Rogue Boeing 737 Max planes ‘with minds of their own’ | 60 Minutes Australia

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Winston

Lorenzo von Matterhorn
Joined
Jan 31, 2009
Messages
9,560
Reaction score
1,748
"No critical system should rely on just one sensor."

In this case, it was an INCREDIBLY critical stall prevention system sensor. This fatal one-sensor choice was made to avoid what would have been FAA-mandated simulator training triggered by the presence of a two-sensor system. One of Boeing's major selling points for the Max over the Airbus A320 which was seriously encroaching on the 737 market was that it required no simulator training for current 737 pilots. Pilots were instead given a DIY training app which could be run on a tablet. The one-sensor stall prevention system wasn't even mentioned in that training app.

Rogue Boeing 737 Max planes ‘with minds of their own’ | 60 Minutes Australia

 
They are in a real mess. The FAA should never have let them certify their own planes, should be TWO external groups with great capability.

I don't see why they even need those AOA sensors on the outside, they seem to be unreliable. I saw a photo of a military transport recently, it had 2 of them on each side. They have an IMU inertial measurement unit on board, 3 axis, the data at least goes to the flight recorder, why not get AOA from that?

That pitch down control was a bad idea from the start, and the means to disconnect it is not intuitive or simple enough. Just have one big red button that reverts everything to manual!
 
A really bad situation and very unbecoming of a world leader in manufacturing airplanes. As a top-of-the-line and world class leader they should have made sure that their product would work under all circumstances. This is really bad because aerospace exports are one of the few bright spots in our balance of trade/exports. With regard to balance of trade and maintaining the U.S. as a go-to leader in commercial airlines this affects all of us.
 
https://www.hideoutnow.com/2019/05/former-boeing-controls-engineer.html

Whistleblower? These planes need a totally independent
EMERGENCY BACKUP AUTOPILOT, that kicks in automatically during
any emergency, like loss of control, hijack, etc. Takes over and just
flies in circles until ATC figures out what is going on. They could fly
the plane back to the airport, then yell at the pilots. Or bust the hijackers! If they can make self driving cars, they can make self flying planes.

Former Boeing controls engineer confirms that 737 MAX switches were indeed modified in a manner that kept pilot from disabling the MCAS system separate from own controls. Pilots were not informed and Ethiopian pilot tried to use switches in vain.
 
I don't see why they even need those AOA sensors on the outside, they seem to be unreliable. I saw a photo of a military transport recently, it had 2 of them on each side. They have an IMU inertial measurement unit on board, 3 axis, the data at least goes to the flight recorder, why not get AOA from that?

Because the IMU has no idea what the air outside the airplane is doing. Analogous to trying to get airspeed from GPS....which also has no idea what the air is doing.

As someone who worked on the MAX for some time before I retired (took a voluntary layoff, then retired, actually) this whole situation makes me angry and sad every time I think about it, including right now. :mad:
 
Well, as a 737 pilot with thousands of hours flying them. I can say the 737 is a VERY safe airliner.
I don't like to speak ill of the dead, but lack of pilot experience and poor training and decisions are what really the
issue here.

MCAS is a new system for the MAX, but the there is a boldface (memory) procedure that deals with this issue. It has been on every 737 for decades. Every pilot trains for it. It's not complicated and would solve the issue in short order.

While it's true that Boeing did not have the right, left, and standby flight instrument systems do not talk to each other, it does give an AOA disagree warning on the cockpit displays.
The aircraft has 3 independent flight instrumentation systems. Its basic training to quickly crosscheck the 3 systems and determine the "odd man out" and act accordingly.

As a professional pilot we are required to be able to fly our aircraft with a major system failed. This is a minor system.
I understand the startle factor, but basic airmanship is expected.

Pitch, power, airspeed. Words to live by in aviation. Set an engine power. Set a pitch angle. An airspeed will follow.
On a 737 its 10 deg nose up with the flaps out and 80% engine power. Set that and you will be safe and climbing at a reasonable rate.
Every pilot should know them by memory.

The Lion Air aircraft had multiple failures of the MCAS system in the days prior. It's criminal that the previous aircrews did not report it to maintenance for repair. The accident crew NEVER did the runaway stabilizer memory procedure.
A simple 3 step procedure that would have solved the problem in short order.

Ethiopian air suffered a bird strike on the runway that caused the AOA vane to separate from the aircraft.
The crew started the Runaway stabilizer procedure, but never finished it and ended up reversing their actions and reengaging the stab trim system. To compound the issue, they left the engines at takeoff thrust and accelerated to 340 knots making manual trimming almost impossible.

The bottom line is I would have no problem putting my family on a 737 MAX tomorrow with 2 qualified pilots.

I have flown the simulator with the stickshaker going off continuously due to a failed sensor while manually trimming the aircraft. Its not fun, but It can be done and done well.

The news media like a villain, Boeing will take some blame. There is no reason for the left and right flight instruments not to talk to each other. It's an easy fix and what is coming, along with limiting the nose down trim authority of MCAS.

The real issue here is qualified, experienced pilots. Many foreign airlines in developing countries don't have the supply of pilots available to them. In the past they would hire qualified pilots from developed countries, but that supply has dried up. The shortage of pilots is very real. Qualified/experienced pilots are leaving the smaller airlines to work at major airlines all over the world because the money/working conditions are so much better.
 
But angle of attack is just the deviation from horizontal on the pitch axis, what does airflow matter? Those moving vane things seem to be unreliable. If they control missiles with IMU's, why not planes?

When they output the data from the flight recorder, it shows in great detail the movement of the aircraft along all 3 axis. These planes were porpoising up and down. If they had that data in real time, it could have been fed into the flight controls to correct the flight path.

And that Airbus that stalled 5 miles down into the ocean, the pitot tubes were frozen. Why could they not use GPS data? They did not even know they were sinking. I remember one case where someone cleaning the plane put tape over the static port and forgot to remove it. On the TV show MAYDAY. Resulted in loss of control and another fatal crash. Maybe they should just give up measuring airflow, use other methods.

MACH: I heard the AOA disagree yellow lite on the max was AN OPTION that was usually not obtained. They are now making it standard. Also see my post above from a Boeing Engineer. I read that 3 times, and still cannot make sense on how the system worked. And I design control systems. Well not for planes. Probably for the best.
 
But angle of attack is just the deviation from horizontal on the pitch axis, what does airflow matter? Those moving vane things seem to be unreliable. If they control missiles with IMU's, why not planes?

No, the angle of attack is the angle of the wing chord plane to the airflow it's operating in. It has NOTHING to do with "horizontal".

Mach7: The one thing that troubles me most (besides the single AOA input to MCAS) is the assumption that was made that MCAS gone awry would be stopped by doing the runaway stab trim procedure. But....as one with the experience that you have, would you recognize a stab trim system that put in some down trim and stopped, then did it again some seconds later and stopped again and so forth be recognized as a trim runaway?

Granted, after the Lion Air crash and the fact that this system was then known to be on the airplane, it might be more so....but what, in your experience triggers that memory item to do the runaway trim procedure? Would a pulsing trim system do it for you?
 
But angle of attack is just the deviation from horizontal on the pitch axis, what does airflow matter?

Angle of attack is by definition the angle relative to the airflow and has nothing to do with horizontal or any other reference plane whatsoever. Bank angle, pitch angle, upside down, right side up...doesn't matter. It's just the airfoil and the flow. IMU can't tell you anything about the wind, and that's the only thing that matters.

Edit: Ha...BEC beat me to it by a split second.
 
Well, check this ref. Airliners normally fly horizontal, I have not flown on any doing aerobatics. I fly hand launch gliders. They glide horizontal. Increasing the angle of attack UP increases lift. Seems simple to me.
https://en.wikipedia.org/wiki/Angle_of_attack

Note in the diagram that AOA is referenced to THE HORIZONTAL PLANE.

Your statement of:

"Angle of attack is by definition the angle relative to the airflow and has nothing to do with horizontal"

does not seem to agree. OK, I am not a pilot, only a model plane flier, but I think the principles are the same.

Also I think it is possible to make an airplane control system that does not measure airflow at all. The IMU tells position in 3D. You can infer speed from engine power and attitude. Add in GPS data and radar altimeter or laser ground ranging. How about optical horizon sensors?

How about putting that F-16 auto GCAS on airliners and stopping the crashes?

Missiles can fly halfway across the world and land within 100 feet, without measuring airflow at all. IMU and computers are the trick.
 
John, AOA is in reference to the relative airflow, not the horizon. A plane can be flying straight up at high speed, and have a very low angle of attack.
 
From that article:
In aerodynamics, angle of attack specifies the angle between the chord line of the wing of a fixed-wing aircraft and the vector representing the relative motion between the aircraft and the atmosphere.

The horizontal plane in the picture is just there for reference. The measurement is between the wind and the airfoil (red and blue lines). The wind could change direction and have an updraft or a down draft and change the angle significantly without affecting the attitude relative to the horizontal plane.

Airplanes and missiles are very different aerodynamic vehicles. Missiles never fly anywhere near stall speed, while airplanes can and do in almost every phase of flight. The control systems are not interchangeable. Principles, yes. Guidance, yes. But you can't land an airplane safely without airspeed and angle of attack information....or else you land like a missile, which they kinda did.
 
An upside-down plane at the same angle of attack with still produce the same lift, it will just be in the downward direction.

Remember that GPS only knows position. Velocity (speed and direction) are inferred.
 
An upside-down plane at the same angle of attack with still produce the same lift, it will just be in the downward direction.
Exactly! This is called a "loop". Edit - well, flying it from horizontal all the way over the top and around at a constant AOA is a loop, for the pedantics out there.
 
Shouldn't there be some vested interest on the part of Boeing in selling this plane overseas and keeping their worldwide reputation. Rather than just let the plane be flown with unqualified pilots with not enough experience and let the chips fall where they may. Once such tragic number of lives are lost, it is impossible to restore things to the way they were.
 
Xolvejohn. No the AOA disagree warnings (like the Altitude and airspeed disagree warnings) are not optional.

What is optional is an AOA display on the primary flight display. My airline has paid for the AOA display in my Heads up display. Its information, but in the airline world we don't really use it. When I was in the Air force years ago we did use it more.

The discussion on AOA vs inertial reference here is spot on. AOA is only RELATIVE wind relation to the wing. There are many reasons the IMU would not give useful information. Some examples are windshear or microbursts. Both very dangerous.

And lets stop talking about the G-LOC recovery system used on fighters. It's totally irrelevant to airliners. It's an automatic system that requires everything working on the aircraft except a pilot that has blacked out due to high G's.
Very limited in scope and if it screws up the pilot has an ejection seat. Not an option for the 180 passengers on an airliner.

What we are talking about is mitigating the effects of failed sensors/computers, so everything is not working properly requiring the pilots to make judgment calls while analyzing rapidly changing conditions.

Airliners are VERY complex, VERY VERY complex. On a good day, with everything operating and good weather it looks like we are a bunch of overpaid bus drivers. But thats not what we are paid for. We are paid to bring the aircraft back when things go wrong, and things go wrong more often than the general public realizes.

Yes aerostadt, Boeing has an interest in the pilots flying their airplanes. But they SELL aircraft for profit. They want to sell aircraft. They also sell training programs for the pilots but they can't create qualified pilots. The demand for qualified pilots is far exceeding the supply.

The fact that to get the education needed to fly an airliner a pilot needs to spend about $250,000 for flight time and ratings or go the military route limits the supply.

There is much more I could post, but I have to go to work. Flying JFK to SFO today.
 
It would be interesting to know the details of how the 787 would handle this problem. I suspect it would check both AoA sensors, and very thoroughly. It has an "Earth Reference System" that melds all the sensor inputs, including pitot static ports, GPS and IMU and works out effectively its location in the world, cross-checking as it goes. It would not surprise me if it checked AoA sensors against each other and also against IMU to determine data veracity. It is a very clever avionics system in its own right. I might have to get out the tech manual again and have a read :).

As mach7 mentioned, there is going to be an enormous shortage of pilots and aircraft technicians in the next 10-20 years. They will need to streamline training and build a far greater of simulators to enable the training. The entire infrastructure is not set up for this currently and ramping to process the number of new staff required for the new airliners will be a logistics and regulatory nightmare. The good news is if you want to become a pilot you are quite likely to land a job! Around 250,000 pilots (yep, quarter of a million :eek:) are needed within ten years. The industry is having to think of new pathways to qualification if they want to meet the numbers successfully. Probably a good industry to be an entrepreneur in!
 
Widebodies always cross check the systems. Narrow bodies typically haven’t.

They need to.
 
Well, as a 737 pilot with thousands of hours flying them. I can say the 737 is a VERY safe airliner.
I don't like to speak ill of the dead, but lack of pilot experience and poor training and decisions are what really the
issue here.

MCAS is a new system for the MAX, but the there is a boldface (memory) procedure that deals with this issue. It has been on every 737 for decades. Every pilot trains for it. It's not complicated and would solve the issue in short order.

While it's true that Boeing did not have the right, left, and standby flight instrument systems do not talk to each other, it does give an AOA disagree warning on the cockpit displays.
The aircraft has 3 independent flight instrumentation systems. Its basic training to quickly crosscheck the 3 systems and determine the "odd man out" and act accordingly.

As a professional pilot we are required to be able to fly our aircraft with a major system failed. This is a minor system.
I understand the startle factor, but basic airmanship is expected.

Pitch, power, airspeed. Words to live by in aviation. Set an engine power. Set a pitch angle. An airspeed will follow.
On a 737 its 10 deg nose up with the flaps out and 80% engine power. Set that and you will be safe and climbing at a reasonable rate.
Every pilot should know them by memory.

The Lion Air aircraft had multiple failures of the MCAS system in the days prior. It's criminal that the previous aircrews did not report it to maintenance for repair. The accident crew NEVER did the runaway stabilizer memory procedure.
A simple 3 step procedure that would have solved the problem in short order.

Ethiopian air suffered a bird strike on the runway that caused the AOA vane to separate from the aircraft.
The crew started the Runaway stabilizer procedure, but never finished it and ended up reversing their actions and reengaging the stab trim system. To compound the issue, they left the engines at takeoff thrust and accelerated to 340 knots making manual trimming almost impossible.

The bottom line is I would have no problem putting my family on a 737 MAX tomorrow with 2 qualified pilots.

I have flown the simulator with the stickshaker going off continuously due to a failed sensor while manually trimming the aircraft. Its not fun, but It can be done and done well.

The news media like a villain, Boeing will take some blame. There is no reason for the left and right flight instruments not to talk to each other. It's an easy fix and what is coming, along with limiting the nose down trim authority of MCAS.

The real issue here is qualified, experienced pilots. Many foreign airlines in developing countries don't have the supply of pilots available to them. In the past they would hire qualified pilots from developed countries, but that supply has dried up. The shortage of pilots is very real. Qualified/experienced pilots are leaving the smaller airlines to work at major airlines all over the world because the money/working conditions are so much better.

I'm not sure I agree with you on the pilot experience. The Lion Air pilot had 6000 hours of flight experience (not clear how many in 737s) and the Ethiopian pilot had 8100 hours, including 1400 hours in 737s. The crash investigators say that the pilots completely followed the MCAS shutoff procedures, and Boeing says that they didn't. I recognize that both of those sides have interests that may sway what is said, but Boeing is also not saying what the pilots didn't do. If you have insight into that, I'd appreciate it.

The reporting I've seen on the AOA disagree alert says that the alert was effectively disabled unless optional features were bought by the airline. American Airlines appears to have had a functional alert, and Southwest didn't though they thought they did. Again, I'd be interested in more info on that if you have better information. Regardless of safety features purchased, there appeared to be no light to alert for a malfunctioning MCAS.

I don't think Boeing has helped themselves here. They've kept to the party line that there's nothing wrong with the original design and training, and that MCAS worked exactly as it was supposed to. I know that Boeing doesn't want to crash planes, but that's the natural response when you hear execs saying it all worked according to plan.
 
1st, you are correct. Boeing is not helping themselves with this issue.

2nd, The AOA disagree warning IS on every comercial airliner sold. I believe it is a certification requirement.

3rd. The confusion is the AOA display. That is an option. American has it on the primary flight display. My airline has it in the HUD. The AOA display is a gauge that continuously shows the AOA of the aircraft. The AOA disagree warning only shows when there is a significant difference between the Captains and F/Os instruments.

4th. I have read the actual preliminary accident report. The Ethiopian pilots started the runaway stabilizer procedure,
never completed it, and then reversed the procedure. Lion air never started it.

5th. There is no need for a separate MCAS alert. It works through the existing speed trim system that has been on 737s for over 30 years. The actions taken are exactly the same.

Pilot experience is an interesting thing. My airline won't hire a first officer unless they have around 5000 hours, that is down from 8000 hours a few years ago. The exception is Military pilots, they get hired with less time because of the outstanding training. Don't confuse minimum flight time for application with who actually gets hired.
In the US the FAA says commuter pilots need a minimum of 1500 hours to sit in the right seat on a the small feeder airlines. It's rare for a Captain at a US airline to have less than 10,000 hours.
Also figure in the quality of training and the quality of the flight hours. That can vary widely.
 
I am somewhat mystified by the whole MCAS design but perhaps the description is lacking.

The problem with the MAX was that due to the new engine location they would cause a pitch up in some conditions. The MCAS system that pushes the nose down in response to angle of attach does nothing about that at all. If the goal was, as has been stated, to keep the plane as similar as possible to previous models, the system should use the trim to compensate for that pitch up tendency. Then let the pilots (along with the usual stall warning/stick shaker) fly the plane. Properly done the pilots would never notice the pitch up tendency and the plane would behave just like any other 737.

The amount of trim would be limited with none of that stupid reset after a few seconds ratchet into uncontrollability.

I expect that this incident is going to be a part of engineering ethics courses as a case study for years to come.
 
1st, you are correct. Boeing is not helping themselves with this issue.

2nd, The AOA disagree warning IS on every comercial airliner sold. I believe it is a certification requirement.

3rd. The confusion is the AOA display. That is an option. American has it on the primary flight display. My airline has it in the HUD. The AOA display is a gauge that continuously shows the AOA of the aircraft. The AOA disagree warning only shows when there is a significant difference between the Captains and F/Os instruments.

4th. I have read the actual preliminary accident report. The Ethiopian pilots started the runaway stabilizer procedure,
never completed it, and then reversed the procedure. Lion air never started it.

5th. There is no need for a separate MCAS alert. It works through the existing speed trim system that has been on 737s for over 30 years. The actions taken are exactly the same.

Pilot experience is an interesting thing. My airline won't hire a first officer unless they have around 5000 hours, that is down from 8000 hours a few years ago. The exception is Military pilots, they get hired with less time because of the outstanding training. Don't confuse minimum flight time for application with who actually gets hired.
In the US the FAA says commuter pilots need a minimum of 1500 hours to sit in the right seat on a the small feeder airlines. It's rare for a Captain at a US airline to have less than 10,000 hours.
Also figure in the quality of training and the quality of the flight hours. That can vary widely.

Thank you for the expertise. 2 doesn't match what I'm reading in this Boeing press release.

The Boeing design requirements for the 737 MAX included the AOA Disagree alert as a standard, standalone feature, in keeping with Boeing’s fundamental design philosophy of retaining commonality with the 737NG. In 2017, within several months after beginning 737 MAX deliveries, engineers at Boeing identified that the 737 MAX display system software did not correctly meet the AOA Disagree alert requirements. The software delivered to Boeing linked the AOA Disagree alert to the AOA indicator, which is an optional feature on the MAX and the NG. Accordingly, the software activated the AOA Disagree alert only if an airline opted for the AOA indicator.

To me, that indicates that due to a software bug, AOA Disagree would not be "illuminated"* unless the optional AOA Indicator package was also installed. Is that how you read it as well? I'm not trying to be argumentative, just trying to understand the issue.

* Illuminated sounds like a light comes on a la Apollo command modules, but this one just pops text up on the screen, again as I understand.
 
2nd, The AOA disagree warning IS on every commercial airliner sold. I believe it is a certification requirement.
Boeing have stated that they found the disagree warning was not functioning on aircraft that were fitted with it if the optional AoA indicator function wasn't purchased. Boeing appeared to be surprised when they delved into the problem, tested and found this. Read about it in quite a few articles.

The MCAS system that pushes the nose down in response to angle of attach does nothing about that at all. If the goal was, as has been stated, to keep the plane as similar as possible to previous models, the system should use the trim to compensate for that pitch up tendency.
Pushing the nose down will counter a high AoA, and the MCAS actually does use the trim to push the nose down.

The amount of trim would be limited with none of that stupid reset after a few seconds ratchet into uncontrollability.
The amount of trim it was originally thought to need (by design), I think, was 0.7degrees. Flight testing showed that 2.5degrees was needed for the feature to work as required, so that change was made. The 737 has a total of 6degrees maximum elevator trim available. The final implementation of the software adds another 2.5 degrees of trim EACH TIME IT WAS ACTIVATED, rather than the intent of only having 2.5degrees of authority total. Even a small blip on the trim switch accidentally gives another 2.5degrees authority to the MCAS. Three shots and your trim is pegged full down.
 
Two conflicting sentences here. Pushing the nose down will counter a high AoA, and the MCAS actually does use the trim to push the nose down.
A tendency to pitch up is not the same as a high AoA.
 
A tendency to pitch up is not the same as a high AoA.

Potentially true, however with the MAX aircraft, control forces were found to decrease as the aircraft's angle of attack increased toward a stall. This is not a certifiable flight characteristic, which necessitated the MCAS function.
 
Potentially true, however with the MAX aircraft, control forces were found to decrease as the aircraft's angle of attack increased toward a stall.
I haven't read that anywhere else, and it entirely makes sense as to why the wanted the MCAS.

I think the 737 has exceeded its life as far as big engine options. Going bigger has its limits. A bit like the spitfire during WW2. Bigger engines when they wanted better performance, but in the end it had some undesirable flight characteristics with the supercharged Merlin engine and five bladed prop (IIRC).
 
If Boeing is saying that a software glitch has caused the AOA disagree warning to not show up, thats news to me. There is nothing in our systems manual about that, but I would guess Boeing would know. My airline has the AOA indicator.
But...... It does not absolve the pilots from flying the aircraft and dealing with a runaway trim situation.

The AOA disagree is just text on the primary flight display.


If you want to get inside baseball here,

ALL aircraft with mid or low wings and the engines under the wings have a pitch/power response due to the engine thrust line being below the C/G of the aircraft. Adding power increases the pitch angle, It's not hard at all to counter.

Fly by wire aircraft deal with this automatically through the flight control computers with no pilot input. The 737's with the CFM engine deal with it through the speed trim system. It is a system that tries to keep the aircraft in trim at high power settings. It is annoying and I do not like it but it is a fact of life with this design.

With the bigger engines on the MAX they had to move the engine forward and up so it would clear the ground. A new nacelle was designed. The new nacelle actually generates lift, this coupled with the higher thrust can create a greater nose up moment. MCAS was designed to counter this. It's obvious at this point that the fix was not a good one. But it does not alter the fact that both accident aircraft should not have crashed because of the failure.

Perhaps the MAX should have had fly by wire for the elevator, but the nose up moment was discovered during flight test and a major redesign of the pitch control system was out of the question. Adding FBW to the elevator would probably voided the original FAA certificate and maybe the common type rating. I'm not an expert in that.
 
Back
Top