Announcement

Collapse
No announcement yet.

Air France Off the Hook on AF447

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

  • #61
    Originally posted by Evan View Post
    Because, further down is something called REC MAX, which apparently you can climb to at just under 7000fpm it you really push it to the edge.
    I'm glad you have memorized the cryptic acronym REC MAX (what the hell does REC stand for?)...

    ...the basic concepts of a measured pull up, reduced performance of altitude, response to stall warnings, the wisdom of a relatively relentless (mildly modulated- but nevertheless stall-causing) pull up, and that you might want to maintain FDnH P+P=P flight when a million alarms go off (without having a named, type-specific, memory procedure) are things you still need to memorize.
    Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

    Comment


    • #62
      I clicked on your link.

      It did not lead me to a chart of speed vs. AOA for an A330 showing that the critical AOA at Mach ~0.8ish was 9-degreesish.

      That is specifically what I asked for, but again, your mind is closed and your comprehension lacking.

      Gabriel- can you be of any help?- What kind of critical AOA's occur with Airbi 330 or other jet aeroplanies when at cruise? I would like to see some actual numbers.

      Thanks in advance.
      Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

      Comment


      • #63
        Originally posted by 3WE View Post
        I'm glad you have memorized the cryptic acronym REC MAX (what the hell does REC stand for?)...
        Recommended Maximum Altitude. It is safely below coffin corner and the performance service ceiling.

        Comment


        • #64
          Originally posted by 3WE View Post
          I clicked on your link.

          It did not lead me to a chart of speed vs. AOA for an A330 showing that the critical AOA at Mach ~0.8ish was 9-degreesish.

          That is specifically what I asked for, but again, your mind is closed and your comprehension lacking.
          Either that or I was assuming you could take the initiative to find it on your own. Failing that:

          Originally posted by BEA 3rd Interim Report
          At 2 h 10 min 51, the stall warning triggered again, at an angle of attack of about 6°, which corresponds to the theoretical stall warning trigger threshold for the Mach which was then at 0.68.
          The actual stall seems to have occurred around 8° AoA. A typical A330 flying at cruise level at M .80 is about 1.5° below stall warning AoA.

          Originally posted by BEA 3rd Interim Report
          In cruise at Mach 0.8, the margin between the flight angle of attack and the angle of attack of the stall warning is of the order of 1.5 degrees, but the stall warning speed displayed on the air speed tape (in alternate or direct law) will be around 40 kt below the actual speed.
          Attached Files

          Comment


          • #65
            Originally posted by Evan View Post
            Yes, but is that a lack of "basic universal airmanship that you learn in the 172? No, it isn't. Did they teach you about mach issues in flight training for the Tomahawk, or did you learn it on your own?
            Evan, come on... You said that you didn't care so much about the stall at the top but about the first 45 seconds that led to it.

            Nobody actually taught me but I know that I cannot climb in my Tomahawk from 9000 to 1000 ft in a similar way that I do from take-off to 1000 ft. Being the service ceiling 12000 ft and the service ceiling being the point where the sustained climb rate drops to 100 fpm. It is basically airmanship.

            Now, if you want to go to the stall now, I can buy that the stall warning at 7 deg of AoA may have been unexpected for the pilot, but pulling up upon the stall warning is the absolutely contrary to basic Tomahawk airmanship.

            --- 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


            • #66
              Originally posted by 3WE View Post
              Gabriel- can you be of any help?- What kind of critical AOA's occur with Airbi 330 or other jet aeroplanies when at cruise? I would like to see some actual numbers.

              Thanks in advance.
              I couldn't find any nice reference that doesn't look like a graph done in MS Paint (like the ones posted by Evan but actually the ones I found in Airbus and Boeing magazines look similar).

              Here is Boeing's take on it:

              Click image for larger version

Name:	attack_fig4.jpg
Views:	1
Size:	18.1 KB
ID:	1042864

              By the way, all those MS Paint - like graphs are all wrong even from the conceptual point of view.

              They show the curve as displaced to the left and shorter, thus with a lower stall AoA and a lower CLmax, but with a higher CL at any AoA lower than stall.

              But actually what changes is the slope of the curve.
              In a first-order-of-magnitude approximation (which is directionally approximate-ish for mid Mach numbers, say up to 0.6 or 0.7, well below the critical Mach), the effect of compressibility can be understood in terms of frame of references, in a similar way that Einstain general relativity, and not surprisingly arrive to a similar formula. There is a somehow complex theory to get to that formula, but we can explain it simplified from a conceptual point of view, and it is still more or less accurate.

              From the wing frame of reference, compressinility makes that the air molecules come packed closer together in the direction of motion. Imagine the drawing of an airfoil in a graph paper. Now imagine that the paper starts to move relative to the table, but by some sheer magic the airfoil stays stationary relative to the table. In other words, we have the graph paper (representing the air) passing by the airfoil. As the Mach increases and the compressibility effect starts to kick in, you can think that the vertical lines (but not the horizontal lines) in the graph paper will start to get closer one to the next, so the little squares in the paper now become little rectangles, narrower than tall. So more vertical lines fit in the length of the airfoil now. Now let's shift to the graph paper frame. From this perspective, the paper is static and the squares are still squares, but since the airfoil now covers more squares, it looks longer (but not thicker). In other words, it looks like a more slender airfoil. This has a couple of effects:

              1- The chord appears to be longer. If you make the Cl vs AoA curve taking into account that chord, the slope will look the same as usual, but the lift (where you need to multiply the Cl times the chord) will be higher due to the longer apparent chord. Now, for the Cl vs AoA curve, when calculating the Cl we don't divide the lift by this "aerodynamically longer" chord but by the real shorter chord, resulting in a higher Cl. The result doesn't change at zero lift (zero divided by any chord is still zero), so the zero lift AoA is still the same. But as the AoA increases the lift, and hence the Cl, will be higher than normal by a factor "apparent longer chord / real chord", thus making the slope of this curve steeper, i.e the Cl and hence the lift will increase faster per each extra degree of AoA.

              2- The airfoil will seem distorted from the graph paper (i.e. air) point of view, by stretching it only chordwise. Obviously the relative thickness of the airfoil will go down (i.e a 16% thick airfoil may look like a 12% thick one). Let's see what happens with the leading edge. Let's see, just for simplicity, that the leading edge was half a circle. Now this half a circle will aslo be stretched in the chord direction resulting in a ellipse where the long axis is horizontal and longer than the original circle's diameter, and the vertical an shorter axis will steel be equal to the original circle's diameter. I hope you can visualize that the radius of the leading edge is smaller now, resulting in a sharper leading edge. And do you know what sharper leading edges do? They stall at lower AoA's.

              As a side note, the factor of the apparent stretch of the airfoil is 1/sqrt(1-M^2) or 1/sqrt(1-v^2/c^2), where v is the airspeed and c is the speed of sound. Interestingly, c is also used for the speed of light, and if you do that and use the same formula you just got the factor for the length contraction (only in the direction of motion), time dilation and mass increase in Einstein's special relativity theory.

              --- 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


              • #67
                Originally posted by Gabriel View Post
                Evan, come on... You said that you didn't care so much about the stall at the top but about the first 45 seconds that led to it.

                Nobody actually taught me but I know that I cannot climb in my Tomahawk from 9000 to 1000 ft in a similar way that I do from take-off to 1000 ft. Being the service ceiling 12000 ft and the service ceiling being the point where the sustained climb rate drops to 100 fpm. It is basically airmanship.

                Now, if you want to go to the stall now, I can buy that the stall warning at 7 deg of AoA may have been unexpected for the pilot, but pulling up upon the stall warning is the absolutely contrary to basic Tomahawk airmanship.
                Yes, I am not concerned with the actual stall but only the first 45 seconds, consisting of what I believe could be an intentional climb to REC MAX, followed by an intention to "go back down". One of the prevailing problems with this theory is "why did he continue to climb despite the two stall warnings?" The answer I am proposing is that he had reason to consider them false. So the question becomes, "why would he think them false in that situation?"

                Well, for one thing there was an obvious malfunction occuring and other indications were false at that time. But also, if he were unaware of the effects of mach, he would certainly not find those warnings reasonable. If, like 3WE (and how many others?!), he had never been educated on this aspect of aerodynamics and, like 3WE, staunchly embraced the idea that "universal, fundamental laws" applied here, he would absolutely have reason to believe a stall warning at 3 deg and 5 deg of pitch at cruise level speed must be erronoeous. This does theoretically answer that question.

                So then, I have to ask, "at what point are pilots given this education?" You, yourself, just confirmed that you were never given this education in basic flight training. I imagine this is something you would only learn in transition to high-altitude jets. But, again, what I suspect Bonin was intending to do, an 'expedited ascent in high-altitude manual flight', does not exist in commercial transport flight training, so the issue would never be addressed through any procedural training or sim session. Pilots are taught to change flight level through the autopilot. So, again, how many pilots up there right now lack an understanding of the relationship between AoA and mach? How many are confidently ignorant and believe that they learned all that is needed regarding stall as part of 'universal airmanship' during their basic flight training?

                Put another way, let's imagine that AF447 DID have an AoA display (without any indication of calculated thresholds) and Bonin was watching it carefully during that first 45 seconds. If he believed stall AoA was an immutable value for that wing, regardless of mach, he would have been intent on keeping it under 10-12 deg, not under 5 deg. The first stall warning came at around 5 deg. The plane ultimately stalled at around 8-10 deg, I think. So, not only would the AoA display have not prevented the stall, it would have convinced Bonin that the warnings were false.

                You keep diverting this to 'airmanship'. It was poor airmanship to change flight level without airspeeds, regardless of altitude or proximity to service ceiling. I've never argued that what he did wasn't poor airmanship or that it could be justified as a correct thing to do. I'm only trying to understand WHY he did it and how, in that compressed moment of time, he might have thought it was the right thing to do.

                Why am I doing that? To highlight and illustrate the danger of improvisation over procedure among pilots who may not possess an adequate, somewhat-Gabriellian understanding of aerodynamics AND a Gabrieliian standard of 'basic universal airmanship'. I suspect there are quite a few up there, right now, who don't, and that isn't going to change unless some very determined and standardized regulation arises throughout the world that forces flight schools and operators to invest more in pilot training. In other words, that's not going to happen anytime soon.

                To safely improvise in this situation, you need a solid understanding of both. To safely fly the airplane in this situation, all you need is 'basic, universal airmanship' and proficiency on a memory procedure.

                Comment


                • #68
                  Originally posted by Evan View Post
                  The actual stall seems to have occurred around 8° AoA. A typical A330 flying at cruise level at M .80 is about 1.5° below stall warning AoA.
                  Wow, I am reeling. Years of ass hat parlour pontification to pull up in an intelligent and measured fashion while paying attention to wing loading, speed and possible stall warnings.

                  Now, I learn this. I knew about the CL shift and high speed tuck, but not stall angle changes...oh, the ignorance.

                  If I become a jet pilot, I better pay attention to the high altitude stuff, and totally change my thinking to where pull ups are done in an intelligent and measured fashion with attention to wing loading, speed and possible stall warnings.

                  To Gabriel: Thank you for confirming that true and accurate data are NOT a quick google search away, and instead conceptual, marginally true cartoons take their place...(teaching value acknowledged)

                  Would you say that if we had an accurate graph, that doing pull ups in an intelligent and measured fashion, with attention to wing loading, speed and possible stall warnings (and AOA indications since I’m asking you ) might be wise?
                  Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

                  Comment


                  • #69
                    Originally posted by Gabriel View Post
                    I couldn't find any nice reference that doesn't look like a graph done in MS Paint (like the ones posted by Evan but actually the ones I found in Airbus and Boeing magazines look similar).
                    The graphs I posted are from the BEA report.

                    Comment


                    • #70
                      Originally posted by Evan View Post
                      The graphs I posted are from the BEA report.
                      I know, and as 3WE said they are "conceptual, marginally true cartoons", very much like the ones from Boeing's Aero 12.

                      If the BEA graph was a true representation, then you would have zero lift at a positive angle of attack.

                      --- 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


                      • #71
                        Originally posted by Evan View Post
                        The actual stall seems to have occurred around 8° AoA.
                        Source or reasoning for that estimation?

                        Some posts ago (#49) I estimated that the stall AoA at the top of the climb in AF447 was 11 deg, and explained my estimation including a graph. But it seems that you didn't agree with that?

                        --- 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


                        • #72
                          Originally posted by Gabriel View Post
                          Source or reasoning for that estimation?

                          Some posts ago (#49) I estimated that the stall AoA at the top of the climb in AF447 was 11 deg, and explained my estimation including a graph. But it seems that you didn't agree with that?
                          I think 11 deg is a reasonable estimate, with the onset of buffet around 8 deg. I believe 11 deg is still lower than 15 deg however.

                          Originally posted by BEA Third Interim Report
                          A change in the recorded normal acceleration behaviour was revealed from 2 h 10 min 53, at an angle of attack about 1 to 2 degrees greater than the warning activation threshold.
                          Attached Files

                          Comment


                          • #73
                            Originally posted by Evan View Post
                            The French magistrates have dropped charges against Air France and Airbus. It was just one of those existential things that happen in life when you place inadequately trained pilots in a commercial transport cockpit with no procedure for a known failure scenario. C'est la vie.

                            [...]
                            That's what we already found out .. when. And then there comes a jetphotos newbie who thinks it would be a good idea to only have 1 F/O in the cockpit, in case of AF #447. Shit. Usually, I am a quite friendly person. I just try to remember what I wrote with only 8 or 18 jetphotos forum entries. Was I so keen back then to say,
                            pah, an Airbus 330 on the longhaul does not need more than 1 unexperienced F/O between SBGL and LFPG. I hope I was not. I am almost sure that I was not!

                            inadequately trained pilots in a commercial transport cockpit with no procedure for a known failure scenario.
                            Yes. And the most evil thing was, the two F/Os in the cockpit said goodbye to the only AF Flight Captain on board, who then went to sleep, as if they had not one question.

                            Intelligent people know what they do not know. Especially on a nonstop flight with almost 6,094 nautical miles (Guarulhos, in the printed LH winter schedule 2010, type LH-B744). Galeao is a little bit shorter, but not that much imho.

                            807 flight hours on type, for the man who killed himself and a full Airbus 330-200. And only 1 Flight Captain, who took his scheduled break outside of the cockpit.

                            What had I done, in case of that 32 year old F/O. If I had never seen a (temporary) failure of one or two cockpit instruments. Shit. I was 30.8 years old when I joined this forum back then in 2008. I assume, I am almost sure that in the year 2008, AF #447 was not the first flight on this planet with temporary instrument failure.

                            Not only inadequately trained pilots on the long haul in high alt, but even inadequately informed pilots. The right seat F/O not only lacked training, he even reacted as if he were completely surprised, completely without knowledge about that phenomenon.

                            What would I do today. With x years in Randazzo's LH-B744 fsx simulator. Almost nothing, or request a slight descent to regain speed. But I also repeat, I have seen longhaul flights on TV with two experienced Flight Captains on board, so that

                            ALWAYS one experienced longhaul Flight Captain sits in the cockpit. Then the only F/O never is able to pull his yoke out of that longhaul flight. Captain Dubois woke up and tried to repair what the young boys destroyed. But it was too late and he died together with all other 227 souls on board. Zero survivors.
                            The German long haul is alive, 65 years and still kicking.
                            The Gold Member in the 747 club, 50 years since the first LH 747.
                            And constantly advanced, 744 and 748 /w upper and lower EICAS.
                            This is Lohausen International airport speaking, echo delta delta lima.

                            Comment


                            • #74
                              Originally posted by LH-B744 View Post
                              Captain Dubois woke up and tried to repair what the young boys destroyed. But it was too late and he died together with all other 227 souls on board.
                              It wasn't too late. The airplane was still responding to control inputs. All the instruments were functioning at the point. A quick look at the PFD told the story: high vertical speed and loss of altitude and no corresponding dive attitude or airspeed. That's a fully developed stall. Remedy: pitch down, trim down, retard thrust, speedbrakes, fly below Vne, get the wings back in lift-generating mode. But he didn't make any such verbal announcement of the situation or any direct, forceful command to facilitate a recovery. The thing missing was situational awareness. But, when he arrived in the cockpit, they still had the time.

                              Comment


                              • #75
                                Trim down? Wasn't the autotrim still active? Holding the stick forward should have provided the commanded nose-down pitch rate or at least the plane would have attempted to do that and because possibly it would not have been able to (against the nose-up pitching moment of the engines) it would have added nose-down trim by itself with no additional pilot intervention other than holding the stick forward (in the same way as it trimmed almost fully nose-up in response to making mostly nose-up inputs).

                                Retard thrust? I am in general against doing that as a first resource but if the plane doesn't quickly respond to nose down (up to full nose-down) and then sensible nose-down trim (which I understand would have been automatic in this case). Now, in theis case, judging from the plane's response when the PNF took the controls and made significant nose-down inputs for some seconds, the plane clearly reduced the AoA(*) but was not fast at all at doing it, so possibly retarding the thrust would have been very much in order in this case.

                                Speedbrakes? SPEEDBRAKES?!?!? Fly below Vne???? They were well below Vne the whole time. Do you mean after the dive resulting from the stall recovery? But then why you say, as the last item, "get wings back into lift-generating mode? As opposed to the stall or as opposed to a zero or very low or negative lift AoA resulting from the stall recovery? Because if you lowered the AoA that much, you way overshoot your AoA reduction. The AoA reduction should be enough to bring it below the critical AoA, not more than that.

                                (*) When the captain arrived to the cockpit, almost simultaneous the stall warning (that had been sounding uninterruptedly for almost a minute since the stall at the top of the climb) stopped sounding because the indicated airspeed went below a very low speed threshold, much lower than the already very low real airspeed, now not because of the icing that had already cleared but because of the extremely high AoA. This very low speed threshold, just double digits in knots, is considered too slow for the AoA vanes to provide relaible indications, so the stall warning is inhibited below that speed. When the PNF took the controls for some seconds and applied significant nose-down inputs, the AoA clearly started to reduce and with that the indicated airspeed started to increase (due to the pitots becoming more aligned with the relative wind) and hence the AoA indication was considered reliable again and, being it still way to high (even if not as high as seconds earlier), the stall AoA activated again. This possibly was very confusing: The stall warning went silent. I push down and it starts again. Sure enough, he pulled up again and the stall warning stopped again, kinda reinforcing that pushing down activated the stall warning and pulling up made it sound. I wonder if they realized of this and made some wrong conclusions.

                                --- 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

                                Working...
                                X