Overview of all doubts on DSB final report

Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

This post is a landing page for all posts published on this website about doubts on the DSB final report.

  1. DSB final report doubts: no bow-tie shaped damage on outer skin
  2. DSB final report doubts: no drift angle
  3. DSB final report doubts: location of explosion
  4. DSB final report doubts: does a 9N314M warhead fit on 9M38 missile ?
  5. DSB final report doubts: missing wreckage of cockpit roof
  6. DSB final report doubts: damage of left wing and left engine nacelle
  7. DSB final report doubts: no mentioning of vertical limits Rostov radar
  8. DSB doubts: mistake in wind speed graph label
Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

54 Comments on Overview of all doubts on DSB final report

  1. Clement Townsend // April 18, 2016 at 5:27 pm // Reply

    Will you be commenting on the DSB’s method of determining the launch site of the alleged buk missile?

    • I am not sure yet how the 320 sq km launch site was determined. There are many variables which are input for modelling of the damage. I hope the JIT will open source that but I guess that will not happen.

    • sotilaspassi // April 20, 2016 at 11:05 am // Reply

      + I would like to see study that explains the position and angle of the warhead vs other information.
      Mainly vs: https://drive.google.com/file/d/0BxNz0P5oVk2wSnMzME12cjdoNkk
      and vs: https://drive.google.com/file/d/0BxNz0P5oVk2wLWt3OFkxSkFKNU0

    • I’m not very happy about it, but I found the combination of parameters giving the launch are similar to the DSB.

      I take my old calculation – https://www.metabunk.org/mh17-missile-plane-intersection-simulation.t6920/

      I adjusted for magnetic declination.
      double beta_nominal=(az+(180-115-8 /*magnetic declination*/));

      I used the coordinates of the point of explosion from albert-lex
      http://albert-lex.livejournal.com/68374.html
      x0 = -1,5…-1,9 м;
      у0 = -0,8…-1,3 м;
      z0 = 1,8…2,2 м.

      On DSB axis:
      Point3D blast_point;
      blast_point.X=1.05; //-0.7;
      blast_point.Y=-1.7; //-3.5
      blast_point.Z=2.; //4.2;

      And I take warhead rear cone angle from IL-86 test.
      http://unit0.livejournal.com/1120.html
      double cone_zero_velocity=120.5;
      #define frag_speed 1617

      Result: http://s017.radikal.ru/i408/1604/64/32f2950401e7.png

      The center of launch area sometimes alleged kilometers south of the Red October village.

      • >#define frag_speed 1617

        Have you used 1617 m/s for fragment speed?

        • Yes. I use data from AA.

          http://s11.radikal.ru/i183/1604/59/19ddfc40ece2.png
          For missle velocity V=600 m/s, maximum ejection angle 105.

          http://s020.radikal.ru/i708/1510/a4/85759f54cff2.png
          For V=0 m/s, maximum ejection angle 126.

          So I get speed of fragments for rear cone 1617 m/s, that give angle changing from 126 to 105.

          It may be wrong value, because I use maximum ejection angle 120.5, but I dont recalculate fragments speed.

          • >speed of fragments for rear cone 1617 m/s

            unit0, Almaz Antey gave the value of 2400 m/s for the fragment speed: http://savepic.ru/9484044.png (for presentation sake AA used a non-uniform axis scales on the slide)

            If you don’t trust them, you can estimate the fragment speed with the Gurney equations
            http://www.un.org/disarmament/un-saferguard/gurney/
            (Enter: cyclotol, 33, 37, cylindrical. AFAIK Buk warhead uses a TNT/RDX mix)

          • http://s020.radikal.ru/i707/1604/04/131f7b2c355a.png

            The speed of the fragments in different directions is not the same, as you can see in the diagram. To find the starting point critical speed is speed of the fragments forming the rear cone of the explosion, they set the border of damage area on the fuselage of the Boeing.

            I found speed of this fragments as early writing.

          • unit0,

            > To find the starting point critical speed is speed of the fragments forming the rear cone of the explosion, they set the border of damage area on the fuselage of the Boeing.

            The rear cone in the static case looks like it has fragments with lower speed, but after adding the relative plane-missile velocity, the fast fragments and the slow fragments get mixed in on the rear edge. See the top right insert in the slide. Anyway, I don’t know much details about your simulator.

          • >>The rear cone in the static case looks like it has fragments with lower speed, but after adding the relative plane-missile velocity, the fast fragments and the slow fragments get mixed in on the rear edge.

            The AA data referred to the static and dynamic event:
            rear cone for static – 126 degree

            rear cone for dinamic (600 m/s) – 105 degree
            http://s11.radikal.ru/i183/1604/59/19ddfc40ece2.png

            The speed of rear cone fragments must be lower then 1617, else the rear cone in dynamic (600 m/s) will be more then 105.

          • >>fast fragments and the slow fragments get mixed in on the rear edge

            The speed of fragments influence on two thing – angle of rear cone of explosion, and the time of fly of fregments from warhead to Boieng. The speed of 1617 make first and must important like true.

      • unit0: your calculation is very interesting but I have trouble to understand how you calculate this. Can you explain to me in high level which parameters you used, what sources you used and what the colours on the map exactly mean? Would be good to be able to verify what the JIT will present in July on the launch location.

        • It is very strange result. Tne main is a new explosion coordinates from Albert-lex that so different to other. I do not think it can be called verification, while JIT will stick to coordinate similar to the DSB report.

          Ask what exactly is unclear.
          About colours, I use radius (R) of sphere with 1m3 for normalization.

          The black colour I use when a difference between the blast coordinats for “best matching” and given blast coordinates lower then 1/2 * R

          Red for lower then 1*R, yellow for 2*R, green for 3*R, white for other.

        • If I understand correctly (there’s a language barrier) unit0 has calculated a launch point at a position a little south of a village named Red October.

          This village (“Chervonyi Zhovten” on some maps) is located 7 km south of Snizhne, in the black zone of the map that unit0 posted at http://s017.radikal.ru/i408/1604/64/32f2950401e7.png

          Per unit0’s calculations the launch point is most likely in the black zone followed in order of probability by the red, yellow, green and white zones. Unit0, please clarify if I have misunderstood.

          I recall that Red October is where Reuters, Bellingcat and Kyivpost say that witnesses saw a missile in flight that, according to one witness “it wiggled around, then some kind of rocket stage separated, and then, somewhere toward Lutuhyne, Torez, I saw the plane fall apart in the air.” Please, consider the sources of this questionable report.

          http://www.reuters.com/article/us-ukraine-crisis-airliner-idUSKBN0M81XF20150312
          https://www.bellingcat.com/news/uk-and-europe/2015/03/13/vladmir-babak-and-reuters-missile-launch-witnesses-how-do-they-fit-with-what-we-know-about-mh17-so-far/
          http://www.kyivpost.com/content/ukraine-abroad/reuters-from-red-october-village-new-evidence-on-downing-of-malaysian-plane-over-ukraine-383268.html

          A serious problem with the albert-lex explosion coordinates is that a 70 kg warhead would have an even more devastating affect than we have seen in the pictures. If albert-lex is right that the coordinates were that close to the cockpit then he is also right that the missile was not a Buk.

          • >A serious problem with the albert-lex explosion coordinates is that a 70 kg warhead would have an even more devastating affect than we have seen in the pictures.

            That’s the dilemma, isn’t it? You either have to accept that it was not a Buk, or move the detonation point away. This was the exact choice the DSB had, and we know what way they’ve gone. The problem is that then it’s very difficult to come up with the plausible reasons why stringing was not used (the DSB’s lame excuse was that the lines were not striaight) and why hits from their point covered a lot greater areas that were observed as damaged on the wreckage.

            As to the unit0 calculations, myself a Russian speaker, did not get much further. At a first glance he uses at least a very incorrect fragment speed (1600 vs 2400 m/s). And I am pretty sure he does not know necessary flight characteristics, because they are not public (I have other questions about the simulation)

          • IsThatSo // April 22, 2016 at 1:06 pm //

            I believe that if the cockpit and business class were reconstructed as accurately and completely as possible then notwithstanding the slight curvature the strings will not lie.

            Eugene, if a machine translation into English of the albert_lex report was available, then would you have the time and inclination to correct the translation errors and put it into readable prose for English speakers?

          • sotilaspassi // April 22, 2016 at 2:06 pm //

            >70 kg warhead would have an even more devastating affect than we have seen in the pictures.

            We see with IL86 what happen if missile detonates 4m+ from the tip of the nose of the airplane. Very little structural damage.

            In MH17 we see that most of the upper part of the forward fuselage was completely blown away, into so small pieces that big part of it is still not found.

            A-A stated: “the entire front end of the cabin would have been blown off”
            That is what we see with MH17. Not with the tiny damage of A-A demonstration.

          • I believe mmost of the rooftop parts were found/seen and or photographed. DSB however failed to recover most part because of the situation being unsafe to perform a recovery.
            Also parts probably were not on the shopping list of DSB and were left in Ukraine.

            It does not make sense to compare the amount of damage seen to MH17 to that of the IL86. The force of forward speed, wind, cabine decompression must have been enourmous.
            The goal of the AA simulation was to show the distribution of the damage (where are holes, where are not) and the size and shape of holes.

          • > The force of forward speed, wind, cabin decompression must have been enormous.

            The force by an upcoming air was indeed big, and caused the cockpit separation. But it’s worthwhile to get a better idea of the forces involved. The dynamic pressure by the upcoming air is around a tenth of an atmosphere. That is if you stick your hand out of the window and turn it across the stream this is the pressure you’ll experience. You may be able to hold your hand perpendicular to the stream, but barely – the force on it will be around 20 kg. You’ll be able to hold your hand quite easily if it is aligned with the air stream. This should give you some feel. There was an interesting near-crash of an airliner, when a pilot was sucked out of the window, but the flight attendant managed to grab him by the legs. The plane landed with the body of the pilot hanging out of the window. He survived.
            https://home.bt.com/news/world-news/june-10-1990-miracle-of-ba-flight-5390-as-captain-is-sucked-out-of-the-cockpit-and-survives-11363985642960

          • > if a machine translation into English of the albert_lex report was available, then would you have the time and inclination to correct the translation errors and put it into readable prose for English speakers?

            I’d love to, but the time is a problem. I have to proofread my own texts to get them in shape (so some replies may look to be in good English while other quick ones in bad) and do occasional translations for my relatives, so I know that it’d take a lot of time to translate the big text by albert-lex.

            I’d like to say that on previous read, alber-lex analysis seemed to be proficient. My mistake then was that I dismissed it because he said that the missile homed in onto the 777’s weather radar. That time I thought it was impossible, but later learned that Ukraine does have a passive seeker on R27 capable of going after radio emission sources such as plane own radars (=not requiring a separate illumination by antennas such as TELAR or Kupol).

          • Yes, but the strange thing is, as I understand correctly, he uses the “close” detonationpoint and gets about the same results as DSB.

          • BTW, this was a reply to the root post from IsThatSo

          • Hector, simulations need to be checked very carefully to be trusted. I’ve dealt a lot with various simulations and can tell this with confidence. You can try checking unit0’s simulator either against common sense or by examining the code. As to the second check – try doing it yourself, you’ll stumble pretty soon. As to the first check – there are questions. For example, how come points on the left of the plane’s track get the same probability of being a launch location as points far on the right, like you see on the picture:
            https://www.metabunk.org/attachments/7fab1450eb4d99a704c5695ab9313440-png.15681/

            The black strip, the strip of the highest launch probability apparently, does not make much sense. An honest author of the simulator should be able to give an appropriate explanation for the phenomenon. If no plausible explanation is given perhaps there is a mistake in a method.

          • sotilaspassi // April 23, 2016 at 3:53 pm //

            thick steel bars and beams were cut into small pieces in mh17

            So please, do not be silly, no wind or cabin pressure does that.

          • Hector, tried to check unit0’s code too. Leaving aside that the code is best suited to scare children, nowhere I found where the data describing the damaged area on the plane is defined. I suspect unit0 uses some very simplified criterion. Unfortunately, the description is very poor. E.g. the description of matching point contains the sentence “let’s assume that match point arbitrarily located on rear cone at some distance (rho) tovertex and at some angle (phi) from down to up”. Ok, his language is uncomprehensible in such critical places, so I hoped Russian comments in the code would help – but there are none. So, the only way to look at his simulator is to take is as a black box: one needs to pass the output past a common sense, without regard to what’s been boiling inside. The first question: why the best lunch location candidates, the black strip, has such a weird shape, e.g. a launch location on the left has the same chances as points quite far on the right? I’ll have more questions if needed.

          • > the code is best suited to scare children
            That was a bit harsh on my part. unit0 is not a professional programmer, my code after uni looked like his.

          • It seemed to me that I made a calculation simple as it possible, I do not even use the dot product.

            There is an easier way to calculate (and fast), but it requires a lot of basic knowledge.

            Russian text available here: http://unit0.livejournal.com/
            Also you may ask me on email unit0#mail.ru.

          • Can you explain in a few steps how your calculation is done?
            What are your assumptions? What is the source?

            For example, how did you determine the minimal and maximum distance of the launch location to the explosion point?
            How come certain area’s are more likely than others (expressed in different colours)

            It is very important as many people as possible understand how the launch location is determined so we can provide some feedback to JIT in July.

          • One should be skeptical if an automated matching method comes up with a missile incoming direction very different from the displayed on the picture http://uploads.ru/kC1dN.gif

          • Eugene: thanks for the animated gif. I included it in the post explaining how DSB determined the launch location.
            Can you tell me why you use a track heading of 123? This should be 119 right?

          • admin,
            >Can you tell me why you use a track heading of 123? This should be 119 right?

            I think, 123 is where the planes body was directed, so is probably right. Check against Brendan’s comment:
            http://www.whathappenedtoflightmh17.com/dsb-final-report-doubts-no-drift-angle/#comment-16919

          • The path over the ground in which MH17 was traveling (its track) was 119 degrees, but the direction in which it was pointed (its heading) was 123 degrees. The drawing in the NRC article that you mentioned shows those angles, although the difference between them looks a bit exagerrated.
            http://webcache.googleusercontent.com/search?q=cache:g30rbBNv9CsJ:www.nrc.nl/nieuws/2016/04/16/boekraket-had-mogelijk-ander-doel-dan-mh17-1609314+&cd=1&hl=en&ct=clnk&gl=nl

            The pilot of MH17 had pointed the aircraft four degrees off-course, around towards the southwest where the wind was coming from. He would have done that deliberately, knowing that the wind would blow it back onto the correct track. So the wind altered its track but not the heading in which its longtitudinal axis had been pointed.

          • The direction the nose of the aircraft was pointing to relative to the ground (not magnetic heading) was 119 according NRC. Or the angle between the nose and the meridian was 119.

            De ware koers (ten opzichte van de meridianen) van de Boeing was daarom 119 (115 + 8 – 4) graden.

            The true course (relative the meridians) of the Boeing was 119 (115+8-4) degrees.

          • Course is different to heading. The course is the intended path to be travelled (a pilot or sailor sets a course at the start of a journey). It’s therefore the same as the track if the calculations are correct, the aircraft is working properly and the wind does not suddenly change speed or direction. Heading is only the same as track and course if there is no wind, or at least none from the side. Normally the heading is slightly different.

          • Another way of putting it. Plane’s compass was measuring the angle of 115 degrees, which got recorded in the black box. The compass does not know anything about the course or wind, the only thing that it can measure is the angle of the plane body wrt magnetic lines. The magnetic lines are additionally rotated by 8 degrees (magnetic declination) relative to the direction to North. Therefore the plane’s body was rotated by 115+8=123 degrees off North. NRC could have been mistaken then. Ole long ago dug up in the documentation that it is the angle wrt magnetic lines that gets recorded by the flight recorder.

          • >>Can you explain in a few steps how your calculation is done? What are your assumptions? What is the source?

            Yes, but it take a lot time. I use only AA and DSB data.

            >>It is very important as many people as possible understand how the launch location is determined so we can provide some feedback to JIT in July.

            You should not think that last result is OK. Now I think you’re wrong about magnetic declination.

            >>The true course (relative the meridians) of the Boeing was 119 (115+8-4) degrees.

            1. I think 115 include magnetic declination already.
            2. You use a wrong sign for drift angle.

            The angle between heading (main axis of Boeing) and track (path on earth) is known as the drift angle.

            Axis-Path=Drift
            115-119=-4

            In your interpretation A is 115+8=123 and Path-Axis=123-119=-4

            So, when in my calculation I added magnetic declination I make a mistake and get path angle 115-(-4)+8=127.

          • 119-123=-4

          • He-he, here is a puzzle for you.

            Suppose we have three numbers 115, 119, and 123, but we cannot decide which one is which. As far as I can see no-one mentioned other numbers in this global confusion, so we are limited to just these three.

            Can we decide which one is the plane body heading, which one is magnetic heading (the compass reading) and which one is the course, just by knowing the fact that the magnetic declination in the area is 8 degrees?

            I’ll leave you to muse about this simple puzzle.

          • There is not puzzle. Right is path(119 known from radar data too), Axis(115).

          • >There is not puzzle. Right is path(119 known from radar data too), Axis(115).

            Ok, we agree on the course – it’s 119. Because the other two should add or subtract to 8. This at least confirms that NRC were incorrect in using 119 for heading (this is not a major mistake though).

            It remains to decide whether the body heading the was 115 and magnetic heading 123, or vice versa.

            We can do this either by looking into the definition of magnetic declination or by looking into documents.
            Check the definition magnetic declination https://en.wikipedia.org/wiki/Magnetic_declination

            Applied to our case:
            heading = magnetic heading + magnetic declination

            Also check the original post of Ole, where he cites the DSB preliminary report
            https://www.metabunk.org/does-damage-to-mh17-indicate-or-exclude-a-particular-buk-launch-location.t6345/page-16#post-159324

          • >>Applied to our case:
            heading = magnetic heading + magnetic declination

            The last calculation was maked with magnetic declination, but now I doubt about it.

            Your interpretation contradicts the drift angle definition. So draft angle must be +4. May be DSB make sign error in report?

          • I don’t know whether NRC made a mistake or whether there was some confusion about the translation of the terms course and heading. But someone who can’t understand Dutch would get the impression that they’re getting the two terms mixed up here:
            “De magnetische koers (‘heading’) was volgens een tabel 115 graden …”.

            However, Google translates this without mentioning the course:
            “Magnetic heading ( ‘heading’) was, according to a table 115 degrees …”

            It translates ‘koers’ as heading in another sentence too.

            I’d be surprised if people as dependent on shipping, like the Dutch, don’t have separate words for course and heading, like in English.

          • I was wrong.
            http://www.ohio.edu/people/uijtdeha/theavionicshandbook_cap_15.pdf

            Drift angle (DA)
            DA=Current Track-Current Heading

          • > So draft angle must be +4. May be DSB make sign error in report?

            The sign of the drift angle is not that strictly defined. Just checked wikipedia and it does not clearly states what gets subtracted from what: course from heading or vice versa. For our case it suffices to know that the wind was blowing from South/South-West to determine that the course was 119 and heading 123.

          • Discussion on drift angle, heading etc in this thread please
            http://www.whathappenedtoflightmh17.com/dsb-final-report-doubts-no-drift-angle/

  2. sotilaspassi // April 20, 2016 at 10:56 am // Reply

    I would love to see a document published with full list & photographs of metal items found that do not belong to Boeing777.
    According to DSB, some of that information is secret because of the criminal investigation, but one day, I wish …

  3. “If I understand correctly (there’s a language barrier) unit0 has calculated a launch point at a position a little south of a village named Red October.
    This village (“Chervonyi Zhovten” on some maps) is located 7 km south of Snizhne”

    Bellingcat decided not to run with that launch point because it would have proven beyond doubt that the famous “launch smoke photos” were not a launch because the smoke was in the wrong place and would therefore proven to have been fakes.

  4. Eugene, thanks for your explanation. I had some doubts because a closer detonationpoint lead to the same direction of the launch and the areas unit0 calcalates give the same probabilities for positions roughly from north to south witcare in fact located kilometers apart from eachother.

    The simple graphic explanation you provide would lead us to a site – the Velyka Shyshivka area – which is interesting on two accounts.

    1. It matches a site of heavy shelling by the Ukrainians after they left the area, according to the report drawn up by “Andrew” also available on this site. They did this possibly to cleanse some awkward things, because the area has no strategic value whatsoever and at the time no separatist presence.

    2. It matches an IMHO logical military strategic position towards the Russian border to shiled off an attack force which was heading from Amvrosievka to the Marinovka border area to win back positions lost on the 16th, as a Ukrainian dispatch service mentioned.

    • > The simple graphic explanation you provide would lead us to a site – the Velyka Shyshivka

      Not necessarily. The matching by aligning the damage area has quite significant error, maybe around 15 degrees. Add to that an additional error due to possible missile maneuvering, which is estimated to be another 10 degrees. That gives quite a range for directions towards the launch site. What I tried to demonstrate is that if someone uses similar methods and comes up with Snezhnoe, they most likely have been tweaking their technique for a predetermined launch location. The graphic shows that it’s easier to stretch your data to get to Z than S, if one wants.

      Trying to be reasonably unbiased, the gif was created during the process, without going back and redoing steps. I encourage other people to repeat the work themselves, so they don’t have to take anyone at their words.

      Rigorous readers might spot a minor mistake. I used the fragment distribution diagram for S, but the direction turned out to be closer to Z. If I used a correct distribution pattern, the resulting blue direction would be even closer to Z by 3 degrees. That mistake was deemed insignificant and had been ignored, it would be dwarfed by the above mentioned errors anyway.

      Also one has to remember that matching on a 2D view, although being a good start, should not replace a 3D matching. 3D matching will give a slightly different heading towards the missile, but is harder to do for us enthusiasts – it needs a professional setting (e.g. DSB would be in a better position to determine the direction with the the greatest density of holes).

      Again, the whole point was to show that if someone comes up with S as a lunch location, they’ve been likely biased. As probabilities go, S cannot be completely ruled, but if someone has to chose between S and Z, Z would be a lot more probable candidate for a launch location.

      PS: I don’t believe that Buk was the kill weapon, but the method applies to AAM missiles too.

    • > the method applies to AAM missiles too.
      A refinement is needed. Should read “the method applies to AAM missiles with fragmentation warheads”, which is our case (rod-based AAMs have a lot slower fragment speed).

  5. sotilaspassi // July 26, 2016 at 6:10 am // Reply

    What is the exact spot of the left wing pictured here:
    http://uploads.ru/HYP7f.jpg
    And do we have photo of the upper side/front of the wing piece?

    (a lot of penetrations it seems + most weapon fragments have remained inside the wing)

Leave a comment

Your email address will not be published.


*