Announcement

Collapse
No announcement yet.

777 Crash and Fire at SFO

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

  • Gabriel,

    You're working on the assumption that they were able to pick up the correct visual aim point (i.e. visual approach path) without electronic guidance.

    Its not the easiest thing to do in a large aircraft, particularly if you are new to it. Its pretty clear that the PF wasn't able to do it (based on not only initial, but later actions).

    Losing speed was never really an issue as you identify. They're at Flap 5 speed at 2000ft - perfectly reasonable. Their issue was that they were 500ft above the path, and not actively trying to get back onto it at that point. The fact that they get onto the path above 500ft, and on speed, demonstrates they really didn't have an energy problem at that stage.

    FWIW, in V/S the thrust levers don't quite sit at IDLE, but just a little above. Just a feature of that mode. The thrust increase over IDLE isn't sufficient to cause you problems - if you're needing less thrust than V/S gives you at its 'idle', then you've got bigger problems.

    Also - calling for Flaps 30 above the appropriate speed is a HUGE indicator of overload. Its one of those things we use as a trigger to demonstrate that either you, or the other pilot (as applicable), is struggling to keep up.

    Comment


    • Originally posted by MCM View Post
      Gabriel,

      You're working on the assumption that they were able to pick up the correct visual aim point (i.e. visual approach path) without electronic guidance.

      Its not the easiest thing to do in a large aircraft, particularly if you are new to it. Its pretty clear that the PF wasn't able to do it (based on not only initial, but later actions).

      Losing speed was never really an issue as you identify. They're at Flap 5 speed at 2000ft - perfectly reasonable. Their issue was that they were 500ft above the path, and not actively trying to get back onto it at that point. The fact that they get onto the path above 500ft, and on speed, demonstrates they really didn't have an energy problem at that stage.
      Okay, I see your point, but have a couple of comments.
      As you say, falps-5 speed and 2000ft looks perfectly reasonable... when you are intercepting from below or crossing 2000ft already established in the glide slope (what will happen at some 6.3 NM of the aiming point in a 3° approach).

      Not when you are 400ft too high at 4.3 NM of the threshold (4.2° of glide path needed to reach the aiming point, much steeper if you want to reach the correct 3° glide path while still more than 500ft above the runway) and want to slow down to Vapp. At 180kts a slope of say 5° means 1600 fpm, which with just flaps 5 will make it hard to slow down quickly enough.

      The fact that they crossed 500ft with the right speed and altitude doesn't mean, in my opinion, that they didn't have an energy problem. The speed "crossed" Vapp at that point, but it was descending, the altitude "crossed" the target glide slope at that time, but the real slope was much steeper, and the thrust was fully idle (and the plane fully dirty), meaning that they were loosing as much energy as it is physically possible. Only then they managed to cross 500ft with the right altitude/speed.

      While a prompt burst of thrust and pitch up at that precise instant would have stabilized the approach shortly after (remember spool up time from idle), I'd say that 500ft was too late for my taste. It's ok in the Tomahawk, but not in a 777.

      So, in my opinion, they did have an energy problem. They had too much energy and, not only that, they were struggling to lose it since the beginning of the animation ("we are a bit high" "I will descend more" and sets VS to -1500).

      A while later, while still too high, the PF sets the VS back to -1000. You say that's because he didn't realize that he was to high. I thought it was because he wanted to slow down. But maybe you are right, and then I see your point.

      So, if your diagnosis is wrong, the medicine will be hardly right.
      If he judged the altitude wrong and didn't realize how much altitude (which is excess energy) he had, he could only have kept flying trying do what his wrong judgement told him until he realizes that this ain't working and goes around.

      So, what would have been the right course of action IF the situation had been properly assessed?

      You say (or so I understood) that increase to 1500fpm until 2 white / 2 red, then 700fpm and then slow down and flaps until Vapp.

      I say that keep 1000fpm, pull back to idle, slow down & add flaps until Vapp, then 1500fpm is needed until 2 white / 2 red, then 700fpm.

      Both might work.
      I bet you a bear that I will reach the stable state (Vapp stable, correct glide slope stable, correct thrust setting stable) farther away from the runway than you.
      And if instead of keeping 1000 I leveled off or even climbed a bit until full flaps & Vapp, and then descended, then I would reach the stabilized approach even earlier, but the pax might not like it.

      The reason is this:

      The mechanical energy of the plane is (1/2*V2 + gravity*altitude)*mass (that is, kinetic + potential)

      It's easy to convert kinetic into potential or potential into kinetic. So it's easy to increase or decrease one of them (at the expense of the other). That's what the elevator is there for.

      What's not so easy is to increase or decrease the sum of both, that is, to change the total mechanical energy.

      The rate at which the sum of both changes (regardless of how they exchange energy between them) is (Thrust-Drag)*V.

      The importance of V in the total energy is always clearly taken into account, but it's importance in the gain/loss of energy is usually overlooked. And not only that: a 1000fpm descent means much more slope at 140kts than at 180kts. While you are loosing energy at the same rate (same "joules per second" or same "watts") you are loosing more energy per nautical mile.

      If I need to lose energy, I want as little thrust, as much drag, and as slow speed as possible.

      So, when you are high and fast and need to go low and slow, first idle (reduce T), then slow down (reduce V, even if that means not descending as much as you'd like to or not descending at all), add gear as flaps a soon (which means as fast) as reasonable (increase D), then descend. If that doesn't make it, nothing will.

      Descending first doesn't allow to reduce V as quickly which is bad in itself and, worse, the delay in the reduction of V means a delay in adding flaps and hence in increasing the drag.

      But yes, this is useful only if you realize that you have too much energy. If you don't think that you have too much, you will not want to lose the excess as quickly as possible.

      --- 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


      • Bracketed material added for clarity

        Originally posted by Gabriel View Post
        As you [MCM]say...

        [but I say]...Not when...
        Hey boss...the total non aeroengineer aggie says that there's more than one way to burn energy (actually there's a broad middle ground of pulling power and adding drag) and that Tommahawk and 172 drivers might want to go a little easy on critiquing the finer points of how a licensed big-iron guy would address a high and fast 777.
        Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

        Comment


        • Originally posted by MCM View Post
          Gabriel,

          You're working on the assumption that they were able to pick up the correct visual aim point (i.e. visual approach path) without electronic guidance.

          Its not the easiest thing to do in a large aircraft, particularly if you are new to it. Its pretty clear that the PF wasn't able to do it (based on not only initial, but later actions).
          Wow- this surprises me- I thought pretty much anyone with 200 hours could do this! And it's personal because I truly could not do it when I was taking real flying lessons. It finally took Microsoft flight sim for me to learn the visual clue of stuff before the TD point gets lower and stuff after it gets higher...

          But after you learn it- it doesn't really matter if you have a nose visible out the front window or have the MSFS view pointed up or down or zoomed...

          Seriously- I'm shocked that someone who has 200 hours and a commercial rating, let alone an ATP and 777 qualified can't "see" where the plane is going. Yeah, it's a skill and while it's not in the FAR's wouldn't you think it would naturally happen?

          Or are we back to fact that the training focuses so so hard on an ILS to minimums and the right way to enter a hold, and doing V1 engine cuts that the basic basic skills suffer at the expense of the specialized specialized, recurrent recurrent training?
          Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

          Comment


          • Originally posted by MCM View Post
            The fact that they get onto the path above 500ft, and on speed, demonstrates they really didn't have an energy problem at that stage.
            Are you just playing devil's advocate now? They were never on the path above 500ft. They dropped through the path five seconds after crossing 500ft, at 1,200 V/S with engines at extended idle. That certainly sounds like an energy problem to me. They had PAPI on-path for five seconds after that, at which point they were at 330ft. I'm sure you mean that they could have arrested the descent but could they could stay on-path in that state by taking action below 500ft?

            Asiana's stated policy is to call "stabilized" or "go around" at the 500ft callout. They were definitely not stabilized at that point. They should have called it out and gone around, being out-of-order on V/S and at dead idle thrust. But as the report states, 97% or unstable approaches are continued to landing...

            I'm sure that stabilized approach criteria was established as a compromise between safety advocates, engineers and industry executives. The compromise is built in. There really should be no further room for compromise on the part of pilots looking to win a tough game of PAPI roulette.

            Note the chart. Note the sharp increase in hard landing accidents on current aircraft (yes, I realize there are a lot more landings involved in the 'current' category, but also compare the figure to all the other incidents)

            Comment


            • Originally posted by Evan View Post
              See MCM's post above
              I did, thank you. Doesn't really answer my question, so perhaps I'll rephrase it: by the time one is upgrading to international widebody command, is it not too late to talk about "automation intuition"(whatever that might be) and the alleged lack thereof in manufacturer's training materials? The person in question had been flying highly automated aircraft for years.

              Comment


              • Originally posted by ATLcrew View Post
                ...The person in question had been flying highly automated aircraft for years...
                Do I respond with:

                However this strange bunch were still uncomfortable hand flying a visual approach on a clear evening with light winds...

                or

                Do I respond with:

                Apparently so much so that the forgot some basics of watching VASI's and airspeeds...

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

                Comment


                • Originally posted by ATLcrew View Post
                  I did, thank you. Doesn't really answer my question, so perhaps I'll rephrase it: by the time one is upgrading to international widebody command, is it not too late to talk about "automation intuition"(whatever that might be) and the alleged lack thereof in manufacturer's training materials? The person in question had been flying highly automated aircraft for years.
                  One possible answer to that is that the automation intuition required to fly a Boeing airplane may be different (at least partially) from that of an Airbus, where had had recently came from.

                  --- 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


                  • Intuition is a powerful thing. I phone apps seem to excel at it while windows 8 hides things...

                    Fms and autopilots and approach procedures have a certain crypticness about them to the point that we fly. 767s into mountains and might inadvertently turn off the speed control...
                    Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

                    Comment


                    • I say they didn't have a big energy problem, not that they weren't high or fast with action required.

                      With the gear down, the aircraft will slow quickly, and will slow down while going down too. At 2000ft the approach was salvageable.

                      Maintaining 1500fpm at flap 5 won't allow a deceleration without assistance - but the gear and flaps were available.

                      I'm not saying they weren't high, just that it was well within the parameters that can be fixed without needing to use extreme measures.

                      You say (or so I understood) that increase to 1500fpm until 2 white / 2 red, then 700fpm and then slow down and flaps until Vapp.

                      I say that keep 1000fpm, pull back to idle, slow down & add flaps until Vapp, then 1500fpm is needed until 2 white / 2 red, then 700fpm.

                      Both might work.
                      I bet you a bear that I will reach the stable state (Vapp stable, correct glide slope stable, correct thrust setting stable) farther away from the runway than you.
                      And if instead of keeping 1000 I leveled off or even climbed a bit until full flaps & Vapp, and then descended, then I would reach the stabilized approach even earlier, but the pax might not like it.
                      I really don't like your last method, and it is effectively banned by a number of airlines when you are down low. It works well when you're up higher. It is far, far preferable to be on a steady (if slightly higher) rate of descent to intercept the approach than to use a high descent rate down low (which is what the level descel then dive method does).

                      I prefer my method, as I know it works for me and my aeroplane. I know what it will do. The method you mention you prefer could well work too, but I know that I can slow down my aeroplane at -1500fpm if required. Many ways to skin a cat. The only one I wouldn't use is FLCH .

                      Comment


                      • Originally posted by Gabriel View Post
                        One possible answer to that is that the automation intuition required to fly a Boeing airplane may be different (at least partially) from that of an Airbus, where had had recently came from.
                        If anything, from the intuition standpoint (in my opinion, anyway), Boeings are easier, since they give you more tactile cues than a Bus does.

                        Comment


                        • Originally posted by ATLcrew View Post
                          If anything, from the intuition standpoint (in my opinion, anyway), Boeings are easier, since they give you more tactile cues than a Bus does.
                          True, but they are all airplanes, click, click and they all fly like an airplane. Some are just larger than others. There is a time that automation must be discontinued and the airplane flown like an airplane!

                          Comment


                          • Originally posted by BoeingBobby View Post
                            True, but they are all airplanes, click, click and they all fly like an airplane. Some are just larger than others. There is a time that automation must be discontinued and the airplane flown like an airplane!
                            I don't disagree.

                            Comment


                            • Originally posted by MCM View Post
                              I really don't like your last method, and it is effectively banned by a number of airlines when you are down low. It works well when you're up higher. It is far, far preferable to be on a steady (if slightly higher) rate of descent to intercept the approach than to use a high descent rate down low (which is what the level descel then dive method does).

                              I prefer my method, as I know it works for me and my aeroplane. I know what it will do. The method you mention you prefer could well work too, but I know that I can slow down my aeroplane at -1500fpm if required. Many ways to skin a cat. The only one I wouldn't use is FLCH .
                              Well, we would actually have to try both methods to see how they would have worked in this particular situation.

                              It's a fact that my method would have had you fully stabilized (in landing config, in speed at Vapp, and in VS and thrust to follow the correct glide path) farther from the runway, and hence higher, than your method. Yes, my method will have steeper descent lower than yours, because you'll stabilize the glide then the speed and I'll do it the other way around. Bun not necessarily higher sink rates. Remember 1000fpm at 140kts is 4.0° while the same sink rate at 180 kts is 3.1°.

                              I don't have it so clear that, with flaps 5 and gear down, the plane will easily descend 1500fpm and slow down at the same time. It didn't seem to work so well. They were loosing just 1 knt every about 3 seconds when they were at -1500 with flaps 5 and gear down and the TLs were a far back as the AT will go.

                              That said, there are a bunch of actions that could have been done while they were at flaps 5, gear down, -1500 and 180 kts:

                              - Manually pull and keep the TLs back to idle (without touching the AT).
                              - Extending a bit of spoilers (I have seen this done a lot of times during approaches, and from Internet comments it seems that it's perfectly acceptable with flaps 5)
                              - Add flaps. Ok, maybe 180 kts is a bit faster than what I'd normally use to set flaps 20 (which would be about minimum maneuver speed for flaps 5 + 20 kts). But I bet my.... ok I don't bet but I strongly suspect that 180 kts is slow enough for flaps 15, probably for 20 too. These settings are take-off settings too, so they should be perfectly ok for V2+20, and 180 kts looks like a reasonable figure for V2+20.

                              So, after thinking it for another while, I'm with you. I'd keep 1500fpm and slow down at the same time, with the help of one or more of the aids mentioned above.

                              Would you also use one of these methods (remember you are at 180 kts)? Which one? (you mentioned the gear, but it was already down).

                              Oh, and by the way, they were at ft when they selected -1500, and still at 2500ft when they went back to -1000, so if 500ft too high and 180 kts would have been easily salvageable at 2000ft as you said, it would have been even more so in this case.

                              So, as you said, many methods, you can use or the other, and you agree it is well within "the parameters". That's why this finding by the NTSB surprised me:

                              Test pilots from Boeing and the FAA, who were type rated and current in the 777, flew
                              multiple runs simulating visual approaches to runway 28L at SFO with the A/P off, beginning
                              5 nm from the runway threshold with two sets of initial conditions. One set of conditions,
                              referred to as the accident profile, modeled the accident flight with the starting altitude at 2,100 ft
                              msl, about 400 ft above a 3° glidepath, and the other set, referred to as the standard profile,
                              modeled a flight that was on a 3° glidepath by starting at 1,650 ft msl. Data from each run were
                              evaluated to determine whether the approach was stable at and below 500 ft agl based on
                              Asiana’s stabilized approach criteria. The pilots had no difficulty in achieving stable approaches
                              on any of the nine runs started at the standard profile conditions (on glidepath), and they were
                              always able to comply with Asiana’s guidance to avoid descent rates in excess of 1,000 fpm
                              below 1,000 ft agl and in excess of 1,500 fpm between 2,000 and 1,000 ft agl.
                              When starting at the accident profile conditions (above glidepath), the pilots experienced
                              difficulty in achieving stabilized approaches, and on 4 of the 10 accident profile runs, they were
                              unable to achieve a stabilized approach
                              . On all 6 runs that resulted in a stabilized approach, the
                              pilots used descent rates in excess of 1,000 fpm below 1,000 ft agl, and on 5 of the 6 runs, the
                              pilots used descent rates in excess of 1,500 fpm between 2,000 and 1,000 ft.
                              We are talking of Boeing and FAA test pilots here, rated and current on the 777!!!

                              --- 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


                              • I'm quite surprised they couldn't get it stable too Gabriel. Very surprised. It must be a very slick aeroplane, and makes it more likely that they needed to give it away earlier.

                                Either way, the methods we have discussed are solid. If we can't get in from that position, then around we go.

                                Not sure on the limitations of the speed brake on the 777, but as long as it was within the usable range, I'd use that. The speed was below the flap 20 limit speed, so I'd use flap 20, then try and get below FL25 limit and extend that. Once you've got FL25 out, the aircraft will slow down quite rapidly as it is a large drag setting.

                                Comment

                                Working...
                                X