Simulation versus facts of MH17 missile damage

Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

I will create a blogpost in the near future about the damage observed to the aircraft and the simulation done by DSB

DSB initially used a method to detect the location of the blast called stringing. This text has been published by BBC.

As the Dutch Safety Board prepared its report, several meetings were held with Russian specialists, who analysed the information they were given and at their last joint meeting questioned the Dutch findings. The Russian team put forward three arguments:

  • The blast location for the warhead was closer to the plane than the DSB said
  • MH17 must have been hit by a different, older type of BUK missile
  • The BUK missile must have been launched from a different area

It was time for the Dutch to respond. And, in the very final pages of the report (p94), they do.

First, the DSB says the Russians had relied on an inappropriate method to decide exactly where the warhead exploded.

The inquiry team used a “stringing method” to determine the trajectory of the missile parts that hit the plane. Specialists used three-dimensional scans and fibreglass rods and then set up a network of lines of string.

While this helped explain the general direction that fragments inside the warhead hit the plane, the Dutch insisted it could not determine exactly where the warhead blew up because that trajectory changed dramatically after impact.


I can imagine stringing did not result in the exact location. Some pieces of the cockpit roof were not used in the reconstruction! I will update this post soon with evidence.

strining method

First some impressions,

The image below shows the circle representing the location of the missile as calculated by the DSB.

However, drawing green lines from all holes in the fuselage lead to a different location.




Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

15 Comments on Simulation versus facts of MH17 missile damage

  1. In the simulation the warhead was too far from the cockpit?

    (btw. the red line is incorrectly in sim picture vs reconstruction)

  2. Maybe that’s why the cockpit roof was not placed in its actual place at the reconstruction, but in front of the engine (6th photo):

  3. admin, have you seen my big email (sent via on a related subject?

  4. Hector Reban // October 16, 2015 at 5:37 am // Reply

    Which simulation is this, admin? The WEST simulation done by the NRL?

    I think we should assess the TNO damage report (appendix X) and NLR impact simulation report (appendix Y) really thoroughly.

    For example the simulation predicts high density elements to the nose tip (in blue), then a stroke with substantially less impacts and then further to the roof low density frag impacts (in red). Can we observe these areas in the remains?

    Furthermore their launch area model – as far as I read it – doesn’t account for high density frags in the left wing. Especially this seems to be hard to reconcile with a south of Snizhne area.

    • > Furthermore their launch area model – as far as I read it – doesn’t account for high density frags in the left wing. Especially this seems to be hard to reconcile with a south of Snizhne area.

      Yes, this is another killer. It’s amazing that this particular issue is given so little attention. Only a few mentions in the comments here and there. I think, the subject deserves a separate post.

      The thing is that the left engine had quite high density of holes with a few holes having distinctive shape made by pre-formed striking elements (pages 12 and 13 of the AA report). One important property of the striking elements’ trajectories is that they are quite predictable and confined. Below certain angles from the forward axis fly simply no striking elements. However, for the Snizhne area launch, which was found by the DSB to be the likely launch location, the left engine nacelle falls right into the center of the no-striking-elements area!

      Hector, you’ve done really great job with your latest article. But it is not an exhaustive list of the report’s fuckups. As we digest, I hope, there will be a place where all the flaws of the report are collected and systematized.

      • Eugene, I think the biggest thing missed there that could account for the damage is the shrapnel from the rear half of the missile when it exploded and the other rear engine parts.
        There will be parts that MH17 flew into that would be the center of the torus or doughnut.
        And then there is no telling the direct direction the nose cone with the radar seeker would travel.
        The rear would have the most shrapnel after the warhead, but the front must have had some as well.

        Fare thee well

  5. The DSB graphic that was base for the second image in the above photo actually is tuned. In the graphic the area of detonation predicted by the simulations must be even higher:

    Table 20 page 142:
    X-axis (- = ahead of nose)
    Y-axis (- = left side)

    TNO 0.0 -2.0 3.7
    NLR -0.25 -3.0 3.7
    Kyiv Research Institute for Forensic Expertise 0.0 -4.0 4.0

    so TNO and NLR predict the area of detonation to be 3.7 m and Kiev predicts the point of detonation to be 4 m above the tip of the nose.

    The fuselage diameter of a 777 is 6.2 m, the image in the link depicts the discrepancy:

    • Wind tunnel man // October 23, 2015 at 12:42 pm // Reply

      We need to see how much shrapnel grazing damage there was on the top of the aircraft’s roof beyond door 1L before making any accurate judgements about the warhead’s position above the aircraft’s centre line.

      Also any marks, caused by warhead shrapnel, further back on the fuselage would be progressively further aft, due to their distances from the warhead and the relative speed of the shrapnel and forward movement of the aircraft, than in any static model. In a dynamic model, when attempting to locate the warhead’s position from a particular frag spread, then the further aft a shrapnel mark is the greater it’s theoretical position needs to be brought forward when compared to any static model due to the forward speed of the aircraft. The calculation is simple enough: speed is based on time and distance and the further aft a shrapnel mark is the greater it’s distance is aft of a mark predicted by a static model.

      Obviously the differences between static and dynamic positions would be small (especially in the case of the effects from a “Snizhne” launch as demonstrated by A-A in their static test,) due to the relative speeds involved, but the furthest marks away from the warhead’s position on the roof could be significant in determining the missile’s orientation and warhead position based on any verifiable frag spread characteristics. Of course the shrapnel penetration and grazing angles would be equally significant.

  6. admin,
    Your statement – Real explosion must be just above this center line – might be true if the explosive material inside the warhead was a uniform cylindrical shape.
    I am not sure it is, if looking at images of the exploded warhead.

    To me it looks like it is a shaped charge made to make the front of the warhead’s shrapnel travel out slower then the rest of the shrapnel to the middle and rear.
    Explosive charge is probably shaped like a tear or maybe even a cone.
    Although it is hard to tell what is filler in this image and what is actual explosive, I think the explosive is in the stainless steel or metal casing and the rest is insulation to keep it stable and from going off accidentally.
    Probably to give a more dense cloud for the target aircraft to fly through.
    Sort of discussed here –
    in AD’s post about a similar warhead.
    If you notice the speeds are slower for the shrapnel at the front of the warhead and faster for center and rear.

    Another item of using censor net’s image is to me it looks like, if you look at the window frames, the left (if looking at the image) frame is significantly further from the center line window frame, then what the right window frame is.
    Measuring on my monitor across the top of the window frame from the center line, I get one and a quarter inches and to the right one inch.
    That can throw off some measurements when using this image for simulation.
    To me it looks like the left side frame needs to be pushed to the right some distance ( 6 inches to a foot is a guess).

    Also, I am not sure your placement is correct, but I could be wrong and you may ave some rational for your positioning.
    To me, I would believe the overlap of two different skins should be further to the right on the plane and the center line should be further to the left of that piece (or maybe not even part of that piece at all)

    A few images where they have put it in different locations.

    And a couple of prepainted images of a Boeing’s roof.

    Fare thee well

    • My comment above is mainly trying to understand the different velocities of shrapnel in the warhead.
      Warhead has shaped explosive charge, insulating material, a detonation cord inside the insulation going to the charge at the base, electronic wiring going to the ‘wings’of the engine to turn the missile, shrapnel pieces, then a outer casing.
      Some of those things are seen, some not in the above image links.

      With the shaped charge being at the base (?) of the warhead, the metabunk jpg image makes it look like it is at the front which I am not sure of but I will accept it with confirmation, or the other end, many pieces will fly with different velocities.
      These different velocities I believe give a denser cloud of shrapnel for the targeted airplane to fly through.
      And a better success rate for the missile.
      I do not think the yellow stuff in the png image is all plastic explosive, but I guess it could be.
      ~ 20 KG of plastic explosive of the 70 kg warhead is a lot, but that much?

      Anyhow, it is a shaped charge and specially made for a cloud and I am guessing those shrapnel pieces travel outward at different speeds.

      Simply the way I see it is the BUK missile ones it explodes it creates a brick wall of shrapnel that is spreading outward in a special shape at twice the combined speed of the missile and plane.
      The plane flies into this ‘brick wall’ with the combined velocities of the plane and the missile (in a near head on collision)
      So some of the ‘slower’ shrapnel might even be going slower then the combined velocity of plane and missile, which could give a more tangent damage pattern than what you suggest above.

      I am not a missile expert and trying to learn as I go, so this is just an observation I make in a search for clarity.
      I could be completely wrong in my ‘guesses’.
      But I think that is why the DSB position of explosion is higher then what you think it should be.

      Fare thee well

  7. On second thought on looking closer at your other page with your images describing pieces of the cockpit roof, I believe your center line is approximately correct on the piece where you placed it IF the labeling on the inside of the panel is placed on the center line and the crossmember piece labels are ‘close’ to center.

    Scale or something else looks incorrect, but I do not know what it is.
    Fare thee well

Leave a comment

Your email address will not be published.