Sad,…
…fascinating, but extremely chilling reading: what happened aboard Air France 447.
And here’s an article written two years earlier by the same author, long before the black boxes for Air France 447 had been found. In it, he speculates on what might have happened to the flight. It’s interesting to compare his guesses at that time about the flight’s scenario to the facts as later revealed by the cockpit recordings:
Without the box’s data, the only physical evidence of the airplane available to investigators was the mangled wreckage. From the way it had been deformed””in particular, the way the floor of the crew’s rest compartment had buckled upward””French investigators determined that the fuselage hit the water more or less intact, belly first, at a high rate of vertical speed. Added to the ACARS messages and the satellite weather data, the evidence began to conform to a possible scenario.
By 10:45 pm, 10 minutes after the last radio transmission, the plane hit the first, small storm cell in the Intertropical Convergence Zone. Fifteen minutes later, it hit a larger, fast””growing system. And then, just before its last ACARS transmissions, the plane hit a whopper, a multicell storm whose roiling thermal energy rose more than 3 miles higher than AF 447’s altitude. Buffeted by turbulence, near the heart of a strong thunderstorm, the pitot tubes froze over. Lacking reliable speed indicators, the airplane’s computerized Flight Management System automatically disengaged the autopilot, forcing the co-pilots to fly the airplane manually.
Without autopilot, the pilots had no envelope protection restrictions, which are designed to keep the pilot from making control inputs that could overstress the aircraft. This is particularly dangerous for airliners at high altitudes. The thin air demands that airplanes fly faster to achieve lift, but they still must remain below speed limits. Flying too fast can create a phenomenon known as mach tuck, when supersonic shock waves along the wings shift the aircraft’s center of pressure aft and can make it pitch into an uncontrollable nose-dive. Flying too slow can cause a plane to stall.
Seems to me his got it pretty close to the truth.
But what was going on in the minds of the hapless pilots when they made their fateful decisions will never really be known.
[NOTE: The comments to the first article are excellent as well, many of them by pilots.]
Much discussion of this at a Neptunus Lex thread, including many military and other pilots.
This isa terrifying account.
The same misplaced reliance on technology occurs on a more prosaic level in scientific instrumentation. At one time a researcher had to understand most instruments fairly well to generate any data whatever. Nowadays (get off my lawn!), the computer takes care of most things; the researcher merely pushes the “activate” (or whatever) button and takes as gospel whatever comes out.
All too often, what comes out is nonsense. It’s tempting – and all too easy – for novice researchers to suppress indications of problems, much like dealing with an infant’s persistent crying by dosing him up with Benadryl.
That is where things inevitably came now: for those who use technology or science they are just a magic. And it getting worse every day, when such basic skills as ability to do arithmetics without calculator are now things of the past. Our BS-meters are going rusty from apparent lack of necessity to use them.
About 3-4 years ago, there was a fatal accident on the Washington MetroRail system…didn’t kill as many people as the 2009 collision, but did kill the train operator. The train was operating under computer control, as is standard practice, but the program did not always compensate properly for the longer braking distance required in icing condition.
The train operator requested permission to switch into manual mode, which was denied—dispatchers had been told to minimize this model of operation because sometimes braking was done too aggressively, flattening the wheels, which was expensive. In this case, the operator was told “let the train do what it’s supposed to do.” What it actually did was to slide into a train stopped at the station, killing the operator of the moving train.
Basically, the assumption made was that both the system designer AND the dispatcher had knowledge/judgment superior to that of the man on the spot.
Imagine the terror that plane crew must have felt.
As a retired airline pilot this is something that is I have expected. I expected it because these advanced flight control and guidance systems are aimed at reducing human error to zero. But in trusting in technology and trying to take the responsibility away from the pilots they created a circumstance where a different type of human error was the cause of the accident.
Being the dinosaur that I am, I was never truly comfortable with the advanced systems. On the DC-10 we were charged by company policy to use the auto-pilot and flight director system almost exclusively. All well and good until you had to hand fly the airplane because you encountered conditions that the flight director/auto-pilot couldn’t handle. Like a 50 knot crosswind at 2500 feet above the the landing area that tapers to a 30 knot crosswind at landing. Then you needed to have those manual flying skills. My policy was to fly every other leg manually.
I read an excellent blog by Captain Dave at Flight Level 390: http://flightlevel390.blogspot.com/
Capt. Dave flies the Airbus airplanes and loves all the bells and whistles. They are far advanced over what we had in the DC-10. The computers are at work monitoring just about everything and flying the airplane in very precise fashion. Captain Dave has given his readers an insight to the way these highly advanced systems work. They are very useful and quite accurate, but demanding of deep understanding and careful monitoring by the pilots. He occasionally hand flies the birds to keep his hand flying skills. (The mark of a true aviator, IMO.) If you read his blog you will get a feel for the kinds of things that are running through the mind of an airline Captain as he battles weather, fuel usage, aircraft weight, slick/short runways, schedule and the Air Traffic Control system. I wonder if a pilot with his insight and experience might not have successfully identified the stall and flown out of it.
The one thing I am sure of is that all airlines with the Airbus aircraft willl train their pilots on this kind of problem. Whether they will allow or encourage more hand flying to maintain those manual skills is the question.
Occam’s Beard,
Now that IS terrifying. The horror story that is worse for being true. When men become mere machines, or is it when machines rule men?
I can more than imagine what goes through, at least the crews’, minds as such things happen. I survived my scrape by the hair on my teeth. I often will pray when I hear such things. In death, no matter who is near, you are naked and alone. It is a raw thing. At least with these it is very quick, when it finally happens.
Let me state what I have sensed from my own previous experience that didn’t turn out this badly, and that of a couple of others.
It is very easy to become disoriented when you are flying over the ocean in complete darkness. Remember John Kennedy Jr. If the pilot on the controls does not stay focused on the instruments in front of him, only allowing for periodic cross checks of the outside world, then it is very likely that disorientation will occur if something other than normal conditions occur.
Spatial disorientation is very difficult to think your way out of. The brain is getting a continuous stream of conflicting information from the senses. It normally involves loss of visual cues because the eyes are such a significant input to how the brain perceives the current reality. The loss of visual cues, coupled with beeping horns and warnings occurring for seemingly no reason, together with unusual inner ear sensations from flight attitudes that don’t match perception, together with a flight control system which yields synthetic feedback, all leave the brain in a mode more suited for the Twilight Zone than the cockpit. Reality becomes surreal as the brain just goes into a dither mode (Tilt!). Clear rational thought becomes impossible. In the midst of this level of confusion, the mind grapples for a singular focus to reset the inputs around it.
In the case of the Air France pilot, it was to pull back on the stick and await the normal indications of a climb to regain orientation. Unfortunately, as is often the case, the indications just didn’t fit the expectations, and the brain went back into the dither mode, probably then emerging next with another thing to refocus on. His brain abandoned the climb scenario when it didn’t work for him, but his hand never released the back stick pressure needed for the climb he was attempting. His brain was too distracted. That was fatal.
Another aspect of this level of confusion is the desire to make things right. No one wants to stay in a confused mode. This can lead to not giving up the controls to another person until it is too late, or taking back the controls well before things have settled out. There is a level of pride which takes over with some. I’ve seen it both ways. If the pilot had of said “Hey, I’m disoriented, take over!” then none of this would have likely happened. Or if he had of just let go of the stick, the airplane would likely not have stalled.
“”When men become mere machines, or is it when machines rule men?””
doom
And the bureaucratic State seems to be the ultimate example of this. It has all the qualities of a machine we’ve made running out of control. If you think about it, we’ve even created a form of autopilot that steadily increases spending so no persons or groups will even be held responsible for the crash.
J.J., as a former pilot, do you think you could help shed some light on something for a layman like me? The pilot “Bonin” apparently was pulling back on the stick for nearly the entire time. But, I had thought that basic training was to push the stick *forward* in a stall. Granted, simply logic tells me there are times when such a thing would be bad – when you’re close to the ground, for example – but they were at altitude. So the thing that’s unclear to me is: Why did Bonin insist on pulling back on the stick the entire time? What combination of inputs and analyses led him to think that that was the best thing to do?
Yes, I do believe that any answer to this would be speculation. But given that you’re a former pilot, I’m hoping that the answer will be *informed* speculation. Which will be better than the average opinion. Anyway, would you have any ideas what he could have been thinking?
Thanks!
Oh, and of course, that PM article had it’s own hypothesis:
… and to a layman like me, that sounds reasonable. But just out of curiosity’s sake, I’m interested to see if there’s a wide variety of opinions or not. Which is why I’m querying J.J. (and any others here who might have some experience to shed light on the topic).
It is very easy to become disoriented when you are flying over the ocean in complete darkness.
Indeed. Right after JFK Jr.’s crash I was sitting next to an airline pilot on a overnight flight to Europe who described it as “flying in an inkwell.” Looking out the window, absolutely nothing was visible. Nothing. It was as though black paper had been taped over the window, or turning off the headlights while driving on a winding mountain road.
It must take extreme discipline and extensive training not to freak at least a bit under such circumstances.
OB, i recall being on a long road trip with an ex female aquaintence when we began discovering there were no gas stations around and we were almost empty. She became quite stressed and disoriented and started speaking gibberish that made no sense. Basically she was in panic mode. I hope she never finds herself alone in similar situations. She couldn’t handle it.
I’ve always felt that it is a bad idea to allow young children to us calculators until they have a good understanding of arithmetic and numbers. You have to be able to spot a bad answer that could result from operator error. Pilots have to spot “bad answers.”
I once had a graduate student add up end of term grades for me with a calculator. There were four test grades to be added and all were below 100. When I got the results, I spotted some totals in excess of 400. I never did that again.
J.J. : did you see the article in Flying mag about the accident? It seems to have parallels to the Buffalo crash also where apparently the pilots failed to recognize and fly out of a stall.
The latest issue has a great article by Martha Lunkin on testing a student pilot (written in her wonderful style) in which the student was extolling the virtues of glass cockpits, GPS, and autopilots. She, of course, during the flight check ride turned those off. As she says, the results were not pretty 🙂
I used to fly, nothing bigger than a C172. Now I do it all on the computer. Still fun, (and cheaper!) and I also tend to fly only those planes in the sim, with the old steam gauges and navigate VOR only. I’m also a dinosaur.
The Air France pilots first surrendered to panic and then surrendered to gravity.
JJ has it.
By the way; as is usually the case, particularly in Europe, there was an acrimonious debate between the pilot’s union, Airbus, and the government over this mishap. It was known that the pitot heat system on the Airbus airplane was unreliable. The subsequent icing over, with the loss of airspeed indications created the cascading loss of computer functions and auto-pilot. The manufacturer and the government obviously did not want to admit this design flaw, and tried to deny the fact. Their response was typical–all pilot error.
At the end of my career I was working for a European aircraft manufacturer. My observations again paralleled the comments of JJ. Their training philosophy was to make maximum use of the computer/autopilot system and de-emphasize manual flying. Why? Because most entry level airline pilots did not have the basic flying skills to fly an airliner manually. This was contrary to the general attitude of American trainers.
In my humble, but somewhat informed opinion, the issue is also compounded by other aspects of pilot training and testing scenarios. In the U.S., and in Europe, all stall training was traditionally oriented toward recognition and corrective action at the first indication by the built in warning system. If the problem proceeded beyond the first stage to the next level built in stall protection, which is a mechanical force that pushes the nose of the airplane down abruptly, it was a failure. The exercise is invariably entered in a very controlled process.
Unfortunately, that is not the way it happens in real life. Pilots were not trained to react to the confusing oral and visual signals and other sensory cues that the approach to a stall with the auto-pilot engaged presented. Long ago, I incorporated a more realistic scenario into training on an informal basis–just for demonstration. The pilot reaction was often confusion, delay and sometimes counter productive reaction. That was years ago, and maybe things have changed since I hung it up.
Sammy Small you present a fanciful scenario. I am sure the crew was confused, but it was not because they were flying over the ocean at night. Their airspeed system was totally screwed up and unreliable. They had not been trained to cope with that, and they flew into the worst possible situation, a stall–or at least a stick pusher– at high altitude. Again, they had no training, nor previous experience to cope with that situation. To compound matters, they had been FORCED throughout training to rely on the systems that had been denied them at this critical moment.
I had a Navy pilot friend experience much the same airspeed sensor situation in a thunderstorm over the Tonkin Gulf. He was trained well enough, and was resourceful enough to use other cues. That ALMOST got him through it without incident. His ability to recover from the next catastrophic situation saved him ultimately. Sadly, he had his crew bail out into a typhoon before he could recover, and they were never seen again.
I saw some comments at the PM article that questioned the side-stick arrangement. With the old-style yoke controls, which were linked together, it would have been obvious to everyone in the cockpit whether they were being pulled back or pushed forward. The sidesticks were apparently independent of one another. Is it possible that nobody else realized that Bonin was hauling back on his stick the whole time? Was there any indicator on the instrument panel that would have shown that?
I believe that the answer to Ricki’s question is–NO.
I have a friend who is still flying the Airbus. I can get an answer from him, if no one comes up with a definitive one in the meantime.
My friend is a techno-freak so he loves all of the bells, whistles and g-whiz in the Airbus; on the other hand he is a throw back pilot, so he does a lot of manual flying despite them.
I don’t like flying on Airbus products. In 1989 I was flying out of New York on an Airbus. The thing headed down the runway and, just before takeoff, aborted. Tried again, same problem. Computer failure, they told us.
Then there is the infamous Airbus Air Show crash.
http://www.youtube.com/watch?v=wBFG3_y6zIg
Just made a trip to England flying Virgin Atlantic Upper Class on an Airbus. Went upper class because of a DVT/Pulmonary Embolism problem I had on an economy flight from Hawaii. Actually enjoyed the experience on Virgin.
My daytime job is software architect. Today, I encountered a problem that made no sense. One program inserted a record into a database. A second program could see the record. The original program couldn’t. Nor could any other program. Any similar records behaved as expected. Eventually, we looked for the record by its assigned number. Once we found it, everything worked as God expects (NOT Allah, Allah is arbitrary).
There is a bug somewhere in the depths of the database software that pops up once in a blue moon. This database system has been around for thirty years or more. It has been exercised by billions of users. Yet it failed when I least suspected it would. It drove me crazy for most of the day. It will tomorrow.
Flight software is extensively tested but there are situations where it fails. If the pilots can’t recognize the failure, disaster ensues.
The fundamental problem is that most software only has one route to a solution. If there is a black swan on that route, it fails. It doesn’t say “Oh Shit, now what do I do?”, and fall back onto more primitive responses.
But lets add a cautionary note before you all decide computers are bad: the number of people dead because of computer error is a hell of a lot less than the number of people dead because of human error.
EMH,
My informed guess is that he froze on the vertical speed indicator. He had full power on and was pulling back on the stick, yet the airplane was descending at a rapid rate. The fact that the stall warning was going off was apparently blocked out of his consciousness while he tried to arrest the descent. For some reason (a good psychologist might be able to explain) he blocked the stall warning and saw only the rate of descent.
My knowledge of the Airbus aircraft is only what I have gleaned from reading FL 390 so the term “alternate law” for controlling the aircraft has a fuzzy meaning. I know it has to do with losing input from two or more sensors. I also know this, it takes a conscious and determined procedure to take control of the Airbus aircraft and fly them manually. It is designed to discourage manual flying.
When the pitot system iced up, the auto-pilot and flight director systems kicked off. However, as they descended into warmer air, the ice apparently came off the pitot system and their instruments were recovered. By that time they were confused and their only thought seemed to be to arrest the descent. When you see that you are descending rapidly and aren’t used to flying the airplane manually……….adding max power and pulling aft on the yoke or the stick is instinctive. To do the opposite is hard to do. To reduce power and push forward will increase your rate of descent, but will also save you by increasing the airspeed. However, you need altitude to do this. They tried it too late.
Safety lies in having pilots thoroughly trained in understanding all the facets of what can go wrong with all the fancy technology. And how to be competent when flying the airplane manually. My guess is that the engineers are hard at work on another tech fix to overcome this error. However, as noted, I put more faith in having the pilots actually fly the airplane on a regular basis.
Believe me this accident has been analyzed and all operators of Airbus aircraft are spending extra time teaching stall recognition and manual recovery techniques. This has been a wake-up call for this particular scenario. I hope it encourages more manual flying.
My question is why was the Captain not in the cockpit when the airplane was entering an area of severe thunderstorms? Had he been in his seat, things may have turned out differently. As Ernest K Gann said, “Fate is the Hunter.”
Response from my Airbus flying friend is that the two controllers are NOT linked. The only alert is an audio warning if both are applying inputs simultaneously.
So, the non-flying pilot probably did not know what the flying pilot was doing; neither did the flying pilot, apparently.
J.J., thank you very much. I appreciate the answer. While I’m neither pilot nor aviation mechanic nor psychologist, simple everyday life has shown me that I and others can occasionally “lock on” to one element of an unusual and urgent situation to the conclusion of all others. And one of the thoughts I had on this was a sneaking suspicion that the pilot did indeed just lock onto one element of the emergency without properly considering others. Without prompting, you speculated in the same direction I did, so I think we’re onto something. 🙂
Either that, or we’re both out in left field together. 😉
—–
At the same time, I have had a parallel thought that’s actually contradictory to the one above, but my absence of knowledge keeps me from determining which one is correct. I wonder if Bonin was actually fully aware of the stall *BUT* was conditioned to trust the jetliner to recover itself. Again, I’m no pilot, so maybe I’m on Mars on this one, but: The military teaches its members that people do not rise to the occasion, they fall back on training. So I’m wondering if his training really conditioned him to believe that no matter what, the jetliners controls would allow him to recover from such a situation if he simply kept hauling back on the stick, simply because the controls programming would supposedly never, ever allow the jetliner to respond in the wrong way to his intent.
Does that make sense? This is based on my layman’s interpretation of what’s been posted online, so everyone should take it with a grain of salt. But the knock seems to be that the Airbus will “think for you” in many cases and not allow you to make the jet exit safe flight parameters. That, I think, is where that whole “Normal Law” and “Alternate Law” set of parameters come into play, and that the pilot was conditioned to trust the jet, but didn’t realize that the jet had put itself into a different set of control parameters that would indeed allow him to screw himself. And he didn’t realize it because of 1. Experience, 2. Conditioning via training, and 3. The urgency of the moment which was time-forcing him to discard deep analysis and rely on training.
Again: That’s all a layman’s speculation. But I’m building it on what I’m reading online from pilots and other aviation “geeks”.
I’ve been running those thoughts through my head, and I don’t know which one’s a better fit. Anyway, if anyone has some thoughts on that, pitch in. Or if I need correcting on anything, again, feel free. I don’t mind being corrected on this; it’s all part of the learning process.
Bah… typo:
To the exclusion of all others. Exclusion. Man, I need a proofreader…
ELM,
You’re thinking very much along the lines of an accident investigator. Deep probing of training and operational procedures along with examining how they impact dealing with emergencies is SOP on every accident. That’s one reason flying is as safe as it is. Evry accident is examined for clues as to what happened, why it happened, and how do we prevent it from happening again. Every new pilot coming along has to be trained and made aware of those causes and how to avoid or remedy them.
There was a DC-10 crash at O’Hare back in the 80s. An engine failed shortly after liftoff. The pl;ane stalled and crashed. At the time the FAA was stressiing noise abatement procedures, which meant very steep climb profiles after takeoff. The profiles were safe if all engines were operating normally, but unsafe with less than takeoff power. But that wasn’t being stressed in the training – only that the steep profile was the goal. Unfortunately, the Captain of the 10 tried to maintain the profile with disastrous results.The exact engine failure was simulated in the simulators and when the pilot lowered the nose slightly, the airplane climbed out successfully. Trying to maintain the profile in the simulator resulted in a low altitude stall.
In the military we spent a lot of time on stall and spin recognition and recovery. In the airlines it was considered to be necessary to be able to recognize and recover from a stall but there was less training because the airplanes were equipped with stall warning devices – both a stick shaker and an aural warning. At the time most airline pilots were former military pilots. It was assumed we knew how dangerous stalls were and that the stall warnings would be heeded. Well, we now know that’s not always the case. I don’t know anything about the Air France pilots’ backgrounds, but I’m guessing it wasn’t ex military.
Just had a lengthy conversation with my Airbus pilot friend.
First I made a classic mistake. I had not read the most recent article, and commented on the comments. I went off on a bit of tangent in my rant about stall recognition and recovery.
He set me straight. Up until this accident they DID NOT TEACH stall recognition and recovery in the Airbus, because it will not stall in the normal and first or second redundant flight control laws. It is also the case that the Airbus does not have the conventional stall warning and protection systems common to other jet airliners.
The airbus will not stall when operating in the flight control regime they were in. Strangely enough, although the aural warning says “stall, stall”, the aircraft is actually in an angle of attack (called Alpha 4) below that of the stall and no matter what the pilot does, it will not increase the angle of attack beyond that point.
In this case they got themselves into a very high drag AOA, in a very heavy airplane, at very high altitude, which is a very untenable situation. It resulted in an extremely high sink rate.
By the time the Captain arrived and realized what was happening it was too late to recover.
Training, or lack of it, may have led to the mysterious actions on the part of the Bodin. They are taught in certain circumstances of high sink rate –near the ground–to simply hold full back stick with maximum power. He apparently applied that nonsensical technique in a totally inappropriate situation–but it was all he knew. I say nonsensical, because my friend tells me that in the airbus they set a “green dot” on their attitude indicator which will show them the max-lift-over-drag attitude at any moment. In other words the optimum attitude. That is of course where he needed to place the nose of the airplane–but that is not procedure. Procedure is to mindlessly hold full back stick and let the computer take over.
So, although Bodin’s actions seem inexplicable, it appears that he was applying what he was taught; and all that he knew. Confirmed by his comment to the Captain that he was holding full nose up controls. The Captain’s response shows that he understood the situation, but too late.
Oldflyer said, “So, although Bodin’s actions seem inexplicable, it appears that he was applying what he was taught; and all that he knew. Confirmed by his comment to the Captain that he was holding full nose up controls. The Captain’s response shows that he understood the situation, but too late.”
Pretty much confirms my opinion that, in their efforts to eliminate pilot judgment and error with technology, they constructed another path for pilot error.
As a former USAF pilot and Airline Pilot for 33 years (USAF Flight Test, Air Sea Rescue, MATS, AF Flt instructor, Airline Captain, Pilot Supervisor, General Mgr of Flying, International 747 Check Captain, and after retirement, a NASA – ASRS program Analyst for 16 years) I have to say that both “J.J. and “oldtimer” are correct and surrounded the problems and solutions correctly.
The lack of “Ye old hand flying experience and Instrument proficiency” was a huge factor here, particularly when the Pilots are NOT ex-military. Add to that the culture of the designers of the Airbus, one that takes the crew out of the loop, like the computers and side stick controllers as mentioned by them. In the older days, when the DC-10 was a primary crash aircraft, the last words on the voice recorders were, “Oh S..t!”. On the AIrbus the last words are, “…what is this airplane doing?”. Some years ago when a Japan AIrlines flight came into SFO without its computer system, the Captain declared “an emergency” when he had to manually fly the aircraft into the approach and landing!
Only one thing was left out, the answer to the primary question of why this sad and needless event even occurred. The basic question WAS addressed by a follow up “J.J.” statement of “My question is why was the Captain not in the cockpit when the airplane was entering an area of severe thunderstorms?”
The crew SHOULD HAVE HAD A WEATHER BRIEFING PRIOR TO DEPARTURE. I don’t know if the NTSB or FAA ever asked anyone in Operations in Brazil or Paris if there was a review of THAT procedure. Why had the Captain NOT planned to be in the cockpit at the controls at a critical time when approaching the front with Thunderstorms over 57,000 feet in altitude? Why did they NOT divert to another “track system” to avoid that area? They had weather radar for reference to these storms and most probably had :45 minutes in which to take action. The planning was OFF FROM THE START. The Captain was “Missing in action” and showed up too late to save his passengers from an unnecessary fate. In would have been interesting to have listened to his crew briefing regarding the weather concerns (if they, as a crew, had any). A post crash review of the storm system indicated parallel, opposite direction, vertical shaft speeds exceeding 100 knots with the presence of super-cooled water that would adhere to and freeze upon contact with any object, particularly with external, and even heated, pitot-static system tubes.
Oh, “J.J.’, by the way, The AA crash at ORD was not a stall induced by the aircraft’s attitude or “ENGINE FAILURE”, but by the loss of the left leading edge flaps when the #1 engine CAME OFF, UP AND OVER THE WING, taking the hydraulic and electrical lines out when crushing the leading edge. Those critical lines were ahead of the main spar, unlike the L-1011 where the lines were BEHIND the main spar. The leading edge flaps then retracted and the aircraft rolled over to the left when the left wing stalled. The Pilot had no chance at all to recover from that roll. The cause of the crash had nothing to do with the sound abatement procedure as they were in their initial climb mode, straight out.
Captain Roy,
Thanks for having a better memory than mine. I should have looked it up. Here’s what I found:
“The first officer raised the nose up to 14 degrees to reduce the airspeed from 165 knots (306 km/h), to the V2 speed of 153 knots (283 km/h), the speed specified in the emergency procedure for engine failure during takeoff. However, the engine separation had severed the hydraulic fluid lines which controlled the leading edge slats on the left wing, causing the slats outboard (towards the tip of the wing) of the separated engine to retract under air load. The retraction of the slats raised the stall speed of the left wing to approximately 159 knots (294 km/h), 6 knots (11 km/h) higher than the prescribed V2 speed. As the aircraft reached V2, the left wing entered a full aerodynamic stall. The resulting asymmetric lift caused the plane to roll rapidly to the left and enter a steep dive from which it could not recover despite maximum opposite control inputs by the First Officer.”
Our instructors on the 10 at brand X began hammering on not worrying about climb profiles (they had been emphasizing maximum climb profiles from lift off until we were above 1500 feet for noise abatement) or V2 speeds, but speed and a positive climb rate. I recall that the profile had been simulated by our instructors and they were convinced that the airplane would not have crashed if the higher airspeed had been maintained. It is, of course, a lot to ask of someone to abandon the SOP when you don’t know you have lost partial lift on one wing. Our answer was keep the speed up, maintain a positive though gentle rate of climb, and get cleaned upas soon as possible.
Then there’s this info under the conclusions:
“The DC-10 incorporates two warning devices which might have alerted the pilots to the impending stall: the slat disagreement warning light, which should have illuminated after the uncommanded retraction of the slats, and the stick shaker on the Captain’s control column, which activates close to the stall speed. Unfortunately, both of these warning devices were powered by an electric generator driven by the number one engine; following the loss of that engine, both systems became inoperative.[1] The First Officer’s control column was not equipped with a stick shaker; the device was offered by McDonnell Douglas as an option for the First Officer, but American Airlines chose not to have it installed on their DC-10 fleet. Stick shakers for both pilots became mandatory in response to this accident.[9]”
My recollection is that our airplanes had stick shakers for both pilot positions before they were required. (Although I would not take a bet on it. My mind started to turn to mush a few years back.) The above quotes come from wicki:
http://en.wikipedia.org/wiki/American_Airlines_Flight_191
Thanks for making me take a closer look..
J.J., you make a lot of sense and you research very well… American airlines did make a bad decision re the “No” on the FO’s stick shaker just like Boeing did on the initial design of the 757 when they elected not to have a warning horn for a situation wherein the throttles were advanced to the TOGO position and the spoilers were deployed during the time the crew had advanced the throttles. That was the final link in the error chain for the AA 757 flight into Cali, Columbia, about 1995/96. That accident was a classroom 101 for the extended error chain of about 10 errors, culminating in a fatal ground contact. Good reading your input fellow flyer. Captain Roy
“Error chain”… that’s a chilling way to put it, but it seems to me that so many of today’s air transportation accidents are not the result of isolated errors, but rather compounded chains of evens. From what few I’ve read (in the news, in periodicals in the library, online, etc.), some out-of-the-ordinary event occurs, suddenly the pilot(s) is(are) forced into an unusual situation, they fall back on what they know but the situation is such that what they know doesn’t correct the situation (or in some cases, like the one this thread is centered on, makes the situation worse), and then errors start compounding beyond the pilot or pilots ability to track.
What’s even more chilling is that for some of the mishaps I’ve studied, these events also didn’t happen in fractions of a second, or in some cases not even single digit minutes. They’ve happened over time. The AF 447 incident, for example, extended over 5 minutes; the crash was not the result of any instantaeous mishap. And for that entire time, those pilots were operating in emergency mode and almost certainly were **very** alert and paying attention to things.
I’m struck by this. I’ve studied 9/11, and am passingly familiar with the FAA and NORAD/NEADS response to the incident. I understand from listening to the NEADS tapes and from reading excerpts from multiple people involved just how much information the poor air traffic controllers had to process on the fly. It’s honestly chilling to put yourself in their seat and then read the timelines, noting that one plane suddenly goes without communication and refuses to acknowledge radio calls, another drops from primary radar when the transponder gets shut down, a 3rd broadcasts a foreign accented message to stay calm and stay in their seats to its passengers over the radio by mistake… and then an explosion of false hijacking reports that confused the hell out of everybody and made tracking which flights were in a state of emergency near impossible (for me as a layman, it becomes clear as day why the ground stop was ordered. Confusion was that rampant). Yet, too many who don’t look at these things in depth ask questions in a way that betrays their ignorance of the fact that such events are not single points of failure, but “error chains” of events that compound each other. Idiot conspiracy theorists *to this day* continue to ask stupid questions like “Why did Flight 77 fly 44 minutes without being intercepted by the Air Force”. They IGNORE the fact that the FAA didn’t figure it out until nearly 30 minutes after the hijacking (procedure forced the Indianapolis ATC center to look for a crash along it’s current flight path, which was westbound), until 40 minutes after the fact. And NO ONE was just sitting on their butts for that entire time; on the contrary, much flurried activity was occuring, first along the east coast, and then spilling across the midwest.
And those are the guys on the ground. Not anyone in the air who were dealing with the emergencies.
Error chains. Many of these events are actually rather complex, and have multiple points where things went wrong. I don’t know what lesson to take from that bit of insight, but it’s rather obvious that such tragedies cannot be distilled down to simple, single paragraph narratives. The truth itself resist such simplification.
Whoops. Another error from poor self-editing.
Corrected paragraph: “… the Indianapolis ATC center to look for a crash along it’s current flight path, which was westbound), and the military wasn’t even notified until 40 minutes after the fact. ”
Again, sorry. I wish posts could be previewed and edited in WordPress.
EMH,
You are onto what causes many accidents in aviation and elsewhere. Namely error chains. Very difficult to overcome. That is one reason why aviation accidents are so thoroughly investigated and the results made known to all people in aviation. Learning from other people’s errors improves safety. The Navy has a mgazine, “Approach,” that is dedicated to putting out information on accidents and incidents so that others can learn from the mistakes that have been made. However, there are always new unknown unkowns (thanks, Don Rumsfeld) cropping up. Thus, accidents will never be eliminated completly.
Your thoughts about 9/11 are interesting. Have you read, “TOUCHING HISTORY,” by Lynn Spencer. If not, I highly recommend it. It details all the confusion at ATC and the NORAD as the events began to unfold. When the full stop order was made, airplanes all over North American airspace were forced to find places to land. That there were no major accidents as a result was amazing. I could not put the book down and found my pulse elevated at times. It’s that good.
Nope. Haven’t read it yet, but I’ve been planning to for some time now. So much to read, so little time… 🙁
Oldflier,
It’s always convenient to blame “training”. And of course one could go through an ever expanded training regimen in airline transports and get the same result in a similar situation. Remember years ago when pilots went through a spate of “unusual attitude” training situations. Didn’t help this guy did it.
Of course, flying during a pitch black night over the ocean with no horizon definition didn’t lend itself to “outside the cockpit” orientation did it.
Or maybe you don’t think that flying with a 41.5 degree nose high pitch attitude while losing altitude at a rate of 10,000 feet per minute is an unusual attitude. Obviously he didn’t recognize those two cues. One because he couldn’t use outside references when his instruments went flaky, and the other…well I guess I would chock that one up to brain lock….to many conflicting instrument indications presenting a totally disorienting picture, not to mention the confusion with the seat of the pants inputs. That is what I call situational disorientation.
There is normally only one way out. Give control to the non-disoriented pilot. Been there, done that (granted in fighters, not transports). He didn’t do that. Disaster ensued.