TAM A320 crash at Congonhas, Brazil
PPRuNe supporter
Join Date: Dec 2003
Location: Planet Earth
Posts: 1,677
Likes: 0
Received 0 Likes
on
0 Posts
1. This significant Thrust differential woudn't cause a left yaw and also make the aircraft "float" and not touchdown on the "spot"?
2. If the Eng 2 TLA was "forgotten" at CLB position, wouldn't be an almost instinctive reaction to reduce Eng 2 TLA to iddle and "kill" excessive yaw?
IMHO, both TLAs were pulled to iddle, and then reverse. That's why the crew seem so surprised when hell broke loose. I can imagine why the delay in applying manual brakes, the aircraft was veering a lot.
Comments, please?
Rob
Comments, please?
Rob
Join Date: Jul 2007
Location: Europe
Posts: 92
Likes: 0
Received 0 Likes
on
0 Posts
IMHO, both TLAs were pulled to iddle, and then reverse
1) Mechanical lock: If the levers were pulled back during flare like SOP and one was stuck in its CLB position, I am sure that a go around would have been initiated immediately. No one would land at a short strip knowing that an engine is pushing forward. At least that engine would have been shut down on the runway.
2) Both transducers of the #2 thrust lever angle (two per lever) failed simultaneously (e. g. broke loose from the lever, got an overvoltage, ...): This is the only case I can imagine apart from human factors that could cause this picture we see on FDR data. They worked as intended at takeoff, so the probability of this to happen at touchdown is really small. I mean - they must show exactly the same error at once.
3) Just one angle sensor of #2 failing or both, but not in exactly the same manner: The FADEC sets thrust to idle automatically if the acquired data show inconsistencies, be it erroneous readouts or discrepancies between the two sensors.
2. If the Eng 2 TLA was "forgotten" at CLB position, wouldn't be an almost instinctive reaction to reduce Eng 2 TLA to iddle and "kill" excessive yaw?
Join Date: Dec 2006
Location: washington,dc
Posts: 486
Likes: 0
Received 0 Likes
on
0 Posts
ray cosmic:
the actual voices of the men flying the plane aren't on you tube or some MP 3 download.
precautions would be taken to allow only investigators to see the video.
I don't have a problem with the idea of cameras in the cockpit...indeed in our simulators we have them to record our actions for later analysis...then the tape is erased in front of both sim partners
the only bad thing it has ever shown...bald spots on the back of pilot's head.
the actual voices of the men flying the plane aren't on you tube or some MP 3 download.
precautions would be taken to allow only investigators to see the video.
I don't have a problem with the idea of cameras in the cockpit...indeed in our simulators we have them to record our actions for later analysis...then the tape is erased in front of both sim partners
the only bad thing it has ever shown...bald spots on the back of pilot's head.
Join Date: Jul 2007
Location: Europe
Posts: 92
Likes: 0
Received 0 Likes
on
0 Posts
Tree:
Yes, good point. What I wanted to express with my previous "no" was mainly, that it is very difficult to introduce new rules in some logic because of unwanted and not foreseen side effects when the circumstances are different.
After all, IMHO the logic should be as simple as possible. Not because of introducing new failures in the software but rather because I think a pilot should be able to understand what's going on. The more complex the rules, the more you are left alone. And I think some of the rules of the A320 logic are at a very complex level already. But anyway, this time there seems to be room for improvement, even when we consider a "forgotten" TL to be very unlikely.
PLEASE. This is NOT 100$-PC-crap-harware. This has nothing to do with your daily windows breakdown, there is no such thing as an IRQ.
How about making one of the dots "both brake pedals depressed beyond the 50% travel point?"
After all, IMHO the logic should be as simple as possible. Not because of introducing new failures in the software but rather because I think a pilot should be able to understand what's going on. The more complex the rules, the more you are left alone. And I think some of the rules of the A320 logic are at a very complex level already. But anyway, this time there seems to be room for improvement, even when we consider a "forgotten" TL to be very unlikely.
IMHO it could (hypothetically) happen if they shared same IRQ.
Join Date: Feb 1999
Location: The Orion Arm
Posts: 118
Likes: 0
Received 0 Likes
on
0 Posts
Cockpit Image Recorders
Yuhp, now that it has been brought up, there's no doubt that if they were able to be used correctly, CIR could be of some assistance with instrument panel views, but any good DFDR will tell you whether one or both thrust levers were moved, you don't need image recorders for that! The price is still too high to have them installed. Cosmic, you hit it on the head - the world's armchair experts would see the video before many of the real experts
Bomarc, there are lots of CVR mp3 downloads on the net, just ask google. Admittedly they appear well after the accident, but they should not be available on the public domain at all. It won't be long before YouTube becomes a sadistic voyeur's haven ... o wait... it is already, how silly of me
Bomarc, there are lots of CVR mp3 downloads on the net, just ask google. Admittedly they appear well after the accident, but they should not be available on the public domain at all. It won't be long before YouTube becomes a sadistic voyeur's haven ... o wait... it is already, how silly of me
Last edited by deadhead; 11th Aug 2007 at 23:32. Reason: Responding to bomarc
Join Date: Dec 2006
Location: washington,dc
Posts: 486
Likes: 0
Received 0 Likes
on
0 Posts
have you guys ever heard of ghost images on ATC radar screens?
There have been cases in which targets are either missing from the screen BUT RECORDED on ATC computers or vice versa, where TARGETS are shown on the screen but not recorded on the ATC computers.
how does a controller know? he doesn't.
having a video camera with a view of the scope would show what the controller saw.
THE SAME could be used to prove pilot actions independently of DFDR's.
And yes, you are right MP3's of CVRs are out there...so far not this one though.
There have been cases in which targets are either missing from the screen BUT RECORDED on ATC computers or vice versa, where TARGETS are shown on the screen but not recorded on the ATC computers.
how does a controller know? he doesn't.
having a video camera with a view of the scope would show what the controller saw.
THE SAME could be used to prove pilot actions independently of DFDR's.
And yes, you are right MP3's of CVRs are out there...so far not this one though.
Join Date: Dec 2006
Location: washington,dc
Posts: 486
Likes: 0
Received 0 Likes
on
0 Posts
IF the pilots didn't pull both thrust levers back, then we have a problem with training and understanding of the airbus systems. This can be addressed.
IF the thrust levers were both pulled back, then we have a problem with the equipment. this can be addressed by changes in the equipment and a last ditch procedure of shutting down engines.
But how will we ever know for sure? Depending solely on the DFDR for thrust lever position will always leave room for a tiny bit of doubt.
Having a camera in the cockpit to have recorded the pilot's actions would clarify the situation.
one picture is worth 1000 words.
IF the thrust levers were both pulled back, then we have a problem with the equipment. this can be addressed by changes in the equipment and a last ditch procedure of shutting down engines.
But how will we ever know for sure? Depending solely on the DFDR for thrust lever position will always leave room for a tiny bit of doubt.
Having a camera in the cockpit to have recorded the pilot's actions would clarify the situation.
one picture is worth 1000 words.
Join Date: Jul 2007
Location: Europe
Posts: 92
Likes: 0
Received 0 Likes
on
0 Posts
No IRQ between TLA sensors and FADEC ?
Once again, this is no cheapo hardware like just one microcontroller doing all the math, so there is no need for such constructions like IRQs.
The engineering is well aware of the importance of treating these things right. Also the test departments are very creative inventing scenarios the engineering departments might have overlooked, kind of competition. Trust me, I have worked in that business, although not especially on thrust levers and engines.
Join Date: Aug 2005
Location: fl
Posts: 2,525
Likes: 0
Received 0 Likes
on
0 Posts
bomarc is correct in that a video camera in the cockpit would have shown if the pilot retarded both TL's to idle or just one. The DFDR says he left the R TL up and didn't retard it to idle. Maybe he did but what if the sensing failed and the DFDR got bad data and thought he didn't retard the R TL to idle. The same information would most likely go into the logic of the automation. I hope pilots don't let thrust levers stay half way up making landings with these automatic airplanes with a simple TR problem that is ok to fly with for 10 more days again. If that is what really happened.
Join Date: Aug 2007
Location: Brazil
Age: 71
Posts: 131
Likes: 0
Received 0 Likes
on
0 Posts
About the Ground Spoiler Status (armed/not armed), on page 4 of the FDR's graphics, at 18:48:36 it jumps from armed to not armed and then to armed again.
On the previous landings I didn't notice that. Could this mean that the Crew tried to deploy Ground Spoilers by cycling It's switches? If so, the Crew wouldn't notice the right TLA at the wrong position?
Am I reading those graphics right?
Rob
On the previous landings I didn't notice that. Could this mean that the Crew tried to deploy Ground Spoilers by cycling It's switches? If so, the Crew wouldn't notice the right TLA at the wrong position?
Am I reading those graphics right?
Rob
Last edited by Rob21; 12th Aug 2007 at 02:23.
The DFDR shows the right TR was not retarded and the other data supports this. So why do you all think the DFDR was suddenly corrupt?
It appears that for some reason 2 experienced Captains landed their A/C and simply forgot to close both TL's
I mean even if you forget the DFDR, weren't there witnesses that saw the A/C going down the runway not slowing down with no spoilers and what appeared to be spray from one engine developing power? That too must add weight to the DFDR recordings.
I'm certainly no expert but trying to say the DFDR was wrong? mmm grasping at straws a little.
It appears that for some reason 2 experienced Captains landed their A/C and simply forgot to close both TL's
I mean even if you forget the DFDR, weren't there witnesses that saw the A/C going down the runway not slowing down with no spoilers and what appeared to be spray from one engine developing power? That too must add weight to the DFDR recordings.
I'm certainly no expert but trying to say the DFDR was wrong? mmm grasping at straws a little.
PPRuNe supporter
Join Date: Dec 2003
Location: Planet Earth
Posts: 1,677
Likes: 0
Received 0 Likes
on
0 Posts
it’s wet, and it is slippery. I will report three five left clear, three zero five four
Join Date: Jul 2007
Location: the City by the Bay
Posts: 547
Likes: 0
Received 0 Likes
on
0 Posts
to ELAC: Thanks. Of course you are much more correct then I am because I was only extrapolating the average speed reduction actually achieved. And of course its much more scientific then that as of course speed reduction is not linear due to various factors as you said. However, it is at best an inexact science, this "prediction" of how much more runway would be needed to stop. My numbers would be worse case. At any rate neither Taipei nor Congonhas had enough runway to stop these two jets in similar predicaments. The much longer Taipei runway and RESA allowed a potentially horrible Congonhas type disaster to be mitigated to a relatively minor problem, as "zero injuries" and "plane still in service" indicates. If Congonhas had the luxury of a much longer runway , then of course we would be looking at a far less serious accident. But we have to go with what we have to work with airport wise.
Which brings me to the runway at Congonhas (re: Nigel on Draft):
Of course not advocating all runways be closed to jets as big or larger then A320/ 737 if they are less then 10,000feet. Just a matter of correct management as appropriate to each individual runway. EAch is unique and has an unique set of circumstances of course.
IMHO, Congonhas has several options:
1. Do nothing physical to the runway and manage it better. Say restrict planes with T/R inop from operating there. Insistence on proper training for overrun situ for all types operating there. Insist on Go Around if such and such a condition exists or happens at such and such a time, etc. In other words, deal with it as a Black Star Airport (like HK's Kai Tak) accordingly. And hope for the best.
2. Keep everything per above but add EMAS to all runway end zones as well as alongside the entire length of the runways (where possible) . JUst EMAS every possible area as much as possible. And hope for the best.
3. Close Congonhas to jets the size of A320/ 737 and only allow planes with proven shorter runway requirements for landing and take off. And hope for the best.
4. Complete rezoning of the areas around the airport. Yes knock down buildings, rezone the whole area by government edict. Build that 10,000 foot runway (12,000 is no doubt better, everyone likes 12,000foot runways) and put in recommended RESA with EMAS. Make Congonhas state of the art. And hope for the best.
Landing is a managed risk. There is always a risk, just we need to put as much of the odds in our favor as we can. Safer airports with better flying and maintanence sure helps and can save the day.
to Discountinvestigator:
1. ok
2. I understand 61 meter runway end zone Per Se.
3. http://www.ifalpa.org0/
Which brings me to the runway at Congonhas (re: Nigel on Draft):
Of course not advocating all runways be closed to jets as big or larger then A320/ 737 if they are less then 10,000feet. Just a matter of correct management as appropriate to each individual runway. EAch is unique and has an unique set of circumstances of course.
IMHO, Congonhas has several options:
1. Do nothing physical to the runway and manage it better. Say restrict planes with T/R inop from operating there. Insistence on proper training for overrun situ for all types operating there. Insist on Go Around if such and such a condition exists or happens at such and such a time, etc. In other words, deal with it as a Black Star Airport (like HK's Kai Tak) accordingly. And hope for the best.
2. Keep everything per above but add EMAS to all runway end zones as well as alongside the entire length of the runways (where possible) . JUst EMAS every possible area as much as possible. And hope for the best.
3. Close Congonhas to jets the size of A320/ 737 and only allow planes with proven shorter runway requirements for landing and take off. And hope for the best.
4. Complete rezoning of the areas around the airport. Yes knock down buildings, rezone the whole area by government edict. Build that 10,000 foot runway (12,000 is no doubt better, everyone likes 12,000foot runways) and put in recommended RESA with EMAS. Make Congonhas state of the art. And hope for the best.
Landing is a managed risk. There is always a risk, just we need to put as much of the odds in our favor as we can. Safer airports with better flying and maintanence sure helps and can save the day.
to Discountinvestigator:
1. ok
2. I understand 61 meter runway end zone Per Se.
3. http://www.ifalpa.org0/
Last edited by armchairpilot94116; 12th Aug 2007 at 16:45.
Join Date: Sep 2002
Location: La Belle Province
Posts: 2,179
Likes: 0
Received 0 Likes
on
0 Posts
@Rob21
I suspect - and this is purely a guess, not knowing the details of the AB GS logic - that the GS only remain armed for a set period of time following touchdown, and if they haven't deployed within that time then they disarm. That would be one way of ensuring they don't deploy during a touch-and-go, say. That *may* be why they disarmed.
Since the reversion to "Armed" occurs at 18:48:48 - which from the first page acceleration traces looks a lot like the start of the fatal impact - I'd hesitate to put any great credence on that behaviour. The last few seconds of data in any accident have to be treated with great caution.
edit: another possibility:
In order to move the speedbrake handle, they would have to push it down, thus disarming the GS. The disarming may indicate an attempt to manually deploy the speedbrakes - which of course wouldn't work in the landing config, but they may have forgotten that in the attempt to do SOMETHING.
Perhaps significant, the "it can't" statement on the CVR has been added at almost the same time as the GS disarm. Perhaps the references people have made to the crew being unable to move a lever (though there's no reference in the CVR which directly fits those early statements) - which many speculated meant a jammed TLA - actually occurs at about this time, and is referring to being unable to either move the Speedbrake lever or, once it was moved, to it not deploying the speedbrakes?
@Sdruvss
You described a dual failure; for the kind of system used to control a FADEC (or other level A systems) to fail requires far more than a dual failure, and even then the system will fail in a passive sense under almost all circumstances. I'd put the chance of a combined hardware/software failure being the reason why the TLA is not retarded to idle as very much less than 1 in a billion.
About the Ground Spoiler Status (armed/not armed), on page 4 of the FDR's graphics, at 18:48:36 it jumps from armed to not armed and then to armed again.
On the previous landings I didn't notice that. Could this mean that the Crew tried to deploy Ground Spoilers by cycling It's switches? If so, the Crew wouldn't notice the right TLA at the wrong position?
Am I reading those graphics right?
On the previous landings I didn't notice that. Could this mean that the Crew tried to deploy Ground Spoilers by cycling It's switches? If so, the Crew wouldn't notice the right TLA at the wrong position?
Am I reading those graphics right?
Since the reversion to "Armed" occurs at 18:48:48 - which from the first page acceleration traces looks a lot like the start of the fatal impact - I'd hesitate to put any great credence on that behaviour. The last few seconds of data in any accident have to be treated with great caution.
edit: another possibility:
In order to move the speedbrake handle, they would have to push it down, thus disarming the GS. The disarming may indicate an attempt to manually deploy the speedbrakes - which of course wouldn't work in the landing config, but they may have forgotten that in the attempt to do SOMETHING.
Perhaps significant, the "it can't" statement on the CVR has been added at almost the same time as the GS disarm. Perhaps the references people have made to the crew being unable to move a lever (though there's no reference in the CVR which directly fits those early statements) - which many speculated meant a jammed TLA - actually occurs at about this time, and is referring to being unable to either move the Speedbrake lever or, once it was moved, to it not deploying the speedbrakes?
@Sdruvss
You described a dual failure; for the kind of system used to control a FADEC (or other level A systems) to fail requires far more than a dual failure, and even then the system will fail in a passive sense under almost all circumstances. I'd put the chance of a combined hardware/software failure being the reason why the TLA is not retarded to idle as very much less than 1 in a billion.
Last edited by Mad (Flt) Scientist; 12th Aug 2007 at 04:11.
Join Date: Sep 2002
Location: La Belle Province
Posts: 2,179
Likes: 0
Received 0 Likes
on
0 Posts
@ACMS
The implication people are making is not that the DFDR data are wrong, but that they don't accurately reflect crew action either because:
(1) the TLA was stuck and the crew didn't 'forget' to retard the throttle, but couldn't; or
(2) the TLA was physically retarded, but some esoteric fault caused the RH engine to behave as if the TLA were left unmoved, which in turn inhibited the various systems dependent on TLA positioning.
I don't consider either very likely scenarios - the simplest explanation is that the crew didn't move the RH TLA - but those are what people are referring to eliminating as possibilities with a cockpit camera, I believe.
The implication people are making is not that the DFDR data are wrong, but that they don't accurately reflect crew action either because:
(1) the TLA was stuck and the crew didn't 'forget' to retard the throttle, but couldn't; or
(2) the TLA was physically retarded, but some esoteric fault caused the RH engine to behave as if the TLA were left unmoved, which in turn inhibited the various systems dependent on TLA positioning.
I don't consider either very likely scenarios - the simplest explanation is that the crew didn't move the RH TLA - but those are what people are referring to eliminating as possibilities with a cockpit camera, I believe.
Join Date: Jul 2007
Location: Hadlow
Age: 60
Posts: 597
Likes: 0
Received 0 Likes
on
0 Posts
Data recorders
Thanks to A310driver and PJ2 for replying to my question.
As I thought, it is technologically feasible, Cost & politics seem to be the main problems. Will leave the politics to the politicians, but a compromise on the cost could be achieved by having live streaming of data only during take-off and landing, which is where most accidents occur.
As I thought, it is technologically feasible, Cost & politics seem to be the main problems. Will leave the politics to the politicians, but a compromise on the cost could be achieved by having live streaming of data only during take-off and landing, which is where most accidents occur.