Hacker News Reader

>MCAS wasn't designed because "autopilot systems cannot cope with AoA disagree." MCAS couldn't handle AoA disagree, that's the whole crux of the problem.

>- MCAS wasn't enabled at all when autopilot was. Only manual flying

Correct.

- MCAS was designed to simulate the non-Max 737's thrust curve to maintain a similar type rating FOR PILOTS. It had nothing to do with autopilot at all.

Incorrect. Ish.

MCAS has squat to do with thrust curves, it has to do wit control force curves. Namely bringing the amount of control column force at high AoA needed to achieve further increases in AoA in line with the expectations of a trained 737 pilot. The rest of that point is accurate though.

>MCAS used trim to counter-act an overeager upwards angle during key parts of the flight like take-off. It used electronic trim to accomplish this and a single AoA sensor to determine when.

Incorrect-ish.

Yes, take-off would be one of the more likely places to run into an MCAS activation, but generally speaking, even that rotation doesn't come anywhere near the AoA MCAS was meant to deal with, and to boot, flaps being engaged disables the MCAS system.

>This forced the Ethiopian Airline pilots into a difficult choice; and they ultimately decided to re-enable electronic trim and use the yoke trim switch (which is more powerful) to fix the trim. Unfortunately MCAS trimmed the aircraft further, they lost control, and the aircraft crashed.

Trim switch is higher priority, but not more powerful per se. the trim switch was clearly overwhelmed in terms of MCAS's ability to command trim changes in 2.5 degree increments five seconds after the release of the yoke trim switch as long as it detected the high AoA. Not sure whether those trim switches are continuous direct drive (I.e. jackscrew turned as long as the switch is held down) or discrete (updates the jackscrew motors to move to next position over time, only to be overridden by MCAS five seconds later).

Other than that, that looks like a fairly good summary.

Autopilot comes up several times in the report, but maybe I'm misunderstanding. Can you clarify?

At 05:38:46 and about 200 ft radio altitude, the Master Caution parameter changed state. The First Officer called out Master Caution Anti-Ice on CVR. Four seconds later, the recorded Left AOA Heat parameter changed state.

At 05:38:58 and about 400 ft radio altitude, the flight director pitch mode changed to VNAV SPEED and Captain called out “Command” (standard call out for autopilot engagement) and an autopilot warning is recorded.

At 05:39:00, Captain called out “Command”.

At 05:39:01 and about 630 ft radio altitude, a second autopilot warning is recorded

At 05:39:22 and about 1,000 feet the left autopilot (AP) was engaged (it disengaged about 33 seconds later), the flaps were retracted and the pitch trim position decreased to 4.6 units. Six seconds after the autopilot engagement, there were small amplitude roll oscillations accompanied by lateral acceleration, rudder oscillations and slight heading changes. These oscillations continued also after the autopilot was disengaged.

At 05:39:55, Autopilot disengaged,

At 05:39:57, the Captain advised again the First-Officer to request to maintain runway heading and that they are having flight control problems.

At 05:40:00 shortly after the autopilot disengaged, the FDR recorded an automatic aircraft nose down (AND) activated for 9.0 seconds and pitch trim moved from 4.60 to 2.1 units. The climb was arrested and the aircraft descended slightly.

Before MCAS started malfunctioning.

Autopilot was enabled for a short period after take-off, then disabled at 05:39:55 due to turbulence and never re-enabled.

If you look at the graph in Appendix 1/page 26 -- FDR Data you'll see that the first MCAS command occurred at 05:39:30. Once autopilot was disabled it remained disabled until impact.

The information you're quoting above is just included for informational reasons, it likely won't play a part in what caused this accident.

And if you look further through the transcripts you'll see that they kept playing with the flight director for reasons unknown.

That's commentary. We can reasonably speculate they were looking at warnings and the flight log, autopilot is only one part of the flight director and as autopilot was never turned back on, pointing at it seems odd.

That's commentary.

Changing settings in the flight director is very much an action (not mere commentary). Plus it's a good distraction from flying the plane — especially while the overspeed alarm is going off.

Maybe, just maybe, when you're in a plane that does not respond to your commands, you want to have a look at the most likely possibility: the autopilot did not disengage, or the autopilot malfunctionned and prevents you to get full control of the airplane. IF hte pilots knew about MCAS, they would likely have checked that instead, but they did not.

Most pilots are rationnal and pretty smart people. Implying that two of them are stupid enough to "play" with the autopilot in this situation is pretty insulting.

This kind of reactions are interesting though. I might tour facebook, twitter and reddit to see if it's widespread.

Overspeed is the result of the mistrim. They were in fact flying the plane, the significant back pressure they used to fight the mistrim makes that clear. On what basis do they anticipate 4 seconds after two successful nose up trim using yoke toggle switches (manually initiated, electric motor turns jackscrew) that MCAS will cause a 40 degree nose down within seconds?

Nothing. There's no basis for expecting such madness. The emergency AD following LNI610 doesn't at all account for significant mistrim, the possibility of heavy jackscrew loading forces that could prevent manual (handcrank) retrim, and the necessity of setting stab trim switches back to normal to solve the mistrim.

It's realy vile to me that this whole mistrim scenario was not thoroughly explored by the FAA and NTSB (independently of each other) in 737 MAX capable simulators following Lion Air 610. Very clearly the emergency AD was still inadequate months after Lion Air 610. Could ET302 have been prevented? Perhaps, but not based on the then published AD.

Overspeed is the result of the mistrim

No, it wasn't. Overspeed was because the pilots basically firewalled the throttles and never once touched them. Look at the Lion Air flights, they never came close to those sorts of speeds.

If I'm trimmed for level flight and go to 100% power, I will not overspeed. Initial airspeed increases, the plane starts to climb, airspeed decreases, and fairly quickly without any other inputs the airspeed will return to level flight airspeed, but I'll be in a climb. It's basic aerodynmics. Primary flight students are taught this, and in more advanced training it's given a name: positive static stability. All FAR 23 and FAR 25 certified airplanes have this behavior.

In the case of ET302, the power settings were completely normal for the phase of flight they were in. What was abnormal was the trim. If the trim were normal, they would have had a normal climb and normal climb speed. On what basis would or should the pilots have made a power reduction?

In the case of LNI034 and LNI610, you're determining speed how? Do you have a page reference? Are you even able to estimate percent Vno? What I see for LNI610 is normal takeoff power, and just before midway through the flight a power reduction. Airspeed isn't affected. And how are you comparing the stabilizer trim setting value between LNI610 and ET302?

In the case of ET302, the power settings were completely normal for the phase of flight they were in.

N1 was nearly at 100% for most of the flight. ET302 had an unreliable airspeed situation, yeah? My understanding is that the memory items for this on the 737 involves setting the throttles to well less than 100% (closer to 80-85% is what I've seen). Honestly I thought climb thrust was around 80% as well.

On what basis would or should the pilots have made a power reduction?

Both the overspeed clackers were going off for much of the flight (granted so was one of the stick shakers — at the same time).

In the case of LNI034 and LNI610, you're determining speed how?

The initial leak of the preliminary report included units with the graphs.

https://ngamotu.nz/images/20181122-jt610-knkt.pdf

Another example of Dunning Kruger effect here. I assume you're competent at something, that's as polite as I can be. But the way you willfully ignore my power and trim explanations and yet continue to armchair pilot by mischaracterizing the importance of the power setting, while also implying the pilots were incompetent, is shameful. But you are so incredibly ignorant, you have no idea, so you persist in being obtuse. That is how a real pilot and flight instructor will view your characterizations. And even as I try to school you, you ignore it. Your defect is significant enough I do not have the time or interest in schooling you anymore.

FAA emergency airworthyiness directive 2018-23-51 directly contradicts much of what you are saying and implying.

The PDF you cite, if authentic, and if the units are the same between LNI610 and ET302, indicate substantially more nose down trim for ET302. And that is consistent with the airspeed differences, not power.

Since you seem eager to accept more rope and hang yourself with it: ET302 05:41:15, take the airplane state at that time, and only do a power reduction, explain what you think will happen and why, in sequence with and emphasis on the force changes that happen including change in attitude.

> MCAS used trim to counter-act an overeager upwards angle during key parts of the flight like take-off.

Great post, but minor correction: At take-off flaps are set which disables MCAS.

MCAS wasn't designed because "autopilot systems cannot cope with AoA disagree."

Reread what I wrote.

MCAS wasn't enabled at all when autopilot was. Only manual flying.

Quite literally what I wrote.

The Ethiopian Airline pilots never tried to enable autopilot, and didn't "keep trying to engage autopilot" that's pure fiction.

I'm not sure what you're going on about. You can see plain as day that the captain engaged the autopilot — there are five AP warnings on the graphs in the preliminary report. You can also see the position of the AP master switch (which, yes, they engaged autopilot with the stick shaker going off). You can even read through the transcripts where they're fiddling with the autopilot throughout the whole flight.

Your own post contradicts itself on that point:

> The autopilot systems cannot cope with AoA disagree, that's why MCAS came into play.

And:

> Unfortunately for them MCAS is only needed when autopilot is disabled.

Pick one. Either MCAS "came into play" to "cope with AoA disagree" because "the autopilot systems cannot" or "MCAS is only needed when autopilot is disabled."

Pick one. Either MCAS "came into play" to "cope with AoA disagree" because "the autopilot systems cannot" or "MCAS is only needed when autopilot is disabled."

The statements aren't mutually exclusive. Not even a little bit.

Autopilot eventually disengaged because there was an AoA disagree. MCAS was active because the autopilot was disabled. That's how MCAS works. That's the only situation in which MCAS could be engaged. And that doesn't mean that MCAS was designed to mitigate the problems related to an AoA disagreement.

> And that doesn't mean that MCAS was designed to mitigate the problems related to an AoA disagreement.

You literally said it was designed to do just that:

> The autopilot systems cannot cope with AoA disagree, that's why MCAS came into play.

I still have no idea what you were trying to say there.

You literally said it was designed to do just that

No, I didn't.

> You literally said it was designed to do just that:

> > The autopilot systems cannot cope with AoA disagree, that's why MCAS came into play.

He meant "The autopilot systems cannot cope with AoA disagree, thus the autopilot automatically disengaged when AoA measurements disagreed, that's why MCAS became active.", not "The autopilot systems cannot cope with AoA disagree, that's what MCAS was designed to handle."

FWIW, I recall reading that Airbus uses three AoA sensors because autopilot becomes disabled when AoA is inoperable, and Boeing gets away with two because it does not disable autopilot on AoA disagree.