3 ms·
One would hope MCAS would not do anything if the AoA input was known bad. This looks really bad for the company. I'd be asking for FMEAs and/or FTAs on this, wh
by phkahler 8y ago
One would hope MCAS would not do anything if the AoA input was known bad. This looks really bad for the company. I'd be asking for FMEAs and/or FTAs on this, what the conclusions were, and how they intended to mitigate this exact failure - because it's got to be in there.
- zaphar 8y agoIf you've only got 2 sensors and you aren't even reading from one of them then "bad" input is hard to know. You can do nothing if they both disagree but you can't actually know which sensor is bad and which is good without a tie breaker.
- Gibbon1 8y agoWho knows but it the chatter makes it sound like the system doesn't have redundant sensors. You have two systems with one sensor each. One system is active and the other is inactive. And no voting logic.
- ddalex 8y agoYou don't need to automatically fix the disagreement. You just need to stop taking nose-down action and inform the human in charge . Problem is, if you need to inform the human, you also need to train the human, which leads to extra costs.