TAM A320 crash at Congonhas, Brazil

[AlexL]

However I still make the point, that in normal ops the 757 (and all other boeings?) the thrust lever matches the actual thrust delivered at any point n time, whereas on the airbus it doesn't.

That may be true but what has been stated in this thread on several occasions, is that: with a manual landing - in any manufacturer's machine - if you leave one TL forward, then it will give the thrust commensurate with that position.

however I make my point again, that in a boeing you would need to retard the lever to land - (you would notice if you forgot because you would be floating half way down the runway still airborne), and you need to push the lever forward yourself for the engine to spool up again.
In the Airbus, it will retard the thrust itself- without moving the lever, and then spool up again, iteself, without pilot intervention, if the lever is not moved - that in my book is a significant gotcha for a human factor error.
In this accident, surely it must have been an autothrottle approach? a manual throttle approach surely cannot have been mishandled thus? the aircraft would have floated enormously had the lever been left open.
yes you could do this on any manufacturers machine, but I say again, in my being you would have to retard the thrust lever to land, and then push it open again, a highly improbable 'mistake' to make. and an impossible one to make if reverse is deployed as the reverse levers lock the thrust levers closed.

[J.O.]

Skallas

Besides that, as a secondary thought I think the interlock preventing forward thrust during at least one engine reverse (and ALLOWING spoilers) and/or vice versa: preventing reverse (and spoilers) during one engine forward thrust - makes perfect sense and should be introduced.

There is one problem with this idea, that being the case of an undesired (i.e. accidental) thrust reverser deployment during takeoff after the go/no go decision speed (V1) has been reached. In this case, the crew would be required to continue the takeoff, because there may not be enough runway left to stop the aircraft, particularly in a "performance limited" case. An undesired thrust reverser deployment would result in the affected engine thrust automatically going to idle, thereby limiting the effects of the deployed reverser, and making the aircraft flyable. So, at a time when the crew would need the thrust from the remaining "normal" engine, you would take that thrust away. This could in fact, result in exactly the same accident as Cononghas, and aircraft at too high a speed overrunning the runway with tragic results.

I am not opposed to taking a careful and detailed look at the circumstances of this accident with an eye to improving technology to reduce the chance of a repeat event. But, as has been said above, this must be taken with great caution, because every "improvement" brings with the possibility of introducing new error-producing conditions.

[BOAC]

It seems we are arriving at a conclusion as to the basics of 'why'. This brings me back to my point about the training system not keeping up with the technology.

It seems that the 'PFM' systems on the AB are in danger of significant confusion and unless the pitfalls are pointed out, these sorts of accidents will continue. If I understand the system correctly, A/T remains 'engaged' even when the a/c is landed 'manually', and then drops out shortly after touchdown. This then places the throttles in a 'direct' position driven regime?

With this at the back of someone's mind, with a short and slippery runway to wrestle with, UNLESS the latter 'reversion' has been stressed to them they could as said above, mentally 'block out' the throttle with the u/s reverser since they are of the opinion that they will not be touching it after landing (subject, of course, to somewhat confused opinions on which MEL was active at the time.)

Some AB 320 pilots seem to know the functions we are discussing, some do not. A chilling 'prophecy' by mcdhu (who did know) on 10 June is on another thread. There appears to be a gap that needs to be closed. Hopefully the MEL revision will help to close the gap, but I feel the closure may not be complete enough with just that.

It makes me uneasy to cut and paste this. It refers to an autoland, but is relevant, I feel. My bolding.

We've got most of it I think. With 'Land' annunciated and an A/P in, the FMGC will command the thrust to Idle and 'Thr Idle' will be annunciated on the FMA. A 'Retard' calll will be made as a reminder at 10' Radalt. If you don't, it will keep shouting 'Retard' at you (with some justification), there will be no spoiler deployment and therefore no Autobrake and, as someone said earlier, you had better tighten your harness because the speed will stabilise at about 100KIAS (A319) and you are going ploughing!!

[threemiles]

Can somebody please explain what the purpose is of flying one dot below the glide slope to land on a slippery runway?
Does this in theory really help to shorten the landing run?
On a slippery runway wouldn't a hard first landing gear contact be desirable?
Thank you.

[PBL]

It is now official. Thrust lever position, EPR, airspeed at touchdown (wind 330/8 we have already heard).

One issue. Some posters here have queried why there was no warning of a thrust mismatch. Good question. In response to a recommendation in the report on the 2004 Taipei incident, Airbus developed a warning consisting of an ECAM message, a continuous chime, and illuminating the red master warning light. Looking at the CVR transcript (only preliminary), there is no chime noted. Was this SB incorporated on the accident aircraft? Anybody know?

PBL

[Right Way Up]

BOAC,
When flying the AB manually, you close the thrust levers as you would on any conventional aircraft. When the T/Ls have reached the idle stop the autothrust disconnects.

[the_hawk]

PBL, while nobody has posted if the SB was incorporated, it was mentioned more than once already together with its categorisation "desirable".

As you say, nothing on the CVR transcript so with some probability not implemented, will be in the report I guess...

[PAXboy]

AlexL

however I make my point again, that in a boeing you would need to retard the lever to land - (you would notice if you forgot because you would be floating half way down the runway still airborne), and you need to push the lever forward yourself for the engine to spool up again.
In the Airbus, it will retard the thrust itself- without moving the lever, and then spool up again, itself, without pilot intervention, if the lever is not moved - that in my book is a significant gotcha for a human factor error.

What has been stated by many in this thread is that:

• Boeing = bring the TLs back to idle at the correct moment, depending upon automatic or manual landing.

• Airbus = bring the TLs back to idle at the correct moment, depending upon automatic or manual landing.

If one TL is left in the forward position - in ANY a/c - then it will directly affect the outcome of the landing. Perhaps this will bring another discussion about two small video cameras being added to the flight deck. Many do not like the idea but they would show exactly what happened (assuming that FDR not damaged etc.)

[fox niner]

threemiles:

It doesn't shorten the landing run, but you do get more runway available in front of the airplane. Always handy on a short runway.

BOAC:

You hit the nail right on its head. as is clear from the CVR transcript, RETARD, RETARD was said by the onboard computer, so apparently they didn't pull back the thrust lever(s) far enough....right?
But why did they not pull back the TL's?
- they forgot?
- they miss-grabbed the right thrust lever, and therefore it stayed too far forward?
- they were focussing too much on the short, slippery runway?
- they had become too accustomed to Airbus autoflight behaviour and became somewhat expectant of the system to solve it for them?


I have the feeling that the airplane was doing exactly what it was designed to do. No computer glitches or latent design flaws. But you do need people in the cockpit who know what an airplane is going to do in certain situations.

It all comes down to good training.

[the_hawk]

as is clear from the CVR transcript, RETARD, RETARD was said by the onboard computer, so apparently they didn't pull back the thrust lever(s) far enough....right?

question to those in the know:

the two "FWC retard" at 21.6 and 23.0 right after the twenty call I'd expect to be standard, but what about the "GPWS retard" at 25.5? and IMO it is likely the pilots didn't hear / recognize this third retard, what do you think?

[skallas]

J.O.

There is one problem with this idea, that being the case of an undesired (i.e. accidental) thrust reverser deployment during takeoff after the go/no go decision speed (V1) has been reached. In this case, the crew would be required to continue the takeoff, because there may not be enough runway left to stop the aircraft, particularly in a "performance limited" case.

As I wrote in my original post, the interlock could work one way or another (lock out a reverser while at least one lever in forward thrust OR lock out forward thrust while one engine in reverse), based on which "decision" is considered to be safer.

Based on your comment it appears that locking out reverse while forward thrust is present is safer - even with limited runway of Congonhas, if the reverser hadn't deployed and PNF had called out "no reverse, no spoilers, no decel", the crew would have all the time in the world to initiate a go-around. Partly because they would have lost noticeably less speed by that time, partly because they wouldn't have had committed to landing because of reverse thrust - they would have had an "out".

It would have taken a while to figure out the reason for what happened, but they wouldn't have crashed. Ditto in the case of uncommanded reverser on takeoff after V1.

* * *

Actually, I was originally tempted to argue a case with statistics that even if the computer would make a bad call and decide to lock out thrust on TO after V1, then it would still be better than the current situation - because statistics apparently shows that uncommanded reverse on takeoff after V1 appears to be much rarer event on A320 series than mishandling of "the other" thrust lever on landing with one reverser inop.

But all this is irrelevant of course because the other decision (no reverse until both thrust levers retarded) (while shouting "retard, retard" at the same time) would probably indeed be safer overall. It would lean towards trying to bring pilots back into the loop instead of deciding for them. But I still think that no decision is worst choice of all.

[bomarc]

if the airbus was making a normal automatic landing (like a cat 3), would the authothrottle disconnect on touchdown automatically?

would the pilot have to RETARD the throttles manually?

IF the autothrottles were left on/engaged while being hand flown, would touchdown trigger auto throttle disengagement?

[GearDown&Locked]

I've read on this thread (can't remember the post #...) and please correct me if I'm wrong, that if you're flying in cruise and one of the T/Ls is moved to MCT and the other remains at its normal cruise position the system will command both engines to MCT. It maybe abnormal but it seems logic.
So why not implement the same logic on the other end of the spectrum: if only one of the T/Ls is retarded to Reverse, either Idle or Max, why not command the other engine to Idle, and let the plane complete its landing logic, deploying the spoilers and activating autobrakes. Its abnormal, but with the plane on the ground and a reverser engaged you're not going around.

[BOAC]

Right Way Up - thanks - I had seen that in previous posts which is why I said the copy related to an autoland. My point was about what the crew might have been subconciously 'expecting'.

threemiles - there is no point if it is done early since as Tyro has said you merely flatten the approach angle making an accurate touchdown more difficult. If you look back, however, you will see that the same a/c the previous day had been landed 500' short of the normal touchdown zone (and only just stopped). This can only be done safely and accurately by a LATE adjustment to steepen the approach, ideally in the last 200' or so, which would then of course generate a 'low on glideslope' indication although probably not a warning (737). This was a standard technique in my day when I flew highly-swept fighters to put the beast 'on the numbers' from an instrument approach.

Originally Posted by fox niner

I have the feeling that the airplane was doing exactly what it was designed to do.

- I agree.

[Rananim]

Excellent post from Skallas.However...

Of course all this doesn't mean this logic caused the accident - the biggest hole in the cheese was of course a human mistake. But lack of "retard, retard" shouts *has* to be the second biggest, preventing them from diagnosing the nature of the problem quickly.

Cant agree that repetition of retard would be a fix.The only valid fix is to deny the pilot use of his good TR.Having one engine in TR and the other producing forward thrust makes no sense.Its bad design.
Its a triple whammy..AB compound the above by having the engine in forward thrust(if the pilots forgotten to retard) actually accelerating to CLB and then deny spoilers and braking.The pilot can hear the TR and thinks retardation is proceeding as normal(in this case he was only expecting to see 1 good TR anyways) until he sees/feels that speed isnt decaying.These valuable seconds on a short wet runway are what cost lives.On a Boeing if you havent quite retarded one of the TL's then you wont get TR and you will either quickly overcome your "finger trouble" or decide to abort the landing.You havent been lured into reverse thrust with forward thrust still being produced(or actually increased as in AB) and so the option to GA is still there.How quickly you overcome your finger trouble and how short/wet the runway is will determine what you actually do.As it turned out,the skipper in SWA's overrun made the wrong decision but at least it was his decision to make.And thats the point.If you're going to go off the runway and crash and burn,you at least want to be able to say is "Holy sh**,I made the wrong decision but the aircraft tried to protect me."You dont want to disappear off the end of the runway saying"My God,what is happening here?"Its the difference between good design and poor design.
Add to this the problem of considerable swing with one engine in rev and one in forward thrust and increasing.This will only complicate an already lethal situation by distracting the pilot.Not only is the poor pilot stuck half way between a landing and an abort but now hes got a severe control problem as well.

[James7]

A320 Tl

As has been said many many times and will say again.

There are no motors attached to the thrust levers.

The thrust levers move by hand and hand alone.

Any modification to the TL regarding them to be powered back and forth will cost absolute millions and is NOT an option.

On landing the retard retard call is often made 2 or 3 times just look at the video of the 320 landing given on a previous post.

It should be heard only ONCE as this is a fixed call. TL should be at idle prior to touch down. But then none of us are perfect.

[DozyWannabe]

Rananim:
AB compound the above by having the engine in forward thrust(if the pilots forgotten to retard) actually accelerating to CLB and then deny spoilers and braking.

Nope, if I understand the logic correctly the A/THR disconnects and if a TL is in an incorrect (forward) position then the FADEC sets the engine thrust on that side to whatever the lever is set to. Initial rumour stated that the unretarded lever was at or near the CL detent, so the thrust was set to CL on that side. The way you put it suggests that no matter what position the TL is in if it is not abaft of IDLE, the computer forces CL power on that side - this is (again as I understand it) not correct.

The whole "AB is taking control away from pilots" meme should have died a death by now, because it is categorically untrue. That both TLs should be set to IDLE before touchdown, and reverse is applied has never changed.

[Edited to add : YouTube of A320 landing at Conghonas here : http://www.youtube.com/watch?v=FZ77mPgJ_Sk . "RETARD" warning is heard precisely once.

[fox niner]

There are no motors attached to the thrust levers.

Thanx James7...

Now I finally think I sort of understand the "autothrottle" system on an airbus. You move the thrust levers to CLB and then you can LET GO of them. The autoflight system regulates the required thrust and of course: there is no movement!

So here follows a translation of the various thrust lever settings on an Airbus, written for Boeing drivers (like me-self)

Idle: Hey computer, don't give me any thrust at all, just keep the engine running.
CLB: Hey computer, just give me enough thrust to keep flying and don't bother telling me exactly what you are doing.
TO/GA: Hey computer, give me all you got.

[James7]

thrust lever

Fox Niner - almost there.
The FMA tells you what is happening and of course a quick look at the N1 guages gives a good clue. Does anyone scan these anymore????
The only action that was not availabe to the pilots were the SPOILERS.
Brakes were MANUAL.
Airbus were obviously aware of this potential problem hence the FIX they put in place.
1. Service Bulletin
2. MEL ammendment.

[GearDown&Locked]

http://www.youtube.com/watch?v=3Otca...elated&search=

Working with those T/Ls...how difficult can it be?