This is how DSB determined the location of launch and the explosion of the BUK missile

Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

This blogpost provide some insight on the way Dutch Safety Board determined the location of the explosion of the BUK missile. Based on the location of the explosion, the location of where the missile was launched from can be calculated.

Around July 2016 the JIT will announce the exactly launch location. This location could be determined by satellite photos, SBIR data and calculations.

To be able to verify the JIT conclusion on where the missile was launched from, we need to understand how JIT likely did the calculation.

The main issue in calculating the location of both explosion and launch is that some parameters are not exactly known. So this is why DSB calculated an area of 320 square km as possilble launch location, while Kiev Research Institute for Forensic Expertise calculated a very small area.

Assessment of damage pattern

DSB performed an assessment of the damage of the aircraft other skin. They looked at which part of the fuselage was hit by fragments and which part not. The border between area’s  hit and area’s not hit is important.

NLR calls this “the search for a  ribbon of perforations over the aircraft contour”.

To establish the border provides information on the location of the missile when it exploded.

Another part of the assessment is trying to find out if certain area’s had specific shapes of holes. Each warhead has a kind of fingerprint. DSB stated that a ribbon on the righthand side of the cockpit was hit by a relative large number of bow-tie shaped fragments.

Also DSB looked at the pitch. The pitch is distance between holes made by fragments. The idea is that because of a warhead with preformed fragments a certain pattern can be observed. The trajectory of the preformed fragments can be calculated. Based on the pitch distance between warhead and aircraft can be calculated.

DSB also studied the damage of the blast caused by the explosion on the aircraft fuselage.

DSB found a couple of bow-tie shaped fragments. DSB concluded based on the bow-tie that the only weapon which could be used was a 9N314M warhead.

Probably DSB then reversed enginered the location of the explosion based on the found bow-ties.

The 9N314M warhead has a couple of characteristics which determine the damage pattern on the warhead.

  • the shape of the warhead. The warhead is barrelshaped. This leads to a certain minimum ejection angle and maximum ejection angle.
  • the location of the detonator. This determines the ejection angle. If the detonator is located in the front of the warhead, fragments will be pushed out more backwards then when using a detonator located in the back or middle
  • preformed fragments. The 9N413M warhead has preformed fragments. This means each fragment has a distinct shape and is put loose into the warhead in a pattern. The alternative is pre-ready fragments. In this situation the fragments are not loose but are fixed to eachother. The force of the explosion creates random shaped fragments.
  • weight of the fragments. Heavy fragments have higher speed of the explosion due to the kinetic energy. Higher speed of the fragment combined with the speed of the missile mean a different distribution of fragments compared to light fragments.
  • number of light and heavy fragments

Other parameters which are important in locating the location of the missile when it exploded are :

  • speed of the missile
  • speed of the aircraft
  • the logic of the proximity fuse. A BUK warhead has a proximity fuse. It detects that it is near the target and will then detonate the explosives. The conditions for detonation tell more about when and at what location the BUK missile explodes.
  • direction the aircraft flew in relative to the ground (not the magnetic course but the true track)
  • direction of the missile relative to the ground
  • type of explosive used in missile
  • density of the air. This has a result on the drag of the fragments and this speed. However the effect is quite low
  • pitting. The damage caused by hot parts of the explosives
  • soot. The black residue of the explosion

The speed and altitude of the aircraft has an additional effect. After the explosion the pressured cabine will put force from the inside to the outside. Additionally the forces of the aircraft caused by the speed and the wind will have an impact on damage patterns.

How to calculate the velocity of the fragments

The speed of the fragments after the explosion can be calculated using the Gurney equations.
(Enter: cyclotol, 33, 37, cylindrical. AFAIK Buk warhead uses a TNT/RDX mix)

Location characteristics

Based on all these parameters the location relative to the nose of the aircraft can be determined. Characteristics of the location of the missile are:

  • x= location relative on the horizontal axis of the aircraft (in front or behind the nose)
  • y= location on the horizontal axis (left or right of the nose)
  • z= location on the vertical axis of the nose (up or below the nose)

Then the elevation and azimuth of the missle are determined by DSB.

The elevation is the angle of the missile when it exploded relative to the horizon.

The azimuth is the angle of the missile relative to the course of the aircraft.

2 D Simulator of fragment distribution

To get a basic understanding of the effect on fragment distribution this is a nice 2D simuation. Mike West made this simulation. It can be used to see the effect of missile spead, aircraft speed, spread on distribution of fragments.


TNO used commercial closed source software named Split-X to calculate the distribution of fragments on a 3D model of a Boeing 777. TNO used three different models of a warhead. Each model had different characteristics like ejection angle and number of light and heavy fragments.

How to determine launch location

A user called Eugene made a nice animated GIF about the process of determination of the launch location. By publishing  I don’t necessarily agree with the finding.

Eugene used a track course of 123 while in fact the aircraft course on a flat map was 119 degrees.

Open source calculation of launch spot

Someone with nickname Unit0 wrote a VisualBasic code to determine the launch area. He used several parameters. In a future post I will try to explain all the parameters. The result of this calculation differers from the results of the Eugene method.

For now I show the thread at Metabunk which shows the calculation.

Unit0 made a new calculation in April 2016. The result is shown in the image below. The black area is the most lilely launch location.

The center of the area is south of Red October/ Chervonyi Zhovten


Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

10 Comments on This is how DSB determined the location of launch and the explosion of the BUK missile

  1. Charles Wood // April 26, 2016 at 12:57 am // Reply

    I’m curious why all models assume a launch from the South.

    What specific reasons exclude the missile being launched from the north on a reverse track?

    I assume the damage pattern would be very similar?

    • This is a good question. It is possible to make practically the same side-arm of the shrapnel beam with a missile coming from the other direction. The difference is that the rear-facing jet of shrapnel is much less intense that the forward one. This is because the missile speed is subtracted from shrapnel speed and because the forward spray is much more intense due to the warhead firing slightly backwards (if we take Buk). The conservation of momentum says that the forward bit of the missile needs to be intensely thrown forward to compensate for the warhead shooting slightly backwards. As we see quite a lot of damage on the left wing, it is more likely that the missile came from ahead or right.

    • The SBU phone intercepts said Cossacks from Chernukhino did it, and they are in the northeast of Hrabove. I wonder why the SBU went with that story initially until they had the Aleynikov Snizhne smoke trail?

      • On 17 July, SBU just used whatever intercepts they had at hand, regardless of the relevance of those intercepts. SBU was anxious to provide “evidence” for the separatists-did-it story and had no time to be choosy. Along with the Chernukhino Cossacks intercept, on 17 July SBU published an intercept about Sapper’s group. It is very likely that the inercept was old, as in early July Sapper’s group retreated from the town of Slovyansk and thus the separatists considered Sapper a traitor.

      • In a 25 July article
        AP journalists Yuras Karmanau and Peter Leonard wrote about a top rebel official (unnamed) who told them that, indeed, it was rebels who had shot down MH17. In particular, that official said that “Sapper had been sent that day [17 July] to inspect three checkpoints – in the towns of Debaltsevo, Chernukhino and Snizhne…At some point in these travels, he joined up with the convoy accompanying the missile launch system.”
        My guess is that the purpose of this alleged “confession” was to combine the in fact incompatible stories of the Chernukhino Cossacks and of Sapper’s group. It would be funny if not too sad: Sapper “joined up with the convoy accompanying the missile launch system.”

    • For a missile to be able to destroy an aircraft, proximity fuse must work certain ways. A-A described insane way would cause misses of real military targets.

      According to A-A, if launched from snizhne the whole coskpit would have been blown away. That exactly happened to MH17.

      Oversimplified: If launched from Z, the proximity fuse react ~17m before the target.
      If launched from snizhne the proximity fuse react ~5m…7m before the target.

      • “Without the full fusing logic/algorithm decoded no one knows how any BUK missile’s proximity fuse will react to such a large fast flying target such as a commercial passenger plane.”

        From Z direction I’m pretty sure BUK would spray the right side engine & fuel with shrapnel. In the most RU favorable (silly)case it would see forward fuselage before the engine.
        It simply does not make sense that the missile would fly 6m by/over/past the target before warhead detonates.
        Even more so when you consider that the shrapnel cone is slightly forward, and more so with approaching target.

      • Hector Reban // April 28, 2016 at 3:59 pm // Reply

        Sotilaspassi: In fact, when taken into account the DSB detonationpoint including Snizhne orientation and height, the right wing, right cockpit canopy and even the right windows should have been targeted.

        Notwithstanding shielding by the cockpit, a Snizhne detonation would spread preformed fragments over the cockpit towards the right wing area.

        Furthermore AA mentioned high-kinetic fragments would penetrate twice through the cockpit hull (not seen in MH17 though) and would smash the right window section (also not seen).

    • “I assume the damage pattern would be very similar?”

      It’s hard to explain without the proper graphics, but you would only get a similar fragmentation pattern from both directions in the static model, where it would be roughly symmetrical in its x, y and z axes.

      In the dynamic pattern, the missile’s velocity pushes forward the two cones that mark the boundaries of the frag spray.

      In Almaz-Antey’s ‘lancet’ dynamic model for the BUK, the angle of the rear cone is approximately 90 degrees, so it looks more like a 2-D plane than a pointy cone. That makes the damage boundary look like a straight line when viewed from the warhead. This gives a good match with the damage on the front of the MH17 cockpit, but it’s hard to analyse exactly because of the cockpit’s complex 3-D shape.

      The DSB produced a different dynamic pattern but they didn’t make it clear where they got it from. It could have been created in order to match a launch from near Snizhne.

      Their model pushes the rear cone further forward (or to look at it in a different way, its more pointy in the backward direction towards the warhead). That would make the damage boundary more curved than straight. The DSB were able to get around that mismatch by moving the detonation much further out from the cockpit, from where the curve looks more linear.

      Getting back to the idea of a launch from the opposite direction, AFAIK nobody has presented a model for its frag spray that could explain the damage to the aircraft.

  2. The map image above…
    Shrapnel match with cockpit damage becomes better, but the left engine might now get no hits from any fragment/shrapnel.

Leave a Reply to Hector Reban Cancel reply

Your email address will not be published.