Plane engineers of ATOT, what would have been a better solution than MCAS for the 737 max?

[Chaotic42]

A button to modify the gravitational constant of the universe. You people are amateurs.

 

[crashtech]

Also giving an automated system unlimited horizontal stabilizer and thrust authority was clearly a blunder in hindsight. Correcting an incipient stall needs a light touch, not a ham robot fist.

Oh, and any sensor disagree warning shouldn't be in a list of optional equipment...

 

[JEDIYoda]

Since we're all family here, give us your take on the issue so us laymen can understand more clearly.

I would bet that it is far more complicated and really can`t be put into laymans terms because to do so would open the explanation open to never ending (noob) questions that require a more advanced knowlwdge than the average ATOT member would have at their disposal!

 

[Number1]

Also giving an automated system unlimited horizontal stabilizer and thrust authority was clearly a blunder in hindsight. Correcting an incipient stall needs a light touch, not a ham robot fist.

Oh, and any sensor disagree warning shouldn't be in a list of optional equipment...

The automated system does not have unlimited horizontal stabilizer and thrust authority. It can ba shut off at any time. RECOGNIZING what the problem is IS the problem.

 

[Number1]

 

[JEDIYoda]

The automated system does not have unlimited horizontal stabilizer and thrust authority. It can ba shut off at any time. RECOGNIZING what the problem is IS the problem.

well DUH....

 

[crashtech]

The automated system does not have unlimited horizontal stabilizer and thrust authority. It can ba shut off at any time. RECOGNIZING what the problem is IS the problem.

You're confusing unlimited with absolute or irrevocable. It appears, in fact, that the MCAS system was able to max out the stab trim in full nose down position, without predefined limitations.

 

[Kaido]

There's an anonymous Federal database that pilots can lodge complaints into for the FAA; there were multiple complaints about the 737 Max months before the second crash:

https://www.theverge.com/2019/3/13/18263751/boeing-737-max-8-pilot-complaint-autopilot-mcas

 

[SKORPI0]

Is Boeing Canceling the Boeing 737 MAX?

But to fit the new LEAP-1B engines, Boeing had to make some changes to the general design of the 737. To sum up the changes, Boeing added new split tip winglets, a new tail cone, revised APU and exhaust inlet. The Seattle based manufacturer also removed the aft-body vortex generators.

Additionally, Boeing included a higher nose gear and new display screens for the pilots. While these changes don’t seem to be very radical, their eventual impact would be very significant.
The new nose gear put up the aircraft’s nose higher than usual. With the new engines generating a lot more power, the nose went up even more. After conducting various tests, Boeing realized that their new jet is a bit too prone to encountering stalls in certain situations.

 

[herm0016]

Is Boeing Canceling the Boeing 737 MAX?

taller nose gear makes the plane stall in the air when the gear is up..... uh hu, got it.

would love to hear from the expert also. even just some snipits. I have been asked because of my background by some friends but refuse to speculate.

 

[s0me0nesmind1]

The decision to move the engines, change the CG, and use MCAS to compensate was fine. (Although these are still links in the error chain that lead to the failure)

For catastrophic failures system design should generally be two fault tolerant. At the bare minimum one fault tolerant with the second level of fault tolerance coming from human or software intervention.

The main egregious design failure as near as I can tell was assuming that two sensors made the system one fault tolerant. Sure it may work to swap from a failed sensor to a good sensor but this was apparently sensor disagreement. Without three sensors in a voting logic scheme it’s extremely difficult to determine which sensor has drifted and which is still true.

What should have happened was upon the two sensors disagreeing the system should have stopped functioning and/or alerted the pilots to take action.

Of course the disagree light was not included on the base Max8.

As it was the system was 0 fault tolerant to a catastrophic hazard.

Which is the second egregious failure. These hazards should have been identified and they weren’t, in part because the FAA allowed Boeing to perform the oversight function.

I’m sure this whole thing is going to be taught in hazard analysis training for years to come.

Any explanations (or rather, theories) as to why we haven't seen this problem with US-Based airliners yet?

Maybe they paid for the disagree light? Or they are just better trained pilots?

 

[herm0016]

Any explanations (or rather, theories) as to why we haven't seen this problem with US-Based airliners yet?

Maybe they paid for the disagree light? Or they are just better trained pilots?

i'll speculate on that one. both. looks like united did not pay for the upgrades, but claimed to have something else that will spot the problem on their planes, southwest and american both have the upgrades.

remember the ANA crash at SFO? I would venture US pilots are better trained.

 

[Number1]

You're confusing unlimited with absolute or irrevocable. It appears, in fact, that the MCAS system was able to max out the stab trim in full nose down position, without predefined limitations.

Of course it did because the pilots did not understand what was going on. They could have saved the plane easily.
+

 

[brainhulk]

https://www.businessinsider.com/lio...&utm_medium=referral&utm_content=allverticals

An off-duty pilot riding in the cockpit of a Boeing 737 Max 8 fixed a malfunction on the second-to-the-last flight for the aircraft before same plane crashed during a different flight in the Java Sea the next day, Bloomberg reported on Tuesday evening.
The pilot reportedly advised the crew to kill the power to a motor that was pointing the aircraft's nose downward. That move helped prevent a catastrophe, according to Bloomberg.
The aircraft was being operated by a different crew the next day, on October 29, 2018, and crashed fewer than 15 minutes after takeoff, killing all 189 passengers and crew on board.

 

[Ken g6]

Obviously this poll is tongue-in-cheek, but the first option shows your complete lack of understanding of basic airplane characteristics. More thrust increases nose-up pitch and excaberbates stall - the condition trying to be avoided in the first place.

Apparently I also have a complete lack of understanding of basic airplane characteristics, because I don't know why that would happen. Unless the engines are mounted too low under the wings. Hm, maybe the root problem is they couldn't redesign the aircraft to mount the engines higher?

Angle-of-attack sensors seem prone to failure in general - they're very mechanical. What about air pressure sensors on top of the wings to detect stall?

Plus it seems like MCAS shouldn't nose the plane down below level flight for altitudes under, say, 10,000 feet. And there's probably a "maximum safe descent rate" somewhere from which they could extrapolate a maximum nose-down angle in all cases.

Say, isn't there a required "terrain - pull up" system in airliners now, that pulls the nose of the plane up when it's too low to the ground? Does MCAS override that? It shouldn't.

 

[Starbuck1975]

There needs to be an option in the poll for not letting executives make engineering decisions

 

[PowerEngineer]

Admitting that KillerCharlie is right about simplistic answers...

It is my understanding that the desire to maintain the 737's height off the ground while adding larger diameter engines has meant that the position of those engines is much more forward than they were originally. More ahead of the wing than under the wing. I can see how this could cause the nose to want to tip up as power is applied.

I am a bit uncomfortable with design decisions that rely on corrective software rather than actual design changes to mitigate undesirable tendencies.

 

[zinfamous]

A button to modify the gravitational constant of the universe. You people are amateurs.

Hey, there's a dude over in the star wars fan fiction thread trying to figure out "a science" for his scientist, and it's taken him about 4 months to do it. Maybe he could use this idea?

 

[Paratus]

There needs to be an option in the poll for not letting executives make engineering decisions

Easier said than done. The overall project is always coming from management. You will always find an element of budget pressure and schedule pressure in any engineering failure.

Remove management pressure and these kind of failures will almost never happens.

Mostly because no project will ever complete without budget and schedule pressure.

The mitigation for this is an independent or external safety review organization. Normally that’s the FAA. Instead, because all regulations are bad, we let Boeing regulate itself.

 

[Hans Gruber]

If every takeoff was manually done with no autopilot to 5,000ft. There would be no problem. Some of these foreign trained pilots rely too much on pushing buttons and autopilot. How does a plane accelerate to 500mph before crashing? Auto throttle. Obviously there are software and sensor issues but how difficult is it to disengage the autopilot and put your life and the plane in your own hands?

 

[Starbuck1975]

Easier said than done. The overall project is always coming from management. You will always find an element of budget pressure and schedule pressure in any engineering failure.

Remove management pressure and these kind of failures will almost never happens.

Mostly because no project will ever complete without budget and schedule pressure.

The mitigation for this is an independent or external safety review organization. Normally that’s the FAA. Instead, because all regulations are bad, we let Boeing regulate itself.

True, but Boeing also made a deliberate strategic move to focus on shareholder value, and lifted and shifted a lot of work from Puget Sound to Charleston, mostly to circumvent the unions. From a stock standpoint, they’ve been wildly succesful. From an engineering, quality and ethical standpoint...

 

[Artorias]

Is Boeing getting sued yet? This is so disturbing and negligent its unbelievable, quite frankly they shouldn't exist as a company anymore for withholding such key information.

Obliviously aviation is complicated, as a passenger I would like to know that the pilot has 100% control of the plane when they need it, there has to be an easier way to kill automated systems and go full manual control.

 

[Hans Gruber]

Is Boeing getting sued yet? This is so disturbing and negligent its unbelievable, quite frankly they shouldn't exist as a company anymore for withholding such key information.

Obliviously aviation is complicated, as a passenger I would like to know that the pilot has 100% control of the plane when they need it, there has to be an easier way to kill automated systems and go full manual control.

It's called the disengage autopilot switch on the right side of the instrument stack in front of the 1st officer. It is known as a sacred button where a pilot gets their wings. With said button disengaged the stick, rudder and throttle belong solely to the pilot in command.

 

[Paratus]

Any explanations (or rather, theories) as to why we haven't seen this problem with US-Based airliners yet?

Maybe they paid for the disagree light? Or they are just better trained pilots?

Well the sensor doesn’t just have to fail. It has to fail in such away as it causes the downward trim. So even ending up in the situation is the luck of the draw.

Then you are right, airlines who have the added safety features would more easily identify the issue and correct it. Those cases would likely have been classified as maintenance issues if the crashes hadn’t just happened.

Finally, as you said, training. This is another place Boeing looks bad. Because of the rush to beat Airbus to market and the desire to keep customers by making the Max as similar to the regular 737, they downplayed the training required.

Since MCAS, (and other design decisions),made the Max fly like a 737, then training could be reduced to about an hour on a tablet from what some stories have reported.

Now some airlines still requires more training. SouthWest actually has the only Boeing Max simulator. So it’s arguable that their pilots would have a better chance identifying and recovering from the failure.

 

[Hans Gruber]

I have spoken to real pilots. The MCAS feature is filled with flaws. The nose drops down and decends 1500-2000ft per minute. Captain disengages the autopilot and manually flies the plane. Pilot makes note of the issue and contacts airline base and documents the issue. Flight continues as normal. In Indonesia and Africa. Pilots rely exclusively on autopilot fully trusting the autopilot and all features of it. Instead of disengaging the autopilot they change settings in autopilot and choose to try to overpower the autopilot with aggressive stick, rudder and trim maneuvers with the autopilot still engaged. They lose that battle over time causing the deaths of 300+ people and the complete loss of two brand new aircraft.