OK, digged out a bit, here is the hypothesis (in french) the source being a
comment on HMC's blog.
Durant le givrage des sondes pitot les vitesses dégringolent et l'altimètre perd 400' ( mais c'est une fausse indication ) On est donc dans le cas ou le bird, le vecteur vitesse, le FPV n'est pas valide comme indiqué partout dans la doc et tout particulièrement celle d'Air France et même son beau FCTM Airbus customizé AF......
Bref le PF voit le bird tout en bas de l'horizon, il cabre pour tenter de le remonter, perdant complètement de vue l'instrument de pilotage normal qu'est l'assiette. Il est obsédé par le bird qui ne veut pas venir vers le haut, et donc il tire il tire il tire......
Au bout de cinquante secondes, l'avion atteint les 37 500' et son énergie potentielle en a pris un sacré coup, mais les vitesses redeviennent normales et le bird est alors sur l'horizon . Fantastique ! les PNT de l'AF 447 ont donc mis l'avion en palier avec le bird sur l'horizon, mais avec une assiette = incidence = 16° situation qui ne peut perdurer mais qui explique l'action continue à cabrer, car le bird s'effondre, la pente passe harmonieusement de zéro à -25 . L'alarme STALL hurle toujours dans le cockpit et la vitesse passe alors sous soixante kts. L'alarme Stall s'arrête alors, car l'incidence n'est plus considérée comme valide, et le bird disparaît car plus valide non plus. C'est le moment ou le message FPV est envoyé par l'ACARS.
Ensuite plus de bird, et une bande rouge noir et blanc ( black and red Barber Pole ) est visible sur toute l'échelle des vitesse.
Mais ce n'est pas la black and red barber pole de VMO /MMO, c'est la bande basse vitesse c'est à dire la black and amber barber pole de la Loi Normale, qui est remplacée par une black and red barber pole de l'alternate law...
Alors évidemment le PF se croit en survitesse .... , car tout est rouge et noir.......
Rough translation
During the probe pitot icing, speeds decay and the altimeter loses 400ft (but it is a false indication). Therefore the bird (velocity vector / FPV) is not valid as shown throughout the doc and especially that of Air France - even AF's customized FCTM......
In short, the PF sees the bird at the bottom of the ADI, he pitched up in an attempt to "raise" the bird, completely forgetting the pitch angle which is the basic indication. He is obsessed with the bird that won't come up, so he pulls, pulls, pulls on his stick.....
After fifty seconds, the aircraft reaches 37,500ft and its potential energy has taken a big hit, but the speeds are back to normal and the bird is then on the horizon. Fantastic! AF 447 crews have therefore the aircraft level with the bird on the horizon, but with a pitch = incidence = 16°.
This situation cannot be maintained but explains the pull-up action.
Then the bird collapses, the slope goes smoothly from zero to -25. The alarm still screams 'STALL' in the cockpit and then the speed passes under sixty knots. Stall alarm then stops, because the incidence is no longer considered valid, and the bird disappears as no longer valid either. This is the time the "FPV" related messages are sent by ACARS.
After that, no more bird available, but a red black and white ("black and red Barber Pole") is visible over the entire speed scale.
But this is not the black and red barber pole VMO / MMO, it is the low speed band that is the black and amber barber pole when in Normal Law, which is replaced by a black and red barber pole in Alternate Law ...
So obviously the PF believes he's overspeeding .... Because everything is red and black .......
I haven't had time to cross-check with the report if such a scenario is possible, given the known facts (but I got doubts following CONF iture's last post).
If someone want to cross check and/or react to this hypothesis, feel free.
I thought it was interesting in that if could explain the initial pitch up reaction better than with saying the pilot followed (but only
partially) the UAS procedure (memory items) for low level flights.