Announcement

Collapse
No announcement yet.

What's It Doing Now, Chapter IV: The Gyro Menace

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • What's It Doing Now, Chapter IV: The Gyro Menace

    3WE should love this one:

    Aviation Herald - News, Incidents and Accidents in Aviation

  • #2
    Originally posted by Evan View Post
    Intresting. Airbus' automation design philosophy sucks. Pilots could fly it like a Cessna, direct law, no automation, steam-gauge altimeter, airspeed, and artificial horizon. Kuddos (beyond the mistakes they could have made to out themselves in that position in the first place).

    --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
    --- Defend what you say with arguments, not by imposing your credentials ---

    Comment


    • #3
      Originally posted by Evan View Post
      1. Fixed.

      2. There's so many damn acronyms in that report that I'm not really sure what it says. Definitely the stuff that Evan loves. I did look at the pictures (more attune to my abilities) and it seems there's no attitude indication. Also, if we expect pilots to learn all of that crap, are we surprised that they forget that relentless pull ups are a good way to stall, and forget that ABS/Antiskid generally increases stopping performance and controllability and might be beneficial for times when intense braking is needed?...???

      3. I have few credentials, but will impart this wisdom: Some dude named Murphy once said that anything that could go wrong will eventually go wrong.

      4. I see Gabe saying something about flying it like a 172...I dunno...doesn't seem like that bad of an idea cuz my laptop bluescreened the other day, and the TV went out a couple years back, and the wife's TPMS went out and we had no idea if the tires were inflated, and the microwave electronic control board fried...If you are flying something that acts like an airplane and it suddenly becomes an airplane...maybe you have some knowledge and recent 'training'. If it changes from a phugging-phugoid-phree-vanilla-drone into an airplane...that's tougher...Gabe also once said that electronic monitoring was a really good thing- but that it was good to have old fashioned backups...human and non-human.

      Thanks providing some outsider pontification inspiration...been slow recently...
      Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

      Comment


      • #4
        Ok, here's what seems to have happened:

        1 ) One of the inertial reference units (IRS2) was defective and drifting.

        2 ) The problem was reported and maintenance replaced the wrong unit because the racks were mis-labled (stoogery).

        3 ) The aircraft was thus dispatched with a still-faulty IRS2 which soon began to drift (not a big deal in itself).

        4 ) One of the flight crew reacted to this by switching the wrong IRS selector (IRS3) from NAV to ATT, then (perhaps realizing the mistake) switched it back to NAV (stoogery). Apparently the unit cannot recover NAV capability simply by switching it back like this. (FCOM procedure is to switch a faulting IRS to ATT if available or OFF if not, and to leave it there).

        5 ) So now perfectly functional IRS3 is out of the game, and IRS2 is still faulty and drifting.

        6 ) Without the redundancy of IRS3, an IR disagree results, which, by design, disconnects autopilot and autothrust.

        7 ) The crew IMPROVISE an in-flight IRS realignment, which doesn't go well and has further ramifications for displays and other forms of guidance (GPS is lost).

        8 ) Flummoxed, they fly it in with the standby instruments and direct law—sort of like a big Cessna—as it is designed to fly in this degraded condition.

        9 ) At least one member of the crew lies to investigators (or 'doesn't recall') about switching IRS3 to ATT mode.

        One other thing: the report included the following...

        A short video taken by the second first officer on descent towards Athens shows all three IRS switches in NAV position and the IR fault lights flashing.
        A steady fault light means the IR is bricked, but a flashing one indicates that the heading and attitude may be recovered by switching to ATT mode. All three selectors were still in the NAV position.

        I get the very strong impression the crew had no useful training on ADIRS panel functions, particularly dealing with faults, system ramifications and recovery procedures.

        Once again, this is the job. The job has changed with the technology, as it must. A former 747 pilot is not going to figure this stuff out intuitively, but a properly trained A330 pilot shouldn't have such trouble with it.

        The crew are there to a) fly the plane and b) monitor, troubleshoot and reconfigure the systems as needed, so that requires piloting fundamentals and systems administration skills.

        I'm sorry guys, but that IS the job description on modern jets.

        Comment


        • #5
          so basically what evan is saying is that pilots of airbi and to an extent boeing and others, have to first become real pilots, with the ability to fly airplanes physically. then after having spent some time flying real airplanes as real pilots, they have to go through software engineering/understanding training, which likely requires much more brain power, not to mention a fairly incredible memory and instant recall ability.

          the systems are so deep and complex, that to understand "what it's doing," one needs to understand the fault steps, e.g., the failsafe process each aircraft system goes through as things progressively go to shyte. without that knowledge, the two stooges upfront will never really understand and will inevitably screw the pooch and kill some self-loading cargo.

          sure they have those wonderful books and ipads with tons of useful info, and in a perfect world, the pnf is flipping through those books or swiping through 100's of pages on an ipad searching for the cause/solution.

          i disagree with the crew responsibilities evan wrote above. they should be: the crew are there to a) monitor, troubleshoot and reconfigure the systems as needed, which requires extensive systems knowledge and administration skills, and b) fly the plane.

          Comment


          • #6
            Originally posted by TeeVee View Post
            so basically what evan is saying is that pilots of airbi and to an extent boeing and others, have to first become real pilots, with the ability to fly airplanes physically. then after having spent some time flying real airplanes as real pilots, they have to go through software engineering/understanding training, which likely requires much more brain power, not to mention a fairly adequate memory and instant recall ability.
            Fixed that for you.

            What Evan is saying is that [somebody] ACCIDENTALLY switched IRS3 to ATT mode and then back to NAV again. This was most likely because the three selector switches are laid out to reflect the architecture (1,3,2). EVERY CERTIFIED AIRBUS PILOT SHOULD BE FAMILIAR WITH THIS. Both 1 and 2 are thus placed onside while the redundant unit (IRS3 which, if needed, will go to either side) is placed in the center. This is logical ergonomics. The MCDU monitor page lists the IRS array as a vertical column, as 1,2,3. To do otherwise would be illogical.

            What Evan is saying is that [somebody] wasn't looking closely at the panel when they did that, and it should be taught very deeply in CTPL training to turn the switch with your eyes AND your hand.

            What Evan is saying is that [somebody] also didn't seem to understand that once the switch is turned from NAV in flight, it can't go back there. Unless I'm mistaken, if [somebody] had made that human error and then just left the switch in ATT, they wouldn't have lost the attitude displays and probably not the autoflight either. Turning it back to NAV seems to have been mistake #2.

            What Evan is saying is that [everybody] didn't seem to follow the relatively simple QRH procedure for in-flight IR relignment, but rather improvised. That's what it looks like from Evan's flightdesk anyway.

            Mainly what Evan is saying is that none of this requires 'fairly incredible memory and instant recall ability'. It just requires solid training, concentration, CRM and time to do things by the book (which they had).

            If [somebody] had simply concentrated on the panel and switched off IRS2, there would have been no incident here. (that's not even 'click-clack', it's just 'click'.)

            Evan is saying stop blaming the aircraft just because it's not a Cessna.

            Comment


            • #7
              thanks for taking the time to try and explain yourself.

              but first you did an in-depth analysis of how the system supposedly functions and the failsafe procedures the system's login follows (as you've done on countless other occasions) and then you attempt to simplify it by saying, "all they had to do is..."

              the bigger issue is this: at times the technology is so deep, it doesn't work to say, RTFM and all will be good. i won't bother looking for how many times you've espoused the necessity for pilots to have in-depth systems knowledge, and i don't disagree with that premise. obviously we are training our pilots the wrong way.

              Comment


              • #8
                Originally posted by TeeVee View Post
                thanks for taking the time to try and explain yourself.

                but first you did an in-depth analysis of how the system supposedly functions and the failsafe procedures the system's login follows (as you've done on countless other occasions) and then you attempt to simplify it by saying, "all they had to do is..."
                I think you're confusing "all they had to do is" with "all they had to know is". I disagree with the idea that pilots should only know procedures and not the underlying reasons. Essentially, just as I agree with Gabriel that pilots should understand the aerodynamics behind stall as well as the recovery procedures, I think today's pilots need to understand the system architectures and interdependencies, and no, I don't think that is too much to ask and I don't think it comes at the expense of basic airmanship.

                But this incident was also about concentration, self-discipline and CRM. Which is to say, culture.

                Comment


                • #9
                  Originally posted by Evan View Post
                  If [somebody] had simply concentrated on the panel and switched off IRS2, there would have been no incident here.
                  I think the same outcome would have happened had everybody done nothing (or nobody done anything) with any IRS.

                  That said, come on, disengaging AP and AT and reverting to direct law because one IRS in in ATT mode (or back from ATT to NAV) and the other 2 disagree in NAV info but all 3 agree in attitude info?

                  And, in any event, switching momentarily the wrong IRS switch is a mistake that CAN happen (I mean, pilots have secured the good engine, or even worse, confused the flaps lever with the landing gear lever or the door switch with the rudder trim, which are controls of different shape, different location, and different tactile feedback). So mistaking two identical switches that are one next to the other is perfectly foreseeable.

                  And, as others said, it is not that pilots need to know this. They need to know all the different systems and all their interactions and what things you can do and what you shouldn't attempt and the ramifications of the tree of "what if" for each of them, we are likely talking about a few hundred of lines of knowledge here. The airbus is very (too) pilot friendly and easy to follow when all goes ok, but when things start to go wrong it too quickly reverts to things that are too counter-intuitive and hard to follow, and punish the pilots with too much degradation. For example disconnecting AP and reverting with abnormal alternate law because of UAS. Or for example requiring the pilot to take the thrust levers out of the climb detent to be able to set climb thrust.

                  When it's my time to fly the Airbus I will take a smartphone with me.


                  --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
                  --- Defend what you say with arguments, not by imposing your credentials ---

                  Comment


                  • #10
                    Originally posted by Gabriel View Post
                    I think the same outcome would have happened had everybody done nothing (or nobody done anything) with any IRS.
                    Probably, but if you know a source of IR is malfunctioning, it's still best to shut it down.

                    That said, come on, disengaging AP and AT and reverting to direct law because one IRS in in ATT mode (or back from ATT to NAV) and the other 2 disagree in NAV info but all 3 agree in attitude info?
                    Again, I think if IRS2 had been selected to ATT, this would have preserved autoflight and attitude displays. But instead IRS3 was taken out by switching it to ATT and then back to NAV. The FMGC cannot (and should not) trust any source of data that does not have agreement with another source. WIth only two IRS sources left, and with a disagreement between them (and both still in NAV), the FMGC can no longer know for certain where things are so it must hand things back to the pilots. This is obvious.

                    So either switch IRS2 to OFF, switch IRS2 to ATT or don't do anything.

                    Look, I'm not saying every Airbus pilot needs to know by heart that losing two ADR's will result in the loss of the rudder travel limiter etc., etc. The ECAM is going to tell them those things anyway. But fundamentals regarding the operation of controls that alter or disable the FMGC's and FCU's on an Airbus jet must be learned. These switches must be used with the same level of caution as when shutting down an engine. The job has changed and the knowledge requirements must change with it.

                    And, in any event, switching momentarily the wrong IRS switch is a mistake that CAN happen (I mean, pilots have secured the good engine, or even worse, confused the flaps lever with the landing gear lever or the door switch with the rudder trim, which are controls of different shape, different location, and different tactile feedback). So mistaking two identical switches that are one next to the other is perfectly foreseeable.
                    Yes, absolutely, this is foreseeable. The way to defend against it is to ingrain the importance of visual concentration ESPECIALLY when switching things that directly affect flight control computers.

                    And I agree that the design could be made safer (although, in this case I'm not sure how). I just disagree that is should have to be made safer when the real problem lies in pilot training. We shouldn't have to design jets that make up for bad fundamental pilot training.

                    This crew showed great skill in flying the plane, but not in CRM, concentration or in dealing with systemic problems. ALL of those skills are required.

                    Comment


                    • #11
                      Originally posted by Evan View Post
                      I'm sorry guys, but that IS the job description on modern jets.
                      1. I see that a nice discussion broke out with some focus on your somewhat extreme and narrow views.

                      2. I see that disclaimers are thrown here and there to try to acknowledge that we understand bits and pieces of each others points.

                      3. I will say that I am not sure if you addressed my comment about some filosofer named Murphy with your proclamation.

                      The thing is an airplane. On rare, but finite occasions Mr. Murphy has amazing power. The laws of fiziks and aereodienamicks tend to be fixed. You can use your fundamental aerodienamical airmanshipmanship daily or you can use it on crazy rare occasions...or you can significantly deprioritize it and then be puzzled when it seems to be missing.
                      Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

                      Comment


                      • #12
                        Originally posted by 3WE View Post
                        3. I will say that I am not sure if you addressed my comment about some filosofer named Murphy with your proclamation.
                        Well, Airbus addressed that with the ADIRS panel (overhead), the SWITCHING panel (pedestal) and the MDCU IRS monitor page. Yes, they expect Murphy's Law to be enforced. Therefore, they provided these controls to either realign the IRS units in flight or to shut down a faulty unit and switch to over to the redundant one (IRS3, the very one that [somebody] shut down here.) But a lot of good that does if [somebody] neglects to concentrate and follow the relatively simple procedure, despite having plenty of time to do so.

                        The thing is an airplane. On rare, but finite occasions Mr. Murphy has amazing power. The laws of fiziks and aereodienamicks tend to be fixed. You can use your fundamental aerodienamical airmanshipmanship daily or you can use it on crazy rare occasions...or you can significantly deprioritize it and then be puzzled when it seems to be missing.
                        You've lost me. Where did you see a lack of 'fundamental aerodienamical airmanshipmanship' here?

                        Comment


                        • #13
                          Originally posted by Evan View Post
                          This crew showed great skill in flying the plane, but not in CRM, concentration or in dealing with systemic problems. ALL of those skills are required.
                          Will you be recommending anyone operating under F- registry be banned from, well, everywhere and/or The Republic of France be downgraded to Cat 7?

                          Comment


                          • #14
                            Originally posted by Gabriel
                            Vanilla report

                            Without the benefit of the examination of the aircraft wreckage and recorded flight data
                            information, the investigation was unable to identify any plausible aircraft or systems
                            failure mode that would lead to the observed systems deactivation, diversion from the
                            filed flight plan route and the subsequent flight path taken by the aircraft. However,
                            the same lack of evidence precluded the investigation from definitely eliminating that
                            possibility.

                            I call that BS.
                            I call that the wrong thread...

                            Comment


                            • #15
                              Originally posted by ATLcrew View Post
                              Will you be recommending anyone operating under F- registry be banned from, well, everywhere and/or The Republic of France be downgraded to Cat 7?
                              Damn, I know I have been out of the cockpit for over a year now but we are up to CAT 7 now!?

                              Comment

                              Working...
                              X