kimoki, from various reports it seems that your doubts about this A330 lacking the FOB ECAM alert are correct. The first sign of trouble seems to have been an ECAM message concerning the oil. Several minutes later the crew received an ECAM fuel imbalance alert when the left-right difference reached 3000kg.
I think, Hugo_pipi, that the chronology may dispell your suspicions about "lazy ass" pilots.
The timeline of the incident, put together from various sources, went something as follows, with the caveat that only the software aspects interest me. I am merely repeating what I have read in trade publications and I don't claim that anything is absolutely accurate. Indeed, any needed corrections are welcome.
0010Z Air Transat Flight 236 departed Toronto.. .0457Z Last check of total fuel before ECAM alert. Fuel normal.
Working backwards from when all fuel was exhausted off-the-cuff calculations have suggested that the leak did not arise more than a few minutes after this time.
0516Z Warning of low oil temperature/high oil pressure in right engine.
This seems to have been the first indication that there was any problem with flight 236. The crew contacted Air Transat maintenance in Montreal. At this point there was no suspicion on ground or onboard of a fuel problem. With hindsight it has been suggested that cold, leaking fuel was the cause of the ECAM warning. I don't see how anyone would have guessed that inflight.. . . .0536Z ECAM Fuel imbalance @ FL390
At this point the crossfeed valve was manually opened, according to reports. I recall reading that the crew was advised to do this - maybe from the QRH, but I seem to recall reading one report that the ECAM display suggested crossfeeding. The QRH does state that the crossflow valve should remain/be closed if a fuel leak is supected. It appears that at this point a leak was not suspected.
0541Z Diversion to Azores
At this time the pilots had not yet determined that the imbalance was due to a fuel leak. OK, so five minutes had passed and so would another seven minutes but the F/O says that they were busy going through the QRH checklists. Based on the investigator's remarks I infer that he is supporting the pilots here.
0548Z Crew determines a fuel leak and declares an emergency.
I guess that after this point questions will arise about not closing the shutoff and crossfeed valves. However, during the flight the crew were not able to determine the source of the leak. Although the transducer for the engine fuel flow guage is downstream of where the leak was, it apparently did not indicate any problem. So how do you determine on which side of the transducer the leak occured? The QRH fuel leak checklist branches into two cases 1) engine leak, or 2) non-engine leak or leak not located. Case 2 calls for descending to under FL200 (to permit gravity-feed without vapor lock) and possibly shutting down an engine. That has to be a scary scenario for a twin over the Atlantic, no?
0613Z Loss of right engine.. .0626Z Loss of left engine @ FL345, about 137km from Lajes Field.. .0646Z Safe landing, under the circumstances.
If published reports are correct, then most of the blame is going to fall on the maintenance supervisor. Air Transat lacked the proper hydraulic line for the replacement RR engine they leased. RR's inspection concluded that the incorrect hydraulic pipe cracked the fuel line. It was reported fairly soon after the incident that the mechanic responsible for the installation advised against putting the plane into service and tape-recorded his supervisor overruling him.
The software problems that seem to be exposed are typical - anticipated events are handled well but other matters are not addressed at all. The software controlling the fuel system does some pretty clever things. It maintains longitudinal balance, shifting the CG around during different phases of the flight. It also attempts to maintain lateral balance despite inevitable discrepancies in burn rate. But from all the data that is available onboard, it should be trivial to determine FOB anomalies. kimoki indicates that the software has been "upgraded" to do this. I cannot think of a reason why the software did not do so in the first place. The fuel burned can be routinely calculated by integrating the measurements of the fuel flow meters. It is then simple to check this result against the total fuel decrease in the tanks as measured by the fuel tank totalizer. . . . .Here is a longer extract from the Frederico Serra interview:
"What we're discussing very carefully is whether the information provided by the computer to the crew is the best information to deal with the problem, and we have some serious doubts with that. There is something wrong with this system. There is something that was not given to crew on time. The crew cannot manage ... the situation because there's no information saying there is a leak. They didn't know what was going on. The computer didn't provide to them the best information. It was too late that the pilots noticed that the fuel was missing and they didn't realize why or where was the cause of the missing fuel."
The reporter conducting the interview asked AI to comment on Mr. Serra's remarks. AI declined, not surprisingly - they are being sued. An unnamed Transport Canada source called Mr. Serra's remarks "musings." Nevertheless, the final report will be interesting.