10 ms·

Does anyone actually know or suspect what's wrong with these planes? Hardware issue? Software issue?

sandworm101 8y ago
All three. It also has to do with pilot training. The general idea is that there is an auto-trim system meant to stop the plane from stalling. But when it gets bad data from a faulty sensor it tries to crash the aircraft (short version). Pilots, all pilots, are trained to recognize this and override the system, but this aircraft requires them to do some things slightly differently. Specifically, they have to shut down the system rather than manually work against it. Difficulties arise where there is a disconnect between what the pilots think is happening, what the systems think and tell them is happening, and what the aircraft is actually experiencing. So this is an interaction between an automated system (software) a potentially faulty sensor (hardware) and pilot training. It is a complex problem that will take a while to fully understand and solve.
- notfromhere 8y ago
  It's a simple problem: don't dodge training requirements by falsely stating that there's no need to retrain pilots on this aircraft. These crashes happened because Boeing tried to market the plane as not requiring retraining in order to boost sales to airlines.
- SmellyGeekBoy 8y ago
  This is a very succint explanation of what's happening here (in the Lion Air case, anyway). Not sure why you're being downvoted - perhaps the claim that this is a "complex problem".
  - sandworm101 8y ago
    HN is slowly walking towards facebook culture. People up/down vote things they like or dislike, with the goal of disappearing opinions with which they even mildly disagree. Valid points and thoughtful discussion are increasingly unwelcome. Lol. Thanks for the downvote all below. Thanks for proving my point about the change of behavior.
    - shaki-dora 8y ago
      Facebook doesn’t even have a downvote option.
      - Sohcahtoa82 8y ago
        It does on some pages' comments. It's a feature they're slowly rolling out.
- dingaling 8y ago
  One amelioration would have beem a simple "MCAS ACTIVE" announcement every time it activated. That would have eliminated a lot of checklisting and guesswork in moments of high stress. "What the hell is happening?" "MCAS ACTIVE" "Flip the cutouts!" Why that wasn't mandated by the FAA I have no idea. Instead the pilots are expected to systematically analyse the options whilst trying to stay airborne.
- linuxftw 8y ago
  How do we know the software works in all conditions, as described? We don't know it had anything to do with pilot training until the investigation is actually finished. To say it's pilot training is pushing Boeing's narrative.
anaphor 8y ago
The engines are larger than previous models and had to be moved forward, which causes it to (potentially) stall, so they implemented an anti-stalling mechanism called MCAS which relies on a particular sensor, which can malfunction potentially. They also didn't inform pilots that this system even existed which means they have trouble diagnosing the problem and will potentially only make it worse by trying other things. Of course that's all speculation since we don't know much about this particular crash, but that's the main issue with them. Source: https://www.youtube.com/watch?v=zfQW0upkVus https://www.youtube.com/watch?v=zfQW0upkVus
- foepys 8y ago
  Didn't Boeing also market the plane as an evolution of the 737 that doesn't require retraining? So pilots with experience on the 737 can automatically fly the 737 Max?
  - alistairSH 8y ago
    Correct.
  - notfromhere 8y ago
    Yep, that was the big selling point since airlines hate retraining pilots. Totally backfired in that regard though.
    - cjrp 8y ago
      Also if an airline is going to have to retrain pilots, they might also look at completely different manufacturers (e.g. Airbus) and play them off against each other to get a better deal.
    - w0utert 8y ago
      It sounds really stupid for Boeing if you put it that way. I can imagine it looks good on the marketing material, 'no pilot retraining required!', but as far as I understand from all the analysis so far, it's actually not that hard to disable the new MCAS system and prevent a crash. As a pilot you only need to know it is there, and what happens if it somehow fails. I would be surprised if they had sold even a single plane less if they advertised it as 'very minimal pilot retraining necessary'.
rbanffy 8y ago
Boeing is issuing a software update in April. https://boeing.mediaroom.com/news-releases-statements?item=130402 https://boeing.mediaroom.com/news-releases-statements?item=1...
- mikro2nd 8y ago
  A software update that was scheduled for distribution (iirc) in January, so already late. As a software practitioner, this scares the living shit out of me. Already late software, now pushed for a deadline-driven release... This can't end well.
  - rbanffy 8y ago
    I would think the software is relatively simple and that the fix, as it's described, is more smoothing inputs and limiting outputs. Or course the combined software that flies a plane is enormous and incredibly complicated, but I'm talking about a relatively small component of that whole. The program reads sensors and actuates surfaces to correct what it incorrectly registered as an abnormally high angle of attack but was, in reality, a misbehaving sensor. The new program, from what's described, tries to figure out the right value even if the sensor disagrees, and limits the output to avoid large corrections. Along the programs, they also need to write new documentation, operating and maintenance manuals, issue corrections for flight simulators and so on. It's also not the first plane to crash because of software reading bad data from bad sensors. Or the last.
swarnie_ 8y ago
The speculation I've seen is around a change to the autopilot system which pilots ave been struggling with: > Following the Lion Air crash, US aviation authorities issued an emergency directive to carriers to update flight manuals with information on what to do when the aircraft’s anti-stall system is triggered by erroneous data from what’s called an “angle-of-attack” sensor. The flight system can react to that data by pointing the plane’s nose sharply downward. Boeing, meanwhile, directed airlines to a checklist in manuals for stabilizing the aircraft. Pilots said the crash and the directives that followed were the first time that they were made aware of these changes to the flight system.
- DannyBee 8y ago
  Traditionally the fly-by-wire systems have at least 3 AoA sensors, which each "vote" on the output value. That way, if you have one faulty sensor, it gets outvoted. The MCAS system in the new 737-MAX's only have 2 AoA sensors, which means a single faulty sensor can cause bad things to happen.
  - linuxftw 8y ago
    > The MCAS system in the new 737-MAX's only have 2 AoA sensors That's pretty damning if true.
    - DannyBee 8y ago
      This is what i was told (it was something like "they have 2 flight computers, and 1 AoA sensor hooked up to each"). The reddit thread linked from elsewhere says the same thing. The preliminary report on the Lion air crash (http://knkt.dephub.go.id/knkt/ntsc_aviation/baru/pre/2018/2018%20-%20035%20-%20PK-LQP%20Preliminary%20Report.pdf http://knkt.dephub.go.id/knkt/ntsc_aviation/baru/pre/2018/20...) says they replaced and tested the AoA sensor 2 days prior. They repeatedly refer to the AoA sensor using only singular nouns everywhere. Given the report says it was giving faulty data and that was used, the only way this can occur is either: 1. Their are multiple sensors and the software is broken 2. There is a single sensor Given all other data, i'm going with #2 :) Nobody has been enterprising enough to publish the MAX8 troubleshooting/etc manuals online, if they did, you could easily verify this.
nck4222 8y ago
The more I read about flight systems and protocols (and I am absolutely a layman when it comes to this), the more it seems like it's very rarely solely isolated to one component. The hardware, software, and human systems are so intertwined that it likely involves all 3, even if the route cause can be isolated to one. That being said, there hasn't been much specific information about the cause released yet, that I've heard.
- linuxftw 8y ago
  Problem is, if they've added so much automation that trained pilots cannot determination the proper course of action 100% of the time, the system is at fault. "You don't have to do anything, the plane will fly itself. Unless there's a catastrophic emergency. Then you better remember everything you haven't practiced from 18 months ago" seems like a failed implementation.
  - ConceitedCode 8y ago
    That's definitely a mischaracterization of what the airlines do. Anyone that has been in a cockpit of a plane knows that you fly by checklists. There's a checklist procedure for almost any scenario they will run into (of course not every). This exact issue was seen by other airlines and the pilots followed the checklist procedures to safely regain control of the plane as expected. In theory, these checklists are optimized to resolve these issues and regain control as quickly as possible while ruling out other causes. It is very rare the correct course of action for the pilot differs from the checklist procedure. There is 0 expectation that the pilot should remember everything. Pilots are trained specifically to communicate with each other to go through these checklists as quickly as possible. That being said, there is a major concern that this issue will popup while taking off and being too low to the ground to properly follow procedure in time to recover control of the aircraft.
    - ReptileMan 8y ago
      The problem is information overload to pilots. And the pipeline how commercial pilots are trained also changed. 50 years ago a lot of pilots were having military backgrounds and training. So they were having more experience with shit going down the drain situations.
jimbru 8y ago
There's speculation that the crash was caused or contributed to by the MCAS system, which is new on the MAX 8. At this point it is just speculation. Here's a good answer from Reddit: https://www.reddit.com/r/aviation/comments/azzp0r/ethiopian_airlines_et302_and_boeing_737_max_8/eibg1sj/ https://www.reddit.com/r/aviation/comments/azzp0r/ethiopian_...
- nas 8y ago
  So much speculation is going on. The angry mob has already concluded that the MCAS must be the cause. Greedy Boeing was too cheap to redesign the plane and made a cheap software hack to fix it. Sneaky computers are overriding the decisions of the poor helpless pilots. The FAA is too incompetent to certify aircraft, etc etc. Based on what I read, the truth is a lot more complicated. The MCAS doesn't work the way most people seem to think it does. Maybe it is a factor in the crashes. We don't know that yet.
  - bambax 8y ago
    We don't know much, if anything, about the Ethiopian crash; but we do know that in the case of Lion air, MCAS was a major factor.
adwww 8y ago
Also training issue - according to a commentator on Radio 4 this morning, the conversion training for this aircraft is a 90 minute course run on an iPad.
BurningFrog 8y ago
Let me also enter the impopular Null Theory: Maybe there is nothing in particular wrong with these planes, and they were just hit with two random accidents.
- maxxxxx 8y ago
  . Maybe this turns out to be true but right now you can't make decisions based on that.
  - Xixi 8y ago
    Indeed: in the aviation world the philosophy is that things have to be proven safe [1] to be allowed to fly, rather than proven unsafe to be banned. In that regard, two similar crashes a few months apart is enough to worry of a design flaw. And if it was just two unrelated pilot errors, the planes will be allowed to fly again quickly. [1] To a reasonable extent, of course, nothing can ever be proven 100% safe.
  - BurningFrog 8y ago
    Yeah, I'm not saying anyone should. Just a reminder of the possibility.
- stevehawk 8y ago
  Malaysian Airlines would say that's always a possibility https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_17 https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_17 https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_370 https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_370 2 Boeing 777-200ERs down within three months of each other (one was shot down by Russia and the other is speculation)
  - akhilcacharya 8y ago
    The fact that the safety record of the 777 was near perfect before 2014 honestly makes me more concerned about what happened to MH370 if it's an inherent fault.
  - 8y ago

All three. It also has to do with pilot training. The general idea is that there is an auto-trim system meant to stop the plane from stalling. But when it gets bad data from a faulty sensor it tries to crash the aircraft (short version). Pilots, all pilots, are trained to recognize this and override the system, but this aircraft requires them to do some things slightly differently. Specifically, they have to shut down the system rather than manually work against it. Difficulties arise where there is a disconnect between what the pilots think is happening, what the systems think and tell them is happening, and what the aircraft is actually experiencing. So this is an interaction between an automated system (software) a potentially faulty sensor (hardware) and pilot training. It is a complex problem that will take a while to fully understand and solve.

notfromhere 8y ago
It's a simple problem: don't dodge training requirements by falsely stating that there's no need to retrain pilots on this aircraft. These crashes happened because Boeing tried to market the plane as not requiring retraining in order to boost sales to airlines.
SmellyGeekBoy 8y ago
This is a very succint explanation of what's happening here (in the Lion Air case, anyway). Not sure why you're being downvoted - perhaps the claim that this is a "complex problem".
- sandworm101 8y ago
  HN is slowly walking towards facebook culture. People up/down vote things they like or dislike, with the goal of disappearing opinions with which they even mildly disagree. Valid points and thoughtful discussion are increasingly unwelcome. Lol. Thanks for the downvote all below. Thanks for proving my point about the change of behavior.
  - shaki-dora 8y ago
    Facebook doesn’t even have a downvote option.
    - Sohcahtoa82 8y ago
      It does on some pages' comments. It's a feature they're slowly rolling out.
dingaling 8y ago
One amelioration would have beem a simple "MCAS ACTIVE" announcement every time it activated. That would have eliminated a lot of checklisting and guesswork in moments of high stress. "What the hell is happening?" "MCAS ACTIVE" "Flip the cutouts!" Why that wasn't mandated by the FAA I have no idea. Instead the pilots are expected to systematically analyse the options whilst trying to stay airborne.
linuxftw 8y ago
How do we know the software works in all conditions, as described? We don't know it had anything to do with pilot training until the investigation is actually finished. To say it's pilot training is pushing Boeing's narrative.

anaphor 8y ago

The engines are larger than previous models and had to be moved forward, which causes it to (potentially) stall, so they implemented an anti-stalling mechanism called MCAS which relies on a particular sensor, which can malfunction potentially. They also didn't inform pilots that this system even existed which means they have trouble diagnosing the problem and will potentially only make it worse by trying other things. Of course that's all speculation since we don't know much about this particular crash, but that's the main issue with them. Source: https://www.youtube.com/watch?v=zfQW0upkVus https://www.youtube.com/watch?v=zfQW0upkVus

foepys 8y ago
Didn't Boeing also market the plane as an evolution of the 737 that doesn't require retraining? So pilots with experience on the 737 can automatically fly the 737 Max?
- alistairSH 8y ago
  Correct.
- notfromhere 8y ago
  Yep, that was the big selling point since airlines hate retraining pilots. Totally backfired in that regard though.
  - cjrp 8y ago
    Also if an airline is going to have to retrain pilots, they might also look at completely different manufacturers (e.g. Airbus) and play them off against each other to get a better deal.
  - w0utert 8y ago
    It sounds really stupid for Boeing if you put it that way. I can imagine it looks good on the marketing material, 'no pilot retraining required!', but as far as I understand from all the analysis so far, it's actually not that hard to disable the new MCAS system and prevent a crash. As a pilot you only need to know it is there, and what happens if it somehow fails. I would be surprised if they had sold even a single plane less if they advertised it as 'very minimal pilot retraining necessary'.

rbanffy 8y ago

Boeing is issuing a software update in April. https://boeing.mediaroom.com/news-releases-statements?item=130402 https://boeing.mediaroom.com/news-releases-statements?item=1...

mikro2nd 8y ago
A software update that was scheduled for distribution (iirc) in January, so already late. As a software practitioner, this scares the living shit out of me. Already late software, now pushed for a deadline-driven release... This can't end well.
- rbanffy 8y ago
  I would think the software is relatively simple and that the fix, as it's described, is more smoothing inputs and limiting outputs. Or course the combined software that flies a plane is enormous and incredibly complicated, but I'm talking about a relatively small component of that whole. The program reads sensors and actuates surfaces to correct what it incorrectly registered as an abnormally high angle of attack but was, in reality, a misbehaving sensor. The new program, from what's described, tries to figure out the right value even if the sensor disagrees, and limits the output to avoid large corrections. Along the programs, they also need to write new documentation, operating and maintenance manuals, issue corrections for flight simulators and so on. It's also not the first plane to crash because of software reading bad data from bad sensors. Or the last.

swarnie_ 8y ago

The speculation I've seen is around a change to the autopilot system which pilots ave been struggling with: > Following the Lion Air crash, US aviation authorities issued an emergency directive to carriers to update flight manuals with information on what to do when the aircraft’s anti-stall system is triggered by erroneous data from what’s called an “angle-of-attack” sensor. The flight system can react to that data by pointing the plane’s nose sharply downward. Boeing, meanwhile, directed airlines to a checklist in manuals for stabilizing the aircraft. Pilots said the crash and the directives that followed were the first time that they were made aware of these changes to the flight system.

DannyBee 8y ago
Traditionally the fly-by-wire systems have at least 3 AoA sensors, which each "vote" on the output value. That way, if you have one faulty sensor, it gets outvoted. The MCAS system in the new 737-MAX's only have 2 AoA sensors, which means a single faulty sensor can cause bad things to happen.
- linuxftw 8y ago
  > The MCAS system in the new 737-MAX's only have 2 AoA sensors That's pretty damning if true.
  - DannyBee 8y ago
    This is what i was told (it was something like "they have 2 flight computers, and 1 AoA sensor hooked up to each"). The reddit thread linked from elsewhere says the same thing. The preliminary report on the Lion air crash (http://knkt.dephub.go.id/knkt/ntsc_aviation/baru/pre/2018/2018%20-%20035%20-%20PK-LQP%20Preliminary%20Report.pdf http://knkt.dephub.go.id/knkt/ntsc_aviation/baru/pre/2018/20...) says they replaced and tested the AoA sensor 2 days prior. They repeatedly refer to the AoA sensor using only singular nouns everywhere. Given the report says it was giving faulty data and that was used, the only way this can occur is either: 1. Their are multiple sensors and the software is broken 2. There is a single sensor Given all other data, i'm going with #2 :) Nobody has been enterprising enough to publish the MAX8 troubleshooting/etc manuals online, if they did, you could easily verify this.

nck4222 8y ago

The more I read about flight systems and protocols (and I am absolutely a layman when it comes to this), the more it seems like it's very rarely solely isolated to one component. The hardware, software, and human systems are so intertwined that it likely involves all 3, even if the route cause can be isolated to one. That being said, there hasn't been much specific information about the cause released yet, that I've heard.

linuxftw 8y ago
Problem is, if they've added so much automation that trained pilots cannot determination the proper course of action 100% of the time, the system is at fault. "You don't have to do anything, the plane will fly itself. Unless there's a catastrophic emergency. Then you better remember everything you haven't practiced from 18 months ago" seems like a failed implementation.
- ConceitedCode 8y ago
  That's definitely a mischaracterization of what the airlines do. Anyone that has been in a cockpit of a plane knows that you fly by checklists. There's a checklist procedure for almost any scenario they will run into (of course not every). This exact issue was seen by other airlines and the pilots followed the checklist procedures to safely regain control of the plane as expected. In theory, these checklists are optimized to resolve these issues and regain control as quickly as possible while ruling out other causes. It is very rare the correct course of action for the pilot differs from the checklist procedure. There is 0 expectation that the pilot should remember everything. Pilots are trained specifically to communicate with each other to go through these checklists as quickly as possible. That being said, there is a major concern that this issue will popup while taking off and being too low to the ground to properly follow procedure in time to recover control of the aircraft.
  - ReptileMan 8y ago
    The problem is information overload to pilots. And the pipeline how commercial pilots are trained also changed. 50 years ago a lot of pilots were having military backgrounds and training. So they were having more experience with shit going down the drain situations.

jimbru 8y ago

There's speculation that the crash was caused or contributed to by the MCAS system, which is new on the MAX 8. At this point it is just speculation. Here's a good answer from Reddit: https://www.reddit.com/r/aviation/comments/azzp0r/ethiopian_airlines_et302_and_boeing_737_max_8/eibg1sj/ https://www.reddit.com/r/aviation/comments/azzp0r/ethiopian_...

nas 8y ago
So much speculation is going on. The angry mob has already concluded that the MCAS must be the cause. Greedy Boeing was too cheap to redesign the plane and made a cheap software hack to fix it. Sneaky computers are overriding the decisions of the poor helpless pilots. The FAA is too incompetent to certify aircraft, etc etc. Based on what I read, the truth is a lot more complicated. The MCAS doesn't work the way most people seem to think it does. Maybe it is a factor in the crashes. We don't know that yet.
- bambax 8y ago
  We don't know much, if anything, about the Ethiopian crash; but we do know that in the case of Lion air, MCAS was a major factor.

adwww 8y ago

Also training issue - according to a commentator on Radio 4 this morning, the conversion training for this aircraft is a 90 minute course run on an iPad.

BurningFrog 8y ago

Let me also enter the impopular Null Theory: Maybe there is nothing in particular wrong with these planes, and they were just hit with two random accidents.

maxxxxx 8y ago
. Maybe this turns out to be true but right now you can't make decisions based on that.
- Xixi 8y ago
  Indeed: in the aviation world the philosophy is that things have to be proven safe [1] to be allowed to fly, rather than proven unsafe to be banned. In that regard, two similar crashes a few months apart is enough to worry of a design flaw. And if it was just two unrelated pilot errors, the planes will be allowed to fly again quickly. [1] To a reasonable extent, of course, nothing can ever be proven 100% safe.
- BurningFrog 8y ago
  Yeah, I'm not saying anyone should. Just a reminder of the possibility.
stevehawk 8y ago
Malaysian Airlines would say that's always a possibility https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_17 https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_17 https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_370 https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_370 2 Boeing 777-200ERs down within three months of each other (one was shot down by Russia and the other is speculation)
- akhilcacharya 8y ago
  The fact that the safety record of the 777 was near perfect before 2014 honestly makes me more concerned about what happened to MH370 if it's an inherent fault.
- 8y ago