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 13th Aug 2007, 00:53
  #1561 (permalink)  

Sun worshipper
 
Join Date: Dec 2001
Location: Paris
Posts: 494
on the capt's sidestick FDR trace

ELAC,
Not too puzzling really. The PF in the left seat had substantial experience on the A320 and was in control of the aircraft. He almost certainly knew that sidestick inputs were irrelevant to the problem at hand, and was probably entirely preoccupied with making brake inputs intended to stop the aircraft.
I would certainly agree with you on principle. Problem is that trace is flat, without even the smallest movement due to acceleration - long or lat -or the involuntary pressure due to the feet efforts on the pedals.
Tried it to-day and even though we have the armrest, it is quite impossible to avoid spurious inputs on the stick.
The only conclusion is that the Capt's hand had left the 'stick.
Lemurian is offline  
Old 13th Aug 2007, 01:31
  #1562 (permalink)  

Sun worshipper
 
Join Date: Dec 2001
Location: Paris
Posts: 494
bsieker

What exactly are the THR TARGET feedback and ATHR EPR TARGET limited to the corresponding EPR TLA? And why exactly would they differ in this situation?
1. You have ATHR on and SPEED managed mode---> Thrust target to maintain the chosen speed
2. you take the thrust lever out of the CLB detent --->TLA reduces the maximum EPR possible between max CLB EPR and Idle EPR = A/THR EPR target.
3. You have an A/THR disconnect after 1.8 seconds if 2 becomes too inferior to 1, meaning that A/THR can no longer maintain the managed speed.
You also get the A/THR Disconnect warning.

That's apparently the disconnection factor on this landing.

Last edited by Lemurian; 13th Aug 2007 at 01:35. Reason: edited for clarity
Lemurian is offline  
Old 13th Aug 2007, 02:18
  #1563 (permalink)  
I support PPRuNe
 
Join Date: Jul 2007
Location: Belo Horizonte, Brazil
Posts: 162
Identification

Paxboy is right. Even thought I am not even close to being a pro fpilot lying A-320s, I am identifying with the pilots (or my, what does a weekend Cherokee pilot have in common with 13000+ hours pilots on B767 and A-320?). So please donīt look at me as being pretentious. I just love aviation.

But in my identification I see myself first trying to decide about landing in Congonhas on those conditions: heavy plane, wet slippery runway without escape, locked down right reverser, etc, etc. Then it is difficult for me to accept that a experienced pilot would move just the left TL to idle when landing, "forgeting" about the right one. They are there, close together, I just donīt believe one would select one and leave the other alone (am I in denial?). Then there is that call do "deccelerate, deccelerate" and the first officer saying "canīt do it, canīt do it". What was that?...

That plane first flew from Confins (Belo Horizonte) to Congonhas and landed there OK. The first officer landed with the right engine on idle.
Then the same two pilots of the disaster flew it from Congonhas to Porto Alegre and landed there OK too. Then they came back.

Yes, I am using my imagination, and I am identifying with the pilots, I guess trying to protect them. Then comes the thought: but this is the third time this happens. Yes, I am also wondering what Congonhas, Airbus and TAM are going to do about it. I understand that TAM has decided that it will no longer fly with a reverse locked out, but I am not sure. I understand that Airbus has an upgrade in the warning system in such situation with light and sound warning the pilot forever until the situation is corrected. But for some reason it did not make it mandatory and TAM did not upgraded it. Congonhas is grooving the runway and decreasing its traffic. Increase the lenght of the runway in the air?...

So, besides human error, I am waiting to know what else will change to prevent a fourth disaster like this.

Thanks to all for allowing me to be here.

Last edited by marciovp; 13th Aug 2007 at 02:30. Reason: to add information
marciovp is offline  
Old 13th Aug 2007, 02:21
  #1564 (permalink)  
 
Join Date: Aug 2005
Location: fl
Posts: 2,561
Several years ago at TGU they had a new A310 in the ditch between the runway and taxiway at the end of then 01 now 02. He didn't use brakes until 500 meters from the end and went off the runway backwards to keep from going off a 70 ft cliff. Guess this would have been another embarassment to Airbus but they got it stopped. I couldn't believe the report saying he didn't use manual brakes until 500 meters from the end. I thought he just landed long and screwed up but maybe there is some reason they don't use manual brakes.

When I landed on that short runway my feet were on the brakes right after touchdown even though we were required to use high auto brakes, position 4.

I wonder how many other unreported incidents have happened with AB landings? This one could have been really bad.
bubbers44 is offline  
Old 13th Aug 2007, 03:05
  #1565 (permalink)  
 
Join Date: Sep 1999
Posts: 542
I have read and re-read the posts from ELAC,PBL and bsieker in a genuine effort to see the other side.Despite their obvious knowledge of AB systems and design,I remain totally unconvinced.It seems to me they're kind of hoping that the design flaws can be camouflaged by convolution and verbosity.I'm sure Airbus itself will try the same game.
I asked ELAC to explain to me the intent of the designers by allowing one engine in rev whilst the other remains in CLB.I hoped for a straight answer and instead got a very erudite side-step.The software detects ambiguity in what the pilot is demanding and cant resolve itself.The machine is really not to blame,it gave the pilot exactly what he asked for.As others have said,human error is not to be tolerated.But humans do make errors.Where is the backup software that automatically retards the forgotten TL when the pilot selects #1 in rev?(even Boeing dont give you that but AB are cleverer arent they?)Or more simply where is the interlock that prevents #1 going into rev whilst #2 is still producing fwd thrust?Surely the software designers must play to the LCD.Assume the pilot is an idiot and hope he's an ace.Thats good design.
Of the three retardation devices in this tragedy,spoilers are key in that the effectiveness of the brakes stand or fall with their activation.However,the one that determines the pilot's mindset with regard to aborting the landing is reverse thrust.Once activated,pilots are naturally reluctant to abort the landing.The AB sets the trap nicely then.Give him rev on #1,but tell him he cant have spoilers/AB and we'll leave fwd thrust on #2 for good measure.That'll fix him nicely.
Lets look impartially at how a 737 would have coped in the same situation.If the pilot is an idiot and forgets to retard the TL,he wont get ANY RETARDATION DEVICES.#1 TR wont work because both TL's must be in idle(simple mechanical interlock).Same for AB(both TL's in idle and wheel spinup).Same for GS.Now,he's got a second(70m)to either detect his error and retard #2 OR he can simply abort the landing.He got no retardation so his speed is okay.All he has to do is stand the TL's up,confirm stable,set a reasonable N1,call flaps 15 and rotate.THREE to FIVE seconds.About 210 m/700'.Aircraft carrier mentality.Same mentality applies to hydroplaning(the only other thing that screws with the logic conditions on a Boeing) and he should be prepared for just such an eventuality everytime he approaches a short wet/contaminated rwy.
Additionally,the lack of moving TL's on the AB must be considered a factor here,despite what the Nintendo die-hards say.And the moving speedbrake lever on a Boeing is a dead giveaway tactile indication that retardation is proceeding normally.It speaks volumes compared to an EICAS annunciation.And then theres the side-stick problem.If you're the skipper and want a go-around and the co-pilot doesnt how do you resolve that one?Would the skipper even know that the co-pilot is demanding something else his side?How many more undetected scenarios are out there that will reveal the weakness in the AB interface?Milllions of hours of normalcy wont reveal a thing but it takes just a few anomolous seconds to reveal something lethal thats been dormant for years.Do they keep tweaking the software everytime they discover a new one or do they sit down and look at the big picture head on?And soon we'll have 380's with God-knows how many people down the back.
ELAC,
You said this in one of your posts:
Looked at another way, imagine if the pilot in this instance had initated immediate manual braking when the lack of deceleration was noted but had still gone off the end at a lower speed with engine #1 producing forward idle thrust instead of the maximum reverse thrust he commanded. Would there not be a howl of outrage focussed on why the autothrust system did not give him the reverse thrust that he commanded? Would we not now be noting that they would have made it if only maximum reverse thrust on at least the one engine was available?
A pilot approaching a short wet runway must have the mindset for a maximum energy stop(ie.all retardation devices...one inop TR notwithstanding).If he doesnt get immediate and clear confirmation that retardation is proceeding as per normal(the moving speedbrake lever on the Boeing which will occur well before manual braking is contemplated/possible) then he must abort.Would you go for max manual brakes on a short wet rwy if you dont get the GS immediately or at least the FS?Kill the lift,then brake.In Sao Paolo,the pilot was led by the nose into the trap by the green-lighting of TR#1 which committed him to the landing.Not Pilot error.Design error facilitated by pilot error.

Last edited by Rananim; 18th Aug 2007 at 10:59.
Rananim is offline  
Old 13th Aug 2007, 03:52
  #1566 (permalink)  
PPRuNe supporter
 
Join Date: Dec 2003
Location: Planet Earth
Posts: 1,679
Wow, if he's been quoted accurately and in context then Estado de Minas (Belo Horizonte) have not got themselves a terribly good "expert". One look at the terrain profile of the approach to CGH would tell you that radio altimeter rate, or vertical speed as measured by changes in the rad alt would be higher than normal due to the upslope of the terrain approaching the runway. I'd bet that correcting for this effect would produce a v/s of approximately 700-800 fpm, a slightly higher vertical rate than normal due to the descent below the g/s which was planned during the last part of the approach. All of which seems to have absolutely nothing to do with the events that followed.
ELAC
Spot on IMO, Bubbers, please read post #1643.
Rananim, most of what you say seems to relate to the fact that the system has to accommodate a pilot making any type of mistake, other points indicate that if it was a Boeing...it's not a Boeing, and in my opinion, I'm very happy about that, the system works pretty good. Please don't take me wrong, I hope we can somehow do something to avoid this situation in the future, what would be your opinion of modifying the system to inhibit the T/R activation until both thrust levers are at the thrust idle position?
D.L.
Dream Land is offline  
Old 13th Aug 2007, 03:59
  #1567 (permalink)  
PPRuNe supporter
 
Join Date: Dec 2003
Location: Planet Earth
Posts: 1,679
I wonder how many other unreported incidents have happened with AB landings? This one could have been really bad
Bubbers, why not give the Airbus bashing a rest, it really doesn't contribute to this thread whatsoever.
Dream Land is offline  
Old 13th Aug 2007, 04:44
  #1568 (permalink)  
 
Join Date: Dec 2006
Location: washington,dc
Posts: 486
as posted in another thread, an overrun in South Korea has just happened and it isn't an airbus.

I don't think bubbers is bashing the airbus. but I think we need to really understand the airbus...and one way is to compare and contrast the boeing. it is familiar to many and provides a contrast.

there have been many boeing 737 over runs for example...and many were pilot error, others not so much.

bashing would be calling the airbus a papa oscar sierra...discussing differences is part of the process of understanding.
bomarc is offline  
Old 13th Aug 2007, 05:52
  #1569 (permalink)  
 
Join Date: Jan 2001
Location: UK
Posts: 2,044
bubbers
Now what is Airbus going to do to prevent this from happening again?
Have you even bothered to read the thread

Airbus have, seemingly, already taken action in that the MEL was amended at some point recently. Whether or not the TAM amendment was in time will be investigated, even if it was "in time" I think most of us can understand why a crew may revert to the previous MEL procedure under pressure / if the amendment was not highlighted / if the reasoning behind the MEL amendment was not highlighted [the TAM accident certainly has now achieved the latter].

If you now re-run the scenario using the new MEL procedure, then most of the "theories" in this thread, whacky or sensible, are addressed and no longer possible and/or, if you subscribe to the "dream up a wheird technical problem" camp, the problem would be more apparent at an earlier stage...

NoD
NigelOnDraft is offline  
Old 13th Aug 2007, 06:17
  #1570 (permalink)  
 
Join Date: Aug 2007
Location: Germany
Posts: 36
Altitude / Control in the last seconds

Hello PPRuNe members, I was following the accident analysis here, only because of my curiosity (I'm a biologist). I got captived here due to the quality of the discussion, congratulations!
He stated that the Airbus had a strange loss of altitude seven seconds before landing.“If the radioaltimeters were correct, he had a ratio of descend from 150 feet (45mg) to 170 feet (51.8m) in the last seven seconds before touching down. This would be equivalent to a ratio of descend of 1.2 thousand feet per minute (365 meters per minute) and this would be a vary rapid descend”.
One look at the terrain profile of the approach to CGH would tell you that radio altimeter rate, or vertical speed as measured by changes in the rad alt would be higher than normal due to the upslope of the terrain approaching the runway. ELAC
If I'm right, the "Altitude Press." on the first page of the FDR transcript is the altitude based on the air pressure, that shows a descent rate of (750f-650f)/40s= 3.125 ft/s=187.5ft/min (57m/min), in a very smoothly way.


Control of the plane:
I'm puzzled about the dialog and the actions of the crew. Maybe I'm totally wrong, but could it be, that the PF call: look this.(18:48:34:4) is meaning something like "look at this" = "take care of the plane"? (It would be a non-conform action,?, but not the first one...) The PF trying to activate the spoilers (disarm, arm again). And the subsequent calling from the PM (HOT-2 go, go, go, turn turn turn turn.18:48:44:6 HOT-2 turn turn to…no, turn turn. 18:48:45:5) was a "loud thinking" or "command to the plane" and not directed to the PF (who was not doing inputs anymore?). Is it possible to tell, who was acting the rudder/brakes?
Regards,
Mauersegler
I'm not sure these two calls were from Hot-2 (in the FDR-transcript they are)

Last edited by Mauersegler; 13th Aug 2007 at 07:03. Reason: add
Mauersegler is offline  
Old 13th Aug 2007, 07:39
  #1571 (permalink)  
 
Join Date: Jul 2007
Location: the City by the Bay
Posts: 528
TL trap

http://www.iasa-intl.com/folders/bel...-throttles.htm
armchairpilot94116 is offline  
Old 13th Aug 2007, 07:53
  #1572 (permalink)  
 
Join Date: Jan 2001
Location: UK
Posts: 2,044
Interesting link...
There are some technical / minor errors in what he says IMHO, but the broad thrust is about right... He is concentrating on a strange HF factor, pretty unpredictable but now repeated a few times. I think that is the more likely explanation for CGH, than whacky technical issues...
The most interesting part in this context (to me anyway) are the phrases '.....all the way through the detent into forward idle. Then retard the lever again against the stop.' 'Through the detent,' and then retard even further, 'against the stop' - not just 'back to'?
He seems to be referring here to getting out of Rev. I certainly find it is difficult to smoothly operate TLs out of Rev to Fwd Idle, and easier, as here, to go somewhat forward of the idle detent / position and then back to idle. Nil effect in the time it takes re actual engine thrust...

NoD
NigelOnDraft is offline  
Old 13th Aug 2007, 07:53
  #1573 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,581
armchair - an interesting insight into the POSSIBLE events. Just one problem in the 'accident scenario' which is instantly obvious is that (to a non-AB eye), I see no increase in TLA or EPR on either engine to indicate an attempt at a g/a?

Furthermore I am indeed puzzled by what appears to be an EPR at around 'idle' on No1 in reverse? I would have expected more to be planned in this landing. Indeed on the previous landing (at Porto Alegre) No 1 appears to be taken up to around approach EPR before being 'idled'.
BOAC is offline  
Old 13th Aug 2007, 08:44
  #1574 (permalink)  
 
Join Date: Jul 2007
Location: the City by the Bay
Posts: 528
to BOAC: I think the Late Go Around theory was quashed soon after it was hashed about. And I wondered why anyone was still harping about that being a possibility. Any other errors I was certain the faithful would catch.
armchairpilot94116 is offline  
Old 13th Aug 2007, 08:55
  #1575 (permalink)  
 
Join Date: Jul 2007
Location: Germany
Posts: 533
Trying to understand Autothrust.

Originally Posted by Lemurian
1. You have ATHR on and SPEED managed mode---> Thrust target to maintain the chosen speed
2. you take the thrust lever out of the CLB detent --->TLA reduces the maximum EPR possible between max CLB EPR and Idle EPR = A/THR EPR target.
3. You have an A/THR disconnect after 1.8 seconds if 2 becomes too inferior to 1, meaning that A/THR can no longer maintain the managed speed.
You also get the A/THR Disconnect warning.
Yes, I figured it was an "involuntary disconnect", as FCOM calls it, with following thrust lock.

Thanks a lot for explaining. I'm trying to write up a short summary about that, but still can't quite figure it out exactly, and I don't like to make ill-informed posts

I understand that at diminishing airspeed there is an increasing EPR target that A/THR tries to set to set on the engine(s), but #2 was in CL detent, and surely one engine at CLIMB power would be enough to maintain 140kts even in CONF FULL.

[Edit: Nonsense about interaction between two FADECs deleted]

Oops, I got it completely wrong.

Thanks Lemurian for catching me! But see my other post, maybe I got it right now.


A/THR is a lot more complex than I imagined.

Bernd

Last edited by bsieker; 13th Aug 2007 at 13:51. Reason: Removed incorrect paragraphs. No need to confuse people further.
bsieker is offline  
Old 13th Aug 2007, 09:34
  #1576 (permalink)  
 
Join Date: Oct 2006
Location: Spain
Posts: 82
Just a couple of things:

1) I donīt think anybody has to blame AB. As I posted earlier, a B737 took of without flaps, and even the loud warnings, the pilots did nothing; do you think then that this is a faulty design from Boeing, alow an aircraft to take off without flaps?

2) Of course it would be a madness, but, Is it possible to put a net to stop planes that overrun, just like in a carrier?
agusaleale is offline  
Old 13th Aug 2007, 09:53
  #1577 (permalink)  
PBL
 
Join Date: Sep 2000
Location: Bielefeld, Germany
Posts: 955
Concerning Waldock's observation reported in Estado de Minas (marciovp in #1642), note the Pressure Altitude is on the same page and shows no sharp variation (as Mauersegler noted). ELAC noted there is rising terrain (#1643). Need we say more?

Comparing RA with PA is so basic that I am wondering whether Estado did not quote Waldock out of context?

Concerning the reliability (or otherwise) of digital avionics systems (in particular comments by flyingnewbie10, Sdrvuss), comparing digital avionics to corporate IT systems or others seems to me to yield about as much insight as comparing the wheel construction and braking systems on (a) your bicycle, (b) your car, (c) your truck fleet, (d) the A380 bogie sitting in the entry hall of the Science Museum in London. Sure they all have similarities, but their differences invalidate the kind of analogical reasoning, especially about failure modes, that I have been reading here.

Back some 70 years ago, Frank Ramsey determined how one may rank convictions in the presence of uncertainty, by considering at what odds one would be undecided in betting on two given events. The ranking he derived is consistent with all future work. (To those who consider betting infra dignitate, I note that this stuff is highly respectable semantics ). Now, investigators don't use it (unfortunately; I have come across some pretty poor reasoning about what is "likely", even in reports from agencies with the highest reputations).

So, given that the evidence for either has disappeared, and thus that we cannot find out definitively, what are the chance of sensor-fault versus TL-really-left-in-climb? Following Ramsey, at what odds are you willing to bet on which one? Note that if you bet askew, someone will be able to offer you a bet which (a) you will take, thinking it to be advantageous, and (b) you are guaranteed to lose (a so-called "Dutch book"). Is anyone arguing for a sensor fault prepared to say what heshe thinks the relative chances are?

And to those who think "likelihood be darned, it is still *possible*", I'd like to ask how far you got in dealing with the possibility that the center wing-tank explosion of TWA 800 was caused by little green men sitting in it and playing with matches at the wrong time. You can't rule that out either.

To slip&turn, I should point out that that analysis in the 1994 braking-logic state machine paper came from FCOM of the Lufthansa machine involved in the Warsaw overrun and is not valid for more recent aircraft, which have a different logic. That current logic may be read in Section 1.32.30 of the A320 FCOM, many "for training purposes only"-copies of which one can find on the WWW.

PBL

Last edited by PBL; 13th Aug 2007 at 10:51. Reason: To get rid of superfluous material; to resolve a confusion
PBL is offline  
Old 13th Aug 2007, 09:58
  #1578 (permalink)  
 
Join Date: Sep 2003
Location: 59°45'36N 10°27'59E
Posts: 999
point he could descend below the glide slope to have more runway available to him at touchdown. A normal 750 fpm rate of descent would then go to 1200 fpm to accomplish that. Most of us have done it many times. I have many times with positive results. It was done below 200 ft. Check the CVR.
As an ATCO for a RWY with 800m displaced papi on one of the RWY ends, but with all 800 meters available for landing (not in LDA) I've seen this a lot.
AND I've seen crews doing this when they have to much energy on short final.
Result: They land long after the PAPI as well, after floating 800m plus.
M609 is offline  
Old 13th Aug 2007, 10:10
  #1579 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,581
Here's where I stand at the moment: assuming that leaving No2 at CLB was a major factor, and that this is 'eliminated' by repetition of 'warnings', I would like to see the following recs out of this:

1) Planned landings on 'slippery' runways barred if T/Rs are not 100% (common practice in my experience)
2) A realistic MEL addition for LDR with one or more u/s, which varies with runway condition (common practice in my experience)
3) A 'standard' (international) downgrading of a runway that has been re-surfaced but not grooved to 'slippery when wet' - enough recent 'evidence' that this is so
4) an increased CVR time made urgent rather than 'convenient'

We can forget all the mumbo-jumbo of logic circuits/sensors etc and with a re-empahsis on 'basic' flying skills plus the above I would hope we will have learnt sufficiently from this tragic accident.
BOAC is offline  
Old 13th Aug 2007, 10:11
  #1580 (permalink)  
 
Join Date: Aug 2007
Location: South Africa / Thailand
Age: 67
Posts: 10
Now that this landing has been simmed (as reported earlier), can anyone report whether continued flight was possible after the operational T/R was engaged. If at 11 seconds, instead of manual breaking, TOGA was initiated (albeit with waiting for the T/R to stow), would the bird have been able to fly in controlled flight?
BoughtMyPoints 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 Đ 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.