TAM A320 crash at Congonhas, Brazil
Join Date: Jul 2007
Location: the City by the Bay
Posts: 547
Likes: 0
Received 0 Likes
on
0 Posts
not a minor error
Point taken. Not a minor error indeed ! What I meant to emphasize though (but in no way could do so as well as ELAC) was that somehow it was possible to miss a throttle position during the flare and to ignor the RETARD command (at least as far as retarding both levers were concerned) on the A320 and this was perhaps much less of a problem on the Boeings.
Another point taken was that apparently there IS a situation where you would want to stow your reversers and take off again (in spite of the risk of not making it).
So what is the consensus of opinion so far (from what could possibly be gathered )
1. This plane performed as expected and you dont fault AB control logic and wouldnt want the plane to come to a full stop for you in such a situ.
2. This plane perhaps did not have the CRC and ECAM and Master WArning Light facility to warn of incorrect TL position. Although suggested by AB.
3. Suggested AB fix for this exact problem of mismatched TL should be made mandatory and would be enough of a fix to solve this issue and prevent its exact recurrence.
thank you for indulging me, I will attempt to keep my mouth closed for the next 1000 postings
Another point taken was that apparently there IS a situation where you would want to stow your reversers and take off again (in spite of the risk of not making it).
So what is the consensus of opinion so far (from what could possibly be gathered )
1. This plane performed as expected and you dont fault AB control logic and wouldnt want the plane to come to a full stop for you in such a situ.
2. This plane perhaps did not have the CRC and ECAM and Master WArning Light facility to warn of incorrect TL position. Although suggested by AB.
3. Suggested AB fix for this exact problem of mismatched TL should be made mandatory and would be enough of a fix to solve this issue and prevent its exact recurrence.
thank you for indulging me, I will attempt to keep my mouth closed for the next 1000 postings
Join Date: May 2000
Location: Camp X-Ray
Posts: 2,135
Likes: 0
Received 0 Likes
on
0 Posts
How many Boeings actually land from a normal approach with the autothrust engaged? We certainly don't do it on the 747 as the pitch power couple destabilises a manually flown approach, hence it's perfectly possible to land with the throttles all above idle or in staggered positions. At least on the Airbus you are required to close both thrust levers as the primary method of disconnecting the A/THR when landing.
Join Date: Aug 2005
Location: Cloudbase
Posts: 149
Likes: 0
Received 0 Likes
on
0 Posts
Hetfield, that would certainly have been my question to HanSolo:
when you're flying manual thrust, you're in the loop, actually in command, when it comes to thrust settings.
Of course, you can still forget to move those (or one of those) TLs back to ilde, but it seems far less probable than to forget the state change "CLB to IDLE to REV" for a lever where "REV" is not the ultimate position you'd want to bring it to.
pj
when you're flying manual thrust, you're in the loop, actually in command, when it comes to thrust settings.
Of course, you can still forget to move those (or one of those) TLs back to ilde, but it seems far less probable than to forget the state change "CLB to IDLE to REV" for a lever where "REV" is not the ultimate position you'd want to bring it to.
pj
Soaring skies,
With the Boeing yes you could manually select spoilers, however the fact that it took 11 seconds to start manually braking would suggest there was complete confusion and would most probably have been missed.
With the Boeing yes you could manually select spoilers, however the fact that it took 11 seconds to start manually braking would suggest there was complete confusion and would most probably have been missed.
Join Date: Aug 2005
Location: Cloudbase
Posts: 149
Likes: 0
Received 0 Likes
on
0 Posts
actuall, the fact that the spoilers didn't deploy was spotted rather early on.
I would assume, but that's just a guess, that those 11 seconds were spent troubleshooting the system (wonder why they didn't spot the TL then...)
A "spoilers not deployed - manual override" might be far easier than the troubleshoot.
But of course those inhibitors serve the task to protect from potentially more disastrous outcomes if the prerequisites aren't met...
pj
I would assume, but that's just a guess, that those 11 seconds were spent troubleshooting the system (wonder why they didn't spot the TL then...)
A "spoilers not deployed - manual override" might be far easier than the troubleshoot.
But of course those inhibitors serve the task to protect from potentially more disastrous outcomes if the prerequisites aren't met...
pj
StS,
I meant if it had been a Boeing. In my sim experience, people miss "no spoilers" in the Airbus everytime, whilst in the Boeing it occasionally was missed. With the circumstances, I believe that their fate was sealed as soon as only one TL was retarded. Boeing or Airbus the same result would have ensued.
I meant if it had been a Boeing. In my sim experience, people miss "no spoilers" in the Airbus everytime, whilst in the Boeing it occasionally was missed. With the circumstances, I believe that their fate was sealed as soon as only one TL was retarded. Boeing or Airbus the same result would have ensued.
Join Date: Jul 2002
Location: Confusio Helvetica
Posts: 311
Likes: 0
Received 0 Likes
on
0 Posts
Okay, to rephrase the logic question:
If <reverse thrust> should mean <committed to landing>, why not have the TRs follow similar logic as the spoilers, and only deploy if the TL was somewhere near IDLE?
TR not working as expected might cause a Tahiti-style overrun (or MDW), but not a 100-kts one.
Now, to something I know something about:
Short answer: not really, but you could squeeze it.
Ockham's razor is more or less an application of the Aristotelian principle of parsimony, and is often cited along the lines of "when multiple explanations are possible, the simplest one is to be preferred."
William of Ockham's formulation would be along the lines of "Do not posit in existence more entities than necessary to explain the phenomenon, except in divine matters."
The exclusion of things like God and the Trinity (exceptis divinis) is often brought up as a limit to Ockham's razor, but there is another limit:
Ockham is talking about scientific knowledge, involving reasoning from principles to conclusions. And, as Aristotle will say, the conclusions of a science are things that are true all the time, or for the most part (epi tou pollou). And they both would exclude from science the interactions created by human action (fortune) or the unpredictable natural interaction of systems (chance). In other words, for Ockham, you can analyze each slice of the Swiss cheese individually, but not how they line up. "Murphy's Law" and "Fate is the Hunter" are not expressions that Ockham would be familiar with (well, maybe the Principle of Plenitude, but that's a different story).
Accident investigation takes as its point of departure things that are the combination of fortune and chance, and which happen somewhere between very rarely and almost never. In just about every aviation accident, the simplest explanation (except for supernatural action) is Pilot Error. But, even if it is the most common cause of accidents, that doesn't make it automatically the cause.
(As an aside, this is why the conspiracy freaks love airliner accidents: they happen so rarely, there's a lot more room for the logical legerdemain that is the bread-and-butter of the black helo crowd.)
So, sure, here you could say, "leaving the TL in CLB caused the accident", but you would still need to respond to three basic questions:
A. Does the design contribute to the frequency with which pilots make this "fundamental mistake"? If I arrange the levers so that a PF who wants to set flaps for landing, sets the AP to GO instead, (or deploys spoilers), is this going to enhance the effects of error?
B. Did other circumstances contribute to making the mistake? ("Not even captain could make such a fine landing!", fatigue, stress, training, poorly written documentation)
C. Was the mistake sufficient for an accident? What role did a short, slippery, ungrooved runway play? Did it increase the attention of the crew to stopping the aircraft such that they did not have time to observe which condition was preventing the spoilers from activating? Were there any slips, skids or hydroplaning events, or did it just increase the braking distance?
If <reverse thrust> should mean <committed to landing>, why not have the TRs follow similar logic as the spoilers, and only deploy if the TL was somewhere near IDLE?
TR not working as expected might cause a Tahiti-style overrun (or MDW), but not a 100-kts one.
Now, to something I know something about:
Does Occam's Razor apply in aircraft investigation?
Ockham's razor is more or less an application of the Aristotelian principle of parsimony, and is often cited along the lines of "when multiple explanations are possible, the simplest one is to be preferred."
William of Ockham's formulation would be along the lines of "Do not posit in existence more entities than necessary to explain the phenomenon, except in divine matters."
The exclusion of things like God and the Trinity (exceptis divinis) is often brought up as a limit to Ockham's razor, but there is another limit:
Ockham is talking about scientific knowledge, involving reasoning from principles to conclusions. And, as Aristotle will say, the conclusions of a science are things that are true all the time, or for the most part (epi tou pollou). And they both would exclude from science the interactions created by human action (fortune) or the unpredictable natural interaction of systems (chance). In other words, for Ockham, you can analyze each slice of the Swiss cheese individually, but not how they line up. "Murphy's Law" and "Fate is the Hunter" are not expressions that Ockham would be familiar with (well, maybe the Principle of Plenitude, but that's a different story).
Accident investigation takes as its point of departure things that are the combination of fortune and chance, and which happen somewhere between very rarely and almost never. In just about every aviation accident, the simplest explanation (except for supernatural action) is Pilot Error. But, even if it is the most common cause of accidents, that doesn't make it automatically the cause.
(As an aside, this is why the conspiracy freaks love airliner accidents: they happen so rarely, there's a lot more room for the logical legerdemain that is the bread-and-butter of the black helo crowd.)
So, sure, here you could say, "leaving the TL in CLB caused the accident", but you would still need to respond to three basic questions:
A. Does the design contribute to the frequency with which pilots make this "fundamental mistake"? If I arrange the levers so that a PF who wants to set flaps for landing, sets the AP to GO instead, (or deploys spoilers), is this going to enhance the effects of error?
B. Did other circumstances contribute to making the mistake? ("Not even captain could make such a fine landing!", fatigue, stress, training, poorly written documentation)
C. Was the mistake sufficient for an accident? What role did a short, slippery, ungrooved runway play? Did it increase the attention of the crew to stopping the aircraft such that they did not have time to observe which condition was preventing the spoilers from activating? Were there any slips, skids or hydroplaning events, or did it just increase the braking distance?
Join Date: Mar 1999
Location: UK
Posts: 367
Likes: 0
Received 0 Likes
on
0 Posts
Guys
we now know that the Number 2 Thrust lever was left in the normal flight Climb detent, and not retarded. Like bmi baby our company has just issued the Airbus telex.as a Safety/Incident notice.
Airbus HAVE to allow the thrust levers to be split, forward and retarded, in order to cater for (pilot commanded) go arounds from any stage of flight. The question that must be answered is why they left it there. Many people have already banged on about normal airmanship being to close the throttles for landing, whatever the aircraft.
Whether the thrust levers move or not in normal operation is a red herring. Remember the Fokker 100 crash (http://desastresaereos.net/acidente_...te_o_laudo.htm Portuguese only, I'm afraid.)
Here the manufacturer had all moving throttles, as well as normal linked flying controls. This was not a dead stick aeroplane. In this accident a thrust reverser became unlocked and physically retarded the appropiate thrust lever. Left alone the aircraft would have climbed away. The crew however chose to advance the thrust lever with the tragic events that followed. The questtion again was why did what they did. Why did they fail to realise the consequence of what they were doing?
Here again we have a very (overall) experienced crew who strayed away from the norm. We have an idea of WHAT they did. But WHY did they do it.
Recently at the end of a long (A320) sim session we finished off with a single engine circuit, approach and single engine go around, virtually from the flare. I couldn't remeber which engine wasn't working (this time) so just advanced BOTH thrust levers to TOGA. Moving the dead one didn't matter. Not moving the good one did. Surely in training the same principle must be hammered home on landing. All thrust levers must be back at least to the idle detent on landing. Preferably Idle reverse. NO MATTER WHAT AIRCRAFT.
Sorry for shouting.
we now know that the Number 2 Thrust lever was left in the normal flight Climb detent, and not retarded. Like bmi baby our company has just issued the Airbus telex.as a Safety/Incident notice.
Airbus HAVE to allow the thrust levers to be split, forward and retarded, in order to cater for (pilot commanded) go arounds from any stage of flight. The question that must be answered is why they left it there. Many people have already banged on about normal airmanship being to close the throttles for landing, whatever the aircraft.
Whether the thrust levers move or not in normal operation is a red herring. Remember the Fokker 100 crash (http://desastresaereos.net/acidente_...te_o_laudo.htm Portuguese only, I'm afraid.)
Here the manufacturer had all moving throttles, as well as normal linked flying controls. This was not a dead stick aeroplane. In this accident a thrust reverser became unlocked and physically retarded the appropiate thrust lever. Left alone the aircraft would have climbed away. The crew however chose to advance the thrust lever with the tragic events that followed. The questtion again was why did what they did. Why did they fail to realise the consequence of what they were doing?
Here again we have a very (overall) experienced crew who strayed away from the norm. We have an idea of WHAT they did. But WHY did they do it.
Recently at the end of a long (A320) sim session we finished off with a single engine circuit, approach and single engine go around, virtually from the flare. I couldn't remeber which engine wasn't working (this time) so just advanced BOTH thrust levers to TOGA. Moving the dead one didn't matter. Not moving the good one did. Surely in training the same principle must be hammered home on landing. All thrust levers must be back at least to the idle detent on landing. Preferably Idle reverse. NO MATTER WHAT AIRCRAFT.
Sorry for shouting.
Join Date: Jan 2004
Location: LPPT
Age: 58
Posts: 431
Likes: 0
Received 0 Likes
on
0 Posts
A310:
I'll take the plane that the pilot can fly the wings off, burn the engines out, deploy all reversers inflight, all spoilers deploy when I say so and the like.
I don't need automatic spoilers or brakes and I will take the runway penalty for not having them.
BUT, I would demand that the pilots are super well trained, and have a full understanding of flight and the moral responsibility of being a pilot.
while the airbus might be more fuel efficient, sometimes you gotta bend some metal to save everyone's life.
you can provide information, but let me move the controls. (that's why the dc9 was known as the last PILOT's airliner)
I'll take the plane that the pilot can fly the wings off, burn the engines out, deploy all reversers inflight, all spoilers deploy when I say so and the like.
I don't need automatic spoilers or brakes and I will take the runway penalty for not having them.
BUT, I would demand that the pilots are super well trained, and have a full understanding of flight and the moral responsibility of being a pilot.
while the airbus might be more fuel efficient, sometimes you gotta bend some metal to save everyone's life.
you can provide information, but let me move the controls. (that's why the dc9 was known as the last PILOT's airliner)
How I would have designed it: if one engine is being reversed with the plane on the ground command the other one to idle if it’s not doing the very same. Then show the following ECAM message: “I’ve just put eng # to Idle cos’ you forgot to pull all levers back to Idle as per design – BTW I’ve just sent the pre-programmed TEA & BISCS email to management etc...”
If this kind of logic seems illogic …well provide a nice little red button anywhere on the panel that would read “DIRECT LAW” (or “Panic Button” for that matter) that would kill any automation on the bird.
Join Date: Jul 2007
Location: Germany
Posts: 556
Likes: 0
Received 0 Likes
on
0 Posts
Memory Items
As has been mentioned, "Loss Of Braking" is one of the very few memory items in A320 "Abnormal and Emergency" procedures.
These are the procedures that have to be performed immediately, without delay, and without referring to paper.
So the TAM crew shouldn't have spent time troubleshooting. It will be a case of intensive study by the investigators why it took the crew so long to begin the procedure, which starts with the obvious:
Brake pedals ................ PRESS
It appears that they did not think they had lost braking, since it never even started.
This will be another matter of looking into training at TAM, if the "Loss of Braking" event also incldes the event "Failure of onset of automatic braking".
How is that handled at other airlines? Is the "Loss of Braking" procedure meant to be applied whenever deceleration is below expected values? (i. e. DECEL light fails to illuminate?)
Another indication that their mind was not on braking is that they also neglected to follow the SOP for landing, which states:
BRAKES .......... AS RQRD
- Monitor the autobrake, if it is on. When required, brake with the pedals.
Does "on" in this case only mean "active", or does that include "armed"? The system description only talks of "arming" and "acitvation", but does not use the word "on".
For completeness: the other memory items are:
- Windshear
- Windshear ahead
- TCAS
- EGPWS 1)
- Beginning of Emergency Descent
- Beginning of Unreliable Speed Indication
1) The A320 accident in May 2006 near Sochi was another one where failure to promptly follow a "memory item" procedure was a causal factor, in that case the EGPWS "pull up" warning.
These are the procedures that have to be performed immediately, without delay, and without referring to paper.
So the TAM crew shouldn't have spent time troubleshooting. It will be a case of intensive study by the investigators why it took the crew so long to begin the procedure, which starts with the obvious:
Brake pedals ................ PRESS
It appears that they did not think they had lost braking, since it never even started.
This will be another matter of looking into training at TAM, if the "Loss of Braking" event also incldes the event "Failure of onset of automatic braking".
How is that handled at other airlines? Is the "Loss of Braking" procedure meant to be applied whenever deceleration is below expected values? (i. e. DECEL light fails to illuminate?)
Another indication that their mind was not on braking is that they also neglected to follow the SOP for landing, which states:
BRAKES .......... AS RQRD
- Monitor the autobrake, if it is on. When required, brake with the pedals.
Does "on" in this case only mean "active", or does that include "armed"? The system description only talks of "arming" and "acitvation", but does not use the word "on".
For completeness: the other memory items are:
- Windshear
- Windshear ahead
- TCAS
- EGPWS 1)
- Beginning of Emergency Descent
- Beginning of Unreliable Speed Indication
1) The A320 accident in May 2006 near Sochi was another one where failure to promptly follow a "memory item" procedure was a causal factor, in that case the EGPWS "pull up" warning.
Join Date: May 2004
Location: Toronto, Canada
Age: 73
Posts: 78
Likes: 0
Received 0 Likes
on
0 Posts
Not a technical observation, but close to on-topic I think: I flew TAM this weekend, GRU-IGU and IGU-CWB-GRU. Read the newly-printed August edition of the in-flight magazine. Near the front, full page RHS, a copy of the "Message from the Amaro family" [controlling shareholders] that you've seen already (at least on the TAM Web site, maybe published elsewhere). Interesting that there is an open acknowledgement to pax of the disaster, quite prominent- at least to those who actually read the magazine.
Join Date: Jun 2001
Location: East of the Sun & West of the Moon
Posts: 286
Likes: 0
Received 0 Likes
on
0 Posts
How I would have designed it: if one engine is being reversed with the plane on the ground command the other one to idle if it’s not doing the very same. Then show the following ECAM message: “I’ve just put eng # to Idle cos’ you forgot to pull all levers back to Idle as per design – BTW I’ve just sent the pre-programmed TEA & BISCS email to management etc...”
one engine is being reversed with the plane on the ground command the other one to idle if it’s not doing the very same + uncommanded reverser deployment after V1 on takeoff = YOU'RE DEAD!
It's not as easy as it might seem to define the correct logic for situations such as this.
ELAC
Paxing All Over The World
armchairpilot94116 those questions have all been asked in the course of this thread and the answers to many of them have also already been given. However the FULL answers will come from the investigation, the report of which will emerge in (typically) 12~24 months.
Join Date: Jan 2004
Location: LPPT
Age: 58
Posts: 431
Likes: 0
Received 0 Likes
on
0 Posts
ELAC
Ok, so let me complicate this logic a little.
Having in mind that Vref+something is not that far from V1 for a specific leg and usually a bit lower, why not INHIBIT that command above V1 (if this speed is entered on the FMGC as it normally should).
Simple. Bellow V1 uncommanded REV deployment would set the other engine to idle and the situation is treated as a RTO. OTOH on a situation of T/L left above idle on landing would let the plane do its magic. Either situation would trigger autobrakes and spoilers.
Ok, so let me complicate this logic a little.
Having in mind that Vref+something is not that far from V1 for a specific leg and usually a bit lower, why not INHIBIT that command above V1 (if this speed is entered on the FMGC as it normally should).
Simple. Bellow V1 uncommanded REV deployment would set the other engine to idle and the situation is treated as a RTO. OTOH on a situation of T/L left above idle on landing would let the plane do its magic. Either situation would trigger autobrakes and spoilers.
Last edited by GearDown&Locked; 6th Aug 2007 at 14:16. Reason: spell check
Join Date: Dec 2006
Location: washington,dc
Posts: 486
Likes: 0
Received 0 Likes
on
0 Posts
I asked a few dozen posts ago why two type rated experienced pilots wouldn't move the throttles to idle.
so far no real answers.
but now I bring up the concept of tunnel vision, brain fog, or whatever you might want to call it.
why did it take southwest/midway 18 seconds to get into reverse?
why did it take air france/toronto 17 seconds to get into reverse? (all times after touchdown).
I've seen excellent pilots, with more than 10,000 hours, some with 20,000 hours not bring throttles back in regular, non fly by wire airplanes.
Sometimes their hands were both on the control wheel in a very difficult x wind situation...and they asked for help with the throttles. Sometimes not.
BUT, there are two pilots who can reach the throttles.
So why did both pilots not bring the throttles back?
I don't know what you call this situation. But sometimes fatigue, or just a very difficult approach followed by the release of touchdown sometimes makes a pilot freeze.
Now, of course, they shouldn't, but we are human. (and there is still the chance that something with the machine wasn't right)
so far no real answers.
but now I bring up the concept of tunnel vision, brain fog, or whatever you might want to call it.
why did it take southwest/midway 18 seconds to get into reverse?
why did it take air france/toronto 17 seconds to get into reverse? (all times after touchdown).
I've seen excellent pilots, with more than 10,000 hours, some with 20,000 hours not bring throttles back in regular, non fly by wire airplanes.
Sometimes their hands were both on the control wheel in a very difficult x wind situation...and they asked for help with the throttles. Sometimes not.
BUT, there are two pilots who can reach the throttles.
So why did both pilots not bring the throttles back?
I don't know what you call this situation. But sometimes fatigue, or just a very difficult approach followed by the release of touchdown sometimes makes a pilot freeze.
Now, of course, they shouldn't, but we are human. (and there is still the chance that something with the machine wasn't right)
Join Date: Aug 2005
Location: Cloudbase
Posts: 149
Likes: 0
Received 0 Likes
on
0 Posts
BOAC,
of course an interesting if not the key question.
One thing I was wondering is this:
going from CLB to IDLE to REV is a series of state changes rather than true thrust commands.
Maybe, and this is wild speculation as I will admit, maybe they had briefed the final stage (CVR recorded "reverse only one") and they have mentally skipped the "IDLE" stage for whatever reason and went right into REV. Of course, they had REV only mentally linked with the #1 engine so they pulled that lever.
Very very very speculative, but the only way of tunnel vision I can think of at this point.
pj
of course an interesting if not the key question.
One thing I was wondering is this:
going from CLB to IDLE to REV is a series of state changes rather than true thrust commands.
Maybe, and this is wild speculation as I will admit, maybe they had briefed the final stage (CVR recorded "reverse only one") and they have mentally skipped the "IDLE" stage for whatever reason and went right into REV. Of course, they had REV only mentally linked with the #1 engine so they pulled that lever.
Very very very speculative, but the only way of tunnel vision I can think of at this point.
pj
Join Date: Jul 2007
Location: UK
Age: 59
Posts: 43
Likes: 0
Received 0 Likes
on
0 Posts
StS,
Can the TR latch be lifted with the TL at CLB? i.e. is one continuous movement possible from CLB to REV? However, The Airbus AIT as posted states "Before touchdown, the engine 1 throttle was retarded to idle."
Can the TR latch be lifted with the TL at CLB? i.e. is one continuous movement possible from CLB to REV? However, The Airbus AIT as posted states "Before touchdown, the engine 1 throttle was retarded to idle."