Boeing 737 Max8 Aircraft Crashes and Investigations [Part 7] 16

[Sparweb]

This post is the continuation from this series of previous threads:

thread815-445840
thread815-450258
thread815-452000
thread815-454283
thread815-457125
thread815-461989

This topic is broken into multiple threads due to the length to be scrolled, and images to load, creating long load times for some users and devices.
If you are NEW to this discussion, please read the above threads prior to posting, to avoid rehashing old discussions.

Thank you everyone for your interest! I have learned a lot from the discussion, too.

Some key references:
Ethiopian CAA preliminary report (Link is now broken. See PDF download below, 3 MB)

Indonesian National Transportation Safety Committee preliminary report

A Boeing 737 Technical Site

Washington Post: When Will Boeing 737 Max Fly Again and More Questions

BBC: Boeing to temporarily halt 737 Max production in January

www.sparweb.ca

 

[3DDave]

So - it's not unsafe? How did they determine that an engine should be shut down? Was it the Atlas Air pilot?

It's still 7th hand. Telling me exactly how they came to use the wrong checklist when they were there for only one reason is where the truth is. Running the wrong checklist is about 99th on my list as long as the pilots keep the plane out of the dirt. And you just said they all did. So now it's not a plane problem. It's a pilots problem.

I don't want a pilot to diagnose anything. That's what has crashed any number planes when crews get to figuring out what's wrong.

A checklist is the opposite of diagnosing things. It's an "if this happens do this thing," to avoid the pilots having to think. Like "If airspeed is greater than VMo, reduce thrust". But I don't know about "If the wheel is pulling too hard and you aren't doing a loop use the trim button, the one under your left thumb, right where it has been for 500 to 20,000 hours. Press until the load is GONE." Is there a manual entry for exactly how far to pull the wheel back at rotation speed?

I want the plane to be operated safely. I know there was greater than zero chance as one crew, without any hints, flew safely, another half crew flew safely until the new guy took the wheel, and a third crew - by the checklist they half-recalled - piled it into the ground. From Atlas Air it's clear the FAA is no longer concerned with incapable pilots getting in the seats. And from PIA 8303 it's clear Airbus has no answer for that problem either, triple redundancy or not.

Maybe it's time for cockpits to have three pilots with duplicate controls, all making inputs, and then a computer cuts out the one that disagrees with the other two, maybe gives a high-voltage shock to let them know to let go.

 

[Alistair_Heaton]

Checklists will direct you towards various procedures. If the checklist you have taken is not correct you may end up doing the wrong thing.

The classic on the Q400 is a bus failure if you don't spot it within 30 seconds then generators start going off line and pretty much half the board lights up and screens drop out and you have a choice of 8 checklists depending which red light takes your fancy. The bus fail is actually a single amber light and if you go for that checklist one switch is set and everything starts working again. I think 4 out of the 8 checklists will end up with a engine shutdown because you loose oil pressure readings and various other things. Its a well known feature which type rated pilots should know about. But even if they miss it and shut an engine down nobody dies.

Its one of the joys of flying a stretched grandfathered machine where they had to change things to keep differences training instead of full type rating. Which in the Q400 is 2 sim sessions and half a day of ground school on performance if you have flown one of the other flavours.

It obviously unsafe to have a single component failure result in a fatal crash what ever the process is after that. The threat should have been trapped by design. That premise has been around since the 737 was first designed in the 60's. There are somethings you can't have two or more of encase of failure such as wings and hull so you build them to not fail. You have others that their failure will not kill you so you don't need to double up on.

You would need 3 computers and at least of 3 of every sensor supplying data to them. One just doesn't cut it. 2 is marginal.

 

[1503-44]

Checklists are not concerned with diagnosis. They have no if, then, else paths. They are the last item on a diagnosis flowchart.

Simplifying a 50-60% success ratio of the man-machine interface is indicative of results obtained from blindly choosing between two alternatives. It needs to be something over 99.75% Any monkey can pull off 50/50. Considering that the pilots are supposedly the highly trained half of a 50/50 combined man-machine system, they typically get acceptably high scores, yet the mission failed, means that the machine isn't contributing to success at all.

A machine with a 50/50 success ratio needs 8 redundancies to get 99.5% chance of success, but then as they say, "A man with 2 watches never knows the time". Is putting your faith in 8 watches the answer we're really looking for? 8 machines at 50% correct voting yes or no still results in a tie. Guess we really only need 7.



“What I told you was true ... from a certain point of view.” - Obi-Wan Kenobi, "Return of the Jedi"

 

[davefitz]

"It obviously unsafe to have a single component failure result in a fatal crash what ever the process is after that. " Yet here we are, with a 737 Max with a single sensore being used to feed the MCAS. It is obvious, yet we persist in believing the emperor is wearing clothes.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[rb1957]

Flight says NPRM coming out for the AD to lift the grounding.

Pt 8 on it's way ?

IMHO if we want to improve safety the best thing is to stop grandfathering. Types have a production life of 20 years, and need to be updated to continue production. That'll stop a 1960s design being "lipstick up" in the 2010s.

another day in paradise, or is paradise one day closer ?

 

[3DDave]

Nothing about the 737 crashes was due to grandfathering. But I bet makers would like to see all 20 year old air frames taken out of service.

It occurs to me that the FAA is covering for a huge foul-up. All software will be under both version control and be stored in a software vault, the sort of vault that can generate messages when the software is updated in the form of e-mail notifications. Why did the FAA reps fail to be notified by automatic methods? Or were they and they ignored the changes to MCAS and are seeking to hide that they were in fact informed? It would be what a quality conscious organization that is delegating would do to ensure that there was a correct communications of changes.

 

[3DDave]

Three pilots, all with hands on duplicate controls would have prevented the PIA crash and the Atlas Air crash. Multiple computers are obviously insufficient.

Alistair, you said the 737 pilots shut down engines in response to a stick shaker and stall warning. I don't care about the Q400 and electrical bus failure. Why did the 737 pilots do that?

 

[Alistair_Heaton]

i have zero clue about the 737 NG systems or checklists similar to main line Airbus systems.. I don't know what system failures require decision points. I do know though that it doesn't have a central CAP which means you have to scout around the cockpit and see what lights are showing and then find the checklist and decide which checklist to run once you have located the book. To be honest single engine is not that extreme compared to having to use standby instruments to fly.

They have been extremely closed mouthed about what was actually thrown at the pilots in those sessions. All we know publicly is that pretty much everyone gave up about the Ipad training differences training and also a lot of the major changes to the avionics after that set of tests. They went dual flight computers quiet soon after that.

Per say grandfathering still has a place but it should have a limit before a large picture look at the complete device to see how all the parts fit together. There should also be a limit when certain technology's can not be omitted due to being limited because of triggering none grandfather testing and work due money and marketing.

Now there is two AOA in the mix with the dual primary FCC in the mix, if there is a disagreement between the two then MCAS won't fire. I presume there is a whole raft of new checklists to deal with the computers not agreeing. This is only a temporary fix to get it flying again. Within a certain time frame they will have to swap at least two of the old school pitot tubes for smart probes which also measure AoA. How many of them they will have to swap I have no clue but I would think at least 2, which will bring it up to 4 AoA sensors but it could be all of them to get rid of the piping.

But again we are just talking about MCAS, it is the primary cause of the grounding and the cause of all the deaths, But in reality it is just one item of many which meant the plane should never have been certified in the first place. The flight tests are only for ungrounding in USA airspace. The training still hasn't been finalised. The other world wide regulators also have not had there say. I know Canada has a few issues which will need resolving around the stick shaker which I can't see the other regulators disagreeing with. But in reality the MAX may look the same on the outside but is completely different beast on the inside with major changes having occurred.

 

[Tomfh]

How long until they bite the bullet and scrap this line?

 

[3DDave]

Nevermind. I see no one cares how the FAA should have dealt with MCAS on a limited number of FAA inspectors.
Weird I thought I had a post about the FAA process. Has it gone missing for some reason?

 

[Alistair_Heaton]

Its a given that the whole Regulatory process was completely flawed producing the MAX. I also agree with you that the world wide standards need addressing as well both maint and pilot standards. Quite how you do that from experience working round the world with the different cultures involved I have no clue. The only way i can see is by limiting the sales market and just not allowing them to operate in certain regions of the world. You have to design for the min standard that will fly it or prevent them flying it.

There was also an element of keeping information from regulators and not doing what the regulations require. That's an internal issue for the USA to sort out.

The line won't be scrapped. They will have to provide support for it now until the end of life cycle even if they don't want to and it will cost money to do so. I would love to be wrong and after this is finished with it turns out to be a successful type.

 

[1503-44]

Imposing sanctions on countries that won't accept the results might cross someone's mind sooner or later. They think it works for pipelines.

https://foreignpolicy.com/2020/07/20/nord-stream-2-pompeo-sanctions-russia-putin-geopoliticsa/

“What I told you was true ... from a certain point of view.” - Obi-Wan Kenobi, "Return of the Jedi"

 

[Alistair_Heaton]

I am sure it was on the table with the old Boeing exec and FAA bosses.

After the second crash then even sanctions couldn't prevent it. Then they had a choice let outside countries control the recertification or retain it internally.

Remember the new 777 is due soon and its market is extremely limited domestically.

Having two product lines domestic only would be killer. Plus it would end up two type designs per product which is back to the old days of the 80's

 

[3DDave]

"keeping information from regulators"

That's just it. The regulators had to willfully avoid getting that information, much the same as the Ethiopian CAA avoided doing their job.

 

[Alistair_Heaton]

Don't have a clue about that side of things, but doesn't change the point that it should never have been certified in the first place.

FAA wasn't doing there job and neither was the Ethiopian. And Boeing certainly wasn't.

 

[Alistair_Heaton]

This came out Yesterday. Expect quite a few of these with all the aircraft groundings recently. No per say the MAX but all of them will have similar issues after being on the ground so long. Expect a heap of gear and flap issues as well. TO note all types will have issues not just the ones mentioned. It will just take time for the issues to surface.

https://rgl.faa.gov/Regulatory_and_...2585ae007cd448/$FILE/2020-16-51_Emergency.pdf

 

[waross]

As an interested observer, I and I suspect others, am wondering what is the function of this check valve that makes it aircraft type specific rather than engine specific.
Thanks.

[link ]Bill[/url]
--------------------
"Why not the best?"
Jimmy Carter

 

[Alistair_Heaton]

The bleed air system is aircraft not engine. The valves on the compressors stages will be engine. But like the hydraulics after the output of the engine driven pump its then aircraft.

The bleed air system basically uses compressed air from the compressor. depending on what power setting the engine depends on which compressor it takes it air from. generally low power settings and it will take from the HP compressor but high power then the low pressure compressor will have enough. The check valve stops air from the HP going through the system back into to Low pressure compressor when the HP bleed is open. If it manages to it upsets the aerodynamics of the LP compressor and you start getting engines stalls or only surges which require a shut down.

It sounds like the check valves haven't changed since the aircraft came out and are a common part. If they work there is no reason to change them in the design. Normally that area is warm so they won't have condensation forming in the pipes. Now they will have sat for months with water in them on the valve.

But per say this isn't Just a Boeing issue all OEM's will be having issues with seals going and valves sticking.

There is a lot of power taken from these bleed systems. They run the hot wing icing system and also power the pressurisation and air conditioning for the whole aircraft. The air coming out of them is hot 500 deg C plus. And is you stick hot compressed air into the LP compressor it will stall it, which will decrease airflow into the combustion chamber which will then spike the exhaust gas temp and melt the rear turbines. If you kill the turbines you have nothing to drive the compressor drive shafts so you stop compressing and very little power can be produced from the engine and usually it flames out.

 

[LittleInch]

The language they use is interesting.

What is a "forced off airport landing"?

Catastrophic crash in normal English?

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

 

[1503-44]

Yeah.
Ridiculous wording!
Obviously has difficulty dealing with truth.
Where do they get these ... these ... Professional Obfuscaters?

“What I told you was true ... from a certain point of view.” - Obi-Wan Kenobi, "Return of the Jedi"