Originally Posted by
GordonR_Cape
ecto
Your analysis is fundamentally flawed because AOA is not valid before takeoff, and any data points prior to that must be ignored. The AOA sensor relies on significant forward speed to align the vane with the airflow, and this is not possible while taxiing.
IMHO,
Any data points before takeoff (or after landing) do not correlate with real AOA values, without significant forward speed the vane will not reflect anything resembling AOA:
Obviously correct.
These data points should be ignored?
I don't agree. Lots and lots of hints there.
For instance, a constant offset theory (like bad indexing) doesn't hold water:
If the offset were there from the power on, as in a software error, it is an amazing coincidence that the right side vane, supposedly ok, has no noise and very minimal changes on the ground while the left one (suspect) has significant noise and/or large changes in both flights. (true I'm only basing this on 3 taxi events, but each is some minutes long). As opposed as in the air, where you could not tell one from the other if it wasn't for the offset. Wind on the ground will move the vanes randomly, of course, but it is an amazing coincidence that only left side dances in all three taxiing events.
Another example. If the offset were a consequence of a wrong correction table based on airspeed being used in the left side computer, it would be an impossible coincidence that the offset remained after landing with airspeed back to zero, as we can see in the data trace of the previous flight.
I would regard any proposed theory better if it is consistent with that behaviour in the air and on the ground seen in both data traces.
In fact, due to Occam's razor, any theory should also be consistent with the rest of the alarms. Specially the FEEL PRESS DIFF. There must be a root cause for all this.
Don't you agree?