PPRuNe Forums - View Single Post - TAM A320 crash at Congonhas, Brazil
View Single Post
Old 28th Sep 2007, 09:15
  #2506 (permalink)  
PBL
 
Join Date: Sep 2000
Location: Bielefeld, Germany
Posts: 955
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by DozyWannabe
In all likelihood either the stenographer or a reporter misinterpreted the actual code of "FWC Standard H2F3" as "FW3". Easily enough done I guess.
"FW3" is in the final edition of the transcript. If the Brazilian parliament works like the British parliament, initial versions of transcripts are circulated to all participants for correction. Aurelio Castro would have seen it and had a chance to correct.

Aurelio Castro will have to reconcile the statement quoted above by marciovp that this SW had been available since end of 2006 with his statement before the CIP on 14 August at which he said that the "system" which he called "FW3" had not been offered to them until that date.

My take is as follows.

AI has five reasons consistent with what M. Malinge said before the CIP not to have offered the SW before.
The two main reasons offered were:
(1) The task was to consider a warning system which would remind the pilots to reduce thrust to idle on flare. But this is not necessary: there is one already.
(2) Reducing thrust to idle on flare is a basic task of basic airmanship that every pilot learns from his first flight. Introducing warnings concerning items of basic airmanship is not good practice.

The subsidiary comments were:
(3) One must ensure that a new warning will be efficacious in its effect. AI could not determine that, when a crew have already ignored one reminder to reduce thrust to idle, they would pay proper attention to a second reminder.
(4) One must consider the effect of introducing new warnings on the entire warning situation, to avoid such issues as cognitive overload (aka the "Christmas tree" effect), which serves to distract a pilot from the actual necessary task. In particular, one must be very wary of duplicating warnings.

The fifth reason would be:
(5) Nobody understands why professional pilots would not reduce thrust to idle on flare/touchdown. Until someone understands why and how this can happen, or until it can be shown to happen repeatedly or more frequently than one supposes, reasons (1), (3) and (4) above mitigate against introducing supplementary warnings.

AI's position on (5) will be that, until the Congonhas accident, there was little evidence that failing to reduce thrust to idle on flare was an oft-repeated error. There were two incidents on record in (what AI claimed in its deposition) over 30 million fleet landings in 19 years.

AI may well have surveyed or will survey their operators who run a FOQA program and keep records and maybe find some more instances which have not resulted in anything noted at the time. They will then argue that A320-31-1267 Rev 3 was offered (a) in response to a customer request (that would be TAM on, let me guess, 10 August 2007); and (b) in response to a specific survey after the TAM accident which showed that the phenomenon of not reducing thrust levers to idle on flare occurred more often than previously thought (if this should indeed be the case; that is, more often than twice in 30 million landings), mostly without incident.

I think that would be a good rationale. It coheres with a lot of what has been said here, for example the incomprehensibility of the pilots' actions in not reducing #2 to idle on flare, and the inadvisability of having duplicate warnings for the same phenomenon, as well as the efficacy of a second warning when a first has not succeeded.

Where it might be weak in a court case is if strict liability prevails in Brazilian torts. I offer the following comments
under the understanding that I am not a legal eagle and might be deficient in my understanding of some general legal principles. I am certainly ignorant of Brazilian law on these matters. Strict liability is a legal concept whereby a manufacturer is held liable simply *if something happens*, and not only if there has been a breach of due diligence or duty of care, which are usually criteria for liability to be established in general civil cases. Most of the 5 points above would serve to establish that AI has discharged its duty of care. They do not serve to establish that its airplane didn't crash for a third time due to phenomena which have occurred before, and that might matter more under strict liability.

PBL
PBL is offline