Boeing 737 Max8 Aircraft Crashes and Investigations [Part 4] 28

[Sparweb]

This is the continuation from:

thread815-445840
thread815-450258
thread815-452000

This topic is broken into multiple threads due to the long length to be scrolled, and many 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.

My personal point of view, since this falls close to (but not exactly within) my discipline, is the same as that expressed by many other aviation authorities: that there were flaws in an on-board system that should have been caught. We can describe the process that "should have happened" in great detail, but the reason the flaws were allowed to persist is unknown. They are probably too complex to reveal by pure reasoning from our position outside of the agencies involved. Rather, an investigation of the process that led to the error inside these agencies will bring new facts to light, and that process is under way, which will make its results public in due time. It may even reveal flaws in the design process that "should have" produced a reliable system. Every failure is an opportunity to learn - which is the mandate of the agencies that examine these accidents.

Some key references:

Ethiopian CAA preliminary report

Indonesian National Transportation Safety Committee preliminary report

The Boeing 737 Technical Site


No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

 

[Alistair_Heaton]

Its meant to be impossible to have any auto pilot function in while the stall protection system is active. Ie the AP kicks out as the stick shaker triggers. in fact there is a long list of things that if they annoy it,it kicks out. The FBW machines are different and they come under a different set of rules and have funny modes which are type specific. Some would say the only way the AP is truely out on FBW is when you in the very bottom emergency control mode.

There can't be a pause or some functions working and others not the whole lot has to come out on none FBW.

The stick shaker indicates incipient stall situation so the plane isn't actually stalled and no pilot input is required to recover if the there is nothing increasing the angle of attack. Due to the certified stability of the aircraft the nose will hold position or drop decreasing the angle of attack. You don't need power to recover from a incipient stall. You only need to maintain or decrease the angle of attack. In real life the nose drops anyway as it disconnects. The only time this falls over is if the trim has run away and is way back driving the nose up. But that's a none normal mode.

The FDR records every 2 seconds. For it to have multiple points with the stick shaker going and AP in means that its not a freak pickup of both being in at the same time due to signal lag.

I forget the chart it was a comment on it was 2 threads back I suspect.

Anyway the email was only the instigator it will have been checked with the relevant raw FDR data and CVR recordings which aren't public. If the stickshaker is heard before the autopilot disconnect alarm then its a fail for certification on the CVR and they will be able to see it on the FDR. It would have taken under 5 mins to check if it was a valid concern. And its turned out to be a major valid issue.

Your not wrong it should be either in or out, and impossible to put in if a long list of requirements are not met. There are 3 modes starting with pitch hold and wings level through to high end function when the FMC is running the show. The FMC has protections in it. The other two levels don't.

They won't be able to tell if it was partially in from the FDR. They will have done system tests and looked at the relevant raw data signals to see what is going on. FDR recorded AP in and Stickshaker on. After the tests it appears it was stickshaker on and AP sort of on.

The risk is that the plane is climbing in the second level mode of vertical speed and as it gets higher altitude the engine power drops with density decrease, density change also requires the angle of attack to be increased to maintain rate of climb. Now this is where there used to be conflicting opinions across the pond. The European regulators required a secondary system of a stick push to force the nose down if you went any further than incipient stall. The FAA never had this requirement and was very against stick pushers and made people remove them before operating on the US register. With FAR 25/JAR 25 this was then dropped but no requirement to have one fitted. Conversely the Europeans dropped the requirement to have one. The ins and outs of the arguments for and against is colossal and centres round relative perceived risks of system failure V it actually helping save a plane. But basically it adds weight to the aircraft which the OEM's don't like because it decreases the traffic load and range. FBW doesn't have one because its meant to have performance fences as protection. And legacy product lines such as Q400 have them because the plane was certified with them with the 100 series and to keep grandfather certification they just kept it.

To add there are certain tail configurations that do require a stick push world wide. High T tails when the pitch control surfaces can be blanked by vortex coming off the wings in a stall do require them. You can get into a thing called superstall when you can't recover even with full power on.


As for pilot training they are all issuing guidance and have been for years.

https://www.faa.gov/other_visit/avi...afety/safo/all_safos/media/2017/SAFO17007.pdf

But due to the way the regulators have devolved the standards compliance world wide there is very little chance that it will have the effect you and I want. If the paperwork is compliant then they are happy. MCAS on paper was compliant so they signed it off. What's actually going in the sims and the level of competence which is deemed acceptable they have virtually no clue about until an accident happens. Old days it was forced active prevention, now its reactive statements and finger pointing with a political input.

The pilots are the product of the system. You need to fix the system. The oem's, airlines and regulators don't want to change the system because that eats into profits and creates more work and expense for the regulator. Also the regulators are risk adverse. They have moved all the risk to the operator and oem's so if something ever goes wrong they can never be pinned for it and held accountable at a legal level.

Pax just want cheap tickets, they will not pay for anything which is over and above the legally required minimum standard. The minimum standard is set by the quality of paperwork thats audited by the regulator which these days is done by people that have never landed an aircraft of any size or changed a brake pack. Old school you had flight ops inspectors who were current on multiple types and Engineering assessors who turned up with dirt under their nails. Very poacher turned gamekeeper and knew and had used personally all the tricks in the book.

Its not saving a few bucks its saving millions if not billions of bucks world wide having the current setup.

 

[waross]

Pilot training and simulator training.
This may be another can of worms.
Up until and after the second accident, Boeing had the only valid Max 8 flight simulator in the world.
The information provided to the simulator manufacturers by Boeing was faulty.
No argument. Boeing has admitted this.
After the first Mac Attack, Boeing issued a procedure to manually trim the aircraft.
Unfortunately the procedure didn't work.
I understand that the procedure did work in Non-Boeing flight simulators.
Dare we hope that the regulators are subjecting simulator performance to the same level of scrutiny as the aircraft?


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

 

[Alistair_Heaton]

They will all have to be certified before any official training can be performed on them .

There is a order backlog of over 250 Sims but until they sort out the trim system they don't know what to put into them.

 

[waross]

They will all have to be certified before any official training can be performed on them .

Will the regulators be looking into this issue or will Boeing be allowed to self certify them again?
The question is not will they have to be certified as much as;
"Who will certify the sims?"

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

 

[Alistair_Heaton]

Boeing can't certify Sims. It only provides the data pack.

I don't know much about the process. But it can take 3-6 months to get the first level approval. Then there are other levels for unusual attitudes, low vis training and zero flight time training. They follow on after it's been running for a period in can take 12 months from the boxes turning up until it has all the ticks in the boxs.

It used to be the OEM of the SIM got the base approval for the flight dynamics from the local regulator. But that's out the window now that the FAA is in disgrace with the other regulators.

 

[Alistair_Heaton]

Seems the neo has issues as well.

https://www.flightglobal.com/news/a...tors-alerted-over-excessive-pitch-ano-459718/

 

[RVAmeche]

Also, United has started buying used planes to make up for the grounding. Add it to Boeing's bill.

Link

 

[3DDave]

Alistair, Good - I was sure I had nothing to do with it. Not sure why you'd drag me into that.

Time to update the FCC to do a pop quiz from the memory item list for planes before allowing the pilots to use them. The time limit is 80% of the time the emergency is to be dealt with in real time, so there's no hesitation.

Would a "D" be an OK grade before being allowed to operate the aircraft or would a 100% be the requirement? And should the pilots get a second chance, unlike their passengers will if they fail? I'd say a 24 hour lockout on the plane would be fair. It would give the airline an incentive to ensure the pilots are trained to operate the plane. It might also weed out the pilots who panic under pressure and won't follow the procedures.

Make safety #1.

 

[Alistair_Heaton]

It was definitely a comment made by yourself which highlighted the AP being at the same time as the stick shaker was running which I had missed. Anyway it doesn't matter how it got on the snags list as long as it gets sorted before the plane flies again. The trim issue is going to be by far the hardest to fix anyway.

Clean sheet designs these days don't have memory items. Because it's universally known that they are prone to human screw up.

My first type had 27 separate memory item checklists.

The cs300 has two which are put the oxygen masks on.

All memory item exams are 100% pass mark in my experience. And failure does result in grounding. And various penalties if you fail.

But due to the human responce not being consistent for the same person hour to hour never mind day to day even if the person passes it's no guarantee that they will perform as expected.

Some of the most spectacular falls from grace are with pilots who can spout every little number that are in the book. Bkiut put them in a sick aircraft they freeze and just can't process data. The only way you find out is the first time they have to deal with a full scale emergency. Which could be 10-15 years into thier careers these days if at all.

I suppose a fix could be increased training and testing for max pilots. Plus also a physical test every 6 months to make sure they are strong enough to deal with the control forces and have the stamina to turn the trim wheel manually.

 

[Alistair_Heaton]

An interesting twist on the subject of pilot training is a development in the AF477 accident.

French prosecutors have started proceedings on Airfrance that they didn't train there crews sufficiently.

They also hadn't passed on warning material from airbus or complied with a mod to replace the pitot tubes which had a know issue in icing conditions.

This has the potential to change things somewhat more than the mcas debate on training.

 

[3DDave]

Grounding a pilot from a classroom doesn't hurt the airline - grounding a plane for 24 hours because the airline didn't train a pilot does. It's more important that, in the moments before a flight, the pilots know what they are doing, rather than weeks or months before when they crammed for a test, after which they forget. And those tests don't cover the ADs.

I see that the French prosecutor agrees with me. No surprise that the airlines and some pilots are concerned. It seems to me that the prosecutor is motivated by the evidence from the MCAS initiated, pilot failure completed, crashes.

As to the comment - you'd been specific about it relating to the FDR diagram. My info was drawn from the text of the preliminary report, which the regulators could have read directly. It's nice that my post, copied and pasted from that document, was so critical to their ability to read.

 

[Alistair_Heaton]

It will never happen that they will ground an aircraft as punishment.

AF pilots have mostly been through the elite school system in France which all the political types go through and also legal types.

The pilots got absolutely hammered for being completely useless. So the case is more political face saving by family connection.

If you get your way about training you will have basically killed the 737 and Boeing might as well not bother even attempting to fix it. It's such a cludge of systems hacked together from the 60's the training cost of keeping a pilot current on it will mean nobody will touch them with a barge pole. In fact it would kill the 747 off as well.

Plus the customers won't be happy paying the surcharge for the training.

But its nice in theory.

You would get better value training the engineers.

 

[3DDave]

If I get my way Airbus is gone. There are too many secret modes for pilots to identify what fails.

Another agency has pulled the tickets of multiple pilots over poor training. It looks like there's a trend.

 

[VEBill]

"...Airbus....too many secret modes for pilots to identify what fails."

In the case of Boeing, that'd be MCAS.

 

[Alistair_Heaton]

Err the 777 and 787 are the same as airbus.
With the different modes and control laws.

All the fbw machines have them to be honest.

It's going to be a feature of all fresh sheet designs. Like it or not.

I haven't seen that little gem which agency was that?

I am all for the number of pilots being reduced. It will just mean I get paid more.

 

[LittleInch]

Apologies if this isn't the best place, but something which might be of interest to those in this forum is an illustrators site I found by accident.

He has hundreds of aircraft including most of the 747s, Alastairs Q400 and its a very interesting site. The original 737 is so small. if you want a particular aircraft click on any one then a search line opens up. Or just use google.

Q400

https://www.norebbo.com/2015/08/bombardier-dhc-8-402-q400-blank-illustration-templates/

737 original

https://www.norebbo.com/2018/10/boeing-737-100-blank-illustration-templates/

737 neo (737 -800)

https://www.norebbo.com/2014/01/alaska-airlines-737-890-illustration/

737 max (737-8)

https://www.norebbo.com/2016/07/boeing-737-max-8-blank-illustration-templates/

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

 

[Alistair_Heaton]

I am now hearing a rumour that if they want to keep mcas then the whole flight control system will have to be certified to DAL A standards. Plus secondary independent powered trim stab system.

Apparently I was wrong in thinking it was a 386 processor which was over loaded. It's actually a 286 processor that they use.

 

[davefitz]

Wow, a 80287 microprocessor. Its comforting to know the plane is controlled by a system slightly more powerful than an Atari, and programmed by contractors that are not familiar with aeronautical engineering.

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

 

[Alistair_Heaton]

They didn't have anything to do with that side of things.

That was just a headline grabber.

 

[3DDave]

It's an 80286 variant, the '87 is a floating point co-processor. Compare to the computer in the LEM that put a spacecraft on the Moon and launched it back to lunar orbit. I'd say the 80286 is sufficient. It has an ugly instruction set, but it was created by the company whose founder also drove the adoption of the IEEE standard for floating point computation so that compilers could produce programs to create identical results for identical calculations and therefore support cross-platform software development.

The contractors work to a specification from the Systems Engineering group. If MCAS failed, it's the Systems Engineers to point the finger at. The Systems guys did the AoA that showed that MCAS was safe as it was implemented. 100% MCAS did what the Systems Engineers specified it should do.

There's no argument for blame about the CPU or the developers.

"Taking the 320 as an example, the ELACs are produced by Thomson-CSF around 68010 microprocessors
and the SECs are produced in cooperation by SFENA/Aerospatiale with a hardware based on the 80186
microprocessor."

https://www.airliners.net/forum/viewtopic.php?t=1362007

ELAC stands for Elevator and Aileron Computer; the 68k was what Jack Trameil put into his bare-bones Atari computers.

Too bad about the 68010; Motorola suckered a lot of people with the 68k series. Motorola could have done better, but they seemingly hated their customers. The 80186 was a dead-end from day-one; a better 8086 in a world that didn't want one. OK for cost-conscious embedded work, but not particularly powerful and less than half the transistors of the 80286. I guess airlines are always looking to shave a few bucks off of safety critical hardware.