Hacker News Reader

Comment by kalleboo | Hacker News Reader

I agree with most of your points in general, but Boeing tried as hard as possible to pretend MCAS needed no training. There WAS no simulator scenario to train on for this failure.

View on HN

Based on accurate descriptions of the intended scenario for MCAS, they would be right. When the system is working, it does what it was designed to do. The failure mode when an AOA sensor goes on the fritz is the problem. The proposed fix is what they should have shipped out with the first plane, that's the real failure here -- someone at Boeing surely saw this coming and raised a flag, and got shouted down. A reckless decision given that the 737 is a cash cow and historically such a successful line of planes that I imagine Boeing would be keen to keep the reputation intact.

Sort of with you, but I think once you're willing to admit that MCAS failure can be catastrophic, airworthiness regulations may require three AoA sensors (as Airbus planes have) so that you get error correction, rather than the two that should have been being used to detect disagreement (but weren't).

I think the issue is that Boeing didn't see a potential MCAS failure as really any different from any other cause of runaway trim, for which there was well-established procedure and training. Clearly at least not all pilots seem to recognize this in time, particularly with other distractions caused by a faulty AoA reading, such as stick shaker, other warnings and alarms, and all of that happening at low altitude.

I don't think that matters here, though I agree with what you said. There's also going to be redundancy on the other mechanisms that can lead to runaway trim, because runaway trim can be catastrophic, because we don't assume that pilots are perfect, and because (as you say) runaway trim can happen in situations where the ground will necessarily arrive before your recovery does.

I think it takes around thirty seconds for the trim motor to move you from one end (where both crashed planes had it) of the jackscrew to the other. And that's with the motor! If you cut out the motor because it was literally trying to kill you -- and as Boeing advises you to -- you'll have to use the hand crank and it will take much longer. Runaway trim at low altitude is death.

It really seems to me that the MAX as shipped was not plausibly airworthy (due to the use of a single sensor). I think even Boeing's going to agree eventually.

A "more right" solution would have been to implement a "stick pusher" system to make it obvious to the pilot what was going on. But that would have required pilot training and punctured the illusion that the plane can be flown by any pilot who has flown any 737 made since the 1960's

Reply on news.ycombinator.com