Analysis of the smoke plume photos

Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

A single person was able to photograph what could be the smoke plume caused by the BUK missile that shot down MH17.

Two  photos were made by someone living in Torez. The Joint Investigation Team has accepted the photos for usage in their investigation. Dutch RTL Nieuws reported in December 2014 about these photos.

In a serie of blogposts I will analyse the photos. I will focus on three aspects:

  • are the photos authentic? How likely is that? Who are the experts who analysed the photos. What methods and tools did they use?
  • could the photos be made at July 17 2014 at the time of the launch?
  • do the photos show a smoke trail caused by a BUK missile?

Lets first start with possibilities.

  1. the two photos are genuine. There has not been any digital manipulation and the photos were made around 16:20 on July 17 2014. What we see is black and white smoke from a BUK missile. This is what Bellingcat believes.
  2. the two photos are genuine. There has not been any digital manipulation and the photos were made around 16:20 but on a different date. The black and white smoke show a smoketrail of a BUK missile
  3. The two photos were made at July 17 2014. The black smoke was indeed visible at around 16:20. The white smoke trail suggesting a missile was added by Photoshop or similar tools
  4. the two photos are genuine. There has not been any digital manipulation and the photos were made around 16:20 but on a different date. The white smoke was caused by a different type of missile or rocket. For example a Strela-10 missile

In the next days I will discuss each possibility.

Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

21 Comments on Analysis of the smoke plume photos

  1. Dark smoke was not from a BUK missile in any case. It was issuing from a stationary source during some period of time.

  2. Mr.Bushkin // June 29, 2016 at 9:03 am // Reply

    Well, here is quite a long video with a Buk originated plume for comparison:

    • There is a better place for your video, and, in fact, it’s already there:

      While I am here, there is a thought. The trail on the lower altitudes (below 3-5 km) is just smoke and disperses a little differently from the trail further up, which, in addition, is formed of ice crystals and is called contrail (condensation trail).

      Contrail stay time up there, as the AA expert says, greatly depends on the weather: it takes time for the ice to sublimate, and this is weather (temperature and humidity) dependent.

      On the other hand, the trail stay time on the ground level, I believe, mostly depends on the air turbulence, which directly depends on wind speed only.

      So, as the day was relatively windy and warm, and the visible part of the trail was just smoke (not condensation trail) we should expect the dispersion time to be rather shorter than longer among those we see on the launch videos.

      • Daniel Been // June 29, 2016 at 2:19 pm // Reply

        Not sure if “just smoke” is the right description here. The resulting chemicals are just different during ignition and flight sections (the dual-mode aspect of the two propellants). During flight it’s more like hot gas and indeed we get a lighter contrail, with a dispersion rate which is ultimately dependent on many factors, as you rightly point out. At ground level we’d have not only different turbulence but also likely a different type of fume, including some obvious dirt and dust violently being blown up but also the more powerful ignition: probably heavier elements. This development has also a tactical reason: you don’t want a big arrow pointing to your launch vehicle hanging around for long in a war zone. The more modern the SAM/AAM, the less distinct or dark plume you’ll notice.

        This could mean we’re looking at wind gradients
        ( example: ) or perhaps some “boundary layer” effect where wind speeds can suddenly change too. Think of climbing a hill, just a few 100 meters up and you could experience significant “jumps” in wind speeds and gushes.

        The darker colored gasses/dust cloud could be heavier and thus less transported even with a similar strength wind. But when taking into account the gradient effect one could see how after a few minutes “two plumes” from the same launch could easily form. Keep in mind that this is a thought experiment just to take into consideration. Not sure how to falsify it with the limited data set.

  3. Thanks.

    > The darker colored gasses/dust cloud could be heavier and thus less transported even with a similar strength wind.

    Heavier dust will still be blown with the local wind speed. The time for a particle to assume the wind speed is minuscule (particle inertia is minuscule). So it’s wrong to think that heavier particles will be less affected by the wind. They will fall off quicker, that’s all.

    Also, while talking about the trail at the ground level, I meant the trail excluding the dust cloud. I don’t think we see a dust cloud on Aleynikov photographs. The black plume has been discussed already and is either a train exhaust or an exhaust from a mine.

    As to the duel mode engine. From what I’ve heard, the two modes use exactly the same fuel, what varies is the grain size.

    • Daniel Been // June 30, 2016 at 8:36 am // Reply

      Hi Eugene, you’re probably right about the inertia. Another possibility is that the dark cloud moved to the *left* of the field of view because of the powerful rocket boost, leaving the starting point possibly way closer to the lighter column. The exact structure of the dark cloud remains unclear considering the quality of the image at this magnification and likely distorted color balance.

      You can see similar effects in this video, especially at the 2nd one (around 55s).

      Even the visual “break” of the light contrail is interesting here in the second clip. Because of this I don’t see any convincing reason supplied by anyone to dismiss at this stage the possibility that these plumes are indeed being SAM related. There’s just not enough data in my opinion.

  4. Michael Kobs // June 29, 2016 at 11:15 pm // Reply

    There is just one propellant in two different mixtures for faster and slower burning. The main exhaust is aluminum oxyd (Al2O3) – a white powder hanging as smoke in the air. There are no heavier or darker gasses but during launch some dirt and dust is blown away from the lauchspot. That’s the often brown colored sphere of “smoke” hanging at the bottom of the trail and pointing in the opposite direction.
    Wind gradients are a possibility like hurricans. In fact we see an alleged missile trail that is about 2 minutes old and still pretty straight up. Wind shear or gradients would have caused a visible kink that is missing for all elevations up to 1000 meters.
    A wind speed of > 11.5m/s is a #6 on the Beaufort Scale. You get a problem to handle an umbrella at that wind speed.

    • Thanks. Micha, I asked a couple of questions in the blog about your recent pdf. Can you please check it?

    • Daniel Been // June 30, 2016 at 8:18 am // Reply

      Hi Michael, do you have more detailed information about the ignition process itself? Any other chemicals involved as catalyst? The solid grains have always two components: propellant and oxidizer. It’s the oxidizer and/or shape which can vary, making dual-mode operation with a monopropellant possible.

      You are right about the lighter trail appearing to be straight up but that would not mean it has been two minutes like that. We simply don’t have that information right now. If the average surface wind (including any short term turbulence effects caused by the boundary layer effect) would indeed be around 4 m/s than with the *always present* increasing gradient starting above a few hundred meters, resulting in around 8 m/s, after two minutes it could never be really straight up right?

      • > It’s the oxidizer and/or shape which can vary, making dual-mode operation with a monopropellant possible.

        Actually the dual mode is implemented very simply. It is the same fuel for two modes, but the profile of the fuel ensures that for the first 4 seconds the burn area is much bigger. Check the picture, the fuel is marked #10.
        There does no seem to be any catalyst involved. The ignition happens at a corner (#5).

        • Hugh Eaven // June 30, 2016 at 7:39 pm // Reply

          Eugene, could you translate that picture or at least source it?

          The “profiling of the fuel” is to my knowledge done by differing grain shape, propellant type and/or oxidizer ratio. Since I’m not aware of a 100% identification of the missile type, I’d be very hesitant to even claim it needs to be single-propellant.

          We do have an igniter part for sure as you pointed out. But we have now in the mixture: fuel, oxidizer, binders, sometimes coatings on grain and so on. And what would go first?

          A nice page:!rocket-engines/c1md5

          Check out the hybrid schematic with the liquid oxidizer, where you can see why right after ignition you might have different gassing (the igniter is often also a “fuel”). Considering the pictures and movies of BUK launches, I’d say it’s not just a dirt cloud they’re showing. This is why it all would depend on very precise specifications of chemicals and internal lay-out involved. Do we have that somewhere?

          • I grabbed that picture from the textbook on Buk for students of military schools by Ieltsin. I interpret “profiling” the following way. The hollow fuel cylinder located near the exit has a thin wall, which fully burns away after the first 4 seconds. The thick walled fuel cylinder near the forward section of the engine continues burning for further 15 seconds (19.35 s is the total burn time).

            The section of the book on the engine is two page big, so does not cover all the details you are talking about. It does not mention what fuel (Al powder according to Micha?) nor oxidizer are used.

            Here it is ( Section 3 is about the engine, the picture is a few pages before.

          • By the way the widely circulated picture of speed-distance graph of the missile
            seems to be not entirely correct:
            The first mode lasts 4 not 3 seconds,
            the total burn time is 19 not 22 seconds,
            and, I think, the speed after the first stage is around 1000-1200 m/s not 600 (AA said that).

          • The image *could* be for a SA-6 system. Originally the image was posted on a simulator forum.
            However the difference with BUK M1-2 capable missiles is small.
            See the max. speed for missiles as advertised by Almaz Antey

            A guy using nickname AA posted this image at Metabunk. This guy has a lot of knowledge on BUK.

          • > However the difference with BUK M1-2 capable missiles is small.

            My thinking too

          • Hugh Eaven // June 30, 2016 at 8:50 pm //

            Admin, AA was to my knowing only posting schematics from 1950’s and 1960’s Soviet material. Cannot find my own references back to that dating. The warhead was for sure ancient and I suspect these rocket engine illustrations are from the initial 60’s models mostly. Student textbooks nearly always use very early models although many principles will of course be the same. But don’t take my word on that! But I do wonder how relevant these schematics are to make a technical point.

          • Hugh Eaven // June 30, 2016 at 9:03 pm //

            Eugene: “the speed after the first stage is around 1000-1200 m/s not 600 (AA said that).”

            Mach 3 then I suppose, around 1200 m/s (3×340.3) at sea level. At 10000m it will be closer to 900 m/s if we keep Mach still as specification. But if AA would be right it will be 600 m/s after boost and then slowly accelerate to 900 m/s, half the acceleration in double the time. That would make a bit more sense at first glance. Way less crazy burn for those first seconds.


        • Hugh Eaven // June 30, 2016 at 8:28 pm // Reply

          Eugene, I’m wondering now about those first 4-5 seconds, where the missile needs to accelerate to 1000 m/s (20G!) and the remaining 10-15 seconds then continues with more or less sustained flight. I’d be very interested in the fuel/oxidizer ratio of that first fuel section. Your diagram would suggest it’s only the burn rate which makes up the difference between acceleration and sustain. That seems hard to believe, 20G for 5 seconds just by some clever fuel profiling.

          • Yes, it’s 20 g or the first 4 seconds. I also think, by watching the videos, that the missile is programmed to keep the rail direction for the first 4 seconds. Then it realigns with the direction memorised by the inertial guidance (and sometime later finally switches to the guidance from the seeker).

            > That seems hard to believe, 20G for 5 seconds just by some clever fuel profiling.

            The book briefly mentions that the fuel has “tube-slit” structure, whatever this means. This may also have something to do with the two modes.

          • Hugh Eaven // July 2, 2016 at 8:20 pm //

            From an engineering perspective the suggested acceleration from 0-600 m/s would make more sense than trying to go to max missile speed in a few seconds. What is the purpose to get to Mach 3 after a few seconds of flight and then stop accelerating altogether? It’s more probable the acceleration goes down 50%-75% which would mean after 10-15 seconds the maximum (Mach 3 for 9М38 at 10km being around 900m/s) is reached upon impact, maximizing the forward fragment velocity.

            In terms of proportional navigation ongoing dynamically controlled acceleration is also assumed, like playing “chase” although the Buk appears to have a couple of approaches for all the possible circumstances. Point is that with only sustained flight and near zero acceleration, no real slowing would be possible and missiles would be too easy to shake off by jets. For military missiles this is different from mainstream rocketry, obviously.

  5. Micha Kobs again lie about MH17 disaster. 9M38 missile have propellant PD-17/18 with
    fuel Polyvinyl Butiral – 18%
    flammable plasticizer – 17%
    oxidizer Ammonium perchlorate – 59%
    stabilizer Aluminum – 6%.
    Main source of smoke trail in such propellant is low oxidation rate (which half from ideal – 3.28, not 7) and then incomplete fuel burning.

Leave a comment

Your email address will not be published.