Ethiopia crash

[DellyPilot 109]

3 hours ago, KevinAu said:

The mistakes and risks of the mcas design are known and understood. Nothing really to discuss at this point in the thread as far as I can tell. 

Not sure that is true, there are plenty of aspects we still don't understand, reading the comments here https://www.satcom.guru/2019/04/stabilizer-trim-loads-and-range.html experts like Peter Lemme who has 38 years in avionics design wonders about many aspects of its design... I won't copy and paste them but its interesting.

3 hours ago, KevinAu said:

But the point I am making about going against a procedure and not verbalising it are examples of lessons we can take away in basic airmanship discipline or lack thereof that apply to any pilot in any plane in any emergency.

I am happy to agree with you that this should have been verbalised, however announcing it might not have made any difference to the outcome especially if it was the captain (more likely IMO). If it was the FO secretly doing so then yes a big error agreed.

As far as I know though we don't have a full translated CVR transcript so lots of speculation around coms. The prelim report leaves many unanswered questions, pretty clever move by Ethiopia and I agree their use of repeatedly was also disingenuous at best.

I see why you propose the FO re-engaged switches but I don't think he would have given the culture (perhaps a factor too) and also we know he initially twice announced STAB TRIM CUTOUT. 

It seems to me the captain was struggling to climb from 05:42:30 onwards applying a lot of back stick but keeping the plane barely level. By 05:43:00 the AC is descending (contrary to your claim) due to higher speed / higher altitude / remaining MCAS -ve trim they never lost.. so at 05:43:04 the captain asks the FO to pitch up together its not enough and he twice trims up.

The stab doesn't (appear) to move (it does a bit) and the AC continues to start descending even with both of them pulling back, they are now trapped.  Options:

1. reduce power and let plane pitch forward but hopefully slows enough to unload stabiliser. (not sure this would work, also Sshaker + nearby high ground)
2. let go of yoke and help FO manually trim -pitch while you struggle to wind. (not sure this would have worked either and training is 1 flies and 1 winds)

It's a bit like the scene at the end of the Italian Job. But you have to make the decision in seconds.

In hindsight in a MAX choice 1 was the only workable choice, whack the speed brakes up too..

3. Re-enable stab trim motors to help electric trim which would have been good in his previous NG plane. 

Well this was the worst option because #!£&ing MCAS in less than 10s had them in a 40degree nose forward dive experiencing -2g while both still pulling full force back on the stick. What a terrifying thought, poor buggers!

Perhaps he didn't think it would be that drastic pitch forward and they could fight MCAS with yoke trim? We will never know his thought process but its a product of both poor training and a terrible hidden design.

Again with hindsight carrying out the IAS DISAGREE Unreliable Air Speed checklist like the Lion Air crew did before or immediately after stab trim cutout might have saved them, ironically maybe the the air worthiness directive about MCAS runaway trim focused them away from 2 equally big MCAS dangers that before this crash we didn't really appreciate:

• you must return trim to near neutral before stab trim cutout
• you must reduce speed to <320kt to use manual trim 

Now the errors they made all seem so obvious. I look forward to the full report and translated voice transcripts.

And I agree training will have to look hard at the choices the crew made so we can provide a clearer checklists for a fairly straight forward failure and not rely on pilots to have pour through internet forums in not their native language to understand which parts of which checklist should be prioritised! 

I certainly can't see the MAX flying anytime soon, it needs to be fully re-certified and by the FAA this time!

 

 

[fluffyflops 1,858]

1 hour ago, DellyPilot said:

I certainly can't see the MAX flying anytime soon, it needs to be fully re-certified and by the FAA this time!

 

 

 

the commercial pressure is going to be huge for it to be brought back into service. 

airlines simply cant canx flights and ultimate pull routes for ever. the cost to wet lease is going to bankrupt some of the airlines. 

Edited April 7, 2019 by tooting

[G-RFRY 1,712]

After reading through this thread is flying safe some say the pilots are not skilled enough others the aircraft, the public may come to the conclusion is flying safe, it`s all right to quote numbers but public perception may be different, and the damage from the fallout to the industry could be costly.

[Garys 209]

Very valid question, and in recent history we have seen the Air france crash, Asiana crash, the lion Air and Ethiopian accidents and lets not forget the Air Canada taxiway incident that almost ended in the largest aviation disaster we would have ever seen. Is the training of the new batch of pilots inadequate for flying the latest generation aircraft or have the manufacturers taken the systems a step too far?  Maybe Niether but as a passenger I want to know the ones at the front are capable of bringing the ship home unless the thing is breaking into pieces or a flock of geese have taken out both engines. 

[skelsey 765]

Flying is extremely safe.

Crashing, on the other hand, is exceptionally dangerous and should be avoided at all costs ☺️

Compared to 50 years ago, there are around 3 x more passengers travelling by air and at least 3 x fewer fatal accidents.

That's a pretty impressive and enviable safety record to the point where we are now getting in to the realms of statistical anomalies.

In 2017 there were no fatal airline accidents anywhere in the world.

Imagine a year where there were no fatal road traffic accidents in your town, let alone the planet?

The causes of accidents over the years has shifted quite naturally toward the human factor simply because 50 years ago technology was a lot less reliable. Everything from engine reliability to weather forecasting has become so much more reliable over the years that pretty much all that is left is the human element.

For this reason it is obviously correct to scrutinise the actions of the crew in any accident and the truth is there is not a single crew in history who have ever handled a real non-normal absolutely perfectly according to the book. The important thing to ask however is why did the crew do what they did, instead of simply saying "well they shouldn't have done X, Y and Z... I wouldn't have done X Y and Z..."

If a crew do not perform a procedure correctly, there are numerous possible reasons.

- Were they trained correctly in the procedure? Did they have recent exposure to that procedure in a training environment?

If not, is it the fault of the crew that they failed to perform to a high level? Think about the last time you tried to do something (anything), particularly a practical task, for the first time without so much as watching a YouTube video tutorial beforehand. How did it go? How fast and accurate were you? If someone had said that your life depended on doing it exactly right, first time, within 30 seconds how confident would you be that you could manage that?

- Did the crew understand what was going on? If not, why not? What information were they being presented with and how would they have interpreted it?

There have been a number of experiments where two separate groups of people have been given the same information and asked what the best course of action would be, except 50% were told the outcome beforehand.

100% of the group that knew the eventual outcome were able to identify the correct course of action to avert it, but only about 50% of the group who had no knowledge of the outcome were able to do the same. Hindsight is indeed 20/20.

- What other factors could havehave reduced the crew's performance? Distractions? Impairment? Other failures? Prior training?

For me in these particular incidents the stick shaker is a BIG deal.

From virtually day one of pilot training the instinctive, automatic response drilled in to pilots is that if the stall warning goes off you must reduce the AoA. How many here would immediately assume instead that the stall warner was faulty? What happens when we have a 737 Max crash because the first thing the crew did in response to a stall warning was shut off the trim instead of recovering in the assumption that the stall warner must be false?

Add to that the fact that the noise of the stick shaker is a tremendous distraction. Someone posted a recording earlier on this thread - I suggest that you try putting some headphones on, turning the lights out and playing that sound on loop at high volume whilst trying to do a mental maths quiz and, say, hand-flying your favourite flight sim aircraft and trying to maintain a steady course with the trim set full nose down. See how many questions you get right (and that doesn't even include the additional distractions of ATC and another crew member talking to you).

Simply dismissing the crew as incompetent is convenient and makes us all feel better (because we can all say "well I/the pilots in <insert home country/favourite airline here> wouldn't do that") but it doesn't solve the problem (and is probably not true). It is far more likely that there are systemic factors at play which have resulted in the crew performing in the manner in which they did and understanding the those factors is what will help make aviation safer everywhere.

[KevinAu 441]

14 hours ago, DellyPilot said:

Not sure that is true, there are plenty of aspects we still don't understand, reading the comments here https://www.satcom.guru/2019/04/stabilizer-trim-loads-and-range.html experts like Peter Lemme who has 38 years in avionics design wonders about many aspects of its design... I won't copy and paste them but its interesting.

I am happy to agree with you that this should have been verbalised, however announcing it might not have made any difference to the outcome especially if it was the captain (more likely IMO). If it was the FO secretly doing so then yes a big error agreed.

As far as I know though we don't have a full translated CVR transcript so lots of speculation around coms. The prelim report leaves many unanswered questions, pretty clever move by Ethiopia and I agree their use of repeatedly was also disingenuous at best.

I see why you propose the FO re-engaged switches but I don't think he would have given the culture (perhaps a factor too) and also we know he initially twice announced STAB TRIM CUTOUT. 

It seems to me the captain was struggling to climb from 05:42:30 onwards applying a lot of back stick but keeping the plane barely level. By 05:43:00 the AC is descending (contrary to your claim) due to higher speed / higher altitude / remaining MCAS -ve trim they never lost.. so at 05:43:04 the captain asks the FO to pitch up together its not enough and he twice trims up.

 

 

I know we are both speculating upon speculation, and your opinion that it was the captain is as good as my guess that it was the fo, but if it was the captain who took one arm off that really heavy yoke and flipped the switch to get his electric trim back, then why did he not hold the trim down afterwards? That is as puzzling as the fo flipping the switch on and not calling out as he did earlier.

Look at the fdr readout you posted, the plane had been climbing even after the first two mcas inputs. They reached an altitude of 13,400 by time 5:43:11 when the trim was turned back on. They had steadily achieved almost 10,000 ra above by the time of the trim was turned back on. If they had stuck with the procedure, they would have struggled but they wouldn’t have gone into the nose dive.

I have no expectation of how long it should have taken them to cut out the trim. All I was saying with points 3 & 4 is that it would have been better if they had gotten to it earlier, before the second running of the trim or had pulled the power back. That’s why I separated it from the first 2 points and put the word ‘understandable’ there, to try to communicate to you that they may be errors in the final report, but nothing that should have been held against them. Just that it would have gone better if those two items happened faster or got done.

Our crm classes have steadily drifted away from pilot behavior to analyzing slices of swiss cheese and accident discussions over the last few years. I have a feeling that as this trend of undisciplined pilot actions during emergencies becomes more apparent with the colgan, ethiopean and atlas crashes, all of which had crews doing something surprising to each other, we are going to have to circle back to a more human behavior centric crm class.

Edited April 8, 2019 by KevinAu

[Bobsk8 12,728]

Bank America downgraded Boeing Stock this morning and it looks like 6-9 months before Max will be back in the air again. https://www.cnbc.com/2019/04/08/boeing-shares-fall-4percent-after-bank-of-america-downgrades-stock-on-737-max-production-delays.html

[DrumsArt 270]

I would be curious to know how many crew have been confronted with this problem and had to deal with the procedure with the Max? Is there any way to find out?

Edited April 8, 2019 by DrumsArt

[Wink207 78]

7 minutes ago, DrumsArt said:

I would be curious to know how many crew have been confronted with this problem and had to deal with the procedure with the Max? Is there any way to find out?

Me too. If you find out, please post.

Also I wonder how many 737 pilots have ever heard the stick shaker in normal ops outside the sim? It's like the question of how many transport pilots have ever seen the low fuel warning outside the sim? How many pilots have ever seen "IAS Disagree" due to a failed AoA sensor? I'll bet someone knows these answers, or their is a database somewhere.

[Bob Scott 7,721]

4 hours ago, Bobsk8 said:

 it looks like 6-9 months before Max will be back in the air again. https://www.cnbc.com/2019/04/08/boeing-shares-fall-4percent-after-bank-of-america-downgrades-stock-on-737-max-production-delays.html

No, that's not what the article says.  The Boeing production line is seeing delivery delays increasing by ~3 months...3-6 month delays are extending out to 6-9 months.

And a downgrade by BoA from "Buy" to "Neutral" is hardly catastrophic news...

 

[threexgreen 3,139]

4 hours ago, DrumsArt said:

I would be curious to know how many crew have been confronted with this problem and had to deal with the procedure with the Max? Is there any way to find out?

 

4 hours ago, Wink207 said:

Me too. If you find out, please post.

AFAIK, there were no further incidents related to what happened in these two crashes.

The incidents reported by the two pilots of the same flight on the ASRS and subsequently by media were not MCAS related since the nose down motion began seconds after autopilot engagement where MCAS doesn't respond.

Remember that the MCAS itself never malfunctioned but worked exactly as it was supposed to. Where the trouble began is the fault of the left AoA vane where MCAS in its current iteration receives its data from. So for an event like we have seen here it will need to be that particular AoA vane malfunctioning in order for MCAS to kick in at all. It would also need to malfunction in such a way that the false data sent actually suggests the aircraft is in a situation close to stall. I don't have any numbers whatsoever as to the frequency that AoA vane failures occur, but I would think it doesn't happen too often and if it does, as I said above, it still takes a more intricate chain of events for MCAS to kick in so most AoA vane failures will result in mere incidents with pilots landing safely and reporting to maintenance for a fix.

Two sources for aviation incidents I use myself out of interest are the ASRS and the Aviation Herald. I did a search on both but didn't find anything.

Edited April 8, 2019 by threegreen

[HighBypass 2,339]

11 hours ago, Garys said:

...or a flock of geese have taken out both engines. 

That's why the pilots should have some glider experience too!  

[DellyPilot 109]

3 hours ago, threegreen said:

Remember that the MCAS itself never malfunctioned but worked exactly as it was supposed to.

Very true, it annoys me when the media say MCAS malfunctioned. Nope did exactly the ridiculous things it was supposed to.

3 hours ago, threegreen said:

I would think it doesn't happen too often and if it does, as I said above, it still takes a more intricate chain of events for MCAS to kick in so.

It has happened 3 times (in 2 planes) in only a few months which translates to pretty frequently for such a failure and it is not a complex chain of events that leads to things going wrong because on all occasions the plane has nearly crashed.

As there was no AOA DISAGREE warning light on the first LionAir flight (which landed safely after nearly crashing) the crew never realised what was happening throughout the entire flight even after the 3rd pilot in the jump sweat suggested TRIM RUNAWAY and flicked STAB TRIM CUTOUT.

Their report makes no mention of AoA inputs only IAS DISAGREE / ALT DISAGREE / DIFF FEEL PRESSURE. They tried turning back on the STAB TRIM but the problem started again so they disabled it and flew on Manual trim for the rest of the flight.

Sadly (and inexplicably) they never mentioned either the STAB TRIM CUTOUT or the left SShaker going off all flight and it was this terrible omission by the crew that meant maintenance also never realised the AoA vane was bust and merely sent the plane back out of after a quick clean of the pitot tubes. There is no ground test to prove the AoA vane is buggered. Disastrous design flaw.

As you say the AoA vane could have been broken by a birdstrike or maybe just a ground impact from a careless service truck.

So there have been 2 events (1 over 2 flights) and in both cases the plane crashed (eventually). 

I believe the speculation that if you whack the thing it breaks in such a way that once and only once the airspeed picks up it points directly up.

Clearly this design means any damage can potentially provide the worst possible information but crucially only once you take off! Doh.

This is yet another flaw and something that should have been thought of surely. Why not ensure that it always returns to 0 degrees when busted? Better still ensure the FCC disregards the AoA inputs that do not match other validated inputs like pitot data (altitude/speed or even GPS data). 

[DellyPilot 109]

8 hours ago, Wink207 said:

How many pilots have ever seen "IAS Disagree" due to a failed AoA sensor? 

7 men and only 3 are still alive, first LionAir crew that landed.

In the NG IAS Disagree was not triggered by AoA input disagree, this is a new feature in the MAX because AoA data is used to adjust various things such as speed and altitude (no idea how or why just what I read).

Probably not something they picked up in their 1hr training course.

Edited April 8, 2019 by DellyPilot

[Wink207 78]

8 minutes ago, DellyPilot said:

7 men and only 3 are still alive, first LionAir crew that landed.

In the NG IAS Disagree was not triggered by AoA input disagree, this is a new feature in the MAX because AoA data is used to adjust various things such as speed and altitude (no idea how or why just what I read).

Probably not something they picked up in their 1hr training course.

Thanks for the info. I'm finding 2 thoughts very hard to wrap my head around, as they say:

1.Prior to the grounding there were 350 some odd Max aircraft in service having safely completed thousands of flights with no apparent AoA related incidents. This is a statistically significant number. Also, there are more thousands of transport aircraft flying around with the same or similar AoA sensors. Yet I have failed in efforts to find any data or reports on failed AoA sensors. There must be some, but it is just strange if indeed Lion Air had a failed AoA sensor, and then Indonesia has a failed AoA sensor, and they both occur on or before the TO roll. It is just too coincidental.

2. I find it hard to believe that the Boeing systems engineers did not complete a FMECA analysis during the MCAS development that included failed AoA on TO as a critical failure mode, with drastic effects. It just seems so obviously a critical case for MCAS.

There is a tremendous amount of "smoke" in the air right now and I'm not confident that we, the informed traveling public, will know the truth of all this until years from now.

Rick