It’s been an exciting few weeks in the search for Air France 447. In just 20 days or so, searchers located the fuselage and both the cockpit and flight data recorders in 13,000 ft. of water on the floor of the South Atlantic. More victims are also still entombed in that fuselage as experts try to figure out how to recover them without destroying the bodies in the process.
Then last Friday came the French BEA transcript of the last few minutes of the flight before it pan caked into the water from seven miles above the ocean killing all 228 people aboard, as well as a few indisputable facts. Essentially, the airplane was out of control — completely stalled actually — almost from the moment the autopilot and autothrottles kicked off due to ice-clogged pitot tubes. The question of course is why were three experienced pilots — admittedly some more experienced than others — unable to make the aircraft fly at some point during the three and half minute drop to the water?
As I explained to a Fox TV News anchor Saturday afternoon, the scene inside the cockpit certainly would have been chaotic with darkness adding to the turmoil of heavy rain, turbulence and an array of warning lights and chimes all threatening certain death at the same moment. It would be tough for anyone to think clearly in that environment. Be that as it may, the pilots were paid and given the responsibility of being able to do just that … or were they?
Some reports claim the Air France pilots were never trained to deal with multiple failures surrounding events like they experienced that night. While it’s important to withhold final judgment of the cause of the 447 crash, it is clear that the pilots were either unable to recognize the attitude of the airplane at any point, or were unable to convince the aircraft to follow their commands. Did they hold the aircraft in the nose up condition then as the data recorder suggests? This is the part of the investigation we’ll need to wait for.
A Little History
Pilots losing control of an airplane is not a new event however. USAir lost a Boeing 737 on approach to Pittsburgh in 1994. The final words of the captain as the airplane nosed into the earth was “Help me pull,” thinking they were diving. Like the Air France accident, the USAir aircraft was stalled all the way to the ground.
An American Eagle crew lost control of their ATR-72 and crashed after their aircraft also stalled during an encounter with ice in 1994. In January 2009, a Colgan Air Dash 8 Q400 crashed on approach to Buffalo when that captain misinterpreted a stick pusher for a stall and caused the aircraft to fall from the sky killing all 50 aboard.
So was this an unfortunate technology-induced accident or was the crew merely as far behind the aircraft as some of these other unfortunate
aviators? Sure these French pilots never recovered from the stall, but was it because they were not properly trained to handle this kind of chaos, or did the Airbus A330 try something on its own that these aviators were unable to remedy?
The question is not whether or not the pilots are responsible for the crash, for surely they are … pilots are always responsible, especially the PIC. A stalled wing is a situation every new pilot sees, almost from their first training flight.
The TV folks asked me, “Was there anything the pilots could have done to make their airplane fly again.” There was only one correct answer. “Absolutely. They could have reduced the angle of attack to a point where the airplane would once again fly.”
The real question is why?
Rob Mark, Publisher
Where in the BEA report did it say the pitot tubes froze? You need to make sure you have the FACTS. In all likelihood they did freezed, but don’t state it as a fact.
Also, the aircraft went into “alternate law”, which eliminates the envelope protection. Therefore, I doubt if the airplane was trying to “outsmart” the flightcrew. The thing I found puzzling was the lack of any angle-of-attack display mentioned in the report (yes I know, AOA display is “normalized” with airspeed for display – but why doesn’t it go to raw AOA with loss of airspeed?). I have to believe the the flightcrew should have noticed the unusual pitch angle in conjunction with the altitude loss and should have connected the dots on that. Stall recovery would be nasty and take lots of altitude, so it may have been hard to convince themselves to get the nose down. Two things I would look at with respect to airplane: 1)why the loss of airspeed information? and 2)why wasn’t the stall warning continuous? From a crew training standpoint: Stall recognition and recovery with partial panels (hummm… basic IFR stuff).
I didn’t mean to imply that the pitot issue was mentioned in the BEA report. That came from another source.
I was actually trying to give the crew the benefit of the doubt Ramos when I mentioned the airplane sneaking up on them. There’s a checklist specifically for a loss of primary attitude information on the A330.
And you’re point about stall recovery costing lots of altitude, I absolutely agree. They had plenty … at least to begin with.
Basic IFR stuff?
Sure looks like it to me too. When I read the BEA report, I remembered something my instrument instructor told me 35 years ago … “When all else fails in a stall, full power (which these guys did) and reduce the angle of attack until the airplane flys (which these guys did not).”
As I said in the story … did they not release back pressure because they were confused by something else that made them think they were flying? The recording says they knew they were losing altitude, so the static sources must have been clean.
How much does an airplane need to fall with full power before you shove the stick forward because you can’t think of anything else to try?
Three questions every pilot should ask (before speculating):
1. Any other aircraft ever have a ADC/AOA fault/damage that resulted in loss of pitch control? Yes. C-17 birdstrike/loss of radome in Oklahoma.
2. Where are the airbus test pilots to weigh in on the flight control laws?
3. If a C-17 crashed in Alaska last July due to a STALL and an airbus crashed due to a STALL, then how might I be fooled and how can I prevent that from happening?
Mark … excellent points all.
I neglected to mention that C-17 crash. Thanks for the nudge.
Here’s a link for readers. http://bit.ly/ifzfZx
Who would have thought we’d be talking about what APPEARS to be such simple “basic IFR stuff,” in accidents on these big airplanes?
My impression of airline pilots is, a lot lack stick and rudder time.
Sitting down with one pilot one day, he couldnt even read a VFR map. Kinda funny considering the thousands of hours he has over me. But thats thousands of hours behind working electronics;-)
No offence to any air line pilots, but it would be shear luck if they got out of it.
Some very simple statements:
1)Why the French BEA, investigating without neutral supervision the mainly French owned Airbus and Air France, has so quickly released such report implicitly blaming the pilots??
2)And if they considered necessary to do so, including some remarks done by the Pilots, why the did not release any interchange in the cockpit supporting that the 2 copilots and the
captain(who returned before the plane began to fall)were so ignorant that they didn’t know
that “nose-up” was 180% contrary to elementary anti-stall rules?? all the three (3) of them ??
4) The logical answer to 3.: : As probably releasing the complete transcript would blame Airbus, they opted to release what they did, so to buy some time, hoping to find some fact helping to exonerate Airbus.
5) By the way” If there were some error by the Pilots, this would not exonerate any deficiency of te
“Fly-by-wire” system, only magnify it!!
6) Further: there was no storm, as they had circumvented it. And that it was dark outside, seems not relevant.
7)FINALLY: why the communication with the outside world broke dawn at the same moment as tha autopilot disengaged? Could be the young copilot DISCONNECTED IT??!! And what were the massive failure signals received just before this break down?
THERE ARE NONE SO BLIND AS THOSE WHICH WISH NOT SEE !! N’est pas ??
Somebody please comment!!
Colgan 3407 at Buffalo, Turkish 1951 at Amsterdam, and now Air France 447. All recent accidents where improper basic stall recovery technique, among other links in the accident chain, brought down an airliner.
Basic airmanship is giving way to what some may think is a lack of experience rearing its ugly head. But, if you look back at the basic training of each of these pilots, I am willing to bet they were all taught the basics of stall recovery: full power, reduce angle of attack, active rudder control to maintain directional control.
IMO, it is the training and conditioning later in their careers that lead to the reliance on automation, and the technique of minimum altitude loss during stall recovery taught for years and years in swept wing jets. Full power, don’t lower the nose too much, and eventually you will power out of the imminent stall.
The fact that it appears in each of these accidents, the Pilot Flying rode the aircraft in to the ground with full aft stick is disturbing.
As promoted by the FAA, the stall recovery maneuver (actually an approach to stall) as taught in the simulator was more of a coordination exercise than a true “get this airplane out of trouble” demonstration.
Lest we think, “Not me, I am not going to get my airplane in a stall situation”, I am willing to bet that was the thought process of each of these aviators as well. Unfortunately, we do not get to test our training for this type of situation until it really, really is a life or death scenario.
No sim instructor giving a grade on this maneuver, boys and girls, it is the ultimate pass/fail. In what sounds like a similar, though not exactly the same, scenario, a Northwest crew over the Pacific in an A330 passed, while the Air France crew, rest their souls, did not.
I was encouraged during our last G550 recurrent with a change in the training scenario. Fully configured on the glide slope, pull the power to idle at 1000 feet AGL and let the airplane slow through shaker to the pusher. Face full of dirt at 600 feet AGL gets your attention! Full power and don’t pull until the pusher is done with its job, or one second, whichever comes first.
“A-thousand-one” takes longer than you think! Big lesson is not to pull through the pusher. I know that seems like basic airmanship, but if it is a surprise to you, you will revert to your training, or muscle-memory. Face full of dirt, the survival instinct says “PULL”, unless your training is to recognize a stall and either push, or allow the airplane to push for you.
I’ve rambled on long enough.
I do believe that the FAA now recognizes this problem (that I believe they have created!) and we will see more realistic training scenarios in the sims soon.
Rob,
You’re writing a book on this? (according to the Fox report)
Questions regarding AOA displays. When airspeed data is lost, no, does not revert to any kind of raw AOA display. (though an AOA display is an option on the airplane – apparently this aircraft did not have it)
I too would like to hear more of the transcript. Surely there was a lot more said then what was in the report.
Still, maintaining a pitch attitude of about 15 degrees during this entire episode is not a recipe for success. The report indicates that they actually got their airspeed indications back, but even then continued to pull back on the stick in an apparent attempt to make the plane go the opposite direction than which it was rapidly going.
Delta teaches their Airbus pilots, that in the event of loss of airpseed data, to fly a given pitch attitude and power setting. As I learned long ago in primary flight training Pitch+Power=Performance. Still true.
BP
A330 captain
As I mentioned above Bill, I was trying to give the pilots the benefit of the doubt that perhaps for some reason they were totally disoriented by something of which we are not aware.
But another item I find very odd and to your point of the transcript is that the captain seemed to be doing little more than supervising the situation.
How much room is there between the back of the cockpit and the panel on the A330?
Wouldn’t you have expected the captain at least to have said, “Push the damned nose down,” or finally grabbed the stick himself? The aircraft was falling and without a seatbelt, perhaps he was in an almost zero G situation, but still little on the CVR from him.
And yes to your comment about the book. I didn’t think anyone watched those TV segments except the family.
BP
“I too would like to hear more of the transcript. Surely there was a lot more said then what was in the report”
Robert Merk
” Wouldn’t you have expected the captain at least to have said, “Push the damned nose down,” or finally grabbed the stick himself? The aircraft was falling and without a seatbelt, perhaps he was in an almost zero G situation, but still little on the CVR from him”
This two remaks tell it all. Factually the term “lot more” used by BP is very charitative, as BEA has released nothing, except some inicuous remarks as “go to the left” and remarks regarding altitude and speed during the fall.
And of course, the interaction between the pilots would disclose most certainly if the pilots were so incredibly incompetent as painted or if they were not able to do the required because of the the inadequacy of the “fly-by-wire” Airbus system.
SO: WHY IS BEA REFUSING THE CALL TO RELEASE THR COCKPIT AUDIO TRANSCRIPT?? You must not be a conspiration fan to smell something rotten in Denmark…. sorry, France!!
Lack of seemingly basic stall recovery techniques in all these accidents tells me one thing: That the pilots don’t believe they’re capable of stalling an aircraft any longer. Colgan probably thought it was tailplane ice, IMO, but it seems like the idea that they were stalled never even crossed their mind.
The basic stick-and-rudder stuff doesn’t happen because they never get to exercise it.
The other thing that I don’t like about the Airbus fly-by-wire systems: The airplane acts differently depending on whether it’s in normal, alternate, or direct law. IMO, that’s too much for the pilots to process in an emergency. The saying “You do not rise to the occasion, you fall to the level of your training” comes to mind, though “training” could be replaced with “experience.” Just like the lack of stick-and-rudder flying time, the pilots are used to normal law, where the plane responds one way (and can’t be stalled – See the first paragraph!) and they’ll keep flying it that way when an emergency happens. Airbus fails at the most simple UI design concepts.
Rob,
There’s quite a bit of room aft of the center pedestal in the A330. Enough for the jumpseat to have swivel and recline and room to move past it.
It is hard enough for the jumpseater to see, much less take over, the sidestick control without being in one of the two pilot seats. I don’t think that’s a likely scenario, as he would have to pretty much be in the lap of one of the other pilots to do so.
I do sympathize on the dark and confusing environment. I find the statement “2:12:02 PF said, “I don’t have any more indications”, and the PNF said, “We have no valid indications.” to be curious. Didn’t they have valid attitude data then? They should have – and the ISIS is a completely separate source of attitude as well. I am only aware of missing airspeed data. Altitude and vertical speed (which is primarily derived from IRS data) should have been available and the report doesn’t seem to contradict that theory, from what I’ve seen.
>> The aircraft was falling and without a seatbelt, perhaps he was in an almost zero G situation,<6) Further: there was no storm, as they had circumvented it. And that it was dark outside, seems not relevant.<<
I have to disagree. While they did do some deviation, the satellite photos and other weather data seem to indicate it was a pretty good line of weather that they probably did not completely avoid.
The comment "..or if they were not able to do the required because of the the inadequacy of the "fly-by-wire" Airbus system." That statement assumes that there is some known "inadequacy of the 'fly-by-wire' Airbus system" which I must protest is not the case. If that is his OPINION, then it should be stated as such and no more than that.
I think Kent Shook overstates the differences between Normal, Alternate, and Direct law (they never got into direct law). The handling is not vastly different. Protections are lost to varying degrees, but not so much that they should loose control – after all it is not normal to depend on the protections of Normal Law to keep one out of trouble. – Though it is true that in Normal Law full back stick will result in Cl max up to but not over max pitch/bank angle and g-load(great for terrain ad other emergency avoidance maneuvers). I think this is a great thing and avoids having to dance around stick-shaker to know where the limit/max performance is.
Hopefully the AF pilots weren't taught that whenever the going gets tough to just pull back on the stick [I say that tongue in cheek] – even though it appears (at this point) as though that's what they did.
To that point, stall training for these aircraft has recently (but prior to this report) been reemphasized – mostly due to the Colgan accident, but obviously relevant here too.
BILL AND KENT
I agree with Kent. Please see my insertions in capitals in Bill’s text.
6) (Oton) Further: there was no storm, as they had circumvented it. And that it was dark outside, seems not relevant.<<
(Bill) I have to disagree. While they did do some deviation, the satellite photos and other weather data seem to indicate it was a pretty good line of weather that they probably did not completely avoid.
(Oton) BUT ONLY NORMAL BAD WEATHE< NOT A CONCERNING STORM!
(Oton)"..or if they were not able to do the required because of the the inadequacy of the "fly-by-wire" Airbus system."
(Bill) That statement assumes that there is some known inadequacy of the 'fly-by-wire' Airbus system" which I must protest is not the case. If that is his OPINION, then it should be stated as such and no more than that.
(Oton) HERE I DISAGREE, IT IS NOT 'MY" OPINION, BUT OF A LOT OF PEOPLE, WHICH ASSESS THAT THE OVER-AUTOMATION AND THE LACK OF EASILY ACCESSIBLE MANUAL REMEDIES IS DANGEROUS, ESPECIALLY IF THE SOFTWARE IS SOMEWHERE FAULTY (Kent ALSO IS WAS IMPLYING SO!!). BUT I WILL CHANGE MY STATEMENT AS FoLLOWS: ""THERE IS A BY MANY PERCEIVED INADEQUACY…."
(Bill) I think Kent Shook overstates the differences between Normal, Alternate, and Direct law (they never got into direct law). The handling is not vastly different. Protections are lost to varying degrees, but not so much that they should loose control – after all it is not normal to depend on the protections of Normal Law to keep one out of trouble. – Though it is true that in Normal Law full back stick will result in Cl max up to but not over max pitch/bank angle and g-load(great for terrain ad other emergency avoidance maneuvers). I think this is a great thing and avoids having to dance around stick-shaker to know where the limit/max performance is.
Hopefully the AF pilots weren't taught that whenever the going gets tough to just pull back on the stick [I say that tongue in cheek] – even though it appears (at this point) as though that's what they did. To that point, stall training for these aircraft has recently (but prior to this report) been reemphasized – mostly due to the Colgan accident, but obviously relevant here too
(Oton) WE SHALL NOT FORGET AN ALWAYS VALID AXIOM: AUTOMATISM IS MEANT TO DO THINGS EASIER AND BETTER, IN ORDER TO AVOID HUMAN ERROR, NOT MORE COMPLICATE AND CONSTITUTING PER SE A FOUNTAIN OF ERRORS.
As I said in my comment of 27th May, the Aerospatiale Electric Flight Control System coupled with the fact that the vertical stabilizer broke off in flight, and the apparent lack of latent pilot skills of the Air France cockpit crew all contributed to the destruction of this Airbus A330
To prioritize; the automated flight control system is the real cause of the accident. Because the airplane cannot be controlled without the electronic system functioning in some minimal manner, and with some level of flight control laws governing, the pilots were faced with the equivalent of a Grand Mal stroke (in a human), and losing the vertical stabilizer they were doomed – to say nothing of their lack of piloting proficiency in hand flying, having been babied in their careers by side sticks, auto-trim, and auto-throttles.
The Airbus A330 design killed these people. Aerospatiale set them up as managers of their machine, essentially passengers, in a marvel of deadly automation.
One point, why is no one paying attention to the loss of the vertical stabilizer?
Bill P. is right.
The pitot tubes (the so-called speed sensors) icing over are only ancillary to the fundamental probable cause. They might have precipitated the chain of events within the Aerospatiale computer operated flight control system, but any FAA or DGAC certified transport category airliner should be able to fly safely with no airspeed indication at all.
It’s not difficult to do using only power references and attitude control. Keep your nose on the horizon and set cruise power. Turn off autothrottles. Steer using the magnetic (whiskey) compass.
I’ve personally hand flown (twice) transport category aircraft through violent weather on international trans-ocean flights for hours at a time, using only attitude reference and power setting alone – disregarding the air speed indication.
You must disregard the erratic air speed indications that are caused by up-drafts and down-drafts within the cumulonimbus storm cells.
Sooner or later in their careers, all professional pilots will be thrust into the situation where a violent storm cannot be avoided. This type of extreme weather flying is a once or twice in a flying career event, but you owe it to your passengers to be an expert at flying like this – or to avoid the weather if you can and divert.
No matter how sophisticated or modern your airplane is, no matter how many computers operate the flight controls, the laws of aerodynamics and physics still apply to the air machine moving through the fierce storm of air currents, turbulence, rain, hail, lightning and icing.
Be brave and resolute. Do not give up as you might have to fly like this for hours at a time in order to arrive safely at your destination.
Never forget, the best computer aboard the airplane is still the one between the pilot’s ears!
Michael
Michael
I didn’t realize that he vertical stabilizer broke off!! Where this info is coming from?? And this happened before or after the stall?? and if the former, WHY, as it is confirmed that the weather, even if not good, was not dangerous(which would explain why the captain, assuming it is true, left for his nap!)
Where this info comes from??
As for your post 9.25 pm, I agree widely
If finally the result od the investigation will confirm the danger of the Airbus Fly-by-wire system, which not only affects the A330-200, what remedies Airbus could put in place??
Being the issue conceptual, I cannot se other than a long drawn solution, preceeded by a new design. Your assessment??
Above could imply tent of billions of costs pluss loss of sales, so it would be understandable, but by no means justifiable, that BEA eventually tries to cover it up!
Oton,
” BUT ONLY NORMAL BAD WEATHER NOT A CONCERNING STORM!” upon what data did you base this claim? Storms in the ITCZ typically have poor reflectivity for their turbulence severity. Have you seen the study on the severity of this line of weather?
Michael, (we must have two Michaels here – since they’re saying wildly different things)
>>To prioritize; the automated flight control system is the real cause of the accident. Because the airplane cannot be controlled without the electronic system functioning in some minimal manner, and with some level of flight control laws governing, the pilots were faced with the equivalent of a Grand Mal stroke (in a human), and losing the vertical stabilizer they were doomed – to say nothing of their lack of piloting proficiency in hand flying, having been babied in their careers by side sticks, auto-trim, and auto-throttles.
The Airbus A330 design killed these people. Aerospatiale set them up as managers of their machine, essentially passengers, in a marvel of deadly automation.
One point, why is no one paying attention to the loss of the vertical stabilizer?<>Because the airplane cannot be controlled without the electronic system functioning in some minimal manner<<
Are you aware of a totally malfunctioning flight control system that nobody else is? Enlighten us please.
As one who is intimately familiar with the design and operation of this aircraft, I think you are way off base in your blame of the flight control system. You may not like it for your own reasons, but that does not mean it "killed these people."
I read nothing in the report about the state of the flight control system that is not a predictable result of the loss of airspeed data. What did you read?
Soon after the accident the anti-composite folks came out of the woodwork. Now, I suppose, it's the anti-flybywire crowd? C'mon facts, facts!
How did you conclude that the automation resulted in a "fountain of errors?" In what way was that airplane performing any differently than a 767 would have (which operates all the time in the equivalent of Direct Law)
If Ken has not personally flown an Airbus FBW aircraft than I submit he is not qualified to assess how differently the airplane flies in each flight control law state.
I agree with Bill. Anything without a fact behind it attributed to some reliable source is just speculation. That won’t really lead the discussion along at all.
Bill flies an A330 for a living. None of us do although I have flown some of the Airbus family.
So what do we know from the BEA report? For sure that these guys flew the airplane down to the water with the wing stalled. But what else?
Oton’s point about the limited CVR discussion is an interesting one though. It does seem as if the data the media has read could only lead one to a conclusion that the pilots are at fault. Doesn’t that seem a bit odd to everyone else too? Did we all fall for a bit of a grand PR plan designed to get us to look closely at the crew and no where else?
We can’t forget that this is the French government reviewing a French airline and a French aircraft manufacturer (for the most part). One of my sources told me that the Brazilians wanted the U.S. NTSB to investigate this accident because of the potential for a major conflict of interest.
Perhaps this is all coincidence and that we’ll find out that the pilots blew it. Like Bill though, I also wonder why no one seemed to think that full power and at least putting the nose on the horizon would have helped … if they had attitude indications. And they should have had some.
But even with no valid horizon we revert to stick and rudder skills … assuming we still/ever had them. If you have full power in there there is only one thing left to affect the flying of the airplane … pitch. If you’re holding it back and the airplane’s not flying you shove the stick forward. If it’s still not flying, push some more.
Unless there was some major control malfunction which does not seem to be indicated right now, the airplane would have flown eventually. Might not have been pretty, but it should have at least flown.
Those comments about no valid indications though also bother me. That is so vague. Here are three people who all probably realize they are about to die and they are speaking as if they’re working some sort of mental math problem.
How come no one seems to be upset (hence the thought about the edited CVR transcript)?
The vertical stabilizer WAS found separated from the rest of the aircraft right after the accident which made me connect it to the AA bus that crashed in 2002 at first.
Not fact on that connection though, merely speculation since I have not been able to learn whether the experts believe the section came loose at impact or earlier as someone else suggested.
It is strange that no one has raised that topic. But perhaps the BEA already knows it came off when the airplane hit the water.
Enough for now. Time for some sleep. Again, I’m getting more concerned about what we don’t know myself.
The BEA report, released on May 27 2011, states: “The trimmable
horizontal stabilizer (THS) passed from 3 to 13 degrees nose-up in about 1 minute and
remained in the latter position until the end of the flight.”
A Flight Global article by David Kaminski-Morrow suggests that this happened automatically (in alternate law the auto-trim continues to function).
With regard to these THS data, can anyone answer the following questions:
1. Is it correct that this would happen automatically, or are there factors not mentioned, that would imply the THS was set manually ?
2. If the THS setting did happen automatically, was the (nose-up) auto-trim action the result of the pilot’s manual nose-up stick input ?
3. If not: what would have automatically triggered an automatic nose-up THS setting ?
4. Suppose the THS got stuck (not necessarily mechanically speaking, but rather “software-stuck”) in the nose-up 13 deg. position, what chances would the crew have had, if they had pushed the sticks full nose-down ? In other words, if the aircraft finds itself in a fully stalled position, is it possible to recover from the stall, with a THS stuck in 13 deg nose-up ?
Thanks in advance for your answers.
MICHAEL
Regarding my questions about the vertical stabilizer,I now realize that it was foubd about 30 miles (or was is Km) from the other floatsam.
Analyzing this.can it be assumed that when the aircraft dropped 11,000 m in 3.5 minutes, the horizontal speed component was low??
Previously, it was climbing at a angle near 13* at full power, the horizontal speed should have been in the range of 700-800 Km/h =
~~ 12 km /min which would mean (if the captioned 30 miles have any relation with the time of the break), that this must have happened BEFORE the Autopilot disengaged!
But if the 30 MILES were really 30 KM, or if the distance was initially about 25-30 KM and magnified later by wind and sea currents, than (could be) precisely at the time of disengagement! A mystery, compounded by the fact that at the same moment, the communications with the outer world were lost!
Hopefully, what I say is understandable!
And, of cause verifiable if BEA release the transcripts, as it obviously must be commented by the pilots,
Comments??
ROB MARK
What you wrote this early morning reflects widely my opinion
ANTON BEECKMAN
As you write:
“”Suppose the THS got stuck (not necessarily mechanically speaking, but rather “software-stuck”) in the nose-up 13 deg. position, what chances would the crew have had, if they had pushed the sticks full nose-down ? In other words, if the aircraft finds itself in a fully stalled position, is it possible to recover from the stall, with a THS stuck in 13 deg nose-up ?””
Yes, this address very clearly an issue which myself and other bloggers commented: possibly a frightening “1984 – Space Odyssey” scenario, i.e. computer trumping man!!
And there is a filtration stating that the captain when returning yelled to the copilot something like
“Nose Down”, the response being “I try, but it doesn’t react”
Again, easily verifiable when and if the voice transcripts are released
ROB
Only a short comment regarding that speculation is not productive. Generally, I would agree.
But as we have very few facts and even some of these seem to be manipulated, human character tries to fill the voids with speculation. And so maintain this excellent blog alive!!
To address Anton’s questions
Some background on the operation of the pitch control and trim first:
In normal and alternate laws, the stick commands a g-load/pitch rate. Essentially, the pilot commands the response (gload), the elevator moves to produce the desired result, and the trimmable horizontal stabilizer (THS) moves to provide for neutral elevator for more efficient flight.
The sidestick has no feel feedback like a conventional yoke. The sidestick is spring-loaded in the neutral position, but other than that has no “feel.” It is important to understand that in other than direct law the sidestick commands a response, not an elevator position. The trim can be set manually as long as hydraulic power to the THS is available – this is very uncommon except in direct law, as it does not affect the way the airplane handles.
In Normal law, when the AOA approaches a threshold, the flight control system will limit the AOA (among other limits) that it will fly. Full back stick will instead command “alpha max.” In Alternate law, this is replaced by “stabilities” – which mimic the typical nose-down tendency in this situation (the nose would pitch over if it were not for additional back stick). This is based on speed and not AOA. However, when there is a dual Air Data Reference failure (which probably applies in this case – since speed was apparently unavailable for some period of time) the stabilities are not available.
1. Is it correct that this would happen automatically, or are there factors not mentioned, that would imply the THS was set manually ?
ANS: The trim is automatic in alternate law. automatic operation of the trim with the stick held back appears consistant with this situation as we know it.
2. If the THS setting did happen automatically, was the (nose-up) auto-trim action the result of the pilot’s manual nose-up stick input ?
ANS:If the pilot was holding the sick back, the pitch trim can be expected to run to a higher nose-up position. This seems consistent with the report.
3. If not: what would have automatically triggered an automatic nose-up THS setting ?
ANS: n/a
4. Suppose the THS got stuck (not necessarily mechanically speaking, but rather “software-stuck”) in the nose-up 13 deg. position, what chances would the crew have had, if they had pushed the sticks full nose-down ? In other words, if the aircraft finds itself in a fully stalled position, is it possible to recover from the stall, with a THS stuck in 13 deg nose-up ?
ANS: If the THS was “stuck” in alternate law, the only result is a limit on the elevator authority. Obviously if the trim is full nose up and you ask for nose-down elevator – at a certain airspeed the elevator will not be able to provide the pitch rate you request.
If the THS were stuck in the nose-up position, I would have expected to see more nose-down stick input. However, the report seems to indicate that nose-up stick was held for much of the descent.
Floatsum /vertical stab position.
Another thing to consider is what parts broke off at surface impact (e.g., the vertical stab), and what parts floated up from 12,000 feet under water (or whatever the correct value is), or broke off on the way down.
I doubt that debris coming up from the depths will come up perfectly vertical, nor very quickly.
In like manner, had the stab broken off in flight (as has been professed by some – though not supported by the report) it is not likely that it would have come straight down either. Had that been the case, the DFDRt should also show a loss of rudder position data if and when that occurred. There’s been no mention of that.
So, this is not like a land-based accident scene where if the vertical stab had been found 30 distance units away from the rest of the wreckage, there would be little other logical conclusion to draw than it broke off in flight.
This presents quite a more complicated 4 dimensional vector problem than simply: the stab was here, the X was there, and drawing some simple conclusion from that.
As I’ve pointed out before, if they hit the water with a vertical speed of 10,000 fpm (which I believe is about what was reported), that’s 113mph in the vertical – talk about a hard landing! Stuff is gonna break off!
Bill P.
you say:
“”If the THS was “stuck” in alternate law, the only result is a limit on the elevator authority. Obviously if the trim is full nose up and you ask for nose-down elevator – at a certain airspeed the elevator will not be able to provide the pitch rate you request.
If the THS were stuck in the nose-up position, I would have expected to see more nose-down stick input. However, the report seems to indicate that nose-up stick was held for much of the descent.”
Reading the complete post,it is clealy visible that this autmatized system,which here is only a timy part of an aircraft systes, is a tongue-twister and makes the flying of an aircraft more difficult instead of easier, unless the condiions are erfectly normal This alone sems to merit a condemnation.
As for your response to Anton above:
1) “”If the THS was “stuck” in alternate law, the only result is a limit on the elevator authority. Obviously if the trim is full nose up and you ask for nose-down elevator – at a certain airspeed the elevator will not be able to provide the pitch rate you request OK!!
1a) QUESTION: If for a false instruction of the alternate system, fed by confusing speed signals (IS THIS POSSIBLE??) the aircraft was nose-up after the supposed Pitot failure, but still not stalling, and the copilot would ordered nose-down, would the aicraft obeyed??
2) You write: “”If the THS were stuck in the nose-up position, I would have expected to see more nose-down stick input. However, the report seems to indicate that nose-up stick was held for much of the descent””
My question: How BEA knows in what position the stick was?? Do they monitor the postion of the stick direcly or only by he consequences?? And even if they were able to know: how do YOU know the are telling the truth?? They have billions of reasons not to do so!!!
Factually, this seems the Gordian Knot: if he stick was down during the process, this means the Pilots were aware of what to do but the computers didn’t allow them to do so, in the best tradition of “1984 the Space Odyssey”
Oton,
The limited pitch authority with a stuck THS (of which there is no evidence in this case, by the way), applies to any aircraft with a THS from a Mooney to a 747. It is neither Airbus, nor fly-by-wire specific.
>> this autmatized system,which here is only a timy part of an aircraft systes, is a tongue-twister and makes the flying of an aircraft more difficult instead of easier, unless the condiions are erfectly normal This alone sems to merit a condemnation.<>1a) QUESTION: If for a false instruction of the alternate system, fed by confusing speed signals (IS THIS POSSIBLE??) the aircraft was nose-up after the supposed Pitot failure, but still not stalling, and the copilot would ordered nose-down, would the aicraft obeyed??<>My question: How BEA knows in what position the stick was?? Do they monitor the postion of the stick direcly or only by he consequences??<>And even if they were able to know: how do YOU know the are telling the truth?? They have billions of reasons not to do so!!!<>Factually, this seems the Gordian Knot: if he stick was down during the process,this means the Pilots were aware of what to do but the computers didn’t allow them to do so<<
ANS:Ironic that YOU would be using the word "factually" when your "theories" are not based on the facts as we know them at all, but only your preconceived notion of a software glitch in an flight control system – of which you apparently know very little!
It's "2001 A Space Odyseey"?
Bill P.
I think you are not getting my point:
– I have the SUSPICION (not enough facts to support it) that the there are very heavy software glitches. You are perfectly right by stating that this is not my specialty!!
– But I have a strong circumstantial evidence that the report issued by BEA is covering up the truth, as they are blaming the pilots of little credible misdeeds and, instead to support such by the directly available transcript of the conversations of the pilots, carefully avoid even to mention them, citing only a few laconic (as Rob Clark rightly says) statements.
– I have also major questions regarding why the communication with the Aircraft were lost seemingly at the same time of the Pitots SIMUTANEOUS(??)ALLEGED (!) failure and why the last communications were a salad of failure alarms, both events MOST inconsistent with Pilot errors!!.
Factually, here we have two (2) “parties”, one of them being the main (possibly not the sole) culprit::
– the pilots
– Airbus
The former, supported by BEA, in this case a biased and unsupervised institution defending national interest, who has available a lot of evidence but refuses to disclose it
The latter, if you wish supported (of course not only) by me, not being able to supply direct evidence because it is hidden by the former!! But I have as stated the logical evidence that who is accusing with one hand and hiding available evidence with the other , does so because he is representing … the culprit!!
I have stated repeatedly that if the evidence is disclosed, I am ready to admit I was wrong if it shows my deductions were so!!
And Paul: I may in many ways have much less specialized software knowledge as you, but on the flip side, 50 years electrical, hydraulical and mechanical experience with engineering endeavors, many very major and in leading position with mainly German, Italian and French first class companies (I am 80 and I think not senile!). And what I learned is that LOGICAL deductions generally (not always) trump specialization!
and Paul P
If you assume that all the world follow the guidelines of Anglosaxon and similar countries and their institutions (as the FAA ot NTSA), I refer what is happening just now in Japan and the FUKUSHIMA Nuclar Plant disaster, where a Mega-company (TOKYO ELECTRIC)and the Japanese Government fed the world outrageous lies and two months of misrepresentations, in the meantime admitted,the former defending its economic survival and the latter a now backfiring national prestige and economy intrest!
I’m wondering too – and Bill might grab this one – about the difference at 35,000 feet and a fairly heavy weight of the angle of attack necessary to maintain cruise and the critical angle of attack … and those relative indicated airspeeds.
My guess is the spread was pretty small which might have meant very little room for pitch changes.
So again, why try to climb?
To BILL P. and ROB CLARK (who is this Clark guy? ed.)
Please read the article
http://blog.seattlepi.com/flyinglessons/2011/06/02/sullenberger-on-the-fallacy-of-pilot-error/
There, capt. Sullenberger states between other two statements which are excatly concodnt with my position, in a better English:
“”We have to wonder whether making airplanes more technologically advanced makes it more difficult for pilots to recover when things go awry,””
Answer: Most probably, because against the pilit wishes, the System forced them to it!!
“”So that if an airplane evolves to a point that it relies on the plusses of the human and fails to take into account the minuses IT IS THE SYSTEM THAT HAS ERRED, NOT THE PILOT””
Rob: You ask very correctly: “”So again, why try to climb?””
Answer: Most probably, because against the pilot wishes, the System forced them to do so!!
oton
Rob Mark
If this is real, it constitute a good new chapter for your book!!
From the APACHE Blog:
Garry to OTON:
Here is something you may be interested in:
http://airbusvertstabs.blogspot.com/
On How Airplane Vertical Stabilizers Are Loaded,
What Happens To Airplanes When Vertical Stabilizers Fail,
And Why Airbus Vertical Stabilizers Are Probably Not Acceptably Strong
(Plus Some Notes About
Fly-By-Wire, Envelope Protection,
And Why Composites Are No Less Safe Than Aluminum
Airbus has a history of problems with composite materials used in critical areas of their aircraft, specifically the rudder.
As to your suggestion of a coverup? Always possible. Don’t forget that the Airbus consortium was created to compete with the former “Big 3” of US manufacturers. I would think the company and the country (France) would do their best to prevent design problems (and they are there) from killing their ‘golden goose’.
JMO
#14.1 Garry on 2011-06-04 23:29 (Reply)
* “…., specifically the vertical stabilizer (which includes the rudder).
#14.1.1 Garry on 2011-06-04 23:32 (Reply)
OTON to Garry
Are you suggesting that the vertical stabilizer break away was the CAUSE of the accident??
this would explain a lot of things, as
– that it was found 30 miles from the other flotsam
– that it caused a multi tude of short circuits, which caused
first the emision of a lot of mafunction signasl and
immediately after the interruption of the communications
– As you read in my post, I didn’t believe that the Pitots were the initiators of the accident, but a convenient argument for the cover up.
– and of cause, the pilots would have commented this “minor” issue and therefore BEA surpressed any mention of the interchanges in the cockpit!!
– I cannot judge this: could this be the reason that the
pilots were not able to put the nose of the aircraft dowm??
PLEASE LET ME KNOW. Thanks in advance
best regards
#14.1.2 oton tisch on 2011-06-05 00:52 (Reply)
Rob,
Sorry, I don’t have enough data to answer the AOA question. However, it seems that it usually it is a performance limit (enough power to climb) that is limiting vs. an aerodynamic limit (stall/Mach buffet limit). The crew had discussed not climbing higher than FL350 due to [higher]temperatures aloft – which would, of course, reduce the max altitude.
[I have not tried to look up what the max altitude would have been for that weight and temp-so I can’t speak to their exact performance situation.]
Why did they climb?
This article may possibly answer that question. (however, according to the report they seem to have far exceeded this guidance – which most of us would not agree with in the first place.).
http://www.businessweek.com/news/2010-06-24/air-france-crash-investigator-examines-airbus-emergency-drill.html
Read the report again:
http://www.bea.aero/fr/enquetes/vol.af.447/point.enquete.af447.27mai2011.en.pdf
Still don’t see anything to support Oto’s “… against the pilit wishes, the System forced them to it” theory.
It was no doubt an extremely wild ride. (“The airplane was subject to roll oscillations that sometimes reached 40 degrees.”)It appears that whenever the crew put the nose down, things got better, and then they stopped doing that. :-( I’d like to see the g-load data for that to get an idea of how much they were getting bounced around during this time. (g-load is displayed on the lower-center display, and I assume recorded)
Questions to look for as more data is available:
Are all of the ACARS failure messages consistent with the loss of airspeed, stall, and rapid descent theory?
Specifically: was the cabin altitude fault because they “caught the cabin” or some other reason?
Were the ADIRS and flight control computer errors consistent with air data issues?
What was the specific fault reported on the ISIS airspeed related? {were there any attitude display issues?}
Why did the crew pull the engines to idle? (yes, the report says the thrust levers themselves were in idle – not just the engines) The A330 thrust levers are not servo driven, the only way for them to be in idle is for someone to put them there. Did the engines compressor stall due to the extreme AOA? and/or did advancing the thrust levers to TOGA induce a pitch-up moment that – well, didn’t help any?
(above Fl255 fuel is typically automatically moved aft [to within 2%MAC of the aft CG limit]for more efficient flight. This is nothing new or special to the 330.)
Did/does this aft CG have any effect on the recovery capability from this extremely deep {40° AOA} stall?
A 40° AOA! That’s like a flat spin – without the spinning. Is this a recoverable situation (in ANY transport category airplane)? { I don’t think they test this, just like they don’t test sustained inverted flight).
This is a question, not a theory – I am not intending to provide any fodder for Oto to go off on another rant.
When we do stall training, we practice recognition and early recovery. Not what these guys were into. (Simulators are not designed to model this part of the flight regime – I doubt there’s any flight- test data for it anyway. – not sure though – any engineering flight test pilots out there?)
I saw your post on Jonathon’s airfrance website. I was hoping you could help me answer the following question.
I am in no way any kind of expert, and have next to no understanding of planes, but I have been confused about something. I have been following the tragedy since it happened, and the last I heard before they found the wreckage this April was that the plane broke up in the air. It was said they came to that conclusion after examining the bodies recovered. So did the plane actually break up in the air? Or just when it hit the water? Would love to know what you all think.
The limited data all of us have seen seems to say the aircraft was intact when it hit the water. Nothing on the voice recorder seems to say anything about the airframe failing at all. On the surface, this will probably be pilot error. Butwhy these pilots failed to react to what was happening is a mystery right now.
How do you come to be interested in this story?
And you mentioned Jonathon’s site. Any idea how I might get hold of him?
The best question yet, “why”. There is a reason that must be found – some infer that it has, predictably by those who hold all the available data.
Until that has been independently analyzed and the findings released in an accident report, we all flounder in a sea of disconnected and incomplete information. That’s probably why since accident investigation began the watchword has always been, “wait till the accident report has been released, never rely on hearsay or pre-leaked information from manufacturers or vested interests, it’s thin ice.”
Fascinating to hear from the experienced A330 captain, mildly interesting to hear conspiracy theorists waffle. The ‘Strimmer’ crash at Basle was mired in accusation and counter claim – in the end the pilots picked up the blame and Airbus quietly moved on.
Trust the French when it comes to national prestige and acceptance of blame? Trafalgar? Non – we never ‘ear of it! LOL
Hi Rob,
Thank you so much for answering my question :)
When I first heard about the crash, I was very intrigued because at that time the plane had just vanished, and they weren’t sure what had happened to it. I am personally terrified of flying, and I like to stay on top of current airfare news to help counteract my fear. For some reason, this specific crash touched me on a level as no other had, and have been following it ever since, hoping that we would find the reason for the crash and use the knowledge to prevent any further similar crashes.
In regards to Jonathon, I unfortunately haven’t been able to get in touch with him either. I couldn’t find any contact info, and the question about the plane being broken up in the air verse hitting the ocean intact was plaguing me for quite some time.
Again thank you so much for the response, I hope we can figure out soon why it seems the pilots did not respond to what was happening.
More victims are also still entombed in that fuselage as experts try to figure out how to recover them without destroying the bodies in the process.
I took way too much logic in college and while you professionals chart a path through this selected limitted info here are 2 cents worth.
I checked the satelite photos within hours of the news and a mean band of storms that came off Africa was sitting in the Atlantic that they had to go through.
The next day the vertical tail was floating around and you could see that it came off at the seams – not torn off. Also if it went down flat at low speed the vertical would have come down with the hull, and sunk with it.
They had problems with the vertical tail falling off in New York remember. And it came off of a charter leaving Havana for Canada.
With was turned around to Cuba and landed. But no giant thunder storm involved.
My respect for you guys is great and am interested since son-in-law flys this machine.
Tom
If you honestly believe that three experienced pilots, who between them had racked up thousands of flying hours had, for 54 seconds ignored an obvious stall warning. And despite everything they had been taught continued to keep the nose in an upright position thus causing, alongside other factors the aerodynamic stall that led AF447 to fall 35,000 feet in 3 minutes. There is no way they would have voluntarily gone against everything they had learnt as training pilots. I can accept pilot error to a certain extent, but everything about the latest report screams cover up. And in all the hype Airbus have, once again snuck out the back door.
On June 4th, Rob asked:
>>I’m wondering too– about the difference at 35,000 feet and a fairly heavy weight of the angle of attack necessary to maintain cruise and the critical angle of attack … and those relative indicated airspeeds.
My guess is the spread was pretty small which might have meant very little room for pitch changes.<<
You are correct. The typical angle of attack at cruise is about 1.5degrees.
The 3rd interim report provides a section that discusses the critical angle of attack under those upper altitude conditions. The critical angle of attack margin is greatly reduced.
RE Pensar Cansa
>>Hardly, a 707 pilot of the ’60s would have been misled by faulty readings <<
If only that were true, but it's not.
Check out:
– NW 6231, 12/1/74
– Aeroperú Flight 603 October 2, 1996.
"Anon" says " There is no way they would have voluntarily gone against everything they had learnt [sic] as training pilots."
Ya know it's funny that none of the A330 pilots I talk to and fly this scenario with in the simulator have a problem believing that the two AF447 FOs somehow screwed up.
If you're going to claim that there's "NO WAY" something could have happened, let's hear why.
Would these be the same people that would claim there's "no way" a surgeon could remove the wrong kidney, leave forceps in a patient, operate on the wrong patient, or the thousand other errors of incompetence we see daily in the medical profession? Oh, if only operating rooms had voice and data recorders in them.
Do I have a hard time believing if these two got themselves into a situation, where they eventually could not get out of? In a word: "Nope." By their own words – they did not know what was happening, for a while the PF didn't even know if he was climbing or descending!
I do not doubt that there are many instances in which mistakes are made which can have fatal consequences in all professions. A surgeon taking out the wrong kidney, or leaving forceps behind after a delivery are instances which are particularly few and very far between. In almost all air disasters I have witnessed the outcome is always the same. Pilot error. Im not saying that there were not errors along the way, but the BEA and the press have focused far too much on what the pilots did or didn’t do in the last 3 minutes of flight 447. In my opinion for the average person who has not taken the time to look into the incident in more detail is going to simply fed the same story again and again. The errors of the 330s automated computers and conflicting evidence have been somewhat overlooked. So much so that the French pilots union are no longer taking part in the investigation.
Anon,
Since you have “taken the time to look into the incident in more detail,” perhaps you would like to elaborate exactly what “the errors of the 330’s automated computers and conflicting evidence” are.
You would not put this accident in the “few and far between” category?
I amd still wanting to hear why “There is no way they would have voluntarily gone against everything they had learnt as training pilots.” Could it be to confusion and forgetting what the sidestick commands (hint: it’s not elevator position)?
I do not appreciate you trying to patronize me. If it wasn’t for people like me who independently take the time to look into these incidents in further detail your article would have simply disappeared into the abyss that is the world wide web.
In terms of errors the A330 can account for, it has been widely speculated the the plane was giving incoherent speed readings, and an Airbus spokesmen was reported to have said ‘there is a programme of replacement, of improvement.’ Alongside this the reports published so far have stated that in AF447s final minutes the most junior of the pilots was in control of the plane. And that the second co-pilot had only taken over the controls for a short amount of time before the most junior co-pilot took back the controls.
Before this though the pilots were being told that they were exceeding normal cruise speed. They had manually tried to slow the plane down by pulling back on the throttles, and perhaps to try and aid the slowing down process had pulled the nose up. It was at this point the stall warning first sounded warning the pilots that they were flying too slow. The plane slowed even more, and the stall warning stopped. The pilots then lowered the nose slightly and the stall warning sounded again, which is the complete opposite to what the pilots were expecting to happen in recovery.
I question how you can criticize my comment that there is ‘no way’ the pilots went against what they had been taught voluntarily yet in your article you have stated that it is ‘ALWAYS’ the pilots fault. In terms of this being in the ‘far, and few between category’. I would say that air crashes are something that are few, and far between, but the reason why they crashed in the first place is nearly always primarily ‘pilot error’.
I do not doubt that in only a short 4 minutes panic would have taken over the pilots, but we are led to believe that in this circumstance the junior pilot in particular made some obvious mistakes which the other two pilots made little effort to correct.
Ok, let’s take a deep breath Anon. I didn’t see Bill’s comments as patronizing myself, but I know him so perhaps we both missed something. It is never our intention since we know that shuts down the conversation. I actually tried to send you a note back to the Hotmail e-mail you registered here with, but it bounced back.
But I can’t disagree with Bill’s point that there is no evidence to suggest this was not pilot error at this point, at least directly. The pilots had no idea what was going on for the final four minutes of the flight.
Now is there anything Airbus could do at this point to prevent this from occurring once again? Sure. Add an Angle-of-Attack indicator in the cockpit. It doesn’t lie and could have been used by the crew for the truth at a critical moment.
Let’s not forget an important element here. The blame – if we choose to call it that – will not be placed only on the crew, no matter what … or at least I hope it won’t.
There was obviously a cockpit resource management issue here too since it was only in the last few moments that the junior pilot relinquished the controls. The captain seems to have been a bystander through most of this.
So pilots yes, Airbus yes, Air France yes. Conspiracy to cover up something critically unsafe about the A330? Possible, but I don’t see anything pointing to that right now which is the point Bill was trying to make, I think.
Anon,
Certainly no offense was intended.
However, to allege that there is an error without providing how you came to that conclusion is like me saying you’re ugly – without seeing your picture. Either might be true, but we have not stated the basis for the allegation, and therefore it lacks credibility.
So, you mentioned:”The errors of the 330s automated computers and conflicting evidence.”
Again, what errors of the automated computers, what conflicting evidence? Give us some specifics of what you are talking about (other than the only source you have referenced: “wide speculation”) of how you came to this conclusion?
>Before this though the pilots were being told that they were exceeding normal cruise speed. <>They had manually tried to slow the plane down by pulling back on the throttles, and perhaps to try and aid the slowing down process had pulled the nose up. It was at this point the stall warning first sounded warning the pilots that they were flying too slow. The plane slowed even more, and the stall warning stopped. The pilots then lowered the nose slightly and the stall warning sounded again, which is the complete opposite to what the pilots were expecting to happen in recovery.<<
That is a perfectly reasonable theory, right in the path of – NW 6231 on 12/1/74. But I'm still looking for the high indicated speed you speak of at the start of the scenario.
However, from the transcript it appears that they get the first stall warning right after pulling the nose up and before leaving FL350.
Also, according to the FDR, they also did not move the trust levers until they were well in the climb. I don't see how your explanation matches with the flight data.
{see Thrust resolver angle parameters on Interim report page 108}