Airbus Flight 447 Crash

Sort of along those lines:

Did the gain schedule for the controls change as a result of incorrect inputs.

Maybe.

In my opinion most likely course is the air speed tube icing up thus making the plane speed up to maintain what the computer things to be the correct air speed.

To make maters worse the plane has entered stormy weather, now the plane is most likely buffeting with turbulence, but the captain thinks this is fine as the plane can handle this type of turbulence.

But of course he is completely unaware that the plane has sped up to most likely its stress limits whilst of course the pilots are still think every thing is ok, then suddenly the plane in counters an up draft giving server turbulence on the already over stressed airbus.

The combined stresses will now start making the air craft unstable and the pilots
relieses some things very wrong, so now he disengages the auto pilot, but its to late the craft has gone pasts the point of no return and brakes up and the cabin depressurises coursing every one to pass out from lack of oxygen including the pilots hence no may day was sent.

But of cores this is just one of about a million possible scenarios but I think at this moment in time with the current evidence this in my opinion fits the best probable scenario, well more likely than a bomb

Pretty good explanation. GA to you. I would make one change.

The rudder has a software stop that prevents it from traveling too far at high speed. Full rudder at normal cruising speed would overstress the vertical stabilizer, possibly causing it to fail. When the pitot tube reported low speed and the plane entered the turbulence, the autopilot may have ordered full rudder causing it to break off.

This would explain finding the vertical stabilizer but no other major parts of the plane.

Well put, its my thinking too.....

]There is historical data showing vertical fin failures in Airbus A-300's.. Lots of prior evidence, according to data from aircraft scrappers, and maintenance people. The problems are in the composites that are engineered and used. The company is using, as a "scapegoat" (according to sources left unidentified), weather as the main factor, when there could be some major structural problems with the aircraft. The plane broke up, mid air, as bodies were found intact, with clothes gone. They are always in pieces, in the fuselage , upon impact with ground or sea. Just data that I am getting from former aviation inspectors , military and former FAA personnel.. The theme is....If it isn't Boeing, it isn't going" -- A bit of pride in that statement I suspect--Just my 2 Cents C-Mac

Besides the aforementioned control laws allowing larger control deflections at slower speeds, there is also the upper Mach limit for the airplane. Once that limit is exceeded, the aircraft is subject to a high-speed upset. This was once known as "Mach tuck" referring to the strong nose-down moment that results from the center of pressure moving aft on the wing. At night, in turbulence, with erroneous instrument readings, at overspeed, likely hitting flutter speed.....makes for a very bad experience.

If a pitot tube sensor is a single point failure, why was it not discovered during Fault Tree Analysis (FTA) or Failure Mode and Effect Criticality Analysis (FMECA) during the design phase. If they did these analysis maybe there would have been a heated pitot tube, three pitot tubes and a voting circuit that would indicate an alarm to the pilot that something is wrong, or three heated pitot tubes and a voting circuit.

Based on what I have read, Airbus knew of the problem with pitot tube freeze up beforehand, if they didn't perform FTA or FMECA during the design phase why didn't they perform these analysis now to discover that it was single point failure and ground the fleet.

If Airbus truly knew of the problem with the frozen pitot and did not do these analysis to determine it was a single point failure and subsequently ground the fleet, in my mind they should be liable for negligence.

It would appear that Aibus knew all about the failures and warned all airlines to replace them in September 2008.

Air France simply did not bother to replace them immediately, until this crash occurred.....now they have all been replaced.....was it so difficult?

As I wrote in a previous post, this was similar to the Concorde crash, BA replaced and update parts to do with prevention of getting water in the engines while taxying and in case debris was shot up at the wings (as happened in the USA once before, but the fuel did not catch fire!).

Air France did not bother.....you know what happened after that!!!

There is no way to certify an aircraft without such analysis, and single point failure would never be allowed- under anticipated failure modes.

The general rule is systems analysis goes two failures deep for Part 23(general aviation), three for Part 25(air transport).

I am NOT an authority on these analysis being a receiver of the output, but if we are talking about multiple failures in multiple systems, we could have hit the limits of analysis.

Not necessarily in defense of Airbus, but if I build an aircraft that CANNOT crash - it is because it cannot fly.

Hello Edignan,

I am not an expert on Reliability, but I would like to reply to "systems analysis goes two failures deep for Part 23(general aviation), three for Part 25(air transport)" is that it is too broad brush.

According to NASA Office of Safety and Mission Assurance; Fault Tree Handbook with Aerospace Applications, Version 1.1; NASA Headquarters; Washington, DC; August, 2002: p110

"When the design fault tree is quantified the importances and sensitivities of the different parts of the design are obtained. This is useful information and shows what parts of the system drive the failure probability and reliability. The designer can then focus on the important and sensitive parts. One of the greatest benefits resulting from carrying out any FTA is the establishment of design priorities for all elements of the fault tree and thereby for the design effort. Often, only a few of the elements, or contributors, will drive the failure probability and reliability. The FTAs that have been performed in the past generally show that less than 20% of the contributors dominate the failure probability and the reliability. Often, in fact, 90% or more of the result is driven by as little as 10% or less of the contributors."

Maybe the FAA/JAA should reconsider their approach to system reliability analysis and then maybe we will not "hit the limit of analysis" and safe lives and maybe even save engineering labor cost for the aircraft manufacturer.

The aircraft in question would have had at least 3 and possibly 4 separate airspeed systems. Pilot, co-pilot and standby. Each is a fully-separate system, connected only by the atmosphere that they measure (We still haven't figured a means to make that dual-redundant). In the event of extreme event such as entering a thunderstorm, it is possible to accumulate ice or encounter rain at such a rate that the systems are all overwhelmed simultaneously.

As to why the systems were not replaced earlier, there are a number of explanations there as well. Replacing the systems of the entire fleet is not an overnight thing. Producing those systems for the entire A330 fleet is not exactly a quick thing, especially when the production is already maxed out. Air France is not the only operator and all the other airlines would be pushing for parts as well. Installing new systems takes away from other maintenance - provided by crews already on the lean side.

The point is, if it was easy, it would have been done.

ladies and gentlemen

if i may:

let's assume that all ) 3 ) pitot tubes and the complete pitot static system failed.

that in itself would not cause the Acars MSG.s that were generated.

all of the msg , over 3 minutes worth, were roll control , except ( 1 ) about the rudder limiter failure. so, with that info , it's possible to see that the autopilot was already tripped off. acars generated no msg about cadc ,{ or whatever Airbus calls it, dfgc , or something,} failing. there is no way that the dfgc , or cadc , can cause a catastrophic de-compressive failure .

even if the vertical stabilizer has sheared off by this time, that wouldn't cause electrical failure, only a few systems in the vertical fin. logo lights, and artificial feel, rudder pitot system.

by design most airliner wings are built to a max of mach.78 ..lets assume the plane is being pushed past that. pieces are gonna start flying off. big pieces.

just like the crash in NY , imho: the vertical stabilizer mount attach casting isn't strong enough. Now, with the vert fin sheared off, in massive turbulence, no yaw control, very difficult to predict flight path after tail separation, but it wouldn't be straight and level. maybe why more debris hasn't been found. ( that aircraft didn't disintegrate in the air )

now lets assume that the turbulence is very severe, winds in excess of 50 knots, as long as hyd. power is available, this isn't a problem . the winds can not override the flight controls.

on the engine : RR Rb211 ..within the fadac system is an engine speed limiter.

lets assume that the plane is on autopilot: we've selected 35 k feet and mach . 65.

as the planes sensed airspeed is evaluated the fadac box configures the engine thrust to keep the plane at mach.65. at the same time , fadac will not allow the engine to overspeed.

So, we've told the autopilot what and how fast. as long as it's engaged , that's what i'm gonna get. now with UPDRAFTS and DOWNDRAFTS, BUFFETTING WINDS FROM THE THUNDERSTORM. acars didn't send msg of wind shear system failure, or pitot static system failure. ( how do you think Airbus knew of the failures )..??visual inspection only?..

imho: the conditions that the aircrew " saw " were what was really going on around them. the pitot system didn't fail.

imho: flight departed under a wea advisory, known thunderstorms , tracking this heading moving this fast ... the flight management computer was input a flight plan , that " flight and dispatch" had agreed upon.( that's how we know how much fuel to put on board) the conditions after 4 hours of flying were different than expected. the aircrew attempts to navigate the thunderbummpers..not the 1st time that'z happened, heck i can ask the fmc ( flight mgmt computer ) to plot a course around them, over them , or through them. the wea radar pictorial depiction is very nice, " the glass " cockpit is visually concise. the conditions change very rapidly now, plane gets caught in the updraft/downdraft buffeting..

don't quite see how the fire gets started, or the rapid catastrophic decompression, but it's possible the aircrew deployed the " yellow jungle " the o2 masks. may have been automatic as well. i do know that the aircraft has " negative pressure relief valve" installed, should the psi INSIDE " THE PLANE BECOME TO GREAT , WE EQUALLIZE THAT PSI BY OPENING A " ONE WAY " VALVE. so blowing off of a hatch or cargo door , or entry door .. not to likely..unless there was tremendous psi in the hold. not uncommon to find these valves popped on preflight workaround's.

i don't have any more guesses. and on the Airbus thread , i discussed some possibilities.

We are all guessing.

We are all finding our possible reason.

Most of us are NOT trying to convince others that we are right....simply becaze none of us know who is right!!

No one is mentioning that every transoceanic-rated aircraft has some sort of INS - inertial navigation system. Nowadays, many of these are updated by a GPS input. Regardless, there are multiple sources for air speed, including the INS, so that losing pitot tubes or erroneous data from same would not in and of itself cause a fatal problem.

Somewhere between Airbus internally and the CAA there is no way the pitot tubes could cause a single point catastrophic failure, just as edignan says, not painting with too broad a brush.

emc_c

GPS is a (over) land speed indication only, an aircraft must have accurate "AIRSPEED" indication to be able to fly correctly, that is the reasons for the Pitot tubes..........

If what you thought was true, the Pitot tubes would be long gone.....they are needed urgently ALL the time.......

I agree GPS would not be taken as a speed source.

I found a link today which is some interesting reading, pointing to the computer system having problems.

http://www.timesonline.co.uk/tol/news/world/us_and_americas/article6523467.ece

nice link thankz

I do not understand some of the statements that were made in that report

"Experts suspect a flaw in the behaviour of the three independent air data inertial reference units which collect raw flight parameters such as speed and altitude." How can three independent systems have a simultanteous flaw in their behaviour? Unless it was from direct or indirect effects of lightning. Which as I remember was abruptly ruled out.

"One such faulty unit was blamed for a near disaster on a Qantas Airbus A330 over Western Australia last October. Confused data caused the flight control computers to register — mistakenly — an imminent stall and to disconnect the automatic pilot." The reason for having three independent systems is to prevent confusion, how can one independent system even if it malfunction cause confusion. Maybe there is something wrong in the Voting circuitry, logic, or programming.

Also I wonder what initiatives Airbus has taken based on the near disaster approximately nine months ago and what corrective measures were implemented.

I personally feel that you may be "nearer the mark" than even you may believe.....I suspect that we have a "logic" problem in the computer system that only makes itself evident, when strange things happen to its inputs, say from Pitot tubes that can simultaneously and together freeze up.......

.....but of course Airbus probably knew this, and this is why they told the airlines to replace them in September 08, so why did Air France only replace them very recently - since the crash in fact????

I suspect that Airbus realized that the software could get confused if the tubes all packed up at either the same time or almost the same time......it fits in with my personal take on the accident, at least until hard evidence can be produced from the flight recorders......or there is another accident on/over land!!!

Its almost like the Comet disasters of the '50`s......

TOTAL BUNK

The only way your statement can be correct is if you know exactly what happened.....

I assume you don't know anything more than the rest of us, so you were actually also commenting on your own knowledge of the subject!

I don't know what has happened either, but I have worked a lot with computers and programmers on a large variety of problems over 35 years, I would not be surprised to (hopefully) to eventually to find out that our guesses about the tubes/computers is not so very far from the truth....

But there is no need to "lose your rag" about it either way, very bad manners!!!

" i assume you don't know "..

well my experience outweighs your doubts, assumptions and your knowledge.

We will probabvly all live long enough to see your "Comeuppence"!!! again!!

guess that leaves me off your Christmas card list , huh , buddy ?

Actually, you were never on it.

I try to maintain a certain minimum level of quality.... or better of course!

But I am sure that you won't be upset by that!!

Have a great day...... I never hold grudges, life is too short for that!!!

I think that my last post shows the attitude to maintenance that Air France has deonstrated in the past quite clearly, its an Air Line that I have only used once before (and never agin!).

I have also made a point for many years in NOT flying on DC10's either.....or any othe MacDonell Douglas aircraft, or Lockheeds L-1011...Too many spectacular accidents for me on all of these.

The Airbus 300 is starting to go in that direction for me too.....at least ones from Air France as I have yet to see clear evidence that the makers of the Airbus have been paricularly negligent - yet......

I hope that the reasons for the crash can be 100% identified in the not too distant future for everyone to read and understand.

Lately there has been much attention payed to pitot tubes and the way the information they provide is downloaded to the various sophisticated flight management systems on a modern airliner.

Pitot tubes have been around for many, many years but the info. they provide is now used in many different ways .

Both Captain and Co/Pilot have there own seperate pitot tubes and readouts.

In the event of a discrepancy , the Captains readout is supposed to prevail if all checks are ok. The problem is that the autopilot recieves its info. from the Captains pitot tubes as well and if a problem occurs with them the autopilot simply flies the plane according to the info. recieved . The Co/Pilot would inform the Captain if he observed a discrepency .

Two problems , " and plane crashes " have now shown themselves ,

1) It has been observed that some airlines from countries that have highly entrenched social hiarchies have a cockpit hiarchie that is the same . In an emergency situation a Co/Pilot will defer to even a completely confused Captain .

2) A plane flying under Autopilot control , even with confused inputs from various sensors such as airspeed etc. will stay in the air as the flight computers compensate. Many millions of dollars of software have tried to ensure this . When a pilot shuts off the Autopilot is were the problems begin . He is now flying a plane with no idea of what is actually happening .

Concerning the vertical stabilizer or rudder , the FAA has requested that all pilots be retrained in the use of rudder pedal actuation in emergency situations. Due to the faster response and quicker movement of the vertical stabilizer in modern airliners the older training of stamp on one side and stamp on the other at speed is now being trained as required to prevent the stabilizer from being ripped off the aircraft.

Although flying is the best and safest way to get there , i still cant get over that darn Cost/Analysist stuff where they figure out how many plane crashes equal how much money they will invest in safety features .?

I mentioned in another CR4 thread on this topic, interesting information on the AF447 is available on the Professional Pilots forum, http://www.pprune.org/rumours-news/376433-af447.html Even then you must bear in mind that there is a lot of conjecture and views surrounding the possible reasons and many of the comments are guesswork, even from these people. But if you are looking for sources of relevant info from the guys who know and fly the business it is worthwhile.

Another interesting one with particular detail on the weather-related conditions is http://www.weathergraphics.com/tim/af447/comments.shtml

If you wondered what AF447 looked like the day before the terrible events, it's last landing in France on the day prior, http://www.airliners.net/photo/Air-France/Airbus-A330-203/1539421/L/

I had been following the AF447 incident from the beginning, then the next day a work colleague told us about his son, who works in Brussels. His son has an office workmate who flew Brussels to Rio on business and on arrival at Rio his visa arrangements weren't correct. He was ordered to return on the next flight back - guess which one. There is one poor soul missing from the Brussels office.

Thanks AussieBob. GA.

I believe that everyone who has posted on this thread should read the pprune thread that you posted and then do some soul searching about the accuracy or the sensitivity of their own posts.

Guest

I fail to see why we need to more sensitive of our posts? Or do any soul searching for that matter!

From what I have read so far I see no one being offensive or disrespectful.

We are only at the end of the day generalising on a tragic event are we not, in fact are discussions on this topic are probably a lot less disrespectful than what the media would broad cast to the world and the relatives of the AF 447 flight to see.

We are only putting forward are thoughts and opinions on the possible faults leading up to the a/c tragic end I don't see how that's being insensitive

Maybe you would like to explain ?

Technical matters such as accident investigation tend to be a little cold anyway. There is no room in an investigation for being too sensitive. Respectful, yes. Sensitive, no. In an investigation the only things that really count are the facts. No opinions. Any gut feel has to be backed by data. Otherwise, the investigation gets clobbered by politics and feelings.

That's not to say you have to be malicious though. Just incisive and honest. The combination is what makes it difficult.