has to learn it by experiencing and learning from small crises. If you do, you can learn to do a brain interrupt that causes to you to realize you are unaware of other just as critical inputs.
I'd bet a real examination would show a lot of "pilot error" has a central fail point, namely loss of awareness due to focus on one oftentimes far less important operation. You don't hear, you auto-acknowledge without really comprehending, forget to double check your entries, or your flight controls.
Your brain is a single processor. It does multiple things by interrupts to use computer lexicon. If one process hogs the processor, you miss a lot or the autonomic systems act independently and without check.
I don't know what happened, but I'm of the opinion that this pilot brain freeze was much to convenient an explanation because it has a little of everything - faulty control design, inadequate pitot tubes, pilot dribbling on himself with a more experienced pilot sitting right next to him - I really am suspicious that it was more a software/control problem than a pilot problem.
That's just me - I'm suspicious of easy explanations when all other explanations fail. I know - Occam's Razor and all that, but there's a lot of money at stake with manufacturing these aircraft and Airbus being essentially a governmental sponsored corporation with it's reputation to protect? Yeah, I gotta think there as more to it than that.
Well, that's what the black boxes indicated; that it was in the 'up' position almost the entire time. But what verifies the computer was working correctly is that it un-ups right after he says it out loud and the other co-pilot takes control.
So, if it wasn't the guy, that pretty much just leaves "faulty wiring" on the suspect list.
As for the other co-pilot being "more experienced", and the pundits asking why he didn't take over the controls from the "less experienced" co-pilot, I don't recall the vid or article saying what kind of "difference" they're talking about. It might be two weeks, in which case the question would be irrelevant. It might even be part of their policy, like "If the dif is less than 6 months, you don't take over control."
Yeah and that could be the explanation - its a kind of "target fixation" that fighter/bomber pilots suffer from occasionally so it is possible. And it does make sense.
I'm just stuck on the pitot tube thing. Those things are critical components of the flight systems and they get really hot drawing a lot of current. I'm having some problems trying to get the whole super cooled water thing with that. From what I know about super cooled water, and I'm not a freakin' expert by any stretch of the imagination, that type of water doesn't occur in huge blobs like some airborne lake floating in the sky - it is dispersed so any super cooling would have to be continual. Anything is possible I guess, this just seems somewhat "unlikely".
But there's nothing simple about brain-freeze. It's extremely common and well-documented, but not at all easy to explain or combat, being a function of some unbelievably complex interactions in our cognitive systems. The tricks used to fight it range from the drill-drill-drill system (which we hope will allow us to carry out some basic safety measures while we fight to free ourselves from the panic and regain higher function) to almost zen-like approaches that help a star athlete overcome a clutch and get back in the "zone."
Person after person, interviewed in the wake of an emergency, reports entering almost a trance state, in which a great deal of information simply will not penetrate. It's not just tunnel vision but tunnel hearing and tunnel thinking. Panic stuns.
A quickie that can work: picture yourself. Throw out an 'omniscient eye'. Can solve performance-anxiety in sports. Can work if you think it can --pretty mechanical, actually; it leaves no time for the "Uh oh, here comes the helpless place".
I'd bet a real examination would show a lot of "pilot error" has a central fail point JKB
faulty control design TF
I agree. Don't forget the French Gov't did the investigation and, while, I don't know "to the tune of how much taxpayer money" helped ensure that Airbus Industries would be able to complete with the "Big 3" in the US, I'm sure "any design faults" would not be allowed to appear in ANY final report.
And, don't forget, there is now just a "Big 1" in the US (Boeing).
As a one-time round-engine, trans-Atlantic transport pilot for many years, I am bowled over by the overall coherence and reasonableness of this article and the comments thus far.
Not 4 me but, 4 the rest,
29+ yrs in CYYZ Twr.
A great experience !!