Boeing 737 Max8 Aircraft Crashes and Investigations [Part 2] 44

[Alistair_Heaton]

This thread is a continuation of:

thread815-445840


****************************
Another 737 max has crashed during departure in Ethiopia.

To note the data in the picture is intally ground 0 then when airborne is GPS altitude above MSL. The airport is extremely high.

The debris is extremely compact and the fuel burned, they reckon it was 400knts plus when it hit the ground.

Here is the radar24 data pulled from there local site.

It's already being discussed if was another AoA issue with the MCAS system for stall protection.

I will let you make your own conclusions.

 

[enginesrus]

The airframe problems with newer aircraft or should I say the flight controls, were essentially a non issue in the older days of aircraft. The main problem is electronics and computers, the lack of manual systems like the days of control cables, and using a motorized or servoized jack screw to pitch the tail plane or horizontal stabilizer. DC-3's and DC-7's and other such aircraft used a controllable trim tab, not the whole tail plane, just like a small aircraft and if for some reason it got stuck the pilot could still over ride the effect. I am thinking even if with this max 8 situation, if the pilot knew what to do to prevent the dive into the ground the flight control system would very likely not allow him to do the required maneuver.
The fix for the dive would have been to invert the plane, scary yes, but still better than diving into the ground.

The main question is: How come even with full nose down tail plane the controls ie the elevator could not over come and counter that effect?
My guess is, the flight control system won't allow it. There is always a simple solution to problems like this. Back to the good old system, control cables and fixed tail planes.

 

[RVAmeche]

If that's the case, its a crappy control scheme in a life or death application. To me that's not a good instance of "back in the good ole days things were simpler", it's just an overall poor design.

 

[Lou Scannon]

I may be inaccurate in detail, but qualitatively speaking, the last two large aircraft, (AFAIK, maybe there were other contemporaries, e.g. from Tupolev), that had unpowered main controls or full manual reversion in the event of power assist failure, were the Bristol Britannia and the original Boeing 707. These concepts did not persist into succeeding generations of large aircraft, as they had their own drawbacks, I suppose, high control effort possibly being one, and lack of direct linkage (i.e. servo controls) being another. Hence the introduction and eventual ubiquity of fully powered main controls on large aircraft.
In general, the large aircraft manufacturers seem to have gotten it right from the late fifties until recently.
I remember the 737 (or was it DC-9?) had some kind of mechanical issue with the elevator jack, several decades ago, but that was a design or material issue, not a conceptual issue, IIRC.

"Schiefgehen wird, was schiefgehen kann" - das Murphygesetz

 

[JoshPlumSE]

Not sure this adds a ton to the discussion. But, for the folks who don't know much about jet planes and such, Juan Brown's channel on YouTube (blancolirio) has some good discussion about these 737 crashes. Juan is a commercial airline pilot (777) here in the US and came to my attention with some good reporting on the Oroville Dam spillway issue and construction.

https://www.youtube.com/user/blancolirio/videos

 

[waross]

I am wondering how those AoA sensors ever got approved.
Not a very good MTBF (Mean Time Before Failure)
And not a very good MTBF (Mean Time Between Fatalities).
The first delivery was a MAX 8 on May 6, 2017,
Second crash was on Mar 10, 2019
That's 344 fatalities in 673 days.
Even factoring in the 350 Max 8s flying it is not good numbers.
Then add in the several failures of the AoA sensor that did not result in a crash.
Are these sensors new to this aircraft? If they are used without problems on other planes, why are they failing on the Max *.

Anecdote. A friend of my wife's was visiting today.
I overheard her mention to my wife that she had phoned WestJet concerning her upcoming Mexican holiday.
She wanted to be reassured that she would not be flying on a Boeing plane.
In the end the market place may deliver the harshest punishment to Boeing.

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

 

[MDEAus]

Thank you Alistar,

Is it possible that MCAS is triggering before any stall warnings are active? I know we are limited on the knowledge of MCAS system but if it is activating only from AoA input (lion air had faulty input for some reason), What does the FCC use to determine stall? Lion air had the stick shaker active the whole flight from memory so they must have been dealing with false stall warnings.

The MCAS system could have been employed effectively to simulate the handling of the NG aircraft. From my perspective MCAS should have greatly limited authority, and a limited range (50% of nose down trim or whatever is counter-able with stick pressure to keep the aircraft level).

This whole situation to me boils down to
[ul]
[li]An underestimation of the risk associated with the MCAS system (which may have been fine if it was limited to 0.6 degrees per activation or been prevented from driving the stabilizer to full nose down)[/li]
[li]Uninhibited authority off the MCAS system leading it to command trim inputs (requiring pilots to counter it and ensure they've arrived back at the correct trim setting) [/li]
[li]Reliance on a procedure for runaway stabilizer (can present much different symptoms?)[/li]
[li]Reliance on only (one) AoA input for MCAS activation[/li]
[li]Pilot confusion over strange symptoms (poor documentation and training)[/li]
[/ul]

There may have been things that the pilots could have done to save the flights (see previous flight of lion air) but a system like this should never have been approved. Another thing that annoys me with the MCAS is that it's to simulate the NG 'feel' during high AoA maneuvers yet it's supposed to only trim nose down, so after any high AoA maneuver ( yeah take off is probably it in a normal flight) Trim needs to be reset by the pilot to what they would consider normal. I'm guessing this is something that MAX pilots must have experienced and noted that it is different to the NG? If not is it safe to say that MCAS isn't actually required or could be significantly limited to low levels of trim adjustment? As normal operations wouldn't normally see it activate and higher AoA conditions should still be within the flight envelop otherwise this would instead be a stall protection system which I'd assume would have a greater risk profile.

 

[LionelHutz]

debodine - You bring up a point that has made me wonder if there is a bigger issue that has remained hidden. I would have expected the pilots to know to throw the trim cutout switches when experiencing runaway trim. So, I'm finding it difficult to explain why they struggled for minutes counteracting runaway trim without throwing the switches.

enginesrus - In this case, Turning the trim cutout switches to the off position and then using the manual trim wheel instead solves the problem. So, the solution is to revert to manual controls, just like you want.

 

[Alistair_Heaton]

I only have access to the same info as you how the MCAS system works... Which is the same as the pilots who were flying them.

Just some points through about flight controls.

I have flown unpowered controlled aircraft the Jetstream 31/32 and 41 the 41 is a 10 ton aircraft. They are firm and relatively slow response and you get a good work out on a high wind approach. When you move to bigger aircraft the control response is approximately the same weight but the forces involved normally are the same. this is done via what's called a feel unit which applies a force to the controls which varies with the airspeed. If you say have a hydraulic failure its a whole different ball game your talking two feet braced against the panel and pulling with both hands to move the elevator backwards to flare to land.

FBW is a different ball game which i have zero experence with outside flying a Airbus sim a couple of times. I don't want to comment about it. Mates that have flown them say the first 3-4 sessions in the sim are a bit wierd but you get used to it and after that it no problme but its a different way of thinking.

Now the reason why they had to have the MCAS system. The leap engines are extremely effecient. We had 737 classics fleet and also CS300 fleet the CS fleet uses 1600kg of fuel an hour the 737's its 2400 kg an hour. The fuel savings are collosal. The leap engines are a geared fan and the fans are huge. This means the engine pod is huge. So when they fitted them to the NG wing they didn;t have enough ground clearance. SO they raised the gear height and to get more room the moved them forward and up.

737 800

737 max 8

 

[Alistair_Heaton]

now if you look at the shots above its pretty obvious the geometry changes that have needed to be done.

The leading lip of the nacelles if significantly forward of the leading edge of the wing on the max.

Now what's apparently happening is when the aircraft gets into high angles of attack the huge pods which have a flat base are generating lift and because they are so far forward of the centre of lift of the main wing they are forming a moment couple in addition to the elevator. This has the effect that as the angle of attack increases the control force decreases as does the stability. This is completely against certification rules. As you get closer to the stall the control forces should increase and the stability increase with the aircraft having a natural inclination to lower the nose to decrease the angle of attack. In all cases in the event of a stall in straight and level flight if the pilot lets go of the controls the planes stability should mean that it self recovers ie the nose drops and the angle of attack is reduced.


This wasn't the case with the max with those engines. So they stuck a computer fix in that altered the trimming of the aircraft to produce forward trim as the angle of attack increased. That in itself I don't have a problem with really. What they didn't do is then complete the risk matrix to look at how the Angle of Attack is measured and the failure modes. We tend to have three of everything for critical systems.


Take Hydraulics for example: we have a separate system powered by each engine with either a backup electrical pump or a PTU from the other system or both they are powered from the main electrical buses with pumps fitted to the ancillary gear box . We also have an emergency system which is completely separate which is powered by a DC pump off the emergency battery bus and on these big jets that will be linked to the ram air turbine which is a little windmill that pops out the side of the plane in the event of power failure of both engines electrical supplies (again 3 systems) it has an accumulator and the pump charges it so you have enough pitch authority to flare to land.

The MCAS took its input from 1 AoA sensor of which there are only two fitted so no way of determining the odd one out in the event of a failure. No comparison indicator warning on the flight deck as standard between the two sensors. All other critical flight instruments again follow the rule of three. And all the rest will trigger comparator warnings if there is conflict in the readings, then the pilots can look at the separate system and two out of three win and the faulty reading can be deselected.

Realistically I can't see a fix for the engine lift issue without a complete redesign of the wing and aircraft geometry.

The instrumentation the minimum I can see is them having to fit a third AoA vane. With that it requires a huge rethink of the avionics software. The retrofitting a 3rd vane will not be easy either.

After that pilot training needs to be revisited and there needs to be more than just a 2 hour CBT. And simulators need to programmed for the MAX currently there is only one in Seattle everyone else does their 6 monthly training on NG 737-800 sims which I believe have MAX performance programmed into them getting rolled out but don't have the MCAS system simulated.


Realistically the safest option is to kill the grandfather approvals based on old type approvals. But that would significantly increase the costs of new types in the same product lines.

 

[Alistair_Heaton]

To note flying an transport cat of aircraft inverted is an extreme solution. As pilots we are not trained to do it. The systems are not designed to operate under prolonged negative G. They are not tested to do it.


For them to be able to do it then the complete interior of the aircraft would need to be redesigned for negative loads the additional metal required for the seats alone would put so much mass onto the airframe it would be uneconomical.


The aircrat should not be designed in such a way that its required.

 

[itsmoked]

Why can the trim greatly exceed the elevator control on a computer controlled flight system?

Answer: So the government can take over the plane and plant-it from the ground if needed. [/conspiracy theory]



Keith Cress
kcress -

http://www.flaminsystems.com

 

[Alistair_Heaton]

Oh lord is that the nonsense that they are coming out with now?

Its bad enough when you get a flat earth or chemtrail nutter on board.

The flat earth sit with a spirt level in flight filming it with their cameras.

And the chemtrail lot try and take samples of everything including the crap that comes out the drain valves at the bottom of the fuselage onto the ground. which is the accumulation of sweat and breath moisture which condenses on the inside of the hull in cruise and then melts and runs down to the bottom and drains out the spring valves when the plane is not pressurised.

I was flying an aircraft with a water meth boost kit for hot high operations. The tank was visible in the rear hold when pax were boarding. I thought one guy was having a sexual experience in his pants walking up the steps trying to get pictures of it.

 

[Alistair_Heaton]

This apparently what it does to the stability

 

[itsmoked]

Yeah, that was insisted at me by an engineer.

BTW: Juan Brown put up a pretty informative vid on the trim equipment showing it running.
YouTube 737 MCAS

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Alistair_Heaton]

first words that entered my head when I read your post was "software...applemac"

 

[LittleInch]

Alistair,

A great explanation as usual.

It seems to me that Boeing are not in a good place with this design and explains why there was no quick fix following the lion air crash.

In order to follow the certification rules and prevent a runaway stall they need to modify the flight controls and their only means seemingly is to actually alter the flight control surfaces and not the flight controls.

To do this though using only one of two AoA sensors and then having to increase the action from 0.6 degrees to 2.5 degrees has been disastrous. Sure - there are ways to prevent this leading to destruction but it is clear that with all hell breaking loose in the cockpit, stick shaking, alarms, the plane seemingly having a mind of its own and one or two quite hard to get to switches a procedure built approach to a sensor or computer fault isn't working. This (the MCAS system kicking in) was not the only thing going on and both planes apparently had to deal with faulty or discrepancy in sensors which could easily lead to confusion as to what was happening.

I don't see any easy way out which will allow a quick return to operation and keep within the rules of certification. Fix one problem and create two more.


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

 

[Alistair_Heaton]

Also I am being told that this idea you can use the trim cut offs is a bit simplistic.

Yes you can shut them off.... but and its a big one.

The manual trim unpowered requires a handle to be folded out from the trim wheel. Then the thing wound for minutes to get it back from the full pitch forward position to something nearing what you actually need and its relatively easy to still control the aircraft.


I am reliably told by the classic pilots at work even if pre prepared for it, the FO built like a grinder on a volvo ocean race crew your going to loose an min of 6000ft until its sorted. Normal people who aren't pissing about in the sim its more likely 10 000ft.

 

[RVAmeche]

Man itsmoked, I was really hoping you were just kidding. That's depressing.

 

[LittleInch]

Alistair - Man that is not good.

I thought it might take longer clearly if there is no electric power, but 6, to 10,000 ft....
So unless you manage to hit the off switch just after you've manually thumbed the trim switch back to near level then you're done for at low altitude??

I saw somewhere that the final nose down on Lion air was because the FO only flicked the trim switch up instead of holding it up to regain trim back to near level and then the MCAS kicked in again.

It seems as though the manual trim up switch on the yoke doesn't work while the MCAS is busy winding it down. Can anyone confirm this?

Maybe the same happened on ET with a low hours FO taking over while the captain tried to work out what was going on.

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

 

[Alistair_Heaton]

Your pretty much done with most control issues at low altitude.

Which is why we tend to head upwards like a fart in the bath to give us room and potential energy to deal with things.

The electric trim switch should kill the MCAS system for a period of time. Then it resets and does its thing again. Sorting out what's actually going on with all the bells and whistles going off is shall we say intense.


Another thing is that if the stick shaker is triggered that usually kicks the auto pilot out. And as soon as the auto pilot is out the MCAS system kicks in if the flaps are up.