[gargravarr]

Okay, I've never sat in the cockpit of an aircraft, but the one thing I don't understand - if the nose of a plane suddenly drops, isn't the pilot's first instinct to pull back on the controls to raise it? From what I've read, in Boeing aircraft at least, the pilot's controls can always override the automatic systems.

[JshWright]

Pulling back on the controls moves small control surfaces (elevators) on the horizontal stabilizer (the "tail wing"). The way the anti-stall measure works on the MAX (and many other planes, to be fair) is by adjusting the pitch trim. Adjusting the trim changes the angle of the entire horizontal stabilizer.

https://en.wikipedia.org/wiki/Stabilizer_(aeronautics)#/medi...

It's very possible for the effect of the extreme trim condition to be more than the elevators can overcome. The pilot can certainly override the trim setting, but the issue here is that they weren't expecting that to change, since they weren't trained on the fact that the plane might do it automatically (in those flight conditions).

Here's a good video explaining how trim works (in normal operation) on a (slightly older) 737:

https://www.youtube.com/watch?v=l62NvkRWa5E

[danielvf]

A 737 and a 737 MAX has big wheels in the cockpit that move as the stabilizer trim changes. This lets the pilots see what the autopilot is doing with the stabilizer, and/or override stabilizer changes by grabbing and holding the wheel or rotating it. You also have two switches to disallow control of the stabilizer, one for blocking other stabilizer controls in the cockpit, and one for blocking control by the autopilot.

In theory then, solving the stabilizer problem that caused the crash is as simple as flipping both switches to cutoff, then using the wheel to set the stabilizer back to a sane value. Handling a stabilizer runaway is a standard part of US 737 training, and the updated Emergency Airworthiness Directive just says to follow the stabilizer runaway checklist. Again, in theory, this should have been a no-brainer, "common" emergency, and following the usual checklist would have fixed the issue.

However, pilot's mental model of the aircraft has been broken. In the previous generation of 737's, there were exactly two things outside the wheel that could control the stabilizers - the cockpit trim switches, and the autopilot, and each had it's own cutoff switch. Now we have three systems that can control the stabilizer wheels, and the new one doesn't have a labeled switch, nor was anyone told it existed.

[VBprogrammer]

> In the previous generation of 737's, there were exactly two things outside the wheel that could control the stabilizers

It doesn't take anything away from what you've said really but there is at least one other system which can control pitch trim. The Mach trim system counteracts changes in the centre of pressure due to speed by making adjustment to the trim.

> Now we have three systems that can control the stabilizer wheels, and the new one doesn't have a labeled switch, nor was anyone told it existed.

Just to add to this point, I believe the reason this was done this way was an attempt to keep the control behaviours of the new aircraft as similar as possible to the old (despite having more powerful lower slung engines) for the sake of maintaining training consistency.

[inferiorhuman]

> In theory then, solving the stabilizer problem that caused the crash is as simple as flipping both switches to cutoff, then using the wheel to set the stabilizer back to a sane value.

https://www.youtube.com/watch?v=rxPa9A-k2xY

The issue for the Lion Air pilots was time. At 5,000 ft you don't have time to adjust the trim before you hit the ground. While I think that the stabilizer trim can be adjusted more quickly than the video shows you'd still be looking at somewhere around the order of 30 seconds end to end.

[VBprogrammer]

The elevators still have some authority, you don't have to wind it from end to end to regain control. Also, the full range of mechanical motion is only available via the trim wheel, the electronic trim doesn't drive it to either extreme (although it can move it back to the centre from an extreme).

[inferiorhuman]

> The elevators still have some authority, you don't have to wind it from end to end to regain control.

If you're at a low altitude and managing other problems will you have enough time to even recognize the problem and then wind the cranks and recover before you hit the ground?

> Also the full range of mechanical motion is only available via the trim wheel, the electronic trim doesn't drive it to either extreme (although it can move it back to the centre from an extreme).

Can MCAS drive the trim full down? What I've read suggests that if you give it time, it can.

[VBprogrammer]

> If you're at a low altitude and managing other problems will you have enough time to even recognize the problem and then wind the cranks and recover before you hit the ground?

My suspicion is that it would probably have required a well drilled crew to have handled the problem correctly. But I don't think we're up there with the extraordinary levels of piloting which we've seen demonstrated previously e.g. Flight 1549 (the miracle on the Hudson).

> Can MCAS drive the trim full down? What I've read suggests that if you give it time, it can.

I'd be surprised if it can drive it outside the normal range of electronic motion. There aren't really enough details out there for me to be sure though.

[JshWright]

> However, pilot's mental model of the aircraft has been broken.

Exactly, and critically that "break" occurred during a very busy phase of flight, when they had a lot of stuff to think about even during "routine" operations. They didn't have a lot of attention to spare to notice the runaway trim condition.

[gargravarr]

That's very informative, I didn't realise the whole horizontal stabiliser itself can be a moving part - I always thought they were fixed and trim was done using the elevator surface. Thanks for the links.

So essentially, trim changes the relative 'zero point' of the elevator, and if past a certain angle, even if the pilot applied full nose-up elevator, the altered airflow over the smaller control surface would be unable to counter the altered airflow by the stabiliser itself having been adjusted. I now see why recovery is so dependent on quickly disabling the trim system and resetting the stabiliser to 'true zero'.

[Steve44]

> That's very informative, I didn't realise the whole horizontal stabiliser itself can be a moving part

On most, if not all, supersonic aircraft they are like that. It was found that the traditional design didn't work well at supersonic speeds, there is debate as to who developed it but the Miles M52 is a likely candidate.

I've had a quick look for a decent video demonstrating it. This one shows it fairly clearly in the 5-10 second segment.

https://www.youtube.com/watch?v=AFk7kEQ5tVo

[mveety]

No, my first instinct is to give it the beans and push the nose down more because more often then not you're in a stall if that happens. Also, as a general "rule", it's pretty common for American aircraft to provide easily accessible overrides to the envelope protection stuff. I've never seen the cockpit of an Airbus, but, from what my friends have said, it's much harder to do. I don't know if Boeing is going in that direction, but I doubt they would.

[laken]

Yes, but this is where it gets a little complicated, and why this accident is so fascinating to many people in aviation...

>> if the nose of a plane suddenly drops, isn't the pilot's first instinct to pull back on the controls to raise it?

Not exactly, as the correct way to get out of a stall is pointing the nose down, in order to pick up airspeed to gain more lift. The stall horns were going off, so the pilots may have thought they were stalling. In addition, as others have mentioned, the yoke alone may be unable to beat out the trim by the autopilot.

>> From what I've read, in Boeing aircraft at least, the pilot's controls can always override the automatic systems.

Historically, you are exactly right. This is the _biggest_ reason pilots are pissed at Boeing right now. It's believed that the autopilot was turned off entirely when the plane started thinking it was stalling, and trying to point itself down. This is not present on any of the original 737 aircraft, only the MAX series, and Boeing didn't inform pilots of this.