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 31st Jul 2007, 14:53
  #761 (permalink)  
 
Join Date: Jul 2007
Location: india
Posts: 16
Angel MEL on 1 REV U/S

Just to put doubts of all to rest again. On airbus, if one Reverser is inop, The MEL clearly asks pilots to first bring BOTH TLs to idle at flare and then BOTH TLs to reverse. there is no ambiguity in this and it is clearly given in the MEL. Trust me, I have checked it personally on my last flight. Now, if someone DOES NOT follow procedures, why blame all and sundry.
Having said that, the official reports are still not released. So, we need to hold our horses and then comment on the systems.
Again for all the B types. If your TLs are at climb setting after landing with manual throtles, what will YOUR aircraft do? will it not accelerate? will the speed not increase?
nada1234 is offline  
Old 31st Jul 2007, 15:04
  #762 (permalink)  
 
Join Date: Sep 2001
Location: Toronto
Posts: 2,256
Differences in Human-Machine Interface

The consensus on this thread is that on while on Autothrottle on:

B The TLs move to match the commanded thrust such that on A/T disconnect there is no change in thrust.

AB The TLs do not move to match the commanded thrust such that on A/T disconnect, the thrust changes to match the TL position.

The AB logic is reasonable on its face as long as the crew remembers to move all TLs to idle on the flare -- but fails on human-machine interface criteria because it can put the crew in a nasty spot when by typical human failing in a non-standard situation, they happen to miss moving all TLs.

Likely the ECAM was displaying messages and the "Retard" call was sounding, but touchdown is a time when attention is directed outside, especially when there's not much runway ahead and it is disappearing fast. This is not a good time for head-down fault diagnosis.
RatherBeFlying is online now  
Old 31st Jul 2007, 15:42
  #763 (permalink)  
 
Join Date: May 2000
Location: Glorious West Sussex
Age: 72
Posts: 1,020
BOAC
1. airbus flap settings are called "config" .. A320 has CONF 1, 1+F, 2,3, and FULL. FULL and 3 are "normal" landing CONFs.
Speedbrakes (in-flight use of spoilers) are inhibited in FULL, but Ground Spoilers and Autobrake are always available on the ground in any config.

2. Ground spoiler indication is on lower ECAM - two lines of arrows (tiny Christmas trees?) pointing upwards. If no operation just two lines of dashes with no arrows.
Autobrake has Blue ON light in pushbutton (p/b), but to assess correct Decel PNF is looking for IAS trend arrow on PFD, feeling the decel, and least important the Green DECEL light in the Autobrake p/b. You really can feel MED Autobrake against your harness.
Thrust increase on No.2 would be shown on N1 and EGT (dials) and N2 (digital value) on upper ECAM. The No.1 N1 gauge will show the word REV (boxed) in amber then green (in centre of dial) as the reverser deploys - when green N1 EGT N2 will increase if MAX REV selected. They will not be synchronised but both engines will show a thrust increase.

RatherBeFlying
PNF should be head down monitoring ground spoilers and reverse - it's not fault analysis it's making sure the systems operate as advertised, calling as such, and alerting PF if they do not. Either PNF or PF would also call a relevant ECAM message.

Any reference to PF/PNF actions are airbus SOP's and may vary in individual airlines.
Cheers, TP
TyroPicard is offline  
Old 31st Jul 2007, 15:59
  #764 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,581
Thanks, Tyro, so #767 is partly wrong for 'Config Full' or is that just the way it has been worded - and does the post actually refer to the TL2 'interference' rather than the 'CONFIG FULL' default?
BOAC is offline  
Old 31st Jul 2007, 16:05
  #765 (permalink)  
 
Join Date: Jul 2007
Location: uk
Age: 63
Posts: 6
Would, in the case of this accident, the "Retard Retard" warning have continued after touch down with the No1 TR in RT if No2 was above Idle? If not what criterior would cancal it?
ghunter is offline  
Old 31st Jul 2007, 16:05
  #766 (permalink)  
 
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,975
@nada1234

Again for all the B types. If your TLs are at climb setting after landing with manual throtles, what will YOUR aircraft do? will it not accelerate? will the speed not increase?

So nada what do you think, why (if true) was TL #2 left in CLB detent (like in another accident with 320)?

I'll tell you why. The pilots didn't expect the fwd thrust at all!

Let's put it that way, on AIRBUS there may be system behaviors in critical phases of flight which may be beyond the capabilities of an average pilot, sometimes beyond capabilities of an AIRBUS test pilot.
hetfield is offline  
Old 31st Jul 2007, 16:07
  #767 (permalink)  
 
Join Date: Jul 2007
Location: india
Posts: 16
Angel Forgot To Put Throttles To Idle?????

RatherBe Flying, Again going back to basic flying on any aircraft. What would your instructor have done if you "FORGOT" to put throttles to idle before landing? Guess you wouldnt be flying as he would have stopped your training then and there. Why would you not "FORGET TO LAND"? the actions are very clear and unambiguous in AB. Flare starts at 30 ft and you start hearing "Retard Retard". You bring and keep both TLs with your right hand to idle and with your left hand (for CM1 flying as PF) you land the aircraft. Then you move the TLs to retard position. The PNF then checks out spoilers, reversers and decel (if autobrake is armed). At 70 kts, you move TLs to reverse idle and before turning off the runway, you move them to idle. Thats it.
Problem occurs if you let pervious aircraft habits interfere here which is uncalled for. Or do something not called for like leaving ONE TL in clb detente. The system is confused as to what do you want to do now. you want to land or maintain speed.
Also, you do not want somethings to happen when you are doing a training touch and go. For that there are a different set of actions - like there is no arming of spoilers,no autobrakes and no selection of reversers after touch down. Then you stand them up by moving TLs forward and on GO from instructor, you move to TOGA to initiate go round.
In TAM accident, based on the current rumours, ONE TL apparently, is in clb detente. Now, they would have been neither here nor there. the system would want to know, are you wanting to land (as one half of TLs, is trying to do by being in reverse) OR you intend to go round (as one half of the TLs is trying to do being in clb detente). No amount of logic can prevent these Murphysque situations.
There is a talk of magic button. In this case, the magic button could be TOGA if action taken was sufficiently early since the spoilers would not have deployed and autobrakes not activated. However, going half way down the runway - wet and short, there is nothing much any braking or any magic button would have done. If the contention is that they hit the building at +94kts, they would have been doing much higher midway anyway. Now to stop that momentum, autobrakes could have been at max or any higher setting (imaginary), nothing could have stopped them. Anyway, i guess the brake pedals would have been to the floor in the panic (which corresponds to max autobrake application) anyway and if it still could get speed down to 94 kts only after overrunning runway, crossing the highway and then reaching the building, guess no magic button could have stopped them.
Guess what we all need to learn is to stick to the basics and know that it could happen to best and most experienced of us.
Happy and safe landings.
nada1234 is offline  
Old 31st Jul 2007, 16:27
  #768 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,581
nada - may I sort out some confusion for you? With the 737 it would be pretty well impossible to have a T/L in a 'CLIMB' setting as the highest it would normally be would be at an approach setting. The 737 does not have a 'CLIMB' setting. There is also no way a throttle would 'advance' to any other setting after landing as Autothrottle is already disconnected - and WILL NOT re-engage. The only time that I can envisage that it could 'activate' is if the landing is made with A/T 'engaged' but 'armed' and a fault occurs in the air/ground sensor, but again it would advance both T/Ls.

We are looking at a different auto system here, and while leaving a T/L at 'approach' power would be pretty nasty in a 737 on a wet slippery runway, it would not be so calamitous as it would with this system. I suppose I could envisage a situation where a sudden loss of speed near touchdown caused the HP to increase power significantly, but here again it would be both throttles and there would be a significant visual and tactile clue. If, as suggested in 4HP's post a way back, the hand is REMOVED from a T/L the tactile input is lost and you are really left with only electronic indicators to tell you.
BOAC is offline  
Old 31st Jul 2007, 16:48
  #769 (permalink)  
 
Join Date: Sep 2001
Location: Toronto
Posts: 2,256
TyroPicard: agreed SOPs normally call for PNF to monitor the ECAM and make callouts. Waiting for the CVR.

Nada1234, of course in manual throttles we all pull the TLs to idle. With A/T, Mr. Boeing does that while Messrs. AB do not.

There is an extra task for the AB pilot which fallible humans can, and do if rarely, miss.

From a human-machine interface engineering viewpoint, I much prefer systems that assist people to do things right to those that sometimes set them up for a fall.
RatherBeFlying is online now  
Old 31st Jul 2007, 19:01
  #770 (permalink)  
 
Join Date: Jul 2007
Location: Switzerland
Age: 74
Posts: 98
Nada1234

My initial idea was not that of a "magic button". SoaringTheSkies picked it up nicely and explained the "override interlocks concept" much better than I would have been able.
Many posts so far have claimed that the danger of commanding a full stop at the wrong moment would be a greater risk than the very rare cases where the automated logic confuses the pilots with disastrous results. Maybe thats the case. I doubt it. I have not yet heard of pilots of military jets to trigger the ejection seat unintentionally, and I have not yet heard of pilots flying GA aircraft equipped with safety parachutes to fire them without reason. I would assume that a pilot hitting the "break interlocks override" or whatever the feature is called at Mach 0.8 at FL 390 might be able to envisage the consequences.
But Murphy rules that if it is physically possible to land an aircraft with the TL's in the climb detent, then someone will eventually do it. If you dont touch the levers during the whole flight, some crew might just forget it at the final flare. We all know they should'nt, but it's human to forget things, particularly when other stress factors and/or fatigue influence crew performance. and particularly if some computer system hides the causes until its too late to do something about it.
I wonder how many crews may have had wet pants after a stop at the very end of a runway because of some similar chain of events, but were lucky enough not to make it into the newspapers.
.
clearedtocross is offline  
Old 31st Jul 2007, 19:29
  #771 (permalink)  
 
Join Date: May 2000
Location: Glorious West Sussex
Age: 72
Posts: 1,020
BOAC
To operate Speedbrakes you push the lever down then pull back to the desired setting.
To arm Ground Spoilers the speedbrake lever is pulled up (must be fully forward in RETract setting to achieve this). Ground spoilers are automatic only, deployed by the flight control computers if armed and all the conditions are met. (See above!)
Re Post 767 Option 1.. I think it's just the wording.. no ground spoilers because of T/L 2 position, and no speedbrakes because of CONF FULL.. though IIRC manual speedbrake selection on landing is not done on the A320, if the ground spoilers are Inop they are Inop.(And that is very rare).

nada1234
What would your instructor have done if you "FORGOT" to put throttles to idle before landing? Guess you wouldnt be flying as he would have stopped your training then and there.
If I was the instructor I would teach the student how to do it right. It is not an uncommon fault, especially when converting to a new type when one is focussed on flare characteristics for the first few landings - one concentrates on one hand and forgets the other.. but it is easily fixed with practice.
Cheers, TP
TyroPicard is offline  
Old 31st Jul 2007, 19:57
  #772 (permalink)  
 
Join Date: Jul 2007
Location: Argentina
Age: 63
Posts: 38
Debug

Hi there...
The system must know your are landing. There is an history of state transitions which clearly point to landing; even more: it seems there was an engine with its reverser deployed... So, why FADEC accelerates the other engine? Merely because the pilot missed or forgot to put the TL at the proper detent? Im sorry, but this sounds very buggy to me.
Given the danger inherent to the procedure, this kind of heuristics should be reviewed and tested even by dummies. In my eyes, if the system gets confused by your actions, then it should be much beter (read safe), to prompt the pilot with a small set of possible steps to take, rather than taking dangerous ("intelligent"), assumptions.

Last edited by TioPablo; 31st Jul 2007 at 20:18.
TioPablo is offline  
Old 31st Jul 2007, 20:29
  #773 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,182
So, why FADEC accelerates the other engine? Merely because the pilot missed or forgot to put the TL at the proper detent? Im sorry, but this sounds very buggy to me.
The specifications for thrust control are posted earlier in this thread and *if* that was the cause of the overrun the system performed exactly as the specification said. Software 'bugs' are when software fails to perform as per specification, so it wouldn't have been 'buggy' at all.

Most of the criticism leveled at the AB FBW in the early days was based on the idea that the system would second guess what pilots were doing, so (again, if this turns out to be the cause) in this case the system was doing exactly what the pilots requested. Adding the 'guessing' component would not only draw ire from pilots, but would unnecessarily complicate the software design, making genuine bugs far more likely.
DozyWannabe is offline  
Old 31st Jul 2007, 20:38
  #774 (permalink)  
 
Join Date: Jan 2005
Location: France
Posts: 2,316
TP,
It is not an uncommon fault, especially when converting to a new type when one is focussed on flare characteristics for the first few landings - one concentrates on one hand and forgets the other.. but it is easily fixed with practice.
You serious, man?
To me it has always been "hands on throttle and stick", even forty years ago in a Piper Cub, since I always came in a bit low over the fence....
Translated, it always has been "power and attitude", if you see what I mean.

It sounds from what you say, that that link has been broken somewhere. Even if you have A/T until touchdown, some part of your awareness should keep track of what the "power" is doing.
ChristiaanJ is offline  
Old 31st Jul 2007, 20:41
  #775 (permalink)  
 
Join Date: Apr 2005
Location: US
Posts: 51
Nada1234,
you wrote a valid post for all Boeing guys like me, but I want to explain you the difference.
1) If the A/T is OFF and I don't retard the thrust lever, the engine will continue to produce the same thrust as it was producing before touchdown. This means that the engine will NOT accelerate to try to maintain any speed or to CLB or TOGA thrust whatever you call it.
2) It is impossible to have any SPEED mode in the Boeing 2 seconds after touchdown.
The main difference between B and A (from what I could understand from this thread, is that if you don't retard one thrust lever in the Airbus (for whatever reason: Pilot error, Mechanical Problem,...)the autothrust will keep that engine in SPD mode and will accelerate the engine up to CLB thrust to maintain the SPD while in the Boeing the engine would continue to produce the same APP thrust (around 55% N1). The engine would never increase thrust in the Boeing after touchdown (2sec)unless you manually advance the thrust levers.

As I said before, I don't want to enter in a discussion about which ACFT is the best. Both Boeing and Airbus have up's and down's but the Airbus autothrust logic is completely diferent from the Boeing (to say the least).
Just my .02cents.
Regards,
pira is offline  
Old 31st Jul 2007, 21:07
  #776 (permalink)  
 
Join Date: May 2000
Location: Glorious West Sussex
Age: 72
Posts: 1,020
ChristiaanJ
Perhaps I should have said "I have seen it a few times in the simulator with trainees converting to tha A320, and once in three years of line flying, by a new co-pilot". But always both thrust levers, never just one.
I was merely trying to make a point about learning, in response to a dogmatic statement by nada1234.
Gosh aren't a lot of pilots perfect!
TP
TyroPicard is offline  
Old 31st Jul 2007, 21:39
  #777 (permalink)  
 
Join Date: Jul 2007
Location: Europe
Posts: 92
Hi there, another one leaving his silent corner... ;-)

I have thought alot about that accident, but I don't see the point of introducing a kind of a "red button" would have saved the day of those dear lost souls. If it was like the rumors have it, it was there: thrust levers. What would you do instictively when you go too fast on ground? Cut power, right? Works also in a Cessna. Any introduction of additional buttons make things even more complicated as they could serve themselves as another human factor in the next, but different stress situation.

What you could do in a DC-9 (? not sure) or in a PC-6 is catastrophic with A320 or B737: selecting reverse or beta while flying. This is why there are protections against deploying reversers in flight. I remember at least three accidents with a hull loss caused by that (e. g. LaudaAir - they didn't even had time for a mayday before breaking up). Same goes for lift dumpers since nobody wants a stall in slow flight.

People say that the aircaft "should have known" that JJ3054 wants to stop there an then. But - what if the reverser (was) deployed inadvertently (as reversers already did in history, see LaudaAir et al.) and they originally wanted to go around? The system cannot tell which power lever position is the "right one". Any automatic system that would have been able to shut down ENG #2 on itself would lead to another catastrophe. Not in THIS case, right, but in another scenario. I personally wouldn't feel easy when I would know that there is some system that is able to just shut down power on its own. Remember, also this system could malfunction.

My bottom line would be that any introduction of new logic would make things more complicated and maybe save one ship but would lead to the loss of another, different one on a different day.

(I don't think it's a tall order for a pilot to operate the two single most important levers in the cockpit according to the book, is it?)

Cheers, BBB
TripleBravo is offline  
Old 31st Jul 2007, 21:50
  #778 (permalink)  
 
Join Date: Oct 2006
Location: Salvador - Brazil
Age: 41
Posts: 19
Now it's OFFICIAL, the government will release FDR/VDR tomorrow !!

http://josiasdesouza.folha.blog.uol...._37-10045644-0

Says "... decided this tuesday (31) that they will make public the data of both black-boxes from tam's airbus. ... The release should happen tomorrow."

Let's hope that they don't change their mind and release both FDR/VDR tomorrow, so we can REALLY see what has happened.
duwde is offline  
Old 31st Jul 2007, 22:51
  #779 (permalink)  
 
Join Date: Apr 2004
Location: Essex
Age: 50
Posts: 118
For those so vociferously defending the airbus logic, just remember one thing.
These pilots, as most / all pilots are, were highly skilled and highly trained. They did not intend to crash, they did not intend to kill themselves. therefore one can assume that whatever they did made absolute sense to them at the time. The fact that they made, what to them at the time, with the data they had was a sound a decision and got bitten for it, means that there is a fault with the man-machine interface. End of story.
You can sit reading a book and say it all makes sense as much as you like, but the fact remains that something in that logic screwed 2 experienced pilots and killed 200 people. The fact that it appears to have happened in at least 2 other accidents, (and from that we can assume that it has also happened on other occasions which didn't lead to accidents, and so didn't get reported) means that this now needs to get fixed, pronto.
I don't have an axe to grind one way or the other about Boeings and airbus's, all makes of aircraft have had 'quirks' that have caused problems. That this particular quirk appears to have caused more than one accident, means it is real, and it exists and it needs to go.

FWIW, Personally I like tactile feedback - I like thrust levers that move and sodding great spoiler handles that ping back to show me the spoilers have deployed
AlexL is offline  
Old 31st Jul 2007, 23:10
  #780 (permalink)  
 
Join Date: Jul 2007
Location: Europe
Posts: 92
If it is like the PF really did not move the #2 lever, then to me it looks more like a transition problem from one design philosophy to another. I really think it is difficult to change "deep" habits and expectations in a few days of training. Maybe that point was taken too lightheartedly until now and it seems that it was the pilot flying who was new to this.

But another question I did not read in the entire thread: How come everybody is sure that the lever sensor wasn't broken? What if they operated the bird by the book but the electronic never sensed that #2 was drawn to idle power? The DFDR cannot show the position of the thrust levers themselves, only the data produced by the levers' sensors. At least to me it looks to be VERY important where the TL physically was in the wrecked cockpit.
TripleBravo 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 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.