Announcement

Collapse
No announcement yet.

Air France 447 - On topic only!

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

  • Originally posted by 3WE View Post
    Gabriel...has anyone made light of the fact that apparently, the vertical stabilizer/rudder survived the flight?

    Heck, maybe that explains the abscence of ITS?
    What? I thought that it had been torn off the plane due to poor cheap plastic crackerbox design and construction?

    --- 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


    • Originally posted by Gabriel View Post
      What? I thought that it had been torn off the plane due to poor cheap plastic crackerbox design and construction?
      At least he can still blame the French, so I'm sure he'll be happy.

      Comment


      • Originally posted by Fear_of_Flying View Post
        At least he can still blame the French, so I'm sure he'll be happy.
        Indeed.

        French pilots with weak airmanship.

        French instruments and control systems, that are poorly and cheaply designed.
        Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

        Comment


        • Originally posted by 3WE View Post
          Indeed.

          French pilots with weak airmanship.

          French instruments and control systems, that are poorly and cheaply designed.
          French denial of individual responsibility.

          Comment


          • And who but the French could have united us in this way? It's the French Revolution all over again.

            Comment


            • FAO: Gabriel- coffin corner estimates.

              I'd like to go on a slight diversion.

              I know that some individuals have incorrectly advocated extremely steep, aggressive dives at the first indication of slow speed or incipient stall, which is a bad suggestion if there is very little altitude to spare, or if you are at a high high altitude where you could also quickly exceed the maximum airspeed because you are operating somewhat close to coffin corner. Clearly, such inputs need to be applied with the highest levels of mental and physical airmanship.

              With that in mind, have you seen what the approximate maximum and stall airspeeds were at the altitudes, environments and loads that are estimated to have existed when the Airbus was stalled.

              Just curious how much "slop factor" an ITS type would have to work with if they faced a situation of no airspeed indication, (or even the situation of a first officer pulling up and slowing to stall warning "speeds").
              Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

              Comment


              • Originally posted by 3WE View Post
                .......................With that in mind, have you seen what the approximate maximum and stall airspeeds were at the altitudes, environments and loads that are estimated to have existed when the Airbus was stalled.
                .............................
                Good question.
                Also what is the safety factor that is used on these speeds?

                Comment


                • Originally posted by 3WE View Post
                  With that in mind, have you seen what the approximate maximum and stall airspeeds were at the altitudes, environments and loads that are estimated to have existed when the Airbus was stalled.
                  Well, I have been "playing" a bit around that. I have no definitive numbers (something that surely Evan will be able to get from several sources like the FCOM and the accident report), but in general I'd say that they had quite a range to play around.

                  To begin with, I've found the Mmo reported as M 0.86. They had been flying at M 0.82 and had slowed down to M 0.80 (270 KIAS, which at those altitudes is some 460 KTAS) for turbulence penetration just before the upset. That gives an initial margin of 0.06 of Mach to an upper bound. 0.06 of Mach can seem like not a lot, but wait, at those altitudes it is about 36 kts of true airspeed. But if you like a comparison, Mmo on the high side is sort of equivalent to Vref on the low side, it is an operative speed not a hard limit life or death one. Mmo is required to allow certain upsets and recoveries that would put the plane quite faster than Vmo/Mmo, up to a design speed known as Vd/Md. Md is the max speed that the plane must fully withstand as normally, meaning that it must withstand the full 2.5 Gs with its standard 1.5 margin and behave with normal handling characteristics. And in fact, Vd/Md still have a margin over the point where the plane really becomes uncontrollable, or suffers flutter, or becomes dangerous in other ways, because the plane must be test flown at Vd/Md, and that test flight must keep a safety margin over destruction.

                  On the low side, well, the plane's stall warning was triggered at some 220kts indicated, that at those altitudes would be some 375 kts true, that is 85 kts true slower than what they were flying at the beginning of the incident. And we know that the stall warning has some margin over the actual stall.

                  All in all, that's a range 120 kts of true airspeed between the stall warning and Mmo, with margin on both sides (especially on the high-speed side).

                  --- 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


                  • Originally posted by Gabriel View Post
                    Well, I have been "playing" a bit around that. I have no definitive numbers (something that surely Evan will be able to get from several sources like the FCOM and the accident report), but in general I'd say that they had quite a range to play around.
                    Ok, I'll give it a shot.

                    First off, I'll calculate for FL350, which is the neighborhood they would have stayed in if they did things the right way. During the sequence SAT was around ISA +10, so their true altitude would have been around 37150 ft ASL.

                    Here are the facts as I have them:

                    ------------------------------------

                    REC MAX 372 - 376
                    TEMP ISA +11
                    STATIC TEMP: -38.8 (ISA is -54.3)
                    HEADWIND 15kts
                    WEIGHT 205t
                    COG: 28.7% MAC
                    STANDARD ALT FL350
                    TRUE ALT: 37146
                    N1: DECREASING FROM 100% to 83% at THR LK
                    SELCTED SPD .80M / 275kts CAS
                    ACTUAL SPD AT A/P DISCONNECT: <275kts CAS(?)

                    2 h 10 min 34: 215kts CAS
                    2 h 10 min 51: 205kts CAS
                    2 h 11 min 30: 160kts CAS

                    ------------------------------------

                    Before turbulence penetration they were cruising FL350 at .82M, which at ISA +10 would be about 279 CAS or 483 TAS. Throughout the sequence, their REC MAX was around FL372 to FL 376. Due to the warmer air, they stayed at FL350, and due to the turbulence they selected a .80M speed target (272 CAS or 477 TAS). N1 then went from 100% down to 83% over 8 seconds, where it then stayed due to thrust lock for the next 23 seconds. Because of this oversight, I'm pretty sure the speed never stabilized at .80M before the initial pull-up. (If they had done nothing but manage pitch around 1-3°, and held it at FL350, they would have eventually stalled because 83% N1 is far too low to hold 1G at FL350 at 205t. This is why I objected to the 'alternate, instinctive approach' over the established procedure.)

                    So they dropped through .80M (272 CAS or 477 TAS), getting a momentary read of 270kts on the ISIS at the point of the first stall warning. But the speed continued to decay, even after thrust lock was defeated. 2 h 10 min 34: 215kts; 2 h 10 min 51: 205kts; 2 h 11 min 30: 160kts.

                    Ok, so... what was the speed envelope if they had stayed near flight level?

                    Mmo is .86M. I calculate that to be 295 CAS or 507 TAS for FL350 and ISA +10. So if they held FL350 at .80M, they would have 35kts to give on the top end before Mmo. As Gabriel pointed out, this is not the point of destruction either.

                    If I'm using the FCOM chart correctly (see attached), low speed buffet onset occurs around .58M, 192 kts CAS or 346kt TAS. So they're about 130kts above the onset of stall. Remember, the stall warning is triggered at what Airbus refers to as an 'appropriate margin' above the speed at which buffet onset occurs.

                    I get a speed envelope between Mmo and stall buffet of 165 kts true airspeed (not the 25-50 kts previously advertised by various 'experts' in the media). And then there's still more margin between Mmo and Md.

                    So Gabriel, how'd I do there?

                    Comment


                    • So Gabriel, how'd I do there?
                      Sounds well, except for this:

                      "If they had done nothing but manage pitch around 1-3°, and held it at FL350, they would have eventually stalled because 83% N1 is far too low to hold 1G at FL350 at 205t."

                      I mean, ok, maybe (and I've said MAYBE) 83% N1 is too low to hold FL350 (ISA+10). But:

                      1) Maybe it's not. Because the parasite drag goes with the speed squared, and at high speeds (and especially at high Mach numbers) the parasite drag is most of the drag, you have less drag when you fly slower, and hence you need less thrust to hold that slower speed. At the same time, the induced drag increases at low speeds, and at very low speeds (when you approach the stall point) the parasite drag increases again, because the increase in drag coefficient offsets the reduction in speed. So there is a point of minimum drag, which typically is at a somehow slow speed but not too slow, which is also the point of minimum thrust required. If at that point you need less than 83% N1 to hold the altitude and speed, then the speed and altitude will be held with 83% N1 at some higher speed than that point of minimum drag (note: that this point is the one that requires the least thrust, and hence least fuel per hour, doesn't meant that it's the optimum long range speed -and it is not-, because you while you are burning less fuel every minute you are advancing fewer NM every minute too).

                      2) Most important: Even if 83% isn't enough to hold FL 350 (ISA+10) at any speed, that doesn't mean that you'll stall if you hold 1-3º nose up pitch. There is a problem with the "because" highlighted above, and also with the "and": You either manage pitch t hold 1-3º nose-up OR (not AND)you manage pitch to hold FL 350, not both. If you hold 1-3º nose-up and 83% N1 isn't enough tho hold FL 350 at those pitch angles what will happen is that you'll descend, NOT that you'll stall. And I think that all of us who mentioned the feasibility of the "alternate, instinctive approach over the established procedure", as you've called it, will agree that it was to manage PITCH and THRUST, with altitude being a by-product, and not the other way around. If you try to keep the altitude and see that you need increasing amounts of nose-up pitch, you either let the plane go down or you add thrust. If you try to keep the altitude despite the plane aiming beyond 5º nose-up (not to mention beyond 10 or beyond 15 as it was the case here), then yes, you'll stall, but that's nowhere close to our "alternate, instinctive approach" that called for typical cruise pitch and thrust settings. Finally, again, I absolutely do not advocate for the use of this alternate approach in lieu of the established procedure. The established procedure should be followed period. What I've always said is that I am amazed that, if that the pilots are not properly trained in the established procedure, or haven't even heard of it, or they "just don't feel like following it" for whatever reason, that they did what they did instead of doing what every pilot should be able to do: To keep the plane at normal attitudes and powers.

                      --- 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


                      • Evan,

                        One interesting thing about that chart is that it shows you where the real coffin corner lies, if you know where to look.

                        In magenta, the real coffin corner (where the max and min operating speeds meet).

                        In blue, the zone near the coffin corner where you have 0.05M of range between max and min operating speed.

                        In green, the altitude and weight limits of that near-coffin-corner zone for 1G, which is unattainable: 260 tons @ FL 395 and 240 tons @ FL 410.

                        In violet, the altitude and weight limits of that near-coffin-corner zone for 1.5G, which accounts for bank angles in excess of 45º: 260 tons @ FL 310 and 165 tons @ FL 410.

                        --- 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


                        • Originally posted by Gabriel View Post
                          2) Most important: Even if 83% isn't enough to hold FL 350 (ISA+10) at any speed, that doesn't mean that you'll stall if you hold 1-3º nose up pitch. There is a problem with the "because" highlighted above, and also with the "and": You either manage pitch t hold 1-3º nose-up OR (not AND)you manage pitch to hold FL 350, not both.
                          Com'on Gabriel, it's not like I've been ignoring you all this time. I realize it's either pitch up or descend at that point. When I said "held it at FL350", I meant held the plane at the altitude, not the pitch at that angle. Meaning that if a pilot chose to go by instinct at the moment of UAS and not go to the procedures, and was flying at, say 2° pitch, and thought it best to maintain that pitch and thrust as a "known value", but was focused on the task of remaining above the weather system (as this pilot was), said pilot would have soon found it necessary to 'manage' pitch to maintain flight level, which would have led to a dangerously high AoA as the a/c continued to slow due to THR LK. I think the pilot instinct is to maintain flight level in this situation, not pitch and thrust values. My point is that the THR LK effect of UAS may not be understood by some Airbus pilots, who may also think that the procedures are unnecessary to an old hand. I think a certain number of pilots believe airmanship alone will suffice for manual flight on the Airbus, and so there goes your proof that it won't. The PF AF447 did not notice THR LK. The PNF had to alert him to it, and was probably only aware of it from the ECAM message "THRUST LEVERS MOVE". Twenty-three critical seconds expired before the thrust levers were moved. And yes, I see that you agree with me on the adherence to procedures. So no argument there.

                          Comment


                          • Originally posted by Evan View Post
                            Com'on Gabriel, it's not like I've been ignoring you all this time. I realize it's either pitch up or descend at that point. When I said "held it at FL350", I meant held the plane at the altitude, not the pitch at that angle. Meaning that if a pilot chose to go by instinct at the moment of UAS and not go to the procedures, and was flying at, say 2° pitch, and thought it best to maintain that pitch and thrust as a "known value", but was focused on the task of remaining above the weather system (as this pilot was), said pilot would have soon found it necessary to 'manage' pitch to maintain flight level, which would have led to a dangerously high AoA as the a/c continued to slow due to THR LK. I think the pilot instinct is to maintain flight level in this situation, not pitch and thrust values. My point is that the THR LK effect of UAS may not be understood by some Airbus pilots, who may also think that the procedures are unnecessary to an old hand. I think a certain number of pilots believe airmanship alone will suffice for manual flight on the Airbus, and so there goes your proof that it won't. The PF AF447 did not notice THR LK. The PNF had to alert him to it, and was probably only aware of it from the ECAM message "THRUST LEVERS MOVE". Twenty-three critical seconds expired before the thrust levers were moved. And yes, I see that you agree with me on the adherence to procedures. So no argument there.
                            As a side note, I think you are both using the word "instinct" in the wrong way. To react by instinct means to react without thinking, to trust one's survival to sheer impulse. When a moth flies into a flame, that is a survival instinct. That is also the sort of behavior that led to this accident. The choices you are discussing are between memorized procedures and applying some minimum thought and common sense to the situation by asking, what is the best way to keep this plane flying?

                            Comment


                            • Summary

                              Gabriel: Thank for the "coffin corner" discussion.

                              May I summarize and say, "they had a wide speed range" where, if someone wanted to parlor talk, they might wonder if normal airmanship skills and knowledge might be used to keep the plane in a healthy flight mode with admittedly some variation in cruise speed from ideal, "book" values.
                              Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

                              Comment


                              • Originally posted by Fear_of_Flying View Post
                                As a side note, I think you are both using the word "instinct" in the wrong way. To react by instinct means to react without thinking, to trust one's survival to sheer impulse. When a moth flies into a flame, that is a survival instinct. That is also the sort of behavior that led to this accident. The choices you are discussing are between memorized procedures and applying some minimum thought and common sense to the situation by asking, what is the best way to keep this plane flying?
                                Well sir, you are literally correct. But I am using the term 'instinctive' as it is commonly used figuratively to indicate reactionary vs. rational thought. I think that the reactionary pilot instinct when traversing convective monsters like this is to not sink. I think the 'instinctive' physical action related to that thought is to pull back on the stick if the aircraft isn't holding altitude during that initial phase of confused situational awareness. The reason I don't think it is also "add thrust" is because the pilot awareness is of the thrust already being set for .80M (and therefore around 95% N1 in his mind).

                                We don't know much detail about the other UAS incidents, but perhaps what makes this incident unique is that the pilot went from managed mode to a selected speed reduction only five seconds prior to the loss of airspeeds. The auto-thrust was therefore in a 'speed-change' mode at that moment, decelerating off kinetic energy at a low N1. That resulted in the power setting being locked in at a much lower setting than it would have been at once speed had stabilized at .80M. The lack of thrust lever servos meant the traditionally 'instinctive' indication of power setting wasn't there. There were indications: the N1 gauges, the donuts on the N1 gauges that indicate virtual TL angle, and the message "THRUST LEVERS MOVE" appears on the ECAM. But these are for the rational thought process, not the 'instinctive' thought process. It took 23 seconds to get to them.

                                In other words, if the other 13 or so incidents had not been complicated by this speed reduction issue, simply holding pitch around the last known value and riding out the minute or so of UAS in THR LK would have maintained flight level (more or less) until airspeeds returned. So for those incidents, foregoing the UAS procedures did not result in departure from the envelope. But in this case, 'winging it' would probably have required a constant increase in pitch due to the much reduced N1, perhaps to the stall warning (which would perhaps have been disrespected) and beyond...

                                Of course, this isn't what happened to AF447. I'm just pointing out the hidden danger of not adhering to the procedures.

                                Comment

                                Working...
                                X