CVR suggests spoilers did not deploy
 

"According to transcripts read before a congressional commission investigating air safety in Brazil, the pilots were unable to activate the spoilers — aerodynamic brakes on the plane's wings — as it sped down the short, rain-slicked runway."
See http://news.yahoo.com/s/ap/20070801/...il_plane_crash for the full news story.

Reading the posts in this thead I get the feeling that the runway state is becoming less and less relevant as to the cause of this accident.

The fact that one thrust lever/engine was producing climb thrust and the other was in reverse, caused the computer logic to disarm the autobrakes, and kept the speedbrakes/spoilers retracted....

All of this on a wet/slippery, short runway.

But suppose this happened on an 11,000 foot runway in dry weather, (with the same pilot actions whatever they were) one engine in CLB and the other in reverse would still lead to a serious situation......It would veer off to one side, unable to stop, (same as at Congonhas) and would continue through the grass until the gear snapped away. A sort of runaway plane.

@fox niner
 

Yes, I fully agree.

Spot on.

Regarding the bus...
 

There's absolutely no way to deploy ground spoilers manually on A320!! One can only arm them and wait for automatics to their work once the activation conditions are met. Manual deployment some PPRuNers wrote about would only deploys spoilers 2,3 and 4 in speedbrake mode and that only if config full (i.e. maximum extension of flaps and slats) was not used for landing. Ergo, this (hypothetical) procedure is unapproved, potentially hazardous and spectacularly ineffective.

In landing phase, if spoilers are armed, they will deploy as soon as both main gears have touched down and both thrust levers are at idle. If they are not armed, they will deploy when at least one thrust lever is moved to reverse but other one has to be at idle or below. Also there's partial extension mode when only one main gear strut is compressed and reverse is selected - it leads to compression of other strut and therefore to full extension. Allegedly this mode was introduced after Warsaw accident and because if it, my company SOP is to always select at least idle reverse.

Christmas trees on wheel and f/ctl pages are not analog indicators - they don't show the angle of deployment of the related spoiler - only if it's out more than 2.5°.

Apart from reverse detent, all other detents on A320 TLs are "soft", pilots need only to apply small to moderate force to move levers out of or through them.

MAX autobrake is RTO mode only because there's would be no delay between touchdown and full brake application. Main risk here is not overheated brakes, but smashing the nosewheel into tarmac and possibly moving NLG assembly into cockpit. For maximum stopping performance procedure is to touchdown, select reverse simultaneously with settling the nosewheel on the runway and use maximum manual braking.

Only time when thrust cannot be controlled by thrust levers is when TOGA LK is activated but to activate TOGA LK you have to fly way below VLS (airbusspeak for Vref) - not very likely to have happened here.

There are two pick-off units for each thrust lever angle. Failure of one has no major effect on thrust control. However if their measured angles disagree or they both fail, FADEC automatically selects idle thrust when airplane is on the ground.

Provided it wasn't disconnected by bringing thrust levers to idle during flare, autothrust automatically disengages on ground after landing and it cannot remain in speed mode. Possible complications of landing with thrust levers out of idle are a) engines trying to match TL demand or b) getting THR LK - engines maintaining last ATHR commanded thrust while the TLs are in CLB detent. During autoland ATHR will comand idle thrust, I'm not sure about its behaviour during manual landing as it's against my company's SOP to use ATHR while flying manually.

For just a moment, disregard fly by wire planes.

imagine, landing a conventional jet transport and the right throttle is jammed at 60percent power

the other throttle is normal.

while you have to stretch your imagination a bit, there is immediate tactile warning that something is wrong.

but let's continue for the sake of argument.


a pilot who realized what was going on, could use the fuel shut off lever/switch to kill the runaway engine...with a larger dryer runway, it still might have been a crackup, but a more survivable one.

I think it will boil down to this:

using configuration full and some problem with air/ground logic (wet runway?) landing spoilers and speed brakes wouldn't work...

less runway to work out the problem


less room on the side of the runway when the assymetric thrust took the plane off to the side.

if this had happened at KIAD for example, I think some would have walked away from this crash.


could something have jammed the throttle in the detent? something as innocuous as a writing pen or pencil? stranger things have caused crashes.



IF one thing is accomplished via this thread, I hope all pilots of all types will consider what to do in a runaway engine situation with limited time to THINK.

(if I may: either firewall both and go around and think it over, or FUEL CUT OFF- OFF on runaway engine)

Very little training if ANY is done for this odd scenario.

The only training I had for this scenario was for a very obscure turboprop some 22 years ago. Nothing for transport jets.

Please understand this is not yet an indictment of the airbus , just something to think about.

I too think its scandalous that the brazilian 'authorities' are conducting the investigation in this manner, but what do you expect - look at what they did in the 737 / embraer mid air - it was caused by air traffic control, and they managed to blame it all on the only non-brazilians involved, before the investigation was even close to being complete. ICAO should indeed look at severe sanctions - If this were africa then that countries airlines would be banned from flying in other countries. Why should this shower be any different.

Anyhoo I digres.
Its fairly clear that the accident was caused by lack of spoilers, which prevented effective braking, but of course the $6m question is why no spoilers?
I still can't see why a pilot would leave a thrust lever forward on landing, it is such an un-instinctive thing to do.
Has anyone actually said that the lever was in the climb detent? or are we assuming that because of the airbus memo and the statements that it was not at idle? i,e did he pull it back, but not quite all the way? If he was only intending to pull one back into reverse, I can see how the twisting moment with the hand on both levers could push the other one forward slightly,
How far out of idle does it have to be for the logic to prevent the spoilers actuating?

I assume that a computer error is highly improbable? I guess there must be multiple independent sensors on the TL positions and some form of 'voting' done by the computer in the event that one sensor was duff.

unfortunately it is the airbus throttle design that gives you no time to think, in such a situation with a pen or something jamming the lever. In any other aircraft with moving levers such a problem would be discovered at TOD when the levers closed -simple to resolve - just open them up agian, fly around a bit and sort it out. Worst case - shut one down and execute a 'standard' one engine out approach and landing.
Discovering such a problem on the runway is only ever going to end one way.

Official preliminary CVR transcript (english)
 

Thrust Lever "Stuck"
 

One explanation for the Pilot to say -"its stuck, it wont move etc" is because the pilot has firmly gripped the 'Stray' No1 TL and is exerting too much of a sideways force on the lever thereby Jamming it in position.

A Flight Attendant on the manchester air crash was unable to open a door because he put too much inward force on the handle. The door handle then became jammed with brute strength. Under the circumstance I would imagine the force to have been way above his normal strength.

The same logic could apply in this case.

Retarding only one Thrust Lever

A BIA aircraft coming into land on runway 08L at Gatwick. The a/c was perfectly lined up with the r/w and on a clear night. The PNF then mentioned that Gatwick was using the Secondary runway. The PF then thought how can this perfectly lit runway be the 'Secondary' runway. He then manoevered the plane left to line up with the green lights thinking that this looked more like a 'Secondary' runway, and landed on the taxyway.

The PF here on getting the call 'Reverse number 1 only' could then have moved his hand off the no2 lever and concentrated on the no1 TL only.

This is just a thought and I guess we will never know the real reason why on touchdown both levers were not retarded. However there must be a reason.

Jim

@James7
 

Please read the preliminary transcript above. Contrary to what has been said there is no such indication of a stuck TL for the time being.

TAM 3054 was a manual ldg in config full with headwind 330/8.

the runway still very much plays a role here
 

The runway still very much plays a role in this disaster. The pilots could have been so focused on the fact that it is very short and dangerous that they made an initial relatively simple error of not retarding both TL. Then not realizing in time exactly what was happening and not being able to take steps to fix the problem in the seconds they had remaining.

The Transasia pilots in the 2004 TAipei mishap never did figure out that TL on number two was not pushed back. The captain just manual braked and at the end of the 9000 some foot runway ran off into the grass still going over 60 knots. The PF (copilot) actually asked the captain what was happening to a reply along the lines of "I dont know". They were saved from a Congonhas type accident simply by being on a runway that was long enough to reduce most of the speed and with having enough over run to come to a complete stop with no injuries.

Problems during landing allow only seconds to fix any problems and live to fly another day.

@config Full
 

Yes you are right - It is only preliminary.

alex l

yes, you understand my concepts...imagine a system that works perfectly, if everything is perfect, but is flawed when a human being gets involved. a hand that is too strong from primal fear and jams something?

I recall a cable breaking on a tailcone exit ...perhaps a flight attendant with super human strength due to fear?

perhaps the engineers have something new to think about.

Note, for those who prefer to obtain data through a more common channel, that the NTSB's involvement in this investigation appears in their list of Foreign Investigations

The NTSB docket number appears on the draft transcript (linked at the top of the page) and appears to be DCA07RA059.

While I doubt the NTSB will release anything that the Brazilian authroities do not, they will of course release info in English, where the Brazilians would probably use Portuguese I guess.

Well, the only thing more disgraceful than the discussion here is the publication of the CVR transcript.

http://www.rothstein.com/data/dr647.htm

Strange there is a discrepancy of +/- 2 mn between this transcript and the time recorded on the cameras in CGH.

It seems to me that all recorded time values are supposed to be in zulu time, or no? :confused:

Some questions answered...
BOAC
No latch, you just move them smoothly back to the Idle stop. It is easy to get a T/L out of a detent.
To get from Fwd Idle to Reverse Idle you lift the reverser latching lever - this sits below your finger ends on the forward face of the T/L. Lifting it allows the T/L to move back beyond the Idle stop. There is not a separate reverse lever.
By moving the T/L back - airbus SOP is either Idle or Max Reverse.
No drive, no lock.. they are easy to move. There are physical stops at TOGA, and Idle; the Idle stop is pins (IIRC) that are lifted by the reverser levers, and there is another stop at Max Reverse.
borghha
No guessing allowed .. The investigation will show what happened to the T/L .. but if both were at idle with the a/c on the ground then the ground spoilers would deploy instantaneously well before mishandling pushed a T/L forward.. your hypothesis does not stand up.
TP

I've just read the CVR transcript as posted in the link above.

I am impressed that this crew was quite professional and sounded (with regard to translation) as any professional pilots might sound.

Note that the captain asked to inhibit the glide slope and flew one dot low...a wise move on a short wet runway...I think they knew what they were doing at that airport. they were acting as any of us might have.

they discussed the reverser situation.

they were cautious about getting a runway report.

they simply sighed at not getting spoilers...this was probably not the first time this happened to them.


but then something went terribly wrong...something that might have been caught with conventional moving throttles...or by simply hand flying and hand throttling on the approach...or something as yet unknown.


I give the benefit of the doubt to the pilots on this one...and hope that we all find out the truth sooner or later.

I say again...
 

Just read the CVR transcript.

I agree, the pilots sound just like any pro's would sound. They were clearly aware of the runway state (i.e. slippery) and aware of the thrust reverser inop. Furthermore, going under the GS makes perfect sense.

The CVR transcript also has "RETARD, RETARD, RETARD" in it. This means that the thrust lever(s) were not idle, causing the autobrakes to disarm upon touchdown, and the spoilers to remain retracted. (computer logic) Therefore no weight-on-wheels while braking!!

-The approach was stabilized, no rushed approach.
-There was no time for the crew to realize what was going on...