4 ms·

Can you point us to a better article?

by sargun 8y ago

LeoPanthera 8y ago
The Wikipedia article is quite good. https://en.wikipedia.org/wiki/Lion_Air_Flight_610 https://en.wikipedia.org/wiki/Lion_Air_Flight_610
trevyn 8y ago
Here’s the actual safety board report, including flight data recorder graphs: 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... Looks like there was a problem with the autopilot trim, likely due to a faulty sensor, and the pilot just needed to switch off the autopilot instead of fighting it — the previous flight of that aircraft experienced a similar issue, and they just turned the autopilot trim off and carried on normally. :( The graphs on PDF page 23 are particularly morbid, and you’ll note that two of the final auto nose-down commands are of longer duration than all the rest.
- trevyn 8y ago
  Ah, this is quite good as well: https://leehamnews.com/2018/11/14/boeings-automatic-trim-for-the-737-max-was-not-disclosed-to-the-pilots/ https://leehamnews.com/2018/11/14/boeings-automatic-trim-for... Apparently it wasn’t the standard autopilot trim, but a special safety program specific to the 737 MAX. Which still is just a switch to disable, if you know that switch is there, or that there is a checklist for dealing with runaway trim, or that you’re aware that auto-trim commands are the reason your plane keeps nosing down...
  - caf 8y ago
    I think the most important point described in that article is that the MCAS trim, unlike the other kinds of elevator trim, isn't defeated by limit switches on the yoke - and that this critical information was not in the training or flight manual.
    - dotancohen 8y ago
      If true, then this right here is the problem root. Why isn't pilot limit respected with regard for the MCAS specifically? Are any other systems so independent? I remember years ago the Boeing folks claiming that the automated Airbuses would be dangerous because the pilot always knows best. So how did it come to this on a Boeing?
      - caf 8y ago
        The answer appears to be something along the lines of "because the whole point of MCAS is to stop the pilot flying the aircraft beyond the limits of stability".
- Stratoscope 8y ago
  > The graphs on PDF page 23 Just a note for anyone looking for these, I think this is a reference to the graphs on the PDF page labeled as Page 14. (The PDF has several pages before Page 1.)
- StreamBright 8y ago
  If the autopilot tries to screw things up it is standard procedure to turn it off. Is failing to do so is human error? I guess so.
  - askmike 8y ago
    This wasn't the default autopilot but a new system, as such they didn't know what was causing the behaviour. On top of that all their sensors gave them different information. > If the autopilot tries to screw things up it is standard procedure to turn it off. Is failing to do so is human error? I guess so. You have a new automatic system pilots are not told about that will deep dive a plane with 200 people into the ocean unless you do things you have learned to apply in a different situation. I don't think you can call that a human error.. Except for maybe the human that decided not to tell the pilots?
    - StreamBright 8y ago
      And in the previous flight a similar team could handle it?

Can you point us to a better article?

by sargun 8y ago

Can you point us to a better article?

The Wikipedia article is quite good. https://en.wikipedia.org/wiki/Lion_Air_Flight_610 https://en.wikipedia.org/wiki/Lion_Air_Flight_610

trevyn 8y ago

Here’s the actual safety board report, including flight data recorder graphs: 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... Looks like there was a problem with the autopilot trim, likely due to a faulty sensor, and the pilot just needed to switch off the autopilot instead of fighting it — the previous flight of that aircraft experienced a similar issue, and they just turned the autopilot trim off and carried on normally. :( The graphs on PDF page 23 are particularly morbid, and you’ll note that two of the final auto nose-down commands are of longer duration than all the rest.

trevyn 8y ago
Ah, this is quite good as well: https://leehamnews.com/2018/11/14/boeings-automatic-trim-for-the-737-max-was-not-disclosed-to-the-pilots/ https://leehamnews.com/2018/11/14/boeings-automatic-trim-for... Apparently it wasn’t the standard autopilot trim, but a special safety program specific to the 737 MAX. Which still is just a switch to disable, if you know that switch is there, or that there is a checklist for dealing with runaway trim, or that you’re aware that auto-trim commands are the reason your plane keeps nosing down...
- caf 8y ago
  I think the most important point described in that article is that the MCAS trim, unlike the other kinds of elevator trim, isn't defeated by limit switches on the yoke - and that this critical information was not in the training or flight manual.
  - dotancohen 8y ago
    If true, then this right here is the problem root. Why isn't pilot limit respected with regard for the MCAS specifically? Are any other systems so independent? I remember years ago the Boeing folks claiming that the automated Airbuses would be dangerous because the pilot always knows best. So how did it come to this on a Boeing?
    - caf 8y ago
      The answer appears to be something along the lines of "because the whole point of MCAS is to stop the pilot flying the aircraft beyond the limits of stability".
Stratoscope 8y ago
> The graphs on PDF page 23 Just a note for anyone looking for these, I think this is a reference to the graphs on the PDF page labeled as Page 14. (The PDF has several pages before Page 1.)
StreamBright 8y ago
If the autopilot tries to screw things up it is standard procedure to turn it off. Is failing to do so is human error? I guess so.
- askmike 8y ago
  This wasn't the default autopilot but a new system, as such they didn't know what was causing the behaviour. On top of that all their sensors gave them different information. > If the autopilot tries to screw things up it is standard procedure to turn it off. Is failing to do so is human error? I guess so. You have a new automatic system pilots are not told about that will deep dive a plane with 200 people into the ocean unless you do things you have learned to apply in a different situation. I don't think you can call that a human error.. Except for maybe the human that decided not to tell the pilots?
  - StreamBright 8y ago
    And in the previous flight a similar team could handle it?