Hello,

I have read that in high altitude, the A330 uses the combined data from the AOA vanes and Mach number in order to predict a stall warning activation. In case the Mach number is not available, the value below Mach 0.3 is used¹, which of course would not represent a realistic stalling AOA in high altitude operation, since such an angle decreases with the increase of Mach number.

As for the accident aircraft, I have read the whole pitot icing event lasted about one minute and five seconds². However, 52 seconds after the speeds dropped for the first time (02:10:52) the stall warning started to sound continuously. And it also sounded briefly right after the autopilot disconnected. I'm confused why the stall warning sounded in those two situations since the pitot tubes still had ice (and thus I suppose: no Mach data). The author of the book I'm using as reference mentions that in AF447 it was never the case of the stall warning sounding incorrectly³, but I became confuse with the Mach value vs the pitot icing issue in order to activate the stall warning. Please, can someone explain this to me?

¹ Palmer, Bill (2013) Understanding Air France 447. v1.07 p. 71
² Palmer, Bill (2013) Understanding Air France 447. v1.07 p. 57
² Palmer, Bill (2013) Understanding Air France 447. v1.07 p. 71

I don't know where the M0.3 number comes from. The PRIMs calculate stall warning speed using data from the AoA vanes only, plus the CG, gross weight and slat/flap position.

As I understand it... In this case, speed data from the pitot tubes was unavailable. This made air data from the ADRs invalid, so the control law degraded to Alternate Law. However the AoA vanes and the PRIMs were still working, hence the stall warnings.

Starlionblue wrote:

The PRIMs calculate stall warning speed using data from the AoA vanes only, plus the CG, gross weight and slat/flap position.

It cannot be like that in high Mach numbers. The early shockwave would necessarily make the stalling AOA lower. The plane needs to have the Mach number in order to predict the stalling AOA:

In high-altitude stabilised flight, shock waves can be seen at certain locations by looking at the upper surface through the cabin windows.

This sonic phenomenon around the wings leads to a degradation of their aerodynamic properties. This, in turn, leads mainly to a reduction in the maximum lift angle of attack as the Mach number increases, which significantly reduces the stall margin.

Source: https://safetyfirst.airbus.com/high-alt ... al-flying/ (page 3 when the document is downloaded in PDF)

But I'm still puzzled as to why there were two short stall warnings right after the AP disconnected since the three tubes were clogged. As for when the stall warning started to sound continuously at 02:10:52, the pitots 1 and 2 were already back according to this animation: https://www.youtube.com/watch?v=c6c-MbrVzs0, so I don't see anything strange in the stall warning working from that point onwards.

Disclaimer: I don't know if the video I posted is indeed an official BEA animation of the accident.

MarceloJenisch wrote:

Starlionblue wrote:

The PRIMs calculate stall warning speed using data from the AoA vanes only, plus the CG, gross weight and slat/flap position.

It cannot be like that in high Mach numbers. The early shockwave would necessarily make the stalling AOA lower. The plane needs to have the Mach number in order to predict the stalling AOA:

I'm not disagreeing, but I read that off the Flight Crew Techniques Manual. Granted there may be a great deal of nuance not included in the pilot manuals. And certainly the PRIMs have access to air data.

MarceloJenisch wrote:

In high-altitude stabilised flight, shock waves can be seen at certain locations by looking at the upper surface through the cabin windows.

This sonic phenomenon around the wings leads to a degradation of their aerodynamic properties. This, in turn, leads mainly to a reduction in the maximum lift angle of attack as the Mach number increases, which significantly reduces the stall margin.

Source: https://safetyfirst.airbus.com/high-alt ... al-flying/ (page 3 when the document is downloaded in PDF)

But I'm still puzzled as to why there were two short stall warnings right after the AP disconnected since the three tubes were clogged. As for when the stall warning started to sound continuously at 02:10:52, the pitots 1 and 2 were already back according to this animation: https://www.youtube.com/watch?v=c6c-MbrVzs0, so I don't see anything strange in the stall warning working from that point onwards.

Disclaimer: I don't know if the video I posted is indeed an official BEA animation of the accident.

The stall warning is produced by the AoA vanes, not the pilot tubes. You will get a stall warning regardless of pitot tube condition. Those stall warnings could have been a result of sharp maneuvering. Given the lost air data, the aircraft was in Alternate law so there was no high AoA protection, and it could be stalled.

Starlionblue wrote:

The stall warning is produced by the AoA vanes, not the pilot tubes.

Certainly. I'm aware of this. I have just mentioned the pitot tubes because of the Mach number that is needed to calculate the stall AOA in high altitude.

You will get a stall warning regardless of pitot tube condition.

Yes. However without Mach data, it would be that 0.3 I have mentioned (in high altitude this would produce a delayed warning). In the official accident report, page 44, this information is also avaliable.