TAM A320 crash at Congonhas, Brazil

[PBL]

Originally Posted by flyingnewbie10

The relatively short length (it is really short and has no escape area) is longer than the one for which the A/C is certified to Land..............
So I would prioritise not reducing thrust (as a factor) and that would be the main (or the real) cause IMHO.

So, OK, there is a prioritisation with an explicit selection criterion. The selection criterion is that a causal factor is prioritised because of what's on a piece of paper.

Isn't that rather a long way from the physical happenings, though? Had all the regulating authorities said the day after the accident "we are recertifying the aircraft only to land on runways longer than 2,400 m LDA", wouldn't the prioritisation criterion have disappeared?

Originally Posted by RWA

With respect, PBL, I did tell you how I prioritise not reducing thrust, by referring to it as the 'primary cause.' By definition, there can only be ONE primary cause.

To me, selecting a specific factor as "primary", and prioritising it, are two ways of saying the same thing. I could as well have asked what your selection criteria are for picking this factor as "primary".

You have partly answered it, I think, in saying

Originally Posted by RWA

But they are not the features of it which deserve the most urgent attention, from the point of view of avoiding any recurrences.

In other words, you choose as "primary" a cause whose prophylaxis deserves the "most urgent attention".

Others seem to think that the factor which deserves the most urgent attention is the state of the runway at Congonhas and the circumstances under which one may land there. This was so even before the accident (I refer to recent comments by marciovp on the regulatory to-and-fro, in which under a certain order the landing with one REV INOP would not have been permitted by the regulatory authority).

So this way of prioritising the causal factors seems to depend highly on one's social beliefs and priorities, not on any objective physical feature of the accident itself. And since people have different social beliefs and priorities, they will choose their "primary" cause differently according to those different beliefs. And then we will have lots of different "primary" causal factors, each one as valid as the next.

But, as you say, your selection of the word "primary" is intended to say that there is just one of them, not that there are lots of equally valid "primary" causes, depending on people's social inclinations. So it looks as if your implicit selection criterion isn't going to fulfil the requirement you make ("by definition") on the notion of "primary cause".

On another matter, concerning the warning mod, the situation is this. TAM has claimed that there is a warning mod incorporated in FWC release H2F3 and they are going to install H2F3. However, one of Airbus's chief safety persons, Yannick Malinge, said before the CIP, if I understand correctly, that the mod to which the Taiwanese report referred had not been implemented because regulatory authorities had reservations about its necessity.
And the H2F3 SB, which describes new features in the release, contains not one word about such a warning. I don't think we are anywhere near a final coherent story about the mod yet.

PBL

[bsieker]

Originally Posted by I-FORD

There have been other events of runway excursion involving A320 in a single reverse configuration.
Same airplane, different runways.
To me the airplane gets priority.

PBL has repeatedly drummed upon (the impossibility of) prioritising root causes.

Just a few simple questions on your professed priority argument:

- How many incidents have there been in which this particular thust-lever handling mistake was a causal factor?

- How many runway excursions have there been, in which a short runway was a causal factor? (It doesn't matter if you call it "primary cause" or "contributing factor", if the accident had been prevented by its absence, it is a necessary causal factor.)

- How many runway excursions have there been on long (2.5+ km?) runways?

So on what basis do you prioritise the thrust lever handling mistake over a short runway?

I believe in the end, the arguement for concentrating on the aircraft is a political one. It seems to much easier to "fix the aircraft", or blame people, than to extend a runway in a densely populated area.


Bernd

[Lemurian]

Flyingnewbie

Should we discard a below average braking system effectiveness ? One of the last maintenance reports for the PR-MBK, if I am not wrong, is related to problems in the braking system.

That maximum brake pedal application is related to *some* deceleration is proof enough that the brakes were working. That the deceleration was so slow is an indication of low friction and the anti-skid system kicking-in to prevent tyre lock. (the illustration of the results are to be constated on the TPE graphs that are more complete and show that with full brakes, the hydraulic pressure applied to the discs oscillates between 1800+ psi and 200). If that is not the indication of sliding wheels the antis-skid tends to prevent, hence a *contaminated* runway, I don't know what is.
Back to the CGH graphs, by time tag :30, a definite deceleration can be seen and is related to the #1 reverser -and to some extent aerodynamic drag- only.
So, the total *thrust budget* is negative, the reverse thrust more than balances the #2 engine thrust.
Note : It is apparent that this discussion is turning on its head. Old arguments keep creeping back and new points, especially bsieker's WBA study, are left un-answered.
What is needed is a sum-up of the previously covered aspects.
Me, I'm a bit tired of using the same arguments I used a month ago for newcomers.
Regards
Lemurian.

[HotDog]

IMHO, what is needed now, is the final investigation report. The A320 in this thread has now been redesigned, rebuilt several times and Airbus SOPs amended a dozen times by people with no knowledge or experience of A320 operations. If Airbus SOPs are deficient in this area, I'm sure Airbus would have informed all A320 operators by now; so give it a rest and read the entire contributions to this thread before suggesting a repetitive theory and how to fix it.

[stickyb]

WBA Graph

a question for bsieker or pbl. Please excuse if not a sensible question.

Does the non implememtation of the recommendation from the Taiwan incident deserve a specific mention in the WBA?

[GMDS]

If Airbus SOPs are deficient in this area, I'm sure Airbus would have informed all A320 operators by now; so give it a rest

Sarcasm in:

... and face all the lawsuits of relatives?? You must still believe in Santa.
We can almost certainly assume that it will be neither the aircraft, nor the company sop's and not the runway ... but only the pilots. It's the most convenient and cheapest solution and thats why we're still up there, even if we will no longer be able to identify and correct any failures, glitches, automations, bad sop's or (too) short runways.

Sarcasm out.

Enjoy you week-end folks.
GMDS

[lomapaseo]

Sarcasm in:
... and face all the lawsuits of relatives?? You must still believe in Santa.
We can almost certainly assume that it will be neither the aircraft, nor the company sop's and not the runway ... but only the pilots. It's the most convenient and cheapest solution and thats why we're still up there, even if we will no longer be able to identify and correct any failures, glitches, automations, bad sop's or (too) short runways.
Sarcasm out.

You seem to be trying to make outlandish statements while hiding behind the sarcasm banner.
I wouldn't wait for the official investigation nor the regulator to issue recommendations for prevention folllowing an accident. The quickest response almost always comes from the designer/manufacturer even if it's a stop gap measure and all the facts are not in yet. Knowledge and understanding are always prefered over distrust and prejudice.
happy flying

[GMDS]

No lomo. I am just beeing sarcastic. Not less, not more.

[RWA]

Originally Posted by bsieker

- How many incidents have there been in which this particular thust-lever handling mistake was a causal factor?

Besides Congonhas, three incidents, all involving runway excursions:-

Taipei - swerved off 9,200-foot runway 300 feet from the end and finished up in a ditch.

Phoenix - swerved off 11,500-foot runway to right. I don't know how far along it the aeroplane got, sorry.

Bacolod - swerved off 5,932-foot runway, then back on again - finally ran off the end, across a river, and into a village.

I don't see any common factors so far as runway length is concerned - looks as if, if the asymmetric thrust causes you to swerve off it anyway, the length is immaterial.

Originally Posted by bsieker

I believe in the end, the arguement for concentrating on the aircraft is a political one. It seems to much easier to "fix the aircraft", or blame people, than to extend a runway in a densely populated area.

Frankly mystified by that comment? What is 'political' about it? Don't you think that the Taiwanese suggestion of a simple warning system covering BOTH levers has some merit? Wouldn't cost much, wouldn't take long, and might very well end the problem once and for all?

[Dani]

HotDog, thank you so much! I recommended this 500 posts and two months before.

Please close this thread!

Moderator, the check please!

I thank every person involved, especially the smart guys from Uni Bielefeld, and all engineers and Airbus pilots around the globe. It was a pleasure to discuss with you. On another occasion, maybe...

Dani

[bsieker]

Originally Posted by stickyb

a question for bsieker or pbl. Please excuse if not a sensible question.

Does the non implememtation of the recommendation from the Taiwan incident deserve a specific mention in the WBA?

It is a good question and maybe deserves a note in the introductory comments.

We have given it some thought, but there are a number of reasons why it is not included:

- Information about the addition of a thrust lever warning in the H2F3 FWC version is almost unavailable, and at best circumstantial. The official description of this FWC update does not mention it. Apparently the authorities and/or operators were not happy with adding another bell and whistle to the system, and it is doubtful if it is installed in any aircraft. Maybe someone could verify or falsify this.

- The inclusion of such "Un-Events" follows some guidelines, mainly that it has to be a clear omission of an action or system feature that should have been performed or installed, but wasn't. Clear indications of this are if the omitted action was (part of) a SOP (such as pulling back both thrust levers), a legal requirement, a system upgrade mandated by an AD, an action that was performed in all previous comparable situations, etc. This FWC update was not mandatory, so the case for including it is not strong.

This guideline, serves to keep analysts from including all sorts of magical/hypthetical "save-me" devices in the graph. Without it one could imagine adding "No manual ground spoiler deployment facility" or "Aircraft does not idle other engine when one reverser is deployed". This is not helpful for the initial analysis.

- it is unclear that an additional warning would have gotten through to the crew in the given situation. Even if the condition for its inclusion above would have been fulfilled, it would have remained an assumption.

Thanks for your comment, good to see that some more people take a look at my analysis. I hope that answers your question.


Bernd

[bsieker]

Originally Posted by bsieker

- How many incidents have there been in which this particular thust-lever handling mistake was a causal factor?

Originally Posted by RWA

Besides Congonhas, three incidents, all involving runway excursions:-

[...]

Phoenix - swerved off 11,500-foot runway to right. I don't know how far along it the aeroplane got, sorry.

Phoenix was different. Both thrust levers were first retarded to idle, only later was one "knocked up" again ...

You still haven't answered my other two questions, to see that short runways cause far more accidents than this kind of thrust lever mishandling. Which leads your criterion for prioritising aircraft-related causes ad absurdum.

If you say "But the aircraft is easier to fix", then this might be a criterion, but it is what I call political. Not looking at what will potentially prevent the most future accidents, but what will be easier to do, and cheaper. And in this case, not involve anything about the country in which the accident happened, but only foreign manufacturers.


Bernd

[RWA]

"Airbus has developed a specific warning when one throttle is set to reverse while the other is above idle. This warning generates an ECAM warning "ENG x THR LEVER ABV IDLE", a continuous repetitive chime (CRC), and lights the red master warning light. This new warning is implemented in the FWC standard "H2F3".

"A Service bulletin will be issued very soon on this subject."

Why did Airbus design and develop it if it was so inadvisable, Bernd?

I honestly can't see what harm it could possibly do. On the other hand, I can see how it might do a great deal of good. We only have fragmentary evidence from the CVR and FDR, but what evidence there is suggests that the pilots never actually knew what was wrong; their effort to get the spoilers to extend by cycling the switch strongly suggests that they thought they had a spoiler malfunction.

The Master Warning Light going off, with a clear and specific warning, might very well have saved a lot of lives.

In my view the Taiwan Report points to a clear area where the design could be improved and made more 'fail-safe,', in that currently, even if only one throttle lever is actually retarded, the 'Retard' warning cuts out. An even more straightforward mod. would be just to have the warning stay on until both levers are retarded; but the Master Warning is probably the better approach.

Tony

PS Sorry, crossed with your next post. You're surely not suggesting that we extend all runways at all airports to Heathrow or Kennedy standards? If a given aeroplane needs a longer runway it's certification should presumably be changed? What we are looking at here are four incidents involving the same (very unusual) mistake being made within quite a short timescale in a single marque of aeroplane. Surely a revised warning system is well worth trying?

[flyingnewbie10]

What is needed is a sum-up of the previously covered aspects.

I aggree.


One thing (I think this is correct): More Airspeed, more Drag. So the lack of "positive thrust" could be explained by the higher drag (or higher airspeed) around time tag 30.

I ask anyone (maybe you know) what is the correlation between N1 and thrust in KN (is it specific to the engine - an IAE V2527-A ?).

[barit1]

Let's think a moment about landing length required per AFM (certification) standards. We all agree that many airports could use more runway, not because they don't meet/exceed AFM requirements for a particular type, but because more r/w (or overrun) is another layer of swiss cheese to help keep the holes from aligning. I'm not sure this would have prevented CGH from happening, but it could have saved lives.

If you predicate an operation on everything "just right", then toss in a slick ungrooved runway, and a locked-out reverser, and a Boeing captain transitioning to a new type...

[IFixPlanes]

HotDog, thank you so much! I recommended this 500 posts and two months before.

Please close this thread!

Moderator, the check please!

I thank every person involved, especially the smart guys from Uni Bielefeld, and all engineers and Airbus pilots around the globe. It was a pleasure to discuss with you. On another occasion, maybe...

Dani

I think in the same direction.
But i must exclude the Uni Bielefeld. (sorry guys)
Studying something ist not the same as knowing something.
Some conclusions are absolutely wrong.
I unsubscribe from this Thread...

[bsieker]

Originally Posted by IFixPlanes

I think in the same direction.
But i must exclude the Uni Bielefeld. (sorry guys)
Studying something ist not the same as knowing something.
Some conclusions are absolutely wrong.

Since I have been attacked in public, I will answer to this, even if IFixPlanes will not read it.

After I published my analysis, in which so far no large errors have been pointed out by highly experienced Pilots and other interested parties, IFixPlanes wrote a private message to PBL, complaining about a big error in the analysis.

For reasons unknown he chose to keep it private, and we respected that. But attacking our analysis now by simply saying he knows better, and we do not know, without publicly discussing the points, seems quite odd.


Bernd

[Clarence Oveur]

It seems to me that some people here are very impressed by themselves. Perhaps they should be, but it seems to affect their opinion of others. "1 up, 2 down" I believe it's called. I will no doubt be corrected if I'm wrong.

As for publishing Private Messages, I am not impressed.

[bsieker]

Why did Airbus design and develop it if it was so inadvisable, Bernd?

I don't know.

It may surprise some people but I do not work for, and I am not paid by, AI.

I thought it might be a good idea, but the information on it, apart from what you quoted, is sketchy. And PBL tried quite hard to find something more substantial.

In my view the Taiwan Report points to a clear area where the design could be improved and made more 'fail-safe,', in that currently, even if only one throttle lever is actually retarded, the 'Retard' warning cuts out. An even more straightforward mod. would be just to have the warning stay on until both levers are retarded; but the Master Warning is probably the better approach.

Yes, I think it was Lemurian who was strongly in favour of this one, augmented by the artificial voice pointing to the respective thrust lever: "RETARD TWO!" But it's really not my call to make.

PS Sorry, crossed with your next post. You're surely not suggesting that we extend all runways at all airports to Heathrow or Kennedy standards? If a given aeroplane needs a longer runway it's certification should presumably be changed? What we are looking at here are four incidents involving the same (very unusual) mistake being made within quite a short timescale in a single marque of aeroplane. Surely a revised warning system is well worth trying?

I agree. I am not trying to prioritise causal factors, so training, warning systems, runways, certifications, should all be examined.

I was trying to make the point that the number of accidents of a type does not, specifically not in this case, point to the aircraft when looking for a "primary" cause. So even seemingly similar criteria for prioritising causes may lead to different results, when done by different people.

So my point was trying to show that one should not prioritise.

I rest my case.


Bernd

[Rob21]

TLA x EPR

I noticed on the graphics (FDR) that there is not a "consistency" between TLA and EPR. I am talking about engine #1 (the good one) TLA x EPR, after touchdown on the final landing and also on the landing at POA.

My questions are (considering only engine #1 / reverse operating):

1- EPR shouldn't be a result of TLA? (after touchdown)
1a- On the last landing, engine #1 TLA = -22,5º constant, against an EPR variation between 0.97 to 1.02.
1b- On the landing at Porto Alegre, engine #1 TLA = -22,5º (during ten seconds), against an EPR variation between 0.97 to 1.07.

2- Why EPR varies so much (0.97 to 1.07) given the same TLA (-22,5º)?

3- Why at the last landing, EPR on engine #1 didn't go over 1.02, and at Porto Alegre EPR got to 1.07 given the same TLA (-22,5º)?

I am not talking about engine performance (density altitude), I am just wondering if EPR shouldn't be a result of a given TLA...