Originally Posted by
Cows getting bigger
Dare I say it, but the differences between A & B philosophy really show here. In the early days of A, I was like many pilots - dubious and wary of automation. Over a generation of flying later we seem to have got our head around they fact that HAL(A) has matured, is now rather good at his job and, generally speaking, our side of the house only gets into trouble when we ignore HAL(A). Meanwhile, HAL(B) appears to still be at school and capable of having hissy-fits, just like my teenage daughter.
I think that is unfair to Boeing - B has had a proper grown up HAL since 777 and many think it is superior to the A model (certainly it came after and learned from it). From an engineering point of view (I accept that pilots may have a preference) there isn't much to choose between C* and C*U, sidesticks or yokes - both systems work and are well proven now.
What
was clear to this engineer was the difference in philosophy between A & B Hals - HAL(A) will protect the pilot from him/herself overriding control inputs because HAL(A) knows best, while HAL(B) will warn pilots and make difficult what it considers as "incorrect" control inputs it will always allow the pilot to stuff it up if they really want because the pilot knows best.
The hissy-fit system HAL(MAX) seems to me to overturn the old B philosophy and head much more towards A, but worse, it decides it knows best (where best is mistrim AND to the stops at high speed and low alt) based on a single AOA sensor rather than the multiply redundant sensor sets of HAL(A) (or B).
That to me is the really sad thing - Boeing knows (or knew) how to do this stuff properly, but for some reason they decided not to for MAX. It's not just doing FBW properly 777-style, it's MCAS itself - it's looking like they ripped MCAS straight off the KC-46, except that there it uses 2 AOA inputs, so at some point in copying the system they
reduced the number of AOA inputs to the MCAS system. Why on earth would anyone do that?