TAM A320 crash at Congonhas, Brazil

[PJ2]

Ignition Override;

re,

There was a lot of debate about whether a given jet has certain penalties on wet runways, whether with spoilers, reversers or whatever.

I don't see the point of such debate

The debate is valuable if for no other reason than it puts information in front of a very large community many of whom may be professional airline pilots flying the same equipment. Such discussion clarifies these matters, hopefully reduces misunderstandings and may put knowledge in a pilot's hands with which to handle future circumstances.

The "debate" (don't think it was quite that) isn't about searching for answers to this accident - not at all, at least that certainly isn't my own interest; - there's investigators for that and most here know enough to wait for their work to unfold. But the interest in these circumstances is clearly very high, almost certainly for the reasons given. We cannot possibly surmise what was on the Captain's mind at this point, if ever, but sufficient research can be done regarding the planning stages of the flight to hopefully gain valuable insight.

There would be no reason not to assume that the crew was as conservative as was demanded by the situation, a situation with which we may be certain they had seen many times before.

The media speculation we see is standard fare, much as the "expert" with the PPL. It soon goes away, leaving the investigators to their sad tasks and the families to grieve in private.

finfly1;

re,

Why is the NTSB going to involve themself in a South American airline crashing at a South American airport, flown presumably by a South American pilot flying a European jet?

Although I cannot point to the direct relationship, likely it has to do with ICAO Annex 13 which provides for varying levels of official status for foreign investigative bodies. The aircraft is certified in the US and this may be enough to provide the connection, or they may be invited for certain expertise on a consultative basis. The Canadians have vast experience with "contaminated" runways (CRFI, for one tool), wet or snow-covered and may also be part of any investigation.

[manrow]

finfly1

While there is intense competition between the aircraft manufacturers across the Atlantic, there is thankfully enormous co-operation between the accident investigation bodies worldwide, and as others have posted there are pockets of expertise scattered all over the world.

Furthermore this accident, like all others, will be looked at by manufacturers and regulators alike to see if lessons can yet again be learnt for the benefit of the flight crews and hence the passengers at large.

[Major Attack]

I thing the answer is easier then that.

According to Annex 13 the representatives from the accident boards of the followin countries should be invited to any accident investigation.
1. The country where the accident happened (obviously)
2. The country where the aircraft was registrated.
3. The country where the aircraft was built
4. The country where the engines were built

MJ

[OverRun]

The two components of wet weather friction on runways are skid resistance and macrotexture. This runway lacked macrotexture because the grooving was not yet done, which would have increased the risk of aquaplaning when there was water on the runway. But it may also have lacked wet skid resistance.

In some cases of new asphalt surfacings, but not all, the wet skid resistance may be unexpectedly low for a short time (Low here is defined as the skid resistance level which would trigger an investigation for higher-risk sites).

Some research has confirmed the suggestion that these “early life” phenomena can be attributed to the presence of a film of binder that can adhere to the surface of the aggregate for a significant period of time (until the traffic wears it off).

The binder film appears to have several effects, of which the key one for this wet runway is:
(i) It prevents the microtexture on the aggregate particles making contact with the tyre, resulting in lower wet friction at higher speeds than would normally be expected.

The fact that the runway was wet for several days leads me to also speculate about some slight emulsification occurring of the bitumen. Depending on the properties of the bitumen, this could also lead to either more of this binder film or even another film on the surface. Accentuating the problem. It would be nice to see some [accurate] wet weather skid resistance measurement results on the runway.

[BOAC]

Assuming that in the second video the film speeds are the same, the speed (in frame) would appear to be over 3 times higher, and it looks as if a single reverser (No 1) is in use.

Does no-one have the relevant Notams for the day in question?

Dreadful to watch.

[skiesfull]

Certification of the aircraft assumes that for a wet or dry runway, reverse is not required. However, in determining the minimum landing distance required for the configuration and conditions, the lack of available reverse thrust means an additional distance should be factored into the calculations. I assume that TAM's MEL refers to dispatch with an u/s T/R as requiring reference to additional landing distance required?
If the weather reported in an earlier post was correct at the time of landing, then the correct landing technique should have been as follows:-
correct land flap setting, approach speed close to Vref, firm touchdown on the touchdown point (and on centreline), immediate deployment of speedbrakes/spoilers, immediate application of wheelbrakes (auto or manual),selection of available reverse. The retardation should aim to bring the speed close to taxi speed before encountering the rubber deposits at the reciprocal touchdown point.
It is the task of the investigating authorities, to determine whether or not the above techniques were not used and if that is the reason for the overrun.They may determine that other factors such as runway surface condition or precipitation/contamination were causal factors.
Too many overruns are happening because of inappropriate technique. Perhaps the professional pilots amongst us, should ask their simulator instructor to allow them 5 minutes to practise a landing on a short runway in conditions such as turbulence with moderate rain. It is the best place to hone such skills - unfortunately simulator sessions are all-too-often box-ticking exercises with no time for developing basic skills.

[Brian Abraham]

Seems as it may not be a over run as in failure to stop, but perhaps a failed attempt to over shoot. Would explain your high speed observation BOAC. From Avweb for what its worth

A320 Crash In Brazil Prompts Calls For Change
Thousands of runways around the world have inadequate overrun safety areas, the International Federation of Air Line Pilots' Associations said on Wednesday, a day after a TAM Airlines Airbus A320 crashed in Brazil, killing at least 189 people. The crew had been trying to land on a wet runway at Sao Paulo Cagonhas Airport, and took off again when they were unable to stop. The airplane cleared the airport fence and a highway but crashed into a gas station and a building before exploding into flames. The 6,362-foot runway has often been criticized as too short. Runway-end safety areas should be established at all airports with airline operations, IFALPA said, with an overrun space at least 800 feet long or an arrestor system that could halt an errant aircraft. This week's crash is now the worst air disaster ever for Brazil, superseding the death toll of 154 in last September's crash of a Gol Airlines 737 that collided in midair with an Embraer Legacy business jet. The NTSB has sent a team to Brazil to assist in the investigation.

Wonder what remarks the investigation into the 737 incident in March last year had to make re the over run.

[cwatters]

> In some cases of new asphalt surfacings, but not all, the wet skid
> resistance may be unexpectedly low for a short time.

This appears to be well known for roads - particularly by motorcyclists. Some documents on the web suggest it can take upto 3 years for the film to oxidise and be worn off on a road. I would hope technology is more advanced for runways.

[TwoOneFour]

Daftest related story: the report in Emirates Today which says EK is not changing plans to serve Sao Paulo, which misses entirely the fact that EK would actually be operating to a different airport.

[Rainboe]

One notes the following:
-It is apparent that the speed was high near the usual turn-off with high reverse on No.1 showing.
-It is thought a go-around was being attempted. There cannot have been much runway left from the point where of usual turn-off, therefore not much time to allow for transit of No.1 from high reverse to reverse idle to forward idle to power up.
-The evidence shows there was a marked swing left off the runway. If indeed a go-around from a short runway landing was being attempted, it seems highly possible only high power from one engine (the right) was likely to be available whilst No.1 would still be transiting or increasing power. At the sub-V1 speed at the time, this would be an undoubtedly doomed scenario with an uncontrollable swing.

I think the maintenance state of the aircraft is totally irrelevant anyway to the cause of this accident. Even with both reversers functioning, it is highly unlikely that a go-around from landing deep on a shortish runway and in reverse power will be a successful alternative.

[RoyHudd]

Of possible relevance, and for info please

How long does it take to clean a runway landing sections (each end) of rubber deposits? Is it an expensive operation, and is expertise generally available around the world.

And has this cleaning been done on the main r/w at Congonhas since the accident? (I presume the r/w is back in use)

There are a lot of black and slippery (when wet) runway ends in Europe, especially at certain Greek islands, which have concerned me in recent years.

[BOAC]

I concur, although judging by the 'speed' of the first a/c I would guess the viewpoint of the camera is well over half-way down the runway, so I would think the 'g/a' theory not likely since reverse is still 'in' on No 1 at that point - and up to 'off-frame'. I think the first may even have been beginning the turn-off at the end of frame, and it certainly did not appear to have any reverse spray at all in the whole visible run.

The concentration on serviceable reversers IS relevant since in all my 737 flying, landings on 'slippery' or 'contaminated' runways are not allowed with one locked out. If you aquaplane, it is only the reversers that will stop you - for a while.

So - was the runway 'slippery when wet'? Was it 'wet'. Was it 'contaminated'?

Obviously the enquiry will hopefully have all these answers but I think it is important to lay this 'reversers don't matter' trend here.

[Sky Wave]

I think Rainbow has hit the nail on the head. From the video the speed is so much higher than the first landing aircraft that I struggle to believe that he had any retardation.
With the No 2 locked out, if he had selected reverse and then changed his mind and decided to go around, I assume you would immediately get thrust on the No 2 engine whilst No 1 is still in reverse. (I know you mustn't make a go around decision after you've selected reverse thrust)

I'm not an Airbus pilot, can you move the thrust levers out of reverse before the reverse doors are closed, or is there some kind of lock?

[gonso]

Excelent post Rainboe.
There are two golden rules when it comes to approach and landing.
1) Once a go-around has been initiated, don't revoke your decision.
2) Once thrust reversers have been deployed, go-around is not an option anymore.
The one reverser in transition and the other at g/a thrust can definitely explain the large change of ground track in such a short distance.
Initially it reminded me the Girona accident. A lot. It was then a suspicion that reversers had been deployed and a go-around was attempted afterwards. We know now that the collapsed on landing nose gear, broke the thrust control cables, causing the left engine to advance to full thrust !
There is always a reason for such a deviation from the centreline or extended centreline. Sometimes it is obvious, some other times...not. We'll see.

[vapilot2004]

I think Rainbow has hit the nail on the head. From the video the speed is so much higher than the first landing aircraft

This link,as previously posted
http://g1.globo.com/Noticias/SaoPaul...4-5605,00.html
loads a page with two embedded video players. The second video player from the top titled "VEJA TAMBÉM Outros vídeos que podem lhe interessar" has a 31 second clip that shows the following:

A male reporter with playback of two aircraft traversing the screen.

The first aircraft takes about thrirteen seconds to move from the right to left side of the frame.

The second aircraft does the same in less than three seconds.

The video playback speed is virtually identical in both clips of the runway. Note the timing of the blinking lights on the buildings in the background.

[BOAC]

I would welcome an input from an AB driver on the system but I would assume the same safety interlocks are in place as on the 737?

I really do think this 'g/a' theory is a red herring, certainly in the part we see on the video.

Firstly, on the 737 the throttles are 'locked' when reverse is deployed so they cannot be advanced until the reversers are 'stowed'.

It follows (737) that g/a thrust cannot be applied until the reversers are both stowed.

There is no sign of cancellation of reverse on the video.

If we are to blame the engine handilng for the swing, is it not far more likely that 'firewall' reverse thrust was being applied on number 1 as the run continued, thus causing control difficulty on a less than ideal surface?

Are we now agreed that reversers ARE important stopping mechanisms on slippery surfaces - or is the AB MEL different? I shudder to think of new pilots thinking it 'doesn't matter'.

Regarding the runway, Rippa in #126 suggests it was 'wet' but not 'contaminated'. Was it 'slippery when wet'?

[NigelOnDraft]

I would welcome an input from an AB driver on the system but I would assume the same safety interlocks are in place as on the 737?

I am sure the interlocks are there in pratice i.e. what you get. However, the physcial interlocks asre just with the rev lever selected the thrust lever can only go in the rev range.

It follows (737) that g/a thrust cannot be applied until the reversers are both stowed.

AB you can select GA Power on both from full Reverse in 0.25s just by shoving everything forward... However it will give you idle in that engine until the reverser buckets are stowed. With 1 Rev locked out and below VMCG this will naturally make your eyes water

Are we now agreed that reversers ARE important stopping mechanisms on slippery surfaces - or is the AB MEL different?

I am sure AB MEL says Wet Ldg Perf affected...

Regarding the runway, Rippa in #126 suggests it was 'wet' but not 'contaminated'. Was it 'slippery when wet'?

How could they know? It was a new surface and had hardly ever been 'wet' before?

[Rainboe]

I was under the impression that 737 thrust reverser interlocking only applied on individual engines. I'm not sure whether you can have one still out of forward idle and one at GA thrust- it's not exactly something you get any practical experience with! A lot of reports appear to substantiate that a go-around was being attempted and the plane was at high speed. It reportedly crossed the major highway (a bit like the M1 next to East Midlands) without touching it and impacted the building more or less directly. A significant left deviation took place, stronger than I would have expected from asymetric reverse only. I can't help thinking they were in such a deep hole they took the only way out they could possibly see- to try and get airborne again. But that is moving into an area I don't like- a bit too speculative maybe.

[ChristiaanJ]

The runway surface was new, not yet grooved, with a rubber deposit and wet, so the runway surface condition is likely to have played a role.

However, the runway had been re-opened a fortnight or so ago, and it had been raining for the last few days.

So it seems unlikely this was the first A320 landing in exactly those conditions. In other words, there must have been concomitant factors.

[stagger]

The BBC have inferred that - since the aircraft appears to be travelling faster than other aircraft along the part of the runway covered by the video - it was therefore a "fast landing"

Brazil jet in 'fast landing' film

Surely the video doesn't tell you much about speed at the threshold - and instead implies lack of retardation?