Announcement

Collapse
No announcement yet.

ANA 737-700 close call on Sept 6 2011

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

  • #31
    Originally posted by TeeVee View Post
    ................bottom line here is that this guy completely lost situational awareness as ..................

    I'm not sure what you mean here but according to Wiener and Nagal (Human Factors in Aviation) situational awareness is applicable this type of event.

    The pilot has to first figure out what is happening and then implement corrective action.

    So in this case I doubt that the pilot completely lost situational awareness; if he had waited much longer this would likely have become a disaster.

    Comment


    • #32
      Originally posted by Highkeas View Post
      I'm not sure what you mean here but according to Wiener and Nagal (Human Factors in Aviation) situational awareness is applicable this type of event.

      The pilot has to first figure out what is happening and then implement corrective action.

      So in this case I doubt that the pilot completely lost situational awareness; if he had waited much longer this would likely have become a disaster.
      I agree: this doesn't seem to be a very strong example of loss of situational awareness. The pilot accidentally pushed the wrong button. In that event, what mistake wouldn't count as loss of situational awareness? I think of loss of situational awareness in aviation to mean a loss of understanding of what the aircraft is doing with respect to its environment, usually when there has been an unrecognized change in that relationship (i.e. the aircraft is losing altitude and you thought you were still in a climb). In this case, as soon as the mishap occurred, it seems the pilot knew what the aircraft was doing and probably why.

      But where are the Airbus bashers on this thread, I wonder? Not only do we have the unfortunate positioning of buttons on this Boeing, but an upset that would have never happened on an Airbus. Maybe more stick time would have helped... oh wait, the pilot managed to pull this off when he wasn't even flying. Imagine the flying circus that would result from pilots hand-flying on a regular basis.

      Comment


      • #33
        Originally posted by Highkeas View Post
        I'm not sure what you mean here but according to Wiener and Nagal (Human Factors in Aviation) situational awareness is applicable this type of event.

        The pilot has to first figure out what is happening and then implement corrective action.

        So in this case I doubt that the pilot completely lost situational awareness; if he had waited much longer this would likely have become a disaster.
        you really believe the plane turned sooooooooo slowly and soooooooo subtly that a pilot concentrating on what he should've been concentrating on would NOT have noticed???

        Comment


        • #34
          Originally posted by Fear_of_Flying View Post
          I agree: this doesn't seem to be a very strong example of loss of situational awareness. The pilot accidentally pushed the wrong button. In that event, what mistake wouldn't count as loss of situational awareness? I think of loss of situational awareness in aviation to mean a loss of understanding of what the aircraft is doing with respect to its environment, usually when there has been an unrecognized change in that relationship (i.e. the aircraft is losing altitude and you thought you were still in a climb). In this case, as soon as the mishap occurred, it seems the pilot knew what the aircraft was doing and probably why.

          But where are the Airbus bashers on this thread, I wonder? Not only do we have the unfortunate positioning of buttons on this Boeing, but an upset that would have never happened on an Airbus. Maybe more stick time would have helped... oh wait, the pilot managed to pull this off when he wasn't even flying. Imagine the flying circus that would result from pilots hand-flying on a regular basis.
          pay attention. it wasn't simply pushing a button.

          your last sentence is completely uninformed and ignorant.

          Comment


          • #35
            Originally posted by TeeVee View Post
            she was trained by the US Navy, where she qualified in helos and some transport aircraft...not sure which. after her active duty enlistment she qualified on the 737 and has flown it commercially for over 10 years.

            and i won't tell her to be less critical of anything..

            That's great, I'm a Navy pilot as well. I did more than one tour and it wasn't in helos. I've also flown commercially for well beyond 10 years and I've got several thousand hours on the 73NG.

            Now that you've tried to impress me with your friend's Quals and failed, I would reiterate that she (and you) should learn to be a little more humble. She sounds like a CRM nightmare.

            As I was discussing with another Forum member, todays' CRM is all about Threat and Error Management and the first challenge is to understand that on every flight there will be threats and there will be aircrew errors. University of Texas/LOSA data indicates an average of just about two errors on every flight...though there was at least one observed flight with 14 errors on it. 85% of those errors are termed inconsequential. 12% will place the aircraft in an undesirable state. This generally means things like excessive speed or unstable approach, but flipping a jet on it's back is definitely an undesirable state......this ANA pilot was one of the unlucky 12%. There are 5 different types of errors and in this case, it would be classified as a 'Procedural' error. He meant to do one thing and did it incorrectly. The best way to manage errors is for crew members to communicate/brief what they plan to do and for the other crewmember to ask pertinent questions. For that to work, both crewmembers need to understand that they are susceptible to making mistakes, they need to be willing to take criticism, and they need to be willing to admit when they've made a mistake. Your friend sounds like she needs to brush up on her CRM/TEM

            Researchers have found that the best way to mitigate known errors that have occured in the past is to change the conditions that caused them or develop new procedures that avoid the error and train flightcrew in their use...strangely enough, they don't subscribe to your friend's method which is to assume the crewmember is an idiot, fire him, and decide to continue on as if it will never happen again. In this case, we could try to eradicate this error by changing the placement of the two switches or we could come up with a new Cockpit entry procedure that would avoid this pitfall. I can tell you that the Entry/Exit procedures we use at my airline don't involve this switch so it would be far less likely to happen. I'm not aware of what they do at ANA and I dont' plan to go in depth concerning what we do

            you really believe the plane turned sooooooooo slowly and soooooooo subtly that a pilot concentrating on what he should've been concentrating on would NOT have noticed???
            Statements like this lead me to believe that you have no idea how the autopilot works which is strange coming from a guy who believes he hasn't made a mistake in 14 years (even though he has no idea how to use capital letters)
            Last edited by Vnav; 2011-10-01, 21:44.
            Parlour Talker Extraordinaire

            Comment


            • #36
              Originally posted by TeeVee View Post
              you really believe the plane turned sooooooooo slowly and soooooooo subtly that a pilot concentrating on what he should've been concentrating on would NOT have noticed???
              I do not know the rate of the yaw/roll but if the flight data info is made public then the timing of events can discerned.

              I found the FDR plots here alomg with another photo of center console
              Tech Log - ANA Japan roll incident. - According to the article ANA said a 38-year-old co-pilot likely hit the rudder trim controls by mistake instead of pushing the door unlock button to let the captain in after he stepped out of the cockpit to go to the restroom.

              The aircraft rolled for approx 28 seconds before reversing roll direction.

              Comment


              • #37
                Originally posted by VViscount View Post
                Assuming this is a photo of the ANA switches it differs from the one shown previously.
                (I don't understand Japanese)

                Comment


                • #38
                  Originally posted by TeeVee View Post
                  you really believe the plane turned sooooooooo slowly and soooooooo subtly that a pilot concentrating on what he should've been concentrating on would NOT have noticed???
                  No. I believe that the plane hardly turned at all until the AP gave up, and then it rolled violently.

                  And what's exactly "what he should've been concentrating on"? Their were at cruise on AP/AT with no turbulence or anything. It was a pretty routine and boring phase of flight. In that circumstance you expect the crew to regularly, but not all the time, monitor the instruments. If you think the opposite then you're just wrong. It just won't happen that you'll have the flying pilot looking for 6 hours to the instrument just seeing how nothing moves.

                  And what this pilot should have been concentrating on during the initial stages of the incident was in grabbing the right f*king knob and not just anything that was at hands reach.

                  I wouldn't call this incident a loss of situational awareness. This pilot had the goal of opening the cockpit door, so he reached the knob (evidently not doing the proper checking and with the "help" of a less than perfect layout) and operated it. He did a mistake. A single and instantaneous mistake. And then had enough situational awareness to detect the symptoms (not that it was hard to tell that the plane was snap-rolling) and act accordingly and mostly correctly in a very dare situation, and it worked. That's not compatible with someone loosing situational awareness. If you don't agree, please tell me which pilot error would not fit into the "loss of situational awareness" definition.

                  And today, this pilot is probably one of the less prone to mistake a switch (again). I would definitely not fire him for this.

                  A final word about your friend. Maybe she is the best of the best. And that's great. Now, pilots come in many shapes and colors. There is a "distribution curve" of qualities, like the normal distribution curve. A left tail contains the few terrible pilots, the middle lobe contains most of the pilots, that are neither terrible nor terrific, and then you have the right tail with the terrific pilots. It's Ok to cut the curve at some point and leave all the pilots left of that line without a job (or train them and bring them right of the line). By all means, please keep away of the pilots in the left tail. But if you are going to keep just the best of the best in the right tail then just there aren't enough of them, not even nearly. Most of the flights just would not happen.

                  --- 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


                  • #39
                    It scares me when I hear pilots say that they couldn't make a mistake like this. It shows a complete lack of human factors understanding. You don't make mistakes like this all the time - you only make it once.

                    TeeVee,

                    You may not have made a critical error in your career yet. Have you made any non-critical ones? Started to do something wrong then realised half way through?

                    Remember the pilot, in this circumstance, is not actually able to be monitoring the instruments actively. He needs to be looking at the monitor displaying the camera image for security, and it is located on the back wall of the cockpit. I'll bet this pilot is one who won't do it again - and moreover, no-one will make this mistake again for a while, as all of us "mortal" pilots who recognise we make errors will be particularly aware of this one.

                    Comment


                    • #40
                      sounds to me like you guys are becoming a bunch of apologists!

                      apparently, the AP didn't/wouldn't give up for quite some time, not mere seconds. on pprune, they're talking about 29 seconds. that's 29 seconds of him what? staring at the camera while holding the ABSOLUTE wrong dial and leaving the door unlocked? Gabriel, this was NOT an instantaneous error. both controls need to be held in position to function. if he let go of the dial before the AP quit, most likely nothing adverse would have happened, other than the plane flying out of trim and burning some extra fuel. but no, he sat there diligently staring at the video screen, holding the wrong dial, while the AP corrected his error until it couldn't. in the meantime though, had he scanned his instruments, he should've noticed that his yoke was off center, and the ball was off (pilot speak apparently). i guess he though letting old boy back in the cockpit took precedence.

                      as for the mistakes i make, yeah, all the time. but i never mistake the gear selector (shifter) for my parking brake even though they are both cylindrical and located right next to each other.

                      MCM, my pilot friend said right from the start, that this happens quite often in the 737. the reason we never hear about it is because the pilots realize well before the AP triggers off that they grabbed the wrong dial.

                      i will give this ANA credit for recovering the aircraft though.

                      lastly, when i said i never made a mistake in my job, i was referring to when i was a paramedic (lawyers screw up every time we open our mouths). did i sometimes grab the wrong vial of drugs when digging through my box of tricks? sure! several of the drugs we used were in the exact same type of vial, differentiated only by the color of the caps. but i always verified which vial i grabbed before pushing the drug. this pilot grabbed the wrong dial, even though the two dial were extremely different physically, and never looked at what he was grabbing. (well, VNAV thinks that knowing the controls blindfolded is a waste or not necessary, hmmmm maybe, just maybe....) next, he stared off at god only knows what for nearly half a minute! and you want me to believe he was aware of the situation? you have got to be kidding me!

                      Comment


                      • #41
                        TeeVee,

                        Have you seen (and looked at) the FDR plots yourself?

                        I don't see those 29 seconds anywhere. And yet, 29 seconds of not looking at the instruments while at cruise with everything Ok doesn't sound like something crazy.

                        In this video what you see is the following:
                        The plane is yawing a bit and banking a bit to the right. That's the pilot holding the heading.
                        Then, at 0:13, the plane starts to bank to the left. That's the AP being unable to keep up with the roll. I don't know at what point the AP gave up. There is a difference between "trying and being unable" and "giving up". If the AP tries and is unable, then it's applying corrective commands to the best of its abilities even if that's not enough. If the AP gives up, it disconnects and the ailerons self-center, so any anti-roll command that was being held up to then suddenly disappears.
                        Another important difference is that when the AP disengages there is an aural warning, which might have been the first clue the pilot had that something was wrong. Another possibility is that the plane keep rolling to the left for a while until the bank angle exceeded 35deg. At that point there is a "bank angle" voice warning which, if the AP was engaged by then, would have been the first warning. The G forces were normal up to then and the roll and yaw motion was very smooth up to then, so it's likely that nobody (including the pilot) felt anything up to then.
                        Unfortunately, the plots made public this far does not show the control inputs other than rudder trim, so it's hard to tell how and when the pilot responded to the upset. You'd have to revers-engineer the motion of the plane and that's beyond my capabilities (ok, if I had the right tools and really wanted do it, maybe...)
                        In any event, at 0:27 (that's 14 seconds after the left roll starts) there is a short roll reversion, and then the roll continues to the left. The only thing that I can think by now (there may be other explanations) is that at that point the pilot started the recovery, but was overwhelmed by the out of trim state, not understanding at first why a good bunch of right aileron was not producing the expected right roll response. Very likely there was a period of confusion, which is perfectly expectable and understandable, but at some point the pilot must have applied not "the rational amount of ailerons" but whatever ailerons plus right rudder pedal it took to roll the plane back to level. That the pilot knew which way to roll, and rolled and waited for the plane to pass back to the "good side" the 90° bank angle before pulling up, shows that both his situational awareness and his upset recovery skills were pretty good. Kudos to whoever trained this pilot in this subject. He did pull up too hard, he should not have exposed the plane to load factors exceeding the limits. But again, there is no "load factor" indicator, so the pilot had to balance between his desired to bring the nose up as soon as possible while not applying excessive loads to break the plane, and he had to do it only by feel. While I think that he should have erred on the safe side of the load factor, he did it pretty well and it worked.

                        http://www.youtube.com/watch?v=FzCw_k_bJvI

                        --- 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


                        • #42
                          Question: when in cruise, are there specific guidelines for how often and what a pilot is supposed to monitor? I imagine this might vary according to airline and/or aircraft.

                          I guess what surprises me here is not that the pilot didn't notice what the aircraft was doing, but that he didn't notice that the door wasn't unlocking. But then, I don't know exactly how that works.

                          But I see no purpose in ranting at the pilot. He made a foolish mistake, as we all do, it wasn't intentional, so far less reprehensible than for example choosing to have several beers and then drive home from the bar, and he was able to correct for the mistake with minimal consequences. I don't think anyone should expect a hair-trigger response from a pilot flying over the ocean on autopilot. If anything, the lesson here once again is that the less that is left up to the pilot, and the more that is safeguarded by automated systems, the better.

                          Comment


                          • #43
                            Originally posted by Gabriel View Post
                            TeeVee,

                            Have you seen (and looked at) the FDR plots yourself?

                            I don't see those 29 seconds anywhere. And yet, 29 seconds of not looking at the instruments while at cruise with everything Ok doesn't sound like something crazy.

                            In this video what you see is the following:
                            The plane is yawing a bit and banking a bit to the right. That's the pilot holding the heading.
                            Then, at 0:13, the plane starts to bank to the left. That's the AP being unable to keep up with the roll. I don't know at what point the AP gave up. There is a difference between "trying and being unable" and "giving up". If the AP tries and is unable, then it's applying corrective commands to the best of its abilities even if that's not enough. If the AP gives up, it disconnects and the ailerons self-center, so any anti-roll command that was being held up to then suddenly disappears.
                            Another important difference is that when the AP disengages there is an aural warning, which might have been the first clue the pilot had that something was wrong. Another possibility is that the plane keep rolling to the left for a while until the bank angle exceeded 35deg. At that point there is a "bank angle" voice warning which, if the AP was engaged by then, would have been the first warning. The G forces were normal up to then and the roll and yaw motion was very smooth up to then, so it's likely that nobody (including the pilot) felt anything up to then.
                            Unfortunately, the plots made public this far does not show the control inputs other than rudder trim, so it's hard to tell how and when the pilot responded to the upset. You'd have to revers-engineer the motion of the plane and that's beyond my capabilities (ok, if I had the right tools and really wanted do it, maybe...)
                            In any event, at 0:27 (that's 14 seconds after the left roll starts) there is a short roll reversion, and then the roll continues to the left. The only thing that I can think by now (there may be other explanations) is that at that point the pilot started the recovery, but was overwhelmed by the out of trim state, not understanding at first why a good bunch of right aileron was not producing the expected right roll response. Very likely there was a period of confusion, which is perfectly expectable and understandable, but at some point the pilot must have applied not "the rational amount of ailerons" but whatever ailerons plus right rudder pedal it took to roll the plane back to level. That the pilot knew which way to roll, and rolled and waited for the plane to pass back to the "good side" the 90° bank angle before pulling up, shows that both his situational awareness and his upset recovery skills were pretty good. Kudos to whoever trained this pilot in this subject. He did pull up too hard, he should not have exposed the plane to load factors exceeding the limits. But again, there is no "load factor" indicator, so the pilot had to balance between his desired to bring the nose up as soon as possible while not applying excessive loads to break the plane, and he had to do it only by feel. While I think that he should have erred on the safe side of the load factor, he did it pretty well and it worked.

                            http://www.youtube.com/watch?v=FzCw_k_bJvI

                            YOU MEAN THIS REPORT: http://www.mlit.go.jp/jtsb/flash/JA1...906-110928.pdf

                            Yeah. and if the pic on page 3 is the aircraft he was flying or setup the same, the pilot should have to go through training from day one. look how far apart the dials are!!!!!! not even in the same section of the console!

                            anyway, i'm assuming to the red deflections in trim are his TWO attempts to unlock the door with the wrong dial. the blue deflections appear to be him trying to save his ass.

                            so maybe it was 20 seconds of trying to unlock the door with the rudder trim. so what? so what if it was 10 seconds? apparently he wasn't paying attention to the instruments or his yoke during that time, since both would have told him right away that the aircraft was doing something it wasn't supposed to be doing.

                            according to my pilot friend, at the point the autopilot gives up in this situation, you will already be screwed since by that time the plane will be in an unsafe attitude. unless you are totally unaware, you should realize that something is wrong BEFORE the autopilot gives up.

                            now, keep in mind that these arguments are not really mine. i'm just parroting what a real life 737 pilot is saying. argue with her if you feel like it. or since vnav has all those hours in the 737 maybe he can comment.

                            Comment


                            • #44
                              Originally posted by TeeVee View Post
                              YOU MEAN THIS REPORT: http://www.mlit.go.jp/jtsb/flash/JA1...906-110928.pdf

                              Yeah. and if the pic on page 3 is the aircraft he was flying or setup the same, the pilot should have to go through training from day one. look how far apart the dials are!!!!!! not even in the same section of the console!

                              anyway, i'm assuming to the red deflections in trim are his TWO attempts to unlock the door with the wrong dial. the blue deflections appear to be him trying to save his ass.

                              so maybe it was 20 seconds of trying to unlock the door with the rudder trim. so what? so what if it was 10 seconds? apparently he wasn't paying attention to the instruments or his yoke during that time, since both would have told him right away that the aircraft was doing something it wasn't supposed to be doing.

                              according to my pilot friend, at the point the autopilot gives up in this situation, you will already be screwed since by that time the plane will be in an unsafe attitude. unless you are totally unaware, you should realize that something is wrong BEFORE the autopilot gives up.

                              now, keep in mind that these arguments are not really mine. i'm just parroting what a real life 737 pilot is saying. argue with her if you feel like it. or since vnav has all those hours in the 737 maybe he can comment.
                              I'm still not entirely sure what your point is - that the pilot screwed up? I don't think there's much disagreement there. But unlike some other pilots who have screwed up in recent times, this pilot actually knew how to fly the plane and how to fly himself out of the mess he created. Isn't that what we have been disparaging so much, the lack of airmanship that has led to so much tragedy? Well, this was a shitty mistake, and obviously the guy was zoned out or whatever, but it wasn't a mistake in that vaunted stick and rudder flying that you've been demanding. When it came to that, this guy was a pro. It was, as has been pointed out, a procedural error, the kind of mistake we all make every day, that 999 times out of 1000 is inconsequential.

                              Here's some lawyer logic for you to mull over. Tell me if this argument is sound: Because he wasn't monitoring when the mishap occurred, therefore the mishap occurred because he wasn't monitoring.

                              I think it's premature to conclude that the pilot was remiss in his monitoring responsibilities. He may have done a check of the instruments just before he tried to open the door. Then there was no aural warning or notable response from the aircraft until it was too late. Is the pilot supposed to train his eyes on the instrument panel the entire trip over the ocean? Is it not likely that he was trying to figure out why the hell the door wasn't opening? Kind of like looking for your sunglasses for five minutes - the ones that have been on your head the whole time.

                              Comment


                              • #45
                                Originally posted by Vnav View Post
                                Researchers have found that the best way to mitigate known errors that have occured in the past is to change the conditions that caused them or develop new procedures that avoid the error and train flightcrew in their use...
                                I wonder if the FAA will ever listen to these researchers. I have seen nothing in the way of an AD or bulletin to address this risk. Hopefully, Boeing has at least put out a bulletin advising pilots of the potential (and somewhat probable) hidden danger.

                                We talk a lot about 'safety culture' here. When I was in college, I had a part time job running a printing press. I worked with some old-timers who taught me to always turn knobs and switches with my eyes as well as my hands. There is a culture of safety amongst machine operators, almost to the point of superstitious fear. Of course, there's always the guy with the missing finger or two who forgot the mantra for a brief moment, there to remind them.

                                I would like to know that pilots are taught a similar lesson about using their eyes to select non-instinctive controls. I would like to know that check pilots monitor this and remind them when they forget.

                                A press or a lathe is a screeching, hand shredding machine, the threat is quite apparent, whilst a 737NG cockpit is quite calm, and the threats are not so apparent. Perhaps that makes a difference.

                                Comment

                                Working...
                                X