Slashdot Mirror
Flight 447 'Black Box' Decoded
fermion writes "An initial report has been released by the BEA concerning the details of the last minutes of Flight 447 en route from Rio de Janeiro to Paris. According the report, the autopilot disengaged and stall warning engaged at 2 hours 10 minutes and 5 seconds into the flight. Less than 2 minutes later the recorded speeds became invalid. At 2 hours 14 minutes and 28 seconds, the recording stopped. The final vertical speed was recorded around 10,912 ft/min."
“So, we think that until impact they did not realize the situation, which for the family is what they want to hear — they did not suffer.”
A three minute decent at 10,000 ft/min over the middle of the ocean?
I'm pretty sure everyone onboard knew exactly how that was going to end about half way in.
= 124 miles/hour
What seems to be remarkable is that the trigger to the catastrophe has indeed been revealed to be the pitot tubes - something that was suspected very soon after the flight went down. To a layman like me, it is amazing that without the benefit of all the data that has been recovered from the flight data recorders, experts were able to get so close to the mark.
Now, one could flip this around and also say that given that so many observers were able to so accurately get to the initial trigger for the failure in the absence of hard data, it must mean that this was a really common failure mechanism that should occurred in the field only as a result of the problem being repeatedly ignored.
It is a triumph of technology that the flight data recorder survived under such extreme conditions for so long. It was a triumph of technology, that it was located and retrieved from such an extreme location. Surely, a species with such (magical?) technical expertise could have expended the effort into preventing such a failure?
There is no such thing as luck. Luck is nothing but an absence of bad luck.
Now the question is, how many air crashes have been due to pilot error versus those due to software error. I suspect software has the better record.
The pilots have to keep the plane within a pretty strict speed range to both keep the plane together and avoid a stall. Even when the speed sensors fail and the pilots have no airspeed indicator, there is a standard procedure that allows them to keep the plane within that narrow range by setting the throttle and controls at specific settings until it unfreezes. Frontline even aired a special where two flight instructors demonstrated this after being presented with this exact scenario in a simulator. The problem happens because the pilots don't have a lot of time to react, and if they panic or misinterpret the situation it's all over.
I trust software more and more in the long run. And I m not a geek!!
According to some media-released transcrips I have read, the plane was stalled for over three minutes and yet the pilots consistently kept the nose of the plane at an upwards angle. Piloting 101 states that if your plane is stalled, the proper maneuvar is to point the nose downwards and dive sharply to pick up enough airspeed so that you can swoop and obtain lift so that you are no longer stalled. Apparently the pilot's actions deviated from this almost universal practice and further doomed the situation.
I am not nor have I ever been the pilot of an aircraft, however I fly remote control planes and I've had to deal with stalls a time or two using such a tactic. If the pilots of 447 had executed such a practice, odds are the stall would have broken. Perhaps that accounts for the 'human error' portion of the blame, but it was significant. I realize that airspeed and altimeter tools are invaluable to flight, but with the loss of those a firm knowledge of aviatic physics can mean the difference between life and death. As it was here.
Of course, in this case, even the computer didn't have the data it would have needed to fly the plane - if the pitot tubes were blocked and not giving a speed reading, the pilots may have attempted to outguess the situation. They may have looked at thrust settings and maybe even GPS speed reports and concluded that there was no stall despite the warnings since they knew the airspeed was far above what was reported.
I trust that an airline pilot has enough training to know how to handle a stall, but if he knows his instruments are lying to him, then he may choose to ignore them.
Do stall warnings use anything other than airspeed and angle of attack to warn about a stall, or are there some type of sensors on the wings to detect airflow and lift?
If it were me, and I only had 3 minutes left, I'd grab a pen from my pocket and carve into my shoulder "4-8-15-16-23-42" to send a message to my family that 'even if the ending sucked, at least I will find happiness in purgatory.'
I'm sure the sharks that ate your body would have been impressed by your ingenuity and would have passed the message on to your family.
1. Unless it's pressurization system was faulty (it wasn't) the pressure change wouldn't have been great.
2. Unless accelerating, you wouldn't know you were going down (or up, or banked or upside down...).
So the claim that the passengers probably didn't think it was anything more than turbulence is not hard to believe.
This pre-supposes that the passengers felt approximately 1G of gravity downward from their point of view, which pilots normally carefully maintain for passenger comfort. A banked turn that maintains 1G of gravity downward by carefully controlling the turn rate feels "normal" as if the plane were flying straight. Pilots can't do this in an emergency when there are systems failures.
Remove that 1G of gravity like in a free-fall, or flying straight down, and the passengers are going to become acutely alarmed very quickly. Normal turbulence of sufficient magnitude can do this -- and the conditions they were in were worse.
So put your thinking cap on for a second, and consider how it might be possible for a plane flying directly downward to somehow create 1G of gravity sideways such that from the point of view of the passengers, they feel 1G "downward". How is that possible? The only way that could be done is through centrifical force, such as if plane were pulling up, and if it's flying strait down, it's not pulling up. At best, the passengers felt gravity at their backs from the acceleration downwards. As soon as they reached terminal velocity downward, the gravity they felt would be 1G to their front. I'm quite sure that as a passenger I'd notice all of that.
I think this was three minutes of terror for everybody on that plane.
The training has proven to be entirely inadequate. This type of accident is far too common, since the earliest days. One guy should always be flying the plane, but all too often everybody's trying to troubleshoot the problem. and even in a storm it is possible to maintain control with a working artificial horizon and a fixed power setting as you point out.. Key word is 'situational awareness'. Lose that, then indeed, all bets are off. A lot more hours in the simulator are needed to burn this into the guy's head.
For justice, we must go to Don Corleone
I have been reading the report and there are some strange interesting passages. Here is a partial summary, focussing largely on pilot control inputs
Copilot is PF. Captain is PNF.
2 h 08 min 07: "...turbulence increased slightly and the crew decided to reduce the speed to about Mach 0.8"
2 h 10 min 05: "...the PF made a left nose-up input. The stall warning sounded twice in a row...Autopilot and auto-thrust remained disengaged for the rest of the flight."
2 h 10 min 16: "...The airplane’s angle of attack increased progressively beyond 10 degrees and the plane started to climb. The PF made nose-down control inputs and alternately left and right roll inputs... The airplane was then at an altitude of about 37,500 ft"
At 2 h 10 min 51: "...The thrust levers were positioned in the TO/GA detent and the PF maintained nose-up inputs...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...The PF continued to make nose-up inputs."
2 h 11 min 40: "...The airplane’s pitch attitude did not exceed 15 degrees and the engines’ N1’s were close to 100%..."
At 2 h 12 min 02: "...At that moment, the thrust levers were in the IDLE detent and the engines’ N1’s were at 55%. Around fifteen seconds later, the PF made pitch-down inputs..."
This much seems clear: the airplane was cleared for flight level 350 (35000 ft), and was likely at that altitude when the trouble started, assuming the altimeter was functioning. Stall warnings went off, and the airplane climbed to 38000 ft. It then descended rapidly, it seems with a monstrously high angle of attack. It also seems from the report that the nose of the plane was mostly pitched up through this, though I am not absolutely sure on this. This would imply a very bad stall...essentially the airplane was falling from the sky. One can speculate that the pilots were doing their best to recover from the stall with imperfect data on their airspeed.
To me the important period was between 2 h 08 min 07 and 2 h 10 min16. There was a decision to reduce speed, which would entail a reduction in thrust. Two minutes later, there was a stall warning, implying that the airplane's airspeed was out of the very narrow range required at that altitude (plus/minus 10 knots according to the Nova documentary, though I'm not sure it's so narrow). The question is, what caused those initial stall warnings? How did the airplane's speed get out of the proper range? Did the pilots forget to increase thrust after the autopilot reduced it? During those two minutes, was the airplane catastrophically slowing down?
This and no other is the root from which a tyrant springs; when first he appears as a protector - Plato (423 to 327 BC)
1. Unless it's pressurization system was faulty (it wasn't) the pressure change wouldn't have been great.
2. Unless accelerating, you wouldn't know you were going down (or up, or banked or upside down...).
Unfortunately, you're right -- because they were stalled but yet relatively level, the passengers felt 1G downward as if they were in normal flight. So they had a brief rough ride during the initial stall, but then after they seemed to "recover" and reached terminal velocity, they felt 1G and as if everything was normal.
What's most interesting in this case is that the systems warned the pilots of an impending stall, but then once they were in a stall, there was no warning at all, as if they had recovered from the stall. That's really unfortunate.
Airbus is run by software. Boeing by pilots.
This is a delusion. Read up on Turkish Airlines Flight 1951. I mean, they programmed 737 to do what? Trust a single radio altimeter known to fail in the worst possible mode, producing false readings? Actually use these readings to automatically enter the landing flare mode with rather subtle indication to the pilots? Autothrottle reverting pilot's inputs? Sounds like scary automation to me.
The reality is, all modern commercial airliners are run by software. If you want an airliner run by pilots, go to Russia and fly some Tupolevs, there are still quite a few left (ehhehe).
My exception safety is -fno-exceptions.
Do stall warnings use anything other than airspeed and angle of attack to warn about a stall, or are there some type of sensors on the wings to detect airflow and lift?
IANA-aeronautical-engineer, but I imagine that you cannot easily determine the actual angle of attack w/o sensors to detect airflow. The actual angle of the wing referenced to the ground (or other fixed reference) doesn't determine the angle of attack, it is relative to the windspeed around the winds. Thus the angle of attack is mostly inferred by sensors that measure windspeeds (mostly). AFAIK stall warnings often additionally use measurement devices that measure dynamic differntial pressure around (or nearby) the lift surfaces (aka LRI), and other secondary indicators (e.g., buffeting dynamics of the plane when it nears the stall conditions).
I disagree with that synopsis, my understanding is the pilot started fighting the autopilot without disengaging it. The plane might have been fine if it had totally disregarded the pilot.
My understanding of what happened here is the autopilot was automatically disengaged (because it didn't believe the airspeed reading it received) and handed control over to a human, who proceeded to stall and crash the plane. It's too bad the autopilot wasn't programmed simply to do the best it could instead of handing over to the pilot (in the errant assumption a person would do a better job coping with the situation).
I trust software more and more in the long run. And I m not a geek!!
That's why.
It's randomly on a car forum, but its worth a read. Some guys that know what they're talking about talk about what they think happened. They also include pics of various airbus cockpits for reference.
http://www.mye28.com/viewtopic.php?t=64381&postdays=0&postorder=asc&start=25
Here's the basic story, as I understand it:
- the pilots flew into a thunderstorm
- they were 100% blind, relying entirely on the glass-screen instruments
- once all 3 pitots froze, the redundant computers started disagreeing and then finally agreed that things were ugly
the effect in the cockpit is that a serious of cascading failures were unfolding, likely overwhelming the pilots.
additionally, there would be NO functional indicators for alt, speed, horizon, etc. Once the computers have faulted, they no longer share that information.
Also, as the computers degrade authority, in an Airbus the pilots get MORE control of the aircraft. This means that controls move through larger ranges.
As flight control reverts to failsafe mode, the controls in the cockpit do not "auto-zero". And the forcefeedback goes off line.
Effectively, the pilots are 100% blind, and the inputs they make have no feedback whatsoever. They cannot even tell if they have _stopped trying_ to turn.
Imagine being blindfolded. Your job is to put the end of a 4 ft long stick inside of a 1" circular hole in the floor. Except the stick is a peice of yarn.
That's what their instruments and control apparatus were like.
Now imagine that everything is beeping at you and you are in a plane in a thunderstorm, over the ocean, at night, and everything outside is total blackness.
You're fucked. Thoroughly and completely fucked.
My opinions are my own, and do not necessarily represent those of my employer.
The actual BEA report, which should be read before commenting, does not assign blame. That will come later.
At one point, the left side airspeed display showed 215 knots, far above stall speed. The backup airspeed indicator showed 185 knots, also above stalling speed. The right side airspeed display value isn't logged. Then all speeds showed as invalid. Given that conflicting information, at night in a thunderstorm over water with no outside visual cues, it's not totally unreasonable that the pilots, finding themselves losing altitude but thinking they had more airspeed than they did, tried to pull up.
It seems very scary that on an aircraft with everything working but the airspeed indicators (and I understand that those are very important), after more than 3 1/2 minutes the aircrew was unable to prevent the plane from hitting the ocean. This was a state of the art aircraft. Makes you wonder how many close calls there have been that luckily didn't result in catastrophe.
I find it highly unlikely that clogged pitot tubes would take out all the static and gyroscopic instrumentation too. That would be a phenomenally stupid design, and would rank the Airbus lower than a Cessna 150 in terms of instrument reliability.
The information in the report is preliminary, and there has to be more to this story. Even the lowliest weekend warrior pilot earning the most basic instrument certification has to demonstrate an ability to fly in IMC with multiple instrument failures.
What I want to know is why the PF pulled the stick back to the stop with the stall warning blaring in his ear. There's no way an airline pilot would make such a stupid mistake—unless there's way more to this story than what we're reading in the preliminary report.
Correct, there were no instrument displays telling them they were level. The computers that decide that information disagreed with each other and gave up.
If you aren't a jet pilot, I would probably stop suggesting how it should have worked. Your suggestion to apply 5% more throttle could have destroyed the aircraft.
One thing I learned from this thread is the "Coffin Corner". Flights are operated very near the intersection of their absolute altitude and their absolute vmax.
This puts them in bad shape -- if they go faster, they exceed the design speed of the aircraft. This can cause things like mach tuck and cause the control surfaces to stop responding correctly. Nobody wants a plane that becomes aerodynamically unstable AND doesn't respond to controls.
If the overspeed is excessive of course the aircraft fails structurally.
If they go too slow, obviously they stall.
For a loaded Airbus operating at say FL350 and mach .8, there may only be a 30 knot range of permissible airspeeds. The auto throttles ("cruise control") will attempt to maintain a set speed within that range. THe upper and lower bounds are constantly changing based on atmospheric and flight parameters..
Note that its not as easy as hitting the gas or the brakes -- aircraft attitude greatly impacts speed, as does its altitude.
The planes fly high because it saves gas, its quiet, and its fast. But there isn't a big margin for error up there.
So without having ANY idea how fast you're going, where you're pointed, or what your altitude is, its grossly irresponsible to just apply more throttle.
The only safe course of action is to try to get to lower atltitude somehow without overspeeding the aircraft. But again, how do you do that if you don't know how high you are, where you're pointing, or how fast you're going?
It's basically bullshit to blame the pilots for what happened here.
My opinions are my own, and do not necessarily represent those of my employer.
WRONG WRONG WRONG FUCKING WRONG!!!
All Airbus cockpits have "old fashioned" analog devices, in particular the artificial horizon. Also, there is no force feedback on the sticks.
Please don't spout crap about an aircraft that you know nothing about.
If you lose the computers the aircraft flies like any other normal aircraft. You are not fucked in anyway.
This is more like a complete loss of situational awareness by the pilots - actually a very common problem that is independent of the aircraft type, manufacturer and whether is the "computer" controlled (whatever that fuck that means). Similar things happened to the Turkish 737 crew who crashed at Amsterdam a few years back, the Cali 757 accident, the Strasbourg A320 crash etc etc
ps: I fly an E170/190 for an airline that also happens to have A320s ....
If Air France's A330-300 are set up the same as SAS' (430KB PDF), there should have been a cluster of mechanical backups just to the right of the pilot's primary glass displays, including an artificial horizon. Even if the backup sink rate, airspeed indicator, and altimeter were returning bogus values, the gyroscope and compass wouldn't, and ergo there would have been at least enough information to know which way the bird was pointing.
Luke, help me take this mask off
Alt. must have worked it is recorded correctly by the black box, I see no reason to believe horizon shouldn't have been working, so only visual and speed was impaired. That is still a difficult situation, so no reason to exaggerate it.
the pilots flew into a thunderstorm
What we can assume is that they expected only a slight increase in turbulence. The pilot in control of flying the plane told the cabin crew: "in two minutes we should enter an area where it’ll move about a bit more than at the moment, you should watch out". They also made a left turn shortly after that in hopes of avoiding even that, and then the turbulence "increased slightly". As for visibility, they probably had no visual references prior to entering the storm. For instance, as far as I can tell with Celestia the moon was not visible from their location at 0200 UTC on 20090601.
once all 3 pitots froze
We only know for sure that 2 of the indicated airspeeds were incorrect at some points, probably due to icing in the pitot tubes. The third was not recorded.
the redundant computers started disagreeing and then finally agreed that things were ugly
the effect in the cockpit is that a serious of cascading failures were unfolding
Yes, the speed inconsistency lead to the following among other things:
- alternate law in effect (see below),
- autopilot disconnecting, and
- autothrust disconnecting (although I understand the engine thrust would have stayed the same at this point until a pilot altered it).
likely overwhelming the pilots
This is pure speculation at this point.
additionally, there would be NO functional indicators for alt, speed, horizon, etc.
Wrong. There's no evidence to support malfunctioning fundamental indicators other than loss of airspeed indication, and perhaps the somewhat inconsistent stall warning.
Once the computers have faulted, they no longer share that information
See previous point.
Also, as the computers degrade authority, in an Airbus the pilots get MORE control of the aircraft.
This is a fair statement. They degraded into alternate law.
This means that controls move through larger ranges.
More or less so (stalling etc. possible), but the computers still have an intermediary role so for instance they could not go beyond max G loads in this particular mode.
As flight control reverts to failsafe mode, the controls in the cockpit do not "auto-zero".
Are you talking about auto trim? Auto trim was still available, as shown in the report. (If the aircraft had degraded into direct law, it would have been disabled.)
And the forcefeedback goes off line.
There's no force feedback that I'm aware of, other than a stick shaker during indicated stall.
Effectively, the pilots are 100% blind, and the inputs they make have no feedback whatsoever.
The report clearly states that the engines responded, and the report seems to be consistent with the aircraft responding to all "mainly nose-up" inputs by one or both pilots. What was displayed to the pilots was inconsistent for a time, yes, but again many of the instruments may never have failed at all.
They cannot even tell if they have _stopped trying_ to turn.
What do you base that on? Again other than the stall warning and airspeed indications there is every reason to believe that many if not all other instruments were working, such as:
- Information from GPS etc. (such as g
My information suggests that the sensors feeding the black box either cannot or were not made available to the primary flight instrument computers, astonishingly enough.
Given the amount of system isolation in aircraft this isn't entirely impossible to beleive. If you tie part B into part A's wiring in any way, you have to do more testing/verification than you would otherwise.
My opinions are my own, and do not necessarily represent those of my employer.
No, because in most jets extra thrust will actually push the nose up since the engines are located under the wing. The most recent procedures actually have you reduce thrust to bring the nose down more quickly.