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.

 

[thebard3]

Waross;
I'm not arguing with your list. I just want to point out that every company I ever worked for suffered from "loss of corporate memory" from one day to the next.
I find it most curious, if my understanding of the MCAS function is correct, is that it just seems to have been poorly envisioned and implemented. Placing so much reliance on a single sensor when another is available, and to give the system full authority over the pilots attempts to countermand it seems to be poorly thought out, and I would assume that the functions of the software went through many layers of peer review, including by experienced pilots who would probably tend to not like that.

Brad Waybright

It's all okay as long as it's okay.

 

[KoachCSR]

I find it most curious, if my understanding of the MCAS function is correct, is that it just seems to have been poorly envisioned and implemented. Placing so much reliance on a single sensor when another is available, and to give the system full authority over the pilots attempts to countermand it seems to be poorly thought out, and I would assume that the functions of the software went through many layers of peer review, including by experienced pilots who would probably tend to not like that.

I believe SparWeb hit the nail on the head:

Refer to last month's Seattle times reports that the original system design of the MCAS included a very small one-time adjustment to trim, but was later expanded in scope after flight test results.

I am not in the aviation industry, but I can empathize on the point that last minute changes on a project on its way out the door often do not receive the same level of scrutiny as they would inside of the "normal" design process. It typically takes a seasoned individual to "catch" the mistakes or shortcomings imposed by late changes and correct them. I would surmise that some degree of pressure from management served to fast track the MCAS changes after the flight tests - as others in this thread have posited as well.

 

[thebard3]

a project on its way out the door often do not receive the same level of scrutiny as they would inside of the "normal" design process.

Hence the need for 'Change Management'. I'm not in the aviation industry either, but in the chemical world all of our processes are subject to review before any changes are made. And usually, if something bad goes wrong the worst thing that happens is that everything stops. Since it's all on the ground, that's usually not a terrible thing.

Brad Waybright

It's all okay as long as it's okay.

 

[IRstuff]

I doubt that Boeing doesn't have change management; it's part of the system engineering process certified by their CMMI rating. However, no such process guarantees quality paperwork, just that paperwork exists and is processed per written procedures, and no such process guarantees that everyone on the sign-off applies their full diligence.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[saikee119]

I find the design philosophy of the MCAS system deeply disturbing.

(1) It accepts input from just one AoA sensor, trusts it to continuously to trim the aircraft to destruction.
(2) It does not monitor/check the result of its own action.
(3) It does not take other information from the plane control to test the need of MCAS.
(4) It has no ability to verify the soundness of the AoA signal like a sudden jump movement that clearly cannot possibly happen to a 70 ton plane flying in mid air. (JT610 jumped 20 degree and ET302 jumped 75 degree are detectable by a simple check on the signal’s rate of change/gradient/first derivative).
(5) It persists activation even the AoA signal indicates the plane is outside the normal operating envelope (unreasonable AoA angles).
(6) MCAS has no other intelligence to tell a nose up from a nose dive and applies the trim indiscriminately.

 

[IRstuff]

"(4) It has no ability to verify the soundness of the AoA signal like a sudden jump movement that clearly cannot possibly happen to a 70 ton plane flying in mid air. (JT610 jumped 20 degree and ET302 jumped 75 degree are detectable by a simple check on the signal’s rate of change/gradient/first derivative)."

This is precisely the same behavior that resulted in death of the pedestrian by the Uber self-driving car. A pedestrian suddenly appeared in front of the car with zero time for even reacting, but the sensors on the car detected the pedestrian well before then, but the software was not written to follow the logical course of induction.

I've seen the exact same lack of design philosophy and derived requirements in a project in my own history.

Sadly, this comes from so many programmers and too few actual and experienced system engineers. Everyone thinks that they can take an image processing class, or read a book, or whatever, and start to program complex systems. Programming is nearly the absolute last thing in the design process.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Alistair_Heaton]

I must admit I have extremely little experience with control systems in fact after 3rd year systems and controls at Uni. I have only recently had to relearn control while renovating a house I am building which I am sticking in a ground heat pump system and underfloor heating and fan coils. And its changed days from my youth with vented hot water cylinders and a 240V triggered gas boiler. So getting my head into TMV3 valves, 2 way diverter valves, 3 way mix valves and smart controls... Anyway back to aircraft in flight....


The problem with stalling is that the books all talk about the stall speed and its very easy to fall into the trap and think that its the only speed the aircraft stalls at. The quoted speed is for a very precise set of conditions ie 1g straight and level flight clean aircraft.


The actual speed which an aircraft stalls is dependent on power setting, angle of bank, weight, CoG, configuration, altitude and what your doing with it. When they quote speeds its actually a speed where the angle of attack hits the critical angle of attack for the defined conditions.

At the beginnning of flight stalling wasn't understood at all. In fact in the First world war the standard procedure for dealing with a stall was to pull back on the stick until you crashed. Then some british combat pilot survived a stall because he lost grip on the controls and the aircraft recovered itself by dropping its nose. He spotted this and then tried it on purpose and pushed the nose down and recovered. He reported this to the powers that be and was instantly court martalled for disobeying the pilot order book. But the stall recovery procedure which we still use today was born. (I don't have a reference for this tale, I had the utter joy of flying with a BAe test pilot in his final job before retirement who told me this. Ex empire test pilot school, Lightning pilot, and was the lead test pilot on the Jetstream 41. So don't have clue if its true or not.. but it does have a ring of how it might have happened.)


The first stall systems where sucktion hooters on slightly the underside of the wing, which when the pressure changed from positive pressure of the underside producing lift changed to negative pressure they gave a sound like a duck getting murdered to the pilot. This system is still used today on the likes of cessna 150's to C172's


It then progressed to a electrical system which used a tab in the airflow which when lifted by the same effect would trigger an alarm and recovery systems. The likes of Jetstream 31/32's shorts 360 straight wing had this sort of system.

With the developement of straight wing jets the above methods didn't work for high altituide and any transonic airflow and they had to look at something else.

Because airspeed can't give you the state of the aircraft apart from in very set conditions the only thing that was left was the measurement of the actual angle of attack. So they started using AoA vanes. But it didn't progress through the whole system. They decided that the pilots didn;t need to know the actual angle of attack and could stick with just nailing the airspeed and getting the values from a book. Which for transport type aircraft works fine. Military came away from this with performance aircraft and started using angle of attack on approaches and the like and the pilots nail the angle and the airspeed looks after itself and they put enough power in to maitain the flight profile they want.


I really don't know why, its still to this day a huge reluctance to allow pilots to see the actual AoA of the aircraft. Most of the flight management systems use it to set cruise speeds for fuel effeciency etc also it does the stall system. On the flight deck with EFIS avionics we can see the AoA varying with the low speed red tape on the airspeed indicator. Most of the time on turbo props we can't see it but on approach to land we come in at approximately 30 knts above the tape. On windy gusty days it bounces up and down and when it gets too close to the Vref we add a bit on so the stick shaker doesn't go off. But if you add to much on, when it comes to the flare to land the aircraft has way to much energy and will float and all your landing performance numbers will be way out. Add in a contaminated runway and you have all the holes in the swiss cheese model lined up for departing the runway at the end or off the side.

It started off as a single job system which only triggered an alerting system. Then the data was used to feed into other systems which wasn't a problem when those system had no direct control over the aircraft apart from say setting a cruise power. Now its being used to directly control the aircraft and its systems. But fundamentally the sensor and data handling and error trapping is the same as it was in the 1950's when it was developed when the limit of its authority was sounding an annoying alarm in the cockpit and triggering a vibrator on the stick.


In my experence AoA vanes are very solid infact its usually the heater which they have to stop ice forming on them that goes which then means you can't fly in icing conditions that causes thier replacement not the actual unit providing dodgy data.

 

[waross]

Hi Alistair. Thank you very much for the information that you have shared with us.
I heard a slightly different version of the WW1 story.
As the story went, there was a tendency to spin out when turning.
One pilot had a theory that you may be able to recover from a spin by diving.
He argued this with fellow pilots.
Then one day he tried to test his theory.
The pilots on the ground observed that he did bring the plane out of the spin.
Unfortunately he did not have enough altitude to recover from the dive and died in the crash.
This may be a different version of your story.
We may never know the true story. grin.

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

 

[WKTaylor]

To an extent these 2-mishaps continue to remind me of the Aeroperú Flight 603 B757-23 that crashed in the ocean very soon after take-off on a dark night with very poor visual references.

https://en.wikipedia.org/wiki/Aeroperú_Flight_603

The aircraft took off 42 minutes after midnight (05:42 UTC) on 2 October,[6](p10) and straight away, the Boeing 757 airliner crew discovered that their basic flight instruments were behaving erratically and reported receiving contradictory serial emergency messages from the flight management computer, including rudder ratio, mach speed trim, overspeed, underspeed and flying too low. The crew declared an emergency and requested an immediate return to the airport.[2]

In this case the pitot tubes had been taped-over for maintenance [wash?] and the tape was inadvertently left in-place. Photos of the wreckage in the ocean clearly showed the Pitots covered with tape. This prevented the ADC from interpreting the flight condition and poor visibility made visual flying almost impossible.

In the case of the 737MAX it appears the compounding factor of the MCAS... working in the background with unreliable pitot/AOA 'bug' [left, right side] may have made the situation even worse than the AeroPeru 757 experienced with blocked pitot's.

I wonder if 'bugs' or other FOD [mud, etc] sources could have contaminated one or more Pitots? I doubt if the wreckage will ever reveal the in-flight condition of these probes.



Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[Alistair_Heaton]

Don't go with that one I am afraid.

When you are in a spin everything is stalled apart from the rudder. You can wiggle and waggle the stick as much as you like it won't come out.

Only thing that will work is opposite direction rudder and when that then bites you get airflow again over the other controls. The aircraft is actually under very little loading while its spinning. You just have to watch you don't pull the wings off when you pull the nose up after the rudder input stops the spin.

 

[waross]

Thanks for debunking that Alistair.
It was a long time ago and memory fades. It was probably something out of a novel, possibly by Nevil Shute, possibly not.
Given Shute's background as an aeronautical engineer he would have known better.
Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[BridgeSmith]

Alistair_Heaton, I think the suggestion was not to have the system ignore the AoA sensors at any particular angle reading (maybe some limit where the angle would not be possible?), but above a certain rate of change in the reading that would not be possible, indicating a fault in the sensor. Would that be a workable improvement to the system?

Also, from some of the previous posts I get that variations between AoA sensors on opposite sides is a normal occurrence, but is there a limit on how much the angle of attack can vary between one side and the other during normal flight for these planes?

 

[Alistair_Heaton]

My point was more there is no other way to obtain if you about to stall or not.

I really don't know how much angle change is required to give a certain amount of speed red tape to change. Or for that matter how much difference you get between sides.

That sort of detail is not required to operate the aircraft. The B2 technicin s avionics will have limits in thier mans but pilots do not get access to them normally.

It doesn't help if I speculate with you lot so I will stick to only the stuff I am sure about.

 

[MacGyverS2000]

Apropos to Alistair's story about recovery...

https://www.historynet.com/spin-control.htm

Dan - Owner

http://www.Hi-TecDesigns.com

 

[btrueblood]

Alistair, the account related by your instructor is pretty spot on. From the wikipedia article on Spin (aerodynamics), the history section says this:

In August 1912, Lieutenant Wilfred Parke RN became the first aviator to recover from an accidental spin when his Avro Type G biplane entered a spin at 700 feet AGL in the traffic pattern at Larkhill. Parke attempted to recover from the spin by increasing engine speed, pulling back on the stick, and turning into the spin, with no effect. The aircraft descended 450 feet, and horrified observers expected a fatal crash. Though disabled by centrifugal forces, Parke still sought an escape. In an effort to neutralize the forces pinning him against the right side of the cockpit, he applied full right rudder, and the aircraft leveled out fifty feet[17] above the ground. With the aircraft now under control, Parke climbed, made another approach, and landed safely.

In spite of the discovery of "Parke's technique" spin-recovery procedures were not a routine part of pilot training until well into World War I. The first documented case of an intentional spin and recovery is that of Harry Hawker.[18] In the summer of 1914, Hawker recovered from an intentional spin over Brooklands, England, by centralizing the controls. Russian aviator Konstantin Artseulov, having independently discovered a recovery technique, somewhat different from Parke's and Hawker's, on the frontlines, demonstrated it in a dramatic display over the Kacha flight school's airfield on September 24, 1916, intentionally driving his Nieuport 21 into a spin and recovering from it twice.[19] Later, Artseulov, at the time an instructor at the school, went on to teach this technique to all of his students, quickly disseminating it among the Russian aviators and beyond.[20]

In 1917, the English physicist Frederick Lindemann conducted a series of experiments in a B.E.2E [21] that led to the first understanding of the aerodynamics of the spin. In Britain, starting in 1917, spin recovery procedures were routinely taught by flight instructors at the Gosport School of Special Flying, while in France, at the School of Acrobacy and Combat, Americans who had volunteered to serve in the famous Lafayette Escadrille were by July 1917 learning how to do what the French called a vrille.[22]

 

[charliealphabravo]

The article also relates the story that @waross may be remembering.

The first chapter of the spin story closes with a tragic irony. Less than four months after his landmark spin recovery above Salisbury Plain, Wilfred Parke set off from Hendon on a cross-country trip, piloting a sleek Handley Page E monoplane. About five miles out over Wembley golf course, the engine was running so rough that he decided to turn back toward Hendon, trusting in the luck and skills that had served him so well. But as he turned downwind over the treetops with a failing engine, his Handley Page stalled—the beginning of Parke’s Dive. He knew what to do, of course, but this time Parke fatally ran out of height and luck.

 

[Alistair_Heaton]

I am not surprised that Tim's account of what happened was more true than not. I think he was RN as well originally and a carrier pilot.

I had a very enjoyable week of flying charters with him between Aberdeen in Scotland and Bergen, Stravanger and OStesund in Norway and Shetland in Scotland. I had been a Captain for 6 months but it was very easy to fall into being a FO again with an old master aviator in the Captains seat.

We spent our time in the cruise and waiting for pax discussing aircraft performance and aerodynamics. The pilot theory taught is basically wrong and simplified to the point of having a clue but no clue. I had the basic stuff from my engineering degree but never really understood terms such as control harmony, dutch roll and the big one which is min control airborne. The last one is for when we are one engine down and the rudder is your life saving control input, not enough air over it and it won't produce enough lift to counter the moment from the live engine and you won't have enough roll authority to stop the aircraft rolling on its back and going into a none recoverable spiral dive at low level. We had plenty of empty sectors as well so he got to demonstrate alot of what we were discussing. So it was the back side of the weather brief covered in maths and theory with pax onboard and flying on one engine empty. I realise now how lucky I was to experence this. Most pilots never get to see these effects and survive never mind get them explained by someone that knows what they are talking about and can waggle a stick.

That generation of aviators who were slightly before Sully's time are completely different to modern day pilots. Tim was friends with Douglas Bader, Eric Bristow and the like and had the personal email address of Niel Armstrong. Niel it was nothing to do with the moon it was to do with glue for fixing a wooden and canvas aircraft. Can't remember what type it was but think Tiger Moth vintage. It involved horses and human piss I seem to remember. But the resultant glue had some extremely good properties for canvas and wood.


Its some 14 years ago now but that week still has day to day effects on my flying and also in the sim checks as well. Even if you wanted to pay for it I susepct it would be very hard for a pilot to get the training I did that week with Tim. It was two blokes with a love of aviation that could both speak maths, one a grandad (Tim) and one a grandchild (me) who was hanging on every word he said, and he loved being listen to and being asked questions and was more than happy to answer them. And we had an aircraft to play with and fuel to burn.


I suppose in Engineering terms it was like getting your first lead engineer job as a naval Architect and the Engineering director used to work on Square riggers and knew how the sail plan and rigging worked and you got stuck in in some ship yard together with a bar for a week waiting for transport and he found out you were building your own boat. Then things developed....


I was feeling guilty then of thread drift but then realised the above yarn is actually the issue with this issue. There are no people like Tim involved with the big picture or just lots of people dealing with thier own little picture and not communicating to the big picture. So we get "we have a problem big picture, the stabilistion performance at high alphas is outside certification limits with current values".... "ok small picture fix it and let me know" "ok big picture we have fixed it with a software change of the limits"........ Now this next bit is where the error wasn't trapped. Was the question asked how and what effect this software fix was going to have on the big picture or was it just accepted as fixed.

 

[JAE]

Boeing CEO says it’s completed 96 test flights with 737 Max software fix

CNBC Link

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[waross]

Big picture, little picture.-
Little picture.
Does the fix work?
Big Picture.
With the radical change in characteristics caused by the new engines, is the extension of type certification valid, or wise?



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

 

[waross]

Alistair; That is an experience to treasure.
Thank you for sharing with us.
It brought to mind an experience of mine.
Very early in my career I had a job driving the truck for the guy who changed the street lights in a big city.
I was young and he was old.
One day he spotted an old work mate on the street and hailed him.
We went for a very long coffee break.
These men had worked together stringing the first rural distribution lines across the Canadian prairies.
I was a fly on the wall as these old timers remembered the old days.
In the hot, dry Canadian summer on dry cedar poles they worked 2300 Volts bare handed.
"It was okay as long as you didn't get in series."
"Ya, you never wanted to get in series."
"Do you ever hear from Joe?"
"No. Didn't you hear? He got in series down by Moose Jaw."
"Ya, you didn't want to get in series."
"What about Harry?""
"Oh, he got in series over by... What about Sam?"
"Sam got in series by..."
I guess that these were the only two left. Every mutual friend that they could remember had 'Got in series' somewhere.
After about an hour, I envy the week that you spent, Alistair.


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