TAM A320 crash at Congonhas, Brazil

[TyroPicard]

STS

Retarding the levers to idle really has nothing to do with bringing the engines to idle, they are already very close to ilde by the AT It's a metaphor for the pilot to tell the plane "we are committed to this landing". It really disconnects the AT and enables the GS and AB system.

For a CFM-56 Ground Idle N1 is about 19.5%, typical appraoch N1 54%. Quite a difference, not "very close to ildle (sic)" at all. As the T/L is retarded it reduces the thrust, if you don't then the A/T would increase thrust to maintain IAS. It has everything to do with bringing the engines to idle, and is not a figure of speech at all.

the same TL also controls the reversers, which might have added to the confusion in this case.

Never found them confusing myself. No need to move your hand to another set of levers - what could be less confusing?

first, if they were moving with AT (or flown manually) there's just no possibility that they would be sitting in the CLB detent. Very simple, no room for error.

IIRC if you go back through this thread you will find examples of that error on some of Mr. Boeing's finest - we are all human whatever type we fly. I assume you fly - your profile does not say?
Cheers, TP

[bsieker]

Be the right term "desconnection" or "deactivation", what I would like to know is if the condition is TLA <= 0 or simply TLA = 0.

I didn't program the thing, so how I could I talk about the actual code lines used?

And I still don't see your point.

FCOM says, A/THR disconnects if both thrust levers are pulled to the IDLE position. And I don't think there is a single indication that it has ever failed to do so.

In addition, A/THR is supposed to stay active if only one lever is pulled to IDLE, so it's irrelevant, anyway.

[Lemurian]

I find it quite extraordinary that with all the data the FDR transcript provided, we are still spending time discussing TLA.
The TLA traces + the EPR traces + the concordance with the spoiler and autobrake deployment logic all lead to the most likely conclusion : the #2 Thrust lever was left in the CLB detent.
The TPE accident, in which the aircraft cockpit was recovered and where investigators could test TLA/TRA values bore that conclusion.
And now comes the boring "Can't happen on a Boeing" argument...Enough of that as it brings no value to this discussion.
I've received a few messages from colleagues flying the '320. Apparently, flying a TAm/CGH-type of landing has become of real interest. Mostly, they have all chosen
- a 1900 m or so runway
- made it wet
- landed at 62 tons
- with one T/R only, the #2 engine left in CLB detent
- They have all observed *no spoiler* and *no auto-brake*.
- an 11 sec delayed brake application
In the above conditions, the landing has led to high speed runway excursions on all reported simulations. Excursion speed ranges between 80 kts and 100.
The published FDR transcripts show a few interesting points :
1. They landed well before the 1000 ft marker, as intended by the captain when he elected to fly one dot-low on the glide. The passing abeam the G/S antenna (and yes,Mr investigator, I stick my neck on this one, by reason of thousands of experiences ), when the glide signal goes up then nuts occurs after touch-down - whichever way you consider it Radalt or WoW.
2. An important call-out is missing on the CVR : " No DECEL !", indicating the auto-brake isn't functioning as set.
3. As PBL wrote, the brake application occurred earlier than the 11 sec already published...Problem is that application seemed very tentative (as if checking the auto-brake was there...), the maximum braking started 6 seconds later. (the 11 sec are there ).
4. This is the most puzzling point : the left sidestick was then left alone, the FDR trace for it remains abnormally flat...on the other hand, the right sidestick becomes quite lively - even going for full nose-down application in the last seconds ... The CVR never showed a control transfer, did it ? This remark is borne by the CVR transcript - if it is to be trusted - that the "vai...vira.." utterrances come from HOT1 (the Captain's mike ).
5. the gradual, then accelerating heading drift to the left is also surprising to the colleagues who simmmed the accident. No one had any directional problem at all. Was the runway more slippery than advertised or did they experience some aquaplaning - IMO too easily dismissed in this case -?

[SoaringTheSkies]

tp, let me answer your last question first:

I do fly, but no big iron, so I guess the correct answer to the intended question is no.

Also, I have never said that there were no such pitfalls in a Boeing. To the contrary, I've stated multiple times that I had no reason to prefer one over the other. I actually wished there was a way to contrast the Airbus system without spelling "Boeing", but they are the two antipodes out there, at least in some aspects.

Actually, if you don't mind me digressing for a bit, actually, I'm pretty amazed how manufacturer sensitive (manufacturer patriotic?) many here seem to be. It's almost as bad as a Windows vs. Mac flamewar where you also cannot mention one without being flamed for bashing the other (or hyping it, depending on the slant of your posting). It's never a good thing to be pro or con something on an emotional basis (other than your g/f or b/f maybe). It gets in our way when we're looking at the thing's faults.

Never found them confusing myself. No need to move your hand to another set of levers - what could be less confusing?

Hmm... you know what? Probably the crew in this accident never found it confusing either at any time before.
My point is: the thrust levers on the Airbus are overloaded (no negative term, a technical term meaning that they serve several purposes) with a) condition presets (IDLE, CLB etc) b) max N1 (or EPR?) in AT, c) manual thrust setting if AT is not active and d) activating reversers. Those are all functions that deal with the same system, the engines, but still they're rather different functions. I think an argument can be made that overloading controls bears a risk of mishandling them due to mode confusion. (not necessarily what happened here).

On the approach N1 vs. ground idle, the whole thing might be irrelevant if (put any other airplane's manufacturer's name here where the TLs are actually moving, this clause inserted to avoid the "B" word) does not use the TL position to inhibit the ground spoiler / brakes logic.

pj

[vapilot2004]

Airbus thrust lever description

The mechanical thrust controls for the A320 consists of three main parts:

The Levers,artificial feel units and control units.

This setup is common for most FADEC equipped aircraft with the only difference between Airbus A320/330/340 and others is the addition of a servo motor to back drive the throttle levers in autothrust mode.


Each thrust lever is connected by a pushrod to the
input of an artificial feel unit. This unit provides friction
and the detents throughout the motion of the thrust levers.

The artificial feel unit's output shaft is connected
to the thrust control unit via a second pushrod.
Within each thrust control unit are 6 potentiometers
and 2 thrust angle resolvers.

The pots provide thrust lever position to the flight control system
while the resolvers are dedicated to the engine thrust control.

A resolver is a sort of rotary transformer that outputs sine & cosine
waves corresponding the the thrust lever angle. The mechanical ratio of the
thrust lever angle and the resolver angle is 1:1.9. For every degree of
movement of the levers, there is a 1.9 degree movement indicated
by the resolver. Resolver units are more accurate and reliable in translating
angular position data than potentiometers and is the reason for their use here.

Each thrust control unit has two resolvers - one for each EEC channel.
A disagreement of more than 15' (1/4 degree) causes the EEC to go into
resolver failure mode.

Each EEC also checks the resolver output for upper and lower limits.
For the IAE engines the limits are from -41 to 88 degrees. Any
value outside of these limits will cause the FADEC system to transfer
engine thrust control to the alternate EEC.



Outside of a jammed feel unit or thrust control unit, a failure of the A320 manual thrust control system seems unlikely.

[bsieker]

Originally Posted by SoaringTheSkies

On the approach N1 vs. ground idle, the whole thing might be irrelevant if [...] does not use the TL position to inhibit the ground spoiler / brakes logic.

Well, as has been posted by "Right Way Up", some other manufacturer, whose name starts with B and ends with "oeing", does inhibit ground spoilers unless both thrust levers are at idle: See post #1259, on page 63.

Just though I'd mention it since not everyone seems to have read the entire thread.

Bernd

[TyroPicard]

STS

My point is: the thrust levers on the Airbus are overloaded (no negative term, a technical term meaning that they serve several purposes) with a) condition presets (IDLE, CLB etc) b) max N1 (or EPR?) in AT, c) manual thrust setting if AT is not active and d) activating reversers.

The TLA sensors sense the angle. All that other stuff is done by FADEC, A/THR etc.

Also, I have never said that there were no such pitfalls in a Boeing.

You said, concerning moving thrust levers..

Very simple, no room for error.

I pointed out previous posts where aircraft with moving thrust levers - which happened to be Boeings - came to grief by human error. Your statement is wrong.

On the approach N1 vs. ground idle, the whole thing might be irrelevant if (put any other airplane's manufacturer's name here where the TLs are actually moving, this clause inserted to avoid the "B" word) does not use the TL position to inhibit the ground spoiler / brakes logic.

TL position is an extremely important part of that logic - all explained in previous posts.
TP

[SoaringTheSkies]

hmm, TP, what I'm really wondering about is this:
what is the point?

We seem to agree that the most possible root cause is human error.

It seems to be general consensus that the crew was very experienced and had performed flawlessly before.

It would therefore seem logical to assume that they were well trained on type and, it being a hub, familiar with the local airport.

Now, by your and some other's reasoning, the aircraft design is somehow off limits at least for consideration here, because neither you nor anyone else has ever experienced any issues with it (maybe part of the reason why you're hear to reason with us). There seems to be a big interest in proving that it's no worse than anybody elses. Fair enough.

What are we left with then?

The occasional human slip that can turn out a disaster?

Not much we can do to prevent that. We can try to make it happen even less frequently, but we'll never completely fix it.

This is why I demand that any complex system involving humans adheres to three rules:

- avoid error
- tolerate error
- fail gracefully

pj

[borghha]

lemurian wrote

This is the most puzzling point : the left sidestick was then left alone, the FDR trace for it remains abnormally flat...on the other hand, the right sidestick becomes quite lively - even going

This has even appeared in the Brazilian press: one of the people involved with the investigation said that the PNF took over during the last 15 or so seconds, describing the switch as an action of despair.

[bomarc]

is someone suggesting the left seat (captain) became incapacitated and the right seat pilot started to try to fly the plane?

add to this, the somewhat odd way of switching seats and switching legs and the perhaps less comfortable right seat flying?

[ELAC]

I cannot explain why engine 2 EPR rose to 1.26, then fell back, and only then did A/THR disconnect.

bsieker

I think the answer can be found in the explanation of the A/THR disconnection logic for the specific condition which is found on page 148 of the taipei accident report.

-If the THR TARGET feedback of one FADEC is different by 0.15 EPR from the
ATHR EPR TARGET limited to the corresponding EPR TLA, the ATHR is
disconnected after 1.8 sec.

The 1.8 seconds from the time that the #1 engine was selected to reverse was probably just about the amount of time needed from the flare for speed to decay sufficiently for A/THR to respond with an increase of power on #2 to stabilize the speed and then reduce again as the speed target was met. At that point A/THR disconnected resulting in the THRUST LOCK condition on #2.

ELAC

[slip and turn]

Inconsistency Alert?

The TLA sensors sense the angle. All that other stuff is done by FADEC, A/THR etc.

TL position is an extremely important part of that logic - all explained in previous posts.
TP

The first statement, TP, looks a bit dismissive. The second (true) statement which I am guessing at least some current Airbus drivers have woken up to only by reading this thread or reports of the accident, seems to fly in the face of the first, don't you think?

[ELAC]

4. This is the most puzzling point : the left sidestick was then left alone, the FDR trace for it remains abnormally flat...on the other hand, the right sidestick becomes quite lively - even going for full nose-down application in the last seconds ... The CVR never showed a control transfer, did it ? This remark is borne by the CVR transcript - if it is to be trusted - that the "vai...vira.." utterrances come from HOT1 (the Captain's mike ).

Lemurian,

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. The PM in the right seat on the other hand was new to the equipment so would likely be less certain of what inputs would be helpful and what not, and also was stuck in the unenviable position of being an observer to a situation that was rapidly becoming fatal. In those circumstances grabbing and moving whatever was at hand in the hopes of altering the result seems like a natural reaction to the situation. That's how I read that bit anyway.

ELAC

[bsieker]

ELAC,

thanks, this seems a reasonable explanation for the thrust variations prior to disconnection of A/THR.

I am still not quite sure I fully understand the explanation in the Taipei-Report. 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?

I also noticed in that report that some terms are different from the terms used in the FCOM, likely this is a translation problem.

Bernd

[marciovp]

An expert

The newspaper Estado de Minas (Belo Horizonte) brought six graphics (DFDR) and voice transcription to a north american expert Mr. William Waldock from the Aeronautics University Embry-Riddle in Prescot, Arizona.

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”.

He did not know the reazon for this.

“I suspect that the investigation will spend much time in these seven seconds and in the effects they produced in the in the control of the automatized flight and the acceleration systems of the Airbus”.

He claims that the TAM airplane had several anomalies in its measurement instruments, especially in the ratio of pressure in the motor (EPR), the angle of the TL (TLA), the vibration of the motor and speed of the plane. The automatic pilot was disconnected at 18h47min53s only 33 seconds before toucing down. For him, the landing wheels touched down at 18h48min24s and the front wheel at 18h48min27s. The system of command of autoacceleration (ATS) was disconnected two seconds later, For the experct this is a sign that the pilots disconnected only a part of the automatation of the Airbus. Both TLs were at 24 degrees, position for taking off, until the touching down At the moment of the touching down the angle of the left TL was brought back to idle but the right one remained at 24 degrees. For him the lack of movement of the right TL would be consistent with an automation still partially active that responded to the disagreement between the TLs.

The response of the EPR, a vibration of the motor of the Airbus and the flux of fuel also raise questions to the expert. The motor 1 (left) indicated idle or low potency. The EPR of the left motor increased 1.27, the vibration increased the flux of fuel also became more intense as if the motor was was responding to na increase in potency three seconds before the system of command of auto acceleration stopped working.

He ends by saying that there are many causes for an accident, that in the end the co-pilot pulled the stick back in na attempt to save the plane. He believes that there may be human error but he also believe that there are operational questions, included that sudden descend seven seconds before touching down,

(Sorry for the translation, in a hurry...Hope this is helpful, at least interesting...)

[ELAC]

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

marciovp

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

[bubbers44]

The reason for the increase in descent rate is quite obvious. The pilot flying asked for the pilot not flying to inhibit the glide slope warning so at that 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.

[marciovp]

Expert

Elac
Coming from you this clarifies the concern of Mr. Waldock. This is not the first time I come accross him from that University. It seems that he indeed is considered an expert and if I am not mistaken he has geen studying the Airbus planes for many years. Will check and see what I find. But what you and bubber said made sense.

Here is Waldock:
http://transcripts.cnn.com/TRANSCRIP.../23/bn.10.html

http://www.projo.com/cgi-bin/include...t990/1102d.htm

[PAXboy]

Lemurian

I find it quite extraordinary that with all the data the FDR transcript provided, we are still spending time discussing TLA.

To the outside observer (such as myself) it looks as if those posting on PPRuNe are being loyal to the pilots and not wishing to have the blame ascribed to the dead men, until due process has been followed. I might also surmise that they can imagine the situation where they have died and would not want their families and friends to hear them being blamed right away.

And now comes the boring "Can't happen on a Boeing" argument ... Enough of that as it brings no value to this discussion.

A further suggestion, everyone wants to believe that they are in an inherently safer machine. That is human nature.

[bubbers44]

I'm not sure what classifies a person as an expert but most of my approaches to my Honduran airport in a B757 were at 1200 fpm descent to flare so don't feel that is as excessive as he does. I think everything was fine until they touched down and the #2 engine spooled up either because the pilot didn't bring the TL back to idle or the on board sensors didn't detect that he had brought it back to idle. They couldn't override the ground spoilers not activating or reduce power on the #2 engine so they crashed. Now what is Airbus going to do to prevent this from happening again?