jcjeant: so, basically, Mr Christian Roger (who is he?) implies the BEA is lying?
I hope he has solid reasons to do so.
So far, the reasons he gives in your quote aren't enough, and show (IMO) a lack of understanding of the aircraft systems.
1/ the BEA didn't say that 3 ADRs were needed for the Abnormal Attitude Law to be invoked. In fact, it seems when reading IR3/Final report that one ADR would be enough.
It said: "Du fait du rejet des trois ADR par les calculateurs de commande de vol (PRIM), la loi d’attitudes inusuelles n’aurait pu se déclencher que sur des critères liés à des paramètres inertiels, mais de telles conditions n’ont jamais été réunies." // "Due to the
rejection of the three ADR
by the flight control computers (PRIM), the abnormal attitudes law could only have been triggered for criteria relating to inertial parameters, but these conditions were never met."
2/
AFAIK, the fact that the PRIMs (part of the EFCS=flight controls computers) reject an ADR's data don't imply that this ADR is OFF/FAIL. Proof of that would be:
- that the said ADRs still provided AoA values to the warning computer, which maintained S/W until the ADRs themselves outputed "NCD" values.
- that the IAS was still displayed on the PFDs, even if the ADR were rejected by a flight computer.
You have some insight of this in the §1.6.9.3 of the final report:
"A l’instar des FMGEC, les PRIM valident les paramètres qu’ils utilisent par des mécanismes de surveillance. En ce qui concerne la vitesse air, c’est la valeur votée qui est utilisée. En fonctionnement normal, il s’agit de la valeur médiane. Lorsque l’une des trois vitesses s’écarte trop des deux autres, elle est automatiquement rejetée par les PRIM et la valeur votée devient alors la moyenne des deux valeurs restantes. Mais si l’écart entre ces deux valeurs restantes devient trop grand, les PRIM les rejettent et la loi de pilotage devient alternate 2."
(it's not about the abnormal attitude law, but about the general mechanism of the agreement/rejection of a value
by the PRIMs)