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

 

[Alistair_Heaton]

I still contend that the "grandfathering" rules and the system design rules are adequate,

The structural side of things I agree with you...

The data flow and human machine interface is some what lacking if not none existent.

 

[Sparweb]

You know, there are enough things going on that there could be some rule changes that come out of this. Even if most rules don't change, and even if the final conclusions show that there was a failure of enforcement, there could still be rule changes in the pipe. Or maybe policy changes or some other high-level reorganization.

From the loads perspective, the total travel in the trim system and the pilot's control are "almost" unchanged since the beginning, possibly not ideal but hard to get yourself into a dangerous amount of trouble. Then the new thing was installed and suddenly you could get in a lot of trouble, fast. The design rules that require system safety analysis and failure mode analysis are intended to reveal these kinds of dangerous situations so that the design can prevent the danger from happening. Concealing that dangerous situation is what makes me say there was an enforcement and integrity failure.

Now that you mention the human-machine interface, I remember a discussion that has since faded. The overwhelming load of warnings and unexpected control forces that the pilots in both cockpits faced led to confusion - becoming an obstacle to both reacting to the situation and accomplishing the emergency procedure. Here is a place where the Joint Task Force report did, I believe, identify a short-coming in the design rules. Cockpits are growing in complexity, and the design rules are not addressing the level of automation that now exists.

http://www.sparweb.ca

 

[waross]

Who remembers that old Nat King Cole song, "Straighten up and fly right"?
While the regulators are trying to justify and certify some sort of bandaid on a bandaid on top of the bandaid called MCAS, I still think that they should be looking at how their regulations allowed the plane to be built in the first place.
They should have tried harder before MCAS not after MCAS.
When I first suggested doing what it would take to move the engines the suggestion was shot down.
Too tough to certify.
Take too long.
Cost too much.
May invoke the need for more training.
Instead, Boeing management and the Boeing culture tried to "manage" a solution.
How much time and money was wasted by Boeing before they realized that honeymoon with the FAA was over and they could no longer game the system?
Consider the loss of both time and money so far and the future losses yet to be incurred.
Remember the start of the cancelled orders.
Can anyone still argue that moving the engines and so fixing the root cause would be too time consuming or expensive?
It's never too late to do the right thing.
Hey Boeing; Straighten up and fly right!
Given the internal memo about designed by clowns, supervised by...

Some of the lyrics are appropriate.
"The buzzard took the monkey for a ride in the air
The monkey thought that everything was on a square
The buzzard tried to throw the monkey off his back
But the monkey grabbed his neck and said "Now listen Jack"
Straighten up and fly right
Straighten up and stay right
Straighten up and fly right"

https://www.youtube.com/watch?v=6fVaP6dM1fs

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Alistair_Heaton]

Spar that's exactly what I am on about.

But with the current grandfather rules there is nothing to enforce adoption of modern tech such as a ECAM system.

The loads on the trim system have changed. The stab is much bigger on the Ng and max and also the trim wheel has got smaller plus has the speed trim system constantly running so the thing never normally stops moving. But I see that as more of an issue with human machine interface and being able to spot there is a problem before it's unrecoverable.

I agree with you that the enforcement and integrity also has issues and needs addressed across the whole industry including training of pilots, ATC, operations maint etc etc.

 

[charliealphabravo]

Instead, Boeing management and the Boeing culture tried to "manage" a solution.

It's more like Boeing management set the engineers to a problem having arbitrary constraints and no solution: Design and ship X number of planes at Y cost on Z date with bigger engines, no pilot training, and no redundant AOA sensors. All that on top of legacy and grandfathering baggage we know little about. When management discovered the problem was unsolvable they used their corporate advantage to push it through the regulators and make the internal engineering objections go away.

I have to believe that there are many successful airframes that are unstable somewhere in the flight envelope without trim tabs or software or pilot intervention. The bigger engines alone did not make the problem unsolvable. If the engineer's had simply been allowed to install redundant AOA systems none of us would ever know anything about MCAS.

 

[Comcokid]

...make the internal engineering objections go away

Usually the objections go away on their own.

You got to know when to hold 'em
Know when to fold 'em
Know when to walk away
Know when to run. - Kenny Rogers

I wonder if there were engineers who sensed issues in the management environment and decided best to be somewhere else.

 

[Sparweb]

The new boss speaks:
New York Times: Boeing’s New C.E.O. Confronts Its Challenges

http://www.sparweb.ca

 

[Sparweb]

This doesn't seem like the right direction to be going:

https://www.bloomberg.com/news/arti...ul-sparks-u-s-canada-rift-over-pilot-guidance

At issue in the article is the suggestion to allow pilots to shut off the stick-shaker.

Since the hazardous situation comes about when erroneous information to several control systems tricks them into mistaking the AoA for an extremely dangerous attitude, triggering the stick-shaker, the solution seems (to me) to fix the inputs and the error-checking in the software, not to let the crew shut off the warning.

This is another point about the reliability and "design assurance level". The odds of a false alarm in a critical system must be extremely low. If anyone in any regulatory agency is still taking about how to just "live with it", then it might be because Boeing is NOT coming up with a solution that guarantees an extremely low false-alarm rate.

Which implies they haven't yet licked the problems of trim excursions or inadequate sensor reliability.

http://www.sparweb.ca

 

[SWComposites]

Just stop it. You cant move the engines or change the landing gear length without redesigning the entire airframe. Boeing should have designed an all new aircraft. But mgmt chose not to so here they are with this mess.

 

[Alistair_Heaton]

Its normal to be able to kill the stick shaker system. In fact I am struggling to think of another aircraft type that doesn't have it. Problem with the 737 is it has no buttons to do it. Q400 we have two buttons one for each side which doubles as attention getters. Press the one flashing and its gone. 737 you have to go find a CB' to pull. I have my doubts you could do it via the CB panel when all hell is letting loose. It would need a couple of buttons, if not 4 infront of the pilots. The 4 buttons is what most other aircraft have. 2 infront of each pilot.

If it is firing incorrectly its a complete pain and a huge distraction. I have had to kill the system maybe 5 times in 17 years. But never on the Q400.

Stick shaker setup in my experience is two completely independent systems one for each side. And when both trigger together it arms the stick push system. If just one system goes you just check your airspeed pitch and power setting and if those all match then you cancel it. In fact its the main method of establishing that there is something wrong with the AoA/stall system. Each side has its own cut off so you even know which side is broken. Once killed you then down grade to only one stick shaker and the stick push is gone. But you still have some redundancy. Stick push is inhibited below 800ft rad alt anyway. 737 doesn't have a stick push system.

There is a bit of history over Stick pushers anyway. The FAA was dead against them for years and they were only fitted to high T tail aircraft which had the issue with super stalls. Now it allows them. Flyby wire don't have them they have software protections called various things such as Alpha floor protection etc. the main reason for this is because the stick pusher shoves the nose down with something like 55 lbs of force and the nose goes way below the horizon and it is utterly impossible to do the old school FAA preferred stall recovery with less than 100ft height loss.

This all revolves around the difference between US thoughts about stall recovery and the rest of the world. But mainly the differences between US and UK thoughts. This was covered in the first thread. IN the US the emphasis on stall recovery was to lose as little height as possible. So what people tended to do was power out the stall and try and hold the nose up as much as possible which then meant 5-10seconds of flying with stick shaker on. Where as UK we were always taught that the only way of recovering was to reduce the angle of attack. Power was a bonus. So for us its pitch/unload, roll wings level then power up in a controlled manner so you don't get a secondary stall. US it was bang the power in and then fight with it to try and stop height loss which usually resulted in a roll coaster ride with multiple secondary triggers of the stick shaker. if you did it just right you would ride the stickshaker trigger with pitch until you were positive climb then let the aircraft accelerate. In the last 15 years after multiple NASA studys and FAA they decided the UK method was actually safer and for the average pilot on a average performance aircraft there was minimal difference in height loss so have gone with that method. But there is still a lot of resistance within the US pilot community of examiners. Usually cantered around what if you stall at 500ft. Which is a load of nonsense because everyone is stabilised at 1000ft these days. So I expect this is the old school FAA procedure advocates fighting to maintain what they think is best. And they are full of shite on the subject and are just fighting a rear guard action that they were teaching nonsense for 50 odd years about stalling and upset recovery.

So most if not all other none Boeing aircraft types on the N reg have it. But the FAA say it would be a bad precedent to allow pilots to be able to kill such a safety critical system. Which they want to allow to be used with a 2 sensor system which is deemed none catastrophic by the simple fact that its not require to be DAL A certified. So is it a safety critical system or isn't it? Sounds to me if they argue to much on the subject they are looking at forcing Boeing to fit a 3rd AoA sensor. I suspect Boeing will tell the FAA to shut up and its easy enough to fit a button and relay to kill the power to the stick shaker motor.


While I agree that critical system design assurance should stop the triggers there is only so much you can do with two sensor voting. Which is why airbus went 3 AoA sensors very quickly. I really can't see how they can get anywhere near DAL A with only 2 sensors. And it will be impossible to get DAL A on the 737 max. And even if they could manage it the amount of changes required would put it firmly into new type rating for all the pilots which will be 2 months of training and two separate training cycles if you want them to fly NG and MAX.

Boeing will be going round in circles now and every solution will just cascade more items down the tree.

The utter silence on the wire loom issue from Boeing and Regulators is the most telling sign for me what the current biggest stop issue is.

 

[Alistair_Heaton]

Just as an example this is my current workplace.

The red circles are the system killers.

Next to the clock we have Sticker shaker and elevator trim kill buttons. Stick shaker I have covered, The trim if it runs for more than 3 seconds continually then a aural alarm sounds and that button lights up. You press it and the trim system dies.

Next along to the right is Gpws test buttons and mute. Which we can use to kill certain EGPWS warnings which occur when we have flap jams and various other none normal configurations.

The 4 horizontal buttons are for killing the spoilers and rudder pcu's and de pressurising those systems.

But all the killer buttons are directly in the field of view you need a fraction of a second to identify and kill and your eyes hardly move away from the primary flight instruments.

The blue are the attention getters to get you to look at the CAP panel one for critical warnings and one for cautions. The black square just above the windows and below the light switches is where all the warnings and cautions come up and we call the CAP . Its normal to operate with this black ie no lights showing. look there and you see what light is showing then look at the panel index in the back of the QRH and it takes you to the right page.

Now the 737 max grandfather cockpit.

It has the caution buttons alerting you to something wrong at the ends of the glare shield but no centralised CAP panel to see what's wrong, you have to search round the cockpit to see which system is showing a light. No 737 version has a CAP panel.

Now the A220 doesn't have a CAP panel either, it has a 5th screen below the flight instrument screens which is the ECAM system which not only brings up the warnings but also brings up the checklist as well.

The 737 MAX really is a geriatric Frankenstein of old tech and new. Items which have been industry standard for 50 years are missing but then its got the latest telly screens. You will notice in the ceiling the analogue pressurisation system circled in red. That contraption has been responsible for more deaths than the max has over the years. Pilots have been wanting rid of it every since the first 737 came out the hanger. Its analogue pressure bellows differential pressure controller. It gets screwed up by water in the system leaks and numerous other things annoy it. Most planes you just set the arrival altitude during the pre start checks. Have a look at it going through 10 000 ft and never think about it. That thing if some one farts in the back it sets up an oscillation in the outflow valve. Just been emailed by a mate that read this and he says that the NG they went digital cabin pressure control. But in true Boeing fashion instead of having one system and a different method for backup it has gone to two systems and they swap over between flights aka the FCU method for the AoA system MCAS 1. Its still an utter pain because each controller is on different electrical buses. So half way through a electrical issue you can suddenly find that you have lost pressurization and the rubber jungle has been dumped in the cabin. But it depends which cabin controller is the live one what happens. But there is no documentation on it so its word of mouth that you have to watch for it during an electrical failure. So its still not great but better than before.


But the main thing is no cap panel which even the Jetstream 31 had which is 45 years old now, Hanley Page who designed the Lancaster bomber put a CAP panel in the heap. Attention getter, look at the panel then straight to the system which has the problem then get the qrh and locate the page using the cap panel index. 737 max attention getter, then complete scan of the panels to find the offending system. Then find the QRH, then find the right page in a 1.5" thick paper book using a list index and no way of killing at least some of the noise and distractions going on.

 

[Alistair_Heaton]

Looks like Boeing is going to get hit with a 19 million fine for fitting uncertified sensors to 800 Ng and Max's. Not alot in the grand scope of things. Only 24k per airframe which I am sure they made more profit on the HUD system than the fine is. And the bonuses will have already been paid. But I suppose it makes the FAA seem like they are doing something. If it wasn't for the max I am sure nothing would have been said or done.

To be honest the FAA should be giving themselves a similar fincial hit.

As it appears Boeing has had absolutely minimal oversight for quiet some time.

And the first of many US investigation reports are out.

Pretty brutal to both Boeing and the FAA.

 

[FacEngrPE]

All of the committee activity on the 737 MAX

https://transportation.house.gov/committee-activity/boeing-737-max-investigation

Press release House Committee on Transportation

https://transportation.house.gov/ne...ee-issues-preliminary-investigative-findings-

Report - The Boeing 737 MAX Aircraft: Costs, Consequences, and Lessons from its Design, Development, and Certification
-Preliminary Investigative Findings- Prepared by the Democratic Staff of the House Committee on Transportation and Infrastructure for Chair Peter A. DeFazio, Subcommittee on Aviation Chair Rick Larsen, and Members of the Committee March 2020

https://files.engineering.com/getfi...gative_Findings_Boeing_737_MAX_March_2020.pdf

Fred

 

[LittleInch]

The bit I don't get from things coming out is that they have now decided that the MCAS activation would only happen "once".

But once what? Every time the AOA sensor goes back into "normal" range and then back again into high AoA? A faulty sensor or wiring break might flip between these?

Once a flight? - But then what if it's needed more than once a flight in correct operation?

How is the overworked FCC going to keep track of all this?

They say they will disable MCAS if the AoA sensors disagree by more than some value. But again, if the aircraft is actually doing a a manoeuvre which needs MCAS to operate how is that going to work?

As ever what seems to be a solution has added complications.

The disable stick shaker article seems to imply pulling circuit breakers - that's not going to work surely. Far too easy to pull the wrong one or for it to become SOP to routinely pull the things.

This saga doesn't seem anywhere near resolution just yet.



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

 

[waross]

Just stop it. You cant move the engines or change the landing gear length without redesigning the entire airframe. Boeing should have designed an all new aircraft. But mgmt chose not to so here they are with this mess.

The first mistake was selecting an engine placement that resulted in unacceptable flight characteristics.
The second mistake was trying to cover up the first mistake.

I understand that the landing gear was redesigned.
I understand that the wings were redesigned.
If the pickle fork is the same as on the NG then it should be redesigned.
The forward engine size and placement significantly increased twisting forces on the wing.

The barrier to doing it right was not technical, it was a marketing decision that there must be no more than I-Pad training.
That barrier is now gone.
Another barrier may have been financial. Don't spend any money.
Well now the largest single shareholder of Boeing is no longer the CEO and the new CEO is blaming the previous CEO for much of the mess.
That barrier may be crumbling.
We are well past the point where it has cost more to cover up the problem than to fix the original problem.

In March 2010, the estimated cost to re-engine the 737 according to Mike Bair, Boeing Commercial Airplanes' vice president of business strategy & marketing, would be $2–3 billion including the CFM engine development. During Boeing's Q2 2011 earnings call, former CFO James Bell said the development cost for the airframe only would be 10–15% of the cost of a new program estimated at $10–12 billion at the time.

So, losses of over $18 billion trying to patch MCAS, when a clean sheet was estimated at $10 B to $12 B.
In hindsight, repositioning the engines, including any redesigns of other components would have been much cheaper.



Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Alistair_Heaton]

In some ways I don't think it was ever about the cost of a clean sheet.

It was purely time and training.

They have succeeded with the time factor because they have the orders and the airbus order book is full for over 7 years so 737 customers are basically screwed and have to stay with Boeing.

The training side of things Air Southwest has a lot to answer for.

They have an extremely specific EFIS setup which nobody else uses.

Its telly screens but analogue instrument representation. Nobody else has it. Its so they used to only do 1 sim session for all types per crew and that covers them for classics and NG. But they got rid of the classics in 2017 so I don't really understand why they want to stick with it. In Europe they used to cycle between the two so one 6 monthly sim session was in the NG and the second in the classic. So they left the EFIS as it was meant to be.

But ASW run over 700 737's with an order for 300 Max's. The 1 million per aircraft or ipad training is no small part of this fiasco. Be interesting to see if they actually enforce it.

An Article in the WSJ seems to indicate a statement about the wiring looms is due form Boeing and the FAA soon. And they will have to move them in the ones already produced and when production starts they will have the new looms. Which is going to penalise the owners of the ones delivered and of those sitting in the Boeing parking lots.

So all the planes produced so far will require.

1. Mcas 2 uploaded.
2. Rewired.
3. FOD inspection to remove all the crap in the fuel tanks etc.
4. All the cowls checked for lightning conductivity check thanks to them hand tooling them to get them to fit.
5. Removal from storage maint.
6. Removal from storage inspection.
7. CoA physical inspection.
8. CoA paper work audit.
9. CoA test flight.
10. There will be snags on all of them after the test flight which will need fixed.
11. Another test flight.
12 CoA issued
13 Customer inspection and test flight.

There may be other things have to happen as well but they haven't released those problems. And the Ethiopian crash report may also highlight other issues.

But realistically I can't see the max being flown for revenue in 2020.

The pulling of CB's to isolate systems unfortunately is quiet common, and your right it is a pain in the bum. The cb panels are set up in a grid and when you get to a point in the checklist it will say LH upper DC essential panel pull CB D8. And you find it and pull it. Some company's put collars on the CB's that are mentioned in the QRH so they are easier to find and easier to pull out. Its not a problem in the q400 because they are all easily reachable and you can see them from your seats. And there is only 3 from memory... and only one major one to be pulled in flight when your doing an emergency landing which very rarely happens, only when you have a gear unsecure normally. Bigger cockpits such as 737 and you have to unstrap to be able to reach a lot of them. During flight your not meant to reset CB's so the only requirement is to be able to see them to see which one is popped.

 

[1503-44]

I have my solution. You won't see me flying in one of them, certified or not.

 

[LittleInch]

So is that 5 screen version a 737 max but rigged for South West Airlines only??

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

 

[Alistair_Heaton]

That's a NG in the picture butchered for southwest.

And it's official now Boeing is going to have to rewire all the Max's that have been produced. They had been playing down the extent of the issue. There 7-8 bundles in the avionics bay and 5-6 in the tail. And two fusalage length bundle's. So that's not a small job to sort out.