Announcement

Collapse
No announcement yet.

Polish President and wife killed in Tu-154 crash

Collapse
This topic is closed.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Originally posted by 3WE View Post
    Man, I keep hoping that you will render your opinion on my thought that maybe the pilots- under some pressure from their high-profile passengers and high-profile mission- tried to fly below their MDA using instruments- hoping for a random "break" to see the runway and land, but that like one or two other cases in the history of aviation, they ran into stuff on the ground instead? Any possibility that's what happened?
    I don't think that the busting of the minimums was intentional.

    I think that the pilot, under perceived pressure, decided to do an impossible approach in conditions that he knew would not lead to a landing, thus taking unnecessary risks, just to show that he did all he could. But I also think that his intention was just that, to go down to the MDA/MAP and then go around. The pilot stated so in several opportunities, including briefing the FO that "at 100 we'll go-around on auto"

    What went wrong, then? A number of things.

    The sterile cockpit was violated, and the CRM (or lack of) was terrible, putting too much share of the workload on the PIC and adding distractions.

    A stabilized approach was never achieved but they did not abort the approach either.

    While the pilots had inside the cockpit all the information and tools they needed not to bust minimums, the ATC didn't help with its less than accurate repeated "on course, on slope" guidance.

    For reasons that I still don't understand, the 100m call (the minimum) was called three times. The PIC and FO called for a go-around more or less simultaneously with one of those calls.

    At that point, they probably were already below 100m, and due to the unstabilized approach the were descending at an abnormally high descent rate, which diminishes the reaction time margins.

    Apparently, they attempted to do the go-around on autopilot, something for which this autopilot is not approved in this type of approach (without a glide slope). Apparently the crew was not aware of this limitation. It's still unclear (for me) if this go-around would still work (even if not allowed), in what circumstances, and how reliably. In any event, the plane didn't rotate and thrust was not added, so the automated go-around didn't work.

    Eventually but too late the pilots realized that, disconnected the automation, added thrust and pulled up. As said, it was too late.

    The rest is history. The plane hit trees below runway elevation even when it never went that down, severed a birch without even touching it, and the birch responded by severing the airplane wing several dozen of meters beyond the point of the encounter. The plane didn't roll inverted but ended upside down and the Russian soldiers shot the survivors before clearing the artificial fog.

    --- 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 Evan View Post
      BTW, Gol Airlines conducted a similar experiment colliding a 737 wing with a comparatively fragile Embraer winglet (not made of solid birch) and half their wing went missing. What exactly are these Tupolev wings made of?
      I intentionally avoided that comparison because, while I feel like you do, it's questionable.

      While a bizjet winglet is undoubtly orders of magnitude weaker that a thick birch, the speed of that impact was also orders of magnitude faster, and energy goes with the sqare of speed.

      --- 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 was able to locate one picture from Binienda's presentation about the findings relating to the flight trajectory. The red dot on the upper pic shows where according to him the wing broke off. The black line on the lower pic shows the new trajectory, the yellow dashed line shows the MAK trajectory.
        The article also repeats that prof. Robert Braun, NASA Chief Technologist, participated in the investigation.

        Comment


        • Originally posted by Gabriel View Post
          The rest is history. The plane hit trees below runway elevation even when it never went that down, severed a birch without even touching it, and the birch responded by severing the airplane wing several dozen of meters beyond the point of the encounter. The plane didn't roll inverted but ended upside down and the Russian soldiers shot the survivors before clearing the artificial fog.
          Wrong, the survivors are being held on that island with the KAL folks.
          Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

          Comment


          • A birch you say?

            Well what kind o' birch was it?

            A Yellow Birch

            A Mountain Paper Birch?

            Perhaps a dwarf American, or Sweet or Cherry birch.

            Bog Birch, River Birch, Paper, Silver or perhaps it was a Gray birch.

            My daddy always said that a birch in the hand was worth a bird in the bush.
            Live, from a grassy knoll somewhere near you.

            Comment


            • Let me throw one more thing in. Binienda maintains that the plane never descended as low as shown in the report. Some parts of the descent below the 100m alt. are irregular enough to warrant a closer look. What you see on the pic is right at the time when F/O says "all normal". The first part is 7.2 m/s - fast already. The second part is 20.8 m/s. It happens in less than a second.
              20.8 - 7.2 = 13.6 m/s. In less than one second. That's an acceleration of about 14 m/ss. The guy who was standing in the cockpit not strapped to a seat must have tested the strength of the ceiling at that point. And the F/O states calmly "all normal". The plane falls 21m (a 7 story building) in one second. The guy who is supposed to test the ceiling says "no visibility" in a normal voice. What am I missing here?

              Comment


              • Originally posted by Northwester View Post
                What am I missing here?
                Probably a matter of granularity of the data.

                --- 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
                  Probably a matter of granularity of the data.
                  Maybe if I was using centimeters and miliseconds...

                  Comment


                  • Apparently there is a parallel investigation taking place focusing on the process of desintegration of the plane. They are trying to figure out what parts separated when and what was the nature of that separation, force required, etc.

                    Comment


                    • Originally posted by Northwester View Post
                      Maybe if I was using centimeters and miliseconds...
                      The problem would not be what you are using but how the data was measured, aquired and recorded.

                      --- 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 think Binienda is wrong about the plane's altitude. There are too many indicators pointing towards the plane flying through the trees: damage to the trees, CVR, radalt readings. He should probably check carefully the data he is using. Was the data fed to TAWS correct? Hopefully he will explain in detail how he arrived at his conclusions. He might be right though about the plane loosing a part of the wing sometime after hitting the birch.

                        I checked the roll of the plane between the first tree collision and the crash. As in many cases here the data is not consistent.
                        I started looking at the radalt readings of the last few seconds (pic 1) and set 6 points for easy reference. "A" is the collision with the birch, "B" is the lowest point (3m above ground), "C" is the first point of altitude gain and the beginning of the gray zone, "D" is probably the highest altitude (slightly above 18m) and the first reading after the gray zone. Between D and E radalt shows a drastic altitude gain at the rate of 34 m/s to the altitude of 37m. It is obvious that this was caused by the rotation of the plane and not the altitude gain. The final reading (point F) shows 42.3m. This is again due to roll but at a slower rate.

                        Pic 2 shows roll and roll rate data as recorded by FDR.

                        There are several sources where roll data is available. Tables in the Polish Report, FDR plots in the Russian report, damaged trees, and implied from the FDR radalt plots. On pic 3 I attempted to reconcile all data. The first row is the data form the Polish Report. It roughly corresponds with the damaged trees, but not at every point. Then I am showing radalt readings from the Polish and Russian reports. Based on the Russian radalt values I can calculate possible roll using the likely actual altitude. When the altitude is 17m but the radalt reading is 42.3m I can calculate the necessary roll angle as shown in the "Probable Roll" section.

                        Here's the first discrepancy. The info in the reports and the tree damage would indicate a roll of over 120 deg. For the radalt to record 42.3m it has to point to the ground, so the roll cannot be more than 90 deg. For the roll over 120 deg it would take some very tall and dense trees that are not there. So if we look at this section (based on radalt readings) the probable roll would be relatively small up to point C, then increase quickly to max value of 86.3.

                        Roll values per FDR are still different. First the 5 deg roll at point B would not be possible (3m above ground), the plane would hit the ground. Then the max value recorded is 65.4 deg. Tu-154 instruments can record higher values, so why did it stop at 65.4?

                        When you look at the roll rate data, it is all over the place. Nothing consistent there. It varies from 7.4 deg/s to 74 deg/s. The FDR roll data stays at 18.2 deg/s for most of the time. Interestingly roll rate increase begins before the plane even hits any trees.

                        When you look at the vertical acceleration plot (red line at the top of the pic) you will notice that the values increase about where the plane first collides with a tree, but the max values are at points 3, 4 and 5. Binienda maintains that the wing separated around where I am showing point 4. This would comfirm his assumption. That's why I tend to agree that most likely the birch did not cause it but it happened later.

                        There is of course the gray zone question, and the alt readings at TAWS 38. Data granularity?

                        Comment


                        • I don't know. Now what?

                          Granularity of the data (in terms of roll angle and time), diference in such granularity between different sources, sync problem between sources, sensors and instruments not working very well while and after clipping trees, Russians and Poles tampering with the data.

                          Maybe we'll know someday. Maybe we'll never know.

                          But I'm much more interested to learn how they got there in the first place (these pilots in these seats, the plane in the known-impossible approach, and below the MDA) and why they failed with the go-around. This is waht might prevent future occurrences. Understanding the dynamics of the plane and sequence of the structural fragmentation after hitting trees won't. "My" investigation ends the instant that the plane hits a ground obstacle.

                          --- 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
                            But I'm much more interested to learn how they got there in the first place (these pilots in these seats, the plane in the known-impossible approach, and below the MDA) and why they failed with the go-around. This is waht might prevent future occurrences.
                            What will prevent future occurrences is the fact that these flights are now conducted by well-trained and experienced LOT pilots, and a very stark lesson has been engraved in the minds of arrogant VIP's who might want to pressure pilots to execute unsafe approaches... and the fact that Lech Kaczyński will never be anywhere near a cockpit again.

                            Comment


                            • You know what I mean. I'm not Northwester.

                              --- 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
                                I don't know. Now what?

                                Granularity of the data (in terms of roll angle and time), diference in such granularity between different sources, sync problem between sources, sensors and instruments not working very well while and after clipping trees, Russians and Poles tampering with the data.

                                Maybe we'll know someday. Maybe we'll never know.

                                But I'm much more interested to learn how they got there in the first place (these pilots in these seats, the plane in the known-impossible approach, and below the MDA) and why they failed with the go-around. This is waht might prevent future occurrences. Understanding the dynamics of the plane and sequence of the structural fragmentation after hitting trees won't. "My" investigation ends the instant that the plane hits a ground obstacle.
                                Unfortunately it seems that in this case the dynamics of the last few seconds holds the key to the understanding of this crash. According to Binienda (he is not working alone though - I mentioned before prof. Braun, also some of Binienda's colleagues from Akron, and some friends from MIT) the trajectory was different. They used the same calculation method as MAK but added TAWS # 38. The plane never touches the trees till after a violent roll when the wing fragment separates. So the damage to the trees on most of the pictures is real, but the birch seems to be fictional. Pic 1 shows the assumptions, pic 2 shows the trajectory (solid line - center of gravity, dashed line - nose of the plane), pic 3 shows the trajectory in equal horizontal and vertical scales.

                                Comment

                                Working...
                                X