6 ms·

Airbus aircraft require three AoA sensors because they're considered safety critical. On the 737 Max they weren't safety critical right up until they added MCAS

Airbus aircraft require three AoA sensors because they're considered safety critical. On the 737 Max they weren't safety critical right up until they added MCAS, at which point that changed, and the regulators should have required it but failed to (in part because Boeing mislead regulators about how much MCAS could adjust the deflection in the horizontal stabilizer, doubling it after regulators had already sign off on it[0]). The fact that Boeing is able to skate by with two and a software off-switch if they disagree, while Airbus continues to be required to have triple redundancy says a lot about their respective regulators (EASA Vs FAA). I'm sure "cost" was discussed given the 350+ aircraft already produced and how impractical it is to add a third AoA sensor, but that's literally putting cost over safety. [0] See "Inaccurate limit": https://www.seattletimes.com/business/boeing-aerospace/failed-certification-faa-missed-safety-issues-in-the-737-max-system-implicated-in-the-lion-air-crash/ https://www.seattletimes.com/business/boeing-aerospace/faile...

CaptainZapp 8y ago
but that's literally putting cost over safety. Which seems to be fully in line with Boeing's business model and their behavior after the Lion Air crash. Even after the Ethiopian crash they did just about everything to deflect, blame and obfuscate until this behavior was just no more maintainable based on the known facts. Switching then to a "We apologise if your dead" marketing shtick. I find it hard to believe, based on all the evidence, that this company put lifes before profit.
carlmr 8y ago
>The fact that Boeing is able to skate by with two and a software off-switch if they disagree, while Airbus continues to be required to have triple redundancy says a lot about their respective regulators (EASA Vs FAA). I'm sure "cost" was discussed given the 350+ aircraft already produced and how impractical it is to add a third AoA sensor, but that's literally putting cost over safety. Looking at this image here: https://qz.com/1461400/pilots-werent-informed-of-risks-of-boeing-737-maxs-new-feature/ https://qz.com/1461400/pilots-werent-informed-of-risks-of-bo... My biggest doubt is regarding the symmetrical left-right placement of the AoA vanes. Isn't it realistic that they may freeze in the same position, thus not disagreeing, but still wrong? I'm not in aviation, so I'm not sure about this, but without deeper knowledge this seems like a realistic reason to ask for a 3rd sensor with a different placement at least.
- danaliv 8y ago
  Theoretically it’s possible that both AoA vanes could ice over, but that would mean failure of both vanes’ heating elements. (Not a jet pilot myself but I teach instrument and charter pilots, and I’m in the clouds a lot!)
- cryptonector 8y ago
  That Air France flight that crashed in the Atlantic had all its pitots freeze. No amount of redundancy will help with that sort of situation -- the sensors were not designed correctly, specifically they did not have heaters with enough power to keep them not frozen. While Boeing should have to add a third AoA sensor, it's important that the system have some additional fail-safes, such as using gyroscopes to estimate AoA relative to the ground -- a poor estimate of actual AoA, but a decent sanity check. Alarming and proper pilot training should be able to take care of the remaining risk. Indeed, alarming and proper pilot training would have been sufficient to avoid the two MAX crashes (I know, that's not a popular opinion around here).
  - carlmr 8y ago
    Proper pilot training of course it's important. From what I've heard it's a very difficult process though, so there's definitely some room for improving the UX here. Using disparate sensors for plausibility is of course also useful and would be possible with a software update.
    - cryptonector 8y ago
      For MCAS there's a button to turn it off.
      - carlmr 8y ago
        From the articles it seemes like you needed to perform an esoteric sequence of 4 things to switch off MCAS and it wasn't clear if you succeeded or not, until it reenabled with a 20s time delay.
  - 8y ago
Spartan-S63 8y ago
So, on the topic of misleading regulators, is Boeing being taken to task for that? I'm following the mainstream media coverage, so I may not be privy to details, but is that being made into a big deal. It feels like it should be because regulators signed off on their impression, then Boeing changed the control authority of MCAS without informing regulators. If this isn't illegal and fineable, I hope the outcome of this makes misleading regulators in this way a criminal act.
- bumby 8y ago
  I'd like to read more on this, do you have any article sources that go into a deeper dive on how they changed the configuration after certification/testing?
  - Spartan-S63 8y ago
    If you look at the comment I replied to, their cited source has a section on the change in MCAS control authority to the rear stabilizer.
    - bumby 8y ago
      That article seems inconclusive to me. The FAA said the software change was news to them but Boeing said the FAA considered the final configuration in their analysis. Either somebody is mis-remembering, outright lying, or they didn't consider the software change a true configuration change.
      - HeyLaughingBoy 8y ago
        Nobody needs to remember anything. The phrase "if it's not documented, it didn't happen" means something.
        bumby 8y ago
        Fair enough from a QA standpoint, except the FAA authorities (according to the article) gave them a pass on not documenting it (at least, not documenting it until later). It seems like the FAA was too trusting in this case.
gsnedders 8y ago
> Airbus continues to be required to have triple redundancy says a lot about their respective regulators (EASA Vs FAA) Remember the 737 MAX has both FAA and EASA type certificates, issued separately.
- CaptainZapp 8y ago
  EASA relied on the FAA certification. Just like any other agency. That was based on FAA's stellar reputation, they built up in decades. And essentially fucked within three days. (That's a simplification, but essentially true). I'm not sure that other certification agencies will rely on the FAA's rubber stamping in the future and see a political storm brewing here.
  - gsnedders 8y ago
    Given the time taken between FAA and EASA certification, was it really just based on the FAA certification?
    - anoncareer0212 8y ago
      Nope, not at all, but don't mention it or the reply guy will keep burying you in replies until you accept his tidier moral universe
    - CaptainZapp 8y ago
      https://airwaysmag.com/industry/easa-transport-canada-to-seek-own-review-of-737max/ https://airwaysmag.com/industry/easa-transport-canada-to-see...
  - anbop 8y ago
    The FAA fucked its reputation in three days, but this was a VERY good thing. It was a completely compromised body for years, and their behavior on the 737 Max is what alerted the world and the public to this fact. If they had grounded immediately, they would have preserved their unearned reputation to the detriment of long term aviation safety.
mhandley 8y ago
It won't just be the FAA that signs off on the Max this time - the review process is to be a collaborative international one. In part, this probably indicates how much the FAA's credibility has been damaged by the 737 Max: https://af.reuters.com/article/worldNews/idAFKCN1RV1A1 https://af.reuters.com/article/worldNews/idAFKCN1RV1A1
- cptskippy 8y ago
  While it's sad and unfortunate that it's come to this, hopefully it will be for the best. Other countries rely too heavily on the US regulatory bodies and as those regulatory bodies are becoming more and more compromised by business interests and interference, it's becoming more important that other countries step up and provide redundant regulatory bodies.
paganel 8y ago
> The fact that Boeing is able to skate by with two and a software off-switch if they disagree, while Airbus continues to be required to have triple redundancy says a lot about their respective regulators (EASA Vs FAA) I wonder if the 737 Max will ever be allowed to fly again in Europe, if they don't add a 3rd AoA sensor, that is.

CaptainZapp 8y ago

but that's literally putting cost over safety. Which seems to be fully in line with Boeing's business model and their behavior after the Lion Air crash. Even after the Ethiopian crash they did just about everything to deflect, blame and obfuscate until this behavior was just no more maintainable based on the known facts. Switching then to a "We apologise if your dead" marketing shtick. I find it hard to believe, based on all the evidence, that this company put lifes before profit.

carlmr 8y ago

>The fact that Boeing is able to skate by with two and a software off-switch if they disagree, while Airbus continues to be required to have triple redundancy says a lot about their respective regulators (EASA Vs FAA). I'm sure "cost" was discussed given the 350+ aircraft already produced and how impractical it is to add a third AoA sensor, but that's literally putting cost over safety. Looking at this image here: https://qz.com/1461400/pilots-werent-informed-of-risks-of-boeing-737-maxs-new-feature/ https://qz.com/1461400/pilots-werent-informed-of-risks-of-bo... My biggest doubt is regarding the symmetrical left-right placement of the AoA vanes. Isn't it realistic that they may freeze in the same position, thus not disagreeing, but still wrong? I'm not in aviation, so I'm not sure about this, but without deeper knowledge this seems like a realistic reason to ask for a 3rd sensor with a different placement at least.

danaliv 8y ago
Theoretically it’s possible that both AoA vanes could ice over, but that would mean failure of both vanes’ heating elements. (Not a jet pilot myself but I teach instrument and charter pilots, and I’m in the clouds a lot!)
cryptonector 8y ago
That Air France flight that crashed in the Atlantic had all its pitots freeze. No amount of redundancy will help with that sort of situation -- the sensors were not designed correctly, specifically they did not have heaters with enough power to keep them not frozen. While Boeing should have to add a third AoA sensor, it's important that the system have some additional fail-safes, such as using gyroscopes to estimate AoA relative to the ground -- a poor estimate of actual AoA, but a decent sanity check. Alarming and proper pilot training should be able to take care of the remaining risk. Indeed, alarming and proper pilot training would have been sufficient to avoid the two MAX crashes (I know, that's not a popular opinion around here).
- carlmr 8y ago
  Proper pilot training of course it's important. From what I've heard it's a very difficult process though, so there's definitely some room for improving the UX here. Using disparate sensors for plausibility is of course also useful and would be possible with a software update.
  - cryptonector 8y ago
    For MCAS there's a button to turn it off.
    - carlmr 8y ago
      From the articles it seemes like you needed to perform an esoteric sequence of 4 things to switch off MCAS and it wasn't clear if you succeeded or not, until it reenabled with a 20s time delay.
- 8y ago

Spartan-S63 8y ago

So, on the topic of misleading regulators, is Boeing being taken to task for that? I'm following the mainstream media coverage, so I may not be privy to details, but is that being made into a big deal. It feels like it should be because regulators signed off on their impression, then Boeing changed the control authority of MCAS without informing regulators. If this isn't illegal and fineable, I hope the outcome of this makes misleading regulators in this way a criminal act.

bumby 8y ago
I'd like to read more on this, do you have any article sources that go into a deeper dive on how they changed the configuration after certification/testing?
- Spartan-S63 8y ago
  If you look at the comment I replied to, their cited source has a section on the change in MCAS control authority to the rear stabilizer.
  - bumby 8y ago
    That article seems inconclusive to me. The FAA said the software change was news to them but Boeing said the FAA considered the final configuration in their analysis. Either somebody is mis-remembering, outright lying, or they didn't consider the software change a true configuration change.
    - HeyLaughingBoy 8y ago
      Nobody needs to remember anything. The phrase "if it's not documented, it didn't happen" means something.
      - bumby 8y ago
        Fair enough from a QA standpoint, except the FAA authorities (according to the article) gave them a pass on not documenting it (at least, not documenting it until later). It seems like the FAA was too trusting in this case.

gsnedders 8y ago

> Airbus continues to be required to have triple redundancy says a lot about their respective regulators (EASA Vs FAA) Remember the 737 MAX has both FAA and EASA type certificates, issued separately.

CaptainZapp 8y ago
EASA relied on the FAA certification. Just like any other agency. That was based on FAA's stellar reputation, they built up in decades. And essentially fucked within three days. (That's a simplification, but essentially true). I'm not sure that other certification agencies will rely on the FAA's rubber stamping in the future and see a political storm brewing here.
- gsnedders 8y ago
  Given the time taken between FAA and EASA certification, was it really just based on the FAA certification?
  - anoncareer0212 8y ago
    Nope, not at all, but don't mention it or the reply guy will keep burying you in replies until you accept his tidier moral universe
  - CaptainZapp 8y ago
    https://airwaysmag.com/industry/easa-transport-canada-to-seek-own-review-of-737max/ https://airwaysmag.com/industry/easa-transport-canada-to-see...
- anbop 8y ago
  The FAA fucked its reputation in three days, but this was a VERY good thing. It was a completely compromised body for years, and their behavior on the 737 Max is what alerted the world and the public to this fact. If they had grounded immediately, they would have preserved their unearned reputation to the detriment of long term aviation safety.

mhandley 8y ago

It won't just be the FAA that signs off on the Max this time - the review process is to be a collaborative international one. In part, this probably indicates how much the FAA's credibility has been damaged by the 737 Max: https://af.reuters.com/article/worldNews/idAFKCN1RV1A1 https://af.reuters.com/article/worldNews/idAFKCN1RV1A1

cptskippy 8y ago
While it's sad and unfortunate that it's come to this, hopefully it will be for the best. Other countries rely too heavily on the US regulatory bodies and as those regulatory bodies are becoming more and more compromised by business interests and interference, it's becoming more important that other countries step up and provide redundant regulatory bodies.

paganel 8y ago