Ethiopian airliner down in Africa

[737 Driver]

Originally Posted by Water pilot

This statement is exactly why I feel that we do not have the complete picture yet. When the trace shows that at least two different pilots did something completely counterintuitive (a few short bursts on the trim rather than holding it down) when they really, really wanted to make the plane go UP instead of DOWN, I think that there is a factor that we do not know about yet.

You can find a fairly comprehensive review at the following link which includes some reasonable speculation as to why the pilots did not trim sufficiently during the final moments:

https://www.satcom.guru/2019/04/what...-on-et302.html​​​​​​

[Water pilot]

Originally Posted by 737 Driver

You can find a fairly comprehensive review at the following link which includes some reasonable speculation as to why the pilots did not trim sufficiently during the final moments:

https://www.satcom.guru/2019/04/what...-on-et302.html​​​​​​

Thanks for the link, but note that they shrug their shoulders as well:

In any case, it is not clear what prevented the crew from continuing to trim airplane nose up to reduce column forces.

This is something that needs to be figured out, and I hope that Boeing's internal practice is different than their public stance that three independent aircrews (the first crew did not respond properly, either) were not skilled enough to fly their plane. They need to test this scenario to failure in the sim, find the worst line pilots imaginable, the ones quite close to being terminated but still qualified, and put them through this scenario and see what they do. If they can't replicate the scenario, something else is wrong.

Boeing's conclusion that pilots are unqualified to fly the MAX -- and that is their conclusion, and the conclusion of many here -- is at odds with their stance that a 15 minute IPAD session will fix everything. Boeing markets the plane as being flyable by pilots that meet industry standards and either the plane is not flyable by pilots who meet those standards (you can't argue with two crashes in six months) or there is something wrong with the plane. The latter is far easier to fix.

[MurphyWasRight]

Originally Posted by 737mgm

Stick shaker only on one side, several caution lights going on, differing airspeed indications create a very confusing situation. For exactly that reason there are memory items in order to deal with that situation and they must be carried out.
...
There was not one procedure done wrong but several done wrong. Each one of these procedures would have saved the aircraft. Carrying out the unreliable airspeed memory items is one procedure
.

Things are never as black and white as one might wish. The ET pilots were not at all perfect but they were also misserved by the system.

This is text from a Boeing UAS flow chart someone posted a while back, not I do not have access to current procedures,although what probably matters as much is what was in effect when the Pilot was trained.
Seems totally reasonable to decide it is false AoA related warning.

AoA sensor failure on Takeoff
If AoA sensor is failed high,stick shaker on failed side will activate on rotation accompanied by IAS/ALT disagree warning flags.

If the pitch power and config are consistent with takeoff and the good side ASI agrees with the Standby ASI,then it is a false warning
There is a side arrow to the side that states "If in any doubt execute the UAS NNC"

The pilot with good side data becomes PF

Land immediately

With a 360 hour co-pilot there may be a reason the Captain retained control, especially with relatively small ASI difference.

Originally Posted by 737mgm

And the Ethiopian Captain did trim out the MCAS induced movement completely the second time it activated. So we could already talk a lot about why they didn't do it the whole time, then lower the flaps for landing (MCAS deactivates) and land.
...
...
even in that case, carrying out the stab trim runaway memory items correctly once MCAS engaged (please stop these ridiculous comments that the MCAS activation cannot be recognized as a stab trim runaway) would also have saved the plane.

Note, the times below are from the FDR chart which is hard to read better than a second or so due to ambiguous sloping edges on binary traces.

The first MCAS trim started at 05:40:00 ending about 9 seconds later, this was as the autopilot dropped out after flap retraction.

The pilot only partially unwound the trim at 05:40:15, 5 seconds later the second MCAS input started but was cancelledby sustained pilot trim starting at 05:40:29.
This second trim attempt lasted until the stab trim was cutout.
From the prelim report:

At 05:40:35, the First-Officer called out “stab trim cut-out” two times. Captain agreed and First-Officer confirmed stab trim cut-out.

Given that Boeing changed the cutout switch functions on the MAX so that either one disabled all electric trim whereas on NG the right switch disabled automatic trim only it is possible/likely that the pilot was still trying to trim.
Much later the (very sparse) partial transcript has this:

At 05:41:46, the Captain asked the First-Officer if the trim is functional. The First-Officer has replied that the trim was not working and asked if he could try it manually. The Captain told him to try. At 05:41:54, the First-Officer replied that it is not working.

This supports the idea that they believed they had only disabled automatic trim and manual electric trim should still function.
This difference is not highlighted, or mentioned at all depending on which MAX type conversion powerpoint the crew used.

Bottom line is that runaway trim nnc was executed but they were (apparently) left with an inability to crank the trim wheel.
"Unloading" maneuvers used to be part of runaway stab training but apparently were dropped some time ago.

To anyone contemplating commenting on the emergency AD stating to trim first if MCAS runway:
A: That was buried as a note -after- action items.
B: ET pilots company may have failed to update manuals.

Of course it is tragic that they did not input sustained trim after an apparent last ditch effort where they re-enabled electric trim. I still wonder if some other factor is at work here since almost exactly the same thing is seen at end of Lion Air traces.

There was less than 45 seconds from start of first MCAS trim to loosing all trim capability.

They might have done better had they not followed the runaway trim procedure.

I do totally agree on your points on the state of training and company policy that result in poor confidence in manual skills, this undoubtedly also played a significant role in this accident.

Note I see several other responses on trim failure while I wrote this, the above is one (of several) possible scenarios.

[Loose rivets]

Water pilot:

This statement is exactly why I feel that we do not have the complete picture yet. When the trace shows that at least two different pilots did something completely counterintuitive (a few short bursts on the trim rather than holding it down) when they really, really wanted to make the plane go UP instead of DOWN, I think that there is a factor that we do not know about yet. I can't imagine anybody not continuously holding the "UP" switch when they see ground in the windshield, it is just not the way humans operate.

I still don't understand that 'noise' where there should be steady nose-up trim. Shaking stick suggested. Not convinced.

[RatherBeFlying]

MCAS Working as (poorly) Designed - AoA fails occasionally

I am wondering how much negative g was produced in the cockpit with the final two MCAS excursions past Vmo. I remember being a front seat passenger when our car rear-ended another car. My legs flew up even though I was strapped in and braced. The driver wasn't and fractured a foot. The accident crews were not military pilots experienced with serious g.

Should we not have a careful look at the stick shaker NNC, given that a failed AoA is setting up the crew for the MCAS knock out combo punch?

I like that at least one pilot suggests pulling the stick shaker breaker once a safe flight path is established.

If failed AoA on one side, is there a way to switch SMYDs - and then perhaps see if the other AoA is correct (sure would like to see three)?

And how about an NNC item to keep flaps down unless switched to a good AOA - that or cut out the automatic trim first?

I am well experienced with non air carrier aviation organisations that sometimes set up pilots to kill themselves (all with the best intentions) and have witnessed a fatality, injury, hull losses, serious incidents and narrow escapes.

A certain paranoia is beneficial, including towards your own attitude.

​​​​In the kingdom of SOPs and multiple checklists, we and management (operator and manufacturer) tend to believe the cotton wool it envelopes us in will protect against all ills we have foreseen.

Works pretty good until something unforseen pops up when the crew better have 737driver's mantra in their back pocket - unless you're in a glider.

[formulaben]

Originally Posted by Water pilot

This statement is exactly why I feel that we do not have the complete picture yet. When the trace shows that at least two different pilots did something completely counterintuitive (a few short bursts on the trim rather than holding it down) when they really, really wanted to make the plane go UP instead of DOWN, I think that there is a factor that we do not know about yet. I can't imagine anybody not continuously holding the "UP" switch when they see ground in the windshield, it is just not the way humans operate. You can see Formula 1 drivers adjusting the steering wheel while their car goes airborne, people try to turn on the lights when the power is out, the helmsman will pull on the throttles as the ship with dead engines inexorably heads to crush the wharf, etc. It defies reason that none of the four pilots held the button down to the last.

I don't know what it could be (can you break the switches by pressing too hard on them?)

Originally Posted by RatherBeFlying

I am wondering how much negative g was produced in the cockpit with the final two MCAS excursions past Vmo.

I could be wrong but I thought it was clear that it was a full negative 2 Gs, partially due to the crew losing control of the yokes with the last nose down input. Additionally, to Water pilot's point I think that due to the fact that the aircraft was well past Vmo it would appear that the short bursts of nose up trim resulted in very high momentary positive G forces. I firmly believe the crew stopped trimming nose up due to the momentarily high positive Gs on the aircraft...and 5 seconds later the unthinkable happened. Remember, the power is still set at TO, and is accelerating all this time, especially when pointing downward. A high speed nose-down trim input at Vmo+ can't be pretty...

[Cows getting bigger]

Trim Runaways

It would appear Boeing's stance is that the MCAS events are basically Trim Runaways, something that is trained regularly. So, there's a very simple question here: how many 737 trim runaways have been successfully handled over the decades? If the answer is 'lots' then there's strength to their argument. Conversely, if there have actually been very few trim runaways, two catastrophic, one may wish to question whether the event is routinely manageable despite what happens in the sims.

[Fortissimo]

Interesting and very carefully chosen words in Boeing's statement https://boeing.mediaroom.com/news-re...ts?item=130426, issued in response to media reports yesterday.

Boeing included the disagree alert as a standard feature on the MAX [...] Boeing did not intentionally or otherwise deactivate the disagree alert on its MAX airplanes. The disagree alert was intended to be a standard, stand-alone feature on MAX airplanes. However, the disagree alert was not operable on all airplanes because the feature was not activated as intended.

It goes on to state that if you did not opt for the AOA display, your disagree alert was "not operable". However, there will be an optional service bulletin to tell you how to make it operable on aircraft already delivered. So that's OK then.

[Less Hair]

If MCAS is both "not needed" and in the future will be de-activated after only one run why not get rid of the whole system and train the pilots how to deal with raw nose up slow flight handling instead?

[Cows getting bigger]

Fortissimo, that is a mind-warping read. It reminds me of a scene from Yes Minister!

[groundbum]

Originally Posted by Cows getting bigger

Fortissimo, that is a mind-warping read. It reminds me of a scene from Yes Minister!

Boeing know full well the future of the company is on the line here, and one mis-spoken phrase could be the nail in the coffin. This is a can of worms with more to come, that's why B is scared.

[skwdenyer]

Originally Posted by Fortissimo

Interesting and very carefully chosen words in Boeing's statement...



It goes on to state that if you did not opt for the AOA display, your disagree alert was "not operable". However, there will be an optional service bulletin to tell you how to make it operable on aircraft already delivered. So that's OK then.

How does that square with the previously-stated position that the AOA disagree was an extra cost option?

[Bend alot]

Originally Posted by 737mgm

Almost 5000 posts later it is clear it doesn't matter what professional 737 pilots will say about this issue. You guys have made up your mind and no facts will get in the way of that opinion. Nonetheless I will give it another try even though what I am about to write has probably already been mentioned several times:

The MCAS system obviously should have been designed better but the press makes it seem as if Boeing knew the system created great risks but they went ahead with it anyway in order to make more profit. That thought in itself is so absurd that arguing with anyone who believes it, is basically pointless. If Boeing had thought these accidents were likely to occur they would have designed the system differently.

What does this MCAS monster as you call it actually do? It lowers the nose of the aircraft by changing the position of the horizontal stabilizer. It doesn't make the wings fall off, it doesn't set the plane on fire... it only moves the horizontal stabilizer. There is a switch on the pilots control column which moves the horizontal stabilizer and it overrides MCAS each time. I know you guys do not want to face it, but it really is that simple. Now if someone is reading this, thinking if it was that simple why didn't the pilots just do it? First of all they did. The Lion Air Captain did it at least twenty times. And the Ethiopian Captain did trim out the MCAS induced movement completely the second time it activated. So we could already talk a lot about why they didn't do it the whole time, then lower the flaps for landing (MCAS deactivates) and land. Aside from the fact that they did use the trim switch a little bit at least. please have a look at Asiana flight 214, Emirates 521, or Turkish 1951. These are only a few examples of flights were pilots failed to do the most basic thing: flying the aircraft. There were other causes as well and in the case of Turkish and Asiana the flights were not stabilized at 1000 Feet which means they should have carried out a go around preventing both accidents even before the mistake of not flying the aircraft (speed control) was made.


What these accidents show you is that pilots will crash aircraft even though it would have been very easy not to. In the case of Asiana and Emirates there were not even any system failures. Why these things happen is a completely different issue and the answers are to be found in the human factors involved when people fly aircraft. The Fact is, planes crash because pilots do not carry out the basic steps involved in flying an aircraft.

Now one important point I would like to make to all the posters with their Skygod and hindsight comments: First of all, it isn't about blaming the pilots. It is about stating facts. If I or others state that the crew did not carry out the unreliable airspeed memory items when in fact Boeing requires a crew to so in the situation they were in, then that is stating a fact and has nothing to do with hindsight or blame. Stick shaker only on one side, several caution lights going on, differing airspeed indications create a very confusing situation. For exactly that reason there are memory items in order to deal with that situation and they must be carried out. This procedure is a life saver because in that situation it is nearly impossible to figure out which system is malfunctioning and what is actually going on. For that reason: AP off, AT off, FD both off, flaps extended 10° pitch 80% N1. This will set the airplane on a safe flight path and you will continue to climb away from the ground. Now you have time to figure out the problem. This is what we are trained for. This is our job. Carrying out these procedures in that situation has nothing to do with being a skygod and pointing out that the crew did not do it doesn't make us armchair pilots with hindsight. If there is an engine fire on take off I carry out my memory items, if there is a rapid depressurization I carry out my memory items. Again this is what we are trained for.

If the crew had done this on any of these flights we wouldn't even be talking about MCAS. I am not blaming the crew when I say they did no do this. The crew is always just a product of the training department and the general procedures of the airline they fly for. For example, in the case of the Emirates crash, the crew wanted to fly a go around. They only pushed the take off/go around button but did not actually move the thrust lever forward in order to increase thrust on the engines. These pilots were not idiots. I can completely understand how this could happen. Normally as long as the automatic thrust control is engaged, pushing the TOGA button will increase thrust. However, shortly after touch down the automatic thrust control is automatically disengaged. Emirates is an airline with procedures that will not allow pilots to fly manually with the flight directors off for example. I can understand how a crew that is trained to rely on automation will not think to actually move the thrust levers and will not check if the thrust is increasing, despite procedures to do so. Basic skills are lost if one does not use them regularly.

The 737 MAX is grounded and will continue to be grounded for a long time because people think the airplane is unsafe. Furthermore the press reported the Ethiopian pilots carried out all the required procedures and yet they crashed. Why are people like me writing on this forum? It simply isn't true that the 737 MAX is inherently unsafe even with the "old" MCAS and it isn't true that the crew carried out the required procedures. That is not my opinion. That is a fact (again why they did not do it is another issue). If the established procedures had been carried out the planes would not have crashed. There was not one procedure done wrong but several done wrong. Each one of these procedures would have saved the aircraft. Carrying out the unreliable airspeed memory items is one procedure. If the pilots do not do this they are already entering an area where they are not flying the airplane according to procedure anymore but even in that case, carrying out the stab trim runaway memory items correctly once MCAS engaged (please stop these ridiculous comments that the MCAS activation cannot be recognized as a stab trim runaway) would also have saved the plane. If the pilots decide for whatever reason to not even do that, they still have the option of just trimming out the control forces each time MCAS activates (basic flying skills). Again when I am stating what they should have done according to procedure I am not blaming the pilots but only stating the facts of what Boeing says pilots must do in these situations and what we are trained for in our manuals and in the simulator. These recommended procedures would have saved the plane and please realize that there was a flight that had exactly the same problems and they did land safely (and they did not carry out unreliable airspeed mem items and also did not accurately carry out the stab trim mem items and still they saved the plane proving there are several options and it is not a react quickly or die situation).

Final comment: In my opinion crews that are not trained well and are lacking basic skills of flying the aircraft due to company policies are a far greater threat to aviation safety than any system design. I believe we still have a long way to go until we will design systems that never fail (if ever). So it will continue to be up to the pilots to save the day when systems malfunction. There needs to be a focus on enabling pilots to do so a lot more than focusing on how MCAS can be improved.

After what 5000 posts - can you answer why :- At 05:42:54 - Both pilots call out "left alpha vane"

This is 4 minutes and 12 seconds after WOW (take off) and they did not have the AoA disagree option I believe - pretty strange for a reasonable time Captain and low total time cadet to have a "snap" on exact words at that stage of flight.

[Loose rivets]

'Damaged wiring' ( quote from somewhere) and a CB popping out?

If the CB was labelled 'Left Alpha Vane', that would clinch it, I'd guess.

[Escape Velocity]

Originally Posted by Bend alot

After what 5000 posts - can you answer why :- At 05:42:54 - Both pilots call out "left alpha vane"

This is 4 minutes and 12 seconds after WOW (take off) and they did not have the AoA disagree option I believe - pretty strange for a reasonable time Captain and low total time cadet to have a "snap" on exact words at that stage of flight.

The Alpha Vane warning light is part of the anti-ice system. It illuminates if the AOA heat malfunctions. It has no bearing on an actual AOA signal failure, only a failure of the vane heating system.

[meleagertoo]

Why don't they just admit it and get on with fixing their problems?!

That is a very good question indeed.

With respect, it's not. Its a very naiive question.

First of all they very clearly are 'getting on with fixing their problems'. What on earth would make anyone dream otherewise?

But more to the point, everything Boeing says and everything they do is governed by lawyers. Every word they utter in public is scrutinised by lawyers before it is spoken or printed. Every action or modification or design approval is swarmend over by lawyers because of the insane levels of liability involved.

A moment's thought would tell you that Boeing cannot admit fault even if it wanted to, because that would render them exposed to unlimited liability. They know they are in for reparations of epic proportions but they must try to limit them out of simple fiancial survival.

In the past Boeing have, I'm told, not changed ambiguous or unclear wording in manuals because doing so would be a legal admission of fault in the original wording, and render them liable post-correction had anyone ever been damaged by the original! That is the sort of legal idiocy Boeing are up against, and I daresay they'd love to be more open - as any decent person would - were they able to. The point is they simply cannot be as open as some would wish, and though their reticence may not be entirely down to this aspect I have no doubt that a great part of it is which leads to statements that appear mealy-mouthed. You have the shysters - and a grossly overdeveloped liability-obsessed legal system - to thank for that.

[meleagertoo]

Originally Posted by Escape Velocity

The Alpha Vane warning light is part of the anti-ice system. It illuminates if the AOA heat malfunctions. It has no bearing on an actual AOA signal failure, only a failure of the vane heating system.

This sort of thing could be significant. If the Alpha vane had been knocked off in a birdstrike I wonder how long it takes to trigger an overheat if there is no vane left to dissipate it? You can bet someone has found out, and if that figure is close to 4min 12 sec it's an indication that might point to the alpha vane physical loss theorey.

[PEI_3721]

Why don't they just admit it and get on with fixing their problems?!

Why don't they just admit it and get on with fixing their problems.
Perhaps Boeing does not know which particular aspect fix. The proposals so far limit the effects of an errant AoA value on trim (MCAS issue), but without specifically identifying the source of error: - ‘well whatever it is, the output is limited’.
But the other systems which use AoA appear not to be protected.
How can you prove (certification) that a system is now reliable if the orrigionating source of the problem is unknow, unreplicatable.
Boeing 737 Max Software Fixes Due to Lion Air Crash Delayed

[MurphyWasRight]

Originally Posted by meleagertoo

This sort of thing could be significant. If the Alpha vane had been knocked off in a birdstrike I wonder how long it takes to trigger an overheat if there is no vane left to dissipate it? You can bet someone has found out, and if that figure is close to 4min 12 sec it's an indication that might point to the alpha vane physical loss theorey.

This was discussed at length somewhere around post 4000 (3000?), a lot of details can be lost in a massive thread, a brief recollection:
The DFR shows initial AoA fault occurs at 5:38:45 followed shortly after by "Primary AOA Heat L" going to off.
Quotes are from the prelim report:

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.

The heater is in the vane so when the vane departs the circuit would open, this would cause the above.

At 05:42:51, the First-Officer mentioned Master Caution Anti-Ice. The Master Caution is recorded on DFDR.
At 05:42:54, both pilots called out “left alpha vane”.

I don't know the system but assume there was a display of Anti ice caution cause.
This was the second master caution, I will leave it to those with more knowledge whether this was a simple repeat of the first (unacknowledged?) or was re-triggered by other events. Totally reasonable for the crew to not prioritize an anti ice caution at takeoff given the conditions, this would only matter at high altitudes.

The other evidence that strongly supports vane departure theory is the last part of the DFR plot where the left AoA sensor can be seen closely following G forces, caused by the internal (vane) counterweight acting as a pendulum.

[737 Driver]

Originally Posted by meleagertoo

With respect, it's not. Its a very naiive question.

First of all they very clearly are 'getting on with fixing their problems'. What on earth would make anyone dream otherewise?

Okay, perhaps my comment was a bit too subtle, because it really wasn't directed at Boeing.

Previously, I listed five broad causal links involving different groups:

1. Boeing
2. FAA and other certificate authorities
3. Maintenance/logistical chain (both the airline and possible third-party repair facility involved with Lion Air faulty AOA vane)
4. Airlines
5. Flight crew members

Let me restate my question differently, perhaps a bit more diplomatically.

Why don't each of these groups simply admit they had a role in these accidents and focus on addressing those issues?

I think that we all know that for legal reasons, anyone in the first four groups is going to be very reticent to say anything publicly that will increase their liability exposure. And to be fair, each of these groups is very likely working in the background to correct their lapses. One concern, of course, is that some of these players may not go far enough because they are concerned that comprehensive remedial action will be tantamount to an admission of guilt (i.e. the initial Boeing response to the Lion Air crash).

What about that fifth group? I'm not talking about the crew members who perished. I'm talking about the rest of the professional pilot corps who has just been handed a great, big warning sign that things may not be well in our collective house. I don't think anyone is going to try to sue anyone here, so we really don't have the same excuse as the others.

Yes, it get the the natural human tendency to defend one of our own. We really don't like contemplating that one of our brethren may have had anything to do with the deaths of hundreds of people. "There but for the grace of God," and so forth. All perfectly understandable sentiments, and all significant obstacles to making any changes to the way we do business.

You can't fix a problem until you recognize a problem exists. I humbly suggest that we collectively recognize the crew competency issues within our own ranks and devote our brain cells to addressing that problem rather than lobbing largely ineffectual grenades over the fence.