Ethiopian plane crash

[PRTVR]

surveyor said:

Octoposse said:

Paraglider pilot here (and a rusty one at that), so I'm clearly not an expert on modern airliners . . .

. . . but puzzled that Pitch / Attitude isn't as significant an input as AoA? I understand that Pitch and AoA are never(?) going to be the same, and that it's AoA that matters in terms of lift vs drag, BUT surely there's a close relationship (with airspeed as a variable)?

Difficult to control pitch when airplane doing it's best to throw itself into the ground pointy end first.

I think he means before it gets to that point, if airspeed increases should the AOA not have a relationship to it, a comparator between say speed altitude and AOA to keep the aircraft safe.

[768]

PRTVR said:

I think he means before it gets to that point, if airspeed increases should the AOA not have a relationship to it, a comparator between say speed altitude and AOA to keep the aircraft safe.

How long before? What counts as before if MCAS has been turned off and on 20 times while they're fighting it?

Some of these suggestions sound simple, but they're not insignificant increases in complexity which is usually a bad idea even in systems that aren't safety critical. It probably just needs (connected) redundant sensors, some better handling of out of range values and a great big off switch that can keep it off in all circumstances.

[PRTVR]

768 said:

PRTVR said:

I think he means before it gets to that point, if airspeed increases should the AOA not have a relationship to it, a comparator between say speed altitude and AOA to keep the aircraft safe.

How long before? What counts as before if MCAS has been turned off and on 20 times while they're fighting it?

Some of these suggestions sound simple, but they're not insignificant increases in complexity which is usually a bad idea even in systems that aren't safety critical. It probably just needs (connected) redundant sensors, some better handling of out of range values and a great big off switch that can keep it off in all circumstances.

I agree about keeping it simple, but if the system works you should not have gotten to the point of switching off and on and be fighting the system.

[tobinen]

More or less what will happen according the latest bulletin

Use both AOA sensors, not just one (crazy idea!)
Cross check values between them
Limit MCAS intervention or system cut-out if above values are not within parameters

[Octoposse]

768 said:

PRTVR said:

I think he means before it gets to that point, if airspeed increases should the AOA not have a relationship to it, a comparator between say speed altitude and AOA to keep the aircraft safe.

How long before? What counts as before if MCAS has been turned off and on 20 times while they're fighting it?

It was more a theoretical point than a practical one - that all the discussion has centred on AoA, yet what pilots actually 'see' and 'feel' are pitch and stall point.

And AoA and pitch are indirectly linked by airspeed within a range of possibilities unique to that aerofoil? The two will only be way, way, different if you're already in full stall?

[M4cruiser]

hutchst said:

It's a huge leap from negligence to criminal intent. I think they're more likely to see the inside of the big house if they try to cover it up (pervert the course of justice in our terms)

But the costs to Boeing are going to be huge. Tens of billions. Damages, rectification costs, lost production, cancellations and lost sales.

I was just browsing the financial settlement paragraph (on Wiki) from Swissair 111. Same type of crash (i.e. high speed nose down) and caused by a deficiency in the aircraft's manufacture:-

https://en.wikipedia.org/wiki/Swissair_Flight_111

So, if I read that correctly, there was no pay-out. Let's hope something similar doesn't happen in this Ethiopian case.

Edited by M4cruiser on Thursday 18th April 13:15

[CrutyRammers]

Octoposse said:

It was more a theoretical point than a practical one - that all the discussion has centred on AoA, yet what pilots actually 'see' and 'feel' are pitch and stall point.

And AoA and pitch are indirectly linked by airspeed within a range of possibilities unique to that aerofoil? The two will only be way, way, different if you're already in full stall?

I think I get what you're saying.
Stall of an aerofoil is due to AoA; the variables which the pilot can use to control AoA are (largely) speed and pitch. In normal flight, pitch, AoA and speed will all be within a defined envelope; pitch will not vary from AoA that much, speed will be within a defined range for each value of pitch.
So in theory, it would be possible to sanity check the AoA reading against the pitch and airspeed readings; if it's outside the norms, you're either already stalled, or one of the sensors is talking rubbish.

That right?

MY guess for why not would be, maybe it doesn't give any extra safety to do it that way over just having redundant sensors. It would add complexity and hence the possibility for unforeseen effects, since you now have to have a model of the wing for 3 variables across all flight conditions. Whereas if stall is just because of AoA, measuring AoA directly cuts out all of that.

[Octoposse]

CrutyRammers said:

I think I get what you're saying.
Stall of an aerofoil is due to AoA; the variables which the pilot can use to control AoA are (largely) speed and pitch. In normal flight, pitch, AoA and speed will all be within a defined envelope; pitch will not vary from AoA that much, speed will be within a defined range for each value of pitch.
So in theory, it would be possible to sanity check the AoA reading against the pitch and airspeed readings; if it's outside the norms, you're either already stalled, or one of the sensors is talking rubbish.

That right?

Yes - you put it better than me! Only 'aircraft' I am qualified to fly have no instruments or sensors whatsoever apart from an audible altimeter bleeping or whistling in your ear!

[Starfighter]

tobinen said:

More or less what will happen according the latest bulletin

Use both AOA sensors, not just one (crazy idea!)
Cross check values between them
Limit MCAS intervention or system cut-out if above values are not within parameters

Someone posted some risk assessment lists related to MCAS early in this thread which said all of this. It appears that these were system design requirements based on the system failure risk level.
It would appear that Boeing did not do this when they developed MCAS and signed this off as compliant. The FAA did not overcheck the system design against the risk level and allowed Boeing to put the aircraft to market. EAA accepted the FAA approval.

[CrutyRammers]

Octoposse said:

Yes - you put it better than me! Only 'aircraft' I am qualified to fly have no instruments or sensors whatsoever apart from an audible altimeter bleeping or whistling in your ear!

Same as me sir. Aint no feeling like it

[tight fart]

Interesting article
https://spectrum.ieee.org/aerospace/aviation/how-t...

[captain_cynic]

tight fart said:

Interesting article
https://spectrum.ieee.org/aerospace/aviation/how-t...

Found that article yesterday, long read but well worth it.

It highlights that it is a hardware problem they're trying to fix with software. Anyone with any experience running production computing knows that you can't fix hardware with software because the software relies on consistent and accurate results from the hardware. Any hardware issues will continue to produce more unexpected and esoteric errors that the software can't handle, inevitably leading to a system failure... Unfortunately for the MAX, "crash" took on a quite literal meaning when software failed.

I sincerely hope there isn't another fatal accident before Boeing or the FAA realises that.

[dvs_dave]

captain_cynic said:

tight fart said:

Interesting article
https://spectrum.ieee.org/aerospace/aviation/how-t...

Found that article yesterday, long read but well worth it.

It highlights that it is a hardware problem they're trying to fix with software. Anyone with any experience running production computing knows that you can't fix hardware with software because the software relies on consistent and accurate results from the hardware. Any hardware issues will continue to produce more unexpected and esoteric errors that the software can't handle, inevitably leading to a system failure... Unfortunately for the MAX, "crash" took on a quite literal meaning when software failed.

I sincerely hope there isn't another fatal accident before Boeing or the FAA realises that.

Another example of someone not knowing what they’re talking about, but making out they do because they’re a “pilot” (but actually only a PPL single engine one). The fact that he’s talking about the MAX engine thrust line position being the main problem makes him loose all credibility. For the umpteenth time, that’s not the problem! The MAX engine centerline and consequent thrust line is actually slightly higher and closer to the wing than it is on the NG. He only mentions the lift from the nacelles at a high AoA as a secondary issue. No, that is the primary problem and what MCAS was (poorly) designed to prevent.

[frisbee]

dvs_dave said:

captain_cynic said:

tight fart said:

Interesting article
https://spectrum.ieee.org/aerospace/aviation/how-t...

Found that article yesterday, long read but well worth it.

It highlights that it is a hardware problem they're trying to fix with software. Anyone with any experience running production computing knows that you can't fix hardware with software because the software relies on consistent and accurate results from the hardware. Any hardware issues will continue to produce more unexpected and esoteric errors that the software can't handle, inevitably leading to a system failure... Unfortunately for the MAX, "crash" took on a quite literal meaning when software failed.

I sincerely hope there isn't another fatal accident before Boeing or the FAA realises that.

Another example of someone not knowing what they’re talking about, but making out they do because they’re a “pilot” (but actually only a PPL single engine one). The fact that he’s talking about the MAX engine thrust line position being the main problem makes him loose all credibility. For the umpteenth time, that’s not the problem! The MAX engine centerline and consequent thrust line is actually slightly higher and closer to the wing than it is on the NG. He only mentions the lift from the nacelles at a high AoA as a secondary issue. No, that is the primary problem and what MCAS was (poorly) designed to prevent.

It is an impressively poor article. Some choice dribbling nonsense:

"Those lines of code were no doubt created by people at the direction of managers. Neither such coders nor their managers are as in touch with the particular culture and mores of the aviation world as much as the people who are down on the factory floor, riveting wings on, designing control yokes, and fitting landing gears. Those people have decades of institutional memory about what has worked in the past and what has not worked. Software people do not."

[tight fart]

Thought this bit was so true for the times we live in, worthy of its own tread.

"Every time a software update gets pushed to my Tesla, to the Garmin flight computers in my Cessna, to my Nest thermostat, and to the TVs in my house, I’m reminded that none of those things were complete when they left the factory—because their builders realized they didn’t have to be complete. The job could be done at any time in the future with a software update."

[lost in espace]

Is the real problem that Boeing can remove MCAS altogether but you have to train thousands of pilots to fly the Max?

[MikeyC]

tight fart said:

Thought this bit was so true for the times we live in, worthy of its own tread.

"Every time a software update gets pushed to my Tesla, to the Garmin flight computers in my Cessna, to my Nest thermostat, and to the TVs in my house, I’m reminded that none of those things were complete when they left the factory—because their builders realized they didn’t have to be complete. The job could be done at any time in the future with a software update."

How do you know they are fixes to serious problems?
They could actually be enhancements/improvements to many different aspects of the software, I'm guessing there's quite alot in a Tesla !
I do agree to some extent, it's near impossible to write code that can handle unforseen conditions that may happen on rare occasions under rare circumstances
The big concern is the fixes/changes don't introduce new issues

[Munter]

MikeyC said:

How do you know they are fixes to serious problems?
They could actually be enhancements/improvements to many different aspects of the software, I'm guessing there's quite alot in a Tesla !
I do agree to some extent, it's near impossible to write code that can handle unforseen conditions that may happen on rare occasions under rare circumstances

I think the main issue is this wasn't an unforeseen condition. Anybody designing this sort of system thinks about a sensor failure, both "dead" and "wrong" options. Then thinks about the safest way to identify and cope with that. Except in Boeing apparently...

[George Smiley]

Dvs_dave did you give your kids beef burgers during the bse crisis? Are you sponsored by Boeing?

It is a hardware fault
It is. A result of using an airframe years beyond the best before date
It is a result of using software to overcome hardware and that always results in failure
Will you take your family on the first max flight?

[tobinen]

lost in espace said:

Is the real problem that Boeing can remove MCAS altogether but you have to train thousands of pilots to fly the Max?

That's my understanding. It's only there to avoid another type-approval for pilots.