Alaska Airlines flight forced to make an emergency landing (Part II)... 26

[AZPete]

Question hasn't been answered: is the grounding warranted?

 

[3DDave]

Is what grounding warranted? As far as I know no planes are grounded.

 

[Alistair_Heaton]

I presume they meant the type certification being withdrawn between March 2019 and December 2020. Resulting in all 737 MAX aircraft being grounded for 20 months which is the recorded for a type being grounded by its original certifying authority.

And BTW I think they could have left it flying in the USA. Although what the insurance situation would have been I have no clue. But when the FAA found out how much they had been lied to they had zero chance.

Plus its an international market. It only being legal in one country makes it worth less. Plus your pax are more media led.

 

[LittleInch]

Well with approx 30,000 commercial aircraft in active service, there must be something like 150,000 or more active pilots (google say > 300,000) in the world.

To get that many and to replace an aging pilot population figure show the industry needs something like 25,000 a year. Quite a lot of the increase will be in Asia and Africa.

The reliance therefore on automation and probable lack of direct flying experience is completely different to what existing in the 1960's when the 737 started flying. The 737 MAX design was, IMHO, fatally flawed with this MCAS addition and introduced an unacceptable level of jeopardy and over reliance on "memory" functions and flying ability which could not be assumed with todays pilot population.

So did the pilots of ET-302 get it badly wrong - evidently yes.
The odds though were being stacked against them by a deficient design which had not properly considered all the possible outcomes from a single sensor failure and a barely believable assumption of the pilots doing something within 3 seconds of being hit by multiple alarms, airplane movement and with no central alarm panel.

I go back to the issue over the silence about the SIM sessions with a variety of pilots not being made public. If 99% of times those pilots managed to control the aircraft and get it back to base then I'm pretty sure we would have heard about it by now...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Alistair_Heaton]

normal crewing is at least 5 crews per aircraft. With a crew needing 2 pilots.

Which is where the > than 300 000 comes from.

 

[waross]

Was Boeing to fault?
Boeing made the decision to lie in order to avoid proper training of pilots.
Boeing made the decision to implement MCAS.
Boeing did not make Max sims available, and this is related to their decision to avoid sim training.
The FAA was complicit with Boeing.
The airlines shared Boeing's cost priorities and were complicit with Boeing.
Were the pilots to blame?
Did they wake up that morning and decide;
"I'm going to die today and take a lot of people with me."
It is pretty obvious that the pilots were not properly trained, and the training that they did receive was inadequate.
Did the pilots make mistakes?
Definitely they made mistakes.
They probably should have been sent for retraining, had not death intervened.
Their mistakes demonstrate their inadequate training, and the fault of that goes back to Boeing, the FAA and the Airline.
The pilots could have done better, but faced with imminent death, they did the best that they could.

But for pilot inaction or incorrect action the planes still allowed full control and both crashes were entirely avoidable and needless. The FAA and all airlines said the Emergency AD was sufficient.

The FAA and all airlines said the Emergency AD was sufficient.
The main responsibility falls on Boeing.
The FAA and the Airlines may be viewed as contributors or enablers, but their responsibility is based Boeing's original actions and Boeing's withholding of information.
Were the pilots to blame or where the pilots victims of corporate greed and inadequate training.

If this is true, Dave:

The FAA and all airlines said the Emergency AD was sufficient.

THEN WHY THIS?

I presume they meant the type certification being withdrawn between March 2019 and December 2020. Resulting in all 737 MAX aircraft being grounded for 20 months which is the recorded for a type being grounded by its original certifying authority.

The grounding seems like an extreme step for what you characterize as "Pilot Error".
Contrast the AD with the grounding of a type for 20 months and I will suggest that an inverse ratio may be appropriate to assess the pilots responsibility.
That is, the pilots share some responsibility, but in comparison, very little.






--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[3DDave]

"Boeing made the decision to lie in order to avoid proper training of pilots."

There's a couple of problems with that statement.

No one in Boeing, so far as evidence has come up, ever identified the case where the AoA sensor would feed bad values into the system that would result in an uncontrollable condition; the symptom was the 30+ year old stabilizer runaway. Nothing about that aspect of training changed. There's no separate training for "Stuck trim switch" and "rat ate a wire and shorted a signal to the trim motor control" and "the computer RAM has a transistor that was dodgy." They don't have separate training for that because it is the symptom that needs to be dealt with, not the cause.

MCAS itself was not a feature the pilots would ever be aware of in normal flight; even the conditions for which it was intended lie far outside the envelope a passenger jet would be used in revenue service. They didn't train for "Speed trim goes wrong" either, even though the first crew speculated to the maintainers that it was speed trim going wrong.

The 20 months was for two reasons.

The first is that everyone was told that a bullet-proof procedure totally failed, and failed in an even higher-energy way than the crash the procedure was based on. This no doubt led to a huge investigation as to how it failed. Of course that investigation would not find anything because the procedure was not used. Turns out, one can look for a long time to find a needle that isn't in a haystack.

The second is that they eventually found the procedure was not only not used, the pilots subverted the procedure at every possible opportunity, meaning that they had to design a system resistant to sabotage. That seems like something that is more difficult to manage a solution for.

"Did they wake up that morning and decide..."
not to review the FCOM or the Emergency AD. Yes, they did decide that. They also decided not to read and walk through the Lion Air crash preliminary report. They also decided not to speak to each other about this possible situation at the gate and which would handle each part. They also decided not to talk to their Chief Pilot. There was a long list of things the pilots decided every morning for the 4 to 5 months from the time all that was issued and the day they welcomed people aboard. None of what they decided appear to be preparing for what they knew could happen.

Or maybe they did skim it, though I cannot see how. Pretty simple - the report laid it out. If there is a stall warning when the plane is not in stall conditions and it's a one-side warning, so probably the AoA subsystem. You are supposed to turn off the autothrottle and set pitch and power and the autopilot goes off because it cannot operate on faulty data. Retracting the flaps fully is the last gate to start MCAS. The Lion Air team made this 100% clear. Is this sequence impossibly difficult to recall?

Between 120 and 150 mornings of deciding they did not need to have any concern about stabilizer runaway following the fatal crash of the same type aircraft, the one they were informed was going to require a software update before the Emergency AD was lifted. They knew the plane had a defect, knew from the Lion Air report it was manageable, but only if they did the stall warning procedure as required for the last 30+ years. Which they decided not to do.

"...Boeing's withholding of information."
As of the Lion Air preliminary report there was no information withheld. The ET-302 crew had far more information available to them than the successful Lion Air crew, and that crew even managed to give it a controlled experiment, similar to ET-302, but the Lion Air crew was prepared to cut it off the instant it misbehaved; and they did.

"Boeing did not make Max sims available"
Ethiopian Air owned at least one Max simulator at the time of the crash.

https://corporate.ethiopianairlines...thiopian-airlines-is-the-new-spirit-of-africa

 

[dik]

You missed 'the government', a big part of the problem.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Alistair_Heaton]

Its hardly a bullet proof procedure, it relies on human reaction times and diagnosis.

That went out the window in the 80's because its well known it doesn't work most of the time.

Its more likely that some one that had a clue about human performance got a vote.

I also think the FAA director going into the sim with it had a huge effect. Along with the other generic pilots.

I checked with some one at Ryan Air and it was a Boeing gag order about what went on with those pilots. Which maybe the reason why Boeing stopped fighting the grounding.

Anyway it was grounded and there are a whole raft of mods in the process before the 10 gets certified. The 8's and 9's also have loads of mods to go through once the 10 is finalised.

Its looking like a new type would have been cheaper to develop and take less time and they would have been setup up for the next 30 years.

For some every accident is the pilots fault, to be honest most of the time they are right. These two its one of the few that the system is stacked against them.

 

[1503-44]

Gov: Remove two old regulations for every new one.

One step forward, two steps back.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[Alistair_Heaton]

the problem they have is its not USA regulations its international.

 

[LittleInch]

"No one in Boeing, so far as evidence has come up, ever identified the case where the AoA sensor would feed bad values into the system that would result in an uncontrollable condition;"

I really don't think that is correct though. They identified this, but decided, based on the ORIGINAL MCAS operation that the likelihood of AoA failure was low but so was the action - initially it was a high speed low angle movement of the trim system. They also decided, like you have done, that existing "memory" procedures would kick in after three seconds and decided this warranted a hazard level below catastrophic. Then the operation of MCAS changed to be a a much higher movement assuming a near stall condition, but AFAIK, they didn't change the evaluation of the fault operation. It is clear that there was an overriding desire to avoid having to have pilots do type certification training and this coloured any discussions.

The other issue being that the original 737 design allowed the elevator to overcome even full trim down operation (I think), but as the plane got larger and larger, this slowly reversed.

Operation of the trim button on the control column for any length of time at higher speeds results in quite violent movement of the aircraft. It's not surprising that the pilots stopped doing this.

The faulty AoA generated a significant number of alarms and issues which helps to mask the trim issue. The biggest issue I see is that when it went off, previously the plane should have been relatively stable, although A/P would have been kicked out. With MCAS, after a small pause, the plane suddenly nose dives without warning. Not surprisingly the first action is to pull on the control column and raise the nose. After that trying to figure out what is happening is far from straightforward. Any pause in operating the electric trim for > 5 seconds and it repeats.

You go on about the original Lion Air flight, but conveniently forget that they had a crucial extra, very experienced pair of eyes who could study all the inputs without having the requirement to fly the plane who actually figured it out. In essence he was the missing central ECAM.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[waross]

It is clear that there was an overriding desire to avoid having to have pilots do type certification training and this coloured any discussions.

The AD as training was lacking in the checks on comprehension and retention.
AFAIK there were no comprehension nor retention check components to the AD and the pilots had not been trained in a MCAS sim.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[zeusfaber]

They identified this, but decided, based on the ORIGINAL MCAS operation that the likelihood of AoA failure was low but so was the action - initially it was a high speed low angle movement of the trim system. They also decided, like you have done, that existing "memory" procedures would kick in after three seconds and decided this warranted a hazard level below catastrophic. Then the operation of MCAS changed to be a a much higher movement assuming a near stall condition, but AFAIK, they didn't change the evaluation of the fault operation

Can't comment on whether or not that is true but, if it is, it serves as a stark reminder that failure modes, failure tolerance and a myriad of other things (like EMC) can all be different in a system that is ostensibly a form, fit and function replacement for its predecessor. If I had a pound in my pocket for every time I've seen that neglected, I'd struggle to walk in a straight line.

A.

 

[Alistair_Heaton]

zeusfaber the wiki has a good timeline of what went on.

https://en.wikipedia.org/wiki/Maneuvering_Characteristics_Augmentation_System

 

[3DDave]

waross, I gave you a link to Ethiopian that claimed they had trained on this procedure in the MAX sim they owned.

Any example verbiage of "checks on comprehension and retention" beyond what every FCOM requires? I'm stunned to hear that pilots are expected to behave as a 13 year old might.

---

"Operation of the trim button on the control column for any length of time at higher speeds results in quite violent movement of the aircraft. It's not surprising that the pilots stopped doing this."

The operation of the trim button would have allowed them to relax the load on the control wheel at the same time. They were maintaining trim by the elevators. It would not have resulted in a violent movement because the trim load on the plane would remain unchanged.

"The faulty AoA generated a significant number of alarms and issues which helps to mask the trim issue."

The trim issue did not become a trim issue until after they failed to deal with the stall warning alarms. The captain did respond to the first MCAS trim in a half-measure way.

"You go on about the original Lion Air flight, but conveniently forget that they had a crucial extra, very experienced pair of eyes who could study all the inputs without having the requirement to fly the plane who actually figured it out. In essence he was the missing central ECAM."

Yes - and learning of this was what was written into the Emergency AD and the FCOM and the Lion Air Preliminary report. You don't need a third person if you understand what you are looking for. All they recalled was "turn off the trim motors." Pretty poor example of what they spent their 120 mornings not doing.

 

[1503-44]

Just think of happy thoughts and you’ll fly. -- Peter Pan

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[Alistair_Heaton]

The risk of pilots not doing something should be in the risk analysis.

As there were two fatal accidents over the same issue the analysis used was proved wrong.

I don't think there was a proper risk analysis for the eventual mcas system.

 

[1503-44]

Those could be some interesting documents.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[Alistair_Heaton]

There was one for the original mcas but they didn't do one for the uprated system.

Also reading the wiki I hadn't realised that the screw jack was slipping due to the control force in both accidents.

"In February 2016, the EASA certified the MAX with the expectation that pilot procedures and training would clearly explain unusual situations in which the seldom used manual trim wheel would be required to trim the plane, i.e. adjust the angle of the nose; however, the original flight manual did not mention those situations.[82] The EASA certification document referred to simulations whereby the electric thumb switches were ineffective to properly trim the MAX under certain conditions. The EASA document said that after flight testing, because the thumb switches could not always control trim on their own, the FAA was concerned by whether the 737 MAX system complied with regulations.[83] The American Airlines flight manual contains a similar notice regarding the thumb switches but does not specify conditions where the manual wheel may be needed.[83]

Boeing's CEO Muilenburg, when asked about the non-disclosure of MCAS, cited the "runaway stabilizer trim" procedure as part of the training manual. He added that Boeing's bulletin pointed to that existing flight procedure. Boeing views the "runaway stabilizer trim" checklist as a memory item for pilots. Mike Sinnett, vice president and general manager for the Boeing New Mid-Market Airplane (NMA) since July 2019, repeatedly described the procedure as a "memory item".[84] However, some airlines view it as an item for the quick reference card.[85] The FAA issued a recommendation about memory items in an Advisory Circular, Standard Operating Procedures and Pilot Monitoring Duties for Flight Deck Crewmembers: "Memory items should be avoided whenever possible. If the procedure must include memory items, they should be clearly identified, emphasized in training, less than three items, and should not contain conditional decision steps."[86]

In November 2018, Boeing told airlines that MCAS could not be overcome by pulling back on the control column to stop a runaway trim as on previous generation 737s.[87] Nevertheless, confusion continued: the safety committee of a major U.S. airline misled its pilots by telling that the MCAS could be overcome by "applying opposite control-column input to activate the column cutout switches".[88] Former pilot and CBS aviation & safety expert Chesley Sullenberger testified, "The logic was that if MCAS activated, it had to be because it was needed, and pulling back on the control wheel shouldn't stop it."[89] In October, Sullenberger wrote, "These emergencies did not present as a classic runaway stabilizer problem, but initially as ambiguous unreliable airspeed and altitude situations, masking MCAS."[90]

In a legal complaint against Boeing, the Southwest Airlines Pilot Association states:[91]

An MCAS failure is not like a runaway stabilizer. A runaway stabilizer has continuous un-commanded movement of the tail, whereas MCAS is not continuous and pilots (theoretically) can counter the nose-down movement, after which MCAS would move the aircraft tail down again. Moreover, unlike runaway stabilizer, MCAS disables the control column response that 737 pilots have grown accustomed to and relied upon in earlier generations of 737 aircraft."

So it was already known it wasn't like a trim runaway in 2016 and the thumb trimming was ineffective