13
$\begingroup$

During Lion Air Flight 610, a Boeing 737-MAX, it seems that the angle of attack sensors caused the software to believe the aircraft was stalling, thus initiating a nose-down pitch to recover from the (non-existent) stall.

I was under the impression that the difference between Airbus' fly-by-wire system and the Boeing types, is that Boeing aircraft can have their computer imposed inputs overridden (as discussed in my previous question).

So in the case of this flight, would pulling up have not recovered the flight, and if so, why not?

$\endgroup$
  • $\begingroup$ Also, there are many variants of the Boeing 737, presumably with different behaviors in specific situations. It would be better if you specify which variant you have in mind. $\endgroup$ – a CVn Dec 6 '18 at 11:55
  • 1
    $\begingroup$ @aCVn The 737 does not have a stick pusher, only a stick shaker. Usually, only T-tail aircraft use stick pushers because they are more susceptible to deep stalls. $\endgroup$ – Bianfable Dec 6 '18 at 12:07
  • 1
    $\begingroup$ @Bianfable I didn't know that. Thanks for the correction. $\endgroup$ – a CVn Dec 6 '18 at 12:27
  • 5
    $\begingroup$ @Cloud We shouldn't need to refer to external material just to know what you're asking. Sure, the information is out there, but it's better to include pertinent information in the question itself. Basically, by making it as easy as possible for people to answer, you're more likely to get (good) answers. $\endgroup$ – a CVn Dec 6 '18 at 13:02
  • 1
    $\begingroup$ An independent analysis of published flight data plots for that accident suggested that the column force plots allegedly reveal that on this flight the pitch disconnect (intended to retain controllability for one crew member in case the other control column jammed) was triggered, rendering the pilot flying in control of one elevator panel only (the other elevator could have been controlled by the other crew member) which then was insufficient to counter stabilizer nose down moment at full aft deflection of a single control column. $\endgroup$ – Cpt Reynolds Dec 6 '18 at 21:44
10
$\begingroup$

The new 737 MAX has a more advanced pitch control system called maneuvering characteristics augmentation system (MCAS), which automatically adjusts the stabilator trim in case of high thrust and high angle of attack (AOA). This makes it very hard to overcome the down force with elevator input alone (pulling up).

In case of a wrong trim adjustment (e.g. due to an erroneous AOA sensor), the pilots have to adjust the trim manually and maybe even turn off the stab trim via the cutout switches in the center pedestal (shown below). This was apparently not conveyed properly in the MAX differences training and the FAA has since issued an Emergency Airworthiness Directive.

737 NG: 737 Stab cutout switches (image source: flaps2approach.com)

737 MAX:
737 MAX Stab cutout switches
(image source: feitoffake.wordpress.com)

So compared to an Airbus, the Boeing pilots can still have all manual control, but they have to use manual pitch trim.

$\endgroup$
  • $\begingroup$ "The new 737 MAX has a more advanced stall protection system called maneuvering characteristics augmentation system (MCAS)" — from what I've read, MCAS has nothing to do with stall protection. It is meant to add more counterforce to the column at high AOA to comply with certification requirements. "So compared to an Airbus, the Boeing pilots can still have all manual control" — compared to the NG, the MAX pilots lose capability to turn off STS/MCAS but to continue using electric trim. The image you attached does not apply to the MAX. $\endgroup$ – Rusty Core May 19 at 1:54
  • $\begingroup$ @RustyCore I agree, the term "stall protection system" is not quite correct. MCAS was added to make the MAX feel like the NG in high thrust and high AOA situations, not to prevent stalls. The image I added does however apply to the MAX (even though this one is from an NG), which has the same cutout switches. $\endgroup$ – Bianfable May 20 at 7:04
  • $\begingroup$ Um, no. The MAX has different switches: aviation.stackexchange.com/questions/64442 $\endgroup$ – Rusty Core May 20 at 15:35
  • $\begingroup$ @RustyCore The switches are identical (labeled differently) and turning both off (which is what the Runaway stabilizer checklist from the QRH says anyway: "4) If the runaway continues: STAB TRIM CUTOUT switches (both) ... CUTOUT") also results in identical behavior in both the NG and the MAX. Only turning one of them off would result in different behavior. I added a second image. $\endgroup$ – Bianfable May 21 at 8:09
  • $\begingroup$ "turning one of them off would result in different behavior" — which means they are different. Different labeling, different function, very different wiring if the schematics found elsewhere is to be believed. Turning any one of the switches on the MAX is like turning both of them on the NG, and turning off STS/MCAS while retaining electric trim is not possible anymore like it was on the NG. Boeing added a very powerful automatic feature and at the same time removed a major tool to defeat this feature. $\endgroup$ – Rusty Core May 21 at 16:23
3
$\begingroup$

So in the case of this flight, would pulling up have not recovered the flight?

"Pulling up" (pulling back on the yoke) would NOT have recovered the flight. In fact, the Lion Air pilots were pulling back on the yoke for most of the flight (see the flight data recorder's Control Column Force graph).

Can computer imposed inputs be overridden on the Boeing 737-MAX?

Yes. Pulling back on the yoke controls the elevator and would override any computer control of the elevator. The problem with the Lion Air flight is that the MCAS (which kicked in due to erroneous AoA sensor data) controls the stabilizer, not the elevator. That's why pulling back on the yoke didn't work.

The stabilizer is the entire "tail wing", whereas the elevator is just the flap on the trailing edge of the stabilizer. When the stabilizer is at full deflection, its effect on pitch will be greater than the elevator (yoke).

The stabilizer trim can easily be overridden as well (but not with the yoke). You can either:

  1. Use the thumb switch to control the stabilizer trim electrically.
  2. Hit the Stabilizer Trim Cutout switches (that would disable the automatic Stabilizer Trim, including MCAS)
  3. Manually/physically turn the Stabilizer Trim Wheel.

The problem with the Lion Air flight is that they probably did not recognize that it was the trim they needed to override, so pulling back on the yoke was futile.

$\endgroup$
  • 1
    $\begingroup$ I am confused again. This picture shows ailerons on the main wing and elevators on the tailplane: dutchops.com/Portfolio_Marcel/Articles/Flight%20Controls/… So, (1) does the whole stabilizer moves on the 737, (2) or only rearmost parts, elevators? (3) Or both? (4) Or elevators are installed on the main wing, along with ailerons? Ah, seems (3) is correct: "The elevator is used for pitching the aircraft and is attached to the horizontal stabilizer, which can also be moved by systems like stabilizer trim and autopilot trim." $\endgroup$ – Rusty Core May 10 at 7:28
  • $\begingroup$ So, when Boeing says that operation of MCAS will "likely go unnoticed by the pilot," they are lying? Because when MCAS is operating, the trim wheels are rotating? This cannot be "unnoticed". $\endgroup$ – Rusty Core May 10 at 7:32
  • 1
    $\begingroup$ @RustyCore, Yes, the trim wheels are moving during MCAS activity, but the trim wheels are around the knees. So if a pilot is looking out the window or at some other instrument, they may not notice the trim wheel moving. Also, STS moves the trim wheel, so the pilot may assume STS is causing the trim wheel activity and not MCAS. $\endgroup$ – mike May 10 at 22:53
  • $\begingroup$ Looking at the videos, the stabilizer trim wheels make such a loud noise, that one has to be completely out to not notice their rotation (I haven't been in a pilot's cabin so cannot judge). Both STS and MCAS causing stabilizer adjustment, so if a pilot notices something funky, it is not really important which subsystem caused the funkiness, because the operations to turn on automatics are the same (stab trim switch + manual crank), and it is not that specific subsystem is turned off. It seems that STS uses the same sensors as MCAS, although I don't know whether it uses both AOA sensors or one. $\endgroup$ – Rusty Core May 10 at 23:06
  • $\begingroup$ @RustyCore The stickshaker (which was going off the entire time) is much louder than the trim wheels. STS also uses only one AOA sensor. $\endgroup$ – mike May 10 at 23:11
3
$\begingroup$

Mike's answer is right. The yoke controls the elevator not the stabilizer.

MCAS trims the stabilizer, which can be overriden through the thumb switch, trim cutout switches (disables MCAS and the thumb switch), or by manually turning the Trim Wheel. (To see what that looks like, refer to the end of this video.)

To expand on Mike's answer, the preliminary report for the Ethiopian Airlines flight has been published.

According to the report, after MCAS activated, the pilots first executed option 2 (hit the Stabilizer Trim Cutout switches). They then attempted to trim manually via the wheel (option 3), which was apparently unsuccessful - this can be due to the aerodynamic forces on the stabilizer making the wheel very hard to turn. At last, they reengaged electric trim, and trimmed in the Aircraft Nose Up (ANU) direction with the thumb switch (option 1).

A few seconds later, MCAS reengaged in the Aircraft Nose Down (AND) direction for 5 seconds. 15 seconds later, the recording ends.

According to the report, they were pulling on the yoke continuously to keep level flight during most of the recording.

$\endgroup$
  • 3
    $\begingroup$ So basically, it's very hard to get rid of what MCAS has done especially when low on altitude. And not so simple as has been suggested. $\endgroup$ – ymb1 Apr 5 at 13:49
1
$\begingroup$

The best is to refer to the following website: https://www.boeing.com/commercial/737max/737-max-software-updates.page

You may read the following in the introduction:

 The Maneuvering Characteristics Augmentation System (MCAS) flight control law was designed and certified for the 737 MAX to enhance the pitch stability of the airplane – so that it feels and flies like other 737s. MCAS is designed to activate in manual flight, with the airplane’s flaps up, at an elevated Angle of Attack (AOA). Boeing has developed an MCAS software update to provide additional layers of protection if the AOA sensors provide erroneous data. The software was put through hundreds of hours of analysis, laboratory testing, verification in a simulator and two test flights, including an in-flight certification test with Federal Aviation Administration (FAA) representatives on board as observers.

To answer your question more precisely, please note pilots are trained to follow procedures. For every failure there is a corresponding checklist. In case of failures requiring urgent response, the pilots will perform the most important actions by memory and will then follow the checklist. On the 737 MAX it is a paper checklist and not an electronic checklist that is immediately displayed like on the Boeing 777.

Concerning the trim behavior related to MCAS, no clear checklist was available because the MCAS function was not available in the FCOM. Normally, in case of runaway stabilizer trim the pilot may instinctively move the column in the opposite direction of the runaway. The next action is to switch off the cutout switches in order to be able to crank the wheel manually, however if the aircraft speed is very high and the angular deflection of the trim is high it becomes very hard to crank the wheel.

$\endgroup$
  • $\begingroup$ It is not merely the lack of a checklist. Pilots of the EA302 did have updated checklist and turned stab trim switches off. They could not adjust the stab manually, so they turned the switches back on, and MCAS flew them into the ground. Supposedly, the updated MCAS is less aggressive, but there is nothing in the blurb about changed functionality of the stab trim cutout switches, meaning that similarly to the EA302 flight, pilots still cannot turn off MCAS while keeping electric trim. The situation will be less likely to happen, but if it happens it will be as deadly as it was for the EA302. $\endgroup$ – Rusty Core May 20 at 20:26
  • $\begingroup$ With the modification in case of AOA disagree the MCAS autotrim is inhibited, so we get into a degraded mode which is not very safe; it would be better to add a third AOA sensor in order to be able to isolate a faulty one. $\endgroup$ – user40476 May 21 at 2:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.