TAM A320 crash at Congonhas, Brazil

[PBL]

Originally Posted by bubbers44

...SWA [Chicago Midway]...The computer wouldn't allow ground spoilers or auto brakes and questionable thrust control on this landing and it is also pilot error?

What computer would that be? This was a FLUF.

PBL

[bubbers44]

FUBAR I understand, FLUF, I don't.

[BOAC]

PBL - you are more than welcome to contradict both me and the NTSB.

[Lemurian]

FUBAR I understand, FLUF, I don't.

Fat little ugly f..ellow, given to the original 737, probably reminding the family inheritance of the BUFF !
It stuck to the rest of the "classics" and the NGs.

[armchairpilot94116]

Running into EMAS is still an over run and therefore an accident/incident. In BOAC's camp on this one.


edit: recognizing though that a too short (for that landing ) runway overrun accident does make the "shortness of the runway" a possible contributing factor and possible causal factor as well.

[DozyWannabe]

PBL:
Bubbers44's second sentence (below) referred to TAM, his first referred to SWA.

bubbers44:
The computer wouldn't allow ground spoilers or auto brakes and questionable thrust control on this landing and it is also pilot error?

The computer 'wouldn't allow'* ground spoilers or autobrakes because in the likeliest scenario, the pilots failed to configure the aircraft correctly for landing. The thrust settings according to the FDR trace are completely consistent with the A/THR remaining in SPEED mode - no 'questionable' about it.

Regardless of whether you're flying an A320, 747, F-16 or a Sopwith bleeding Camel, you retard all throttle controls in the flare - and if you don't you're in trouble, especially when a short field is involved. The technology has bugger all to do with it.

* - I don't like this turn of phrase at all, smacks of the misinformation about the A320 system being there to inhibit pilots again.

[bsieker]

How many Camels have you spotted so far?

[...] or a Sopwith bleeding Camel, you retard all throttle controls in the flare [...]

Except of course that the Sopwith Camel didn't have a throttle. You had to turn the ignition on and off for landing. Unless you had the Gnome engine, which had a special switch to turn ignitiion off on all cylinders but one.



Bernd

[Right Way Up]

The computer 'wouldn't allow'* ground spoilers or autobrakes

This is a bit like like saying you tried to pull a door open but forgot to twist the handle. Those damn door handles are always stopping us from doing what we want the door to do!

[DozyWannabe]

Bernd:

Picky, picky picky...

[PBL]

Originally Posted by armchairpilot94116

Running into EMAS is still an over run and therefore an accident/incident. In BOAC's camp on this one.

Then check out the definition of "Aircraft Accident" on p2 of FAA Order 8020.16, at
http://www.faa.gov/airports_airtraff.../media/AAI.pdf
Designation as an accident requires death or serious injury to at least one person, and/or substantial damage to the aircraft. What constitutes "substantial damage" is determined a posteriori. For example, the pictures of the wrecked wheels on the United AC that landed with parking brake set, referred to on the recent BMI/parking brake thread near the beginning, do not count as "substantial damage". Neither did the incident with the A320 that landed at LAX with the nose wheel at 90° to the runway (videos on YouTube). I doubt an EMAS arrestor incident would could as an accident per se.

It doesn't suffice just to have an "opinion" on something. Where there are documents, precedents and protocols and substantial engineering knowledge it is unhelpful to discussion to just go around making up one's own meanings for terms.

PBL

[BOAC]

PBL - those of us involved in the PRACTICAL side of aircraft operation do not need book references to know that referring to EMAS in terms of preventing an 'accident' in the prevailing conditions at CGH, bearing in mind the terrain and energy levels involved, is farcical.

I expect your work is stunningly clever and relevant to theorists but I do not, at the moment, see it having much bearing on the aftermath of this accident. Even with any EMAS that could possibly have been installed at CGH there would have been severe injury and aircraft damage in this accident, thereby satisfying your definition - and the poster did say "accident/incident", by the way.

No, we don't need a definition of an incident.

[GlueBall]

EMAS would not have affected this accident as the A320 had departed to the left side of the runway, not the end overrun where EMAS is typically installed.

[PBL]

BOAC,

you are misreading me. I wonder, wilfully?

I objected to you using a "primary cause/contributing factor" distinction to contradict a point I made, and I also objected to you using a common but confused notion of "accident" to the same end. That is neither THEORY nor PRACTICE but SOPHISTRY. And how you get from there to read me as apparently suggesting that an EMAS would have stopped the TAM aircraft is neither THEORY nor PRACTICE but FANTASY.

So there

PBL

[3Ten]

Accident (pilot error related):

There's allways someone pouring gasoline on the ground. Sometimes there's lots pouring gasoline on the ground. These are contributing factors, casual factors and all that jazz.

Many times pilots have the zippo in their hands. When pilots light the zippo and throw it over those gallons of gas, we have pilot error primary cause. When pilots decide to throw the zippo in the pocket, we probbably have a non-accident. If pilots manage to keep away large quantities of gasoline, we have a non-event.

For one accident, we have hundreds of non-accidents, and thousands of non-events. Non-accidents may be reported, but often they go unreported (unfortunately), and many times even unnoticed.

Primary cause is the spark, the one that if didn't occur the accident wouldn't happen. That's probbably why pilot error is so commonly the primary cause, because pilots are there. Because seating in front of this computer doesn't allow me to act on a primary cause.

If our TAM coleagues acctualy failed to retard #2 throttle, that was in fact primary cause.

[RWA]

Originally Posted by Rob21

learned here that there are two conditons when "Retard" warning will cease:

1. Both TLs at idle

2. At least one TL at reverse.

If this is correct, why "Retard" warning ceases before condition 2 is met?
(per CVR @ 18:48:25.5)

Thanks for the question, Rob - cost me a bit of researching time, but it's actually uncovered something of a new angle. Before I explain, apologies all round; I usually make an effort to keep my posts short, but this one has to be a bit longer than usual.

At first I thought that the Philippines investigators might have got it wrong; or that the system logic might have been changed since Bacolod in 1998 to 'at least one lever to idle' only.

But then I consulted Captain Chen's excellent report on Taipei, and discovered that not only did he pick up and discuss the point; he specifically questioned the Airbus reps. to establish the position.

This is from the body of his report:-

"(Page 95) When auto throttles are used under normal conditions, if any of thrust control levers is not pulled to idle position, no matter using the automatic landing or the manual landing, at 10 ft or 20 ft RA the FWC will deliver “RETARD” aural alert and then automatically stop when both thrust control levers are pulled back to idle position. From 1959:23(at 23 ft RA), the FWC began to deliver “RETARD” aural alert. After four times alert, it automatically stopped but not caused by the both thrust control levers at idle position. According to the documents provided by the aircraft manufacturer, the reason of “RETARD” stop was caused that the thrust control lever no.1 was at reverse position. When the FWC detected an internal signal of TLA inhibition which includes either thrust control lever at REV position and then FWC stopped the aural “RETARD” alert.

"During landing, the FWC has delivered four aural “RETARD” alerts and then after touchdown two seconds it came to stop. But at this moment the thrust control lever no.2 was not pulled back to idle or reverser position yet. In this situation, one of thrust control lever is not in proper position but the aural “RETARD” alert already stopped. The aural “RETARD” alert should continue or there should be other ways to remind the pilots of pulling back thrust control lever to reduce the probability of an accident caused by human error."


He asked Airbus a direct question about this, and received a response:-

"(Page 146) Investigator 2/ Why does the message "Retard" stop 2 s after the nose gear touched down? Is it as per design?

"Airbus - As soon as one TLA is set to REVERSE (whatever the other TLA position), the internal FWC signal "TLA inhibition" becomes true. If "TLA inhibition" is true, RETARD is inhibited."

So it appears that the logic applying at Bacolod in 1998 still applied at Taipei in 2004; namely that the 'Retard' call is designed to continue until both TLs are retarded to idle, OR at least one TL is put into reverse, OR speed is reduced below 80 knots.

But in both cases (Taipei AND Congonhas) the call SEEMS seems to have ceased before the reverser was deployed?

I think the CVR transcript may provide a clue as to what happened:-


"18:48:14.9 HOT-2 one dot now. okay.
18:48:16.8 HOT-1 okay.
18:48:21.0 FWC twenty.
18:48:21.6 FWC retard.
18:48:23.0 FWC retard.
18:48:24.5 CAM [sound of thrust lever movement]
18:48:24.9 CAM [sound of increasing engine noise]
18:48:25.5 GPWS retard
18:48:26.3 CAM [sound similar to touchdown]
18:48:26.7 HOT-2 reverse number one only.
18:48:29.5 HOT-2 spoilers nothing.
18:48:30.8 HOT-1 aaiii. [sigh]"

Note that the 'retard' call occurs at intervals of about 1.5 seconds - and that the last one recorded occurs virtually at the moment of touchdown.

I occurs to me that the pilot may have been so conscious of the need to deploy the single reverser ASAP that he kept his hand on the No. 1 lever - and slammed it into reverse the first moment he could, as soon as the main wheels were down. The comment on HOT-2, 'reverse number one only,' might just have been a reminder from the PNF, not a callout indicating the time of deployment. In any case, the computers would have been working on the time the lever began moving through the gate, not the time the actual deployment took place.

Having found all that I was still left wondering whether it was possible to move the TL into reverse in the matter of a second or two of touchdown. But then I looked again at Captain Chen's Taipei report; and found that he had provided at least part of the answer:-

"(Page 87) The analysis of FDR recorded data shows that thrust lever 1 was at the positions of 0, -19.7 and -22.5 degrees in sequence within 1 to 3 seconds after the main landing gear touched down. 1 second later, i.e. at 1159:32 hr, thrust lever 2 was still remained at the position of 22.5 degrees."

Not sure what to make of it all - but I hope it's interesting information.

All rather confirms the 'logic' problem in the system, though. It's one thing to deny the pilot ANY stopping devices if he fails to retard both levers properly; quite ANOTHER to give him reverse thrust on the strength of just one lever having been moved, and use that event to cancel the warning callout while STILL denying him spoilers and autobrakes..........

[DozyWannabe]

Oh, give this "computer denying" schtick a rest!

If the pilots didn't configure the aircraft correctly, then they were responsible for crashing the plane. The A320 gives pilots more protection from their mistakes than many types but it's not designed to hold their hand all the way to the gate. Regardless of the opinions of some, the "RETARD" call is meant to be a reminder to perform an action (bring *both* levers to idle) if it has not already been performed - *not* an instruction from the computer to the pilot to do so.

Again, Rananim claimed that the A320's ability to give a pilot reverse on one side in the situation these pilots were in was an example of an Airbus design flaw, until he discovered that the 777 will allow the exact same thing.

I reckon it's this callout:

18:48:26.7 HOT-2 reverse number one only.

that (dependent on translation) is the key. It sounds like for whatever reason the pilot manipulating the throttles perceived that as "bring only one lever back".

[PBL]

Originally Posted by 3Ten

Primary cause is the spark, the one that if didn't occur the accident wouldn't happen.

You will find that in any accident there are lots of factors that satisfy your criterion, not just one, so your explanation is ill-defined. If you want to call them all "primary causes", I guess that's fine. The rest of us call them causes.

Originally Posted by 3Ten

If our TAM coleagues acctualy failed to retard #2 throttle, that was in fact primary cause.

And what about all those others?

There was some discussion of prioritising causes in the middle of August, primarily between myself, TopBunk, Dani, TripleBravo, with contributions also by lomapaseo, slip and turn, Right Way Up, on 18-19 August. Try starting with
[email protected]:15.55 and go at least to [email protected]:20.58
Before we start the discussion again, might I ask you to read those first?

PBL

[xulabias_bent]

The A/THR involuntary disconnection in this case was not "faulty". It was due to a failure condition, but one that is handled in the software. After that, A/THR is simply disengaged and disarmed, with a thrust lock situation, as if it had been disconnected with the FCU pushbutton.

I haved described the complete behaviour in case of the thrust lever angle sensor problems in an earlier post, but here are the relevant bits:

- With TLA (at time of failure) below FLEX/MCT, on the ground or in flight with slats extended, FADEC commands idle power, regardless of autothrust engagement status.

Bsieker,

Excuse me about the "faulty". I just wanted to mean an involuntary disconnection of the A/THR (in this case by a difference between target EPR's)

You are right about the TLA disagree. FADEC sets even the TLA reading itself to idle in this case.

But it is also true that we are dealing with general statements from the FCOM here. Complete logic diagrams like that of the warning previously discussed might help us to be sure about the applicability of the TLA disagrement procedures to this specific case of A/THR disconnection.

Anyway one thing would certainly rule out the hypothesis of a software caused failure: The independent SEC reading of the TLA position !

If the SECs read a distinct set of sensors (pots) then the fact that ground spoilers were not deployed proves beyond reasonable doubt that the TL was actually left by the crew at CLIMB detent.

That if the TRA (resolver) readings are not queried simultaneously to the pots readings by the SEC's spoiler deployment logic.

[Rob21]

RWA,

I feel that there is so much "noise" here about adding a "new" warning to the system, when MAYBE the flaw is in the current TL handling warning system.

DozyWannabe said: "...retard call is meant to be a reminder to perform an action (move both TLs to idle) if it has not already been performed".

So this action (move both TLs to idle) should only be considered complete when both TLs are at idle.

So moving just one TL, the action is not complete. Again, I believe we all agree that "RETARD" is a reminder to move BOTH TLs to idle.

So if I move only one TL to reverse, the action (both TLs to idle) has been performed? No, should be the answer.

So why the retard "reminder" ceases before the action is completed?

So MAYBE there is a chance that the pilots "thought" that since the "reminder" ceased, the action was performed?

What is the logic of having a "reminder device" that quits reminding you while the reason for "reminding" still exists?. If moving only one TL to reverse would produce the same result as moving both TLs to idle, then I can understand this logic.

And I don't need to be a jetliner pilot to understand that one TL moved to reverse in not the same that moving both TLs to idle.

So the "reminder" RETARD call is important. That's why it is there. But when it ceases to "remind" the pilots, I can understand that the action was performed.

When you get the "WHOOP, WHOOP, PULL UP warning, you pull up. When the warning ceases, it means you cleared the obstacle. This is the logic of warnings (or "reminders"), IMHO.

Rob

[bsieker]

Originally Posted by xulabias_bent

Complete logic diagrams like that of the warning previously discussed might help us to be sure about the applicability of the TLA disagrement procedures to this specific case of A/THR disconnection.

I'm not sure what you are trying to say here.

The thrust lever angle disagree failure has nothing to do with differing angles between the two thrust levers, but with a discrepancy between the values read by the two redundant, independent sensors of one thrust lever.

The A/THR disconnection here was triggered by one thrust lever being in reverse, and the subsequent EPR discrepancy.

The two things have nothing to do with each other.


Bernd