@sensor_validation (re
post#6628318)
Very interesting AD, thanks for the link! I haven't seen any mention of the specific fitted AoA sensors, but that AD would certainly explain AoA "stickiness" wouldn't it.
It's a relatively recent AD (originally issued Jan 2010), but they are not limiting the list of potentially affected Thales p/n with any specific manufacturing date range - so
perhaps that AD could apply to the AoA #1 sensor on AF447,
if those Thales AoA sensors were fitted.
(IMHO it also goes to show that any AoA display should display the measured values for all 3 sensors, to allow the crew to take a decision abut which are most believable, in case of discrepancies - it must not display a single computed mean / median, as that calculation could be wrong, as the AD points out.)
@A33Zab - Many thanks for confirming that this erroneous AoA #1 value wouldn't have been used by the FCPC. We seem to be lucky that only 1 (and not 2) AoA sensors were producing erroneous values here - otherwise there's obviously a
potential risk of another Perpignan-type situation, where the one "correct" AoA value gets voted out
