AF447 Thread No. 3
Join Date: Jun 2011
Location: Canada
Posts: 2
Likes: 0
Received 0 Likes
on
0 Posts
@ twistedenginestarter
Did vicious vertical air movements trigger an upset which the intended behaviour of the computers unfortunately led to the crew getting disoriented?
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes
on
0 Posts
twistedenginestarter,
There are several BEA reports to read:
The first BEA Interim report you can find here:
http://www.bea.aero/docspa/2009/f-cp...90601e1.en.pdf
It speaks of pitot tubes and gives the information on how they function and how they are connected to the flight system.
The second BEA Interim report you can find here:
http://www.bea.aero/docspa/2009/f-cp...90601e2.en.pdf
This report gets deeper into the pitot tubes. It discusses how they are certified and examines the experiences of pitot tube icing and observed differences of icing between pitot tubes manufactured by Thales and those manufactured by Goodrich on Airbus aircraft. Refer to Appendix 7 for a listing of events. I think the BEA is very suspicious of pitot tube icing at the time of this report as Thales pitot tubes were the installed on AF447, which were to be replaced by a EASA AD, but were not yet changed out on AF447.
The most recent BEA report is an Investigation Update you can find here:
http://www.bea.aero/fr/enquetes/vol....mai2011.en.pdf
It makes no mention of the pitot tubes, but gives some information from the FDR and CVR regarding the last moments of the flight.
Hope this helps your understanding.
There are several BEA reports to read:
The first BEA Interim report you can find here:
http://www.bea.aero/docspa/2009/f-cp...90601e1.en.pdf
It speaks of pitot tubes and gives the information on how they function and how they are connected to the flight system.
The second BEA Interim report you can find here:
http://www.bea.aero/docspa/2009/f-cp...90601e2.en.pdf
This report gets deeper into the pitot tubes. It discusses how they are certified and examines the experiences of pitot tube icing and observed differences of icing between pitot tubes manufactured by Thales and those manufactured by Goodrich on Airbus aircraft. Refer to Appendix 7 for a listing of events. I think the BEA is very suspicious of pitot tube icing at the time of this report as Thales pitot tubes were the installed on AF447, which were to be replaced by a EASA AD, but were not yet changed out on AF447.
The most recent BEA report is an Investigation Update you can find here:
http://www.bea.aero/fr/enquetes/vol....mai2011.en.pdf
It makes no mention of the pitot tubes, but gives some information from the FDR and CVR regarding the last moments of the flight.
Hope this helps your understanding.
Last edited by Turbine D; 7th Jun 2011 at 00:53. Reason: spelling correction
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by antpp
pitch angle = aoa in horizontal flight at FL 380 which has no other explanation than downward wind strong enough to change aoa by about 4 degrees.
Do not confuse airplane AoA with the local AoA of a wing section at any spanwise station, which varies considerably between wing root and tip.
Last edited by HazelNuts39; 6th Jun 2011 at 22:09.
THS autotrim mechanization
Thanks for the reference, Doze. Already had same for another 'bus variant, but nice to read the real deal.
I still can't see a diagram of how the THS moves with respect to the pilot or autopilot inputs to the "system".
I don't like the term "autotrim", BTW. We tried to eliminate that term in the Viper, as all it meant was the jet would try to maintain the last TRIMMED gee command. Unlike the 'bus, we could trim the gee with the hat switch on the stick or the roller gizmo like the 'bus has. If we had trimmed for zero gee ( some guys did that for air-to-air maneuvers so all they had to do to get energy in a hurry was relaz pressure on the stick), the thing tried to get to zero gee when you let go, duhhhh? We did not have the pitch attitude correction or the bank angle correction - a stable 30 deg climb only required about 0.87 normal gee, for exampl. So our jet would continue to slowly pull up if you let go and it had been trimmed for one gee. As with our little jet, the Airbus is not an attitude command system.
So what gee does the Airbus try to reach when I let go of the stick and the THS is at 10 degrees or 2 degrees or....??? The jet is a gee command as the basic Viper is, and has different gains and rates and such, but it is not an AoA -gee system like we had. Ours gave you 9 gees at 15 degrees AoA, then the gee went down until 25 degrees AoA and it was one gee. Our manuals showed this and we demonstrated it first hop with student studly.
So what does the THS do if I have a constant back stick that moves the elevator beyond "neutral"? Another post implies that the THS does not move unless the elevator is something other than the "neutral" position. Surely someone here can show how the thing works.
Inquiring FBW pioneers wanna know!
I still can't see a diagram of how the THS moves with respect to the pilot or autopilot inputs to the "system".
I don't like the term "autotrim", BTW. We tried to eliminate that term in the Viper, as all it meant was the jet would try to maintain the last TRIMMED gee command. Unlike the 'bus, we could trim the gee with the hat switch on the stick or the roller gizmo like the 'bus has. If we had trimmed for zero gee ( some guys did that for air-to-air maneuvers so all they had to do to get energy in a hurry was relaz pressure on the stick), the thing tried to get to zero gee when you let go, duhhhh? We did not have the pitch attitude correction or the bank angle correction - a stable 30 deg climb only required about 0.87 normal gee, for exampl. So our jet would continue to slowly pull up if you let go and it had been trimmed for one gee. As with our little jet, the Airbus is not an attitude command system.
So what gee does the Airbus try to reach when I let go of the stick and the THS is at 10 degrees or 2 degrees or....??? The jet is a gee command as the basic Viper is, and has different gains and rates and such, but it is not an AoA -gee system like we had. Ours gave you 9 gees at 15 degrees AoA, then the gee went down until 25 degrees AoA and it was one gee. Our manuals showed this and we demonstrated it first hop with student studly.
So what does the THS do if I have a constant back stick that moves the elevator beyond "neutral"? Another post implies that the THS does not move unless the elevator is something other than the "neutral" position. Surely someone here can show how the thing works.
Inquiring FBW pioneers wanna know!
Join Date: Aug 2005
Location: fairly close to the colonial capitol
Age: 55
Posts: 1,693
Likes: 0
Received 0 Likes
on
0 Posts
By the way, the Braniff example you just cited was not true. A Boeing pilot was in the seat, but not at the controls and it was not a developmental test flight. The Braniff pilot managed to remove 3 engines before they had to crash land the aircraft. In the sense of flying a 707 with 1 engine, that surely would have been in the realm of a flight "test. "
cheers!
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes
on
0 Posts
Side-stick Position
Having noted a few comments about the PNF not knowing what the PF is doing with the side-stick, it is worth noting that in the Ground Mode after start-up the side-stick position is shown on each PFD.
My suggestion is that in Flight Mode that this option is selectable, and when in other than Normal Law and on auto A/P disconnect, that the side-stick summed position be shown automatically on both displays. Both pilots and the 3rd or 4th seat occupants will also see what is happening.
My suggestion is that in Flight Mode that this option is selectable, and when in other than Normal Law and on auto A/P disconnect, that the side-stick summed position be shown automatically on both displays. Both pilots and the 3rd or 4th seat occupants will also see what is happening.
Join Date: May 2010
Location: Boston
Age: 73
Posts: 443
Likes: 0
Received 0 Likes
on
0 Posts
DW:
What is the nature of the artificial feel you mention?
If not to the sidestick then where?
Lack of back drive (what the other pilot is doing) is a seperate issue than tactile feedback (what the airplane is doing) to the controls.
The lack of "what the other pilot is doing" feedback seems to me to be more questionable than "what the airplane is" doing since that data is available in many other forms but the PNF appears to not have a way to tune in to "input vs response" of the aircraft.
Airbus aircraft do retain a degree of artificial feel to allow the pilot to know what the *aircraft* is doing, but what CONF and the others are banging on about is the fact that on the Airbus you can't directly feel what the other *pilot* is doing. That's the issue, and that's what I'm talking about when it comes to back-drive.
If not to the sidestick then where?
Lack of back drive (what the other pilot is doing) is a seperate issue than tactile feedback (what the airplane is doing) to the controls.
The lack of "what the other pilot is doing" feedback seems to me to be more questionable than "what the airplane is" doing since that data is available in many other forms but the PNF appears to not have a way to tune in to "input vs response" of the aircraft.
Join Date: Feb 2000
Location: Pacific
Posts: 731
Likes: 0
Received 0 Likes
on
0 Posts
ChristianJ.. The artificial feel built in to the Boeing flight control needs speed input and from memory it used a different set of pitot tubes for that. It did not use the regular pitot information provided to the flight instruments. Those tubes were also heated, but my suggestion is that if the flight instrument tubes iced up, then the feel system pitot might also be iced up, thus causing the feel to be incorrect. Hence the feel might be wildly inaccurate under those conditions.
I am not aware of any way to turn this system off, but in the old 737 you could turn off the hydraulic system that powered the feel and it was a strange experience to fly an airplane without any feedback. As some have said, like a video game. I never did hear of any problems with the feel system on the Boeing, but it might have had some effect on the two 757 accidents. Since the 'Bus does not have the system it would not matter.
The Boeing checklist has a table of thrust and attitude for various configurations and speeds, to cover this type of emergency. Does the 'Bus have that, and is it accessible in flight? (paper or plastic)
Would the crew have been trained on this?
I am not aware of any way to turn this system off, but in the old 737 you could turn off the hydraulic system that powered the feel and it was a strange experience to fly an airplane without any feedback. As some have said, like a video game. I never did hear of any problems with the feel system on the Boeing, but it might have had some effect on the two 757 accidents. Since the 'Bus does not have the system it would not matter.
The Boeing checklist has a table of thrust and attitude for various configurations and speeds, to cover this type of emergency. Does the 'Bus have that, and is it accessible in flight? (paper or plastic)
Would the crew have been trained on this?
Join Date: Aug 2005
Location: fairly close to the colonial capitol
Age: 55
Posts: 1,693
Likes: 0
Received 0 Likes
on
0 Posts
The Boeing checklist has a table of thrust and attitude for various configurations and speeds, to cover this type of emergency. Does the 'Bus have that, and is it accessible in flight? (paper or plastic)
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes
on
0 Posts
PJ2,
PJ2's quote:
I agree, there isn't much left to go relative to thrust! A couple of questions:
1. What would be the N1% at cruise at 35K for the CF6-80E's at Mach 0.83?
2. What would it be if the Mach was reduced to 0.80?
Also, I noted the BEA changed the estimated COG to 29% aft from 38-39% aft on AF 447 at the time of the incident. If true, what does that do to N1% as I would think the N1% required would increase to maintain speed due to more drag?
Your thoughts? I am interested in this among other thing I am not familiar with.
PJ2's quote:
For those who seem to believe that CLB or even MCT at FL350 on the CFs or RRs is cause for a huge pitch-up followed by an immediate overspeed or an uncontrolled climb for the A330, it needs to be understood that there is not much residual thrust remaining and the aircraft response from what "boost" there is, is eminently controllable.
1. What would be the N1% at cruise at 35K for the CF6-80E's at Mach 0.83?
2. What would it be if the Mach was reduced to 0.80?
Also, I noted the BEA changed the estimated COG to 29% aft from 38-39% aft on AF 447 at the time of the incident. If true, what does that do to N1% as I would think the N1% required would increase to maintain speed due to more drag?
Your thoughts? I am interested in this among other thing I am not familiar with.
Last edited by Jetdriver; 6th Jun 2011 at 22:52.
Join Date: Jun 2009
Location: Dorset
Posts: 31
Likes: 0
Received 0 Likes
on
0 Posts
Possible Updraft Effects
There are a number of unresolved (at least for me) queries based mainly on the information in the BEA Interim Report 2 and the Update on Investigation report of 27 May 2011. A possible scenario that might answer these queries is rather than one of blocked pitots, instead a moderate to severe updraft.
Queries:-
a) A/C rolls to right: The “go left a little” at 02:08:07 would put the turbulence to the right of a/c, with the a/c flying at a tangent to any column of rising air. A moderate to severe updraft under the right wing would probably then cause EFCS to respond with right wing down.
b) PF inputs left and nose-up: In an Updraft air is being forced up by the air beneath so static air pressure would increase. This would cause Baro Alt to very rapidly decrease, giving the impression the a/c was descending, possibly then causing the PF to pull the nose up. Also a “barometric vertical speed higher, as an absolute value, than 20,000 ft/min”would cause the FPVs to disappear.
c) TCAS fail: this could be due to failure of the altitude credibility check
d) A/S "sharp fall" from 275 > 60kts Captain's PFD: In an updraft the raw reading of dynamic pressure would be affected(as the airflow would be partially across the end of the tube) by cosine of AoA plus any venturi effect. Also in an updraft, with a rapid increase in local static pressure, calculated AS would rapidly decrease. The subsequent calculations of AS would be affected by the various changes in AoA and static pressure causing a number of fluctuations. Other air speed related faults could also be due to the effects of an updraft, e.g. ISIS and ADIRU.
e) PRIM 1 and SEC 1 fault:Possibly the PF turns them off in desperation, believing that the apparent erratic behaviour of the a/c was because the EFCS was faulty.
Queries:-
a) A/C rolls to right: The “go left a little” at 02:08:07 would put the turbulence to the right of a/c, with the a/c flying at a tangent to any column of rising air. A moderate to severe updraft under the right wing would probably then cause EFCS to respond with right wing down.
b) PF inputs left and nose-up: In an Updraft air is being forced up by the air beneath so static air pressure would increase. This would cause Baro Alt to very rapidly decrease, giving the impression the a/c was descending, possibly then causing the PF to pull the nose up. Also a “barometric vertical speed higher, as an absolute value, than 20,000 ft/min”would cause the FPVs to disappear.
c) TCAS fail: this could be due to failure of the altitude credibility check
d) A/S "sharp fall" from 275 > 60kts Captain's PFD: In an updraft the raw reading of dynamic pressure would be affected(as the airflow would be partially across the end of the tube) by cosine of AoA plus any venturi effect. Also in an updraft, with a rapid increase in local static pressure, calculated AS would rapidly decrease. The subsequent calculations of AS would be affected by the various changes in AoA and static pressure causing a number of fluctuations. Other air speed related faults could also be due to the effects of an updraft, e.g. ISIS and ADIRU.
e) PRIM 1 and SEC 1 fault:Possibly the PF turns them off in desperation, believing that the apparent erratic behaviour of the a/c was because the EFCS was faulty.
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes
on
0 Posts
By the way, the Braniff example you just cited was not true. A Boeing pilot was in the seat, but not at the controls and it was not a developmental test flight. The Braniff pilot managed to remove 3 engines before they had to crash land the aircraft. In the sense of flying a 707 with 1 engine, that surely would have been in the realm of a flight "test. "
cheers!
cheers!
Join Date: Aug 2005
Location: fairly close to the colonial capitol
Age: 55
Posts: 1,693
Likes: 0
Received 0 Likes
on
0 Posts
No problem, thanks for the correction. However part of me suspects that if PPRuNe had existed back then, we'd have heard all about how Douglas had never lost a civilian test pilot in the same amount of time, how Boeing was messing with time honoured pilot tradition with these jet aircraft, and just what the hell was wrong with good old propliners anyway?
Join Date: Oct 2010
Location: Germany
Age: 47
Posts: 402
Likes: 0
Received 0 Likes
on
0 Posts
pj2
very nice post and you,re right ! on an anonymous forum its hard and sometimes dangerous to follow expreriences from self named experts.i can remember an another discussion and dangerous hints from a now banned member...
well,i am not an airbus pilot, but discussing terms like moving/non moving throttle levers for real life is silly. it is a non event for anybody who is REALLY rated on this plane (and not a computer game) . people need to understand that this planes fly everyday and all we need is the understanding that you "notch"the levers for a certain part of flight and let the computer help you. every truly real pilot will understand that in times where you deal with several other things like ATC, delay problems , unexpected route changes especially at a complicated departure, the computer will help you, not harm you, but the plane still will be limited to physical rules. the same applies to the migty TOGA. i wrote like you also several pages before that toga in fl 350 will not give you such a burst that its in regards to trim its an event.
well. the critical question for all this discussion, on an open forum like it seems not very sufficient discussion, is what happened and what can we learn?
it seems strange for me that he pilot climbed out and stalled ( in alt law) until impact. even when airbus is blamed for an failure on IAS indication... keeping power and pitch could help. maybe it was like many incidents: a massive pilot error who wasin panic. this happens, and every time it happens it is sad.
best regards to all !
very nice post and you,re right ! on an anonymous forum its hard and sometimes dangerous to follow expreriences from self named experts.i can remember an another discussion and dangerous hints from a now banned member...
well,i am not an airbus pilot, but discussing terms like moving/non moving throttle levers for real life is silly. it is a non event for anybody who is REALLY rated on this plane (and not a computer game) . people need to understand that this planes fly everyday and all we need is the understanding that you "notch"the levers for a certain part of flight and let the computer help you. every truly real pilot will understand that in times where you deal with several other things like ATC, delay problems , unexpected route changes especially at a complicated departure, the computer will help you, not harm you, but the plane still will be limited to physical rules. the same applies to the migty TOGA. i wrote like you also several pages before that toga in fl 350 will not give you such a burst that its in regards to trim its an event.
well. the critical question for all this discussion, on an open forum like it seems not very sufficient discussion, is what happened and what can we learn?
it seems strange for me that he pilot climbed out and stalled ( in alt law) until impact. even when airbus is blamed for an failure on IAS indication... keeping power and pitch could help. maybe it was like many incidents: a massive pilot error who wasin panic. this happens, and every time it happens it is sad.
best regards to all !
Join Date: Jan 2008
Location: Herts, UK
Posts: 748
Likes: 0
Received 0 Likes
on
0 Posts
HazelNuts
I think you might find a common definition of AoA is freestream vector Vs Wing Zero Lift Angle, not fuselage axis datum
or even, Vs Wing Chordline Datum
After all, when w/t testing wing sections, there is no fusleage or wing chord datum setting angle to refer. to
I think you might find a common definition of AoA is freestream vector Vs Wing Zero Lift Angle, not fuselage axis datum
or even, Vs Wing Chordline Datum
After all, when w/t testing wing sections, there is no fusleage or wing chord datum setting angle to refer. to
Turbine D;
I don't have the numbers for a CF installation on a 332. The Trent installation on a 333 requires a cruise N1% of 78.1, with a CG of 37% - the chart doesn't provide for corrections but I think your observation of a forward CG requiring a bit higher N1 is sound...certainly it would again be a bit higher at M0.83, and quite a bit less (in terms of fuel flow) at M0.80). Max CLB might be 83-87% N1, MCT might be around 92-94% N1. As a good friend described it, the L/D curve is like a marble on a plate...
My personal experience with the A340 (CFM-56s) was a much higher fuel flow required at M0.83 than M0.82. We could often save 3000kg on a ten-hr flight by bringing the Mach back by .01, (about 8-9kts), with ATC clearance. M0.80 wouldn't normally be filed of course as that's quite slow on either ocean but that's the Turbulence Penetration speed, so used as needed. Perhaps someone here might have those exact numbers for you.
I don't have the numbers for a CF installation on a 332. The Trent installation on a 333 requires a cruise N1% of 78.1, with a CG of 37% - the chart doesn't provide for corrections but I think your observation of a forward CG requiring a bit higher N1 is sound...certainly it would again be a bit higher at M0.83, and quite a bit less (in terms of fuel flow) at M0.80). Max CLB might be 83-87% N1, MCT might be around 92-94% N1. As a good friend described it, the L/D curve is like a marble on a plate...
My personal experience with the A340 (CFM-56s) was a much higher fuel flow required at M0.83 than M0.82. We could often save 3000kg on a ten-hr flight by bringing the Mach back by .01, (about 8-9kts), with ATC clearance. M0.80 wouldn't normally be filed of course as that's quite slow on either ocean but that's the Turbulence Penetration speed, so used as needed. Perhaps someone here might have those exact numbers for you.
Join Date: Jul 2009
Location: DFW
Age: 61
Posts: 221
Likes: 0
Received 0 Likes
on
0 Posts
Way back on page thirty-something, it was theorized that none of the three pilots recognized the stall because none of them said anything about getting the nose down.
I have been in a full stall in a swept wing narrowbody (MD80 on a post m/x flight check) and I can add my agreement to that theory. In my case, I was the pilot monitoring the system (stall warning) and was surprised by the unexpected break. Rest assured that I immediately started yelling "PUSH" at the top of my lungs.
To all: As a current A320 Capt, these threads on PPRuNe have been a valuable learning experience. Thank You to everyone.
PS. My carrier produced and trained a fine scenario for Sim Tng of the UAS scenario. But it was based on ADIRU issues not external issues that could cause confusion in the FBW system. This forum has added far more to my knowledge of UAS issues than the airline tng ever did. Thanks again to everyone.
I have been in a full stall in a swept wing narrowbody (MD80 on a post m/x flight check) and I can add my agreement to that theory. In my case, I was the pilot monitoring the system (stall warning) and was surprised by the unexpected break. Rest assured that I immediately started yelling "PUSH" at the top of my lungs.
To all: As a current A320 Capt, these threads on PPRuNe have been a valuable learning experience. Thank You to everyone.
PS. My carrier produced and trained a fine scenario for Sim Tng of the UAS scenario. But it was based on ADIRU issues not external issues that could cause confusion in the FBW system. This forum has added far more to my knowledge of UAS issues than the airline tng ever did. Thanks again to everyone.
Join Date: Jul 2009
Location: Tranquility Base
Age: 68
Posts: 53
Received 0 Likes
on
0 Posts
Altimeter Data Available?
Have not seen much mention if any of whether altimeter data was available. It appears to me from looking at the releases that it should have been there, and would have been unreeling downward and would have been part of the instrument cross-check. Anyone knowledgeable on the -330 want to comment on this?
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes
on
0 Posts
alex_brin, relax a little and consider your sources. He who lives by the news media dies by the news media, or something like that.
The news media is in business to make money. "Extrapolating" and outright "lying" make money.
The "I do not understand" quote does not exist in the BEA report. The closest it comes is this:
The plane's systems were feeding them gibberish for data is what they seem to be saying there. It also implies that at that time they both saw the same thing and were concentrating on their training, which at that time suggested some stall warnings were spurious and in any case recovery meant applying a little more power. We really do not know what data was presented to the PF. He was apparently sitting on the right hand side and the RHS sensors were not recorded the way Air France had the FDR setup.
We can infer from the data presented that a cluster of deficiencies all got together to conspire to bring this plane down. That's not fact. It's simply extracted from the data. Pitot tubes can ice up under conditions they experienced and display false, very low, values. The conditions do not turn up on radar which works best with water and works poorly on frozen water. The pilot training left out some critical "complexities" about high speed stalls and trim tabs. The warnings presented to the pilots MAY not have distinguished between "about to stall" and "stalled" which CAN but don't necessarily have differing recovery modes. The software presumed at 60 knots the aircraft had to be on the ground so stall warnings are superfluous, even at 37,000' indicated altitude. This is not quite accurate as this flight shows. And the pilots seem to have done inexplicable things given the data cluster fed to them when the auto-throttle and autopilot both disconnected.
The above list includes hardware, training, and software design deficiencies.
Regarding the software deficiencies no coding or programmer errors are evident here. They built the design given to them very well. And the particular design decisions may represent engineers getting overruled by management.
So there are enough issues here that courts are going to have a field day trying to affix blame. And some aspects of the plane and training probably should get revisited. How much is a good question. Perfect safety requires infinite effort which requires infinite knowledge and infinite money, none of which exist. And this plane is also far more reliable and safe than prior choices such as the DC-6s, Connies, and 707s I rode on as a child and adolescent. Major improvements may be difficult and, frankly, cost more than is sustainable for either AirBus, Boeing, or the airlines involved. (Some of these issues no doubt infect Boeing aircraft, too. Perfect is for God. All mankind can do is strive for perfection within our limitations.) Simple things like modifying training and using two different types of pitot tubes on the same plane may help. Modifying the software is a very expensive and finicky process. That may not be possible and training extensions to cover this may have to take place.
All of this is without the extended analysis I expect BEA to make. These are my relatively untutored observations with a subset of the data BEA has. I am willing to stick my neck out and suggest if it is pinned on the pilots that will implicate deficiencies their training.
And I've rattled on long enough so it's time to quit with a little observation to another member here (who knows who he is) that the Vertical Stabilizer appears to have really been a part of the plane affixed to the plane until it interfaced with the ocean. And note that I'm trying to determine potential causes and in no way apportion blame.
The news media is in business to make money. "Extrapolating" and outright "lying" make money.
The "I do not understand" quote does not exist in the BEA report. The closest it comes is this:
At 2 h 12 min 02, the PF said "I don’t have any more indications", and the PNF said "we have no valid indications".
We can infer from the data presented that a cluster of deficiencies all got together to conspire to bring this plane down. That's not fact. It's simply extracted from the data. Pitot tubes can ice up under conditions they experienced and display false, very low, values. The conditions do not turn up on radar which works best with water and works poorly on frozen water. The pilot training left out some critical "complexities" about high speed stalls and trim tabs. The warnings presented to the pilots MAY not have distinguished between "about to stall" and "stalled" which CAN but don't necessarily have differing recovery modes. The software presumed at 60 knots the aircraft had to be on the ground so stall warnings are superfluous, even at 37,000' indicated altitude. This is not quite accurate as this flight shows. And the pilots seem to have done inexplicable things given the data cluster fed to them when the auto-throttle and autopilot both disconnected.
The above list includes hardware, training, and software design deficiencies.
Regarding the software deficiencies no coding or programmer errors are evident here. They built the design given to them very well. And the particular design decisions may represent engineers getting overruled by management.
So there are enough issues here that courts are going to have a field day trying to affix blame. And some aspects of the plane and training probably should get revisited. How much is a good question. Perfect safety requires infinite effort which requires infinite knowledge and infinite money, none of which exist. And this plane is also far more reliable and safe than prior choices such as the DC-6s, Connies, and 707s I rode on as a child and adolescent. Major improvements may be difficult and, frankly, cost more than is sustainable for either AirBus, Boeing, or the airlines involved. (Some of these issues no doubt infect Boeing aircraft, too. Perfect is for God. All mankind can do is strive for perfection within our limitations.) Simple things like modifying training and using two different types of pitot tubes on the same plane may help. Modifying the software is a very expensive and finicky process. That may not be possible and training extensions to cover this may have to take place.
All of this is without the extended analysis I expect BEA to make. These are my relatively untutored observations with a subset of the data BEA has. I am willing to stick my neck out and suggest if it is pinned on the pilots that will implicate deficiencies their training.
And I've rattled on long enough so it's time to quit with a little observation to another member here (who knows who he is) that the Vertical Stabilizer appears to have really been a part of the plane affixed to the plane until it interfaced with the ocean. And note that I'm trying to determine potential causes and in no way apportion blame.
My carrier produced and trained a fine scenario for Sim Tng of the UAS scenario. But it was based on ADIRU issues not external issues that could cause confusion in the FBW system.