Lion Air crash: Jakarta Boeing 737 'had prior instrument error'

[SkyMaster]

100% sure.

My point is that those switches are probably inputs to a logic system, who will later decide what to do based on 20 other inputs, all controlled by software

GeorgeOfTheJungle is 1000% correct (yes one thousand). These switchs cut the electrical power to the Horizontal Stabilizer trim mechanism.

[KL27x]

https://leehamnews.com/2019/04/03/et302-used-the-cut-out-switches-to-stop-mcas/

Pilots apparently had an air speed indicator disagree. There were following protocol for that issue when the MCAS fired.
Pilots applied runaway stab trim protocol, as proscribed by beoing, in order to disable MCAS. They cut out stab trim, according to black box analysis.

At higher speeds, an MCAS'd stabilizer can not be moved back up by hand. The pilot could not release the yoke to relieve pressure on the stabilizer in order to re-trim the plane, because this would have crashed the plane, faster. They were at only 1000 feet, shortly after takeoff.

Pilot actually broke procedure by flipping the stab trim back on, in an attempt to move the stabilizer back up with the motor into some position that would not crash the plane. But MCAS fired again. And they could not correct it in time.

The button on the yoke overrides the MCAS commands. The MCAS kicks in again (and again, that's the flaw) 5 seconds after the pilot releases the yoke's trim button, but that's plenty of time to flip the cutout switches.

These pilots turned the stab trim back on only to move the stabilizer up. They would have been pressing the up button. But MCAS fired again, and this did something.. I'm sure they didn't wait 10 seconds, watching MCAS do the opposite of what they were desperately trying to do without some combination of pressing or repressing the trim up button to try to get it to respond to their manual input while also pulling back on the yoke with all their weight to try to keep the plane from crashing.

Many (well, a few, or some) pilots in previous 737 MAX flights have saved the day by doing just that. Why? How did they know? Because after the lion air 610 crash Boeing put out a safety technical bulletin explaining just that. We're going in circles in this thread...

According to the airline, the Ethiopian Air crew received all updated training after the Lion Air crash. Per what has been discovered from the black box, it looks like the pilots followed the recommended protocol in the way they had been trained. They cut out stab trim. The erroneous AOA/MCAS might have put the plane in an unrecoverable situation, stuck between a relatively slower crash and a quicker one. Manual trim not an option. Losing altitude; can't get the nose back up.

I said it a couple days ago, that MCAS needs to be able to be deactivated without cutting out stab trim. I would also not be surprised to learn that MCAS will override the trim up button, if it fires while the button is being pressed. Whether by bug or by intention. It is, afterall, an automatic action that is supposed to prevent the pilot from doing something he is not even supposed to be aware of, that the MAX handles differently than the other 737's in a way that makes it more prone to nosing up and stalling when in high AOA. If you are supposed to be able to fly the plane the same, then a 737 pilot might be intentionally dialing in manual trim up when he accidentally stalls a MAX, in which case MCAS would be useless to prevent this scenario if it doesn't override the pilot's manual trim up button press.

If there was a way for the pilots to have saved the plane by "being smarter," then whatever it is they were supposed to have done should probably be in the emergency procedure as a memory item and practiced in this kind of scenario. And if it's not 100% reliable without a simple procedure... no complex algorithm of  if/and/then branches or "waiting to see what happens after you do w/e".... then it is probably too complicated to deal with this kind of scenario. The scenario appears to be pilots being dumped into a situation where they had maybe 40 seconds to figure out a way to get the the nose up... while at least one of them is fully preoccupied with heaving up on the yoke.

[GeorgeOfTheJungle]

Reading that, it would seem as if the trim (nose up) button on the yoke doesn't override the MCAS (nose down) commands. But reading Boeing's TB19 it would seem it does. That would be an additional undisclosed surprise/problem to confuse the pilots.

But still they could have grabbed and stopped the trim wheel by hand to avoid more nose down trim: there's a reason those wheels are there next to them and have a clutch.

The trim wheels have retractable handles (for a reason too!), if it's true that not even both pilots in unison using the handles can apply enough force to trim back to nose up, then that plane has even more serious problems yet.

[KL27x]

if it's true that not even both pilots in unison using the handles can apply enough force to trim back to nose up, then that plane has even more serious problems yet.

I think it's a matter of altitude. At a higher altitude you could let go the yoke/elevators and allow the plane to nose down for a bit while trimming, to take some of the air pressure off the stabilizer. An AOA/MCAS malfunction at low altitude can perhaps put the plane in a spot where you're stuck between a rock and a hard place.

But still they could have grabbed and stopped the trim wheel by hand to avoid more nose down trim: there's a reason those wheels are there next to them and have a clutch.

The first time this happened, the pilots were already in an unusual situation, right after takeoff and having an airspeed indicator malfunction. They were applying protocol for that issue. They perhaps didn't notice the MCAS misfire for the first several seconds. Now they are stuck with a low nose at 1000 feet and dropping; say they had 40 seconds to get the nose back up. They cut the stab trim. They try to re-trim the plane, manually. No go. They are still nose down and have 20 seconds left to figure something out. "Turning stab trim back on is against protocol, but do you have a better idea?" They turn the stab trim back on and press up... but it goes the wrong way. Say they successfully grab the wheel to stop it from turning down further. They're still up shit creek without a paddle. 10, 9, 8...

I think the aerodynamic issue might be worse than many of us imagine for Boeing to increase the MCAS response from 0.6 degrees to 2.5 degrees just to avoid stalls. I wonder if the MAX is recoverable at all, in the case if it ever did stall (at say high altitude).

[Nusa]

https://leehamnews.com/2019/04/03/et302-used-the-cut-out-switches-to-stop-mcas/

If only people were intelligent and respectful enough to actually pay attention to the most important statement in that article, rather than spouting their own less-expert speculation:
We refrain from speculating more on the limited information we have available. What exactly happened in ET302 will be revealed by the preliminary report, which should be issued within days.

The trim wheels have retractable handles (for a reason too!), if it's true that not even both pilots in unison using the handles can apply enough force to trim back to nose up, then that plane has even more serious problems yet.

If that's actually true, that would be a fault with ALL 737's, not just the MAX. Remember the runaway trim procedure was written long before the MAX existed. There's many thousands of them out there, so I doubt we'll see a precautionary grounding. Southwest would be out of business, and other US airlines would be seriously crippled.

[Gyro]

https://www.bbc.co.uk/news/business-47812225

Ethiopian Airlines Boeing 737 pilots 'could not stop nosedive'
...
Pilots "repeatedly" followed procedures recommended by Boeing before the crash, according to the first official report into the disaster.
...
"The crew performed all the procedures repeatedly [that were] provided by the manufacturer but were not able to control the aircraft," Ms Dagmawit said in a news conference in Addis Ababa.

[djacobow]

https://www.bbc.co.uk/news/business-47812225

Ethiopian Airlines Boeing 737 pilots 'could not stop nosedive'
...
Pilots "repeatedly" followed procedures recommended by Boeing before the crash, according to the first official report into the disaster.
...
"The crew performed all the procedures repeatedly [that were] provided by the manufacturer but were not able to control the aircraft," Ms Dagmawit said in a news conference in Addis Ababa.

The ET investigators definitely are saying this, and they have the data and we don't. However, it doesn't quite make sense. Why would anyone repeatedly put the stab trim cutout switches into cutout if someone was not switching them back on?

Story is more complex than that statement and, per usual, press is not asking informed questions.

[Nusa]

Read the 33 pages of the actual report, rather than trying to drag facts out of the news summary:

http://www.ecaa.gov.et/documents/20435/0/Preliminary+Report+B737-800MAX+%2C%28ET-AVJ%29.pdf

[MT]

Im not ranting besides at armchair pilots like your self

I've been an actual pilot since age 14, but not as a job. Also, I was once involved in MD-11 cockpit training of airline pilots. I suspect I have a better handle on reality here than actual armchair pilots.

Thats worse then i thought, you "almost" a real pilot arguing like armchair pilot, i prefer what real 737 pilots reports in the NASA pilot database! I also notice how you move goal posts and avoid debate the real issues brought up.

So called preliminary crash reports has been done in the past and been found utterly faulty and politicized, fat lady sings when final report is done and then barely.

[floobydust]

... However, it doesn't quite make sense. Why would anyone repeatedly put the stab trim cutout switches into cutout if someone was not switching them back on?

The switches might have gotten flipped back and forth, if the crew thought the cutout was not working or they couldn't move the trim wheel. There's not much else you can try when the emergency procedure is not working and you are plummeting to your death.

I wonder what the MCAS software does when things are not responding due to the cutout.
I did see a feedback block where MCAS looks at stabilizer position. If stabilizer position information goes unavailable or the motors are off-line, there's another S/W path to investigate. Integrator windup would be another.
It looks like flipping the switches back on just repeats the entire routine, starts a delay timer and then it moves another 2.5 degrees. Perhaps during the MCAS timer running, pilots have some ability to move the stabilizer and that would be a reason to keep flipping switches off and on. Nothing else is moving them out of a dive.

[Nusa]

Im not ranting besides at armchair pilots like your self

I've been an actual pilot since age 14, but not as a job. Also, I was once involved in MD-11 cockpit training of airline pilots. I suspect I have a better handle on reality here than actual armchair pilots.

Thats worse then i thought, you "almost" a real pilot arguing like armchair pilot, i prefer what real 737 pilots reports in the NASA pilot database! I also notice how you move goal posts and avoid debate the real issues brought up.

So called preliminary crash reports has been done in the past and been found utterly faulty and politicized, fat lady sings when final report is done and then barely.

The word preliminary is in there for a reason. However, it's still better information than 99% of the content of this thread concerning that flight. Have you read it yet?

And you are correct in part: I don't want to debate all the made-up theories based on conjecture and fantasy. I'd rather have facts and expert opinions.

[AG6QR]

The ET investigators definitely are saying this, and they have the data and we don't. However, it doesn't quite make sense. Why would anyone repeatedly put the stab trim cutout switches into cutout if someone was not switching them back on?

Agree that it's hard to fathom the crew toggling those switches in and out of cutout repeatedly.

I didn't see where anyone claimed they ever put the stab trim cutout switches into cutout.  The only claim I heard was that they followed the procedures.

I don't know the details of the published procedures then in effect.  Under exactly what circumstances do they require putting those switches into cutout position, and what other things are supposed to be tried first?  How much were the official procedures modified after the Lion Air crash?

[floobydust]

The final report will take 6-12 months, and Boeing is in crisis to get the planes up as quickly as possible. They will soon run out of storage space for the produced new planes, unless production is cut back which will put a further dent in the GDP of the United States. It's that big of a corporation.

This thread is full of armchair pilots, myself included.
If people want expert opinions on 737-max aviation, and exact facts pertaining to these crashes, why expect them here on an EE forum?

[Gyro]

I didn't see where anyone claimed they ever put the stab trim cutout switches into cutout.  The only claim I heard was that they followed the procedures.

From the Report link that Nusa posted (Initial findings section, last bullet point):

The crew performed runaway stabilizer checklist and put the stab trim cutout switch to cutout position and confirmed that the manual trim operation was not working.

[djacobow]

The ET investigators definitely are saying this, and they have the data and we don't. However, it doesn't quite make sense. Why would anyone repeatedly put the stab trim cutout switches into cutout if someone was not switching them back on?

Agree that it's hard to fathom the crew toggling those switches in and out of cutout repeatedly.

I didn't see where anyone claimed they ever put the stab trim cutout switches into cutout.  The only claim I heard was that they followed the procedures.

I don't know the details of the published procedures then in effect.  Under exactly what circumstances do they require putting those switches into cutout position, and what other things are supposed to be tried first?  How much were the official procedures modified after the Lion Air crash?

Well, I think there is a good question regarding what procedures the Ethiopian investigators are referring to. The stab trim runaway procedure has been posted here before and is also in the report on page 30. Pages 32 and 33 include supplemental information provided by Boeing to ET for these jets that amplify the basic procedure. This was provided in wake of the LionAir accident.

The FDR summary in Appendix 1, page 26 does not show the position of the cutout switches, sadly. I wonder if it is in the complete FDR data or because they are switches that directly cut out the motors, they might not be recorded.

However the report says this:

"
The crew performed runaway stabilizer checklist and put the stab trim cutout switch to
cutout position and confirmed that the manual trim operation was not working.
" (p25)

It's not entirely clear what they are saying. If they are saying that attempts to trim the aircraft with the trim wheel did not work after the trim motors were put in cutout, then Boeing has a much bigger problem on its hands than MCAS.

If they are saying that manual trim switches on the yoke didn't work after the trim motors were put into cutout, then, yeah, duh. The question then becomes whether the pilots tried to trim the aircraft with the wheel, and if they did, then was there just not enough time to make the adjustment, etc.

As a pilot myself, I would not be particularly scared of an automated system that moves the trim, as long as I could disable it. On the other hand, there's no way in hell I'd launch in an aircraft where I felt that manual trim might not work.

[KL27x]

I'm curious how far Airbus has gone towards full fly-by-wire. Are there any planes without manual access to trim?
In a big enough plane, it would presumably become ever more difficult to move the trim by hand. Bigger plane means bigger forces on the stabilzer. Either you have to gear it down and make a zillion turns per degree or you can't physically do it. At least not while there's significant air pressure on the stabilizer.

And you are correct in part: I don't want to debate all the made-up theories based on conjecture and fantasy. I'd rather have facts and expert opinions.

"made-up theories" Then why are you reading this thread or posting, at all?

Even if it turns out the pilots made a very obvious and egregious error that is very unlikely to ever occur again in the history of man... If there's any possible combination of events where MCAS might cause a problem despite the pilots acting appropriately, it should still be interesting to know about it. I, for one, am merely speculating scenarios that assume the pilots acted correctly. Whether or not that turns out to be the case (or if we will ever know), I would still do it.

It should also be interesting to know how dangerous it is to fly a MAX with neither power trim nor MCAS. How often does MCAS ever activate in normal flight? Never? How many times has it ever activated (not-erroneously) since the plane has been introduced. Ever? Does Boeing have some prediction or actual statistics?

[Macbeth]

Boeing admits its software to blame for 737 MAX crashes, says ‘sorry for lives lost’

https://www.rt.com/news/455587-boeing-sorry-lives-lost-crashes/

[Macbeth]

Boeing IT director: "Hey we can get all these H1-B's and offshore software devs that have dodgy qualifications and can pass cert's with a TestKing multichoice exam, it will save us a few million dollars. So worth it!"

Boeing Accountant: "A few million you say? That's a lot. Divided by a few billions, it's like a whole 10 cents! Bring in the superior offshore software developers that don't know the business model and never ever say 'no' and 'you cant do this, let me explain...' I fucking hate those guys that tell me to turn the PC off and on again. I want yes-men in my department"

Boeing Legal: "Guys, we now have a 10 trillion dollar lawsuit and hundreds of lives lost because you got the mass offshore devs and a few H1-B's in to save 10 cents"

[KL27x]

I still don't see where Boeing has admitted a software failure/error. As far as we know, it was working as was originally intended. The software engineers didn't come up with the parameters. *

The apology video was pretty good but it did not own up to anything specific.
 
*I just made myself laugh. If my clients were Boeing and I was their software guy, Boeing would come to me and say "My plane is acting funny. When you put the nose up, it like goes up and stalls. Here's the blueprint. Can you fix it with software? I can make a video if that helps. Great, you da best. Not my problem, anymore. I'll be sure to sell the plane immediately, with no testing by airplane experts or people who have otherwise ever flown an airplane, because I trust you so much and this is now a software issue, not an airplane issue."

[Nusa]

I'm curious how far Airbus has gone towards full fly-by-wire.

If fly-by-wire is the subject that scares you most, don't get on Airbus. Even the primary flight controls are fly-by-wire. Lots of software.

[KL27x]

^I'm more curious that if you can't even trim, manually, on some planes.... then the manual trimming of a 737 locking up under specific conditions might not be considered an issue in the least, if it weren't for the MCAS thing. 

[djacobow]

100% sure.

My point is that those switches are probably inputs to a logic system, who will later decide what to do based on 20 other inputs, all controlled by software

GeorgeOfTheJungle is 1000% correct (yes one thousand). These switchs cut the electrical power to the Horizontal Stabilizer trim mechanism.

Still 10000000% correct?

Yes. The pictures you show say that the MCAS (or something) commanded nose down trim, but there was no "corresponding motion" because the switches were in cutout. That's what we've all been saying.

There seems to be the possibility that for reasons still unknown, the manual trim (not manually actuated electric trim) wheel did not work. If true, I think that's a big deal -- even bigger than MCAS fucking up. It's one thing to have an automatic system mess with the trim of your aircraft, quite another not be able to do anything about it.

[KL27x]

I wonder if a strength requirement might be introduced.

Pilot must demonstrate ability to curl 10 reps of 50 lbs, 10 reps of 200 lbs of tricep extensions, peak grip strength of 100 lbs. Lat pull is tested by duration, not reps. Pilot must be able to pull back on a yoke with a minimum of 40 lbs of force for 5 minutes. 

[KL27x]

There seems to be the possibility that for reasons still unknown, the manual trim (not manually actuated electric trim) wheel did not work. If true, I think that's a big deal -- even bigger than MCAS fucking up. It's one thing to have an automatic system mess with the trim of your aircraft, quite another not to do anything about it.

According to the article I linked earlier, Mentour Pilot has acknowledged that the forces on the stabilizer can be so great that it might be possible that manual trim cannot be performed.

Apparently, the farther the stabilizer is positioned out of neutral, the harder the pilot fights against the stabilizer with the elevators, and the faster the plane's airspeed, the harder it gets. Even in the video MP posted showing the proper response to faulty MCAS, the copilot could not move the stabilizer wheel by himself. It didn't move, at all, until MP started cranking on it, too.

And this is apparently based on MP's experience as a teacher, using a simulator. I'm sure the simulator is pretty good, but it might not be calibrated or capable of providing the true/full amount of resistance.

The article I linked also shared the opinion of a 737 pilot who has apparently had the task of testing actual planes after elevator/stabilizer adjustments have been made, and he states that under certain conditions, he has to push the stick forward to intentionally dive harder in order to be able to manually adjust the trim back up. This might not be a reasonable option if a temporary nose dive results in a crash.

Hence, the context of my preceding post. MP pilot did not even acknowledge the copilot's failure to manually trim the plane when commanded/dictated.* I suppose he passed the simulation. When MP helps, he takes one hand off the yoke, so he might be relieving some force on the elevators at the same time. The plane in the simulation might have had more room/altitude for this action and/or less speed and less MCAS adjustment to fix. Under a worse condition, the two pilots might have to coordinate with a Lethal Weapon style "on 3," and maybe that wouldn't even be enough. The success might be limited by the physical attributes of the pilot/s.

*We all can imagine "don't wanna die, strength." I wonder if some fear of breaking the trim wheel stopped them from going all out, if it was really not possible. I can see either case being realistic, especially if the limits/realities of manually trimming are not explored in detail during training. I am reminded of the time I tried to teach someone how to ride a motorcycle. When I asked why he didn't let off the gas or use the brakes when gunning straight into a wall, his response was that he didn't want to drop the motorcycle. So he crashed it and put his head into a wall, instead. And after crashing, whadya know? The bike fell.

[Bud]

I wonder if a strength requirement might be introduced.

Pilot must demonstrate ability to curl 10 reps of 50 lbs, 10 reps of 200 lbs of tricep extensions, peak grip strength of 100 lbs. Lat pull is tested by duration, not reps. Pilot must be able to pull back on a yoke with a minimum of 40 lbs of force for 5 minutes.

What would female pilots do ?....