Boeing admits flaw in 737 Max flight simulator

[Kal Niranjan]

Please use the sharing tools found via the share button at the top or side of articles. Copying articles to share with others is a breach ofFT.com (https://www.ft.com/)T&Cs (https://help.ft.com/help/legal-privacy/terms-conditions/)and Copyright Policy (https://help.ft.com/help/legal-privacy/copyright/copyright-policy/). Email [email protected] buy additional rights. Subscribers may share up to 10 or 20 articles per month using the gift article service. More information can be found here (https://www.ft.com/tour).
https://www.ft.com/content/494354da-799e-11e9-81d2-f785092ab560

Boeing has been forced to correct a flaw in the software of flight training simulators that are meant to reproduce the flying conditions of the 737 Max aircraft involved in two deadly crashes in the past six months. The disclosure of a problem with the simulator software is a further blow to the credibility of the Boeing brand, which has been seriously damaged in recent months by the two crashes, in which 346 people died. Subsequent disclosures of serious design flaws both in the anti-stall system of the Max, the manoeuvring characteristics augmentation system (MCAS), as well as errors involving other safety systems have further undermined the reputation of the world’s largest commercial aircraft maker. Boeing revealed at the weekend that the software used on the Max training simulator was unable to reproduce some flight conditions, including the conditions which led to the crash of Ethiopian Airlines flight ET302 on March 10. The preliminary crash report from the Ethiopian authorities revealed that the pilots of that flight were flying at high speed and were unable to overcome the power of the MCAS system as it drove the plane’s nose down, by doing what Boeing had told them to do in such circumstances: use a manual wheel in the cockpit to bring the nose back up.

Read further below:

https://www.ft.com/content/494354da-799e-11e9-81d2-f785092ab560

[Dee Vee]

same thing was reported in the Seattle Times a couple of days ago

https://www.seattletimes.com/business/boeing-discovers-flaw-in-sought-after-737-max-simulator-the-same-kind-that-ethiopian-airlines-had/

Boeing recently discovered that the simulators could not accurately replicate the difficult conditions created by a malfunctioning anti-stall system, which played a role in both disasters. The simulators did not reflect the immense force that it would take for pilots to regain control of the aircraft once the system activated on a plane traveling at a high speed.

On Ethiopian Airlines Flight 302, the pilots tried to follow Boeing’s instructions and manually adjust or trim the horizontal tail — called the stabilizer — using a large mechanical wheel beside the pilot seat. However the heavy forces on the tail made it impossible to move the wheel. It’s this condition the flight simulators failed to reproduce.

In a statement, an Federal Aviation Administration (FAA) spokesman said the safety agency “is aware that Boeing Co. is working with the manufacturers of Boeing 737 MAX flight simulators to adjust the fidelity of the stabilizer trim wheel.”

The FAA statement added that it “will review any proposed adjustments as part of its ongoing oversight of the company’s efforts to address safety concerns.”

[ironbutt57]

the media will NEVER be corrected on "anti stall" will they? good grief

[Gove N.T.]

the media will NEVER be corrected on "anti stall" will they? good grief
Please remember that the media are describing what the system is designed to correct in certain circumstances in simple language for the poor, ill-educated masses, including some politicians it seems, who don’t have the benefits of your inside knowledge.

[A and C]

If the MCAS system failures resulted from an unknown and unintended flaw in the software it is hardly surprising that the simulators had not been programmed with this unknown and unintended flaw.

[Bend alot]

If the MCAS system failures resulted from an unknown and unintended flaw in the software it is hardly surprising that the simulators had not been programmed with this unknown and unintended flaw.



I think a whistle blower has stated the single AoA reference was a known flaw - but using both sensors, would then lead to "extra training" not a iPad conversion training platform.

If that turns out to be fact it is an intended flaw in the software. MAX simulators are low in numbers for a reason and that reason is there is nothing new here. Certainly not 3 things that play with flight controls.

It should be checked as to the other new items, if they also are accurate on the MAX simulators.

Given the events in Russia recently,that nose wheel one could play a factor.

[MemberBerry]

If the MCAS system failures resulted from an unknown and unintended flaw in the software it is hardly surprising that the simulators had not been programmed with this unknown and unintended flaw.


Yes, but that's not the issue they found with the simulators. The issue they found was that it was too easy to move the trim wheels manually in the simulator.

That is a problem regardless of the reason causing you to use the manual trim wheels, and it's not just MCAS that can get you in this situation.

[safetypee]

MemberBerry, # correct, to the point :ok:

Of greater concern is the origin of the error.

Based on what is now known, it’s is unlikely that Boeing had flight tested the extreme condition, an offset stab (unable to control the aircraft). Thus the simulation may have extrapolated steady-state data from operation in the normal flight envelope, which was alluded to in the EASA query.
If so, this might clarify the choice of wording of the emergency AD drill; the belief that the trim wheel could be moved to recover the aircraft. A caveat on this is circumstance where the abnormal ‘yo-yo’ manoeuvre is required - when the trim could not be moved without relief. Furthermore, whether this circumstance is the same as previous variants (assumption in that what worked in the -200, would also work in the -700, NG, etc).

This point is also at the crux of the training / pilot error debate; how much time would the crew have before encountering excessive stick forces, and/or the trim not being available. (MCAS ‘failure’ was a pulsed input spread over a longer time period than might be assumed for a trim runaway - depending on trim rate).
The simulator demos in the media might correctly replicate the difficulty of control and trim, even with lower forces, but would be inaccurate because the difficulties are encountered earlier, a lower stab displacement, and at the stab limit the forces would be much higher.

An associated issue is if the simulated contribution of elevator stick forces were similarly too low; tail trim load is a combination of stab trim and elevator. This can be identified in flight test by mis-trimming the stab and counteracting with elevator, but only up to the point of excessive control force. Of interest this type of test is similar to that which identifies the longitudinal stability compliance - too little at low speed (need MCAS), too much at higher speed (need to reduce the stick force).

A conclusion from the simulator error could be that the 737 Max differs considerably from previous 737 variants, and that these differences are more than Boeing knew of, or had anticipated, - what was the extent of their assumptions.

[Fly Aiprt]

Yes, but that's not the issue they found with the simulators. The issue they found was that it was too easy to move the trim wheels manually in the simulator.

This raises a question : were the sim demonstrations last month conducted on updated simulators, or not ?

[1_of_600]

It's not a flaw, it's an omission!
You need to look into the source of the programming of the MAX simulators. Shouldn't take very long to track it down.
Delivered as a "binary" to the sim manufacturers, who are essentially just box integrators.
There was a very definite decision on the part of the software provider to include ONLY what is referenced in the FCOM and QRH.

Just sayin.....

[safetypee]

Fly Aiprt,
‘… were the sim demonstrations last month conducted on updated simulators, or not ?’

See my comments #8, “The simulator demos…”
It might not be possible to establish the standard of simulator used, but the timescales between demo and public notification of the problem suggest that the demo simulator was not updated. However, the standard of simulation might be irrelevant, particularly as the principle of the problem was identified during the demo.

The future relevance of a modified simulator is in the realism of a trim runaway. Particularly the crew’s ability to recognise the failure amongst potential confusion of ‘recently’ added trimming systems (STS, MCAS) and lack of ‘failure’ annunciation, and then physically recover the aircraft to a trimmed condition. The latter even more important if the point of ‘inability’ is significantly before the limit of the stab screw jack, different to that previously thought - less time for recognition and action.

A more speculative problem would be if the Boeing engineering simulator had a similar weakness, only identifiable when compared with the accident FDRs.

1_of_600, an ‘omission’ if the aircraft characteristic was previously known. If unknown then its a ‘flaw’, a big flaw in aircraft certification.

[DaveReidUK]

A conclusion from the simulator error could be that the 737 Max differs considerably from previous 737 variants, and that these differences are more than Boeing knew of, or had anticipated, - what was the extent of their assumptions.

I'd be surprised if the real-life trim wheel resistance varies significantly between the Max and the NG in any configuration, given the similar aerodynamics.

Of course on the NG, in the absence of MCAS, having to model accurately the effect of the combination of full AND stab trim and Up elevator in the sim probably sounded a crazy idea that would never happen in real life.

[1_of_600]

1_of_600, an ‘omission’ if the aircraft characteristic was previously known. If unknown then its a ‘flaw’, a big flaw in aircraft certification.


You're right, semantics do matter sometimes...

So, by omitting data and information about a known characteristic, did Boeing produce a flawed certification, and a flawed training simulation model? If MCAS "runaway" wasn't a known characteristic, it will not be in the simulation.

Regarding the forces necessary to move the stab trim wheel, I think you mentioned extrapolation in an earlier post... You're probably right about that. However the sim manufacturers also have limitations set by the characteristics of the hardware they install in the sim. They probably would not include a simulated trim wheel system capable of producing that level of force without expensive design to limit it's own potential runaway, purely from safety considerations. Probably one of the many things being re-thought now.

[Smythe]

I'd be surprised if the real-life trim wheel resistance varies significantly between the Max and the NG in any configuration, given the similar aerodynamics.

My concern would be that at 250kts, they could not manually trim the ac. (in the sim)

On the NG, you had 2 switches that operated elec trim and AP trim. On the MAX, the 2 switches are redundant, and either shuts stab trim completely off, going to manual wheel input only.
So far, it looks like MCAS has only initiated on AoA faults, so assume MCAS does initialize, you shut down the stab trim, and now have to rely on manual trim with faulty AoA measurements?

[safetypee]

Dave #12,
Agreed, the force characteristic of the trim wheel might not differ that much. However, the differences in the ‘not so similar’ aerodynamics (why was MCAS required) could result in the Max reaching the same value of force with a lower stab displacement, thus the inference is that the Max would have higher forces at the limit of the screw jack. Even a small difference could have a much greater effect if the elevator feel system was active due to AoA error.

Ignoring AoA (and MCAS, it must not fail the same way again), then the assumption in the trim runaway drill that there would be some elevator available to aid recovery from trim runaway - nose up ‘yo-yo’, could be severely challenged. (see FAA investigates drill)

Crazy (and surprising) things happen every day, :ok:
Even crazier - the 737 NG could differ in these characteristics from previous variants.
The more that is established from the investigations and ‘fall-out’ from these accidents, the more we don’t know. So who knew, who did not know, who needs to know, now.

In reality, a simple view could be that the sim might not have triggered the 4x force from the elevator feel shift at the same time as the MCAS induced trim movement, because MCAS was not expected to fail.
Trim runaway and AoA fail might also be an excluded failure in certification due to extreme improbability.

[bill fly]

I just wonder, presuming that all airliners are certified to the same parameters, how other (jet) transport aircraft stabilisers perform at high speed and extreme trim ranges.

Might be worth checking - and whether their sim software accurately reflects the true case in terms of
a. Elevator load and
b. Stab mechanism stall

Of course, we never go there - but if we do...

[thcrozier]

the media will NEVER be corrected on "anti stall" will they? good grief

Maybe it really is an anti-stall device, with a clever name to hide the fact.

[Byros]

If the 737 MAX simulator cannot correlate to real aircraft, then no amount of sim training could have prevented the outcome.

Boeing is in deep trouble.

[fgrieu]

The issue they found was that it was too easy to move the trim wheels manually in the simulator.


What substantiates too easy ? The only source I know is Boeing's statement (https://www.afp.com/en/news/15/boeing-acknowledges-flaw-737-max-simulator-software-doc-1gm9761) "changes will improve the simulation of force loads on the manual trim wheel". I have yet to find any authoritative source on if the change makes turning the wheel of the simulator easier or harder, and in which part of the flight envelope / position of the ailerons.

[OldnGrounded]

What substantiates too easy ? The only source I know is Boeing's statement (https://www.afp.com/en/news/15/boeing-acknowledges-flaw-737-max-simulator-software-doc-1gm9761) "changes will improve the simulation of force loads on the manual trim wheel". I have yet to find any authoritative source on if the change makes turning the wheel of the simulator easier or harder, and in which part of the flight envelope / position of the ailerons.

Virtually all of the news reports say that the flaw is that the trim wheels are too easy to turn, thus not actually simulating a situation such as the ones the accident flight crews faced, with extreme loading of the H-stab.

The NY Times story says:

Boeing recently discovered that the simulators could not accurately replicate the difficult conditions created by a malfunctioning anti-stall system, which played a role in both disasters. The simulators did not reflect the immense force that it would take for pilots to regain control of the aircraft once the system activated on a plane traveling at a high speed.

Emphasis added.

[MemberBerry]

What substantiates too easy ? The only source I know is Boeing's statement (https://www.afp.com/en/news/15/boeing-acknowledges-flaw-737-max-simulator-software-doc-1gm9761) "changes will improve the simulation of force loads on the manual trim wheel". I have yet to find any authoritative source on if the change makes turning the wheel of the simulator easier or harder, and in which part of the flight envelope / position of the ailerons.

From what I read Boeing's Saturday statement is in response to a New York Times article from Friday (later edit: actually, upon further reading, Boeing's statement is from Friday as well):

https://www.nytimes.com/2019/05/17/business/boeing-737-max-simulators.html

From the article:


The simulators did not reflect the immense force that it would take for pilots to regain control of the aircraft once the system activated on a plane traveling at a high speed.
[...]
On the Ethiopian flight, the pilots struggled to turn the wheel while the plane was moving at a high speed, when there is immense pressure on the tail. The simulators did not properly match those conditions, and Boeing pilots found that the wheel was far easier to turn than it should have been.


If the article was inaccurate, I doubt Boeing's wouldn't have pointed it out in their statement.

[Callsign Kilo]

There are several things a simulator won’t replicate. In some cases fidelity is no way comparable to the aircraft - for example, a manual reversion. I find it pretty unimaginable that the way a simulator characterises MCAS would mirror that of the aircraft. Certainly not in it’s present form.

[MemberBerry]

It's not about MCAS, it's about using the manual trim wheels at high speed, while pulling on the control column, with the aircraft severely out of trim nose down. And it's not just MCAS that can bring the aircraft in that situation.

And from what I've seen in Mentour's YouTube video, where he attempts to replicate that condition in a level D simulator for the 737 NG, the old simulator replicated the condition quite well.

So it seems the issue is specific to the MAX simulators. One requirement for a level D simulator is that it should accurately provide force feedback for the pilot's flight controls through a system called "control loading".

[b1lanc]

It's not about MCAS, it's about using the manual trim wheels at high speed, while pulling on the control column, with the aircraft severely out of trim nose down. And it's not just MCAS that can bring the aircraft in that situation.

And from what I've seen in Mentour's YouTube video, where he attempts to replicate that condition in a level D simulator for the 737 NG, the old simulator replicated the condition quite well.

So it seems the issue is specific to the MAX simulators. One requirement for a level D simulator is that it should accurately provide force feedback for the pilot's flight controls through a system called "control loading".

I'm wondering if downsizing the trim wheel on the Max was overlooked in the sim.

[YYZjim]

MCAS: Stick-force corrector or anti-stall device?

We are constantly reminded that MCAS is not an anti-stall device. It was introduced solely to cause the forces pilots experience from the yoke are more intuitive. So we are told.

But, is it possible that MCAS really is, and really is required, as an anti-stall device? I have not heard of anyone flying the MAX right up to stall other than Boeing test pilots, so all the information about what happens near that cliff has passed through Boeing management. Perhaps passing off MCAS as a minor change was intended to divert attention away from a more serious problem.

[GlobalNav]

Didn’t Boeing put several pilots through the sim to show them how good the “fix” is? Perhaps I have that wrong. But if so, how could the pilots make a reasonable judgement?

[safetypee]

From #91 , https://www.pprune.org/10475116-post91.html (RVF750)

‘I can confirm that Mechatronix NG simulator we have does not have loading to the trim wheels. Very disconcerting when you try the scenario and the F/O can wind in full Nose down on you.
And no, it's not recoverable in that state. Very sobering.’

Huummm. Mix that in with some earlier speculative assessment at #8, #11, #16.

Re MCAS purpose; from Boeing original statement, definitions.
“Maneuvering Characteristics Augmentation System (MCAS) – flight controlaw implemented on the 737 MAX to improve aircraft handling characteristics and decrease pitch-up tendency at elevated angles of attack.”
https://www.boeing.com/commercial/737max/737-max-software-updates.page

Also; “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.”

[gums]

Salute!

Good friggin' grief!!

The damned thing is not "anti-stall" or "stall prevention". It had to be added to the other kludges to keep "control" forces higher at higher AoA. Stall AoA is higher than MCAS activation, but if the plane requires less and less "control" force to increase AoA, then it will not satisfy the FAR requirements.
I am not sure where "they" are measuring "control" forces, so maybe FCeng or "racer" could add to the discussion versus all the "fly the plane" folks. Using control column force is not a valid test because Boeing already has sfwe and hdwe in the way between the elevator and yoke connections. Maybe PEI could also contribute, as he has some commercial plane test experience.
If we are talking about what the pilot feels, then all bets are off, because the plane has crapola gettin in the way already. If we are talking about the 737block 1, mod 0 then we have actual cables and such to measure the tension and so forth.
If we are measuring the tension back at the elevator hinge ( see diagrams we had from back in November), I would be more comfortable. In other words, we are seeing a changed pitch moment resulting from the new engine mounts and who knows what else compared to the grandfather plane so many claim is easy to fly and any "competent" pilot could have avoided the crashes.

Back to the BBQ now.

Gums.....

[gums]

Salute!

Before back to the seafood gumbo on the stove......
the FAR requirements were written to keep the airplane from getting too close to a stall.

My examination of the FAR requirements do not reference stall prevention, just a requirement to have increased back stick force or actual elevator force as AoA increases. They could have done the trim kludge at a much lower AoA.
The regulations were written with the intent of preventing a stall
Somehow, I cannot find "intent" on the applicable requirement sections. But on a personal level, I would always like for the plane to be harder to increase AoA the closer I got to the stall AoA. Maybe it's just me.

If you want to "prevent" a stall it is not easy for the engineers working with a 60 year old plane that has many changes in aerodynamic configuration and avionics. A FBW inmplementation that had AoA as a prime input might have made it thru the FAR process, but I doubt it. Even the 'bus had good longitudinal stability and such that the FBW laws could use. That was not the case with the Max. The thing did not meet the requirements and Boeing had to use the stab trim in a way most of us would not have anticipated.

Gums...

[fergusd]

Salute!

Before back to the seafood gumbo on the stove......


My examination of the FAR requirements do not reference stall prevention, just a requirement to have increased back stick force or actual elevator force as AoA increases. They could have done the trim kludge at a much lower AoA.

Somehow, I cannot find "intent" on the applicable requirement sections. But on a personal level, I would always like for the plane to be harder to increase AoA the closer I got to the stall AoA. Maybe it's just me.

If you want to "prevent" a stall it is not easy for the engineers working with a 60 year old plane that has many changes in aerodynamic configuration and avionics. A FBW inmplementation that had AoA as a prime input might have made it thru the FAR process, but I doubt it. Even the 'bus had good longitudinal stability and such that the FBW laws could use. That was not the case with the Max. The thing did not meet the requirements and Boeing had to use the stab trim in a way most of us would not have anticipated.

Gums...

And so the layers of deceit are peeling off boeings 'story' . . . the closer one looks into these things the worse it tends to get . . . sadly (for all the people that got killed) . . .

[thcrozier]

Salute!

Good friggin' grief!!

Back to the BBQ now.

Gums.....

Gums I always read your posts with great interest because you are far more knowledgeable than I, and many others here, about airplanes. I’m just a lowly Commercial SEL/IR with a thousand hours who quit flying myself around in 1993.

On the the other hand, I know a lot about marketing, risk assessment, and the lengths to which management will go to sell a buyer on the reliability of a system, while at the same time the engineers are telling them their expectations are impossible to meet. The world is full of examples of this phenomenon. Challenger, Columbia, and the FIU bridge collapse being just a few.

The MBAs are rarely able assess risk and come up with all sorts of clever vocabulary to hide it. I’m a Boeing shareholder and have faith in the long term future of the company. Short-term they have blown it big time. Their reluctance to ground the fleet even after it had essentially self-grounded worldwide is clear evidence of management in denial. I suspect that internally the engineers were waving red flags for a long time.

None of us know what was discussed at Boeing, but we do know that control forces became increasingly light as AoA increased, possibly to the point of needing an augmentation system to help the pilot push the nose down.

To me, MCAS was designed to prevent situations which might lead to a stall. As such public perception is going to stay “stall prevention”. A plane requiring a stall prevention system will scare the public. A system augmenting pilot control is an easy sell, like power steering.

[RickNRoll]

Salute!

Before back to the seafood gumbo on the stove......


My examination of the FAR requirements do not reference stall prevention, just a requirement to have increased back stick force or actual elevator force as AoA increases. They could have done the trim kludge at a much lower AoA.

Somehow, I cannot find "intent" on the applicable requirement sections. But on a personal level, I would always like for the plane to be harder to increase AoA the closer I got to the stall AoA. Maybe it's just me.

If you want to "prevent" a stall it is not easy for the engineers working with a 60 year old plane that has many changes in aerodynamic configuration and avionics. A FBW inmplementation that had AoA as a prime input might have made it thru the FAR process, but I doubt it. Even the 'bus had good longitudinal stability and such that the FBW laws could use. That was not the case with the Max. The thing did not meet the requirements and Boeing had to use the stab trim in a way most of us would not have anticipated.

Gums...

What is also puzzling is why, having used something as powerful as stab trim to increase the stick force, they had to go from an initial expectation of a much smaller 0.6 degrees max movement to a significanlty larger 2.5 degrees.

[gums]

Salute!

First of all, this is serious!!! @ Tak
I may not know as much about aerodynamics as some, but I had natural "touch" and survived........ Grew up on the Gulf coast and I enjoy nature's bounty while I can still move about and cook, ingest, digest and excrete.
- Use a blue crab or two that your break up and simmer in a quart of water for an hour
- make your own roux with flour and oil, or use one of the Louisiana blends. Many seafood spice packages here, so just watch the heat index if you are serving a wimp! Add this after the broth is nice
- veggie mix should be okra, sauteed celery and onion. Bell peppers are good, but not essential
- dump veggies in and simmer a long time or until the celery is tender.
- Then add a lotta popcorn shrimp and some fish fillet chunks.
- Simmer for maybe 30 minutes and dine!
++++++++++++++++++++++++++++++++++++++
BTW, I flew two jets with actual AoA/Gee limiters, and another one with a rudder shaker when getting high AoA.
I prefer the actual limiter as the 'bus has and maybe some kinda vibrator if getting too close to the corner of the envelope.
The Voodoo pusher yanked the stick outta your hand, and since it was connected to the elevator the AoA decreased for a second or two. You could still defeat it, but had to overcome about 28 pounds of force and then another 60 pounds for the limiter ( manual command signal limiter, if you had it on). If you tried that hard to pitch-up, then explain it to your boss and the accident board.
Remember that all this was back in the mid 60's with an airframe designed in late 50's.

Gums sends...

[thcrozier]

Salute!

First of all, this is serious!!! @ Tak
I may not know as much about aerodynamics as some, but I had natural "touch" and survived........ Grew up on the Gulf coast and I enjoy nature's bounty while I can still move about and cook, ingest, digest and excrete.
- Use a blue crab or two that your break up and simmer in a quart of water for an hour
- make your own roux with flour and oil, or use one of the Louisiana blends. Many seafood spice packages here, so just watch the heat index if you are serving a wimp! Add this after the broth is nice
- veggie mix should be okra, sauteed celery and onion. Bell peppers are good, but not essential
- dump veggies in and simmer a long time or until the celery is tender.
- Then add a lotta popcorn shrimp and some fish fillet chunks.
- Simmer for maybe 30 minutes and dine!
++++++++++++++++++++++++++++++++++++++
Gums sends...

This sounds good! Any recipe with the instructions “dump” in it has to be on par with a Clemenza Spaghetti.

[Takwis]

Gums, I'll bet you know more about aerodynamics than most.


But I copied your gumbo recipe, anyway.

I had to explain a few things to my boss. Mid-air collision with an F-117, for one... I wish I had some AoA stories, but I always managed to keep that pretty sane. Vmo...well, that's another story.

[bill fly]

...
Fun fact about stall prevention/recovery: "Apparently in the Gloster Javelin (a two seat delta with a T tail) if the aircraft got into a deep stall, the recovery technique was for the navigator to eject (thus shifting the CG forwards

Which proves that aeroplanes are more important than navigators.
Mind you, if that didn’t work the old man got out too...

[Water pilot]

And the hits keep coming. Did nobody test this thing before releasing it to the public?

This is rather concerning, because the simulator designers are going to be using the same engineering calculations used to inform the designers of the hardware. If the maximum forces involved in this system are much larger than calculated then that calls into question all of the engineering that went into it, from bolt sizes to the number of strands in the cable. This is at least the second indication that the engineering projections related to relocating the engines did not match reality.

[ZFT]

And the hits keep coming. Did nobody test this thing before releasing it to the public?

This is rather concerning, because the simulator designers are going to be using the same engineering calculations used to inform the designers of the hardware. If the maximum forces involved in this system are much larger than calculated then that calls into question all of the engineering that went into it, from bolt sizes to the number of strands in the cable. This is at least the second indication that the engineering projections related to relocating the engines did not match reality.

As other posters have stated before, Boeing’s supply binaries to the TDMs for integration. The TDMs are no longer involved in design.
​

[derjodel]

My examination of the FAR requirements do not reference stall prevention, just a requirement to have increased back stick force or actual elevator force as AoA increases. They could have done the trim kludge at a much lower AoA.


Gums,

to accept MCAS is there only to increase back stick force, we need to accept that Boeing engineers decided it was a good idea to repeatedly move the stab, probably one of the most critical control surfaces, just in oder to simulate increased stick force.

We need to accept that noone asked if this might by any chance be a bad idea. Like, you know, trimming all the way nose down in close proximity to the mother Earth. Noone remembered AF447 where pilot error resulted in full up trim, making stall unrecoverable - so nobody said "uhm guys, if something goes wrong and this system fails, it could mean trouble".

Again - all with the goal to produce some back stick force? I mean, for sure there must be better solutions for that alone?

What if there is another explanation? Perhaps the nacelles are so far out that it's a serious problem? Perhaps they act to some degree as canards when approaching stall? Imagine the wing stalls but the nacelles still produce lift... you'd be in big, big trouble.

Now this big big trouble would be something to justify MCAS. But due to competition and regulation requirements Boeing could not be open about it, or it would have meant entierly new certification. That's why it's half baked, hidden and relies on a single AoA.

Test flight data should show if this is indeed the case. Requirement for 2.5 units trim every 5s surely does seem like a smoking gun. Testimony from test pilot would be interesting. I'm sure we will hear about it in the coming years.

Then another question is, why were the cutout switches changed to prevent manual electric trim along with MCAS. There must be a reason. Without a reason it makes no sense. Did Boeing by any chance estimated that keeping the electric trim on without MCAS could easily put the airplane in the high AoA situation as described above? That could be another smoking gun. If that is not the reason to change the switches, then what was it? Any reasonably ideas?

[PEI_3721]

Gums et al, https://www.pprune.org/showpost.php?p=10475209&postcount=36 (https://www.pprune.org/showpost.php?p=10475209&postcount=36)
Re ‘like other 737’, stability ‘enhancement’, or anti stall.

First point, this discussion is probably in the wrong thread. Second, there is very little definitive information on these aspects (see Boeing refs), thus supposition reigns. Third, using what is known about an aircraft to build a simulator is not the same as taking a simulator and inferring what the actual aircraft characteristics are, but there ominous indications in this instance.

‘Like other 737’, is more of a marketing or commercial argument; as such it could be managed by other means (even if not saving $1M per a/c).
‘Stability vs stall’; the critical points are when and where. Any problems with stall ident - the point of stall, or characteristics at that point, or recovery capability, requires quick and forceful action, loud and positive, thus stick push or similar.
‘Anti stall’, not a good term, is more associated with stability, usually prior to stall warning - stick shake; awareness, but not always with action.

I conclude that MCAS is a requirement of stability certification, quiet and subtle background enhancement, ‘dressed-up’ to placate customers. Note the use of AoA vs speed input, flaps up only, and Mach input to manage what appears to be a complex problem in a small part of the flight envelope - wt, cg.

The trim input was intended to be small, incremental, and unobtrusive. The design theory appears to be OK, the engineering implementation, checking, and certification a disaster.
MCAS should not be compared with STS, or even Mach trim; nor its implementation as FBY.
I agree with gums # 36 …

Stability involves ‘feel’, measured as stick force in certification.
As an explanatory description only: consider trimmed flight at constant speed, constant thrust. Enter a turn without changing trim position or thrust (aircraft will descend). The ‘feel’ increases as a pull force; this can be assessed at various bank angles - different AoA.
Repeating these tests for a range of speeds (different trim speed) there should be a consistent relationship between bank angle and pull force. Inconsistent force - reduced pull force with bank angle, or with speed decrease at constant bank angle, could add workload in manual flight.
In extreme, slowing down in a turn or level flight at constant trim should always require more pull force, if not the aircraft feel is an unnatural ‘push to turn’. In some cases reducing forces could tend to oscillation or pitchup. There could be similar interactions with thrust - altitude.

MCAS “will not be a significant issue after modification”, nor require hands-on training except if the handling in the failed state is significantly degraded - cruise flight only. [opinion]

The important issues arising from the simulator mistake are in the processes of checking and certification of training simulators, and the possibly of a shortfall in knowledge about the aircraft, particularly the effect of trim failure.

Gums, [engineering or aerodynamic fix] Instead of lengthening the undercarriage, Boeing should have put the engines on top of the wing [/ fix] ;)

[gums]

Salute!
Thanks PEI and Derjo. Two good points.
Way back I mentioned the "canard" effect of moving the motors forward. I even propose using vanes or vortex doofers on the motors to keep the pitch moments close to the older model planes. A far superior solution than a half-assed sftwe thing embedded in an existing black box. End of rant
.
Secondly, I like having the discussion here, as it seems civilized and less visited by newbies that have not kept up with all the posts. I still have trouble with the control force versus the AoA as a criteria. This is from cert of the Airbus 320 and subsequent. My 'bus friends here will gladly help and confirm that there is zero feedback to those independent sticks. I am concerned here because the 737 and other Boeing planes have had artifical "feel" for decades. Looks like when all the "help" quits, the 737 still had a cable for the elevator and the little wheel for trim. Attaboy for the folks in the "just fly the plane" camp.

However, and this is important, those days are going fast, and I loved them as a wannabe astronaut. But we can't go back, and shouldn't, except to remind us of our roots.

We may never get to the xBox or Nintendo or Playstation world except the folks at Creech AFB. That's because in real aerospace vehicles you can sense gees and body rates. And so far, the carbon life-form creatures have done well in a few military jets and in the Airbus with zero feedback from the control surfaces. Besides, if you get high enuf there ain't no aero loads on the surfaces unless you are going the speed of stink ( ask my Blackbird friends).

Gums sends...

[BDAttitude]

Well, the side stick is spring loaded and therefore provides exactly what is needed. Feedback increases linear with angukar displacement which is G command.

[GarageYears]

The crux of this discussion revolves around the simulator accurately representing the aircraft.

How does any simulator do that?

Well, firstly it starts with DATA. Any simulator is essentially constrained to operate within the 'box' defined by the data provided by the airframe manufacturer as part of the data package.
(There is one caveat surrounding upset and recovery training, as recently introduced, which typically is based on engineering data or similar).

However, there is also the issue as to whether the simulator manufacturer implemented the data correctly. Aside from engineering simulators within Boeing (of which I know nothing), there are five Level D 737MAX simulators that have been certified - four manufactured by TRU Simulation + Training and one by CAE which I believe is currently operating with a Level C certification.

Was adequate data provided for the trim wheel loading? My guess is the data for the trim loading at high speed was NOT provided at low altitude at very high speed...

I suspect that is true of all simulators, whether 737MAX. NG or whatever.

- GY

[1_of_600]

The crux of this discussion revolves around the simulator accurately representing the aircraft.

How does any simulator do that?

Well, firstly it starts with DATA. Any simulator is essentially constrained to operate within the 'box' defined by the data provided by the airframe manufacturer as part of the data package.
(There is one caveat surrounding upset and recovery training, as recently introduced, which typically is based on engineering data or similar).

However, there is also the issue as to whether the simulator manufacturer implemented the data correctly. Aside from engineering simulators within Boeing (of which I know nothing), there are five Level D 737MAX simulators that have been certified - four manufactured by TRU Simulation + Training and one by CAE which I believe is currently operating with a Level C certification.

Was adequate data provided for the trim wheel loading? My guess is the data for the trim loading at high speed was NOT provided at low altitude at very high speed...

I suspect that is true of all simulators, whether 737MAX. NG or whatever.

- GY
You've said it well but still missed the main point....

The MAX simulator software is developed and delivered in binary form by Boeing.

The sim builders (both of them) just plug it in to the box. It does what Boeing decided it would do. No more, no less. Boeing decided what training was and was not required, what data and code was or was not necessary and what the course content would be.

All in the name of "I. P. protection". Yeah right.

And..... Guess what? The 777x / 777-9 will be done exactly the same way.

[RickNRoll]

Then another question is, why were the cutout switches changed to prevent manual electric trim along with MCAS. There must be a reason. Without a reason it makes no sense. Did Boeing by any chance estimated that keeping the electric trim on without MCAS could easily put the airplane in the high AoA situation as described above? That could be another smoking gun. If that is not the reason to change the switches, then what was it? Any reasonably ideas?



If the switches weren't changed then you could be flying the plane in manual mode with MCAS disabled. Kind of defeats the whole purpose of it. Much easier for Boeing to just fudge that too.

[Smythe]

Again, if you disable MCAS with the stab trim switches, you are left with manual trim. No AP option.
Manual stab trim with no AoA input?

How long is the flight?

[yoko1]

Again, if you disable MCAS with the stab trim switches, you are left with manual trim. No AP option.
Manual stab trim with no AoA input?

How long is the flight?

This statement makes no sense. I assume you aren't a pilot? AoA and manual stab trim aren't related. You trim the aircraft to relieve control pressures, not because of whatever the AOA indicates.

[yanrair]

If the 737 MAX simulator cannot correlate to real aircraft, then no amount of sim training could have prevented the outcome.

Boeing is in deep trouble.
Byros
there are dozens of scenarios which are explained by instruction rather than actually having to do them in the sim. Ditching for example which Sully did rather well.
. I was trained 30 years ago when sims never really represented the real plane that the STAB was a killer and that if you let it get way out of trim at high speed you have to “yo yo” to get it back in trim. So- the trick was and still is, don’t let it get out of trim. There is rarely a need for total authenticity in replicating many situations since they are usually demonstrated only once during type rating- manual reversion for example.
Happy flying
y

[robbiecando]

Please remember that the media are describing what the system is designed to correct in certain circumstances in simple language for the poor, ill-educated masses, including some politicians it seems, who don’t have the benefits of your inside knowledge.

Not for 'ill-educated masses', simply for non-pilots

[PEI_3721]

wilbert33 # 58,
Your summary represents the emerging understanding.
The important issue is why these aspects could have been overlooked, unknown, or unaccounted for.

The tail force-to-speed relationship might breakdown at larger angles; the tail could be adversely affected by wing downwash. The measured stick forces during test were probably limited by high forces, thus the extremes were extrapolated (non linear), supported by aerodynamic / structural theory (prediction). This may not have included the elevator, but presumably some aspect was considered in identifying the need for the yo-yo manoeuvre - when; 737-200 vintage extrapolated to new variants.
Longer fuselage, higher thrust engines, increased tail area ?

Similarly the additive effect of nose up recovery elevator which would be an opposing force (couple) on the tail mechanism could have been misjudged - not a normal evaluation of cross-control; possibly no data or extrapolation.

There has also been discussions that with large deviations of the combined tail angle and elevator, the elevator control suffers jack stall, thus is less effective.

The elevator feel adjustment is a well established device used with stick shake (AoA), it would be an unwanted additive to tail trim offset, but not a problem in isolation - AoA fail, no MCAS trim input.

Do we need superman ? No, we require humans in design, evaluation, test, and certification, who can foresee relevant eventualities whilst constrained by management and working practices, and subject to normal human variability due to the working environment. (Reasonable judgement)
The industry requires a robust processes of design - certification, something which will tolerate the variabilities above, so that pilots, with similar human limitations, can manage what is expected of them; manage the residual uncertainty.


yanrair, # 60,
‘So - the trick was and still is, don’t let it get out of trim.’
Aviation safety is not founded on “tricks”, particularly those which might not be described or repeated.
The are significant differences between what you ‘know’ (know what), and ‘how’ to use what you know; knowhow.

[yoko1]

wilbert33 # 58,

‘So - the trick was and still is, don’t let it get out of trim.’

Aviation safety is not founded on “tricks”, particularly those which might not be described or repeated.
The are significant differences between what you ‘know’ (know what), and ‘how’ to use what you know; knowhow.

I think you are misreading what he said. Proper trimming technique for any aircraft is not a "trick," it is a basic skill learned from day 1. Not defending the horrible design of MCAS, but aircraft manufacturers have to make some underlying assumptions regarding the pilots who will be operating the aircraft they build. Believing the pilot will apply normal operating techniques like trimming away undesired control forces is not an unwarranted assumption.

[yanrair]

wilbert33 # 58,
Your summary represents the emerging understanding.
The important issue is why these aspects could have been overlooked, unknown, or unaccounted for.

The tail force-to-speed relationship might breakdown at larger angles; the tail could be adversely affected by wing downwash. The measured stick forces during test were probably limited by high forces, thus the extremes were extrapolated (non linear), supported by aerodynamic / structural theory (prediction). This may not have included the elevator, but presumably some aspect was considered in identifying the need for the yo-yo manoeuvre - when; 737-200 vintage extrapolated to new variants.
Longer fuselage, higher thrust engines, increased tail area ?

Similarly the additive effect of nose up recovery elevator which would be an opposing force (couple) on the tail mechanism could have been misjudged - not a normal evaluation of cross-control; possibly no data or extrapolation.

There has also been discussions that with large deviations of the combined tail angle and elevator, the elevator control suffers jack stall, thus is less effective.

The elevator feel adjustment is a well established device used with stick shake (AoA), it would be an unwanted additive to tail trim offset, but not a problem in isolation - AoA fail, no MCAS trim input.

Do we need superman ? No, we require humans in design, evaluation, test, and certification, who can foresee relevant eventualities whilst constrained by management and working practices, and subject to normal human variability due to the working environment. (Reasonable judgement)
The industry requires a robust processes of design - certification, something which will tolerate the variabilities above, so that pilots, with similar human limitations, can manage what is expected of them; manage the residual uncertainty.


yanrair, # 60,
‘So - the trick was and still is, don’t let it get out of trim.’
Aviation safety is not founded on “tricks”, particularly those which might not be described or repeated.
The are significant differences between what you ‘know’ (know what), and ‘how’ to use what you know; knowhow.
Apologies for poor use of English Wilbert- for "trick " read "correct procedure/technique". Since STAB loads increase rapidly with airspeed changes it is not recommended to change speed. Pariculary with Jammed Stabiliser where it won't budge. Stay at same speed and plane stays in trim. Until of course you have to slow down to land but that is a different trick - sorry, procedure!
Cheers and thanks for your post.
Y

[yanrair]

Trick or treat?
Actually one of my major concerns about these two crashes and others like them, and the trend in general, is that the "tricks" - techniques, airmanship, generations of wisdom, are not being passed on to the next generation of trainers who in turn train the pilots. My last job involved doing just that - helping to pass on the gene pool of learning that can so easily be forgotten. So, my generation have massive respect if not fear of a runaway stabiliser - not fear because we know how to fix it - let's call it respect. Then earlier on in this thread we hear from a 737 pilot who wasn't even told that there were handles on the stabiliser wheels and what they were for!

I do agree that it is no good having a really good training captain full or "tricks" passing these onto those lucky enough to be with him during training since this only glides over the problem. If a trainer has a good idea that is not out there - tell us all about it. Let it then become airline policy.
Y
Happy flying.

[Ian W]

I think you are misreading what he said. Proper trimming technique for any aircraft is not a "trick," it is a basic skill learned from day 1. Not defending the horrible design of MCAS, but aircraft manufacturers have to make some underlying assumptions regarding the pilots who will be operating the aircraft they build. Believing the pilot will apply normal operating techniques like trimming away undesired control forces is not an unwarranted assumption.

And that is a good summation of the problem - one that is highlighted repeatedly here.
Boeing is being castigated for assuming that pilots would do what in the past would be considered "a basic skill learned from day 1". In fact it would have been deemed insulting to not assume it is an innate skill. Yet in multiple postings here and other threads it has been said that trimming to unload the control column is something special that needs to be highlighted in NNCs (what are those Ns - oh yes...). Is special training really required for experienced pilots in _trimming_?? Apparently yes. What would have happened to both accident aircraft if the pilots had kept them in trim? -A clue - that is what the crew of the penultimate Lion Air flight did.

So the real question is what _other_ assumptions based on past expected crew capabilities have been made by A or B? With training both initial and continuation cut to the tick box bone - and real hands-on experience being limited there surely are other assumptions that have been invalidly made.

Which approach will the bean counters favor? Automate out the pilots or (re)train them and increase continuation training? Does automation require a pension/401K?

[tdracer]

Boeing has already done that trick, back in 1984. When replacing the underslung JT-8Ds with the new CFM-56s, they had to push them out front (canard style), and then, yes, had to put a hefty strake on the side of the engine, to direct airflow where it needed to go. The engines got bigger at the NG stage, and the strake got a little bigger, too. I suspect there is a limit to how much you can do with that same solution...and the MAX (aptly named, for a number of reasons) reached, or passed, that limit. Hence the MCAS. I rather like PEI_3721's, solution, in that, at least it would be better than MCAS!

I still believe that one can still build a 707 type airplane, with cables, and pulleys, and servo tabs and balance bays...clean it up with more modern aerodynamics, new engines (only two of them, alas), stick 'em under the wings for nostalgia's sake (sorry PEI), and have it behave more predictably, more safely, more comfortably, more intuitively than an electric jet. But then, I'm getting close to retirement, and so are my ideas.

Tak, that's NOT why the strake is there, The reason for the strake is engine out approach - basically if the engine is not running, at higher angles of attack all that airflow spilling out of the inlet can cause flow separation on the wing - the strake acts as a big vortex generator to re-energize the flow and keep it attached over the wing. That's why you'll see strakes on the inboard side of the nacelles of most wing mounted big turbofan engines. Not to say they may not help in the way you're suggesting, but that's not why we put them there.

As for the engines being hung below the wings instead of out front - there are two main reasons for that unrelated to ground clearance (granted, the MAX is a pretty extreme case due to ground clearance). One is rotorburst - sticking the engine out front limits the amount of wing (and associated systems) that are exposed to damage due to a rotorburst. The other is that it's aerodynamically better - if the engine is too close to the wing, at cruise speeds and power settings you get interference drag due to the interaction of the fan flow with the free stream airflow. The drag penalty can be substantial - as much as 2% fuel burn. On the 747-8, they couldn't mount the engines as far forward as they wanted due to flutter issues and had to pay a not insignificant drag penalty as a result.

[tdracer]

Boeing is being castigated for assuming that pilots would do what in the past would be considered "a basic skill learned from day 1". In fact it would have been deemed insulting to not assume it is an innate skill. Yet in multiple postings here and other threads it has been said that trimming to unload the control column is something special that needs to be highlighted in NNCs (what are those Ns - oh yes...). Is special training really required for experienced pilots in _trimming_?? Apparently yes. What would have happened to both accident aircraft if the pilots had kept them in trim? -A clue - that is what the crew of the penultimate Lion Air flight did.

Posted about this on one of the countless other MAX threads, but it probably bears repeating.
Not too long ago I was at special event at the Museum of Flight - not only was I seated at a table with a bunch of current and retired Alaska Air pilots, during the cocktail hour I ran into a flight test pilot friend who'd been involved in the MAX development (Alaska is an all 737 operator - classics and NGs). To a man, they all agreed that if the stab tirm started doing something you didn't understand or like, the very first thing they'd do is turn it off and trim it manually. Hence the reason Boeing didn't treat MCAS as a flight critical system. However these were all older, high time experienced pilots
That being said, they also all agreed that no sim training for MCAS (or any of the other MAX differences) was a huge miss...

[Fly Aiprt]

To a man, they all agreed that if the stab tirm started doing something you didn't understand or like, the very first thing they'd do is turn it off and trim it manually. Hence the reason Boeing didn't treat MCAS as a flight critical system. However these were all older, high time experienced pilots
That being said, they also all agreed that no sim training for MCAS (or any of the other MAX differences) was a huge miss...

I suppose no one here will disagree with the above.
The unknown is and -remains -, when an how do you notice that the stab trim is doing something that you don't understand ?
In the real aiplane, with a continuous stickshaker, STS spinning the wheels and while running NNC's, how long does it take to notice the stab trim is acting up ?
One, two, ten, twenty seconds ?
The other unknown is, after the above number of seconds, is the out-of-trim recoverable with the manual wheels ? Or, at what speed and out-of-trim degree does the airplane become manually un-trimmable ?
Does someone really know, or do people just "believe", "suppose", "imagine" ?

Yeah, I know, "they shouldn't have..", "the trick is..".
But in the real world of real airplanes ?

[MurphyWasRight]

Posted about this on one of the countless other MAX threads, but it probably bears repeating.
Not too long ago I was at special event at the Museum of Flight - not only was I seated at a table with a bunch of current and retired Alaska Air pilots, during the cocktail hour I ran into a flight test pilot friend who'd been involved in the MAX development (Alaska is an all 737 operator - classics and NGs). To a man, they all agreed that if the stab tirm started doing something you didn't understand or like, the very first thing they'd do is turn it off and trim it manually. Hence the reason Boeing didn't treat MCAS as a flight critical system. However these were all older, high time experienced pilots
That being said, they also all agreed that no sim training for MCAS (or any of the other MAX differences) was a huge miss...

My bolding in above.
One of the criticisms I see of the ET crew is that they did not first trim with electric trim before using cutouts as hinted at in the Lion Air triggered emergency AD.

It is not clear whether lack of manual trim after cutout was due to lack of familiarity with the flip out handles or aero loads or both.
Something had to be happening in the period while electric trim was disabled, we don't know what since the CVR transcript has not been released.

[Uncle Fred]

Tak, that's NOT why the strake is there, The reason for the strake is is engine out approach - basically if the engine is not running, at higher angles of attack all that airflow spilling out of the inlet can cause flow separation on the wing - the strake acts as a big vortex generator to re-energize the flow and keep it attached over the wing. That's why you'll see strakes on the inboard side of the nacelles of most wing mounted big turbofan engines. Not to say they may not help in the way you're suggesting, but that's not why we put them there.

As for the wings being hung below the wings instead of out front - there are two main reasons for that unrelated to ground clearance (granted, the MAX is a pretty extreme case due to ground clearance). One is rotorburst - sticking the engine out front limits the amount of wing (and associated systems) that are exposed to damage due to a rotorburst. The other is that it's aerodynamically better - if the engine is too close to the wing, at cruise speeds and power settings you get interference drag due to the interaction of the fan flow with the free stream airflow. The drag penalty can be substantial - as much as 2% fuel burn. On the 747-8, they couldn't mount the engines as far forward as they wanted due to flutter issues and had to pay a not insignificant drag penalty as a result.

Nice explanation TD. I have to admit that in my flying years, that go back to the late 80's , I did not know that. In general yes, but not specifically like that. Always appreciate your insights from the engineering side.

[Ian W]

My bolding in above.
One of the criticisms I see of the ET crew is that they did not first trim with electric trim before using cutouts as hinted at in the Lion Air triggered emergency AD.

It is not clear whether lack of manual trim after cutout was due to lack of familiarity with the flip out handles or aero loads or both.
Something had to be happening in the period while electric trim was disabled, we don't know what since the CVR transcript has not been released.

The implication of your post is that the 737 pilot is happy to accept a significantly out of trim aircraft.
What was being assumed is that an experienced pilot would not let the automatics whatever they were put the trim where they didn't want it.- so when the Stab Trim cutout occurred there would not be a large out of trim condition to wind back.

There is also a lot of confusion between 'manual trim' with the column electric trim switch and manual trim with the non-powered trim wheel.

[pilotmike]

Not too long ago I was at special event at the Museum of Flight - not only was I seated at a table with a bunch of current and retired Alaska Air pilots, during the cocktail hour....

And there's the rub - anyone can spout off about what they'd do from the comfort of a chair at a table during cocktail hour, loaded with a beverage or 2.

It is all rather removed from the confines of a flightdeck with the stick shaker rattling your vision, doubling your pulse rate, and drowning out any noise from the trim wheel which is highly likely to be doing exactly what you want and need it to be doing for you... until you realise that it is no longer your friend, and it has put you in an in-recoverable corner in a matter of seconds, while you're gathering your ***** together over the stick shaker and other unbelievable symptoms which just don't add up.

Pass me another bevvy, 'cos the alcohol is beginning to work its magic on me and I've got a few more views on how bad pilots are these days to get off my chest while I have an audience... now sit yourselves down, make yourselves comfortable, and recharge your glasses!

[MurphyWasRight]

The implication of your post is that the 737 pilot is happy to accept a significantly out of trim aircraft.
What was being assumed is that an experienced pilot would not let the automatics whatever they were put the trim where they didn't want it.- so when the Stab Trim cutout occurred there would not be a large out of trim condition to wind back.

There is also a lot of confusion between 'manual trim' with the column electric trim switch and manual trim with the non-powered trim wheel.

What I was trying to point out is that the "greybeards" statement was that the first thing they would instinctively do when trim acted up for whatever reason was to hit the cutout switches then use the manual wheel to trim, not first attempt to correct the trim using thumb switches.

Up until MCAS probably the most likely cause of runaway trim was a HW failure of some sort such as stuck relay which the thumb switches might/might not override so would be logical to hit cutout as soon as possible before stab ran to the limits. Only takes seconds of runaway to get to a bad state.

An unanswered question is why the initial re-trim by ET pilot only removed about 1/2 of the MCAS trim, that and failure to manually trim after cutout are the final causes.

[GarageYears]

And there's the rub - anyone can spout off about what they'd do from the comfort of a chair at a table during cocktail hour, loaded with a beverage or 2.

It is all rather removed from the confines of a flightdeck with the stick shaker rattling your vision, doubling your pulse rate, and drowning out any noise from the trim wheel which is highly likely to be doing exactly what you want and need it to be doing for you... until you realise that it is no longer your friend, and it has put you in an in-recoverable corner in a matter of seconds, while you're gathering your ***** together over the stick shaker and other unbelievable symptoms which just don't add up.

Pass me another bevvy, 'cos the alcohol is beginning to work its magic on me and I've got a few more views on how bad pilots are these days to get off my chest while I have an audience... now sit yourselves down, make yourselves comfortable, and recharge your glasses!

I'm beginning to wonder if the pilot-less aircraft isn't really the answer then. Given that the folks up front apparently can't handle a real emergency, might as well not be there at all.

The "unbelievable symptoms" were discussed in great detail after the Lion Air loss, were they not? Stickshaker on one side, trim goes wonky after the flaps are retracted and the AP won't engage.... meanwhile let's leave the thrust at T/O and off we go.

- GY

[Fly Aiprt]

What I was trying to point out is that the "greybeards" statement was that the first thing they would instinctively do when trim acted up for whatever reason was to hit the cutout switches then use the manual wheel to trim, not first attempt to correct the trim using thumb switches.

Thank you for this clarification.
So, after all it appears that for some experienced US pilots, the first thing to do was throw the cutout switches ASAP and then use the manual trim wheel.

[infrequentflyer789]

What I was trying to point out is that the "greybeards" statement was that the first thing they would instinctively do when trim acted up for whatever reason was to hit the cutout switches then use the manual wheel to trim, not first attempt to correct the trim using thumb switches.


The greybeards may have another reason - in the "old" days it was specifically advised not to use main electric trim to counter a runaway.

Peter Lemme (satcom.guru) has tweeted parts of 727 manuals and found the procedure was to hit cutouts "immediately" with no mention of trimming in opposition (see image attached). He also shows other bits of the manual that were specifically advising against it.

Of course the greybeards also had the rollercoaster technique in their manuals and were possibly in their training too. It is possible that everything changed at the same point between classic and NG:

1. "Control airplane pitch attitude manually with control column and main electric trim as needed" added to procedure
2. trim wheel made smaller so manual trim less likely to work if mistrimmed
3. rollercoaster taken out of manuals
EDIT: and quite possibly related: 4. Stab trim actuators combined - classic had AP servo and main trim motor, NG has just single motor driven by AP or main elec trim

Then the MAX removed the separate autopilot-trim cutout too.

[Fly Aiprt]

I'm beginning to wonder if the pilot-less aircraft isn't really the answer then. Given that the folks up front apparently can't handle a real emergency, might as well not be there at all.

- GY

Not sure of what would have happened of the Lion Air and Ethiopian flights had there be no pilots on board ?
Maybe the flights might have ended sooner, so the people on board would have had less time to get bounced around before the crash ?


The "unbelievable symptoms" were discussed in great detail after the Lion Air loss, were they not? Stickshaker on one side, trim goes wonky after the flaps are retracted and the AP won't engage.... meanwhile let's leave the thrust at T/O and off we go.




Not sure of what procedure you are referring to with "Stickshaker on one side".
What would it look like ?
"In case stickshaker activation, first reach over to the other yoke to make sure only one shaker motor is running.
If so, then stickshaker alarm is spurious, ignore it and consider MCAS .
If not, then do stickshaker memory items"
Are you sure you would be able to diagnose the number of shaker motors running before running memory items ?
Really ?

[pilotmike]

I'm beginning to wonder if the pilot-less aircraft isn't really the answer then. Given that the folks up front apparently can't handle a real emergency, might as well not be there at all.

The "unbelievable symptoms" were discussed in great detail after the Lion Air loss, were they not? Stickshaker on one side, trim goes wonky after the flaps are retracted and the AP won't engage.... meanwhile let's leave the thrust at T/O and off we go.

- GY

Seriously? Your conclusion is that pilotless is better? So when the Lion Air 737Max was doing its best to kill everybody on board the day before the Lion Air crash directly caused by MCAS, what - or rather WHO - do you believe saved the day and the lives of all those souls on board?

Was it the aircraft?
Was it automation?
Was it MCAS?
Was it a computer?

Wrong, wrong, wrong and WRONG again! The automation was going beserk and was trying to kill them all. The correct answer is, it was....

...

...

a pilot!

Yes, a pilot saved everyone from the unthinking actions of an ill-conceived and very badly implemented computerised system.

Pilots 1, automated system gone wrong NIL on that occasion.

Unfortunately, MCAS equalised very soon after. And then MCAS went on to score the decider some months later, against Ethiopian. Finally, the ref has called FOUL, and has had the good sense to stop the match. But your solution is to remove the pilot? Incredible, given all the evidence!

[fizz57]

The automation was going beserk and was trying to kill them all.
...

Yes, a pilot saved everyone from the unthinking actions of an ill-conceived and very badly implemented computerised system.


Rather poor choice of example seeing that MCAS is not active during automatic flight. In fact the sole reason for MCAS is to provide pilot-friendly handling characteristics, and would not have been necessary in a pilotless aircraft :)

[Fly Aiprt]

Rather poor choice of example seeing that MCAS is not active during automatic flight.

Hmm, automatic flight ?
With a faulty AOA sensor ?
Really ?

My bet is, as soon as the AOA vane/synchro/whatever would have failed, the autopilot would have called it a day, and another mode of auto flight would have taken place : free fall...

[tdracer]

Pilot Mike and Fly Aiprt, you're rather missing the point. MCAS (and countless other systems) are less than 100% robust because it's assumed that if they do something stupid, the carbon based life forms sitting in row zero will take over and fly the aircraft. However, it's been repeatedly demonstrated that at least some of those carbon based life forms - otherwise known as pilots - are not completely up to the task (forget MCAS, think AF 447). If the designers have to design all the systems on the aircraft assuming the pilots won't take over and do the right thing when something fails, then why have pilots?
Of course this would take a complete re-think of how aircraft and automatics/avionics are designed, with many levels of redundancy and backups. The 'we must have pilots' side keeps pointing to cases where the pilots stepped in and saved the day when the automatics went south - but that's how the system is designed, assuming that when things go wrong the pilots will step in and save the day. The designers would not design the automatics that way if they couldn't count on the pilots to step in.
Now, I'm not suggesting we're to the point where we can design the aircraft so we don't need pilots - that's still decades away. But when pilots become completely overwhelmed and demonstrate the inability to even remember to pull the throttles back so they don't overspeed when something goes wrong, it moves us one more step in that direction.

[Fly Aiprt]

My experience in designing aircraft systems (not for Boeing ;-) is, the key is research for graceful failure.
If when your system fails, average crews get overwhelmed, then something is wrong with the system, not the crew...

That is, provided you actually test your system in real life in real airplanes, which we are not sure Boeing actually did.
And there is no need to this day, to resort to full autonomous commercial flight the technology is not mature to envisage yet.

As a designer, just do your homework, do study ergonomics and human factors, and you'll be the guy up to the task.

As to considering the crews being overwhelmed as not being "up to the task", well the people I know of, that have actually encountered life threatening situations with complex multiple alarms/failures, are far more tolerant than you are.
Because they know that they also have been overwhelmed, like anyone of us would have been, whatever our abilities, real or supposed.

[CurtainTwitcher]

My experience in designing aircraft systems (not for Boeing ;-) is, the key is research for graceful failure.
If when your system fails, average crews get overwhelmed, then something is wrong with the system, not the crew...

That is, provided you actually test your system in real life in real airplanes, which we are not sure Boeing actually did.
And there is no need to this day, to resort to full autonomous commercial flight the technology is not mature to envisage yet.

As a designer, just do your homework, do study ergonomics and human factors, and you'll be the guy up to the task.

As to considering the crews being overwhelmed as not being "up to the task", well the people I know of, that have actually encountered life threatening situations with complex multiple alarms/failures, are far more tolerant than you are.
Because they know that they also have been overwhelmed, like anyone of us would have been, whatever our abilities, real or supposed.

An A330 crew recently had a major electrical failure (https://thewest.com.au/news/aviation/qantas-bali-flight-diverts-to-broome-after-electrical-fault-ng-b881198138z) that generated 62 ECAM's (if my sources are correct) and landed in the the middle of nowhere at 0345 am body clock time on a short runway. I suspect when the facts come out about this it will be seen as one of the remarkable aviation achievements in spite of the automaton, bells & whistles.

Computers, automating & system design have a long long way to go before we can even begin to think eliminating carbon based pilots.

[Fly Aiprt]

But when pilots become completely overwhelmed and demonstrate the inability to even remember to pull the throttles back so they don't overspeed when something goes wrong, it moves us one more step in that direction.

As an aerobatics instructor, I can say with confidence that confronted with a difficult upset, every trainee WILL not even remember to pull the throttle back unless specifically trained for that.
We're talking of flying real airplanes, experiencing real spinning, gees, etc. not calibrated sim exercices with a warned crew.
And my trainees ranged from low time students to multi kilo-hour F117 pilots or airline captains. So...

Besides, confronted with a stickshaker alarm and severe AND trim at low altitude, retarding the throttles seems a rather odd idea... But I may be wrong.

[derjodel]

My experience in designing aircraft systems (not for Boeing ;-) is, the key is research for graceful failure.
If when your system fails, average crews get overwhelmed, then something is wrong with the system, not the crew...


This is a very important point, and people should really understand it. To simplify, if we assume "piloting skill" is something we can measure with a single number, pilots would still have a different level of skill. Normal distribution is defined by the mean and standard deviation, the latter simply telling us how big is the difference between "top guns" and "barely passed the certification".

https://cimg2.ibsrv.net/gimg/pprune.org-vbulletin/500x209/threenormaldistributions_497e1c40eb6f45b1b77327fe4712e85a664 9a7ec.png
Standard deviation


Lots of posters here assume they themselves are, and everybody else should also be nothing but a "top gun" (the blue graph above). Does that reflect reality? I'll let yourself answer that.

But of course the reality is even more complex. Top guns have bad days. Could be a headache, could be a divorce. That stuff will kick you out of your best, within the magenta graph. In essence, even a top gan is only top gun on their best days. Then of course we'd need to measure different parameters. Somebody might be an ace pilot, but will miss a detail, like say trim spinning in the wrong direction (while somebody else might suck at hand flying but would thought, "hmm, this is weird").

So what can you do? Two things:
- stringent certification requirements which measure not only the knowledge and skill, but also "soft skills*" like "ability to keep calm while overspeeding 1000 feet AGL with severe trim down"; in essence making sure nobody by say level 90 on the magenta graph passes (which would run the industry crazy due to "shortage**" of pilots)
- design the systems so that not an average crew can handle it, but any certified crew can handle it. There is a big difference between the two, but I'm sure everyone can agree a certified crew should be able to handle a plane, or there is a problem.

*not really skills as those can't be learned.
**perhaps a big part of a problem - HR market requirements allow subpar pilots to be certified?

[pilotmike]

Rather poor choice of example... In fact the sole reason for MCAS is to provide pilot-friendly handling characteristics,

And an unbelievably inappropriate choice of words by you. MCAS turned out to be anything but "pilot-friendly" when it went rogue on 3 occasions. The results speak for themselves. Nothing friendly about that, at all.

The fact is, MCAS was an ill-considered and very badly bodged attempt to fix another much more serious problem, purely for financial reasons, and the un-intended consequences had horrific results. If it was anything other than a totally unsafe and unsatisfactory system, why would the aircraft be banned by every administration, with no satisfactory fix or end to the desperately sorry saga in sight?

It was nothing whatsoever about being "pilot-friendly" - rather it was a cynical, cheapskate fix for a bigger problem, which was pushed through the regulatory system inappropriately, many would say, and it has killed far too many people.

[Alchad]

It was nothing whatsoever about being "pilot-friendly" - rather it was a cynical, cheapskate fix for a bigger problem, which was pushed through the regulatory system inappropriately, many would say, and it has killed far too many people.

Absolutely, sums it up in a nutshell. This could well be the prosecution closing argument.

Regards

Alchad

[bill fly]

Hi Jodel,
I think we should distinguish between average folk and average professional pilots in this argument.
Pilots only get (and stay) where they are after stringent selection, training and repeated checking. They are a small part of the community.
That doesn’t mean they have to be especially clever or elite - but just ideal people for their job. Other people are ideally suited to other professions, at which many pilots may fail.
A professional pilot should be well up on your scale - if that scale applies to flying skills/aptitude as applied to the general community. The bad ones get found out with very few exceptions.
A doctor or an architect doesn’t go through this process - has to prove him(her)self to stay in business in other ways...
So an aircraft designer has the right to expect a certain level of skill. I seem to remember in my manual, that Boeing were pretty clear on the skill tests their pilots were expected to pass. That needs to be borne in mind when using terms like “average”.
On the other hand, a professional pilot has the right to expect a level of skill from the manufacturer. This includes honest self criticism such as is required of a professional pilot.
B

[FrequentSLF]

Hi Jodel,
I think we should distinguish between average folk and average professional pilots in this argument.
Pilots only get (and stay) where they are after stringent selection, training and repeated checking. They are a small part of the community.
That doesn’t mean they have to be especially clever or elite - but just ideal people for their job. Other people are ideally suited to other professions, at which many pilots may fail.
A professional pilot should be well up on your scale - if that scale applies to flying skills/aptitude as applied to the general community. The bad ones get found out with very few exceptions.
A doctor or an architect doesn’t go through this process - has to prove him(her)self to stay in business in other ways...
So an aircraft designer has the right to expect a certain level of skill. I seem to remember in my manual, that Boeing were pretty clear on the skill tests their pilots were expected to pass. That needs to be borne in mind when using terms like “average”.
On the other hand, a professional pilot has the right to expect a level of skill from the manufacturer. This includes honest self criticism such as is required of a professional pilot.
B
Jodel in his graph made clear that represent from "barely pass certification" to "top gun", which if my understanding correctly represent the total number of professional pilots.
He is already distinguishing between average folk (me) and average professional pilots.
There will be always an average, no matter how high you set the bar, some will be above the average and other below the average.
In fact, in my opinion, he has a very good point. Aircraft certification requires that the design is made for average certified professional pilot, which implies that a number (no matter how stringent is the certification process) of professional pilots will not fall in that category, therefore he is suggesting to have system designed (for aircraft certification purpose) for every certified professional pilot (not the average).

[derjodel]

Hi Jodel,
I think we should distinguish between average folk and average professional pilots in this argument.


We are.


Pilots only get (and stay) where they are after stringent selection, training and repeated checking. They are a small part of the community.


According to wikipedia (https://en.wikipedia.org/wiki/Aircraft_pilot), there are 290,000 airline pilots worldwide.


That doesn’t mean they have to be especially clever or elite - but just ideal people for their job. Other people are ideally suited to other professions, at which many pilots may fail.


Yes, that's why they are pilots.


A professional pilot should be well up on your scale - if that scale applies to flying skills/aptitude as applied to the general community. The bad ones get found out with very few exceptions.


This paragraphs seems to indicate either:

a) lack of understanding of basic statistics / probability

First of all, the picture above is not a scale, it's a distribution where 100 represents the average pilot. There will always, always be pilots who are below average and pilots who are above average. The question is, how below average is the worse commercial pilot currently flying (and how good is the best one on the opposite side).

The only thing that could be up to debate is weather the standard deviation is small (blue graph) and all pilots are almost equally good in skills, or is it (in the extreme) like orange graph and some pilots are way worse than the average pilot.

I don't have the data to support either of the cases. I hope we are closer to the blue graph, but... are we?

Interestingly, Boeing - by blaming the pilots for recent crashes - is saying that there is a wide gap between average and sub-average pilots AND that their airplanes are only meant for average skilled pilots, which in essence means, Boeing is saying the current certification is to permissive (you can't accept the current pilot certification as acceptable AND blame the pilots, it's a contradiction).

b) all pilots are exactly equal

You need to believe that all pilots are exactly the same if you are saying there is no distribution. Oh, I hear you, you are thinking "but that's not what I said, what I said is that the bad ones are filtered out".

Then we are back to a). Because, if you have pilots who have skills 89,93,100,101,110, the average is 98.6. You filter out the bad ones and end up with 93,100,101,110. Now your average is 101, but that is not above average, that's just the new average. Now the pilot with skill of 100 is below average. By pruning the bad ones you are improving the average, but the distribution is still there

There is another phenomena, if the airplane manufacturers are truly targeting "average" pilot, and that average goes up, the number of pilots who can handle the plane properly will decrease, simply because there are just so many pilots who are at or above average, which is a problem.

According to wiki quoted above, Boeing expects 790,000 new pilots in 20 years from 2018. What will their skills be compared to the current population, where we still have pilots who are not "children of magenta"? Where will the average move in the next 20 years with so many fresh pilots AND old pilots retireing? My bet is, it's going to go down cosiderably, while the airplane manufacturers are building planes for just the best of them. If that is the case, perhaps MCAS (hidden system which below average but certified pilots could not handle) is just showing us the future where we are going.


A doctor or an architect doesn’t go through this process - has to prove him(her)self to stay in business in other ways...


Everybody goes through the process, including (or especially) doctors. But yes, some fields have wider variations that others.


So an aircraft designer has the right to expect a certain level of skill.


I would argue they need to support any pilot with a valid license. If they think licensed pilots are not good enough for their machines, we have a big, big problem.



I seem to remember in my manual, that Boeing were pretty clear on the skill tests their pilots were expected to pass. That needs to be borne in mind when using terms like “average”.


Boeing can only dictate the minimum. Average is outside of their control. That said, do they specify that "a pilot should at minimum be able to identify a problem with an undocumented system, while flying close to the ground at high speed with stick shaker, no indication that AoA diasagree, and trim behaving like it's just the opposite of STS but not quite runaway stab trim". Do they? No really, do they? We are talking about a minimum skill required here.



On the other hand, a professional pilot has the right to expect a level of skill from the manufacturer. This includes honest self criticism such as is required of a professional pilot.
B


Challenger didn't happen due to engineering skill problem. On the contrary, engineering told the managers exactly what the risk was. These problems are created by greed.

[DaveReidUK]

First of all, the picture above is not a scale, it's a distribution where 100 represents the average pilot. There will always, always be pilots who are below average and pilots who are above average. The question is, how below average is the worse commercial pilot currently flying (and how good is the best one on the opposite side).

An equally valid question would be why ability (at anything, not necessarily confined to piloting) should be assumed to follow a normal distribution (normal in the mathematical sense, i.e. symmetrical about the mean/median/mode).

It's perfectly possible that the ability distribution is actually skewed, in other words the peak isn't halfway between the two extremes, so that for example there could be a concentration of pilots towards the upper end of the ability scale.

Or, perish the thought, the lower end. :O

[Fly Aiprt]

It's perfectly possible that the ability distribution is actually skewed, in other words the peak isn't halfway between the two extremes, so that for example there could be a concentration of pilots towards the upper end of the ability scale.

Or, perish the thought, the lower end. :O

Possible with a small number.
With a sufficient number of thousands of individuals, the distribution always seems to more or less follow the above curve, whatever the criteria, be it body height, fish size, driving or piloting "abilities".

The notion that some professional pilots could not be up to "average" may seem a bit disturbing, and yet we all know some colleagues that are "outstanding". And unfortunately, some others are not as good, so this de facto creates a distribution where some are "above average" and some are below...

[fergusd]

Now, I'm not suggesting we're to the point where we can design the aircraft so we don't need pilots - that's still decades away. But when pilots become completely overwhelmed and demonstrate the inability to even remember to pull the throttles back so they don't overspeed when something goes wrong, it moves us one more step in that direction.

Interesting, I think that there are challenges remaining, but would expect to see fully autonomous aircraft as a more realistic proposition than genuinely fully autonomous cars. The technology will be there within a decade (about the same time as I will never have to fly again I hope).

However, Do I have the confidence in the implementation quality and regulatory enforcement which makes that technology actually safe ? . . . No chance.

There is a 'race to the bottom' in the software business which largely makes it fundamentally incompatible with the creation of high end safety critical solutions.

The more complex and critical those systems are the more incompatible that approach becomes. I see it regularly in many industries and it is both concerning and somewhat eye opening when you see what is really going on under the skin, as may be becoming apparent with this aircraft.

[derjodel]

An equally valid question would be why ability (at anything, not necessarily confined to piloting) should be assumed to follow a normal distribution (normal in the mathematical sense, i.e. symmetrical about the mean/median/mode).

It's perfectly possible that the ability distribution is actually skewed, in other words the peak isn't halfway between the two extremes, so that for example there could be a concentration of pilots towards the upper end of the ability scale.

Or, perish the thought, the lower end. :O

Well, first of all, it doesn't really matter. The whole argument still stands, even if the distribution is skewed. Worse even, if it's skewed, then you have either some really, really bad pilots out there, or most pilots are below average, with a few aces flying around. Which one would you choose ;-)?

But in reality, it's very likely very close to normal. Take an example: how is the height of NBA players distributed? We pick the best, tallest athletes from the whole population, for sure they will all be best of the best, right? There are just 500 NBA players, out of the population of 300,000,000, that's top 0.000167%. For sure their height can't be normally distributed??

Well, spoiler alert, their height is basically normally distributed.
https://cimg9.ibsrv.net/gimg/pprune.org-vbulletin/610x500/cwxo8cgwsae6e_v_0eca02870f22d55495b357df32c7d19b15db5924.png
Ok, actually this is showing % of minutes played by different heights, so one could expect it to be even more skewed, but it's not.

And that's just 500 players. There are 600x more airline pilots, and with the sample size like that, the distribution is going to be normal.

[DaveReidUK]

And that's just 500 players. There are 600x more airline pilots, and with the sample size like that, the distribution is going to be normal.

OK, I must have missed the lesson in statistics class demonstrating that large distributions are always normal.

[derjodel]

OK, I must have missed the lesson in statistics class demonstrating that large distributions are always normal.

You are trying to make the whole arument invalid on a premise of "pilot skills don't necesseraly follow normal distribution.

1. How exactly does the type of distribution (https://blog.cloudera.com/blog/2015/12/common-probability-distributions-the-data-scientists-crib-sheet/) make my argument invalid? Plug in whatever you like, it's still the same.

2. Yes, indeed you might have missed the Central Limit Theorem. And because skill is sum of many different factors, the pilot skill is even more likely to be normally distributed.

[compiforce]

I am not a pilot, but I am a Chartered Engineer, Chartered Statistician, a Chartered IT Professional, a European Engineer, and a Fellow of the Royal Statistical Society, with a Doctorate in Operational Research and Mathematical Statistics.

Not all natural distributions are Normal. However, very many are, because of the Central Limit Theorem. In essence, if a physical measurement is the sum of smaller figures, it is very likely to be Normal. That's why heights of people are approximately Normal, IQs are roughly Normal, etc.
It would be very surprising if the performance of pilots differed much from a Normal distribution.
A Normal (or Gaussian) Distribution is entirely described by its mean and variance. So, unless the distribution of performance of pilots is very unusual (such as a Laplace Distribution), the only relevant features of the distribution of pilot performance will be the mean (or average) and the variance (which determines how many are in the "tails" and how long the "tails" are).

[FlightDetent]

The word average (skilled) pilot should be avoided for the purpose of this discussion. If we are true to the mathematical meaning, adding even a small number of extremely well proficient people to a group, the previously "average" pilots of that group become less-than-average.

Less than average skilled sounds like "not competent enough to fly an aircraft designed for an average pilot". Which itself is an element of the debate, thus the word "average" makes a big mess of everything here.

My suggestion is that "the average pilot" in the certification realm means "any pilot who passed the licence proficiency criteria with at least bit of a margin, is able to achieve that or better performance continuously (as his experience builds but memory fades), and is not having a bad day"

[alf5071h]

Recent discussions have revolved around the man and machine. The inconclusive (different) views reflect the difficulty in defining the problem, together with the natural human dislike of uncertainty. These are further restricted in forum communication, how best to convey judgements with text, especially where these depend on the task being assessed and operating environment, each with a range of interpretations.

There are views of the man or machine, and then the ‘vague’ interface between them. Each attempts to quantify parameters of reliability and/or performance; similarly the concept of average seeks quantity, but most cases the real world the divisions are subjective - uncertain - qualitative judgement. (Skill is not a number, it’s a subjective rating). Numerical or statistical assessments are meaningless without first constraining the contributing parameters.

This is like an experiment judging human performance for a given task (detect trim runaway and act), but where the environment defining the situation is primarily determined by the test subject - the experiment is unbounded. We cannot measure what the crew perceived or ‘felt’ about the situation, or know what they knew - or recalled, or why they acted as they did.

All that might be concluded is that the performance of the single ‘entity’, the human and machine (humach?) in the circumstance was insufficient for the task. The balance of contributions in improving ‘humach’ capability is a judgement, best made within the guidelines of certification and the wide range of experience in that process.
The recent accidents suggest that the overall process for this judgement (certification, design, and test) was flawed; the resultant uncertainty in system failure could not be managed by ‘humach’.

The certification process is being reviewed, thus should identify technical and human aspects to be reassessed, as should be the resultant uncertainty from system failure.

We may not be able to judge any of the above without greater understanding of the original certification, the accident system failure, proposed modification, and most of all the associated justification.
With that, the subjective discussion in this forum and elsewhere could be founded on fact - reducing some of uncomfortable ‘uncertainty’ and need to resort to quantities to describe uncertainty.

[bill fly]

Sorry, I don’t buy that graph. Too convenient. If we have over 7 billion people in the world and just over 50% are in working age, and there are 300,000 odd pilots in the world, then there is one pilot per 14,000 people roughly - a small number, even if the figures are da-neben.
Let’s say there is a similar graph of 7 odd billion people. Where do you put these few pilots on it? And I am talking about selected airline employed professionals.
What do you measure on such a graph?
Intelligence? Co-ordination? Leadership? Team spirit? Drive? Fitness? Imagination?
All these factors bear on a pilot’s suitability as well as other, conflicting things such as:
Ability to concentrate on an item vs ability to keep the big picture
Ability to stick to procedure vs ability to throw away the rules and improvise
Ability to command vs ability to take advice
Ability to insist vs ability to be patient.
Ability to consider carefully vs ability to act fast
and many other conundrums which don’t mean you are especially clever - but suitable.
If you could quantify and measure all those qualities I very much doubt you would get a nice neat rounded peak for the general population as you would for an intelligence alone graph and even if you did, if you take a bunch of people out of it and scan them, the shape of the new graph would depend on where they were on the old one. Still a nice hill in the middle? I don’t think so.
Now about getting a licence - that is one small part of getting an airline job - one which you have to have but not one which will cause an airline to take you - they are looking for the items above.
So to design a machine for anyone who has managed to get a licence is not what is required. The operative word is professional. Design a machine for the pros and train them on it, then you will have safety.
Anyway the Jodel is a good ship - and as we know old men think there are lies, damn lies and statistics! ^_^
B

[edmundronald]

You are trying to make the whole arument invalid on a premise of "pilot skills don't necesseraly follow normal distribution.

1. How exactly does the type of distribution (https://blog.cloudera.com/blog/2015/12/common-probability-distributions-the-data-scientists-crib-sheet/) make my argument invalid? Plug in whatever you like, it's still the same.

2. Yes, indeed you might have missed the Central Limit Theorem. And because skill is sum of many different factors, the pilot skill is even more likely to be normally distributed.

The question is how bad is the cutoff across the skill axis which is necessary in an emergency? Is your cloud really spherical or is it eccentric?

To become a *commercial* pilot you need ability to take orders, ability to learn, to do CRM, to be methodical etc etc. And a certain ability to fly the plane. And all these factors are tested for. But on the day HAL decides to take a vacation or turn hostile, you suddenly need grace under pressure, a capacity to retain situational awareness, an ability to hand fly the plane in a configuration which has *not* been trained for, a will to find some way to survive. See Sully.. The problem with the aircraft design may be that the designers assume that the pilots have been selected adequately because of the shape of the large distribution, and find out that the distribution of skills necessary in an emergency is actually a bell shaped but much sharper curve with few individuals matching the criteria necessary for survival.

Edmund

[derjodel]

The word average (skilled) pilot should be avoided for the purpose of this discussion. If we are true to the mathematical meaning, adding even a small number of extremely well proficient people to a group, the previously "average" pilots of that group become less-than-average.

Less than average skilled sounds like "not competent enough to fly an aircraft designed for an average pilot". Which itself is an element of the debate, thus the word "average" makes a big mess of everything here.

My suggestion is that "the average pilot" in the certification realm means "any pilot who passed the licence proficiency criteria with at least bit of a margin, is able to achieve that or better performance continuously (as his experience builds but memory fades), and is not having a bad day"

You are spot on. "Average" and "Subaverage" are usually interpreted very, very wrong. Not just by common people, even by people who really should know, like doctors (e.g., they would like to get rid of all subaverage tall kids, and go as far as even giving them growth hormones, producing just a smaller standard deviation and even more pressure on the small kids).

That's why I'm saying, designing airplanes for average pilot is very, very bad idea. Planes need to be designed for any certified pilot. And to make discussion clearer, perhaps it would be good idea to stick to those terms.

Because once you accept that planes must be designed for all certified pilots, then you can not, by definition, blame pilots to not react properly, specially in this situation.

It's actually the other way around - this crash is giving us a data point how a certain crew of certified pilots reacted in under specific circumstances. Actually we have a few more data points: penultimate Lion Air crash, ultimate Lion Air crash and ET crash. All are certified pilots. They reacted differently. Was the system designed to accommodate any certified crew?

And I'll go further and ask: isn't it the fact that the crew can not be responsible. Either they should not have been cleared to fly (e.g. stripped of their license, fail the MAX rating). Either they meet the minimums or they don't. If they don't and they still got licensed, it's not their fault. It's the regulator's. If they crashed because the plane was not flyable by any certified crew, it's boeing's fault.

The crew could only be at fault if they were under influence, if they cheated to pass the tests or hide any medical conditions etc...

[FlightDetent]

d.j., your last paragraphs are missing the case of a properly qualified and competent pilot severely underperforming on a certain occasion. At this point, I am not connecting any dots to the MAX incidents and crashes as far as the crew performance goes, though a general remark that a poorly designed and implemented aircraft system (such as the MCAS on the MAX) will tip the scales.

[hans brinker]

I am not a pilot, but I am a Chartered Engineer, Chartered Statistician, a Chartered IT Professional, a European Engineer, and a Fellow of the Royal Statistical Society, with a Doctorate in Operational Research and Mathematical Statistics.

Not all natural distributions are Normal. However, very many are, because of the Central Limit Theorem. In essence, if a physical measurement is the sum of smaller figures, it is very likely to be Normal. That's why heights of people are approximately Normal, IQs are roughly Normal, etc.
It would be very surprising if the performance of pilots differed much from a Normal distribution.
A Normal (or Gaussian) Distribution is entirely described by its mean and variance. So, unless the distribution of performance of pilots is very unusual (such as a Laplace Distribution), the only relevant features of the distribution of pilot performance will be the mean (or average) and the variance (which determines how many are in the "tails" and how long the "tails" are).

Q: If the average applicant scores 100 points on the pilot qualification test, made up of theoretical questions, and a simulator check, you say they should probably normally distributed. Lets say 60 % score between 90 and a 110 points, and the airlines only hire the pilots with over 90 points. With the 20% percent who didn't cut it missing, wouldn't the resulting curve be non-normally distributed? (and seeing hiring rates around 50% at interviews makes me think the skewing would be worse)

Had statistics class in college about 3 decades ago.....

[pilotmike]

Statistics are used by most people - especially politicians - just as drunks use lamp posts:-

... more for support than for the light they shed!

[Fly Aiprt]

That's why I'm saying, designing airplanes for average pilot is very, very bad idea. Planes need to be designed for any certified pilot. And to make discussion clearer, perhaps it would be good idea to stick to those terms.

Because once you accept that planes must be designed for all certified pilots, then you can not, by definition, blame pilots to not react properly, specially in this situation.

It's actually the other way around - this crash is giving us a data point how a certain crew of certified pilots reacted in under specific circumstances. Actually we have a few more data points: penultimate Lion Air crash, ultimate Lion Air crash and ET crash. All are certified pilots. They reacted differently. Was the system designed to accommodate any certified crew?

And I'll go further and ask: isn't it the fact that the crew can not be responsible. Either they should not have been cleared to fly (e.g. stripped of their license, fail the MAX rating). Either they meet the minimums or they don't. If they don't and they still got licensed, it's not their fault. It's the regulator's. If they crashed because the plane was not flyable by any certified crew, it's boeing's fault.

The crew could only be at fault if they were under influence, if they cheated to pass the tests or hide any medical conditions etc...

Excellent point, and excellent post !

[TehDehZeh]

Q: If the average applicant scores 100 points on the pilot qualification test, made up of theoretical questions, and a simulator check, you say they should probably normally distributed. Lets say 60 % score between 90 and a 110 points, and the airlines only hire the pilots with over 90 points. With the 20% percent who didn't cut it missing, wouldn't the resulting curve be non-normally distributed? (and seeing hiring rates around 50% at interviews makes me think the skewing would be worse)

Had statistics class in college about 3 decades ago.....
Now consider that every airline has a different hiring process leading to a different cutoff. Maybe the cutoffs are normally distributed across the world..
You could also think about how consistent such a cutoff would be implemented, probably the hiring decision also depends on non-piloting factors.
In the end, the reason a pilot is given a license is because it is thought he is able to fly a plane. Obviously this needs to be consistent with what is required to actually do it, or there is a problem.

[yoko1]

In the end, the reason a pilot is given a license is because it is thought he is able to fly a plane. Obviously this needs to be consistent with what is required to actually do it, or there is a problem.

If you want the statistical argument to work in your favor, then it is important that both pilots in the cockpit meet these standards. That way you have a greater chance that one of them will be up to the task on any given day.

[FlightlessParrot]

Q: If the average applicant scores 100 points on the pilot qualification test, made up of theoretical questions, and a simulator check, you say they should probably normally distributed. Lets say 60 % score between 90 and a 110 points, and the airlines only hire the pilots with over 90 points. With the 20% percent who didn't cut it missing, wouldn't the resulting curve be non-normally distributed? (and seeing hiring rates around 50% at interviews makes me think the skewing would be worse)

Quite so. And while there will be an average level of skill for qualified pilots, the distribution is not at all likely to be normal, because you've selected the group, and the value will be pulled to the left by the sky gods.

Anyway, average is probably the wrong measure of central tendency. Better is almost certainly the median: by definition, half the population will be at or below the median value. Clearly, aircraft have to be readily manageable by the half of qualified pilots who fall below the median--especially aircraft like narrow-body twins, which are likely to fly the bread-and-butter routes that don't attract the profoundly gifted.

The issue, surely, is not to do with statistics, but with what the cut-off level is for getting, and maintaining, qualification (and it would be good if there were some measure of variability--you don't really want to have in charge of an airliner someone whose flying is sublime some days, but a total mare on others.)

[bill fly]

In past years in my airline pilots were qualified after each check as
Above (later changed to High) average, Average, Below average (possibly requires extra training) and Failed (out if happens twice).
The average was supposed to be the average for that particular airline, which was supposed to be pretty good.
However this changed to Qualified and Not qualified (which required remedial action). I think that is what we have to look at - not whether someone has a licence - we all have a licence - but whether one is qualified to the professional standard demanded by the airline and stipulated by the manufacturer.
I for one would not like to be a pax in an airliner piloted by two guys who had a licence but had not passed stringent company requirements.
And by the way this qualification was applied on check rides and sim checks but there were also training (refresher) flights where all hell could be let loose to train combination failures. These were not qualified sessions but if someone screwed up further action (training) was given.

[abdunbar]

The crux of this discussion revolves around the simulator accurately representing the aircraft.

Was adequate data provided for the trim wheel loading? My guess is the data for the trim loading at high speed was NOT provided at low altitude at very high speed...

I suspect that is true of all simulators, whether 737MAX. NG or whatever.

- GY

Worse, I doubt that those test point were ever even flown in the actual aircraft. I suspect that at some point they said that they had tested recoverability at a high enough speed with enough nose down trim and called it a day.

similar to demonstrated cross wind capability that becomes the cross wind limit. It is the limit because they had no incentive to go higher and risk an airframe.

[Dave Therhino]

Worse, I doubt that those test point were ever even flown in the actual aircraft. I suspect that at some point they said that they had tested recoverability at a high enough speed with enough nose down trim and called it a day.

similar to demonstrated cross wind capability that becomes the cross wind limit. It is the limit because they had no incentive to go higher and risk an airframe.



I know it's a thread drift, but just to clarify: The Boeing AFMs state the level of crosswind for which landings were actually demonstrated. I don't believe it is stated as, or intended to be, a hard limit. The level to which they demonstrate it is typically driven by customer contract and sometimes is limited by the conditions they can find during the test program. An exception to this not being a limit exists in the case of a few airplane/engine combinations where there is, for takeoff, an engine crosswind limit due to fan blade stress.

[MechEngr]

MCAS has nothing to do with acting like previous 737s. It is entirely to meet a linearity requirement between control force and aircraft response for pitch. The MAX would have not been approved had it been the only plane from Boeing, but had a non-linear response.
(fixed typo 7377->737)

[ARealTimTuffy]

MCAS has nothing to do with acting like previous 737s. It is entirely to meet a linearity requirement between control force and aircraft response for pitch. The MAX would have not been approved had it been the only plane from Boeing, but had a non-linear response.
(fixed typo 7377->737)

According to the testimony yesterday and today, as well as other information, it is both. Linear gradient stick force and provide pilots consistent feel between NG and MAX models.

[MechEngr]

According to the testimony yesterday and today, as well as other information, it is both. Linear gradient stick force and provide pilots consistent feel between NG and MAX models.

Since the NG was linearization compliant, it is a default that the MAX would handle the same since it met the same requirement. MCAS was intended to not expose a new behavior to pilots to avoid new training, What I know is that it could not be certified to fly without MCAS**, but it could be certified if the feel was different, just as a new type. I'd say the higher control on the ability to sell the plane is that it is basically certifiable. The ability to sell this plane model is predicated on it being certifiable; it is just more marketable if it's not a new type. Since cert >> marketable, and since linearization does both, then linearization is the overriding factor and handling is a fallout.

If the FAA had grandfathered non-linear response and MCAS was created to enforce that non-linear response, then that would be different.

I just don't expect that the stick-force gradient is identical in all corners of the MAX envelope to the NG; but I do expect that it is sufficiently linear to meet the CFR.

**Not MCAS specifically, but Boeing had to deal with the lift-curve behavior somehow so some change was going to be required and MCAS could produce that change.

[Grebe]

And now we have the Hogans Heros Sgt Shultz meme - I see nutting i know nutting ( I'm just the CEO )
From Seattle times re today dog and pony show.... a few extracts

.... Committee members noted that Boeing had an economic motive to downplay MCAS and omit information about it: to cut costs to Boeing’s airline customers, who were demanding no changes from the previous model so that pilots could transition to the MAX without simulator training.

Muilenburg acknowledged that when Southwest Airlines signed a launch order for up to 280 MAXs, Boeing promised the carrier a $1 million rebate per airplane if simulator training were required.

But he insisted that “we don’t make training decisions based on economics.” The panel members were not convinced.

Riiiight but then we get this real face plant

...During the hearing, DeFazio asked Muilenburg about Boeing’s push to move some Lion Air crash lawsuits to Indonesian courts.

When the CEO said he wasn’t aware of those court motions, DeFazio was incredulous.

“Seriously? You don’t know that as a fact?” he asked. “You’re telling me that you don’t know your legal strategy in regards to Indonesia?...

Cheaper to settle- no discovery- great legal defense- goes with the meme of third world pilots being the problem - not quite like buying manhatten with a few beads ...but close

And of course we have

“It’s the purposeful concealment that bothers so many of us, with an obvious financial drive behind it,” said Rep. Colin Allred, D-Texas.

Muilenburg was blasted for taking home $30 million last year, and for providing pay incentives to managers to meet cost and schedule targets.

“You are driving profit,” said Rep. John Garamendi, D-Calif. “You sure as heck aren’t driving for safety.”

Who Me ?? Transition to " Who’s on first, What’s on second, I Don’t Know’s on third. ..."