Page 13 of 14 FirstFirst ... 311121314 LastLast
Results 121 to 130 of 137

Thread: OT: not about 737MAX, but yeah, it kinda IS about that.

  1. #121
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    30,947

    Default

    Quote Originally Posted by RB211 View Post
    Airspeed has nothing to do with angle of attack. It has to do with angle of the relative wind to the wing, which is the simple definition of AOA. Accelerated stalls are an example of a stall occurring at any airspeed.
    Do not confuse AOA to pitch angle. If you have a thrust to weight ratio greater than 1:1, and your pitching vertically straight up and accelerating, you have a very low AOA.
    I am not confusing them.

    I am indeed using the term AoA as the angle of the wing to relative wind, and presuming normal "flight", not vertical climb, which is not applicable to the 737 as far as I know. Things may differ with altitude, due to air density, because the "supporting force" on the wings comes from the mass of air that is "moved downward" per second by the wings. The manufacturer will have data characterizing the aircraft in all sorts of situations.

    But I think the basic point is that if you are in level flight, and add power with no other adjustment (presuming here that thrust is neutral as far as pitch), your tendency is to climb, and if you cut power, it is to descend. So there are relations between the various flight parameters for any given aircraft, which the manufacturer will have lots of data about.

    If altitude is steady, power setting is normal, no adjustments have been made, airspeed is steady. but a sensor is claiming a dangerously high angle of attack, one that does not correspond to the known aircraft characteristics given the other data, then to my understanding, there is reason to question the data.

    If there is confirmation for the other data, which should be available, then it would seem that the sensor must be wrong, and there is justification for reporting that it appears to be "anomalous", although I would not necessarily go so far as to suggest that any automatic correction be canceled. The monitoring system may be itself bad, and it should be ultimately the pilot's decision what to trust.

    I would suppose that other things would be checkable the same way, to give some guidance as to whether something in the system appears to be fouled up. Then the pilots can decide if that is good data or not, and whether to tell the system to disregard it or not (assuming there is actually a useful way to do that).
    Last edited by J Tiers; 04-15-2019 at 12:54 AM.
    1601

    Keep eye on ball.
    Hashim Khan

  2. #122
    Join Date
    May 2006
    Posts
    14,646

    Default

    Quote Originally Posted by RB211 View Post
    Airspeed has nothing to do with angle of attack. It has to do with angle of the relative wind to the wing, which is the simple definition of AOA. Accelerated stalls are an example of a stall occurring at any airspeed.
    Do not confuse AOA to pitch angle. If you have a thrust to weight ratio greater than 1:1, and your pitching vertically straight up and accelerating, you have a very low AOA.
    So few of words and so well put...

  3. #123
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    30,947

    Default

    More sensor info.

    Apparently they can be the vane type, or a type similar to a pitot tube. I have no idea which is more reliable, but the vane looks more susceptible to damage.

    https://www.heraldnet.com/nation-wor...-had-problems/

    https://www.apizani.it/index.php/pro...attack-sensors

    https://www.flyingmag.com/how-it-wor...tack-indicator
    Last edited by J Tiers; 04-15-2019 at 03:14 AM.
    1601

    Keep eye on ball.
    Hashim Khan

  4. #124
    Join Date
    May 2006
    Posts
    14,646

    Default

    I watched a vid the other day that showed a AoA sensor taken apart and how it worked - I was surprised to see it was just a glorified old school fuel gauge sensor style - potentiometer but looks like it had a redundancy built into it as there were two gliding tangs that scooted across two different coils, probably thick coils that would take some time to wear through,,,

    the outside was built like a brick $hithouse, I pity the bird that would tangle with it as it would get chopped in two --- still when at speed anything is vulnerable,,,

    I replaced a throttle pedal position sensor on a toyota the other day - the "drive by wire" set up, had to take the old one apart just to see as I do with everything unless it has that triple propeller radioactive sign im going in,

    was surprised to see NO moving parts - the only thing that moved was the spring loaded throttle pedal itself - it looks like it has a perm.mag in it that scoots close to a solid state housing and changes values inside - probably just milliamp that then leads to the ECU... no parts to wear out...

  5. #125
    Join Date
    Mar 2015
    Posts
    2,965

    Default

    Quote Originally Posted by J Tiers View Post
    I am not confusing them.

    I am indeed using the term AoA as the angle of the wing to relative wind, and presuming normal "flight", not vertical climb, which is not applicable to the 737 as far as I know. Things may differ with altitude, due to air density, because the "supporting force" on the wings comes from the mass of air that is "moved downward" per second by the wings. The manufacturer will have data characterizing the aircraft in all sorts of situations.

    But I think the basic point is that if you are in level flight, and add power with no other adjustment (presuming here that thrust is neutral as far as pitch), your tendency is to climb, and if you cut power, it is to descend. So there are relations between the various flight parameters for any given aircraft, which the manufacturer will have lots of data about.

    If altitude is steady, power setting is normal, no adjustments have been made, airspeed is steady. but a sensor is claiming a dangerously high angle of attack, one that does not correspond to the known aircraft characteristics given the other data, then to my understanding, there is reason to question the data.

    If there is confirmation for the other data, which should be available, then it would seem that the sensor must be wrong, and there is justification for reporting that it appears to be "anomalous", although I would not necessarily go so far as to suggest that any automatic correction be canceled. The monitoring system may be itself bad, and it should be ultimately the pilot's decision what to trust.

    I would suppose that other things would be checkable the same way, to give some guidance as to whether something in the system appears to be fouled up. Then the pilots can decide if that is good data or not, and whether to tell the system to disregard it or not (assuming there is actually a useful way to do that).
    The ADC or Air Data Computer, which there is three of, on a 747, probably two on a 737, is supposed to take in all the air data sensors which include pitot tubes, static ports, AOA, TAT, SAT, etc. It then interprets the data for display to the pilots. If there was any logic in place to flag a failed instrument and alert the pilot, it would be there.

  6. #126
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    30,947

    Default

    Quote Originally Posted by RB211 View Post
    The ADC or Air Data Computer, which there is three of, on a 747, probably two on a 737, is supposed to take in all the air data sensors which include pitot tubes, static ports, AOA, TAT, SAT, etc. It then interprets the data for display to the pilots. If there was any logic in place to flag a failed instrument and alert the pilot, it would be there.
    Does it actually DO that?

    It makes total sense to have that, but the question is whether it is looking for reported conditions that are not harmonizing with the aircraft characterization, or if it is relying on other means that may not always work (probably NO means is 100% good). And, of course if it is actually giving a helpful display. It would appear that if there is such a clear display, it should presumably be a place to "look soon (if not first) if there is odd stuff happening".

    That does suppose that effective remedies are available. For the A321 there apparently were, for the 737, possibly not so much.....(or, media reporting may just be as poor as usual).

    I have no idea, and defer to you folks who fly them, or other modern aircraft. Or to any engineering types who know and are allowed to say.
    1601

    Keep eye on ball.
    Hashim Khan

  7. #127
    Join Date
    Mar 2015
    Posts
    2,965

    Default

    Quote Originally Posted by J Tiers View Post
    Does it actually DO that?

    It makes total sense to have that, but the question is whether it is looking for reported conditions that are not harmonizing with the aircraft characterization, or if it is relying on other means that may not always work (probably NO means is 100% good). And, of course if it is actually giving a helpful display. It would appear that if there is such a clear display, it should presumably be a place to "look soon (if not first) if there is odd stuff happening".

    That does suppose that effective remedies are available. For the A321 there apparently were, for the 737, possibly not so much.....(or, media reporting may just be as poor as usual).

    I have no idea, and defer to you folks who fly them, or other modern aircraft. Or to any engineering types who know and are allowed to say.
    I don't know if it kicks out a bad instrument, or simply throws a flag if it receives no data? Does it throw an error to the CMC? Or does the instrument itself throw the error? I do not know. This goes beyond the training and knowledge I am expected to know, and is in the realm of the Avionic Techs and Flight Mechanics.

  8. #128
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    30,947

    Default

    From your perspective, it would be a question of what information it presents to you. And how it presents the information.

    The ideal arrangement would identify combinations of input and aircraft behavior that should never occur with the particular aircraft, and suggest the origin of the "mismatch", so that the pilots can decide what to do about it. Suggesting a solution is good, actually having the system acting to apply a fix might be a problem. I do not think it should automatically cut out the assumed bad sensor, for instance, because of the chance of a big problem if the supervisory system itself was bad. It should probably be advisory only, an assistant system to help decide what the problem is, and what to do about it. A way to make sense of a mess of alarms going off at once in a system with complex interactions.

    Even the "voting" among three sensors is flawed. The Airbus incident link in post #1 shows that. Apparently 2 out of three angle of attack sensors iced up, and since they agreed, they were "voted to be accurate", even though they were not. The third , which really was accurate, was apparently voted out as 'anomalous". So a supervisory system should probably not be trusted to automatically cut off what it thinks are bad inputs. Doing that could have crashed the airbus aircraft in the incident, the pilots had to cut off the input from the two that agreed (but were wrong).
    Last edited by J Tiers; 04-16-2019 at 02:18 AM.
    1601

    Keep eye on ball.
    Hashim Khan

  9. #129
    Join Date
    Mar 2015
    Posts
    2,965

    Default

    This guy released another update, the fixes Boeing is doing.
    They are now using both AOA sensors, and with the use of filters, making it triple redundant. I can only assume those filters are looking at data from other instruments, so it now has the ability to kick out both AOA sensors.
    The biggest question of all is why did Boeing half assed the original implementation so horribly? It cost them billions, and 300 people died.
    https://youtu.be/zGM0V7zEKEQ

  10. #130
    Join Date
    May 2006
    Posts
    14,646

    Default

    Wow - very good seems like much of what we were talking about --- better late then never but agreed on the initial design and have no idea what they were thinking,,,

    now as long as it does not mess with them being "stall happy" in other incidences it sounds like this is going to be the ticket...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •