Read this: I still think it's autopilot +/- training issue instigated by this faulty stall avoidance system. Either way Boeing designed and built a faulty plane and/or provided faulty training. Boeing is at fault in some way. You cannot have two planes that just dropped out of the sky in the same way killing 346 people and the manufacturer completely escapes fault. How much responsibility can be pinned on Boeing is going to depend on its PR and its "backward dealing" ability. How much responsibility should be pinned on Boeing is going to depend on the results of the investigation which I hope will be fully revealed to the public.
So I've had a pilot specifically enter gripes on an autopilot system that was behaving in an unexpected manner, tried to replicate it in a couple of test flights, taken the pilot who wrote up the gripe up with me to replicate it, and found that they were putting enough pressure on the controls in a certain flight regime that the autopilot was pushing back on them, and if they then let go of the controls the "pushing back" resulted in the controls doing what to them seemed like an uncommanded movement but in reality was just the autopilot working as designed to correct for the error that the "loose nut in chair behind controls" was putting into the system. So the only correction required was "tightening the loose nut". I'm guessing you could find an identical gripe to the one reported by the Max pilot in some maintenance log somewhere for every single airframe flying today. You obviously try to track those to ground every time, but it's not negligence to not assume you have an entire fleet problem when one pilot writes it up. It would actually cause groundings of every fleet all the time if you did make that assumption.
Yeah but the nose-diving that happened during the autopilot shouldn't have happened still. And there were 11 "gripes" about it not just one like you sited in your situation with this particular pilot. And even a Boeing engineer admitted that this could happen during flight. It's just that what Boeing thought "unusual circumstances" brought about two crashes and claimed 346 lives.
I represented less than .01% of our fleet, if I saw it then it happened more than once! I'd have to see what the actual Boeing engineer said, that sounds like a typical garbled output when a non-technical WSJ reporter who doesn't know an autopilot from a vacuum cleaner reports on something like this. Again, NTSB does a good job. Anything else is rank speculation.
How can you say it shouldn't have happened? If the plane is trying to prevent a stall, and the pilot keeps bringing it closer to stall by pulling up while the stall avoidance system is active....the systems response going to become more and more severe. Pilot induced oscillations are a thing. From wiki: Pilot-induced oscillations, as defined by MIL-HDBK-1797A,[1] are sustained or uncontrollable oscillations resulting from efforts of the pilot to control the aircraft and occurs when the pilot of an aircraft inadvertently commands an often increasing series of corrections in opposite directions, each an attempt to cover the aircraft's reaction to the previous input with an over correction in the opposite direction. An aircraft in such a condition can appear to be "porpoising" switching between upward and downward directions.
I don't think most non-aviation folks understand that it's possible to get into a situation close to the ground where pulling up makes you crash, or that it's possible to be in an unrecoverable position while still in the air where nothing any pilot or computer does can prevent a crash.
Too bad they couldn't come up with a stall warning horn for jets like they have in Cessna bugsmashers. Cheap, effective, easy. Hell, it's how I knew I was landing properly. Just let the plane settle into the stall based on the sound of the whine of the horn, adding and reducing back power as appropriate to maintain the sound. Yes, I know, huge jets have much more momentum behind them, so they are not as snappy. Just saying. No autopilot should be allowed to over-ride a pilot's input. It should be like the cruise control in your car. Once you make an input, the auto-pilot disengages so you have full control of the craft.
Actually you get the horn and the stick shakes. Plus you get an AOA indicator so you know exactly where you are on angle of attack and how fast it's moving one way or the other, far superior to just knowing when you're in extremis. I sure as hell hope no-one ever lands a transport size aircraft with the stick shaker going! Even flying a light twin you're doing everything to airspeeds and you'll fail a check-ride sure as shit if your stall horn goes off any time besides a stall recovery demonstration. The FE will probably get out the plane and terminate the ride if it goes off when you're too close to the ground to recover from a stall! And as I've said over and over, all auto-pilots can be flown through by the pilot. You don't want it to disengage if you touch the controls, imagine flying a 0/0 approach to landing and you bump the yoke and the autopilot disengages at 25' HAT? Or you disengage by mistake in stable flight and don't realize it, automation complacency is a big deal and thinking the autopilot is on when it isn't is something that's all of us have probably experienced at least once, not fun. They generally all have bat handles such that you can swipe in their general direction and disengage as well as a disengage button on the yoke, that's worked just fine for the several billion miles flow under auto-pilot over the past few decades!
Indeed, I don't mean to infer all that about anything above the basic stuff from my miniscule amount of flight time in light singles. But the basic principles should still apply to most airframes, no matter the size. As for disengaging the autopilot by accident, it should only be disengaged when the pilot makes a "command" input, not a general "ooops I brushed the controls" thing. But you also should not have to move the yoke to the limits to do it, either. I dunno' what bat handles are, I guess that is a similar thing?
Well if they disengage when you're not expecting it while flying through it's a pretty bad thing as well. Think pushing against someone and suddenly they move aside and you fall over. You end up with pretty significant control movements, which with hydraulically controlled systems can lead to AA 587 style failures.