Go Back  PPRuNe Forums > Flight Deck Forums > Rumours & News
Reload this Page >

TAM A320 crash at Congonhas, Brazil

Rumours & News Reporting Points that may affect our jobs or lives as professional pilots. Also, items that may be of interest to professional pilots.

TAM A320 crash at Congonhas, Brazil

Old 29th Sep 2007, 18:00
  #2541 (permalink)  
 
Join Date: Sep 2007
Location: Switzerland
Age: 67
Posts: 62
Cool Statistics?

Hi,

I don't like the statistics.
Why?
The most dangerous place for the humans are the bed.
Statistically the bed is the place were the most humans die..
So ..never go in a bed !
Statistics are only numbers .

Salute.
NotPilotAtALL is offline  
Old 29th Sep 2007, 18:47
  #2542 (permalink)  
 
Join Date: Aug 2007
Location: Brazil
Age: 69
Posts: 131
Way to go, RWA, ChristiaanJ, armchairpilot94116 and NotPilotAtAll.

Statistics, for now, are useless. They will only show the accident didn't happen, but it did.
Why?
We don't know. But I believe that using simple methods like RWA described can lead to procedures or even software mods. Not all good improvements came from scientific research. Common sense is a good point to start.

And we all want to help, if we can...

Last edited by Rob21; 29th Sep 2007 at 21:15. Reason: typo/spelling
Rob21 is offline  
Old 29th Sep 2007, 19:36
  #2543 (permalink)  
 
Join Date: Aug 2007
Location: The Netherlands
Age: 64
Posts: 288
On post #2469, Causality

One factor that is definitively causal and that I keep missing in the shortlists of causal factors is the fact that all accident approaches were carried out with AutoThrust engaged untill touchdown.

It is relatively easy to understand that a hand makes a mistake at one critical moment of action (just half a second action time) when it has been sitting lame and passive for many minutes and if in the brain there are several ideas floating around about the action to be undertaken (do something not completely equal to the thrust levers because, ....).

An approach using manual thrust could not result in the same accident sequence, because, even if a thrust lever was left at approach TLA, that would only be 1.05 EPR and would not increase during the flare without active involvement of the pilot flying.
It is highly unlikely that all of a sudden, during the flare, one T/L would be left behind (forward, to be correct) when, for many minutes prior to that, they were actively moved together.

For the statistic sample, that would cut out a lot of landings - I estimate that 30 to 50 % of pilots always use manual thrust when they fly a manual approach.
EMIT is offline  
Old 29th Sep 2007, 20:11
  #2544 (permalink)  
 
Join Date: Aug 2007
Location: Brazil
Age: 69
Posts: 131
Talking about AutoThrust, I have two questions.

1 - When A/T disconnects, there is an ECAM note in amber and a chime (CRC?). This single shime shouldn't sound again while amber is still on the ECAM?

2 - On the CVR transcripts we (I) don't see this chime after touchdown. Why is that?

Rob
Rob21 is offline  
Old 29th Sep 2007, 20:17
  #2545 (permalink)  
 
Join Date: Jan 2005
Location: France
Posts: 2,316
RWA,
Weird... we were writing at the same time.
I quoted a possible scenario, and you were already quoting Potholer with a real identical case......
ChristiaanJ is offline  
Old 29th Sep 2007, 21:55
  #2546 (permalink)  
PJ2
 
Join Date: Mar 2003
Location: BC
Age: 73
Posts: 2,460
EMIT;

One factor that is definitively causal and that I keep missing in the shortlists of causal factors is the fact that all accident approaches were carried out with AutoThrust engaged untill touchdown.
Couple of items...

1. Re, "until touchdown"...do you mean just before touchdown or really "right until touchdown"? I ask because the Airbus SOP requires that the thrust levers (both) be reduced to IDLE by 20ft, (10ft on an autoland). Some long landings are caused by keeping the TL's in the CLB detent until touchdown among other inappropriate events.

2. I don't think you can use the words 'causal factor' when referring to the fact that autothrust was engaged 'at' touchdown on 'all accident approaches'. That might be close (though not exactly!) like saying the Nav lights were 'on' on 'all accident approaches'. The fact that the autothrust was on, on all accident approaches, cannot be cited as a primary 'cause' the accident.

But I think I know what you are saying - that there may be a relationship between the autothrust being on and the PF not being "engaged" or cognizant of the lever positions (and engine power being developed) in the same way that flying a manual thrust approach provides the PF.

This homes in on one Airbus controversy which many who otherwise love flying this type have commented on - that "manual flight", or "keeping one's hand in", is and remains a high priority for pilots.

Presently, our manual prohibits the use of manual thrust except in extraordinary circumstances such as a clear intent to "practice" in a benign ATC/operational environment. Many believe this to be a fundamental mistake and have fought such prohibition for years. We finally lost. Now, one disconnects the autothrust at one's peril should anything happen. Many crews used to fly approaches manually, including manual thrust levers, all the time and I personally taught it in instructional situations and encouraged it in regular line flying. No longer; the "risk" of blame is too high. In my view, competency with the machine is thereby lost however and, with such AOM restrictions, the relative loss of competency becomes self-fulfilling problem or a vicious circle. It is a big issue, as you likely know.

PJ2
PJ2 is offline  
Old 29th Sep 2007, 22:45
  #2547 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,182
Rob21:
I don't remember the date AI changed the procedure on handling TLs with a locked reverse. Once you left the affected engine TL at idle. Now, you bring both to reverse. What was always, it is not any more. OK, now is just as none of the engines had a reverse inop. But we have one. This is not the same landing. This is not like 30 million landings.
Fair enough, although the point I was making was that AI's standard procedures *always* called for *both* levers to be reduced to idle before selecting either one or both to reverse. Maybe that distinction could have been made clearer in training (as in emphasising *both* to idle), but I don't know.
DozyWannabe is offline  
Old 30th Sep 2007, 00:37
  #2548 (permalink)  
 
Join Date: Jul 2003
Location: An Island Province
Posts: 1,184
EMIT, PJ2, good points, and relevant to the quest of gaining an understanding as to why crews might forget or overlook checks and actions.
I support manual flight = manual thrust. Having hands-on provides a vital link with the brain, and even if the TLs are not moving (auto thrust) the hand/brain is positioned for the next activity, this contributes to situation awareness – projecting ahead, and can be a reminder ‘to do something’.

RWA Re our recent posts, I believe that we are in agreement.
Re your questions on the ‘human factor’ angle. Several researchers have shown that experience defined by flight hour has little relationship to the likelihood of experiencing error. Furthermore many accident scenarios identify simultaneous failures by both crew members (an observation on CRM and monitoring, or that both are experiencing the same situation?).
Several mainstream researchers relate human behaviour and thus the susceptibility to error, directly to the situation that they are experiencing; this also includes how the situation developed, the assumptions, biases, and external pressures (context) – including what James Reason refers to as latent factors.
Therefore an answer, or at least another view on your question (#2423), might come from understanding the situation from the pilot’s point of view. Factually this may be limited; however, with reasonable assumption (meaningful speculation) and avoiding hindsight bias, we should be able to identify some areas of interest (as above).

For those seeking statistics; if the number of TL ‘hang ups’ (errors) is quite high then the lack of associated incidents/accidents might tell us something about error management.
Most people have been looking at the technology end of the human-system interface; a few are now enquiring about the human aspects, but error management provides an opportunity to consider the actual link between the larger ‘system’ and the human and the many things which affect it, the situation, external pressures etc.
This view could be developed by considering that in the two most recent accidents the error management appears to have failed, if so why?
alf5071h is offline  
Old 30th Sep 2007, 04:08
  #2549 (permalink)  
RWA
 
Join Date: Sep 2007
Location: Melbourne
Posts: 180
Originally Posted by ChristiaanJ
RWA,
Weird... we were writing at the same time.
Absolutely, ChristiaanJ - thanks are due to BOAC for mentioning the Potholer post, I'd forgotten it. Can't help feeling that we're on the right track - 'gut feeling' only, though.

Couple more points for the 'picture' of what possibly happened:-

1. The investigators didn't release the full CVR transcript with the before-landing briefing. Very possibly that includes the captain deciding that, given the wet runway etc., he couldn't risk even the moderate amount of forward thrust which would result from putting No. 2 into reverse.

2. Both CVR transcripts (Taipei and Congonhas), and that Madeira landing video, show the 'Retard' call being heard two or three times before the pilots actually pulled the levers back. Presumably the call is triggered at or around say 20' Radio Altitude? Possibly it's timed a bit early - and sounds rather before the pilot's eye tells him that it's time to flare and retard? No big deal, 'one of those things' on the face of it - but if that's so, it's yet another reason why the 'Retard' call may be treated as a mere 'cue' rather than any sort of warning? Especially since it then cuts out.

And a question for A320 pilots. From the Madeira video, and what I know of the manual, and Eric Parkes' notes, I gather that when coming OUT of reverse you have to make sure that you push the levers right through the 'idle detent,' and then pull them back again, to ensure positive location IN the detent as opposed to just 'nearly' in it.

Is it the same when retarding to idle in the flare - that is, you have to pull the levers firmly back, against the resistance of the 'detent' mechanism, to make sure that they are firmly seated? How much 'feel' do you actually get? Is it, for example, the same sort of 'feel' you get when pulling a car's gearshift back through reverse and neutral into 'drive'?

Last edited by RWA; 30th Sep 2007 at 04:17. Reason: Typo
RWA is offline  
Old 30th Sep 2007, 08:03
  #2550 (permalink)  
PBL
 
Join Date: Sep 2000
Location: Bielefeld, Germany
Posts: 955
Originally Posted by PBL
* non-X: failure to retard all thrust levers to idle on landing
Originally Posted by ChristiaanJ
Sounds nice, but it has a glaring hole in it.
We're not interested in the "two known non-occurrences of X", we're interested in the real figure for "non-X".
Oh, dear. Clearly my elementary intro to statistics didn't go quite far enough. RWA and BOAC might like to listen in, also.

You are interested in an event, let us call it T.
T: Thrust levers not immediately reduced by PF to idle on landing; reduced by PM or by PF after PM notifies.

This is a different event from non-X. Why? Because there are no instances of (T & non-X). Why? Because their descriptions contradict each other. (There, I've been wanting to use that smiley for some time. But who would have thought in answer to ChristiaanJ?)

You're more than welcome to be interested in T. In particular, there might well be useful statistics about T across those with FDA programs, although it may be hard to tease them out.

NotPilotAtAll,

we may infer that not only do you not fly planes, but that you don't work anywhere near the procurement, building, or certification of safety-related systems of any sort, for which some facility with statistics is necessary, whether they are airplanes or not. Please try to make your interventions *thoughtful*.

PBL
PBL is offline  
Old 30th Sep 2007, 08:30
  #2551 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,581
Thank you PBL - received and understood. I note you have now introduced another event into the stats, namely 'T'. I am reasonably happy with my somewhat 'dated' working knowledge of stats, but I am more concerned with the practical rather than the theoretical points here. It is primarily 'T' in which I am interested and like many pilots here cannot understand, and which I fear statistics will do nothing to clear.

How you, your fellows and AI sort out 'non-x' in technical terms is of interest, of course, but the root problem lies in the mindset of day-to-day operation of this system and the training for such. 'T' must be eliminated as far as is humanly possible and assuming no aircraft or extra human system failure in CGH would have probably prevented this accident.
BOAC is offline  
Old 30th Sep 2007, 09:54
  #2552 (permalink)  
PBL
 
Join Date: Sep 2000
Location: Bielefeld, Germany
Posts: 955
Originally Posted by BOAC
I note you have now introduced another event into the stats, namely 'T'.
Not me, I just named it.

I agree wholeheartedly that T is of interest. However, I
don't think it's wise just to assume a correlation with non-X. There are lots of things which don't happen in two-person cockpits which may well happen in one-person cockpits. In a one-person cockpit, a T would likely become a non-X; in a two-person cockpit it is (by definition) not. That said, there are many reasons why T events would be of interest. I'm sure you know most of them but I think it's a good idea to ferret them out explicitly.

First, I want to split T and non-X events, which is what ChristiaanJ could have done to make his point (but I'm glad he didn't, for then he would have deprived me of the opportunity to use the smiley ).

P: PF fails to retard all thrust levers to idle on landing
Q: PM notices, draws attention to it
R. All thrust levers are then retarded

Here are the definitions of non-X and T in terms of P,Q,R.
* non-X = (P & not-Q & not-R)
* T = (P & Q & R)

The factor in common to both T and non-X events is P.

There is also another possible event to consider, another of those "how could anyone possibly do that?" events that we thought/think non-X is. This event is
* V = (P & Q and not-R)

Those who think that V is extremely unlikely are right. But V-like events have already occurred. In the Nagoya accident in 1994, the PF inadvertently triggered the TOGA switch. The PM (the CAP) notified him of that, and told him to turn it off. Which the PF didn't do and the CAP didn't insist. The PF was pushing the column for all he was worth; the AP was trimming against it, to full nose-up trim. When the CAP belatedly took over, the AC balloon-climbed, stalled, and landed tail first.

Let's get back to T events. First I observe that the common factor to both T events and non-X events is P. P is also common to V events, of which there haven't been any.

First, a P event is a necessary causal precursor of a non-X event. Thus there is an opportunity for prophylaxis of non-X events: stop all P's, you necessarily stop all non-X's.

Second, there may be many P events, so one might be able to calculate some useful statistics, unlike with non-X events.

Third, a P event is also an instance of failure of the "Retard" warning. The PF is to retard: heshe failed to do so in a P event despite the warning.

The AI argument I labelled no. (1), produced by Malinge on 9 August, can be countered using this third point. The argument was that the "Taipei" warning was redundant, because the "Retard" warning already advised that thrust should be reduced to idle. If there are relatively many P events in comparison with the 3 non-X events, we would have relatively many instances in which the "Retard" warning has failed, thus suggesting one indeed needs a different means of advising that thrust has not been reduced to idle, contrary to what AI said they concluded with the regulators.

Whether that should be a "Taipei" warning in addition to "Retard", or an altered form of warning altogether, does not follow from this, but it has already been decided by H2F3 Revision 3 so this consideration is moot.

The downside of this may be that if AI realises that P events are likely to weaken their argument, they may be less motivated to collect T-event statistics from their fleet operators until after any Congonhas legal proceedings have finished.

BTW, AI won the civil case associated with the V-like event at Nagoya.

PBL

Last edited by PBL; 30th Sep 2007 at 10:19. Reason: Hoisted on my own petard
PBL is offline  
Old 30th Sep 2007, 11:11
  #2553 (permalink)  
 
Join Date: Jul 2007
Location: Germany
Posts: 550
Is it the same when retarding to idle in the flare - that is, you have to pull the levers firmly back, against the resistance of the 'detent' mechanism, to make sure that they are firmly seated?
No.

Idle has no detent. It is merely a stop, although it is sometimes referred to as a detent, even in the FCOM. Look at the schematics posted here a while back of the entire thrust lever mechanism. The detents also have no resistance getting into, only getting out of.

The non-detent nature of the IDLE stop is also the reason for the observed behaviour coming out of reverse. Going slightly beyond (forward) idle, and pulling the levers firmly back to the stop. This has also been quoted from some training procedures quite early in the thread.

The reason for pulling the levers back quickly is to cut thrust quickly to avoid a prolonged float before touchdown.


Bernd
bsieker is offline  
Old 30th Sep 2007, 13:28
  #2554 (permalink)  
 
Join Date: Sep 2007
Location: Portugal
Posts: 39
After seeing the pilot's photos posted here, I finaly knew that they aren't the ones I flew with a couple of weeks before the accident.

This discussion is very complex. I think the pilots here can enlighten the non-pilots on some airmanship concepts. I think an important concept here, that's beeing overlooked, is that flight progresses through "gateways". In final approach, the primary gateway is to stabilize the approach in the desired profile, speed and configuration. The following gateway is, if no go around is performed, to touch down, and retarding the throttles is part of the process to this gateway. The next gateway is to control and deccelerate the a/c on the rwy, and reverse thrust is part of this gateway. So retarding and reverse, although normaly acomplished in sequence, are not part of the same process, in the mind set of a pilot.

Introducing another RED WARNING in this stage of flight is probbably a bad strategy, considering that it's procedural to overlook temporarily the warnings, below 400 feet AGL, and postpone actions untill the landing is acomplished. So this so much discussed warning, would probably be overlooked for a few seconds, after touchdown, untill the a/c was controlled, so it would miss it's objective, both pilots attention could easily be "distracted" by the task of the landing itself. These few seconds are a lot of time in a landing.

If AI wants to introduce something usefull, perhaps they should design a way that, if one thrust reverser is deployed, the other engine is automaticaly commanded to idle. Care should be taken, so that in case of uncommanded TR deployement airborne, this idle action shouldn't occur, based in a/c air-ground logic. Although all this software stuff still gives me the creeps.

Don't forget that in aviation, the "KISS" rule strongly applies: KEEP IT SIMPLE STUPID

Regards
3Ten is offline  
Old 30th Sep 2007, 13:44
  #2555 (permalink)  
 
Join Date: Jul 2007
Location: Germany
Posts: 550
3-10,

thanks for your description of the idea of "gateways",

Originally Posted by 3Ten
Introducing another RED WARNING in this stage of flight is probbably a bad strategy, considering that it's procedural to overlook temporarily the warnings, below 400 feet AGL, and postpone actions untill the landing is acomplished. So this so much discussed warning, would probably be overlooked for a few seconds, after touchdown, untill the a/c was controlled, so it would miss it's objective, both pilots attention could easily be "distracted" by the task of the landing itself. These few seconds are a lot of time in a landing.
I don't know about procedures at different airlines, but in the FCOM a Level 3 Warning, such as the warning we're talking about (including CRC, Master WARN light and Red ECAM text), means:

Originally Posted by A320 FCOM 1.31.10, P2
Red Warning : The configuration, or failure requires immediate action :
- Aircraft in dangerous configuration, or limit flight conditions (eg: stall, o/speed)
- System failure altering flight safety (eg: Eng fire, excess cabin alt)
Unlike Level 2 "Amber Caution", where action should be taken as soon as time and situation permit.

Other pilots have agreed that they think another Red Warning in this case is a bad idea, but is it really because they tend to be deferred? Maybe the FCOM needs some serious rewriting, then.


Bernd
bsieker is offline  
Old 30th Sep 2007, 13:49
  #2556 (permalink)  
The Reverend
 
Join Date: Oct 1999
Location: Sydney,NSW,Australia
Posts: 2,020
Has anybody ever seen statistics of motor accidents where disorientated drivers stomp on the accelerator pedal instead of the brakes?
HotDog is offline  
Old 30th Sep 2007, 14:59
  #2557 (permalink)  
PBL
 
Join Date: Sep 2000
Location: Bielefeld, Germany
Posts: 955
Originally Posted by HotDog
Has anybody ever seen statistics of motor accidents where disorientated drivers stomp on the accelerator pedal instead of the brakes?
Exactly how would such statistics be collected? If they were somehow, what would lead you to think that they would be reliable?

PBL
PBL is offline  
Old 30th Sep 2007, 15:20
  #2558 (permalink)  
 
Join Date: Aug 2007
Location: The Netherlands
Age: 64
Posts: 288
To PJ2, post #2484

Re your point 1 – I mean the normal procedure for use of A/T, that is to leave it engaged until the flare and to reduce the T/L’s to idle at the appropriate moment.
I used the term “until touchdown” to distinguish this from an approach where A/T is disconnected, let’s say at 1.200 ft. To pilots, disconnecting at 1.200 ft would constitute a landing with manual thrust, whereas many of our interested bystanders might call it an approach with autothrust.
With regards to SOP’s, I would say that thrust has to be reduced at the appropriate moment – to aid the pilot in this, the aircraft provides a reminder (RETARD) at 20 ft (autopilot off) or 10 ft (A/P on). Leaving the T/L’s in CLB detent too long certainly is inappropriate, because A/T then starts to increase thrust during the flare manoeuvre. When using manual thrust though, it may well happen that the T/L’s hit the idle stop just as the wheels touch the ground, it very much depends on airspeed and wind situation during the flare – on some Greek islands for instance, you can have pretty weird wind effects at very short finals due to wind / obstacle interaction.

Re 2. The autothrust engaged would be a “required factor” for this accident sequence - without A/T, there is no auto spool-up of thrust. With manual thrust, a manual increase of one T/L to CLB position is very unlikely. Whether it is then allowed to call A/T a “causal factor” is more a linguistic matter. The important point is, without A/T engaged, the 3 accidents would not have happened in this way – of course one could interject that, without A/T, who knows, they would have stalled in the turn to final. As for the comparison with, e.g. the NAV lights being ON or OFF, or the aircraft tails being yellow or red, A/T being engaged or not, is vastly more relevant.


What you say about your company’s SOP’s is worrisome – to prohibit the use of manual thrust is crazy. I am glad that in my company, that choice is left to the discretion of the pilot. An advice that A/T is a wonderful thing, a recommendation to use it in manual flight as well, certainly, but the choice is with the pilot!
EMIT is offline  
Old 30th Sep 2007, 15:43
  #2559 (permalink)  
 
Join Date: Aug 2005
Location: Albuquerque USA
Posts: 171
automobile accelerator brake confusion

Has anybody ever seen statistics of motor accidents where disorientated drivers stomp on the accelerator pedal instead of the brakes?
There was much prominence given in the United States in the middle 1980s to incidents in which drivers asserted that their cars accelerated wildly while the driver was "standing on the brakes".

A feast of injury lawyers and "investigative journalists" pushed particularly the theory that an Audi model was flawed, with an attendant plummet in sales and resale values.

As anyone familiar with the relative capabilities of passenger car brakes and engines would assume, these cases almost certainly represented driver confusion of the sort you specify. The most common venue was in parking lots and driveways, not the open road, as people most often made this mistake on first disengaging "PARK" in an automatic transmission car.

Provision of an interlock which required brake pedal depression in order to disengage PARK removed much of this class of event.

True, there was nothing wrong with the cars, but there was an achievable improvement.
archae86 is offline  
Old 30th Sep 2007, 15:45
  #2560 (permalink)  
 
Join Date: Jul 2007
Location: Germany
Posts: 550
EMIT,

thanks for your reply. I hope you have been able to sort out your copy/paste problems.

I think you are right that the choice to use A/THR for this approach is a necessary causal factor, and I shall add it to the graph. It will be an NCF at least for node (19), and perhaps, possibly with interims nodes, to (42) as well.

It will be very similar to (23).


Bernd
bsieker is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information -

Copyright © 2021 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.