157 Dead Ethipian Airlines Crash in New Boeing 737 Max

Miscellaneous and Personal Stuff I Must Share
201

Um… Magiver, where exactly did you get this idea that they were not instantly aware of the problems with the trim? And that they were not actively and fiercely fighting to fix it within seconds? They were.

You keep repeating again and again that they were not aware of the problem with the trim, or not doing anything to fix it. That’s just false.

Can you cite any actual evidence to support what you are saying?

Have you still not even bothered to read the cockpit transcript?

Are the facts a little too inconvenient for you?

202

My extracts from the report and summary of what happened:
At 05:38:44, shortly after liftoff, the left and right recorded AOA values deviate. Left AOA decreases to 11.1° then increases to 35.7° while value of right AOA indicates 14.94°. Then the left AOA value reaches 74.5° in ¾ seconds while the right AOA reaches a maximum value of 15.3°.

The airspeed, altitude and flight director pitch bar values on the right and left side disagree.

(Stick shaker activates, warning the pilots, and remains active throughout.)

Pilots perform “IAS disagree” Emergency Checklist. This has the effect of increasing airspeed far beyond normal.

(Pitch is still up.)

“IAS disagree” Emergency Checklist doesn’t work. Captain notifies “flight control problems”, and begins process of turning back to the airport.

At 05:40:00 MCAS cuts in for 9 seconds with aircraft nose down (AND) command. Pitch trim moves from 4.60 to 2.1 units. Climb stops and aircraft descends slightly.

At 05:40:03 Ground Proximity Warning System (GPWS) “DON’T SINK” alert occurs.

While MCAS is active, pilots are fighting to re-establish positive climb, and succeed, even before the MCAS activation ends.

At 05:40:20, MCAS cuts in a second time and the stabilizer moves down and reaches 0.4 units.

Three Ground Proximity Warning System (GPWS) “DON’T SINK” alerts occur.

Captain calls on First-Officer to trim up with him. Manual electric trim in the ANU direction is recorded and the stabilizer reversed. Trim improves to 2.3 units.

First-Officer calls “stab trim cut-out” twice. Captain agrees and First-Officer switches off automatic stab trim.

(They are now following the Boeing-recommended procedure for MCAS failure after the Lion Air crash.)

MCAS activates again, but it has no effect because they have cut the stab trim.

Captain calls out three times “Pull up” and the First-Officer acknowledges.

At 05:41:46, the Captain asks the First-Officer if the trim is functional. The First-Officer replies that it is not and asks to try it manually. The Captain agrees.

At 05:41:54, the First-Officer reports that manual trim is not working.

They continue fighting with all their strength to pull up, while interacting with ATC and turning back to the airport. Aircraft is still gradually going more nose down. They do not have the height to try the ‘bunt’ maneuver.

At 05:43:04, the Captain calls to First Officer to pitch up together and says that pitch is not enough.

(As a last resort they switch on the stab trim again.)

MCAS immediately cuts in and stabilizer moves in the AND direction from 2.3 to 1.0 units in approximately 5 seconds.

Additional simultaneous aft column force is applied, but the nose down pitch continues, and reaches 40° nose down.

(Recording ends.)

203

I seem to recall you’re a pilot, but I don’t remember what you fly. Ever flown a modern jet? With auto-throttles, thrust reversers, a tiller and the rest of the bells and whistles? It’s actually a much easier workload to fly an instrument approach in that sort of aircraft than a purely visual one. This is one of the subtle truisms in aviation that’s lost on the general public.

These aircraft are designed to be flown through their automation. They are optimized to fly instrument approaches with both lateral and vertical guidance, and we are trained to do things methodically, in a predictable sequence. When I transitioned from general aviation airplanes to jets and airliners this was challenging for me to absorb. I was used to eyeballing a runway, flying it down by hand and touching down on the first brick. I was immediately told, “No, no, no - use the instrument approach even when in visual conditions, touch down at the 1000’ marker and avoid any major thrust or pitch changes.”

Therein lies the conflict at the heart of modern airmanship. It’s true, people do become too dependent on the automation. And companies will say they encourage pilots to fly by hand when appropriate to maintain proficiency. But the fact is there are fewer and fewer occasions where that’s appropriate, and pilots are further encouraged to give passengers the smoothest ride possible. That means using automation. And as the automation advances in capability, it also goes up in complexity.

I’ve seen pilots get in trouble trying to show they’ve “still got it”. I was in the jumpseat of a Boeing once and the captain decided to turn off the autopilot and auto-throttles for approach and landing. He bounced the living crap out of it - hardest landing I’ve ever seen in a jet. Was he a bad pilot? No. I have no doubt he was perfectly capable of doing that sort of landing with a few repetitions, but the odds were against him - the plane doesn’t want it, and he rarely had the opportunity to perform it.

The Asiana pilots had over 12,000 and 9,000 hours respectively. They were not inexperienced. They made mistakes. Those types of mistakes are sometimes understandable in context. Having been on the wrong side of a visual approach in a turbine aircraft myself, I get it. It doesn’t excuse it, especially when people die, but I get it.

I also get it when pilots are not able to solve a problem with malfunctioning automation, as with the Max. You and I weren’t there. You might consider viewing it from their POV with a bit more empathy.

204

My understanding is that:

  • The trim wheel spins all the time by itself, even when not on autopilot, because it automatically compensates for various parameters. Just seeing it spin by itself is not a cause for concern.

  • The way MCAS adjust trim does not look like typical “runaway trim.” It is not “spinning unrestrained.”

205

I’m not a pilot, but the author of this recent New York Times Magazine article reportedly is. Some of the points the author made that stuck with me when I read it over the weekend include the following:

He then goes through the likely sequence of what caused each of the crashes, including the fact that as velocity increased (because of the throttle settings), the aerodynamic forces got to the point that the trim could no longer be manually adjusted.

The pilots also reportedly spent much of their time trying to get the autopilot to re-engage instead of taking control and flying the plane.

The article also goes into considerable detail about the role that the increasing number of discount carriers has played, as well as pilot training programs in developing countries.

206

That New York Times Magazine article was previously cited here and I provided a partial response to it in post #180.

I’m not sure it’s particularly useful to muse on whether different pilots might have been able to save the flight – not necessarily “better” pilots, but just pilots that acted sooner or took different actions. The answer is almost certainly “yes”, but it doesn’t detract from the fact that Boeing produced an airplane with an extremely dangerous design feature and compounded the problem by failing to even tell anyone about it, let alone provide simulator training on it. The MCAS produces behavior completely different from any kind of malfunction that a pilot would normally be trained to handle, only superficially resembling runaway stabilizer trim. Whether pilot training in developing countries is not as good as in the first world is rather irrelevant as it’s beyond the bounds of all reason to attribute this accident (or the virtually identical Lion Air accident) to “pilot error”. The real cause is the defect that prompted the aviation authorities of two dozen countries to ground the 737 MAX indefinitely.

207

It doesn’t appear as if you have any background in aviation so it’s difficult to explain it to you but I’ll try.

A typical commercial flight involves a crew engaging the appropriate settings and monitoring them. So there is the automation factor discussed by Llama Llogophile. However, on climb out to say 10,000 feet above ground level They’re specifically focused on maintaining the planned flight parameters. It’s a sterile environment devoted entirely to the flight. It’s a different environment when they are at cruise at 35,000 feet.

A passenger with no flight experience would not know when a trim emergency starts. they can’t the gauges. However, if the plane started rolling unabated every passenger would understand it was not a normal aspect of the flight. they don’t need to know anything about flight to understand something is wrong. Long before they realized the flight was going south the crew would understand it because they’re monitoring the flight.

If the plane’s pitch changes unabated it’s the same scenario. When this happens in conjunction with an obnoxiously obvious trim wheel continuously spinning then there really isn’t any question what is going on. it’s not a deep mystery. Its not high tech. It’s a runaway trim. It doesn’t require digging out an emergency manual. THIS particular plane has a prominent trim system that is painfully obvious. Even if you were staring out the window wondering WTH is going on you’d hear the trim continuously moving and make the connection.

A runaway trim is solved by shutting it off. It has to be done before the horizontal tail plane reaches it’s limits or it gets much worse. As the tail plane moves away from center-line the air-stream pushes it toward the limits. That’s why it requires more force to pull it back toward center-line.

208

If you read GreenWyvern’s post showing the timeline you will note 3 things. The crew was aware that the angle of attack sensors were in disagreement, their trim switch worked and the plane’s computer was attempting to trim in the wrong direction.

The trim wheel will make minor adjustments back and forth. It does not spin “all the time” and more importantly, if it’s spinning in the opposite direction of the pitch control you’re trying to engage then it’s a runaway situation. If you have input using the electrical trim switch then retrim to a neutral setting and shut off the electric trim.

It’s about as basic as if your cruise control kept raising the speed. It wouldn’t take you long to realize this and turn it off.

209

That much seems clear. Bet they’re wishing now they’d taken the cost/time hit of notifying the changes and having the crews take the retraining.

210

If I understand it correctly, it was purely a function of maintaining the same flight characteristics as all the other 737 variants. This was a logical selling point. It might have been the main selling point.

They could have left the MCAS off the plane entirely and called it something else. It would require a separate rating/training.

211

Look, I believe in not making it personal on the boards. But I have to tell you, I don’t know what you’re saying in this paragraph. “Engaging the appropriate settings and monitoring them”? You mean, flying the plane? Even when using the automation, pilots are still flying the airplane, we’re just not doing it by hand. And “specifically focused on maintaining the planned flight parameters”? Again, this seems to describe simply flying the airplane.

“Sterile environment devoted entirely to flight” is confusing and lacks context. The term is usually “sterile cockpit”, and it just means no chatter or actions beyond what’s necessary, and is typically between ground level and 10,000’ (and / or critical phases of flight).

You seem to be using a lot of words and effort to say the Max pilots should have been able to solve the issue by treating it as runaway trim. I disagree based on everything I’ve read, and my experience in turbine aircraft. I’m open to a reasonable critique of the pilots, but the context should be remembered: The event was unexpected, happened at low level and the MCAS may very well have interfered with the pilots’ efforts. To say nothing of Boeing’s possible malfeasance.

Can I suggest we leave it there until more information comes out?

212

Strange how something so “obnoxiously obvious” wasn’t noticed by either of the pilots. :rolleyes: Or maybe you’re wrong, as just about everyone posting here including two commercial pilots have been telling you, but, as Richard Pearse correctly surmised, no amount of discussion will change your views. Nor any amount of evidence, apparently.

First of all, MCAS behavior is not “runaway trim”, which is an entirely different problem where the horizontal stabilizer fails to stop at the required set point. The MCAS merely shares some superficial characteristics with it, but it’s an active flight control system that repeatedly engages. And I will point out once again that MCAS malfunction was never part of pilot training, nor was it even documented. Boeing in fact asked the FAA to drop MCAS from its certification documents and removed all mention of it from the aircraft operations manual, where it appears only as an unexplained entry in the glossary.

Secondly, the retrospective armchair quarterbacking of what the pilots “should” have done, while it may be of some academic interest, has little relevance to the question of what the principal learnings are from these accidents, and how to best go forward. The preliminary report on Ethiopian 302 produced a summary of findings and safety recommendations; not one of them stated or even remotely suggested “pilot error” or said anything about pilot training. Your opinion notwithstanding, the official report stated that Boeing must review the flight control system responsible for the “repetitive un-commanded aircraft nose down condition”, and that aviation authorities must verify that Boeing has addressed the issue before the 737 MAX is allowed to fly again.

213

Every report I’ve read says that MCAS was originally designed for a specific set of conditions that you should encounter rarely, but that then as other handling issues were detected during testing they just kept tweaking MCAS to handle them too, expanding its scope of action. One can believe that at some point they felt they were stuck with it being too late to back down from the promise of a seamless transition but confident that they could make it good enough. And then on top of that decided to not even mention it to the customer.
(BTW though nowhere so lethally spectacular, the Airbus 321neo does seem to also have had its own set of pitch control issues detected in use with the new engines, apparently being dealt with also through software.)

214

It’s difficult in discussions like this to understand that one can have empathy and still assign blame. They are not the opposite sides of the same coin.
I do empathize with the horror of what they went through.

215

If you want to split hairs for the sake of argument knock yourself out. The MCAS system was reacting to false input and only stopped trimming when pilot electrical trim was engaged or the trim was shut off. It failed to stop at any required set point because it was fed wrong data.

216

I just want to note that the above quote was hugely screwed up, though I’m sure unintentionally. It managed not only to include stuff I’d never said, but failed to note the omissions in between the stuff I said, the stuff I never said, and the stuff I said later.

This is what I actually said:

My statements remain correct, AFAICT.

217

they did eventually treat it as a runaway trim. the 9 seconds of initial MCAS trim seems a likely flag toward that realization.

We can but it’s not tribunal in need of a good lawyer. Empathy is always available.

218

Just a note on the trim “runaway” being obvious.

As I understand it (I don’t fly the 737), the trim is working a lot during the initial climb. It does it automatically in response to speed changes even when being manually flown. The fact that the trim is running uncommanded is not necessarily a sign of anything wrong and it may not be as immediately obvious what’s going on. Particularly when you factor in that they didn’t just have the trim running, they also had the stick shaker going (which normally means you’re too slow and should lower the nose!). I wonder if the stick shaker may have made it difficult to operate the electric trim thumb switches consistently enough to counter the MCAS.

219

I’ve been following this, but not closely enough to know the answer to this basic question: at this point in the flight, is the problem that they are in a dive (previously initiated/exacerbated by MCAS, which is now switched off) so steep that the airspeed over the control surfaces is too high for them to be able to pull out? In other words, by this point they are basically doomed, largely thanks to the previous MCAS interventions?

220

yes the trim is making minor corrections all the time. But they were fighting a downward trim the whole flight. It was always a trim issue.

Allowing for the confusion of what they may have thought was the cause, the plane continued to trim until it was extremely difficult to handle. It wasn’t just a function of realizing it at the last minute. there were pilot re-trims throughout the flight. I’ll post the graphs when I find them again.