Geoff FordenNow It Can Be Told: Inside BOB

(posted at 2:01 GMT 4 June 2010)

One of the great things about writing for the Wonk is that people tell you things, including what’s inside what has been called here the “Big Odd Box” in Burma. Last January, I was invited to join a group of experts in Oslo, Norway, to review a ton of electronic documents smuggled out of Burma to the Democratic Voice of Burma (DVB). (There is a great documentary about the DVB that was nominated for an Oscar in 2010. You can watch it on YouTube here.) Now that DVB has released its latest documentary, I can tell about my part and the information I learned about Burma’s nascent missile development program. Other experts can address any nuclear connections.

These documents contain a large number of images taken by elements of the Burmese military as they constructed the two BOBs and then installed an amazingly sophisticated numerically controlled machine shop. Such documentation is a normal part of any construction project today much like the photos taken of Syria’s reactor when it was being built. And like the Syrian photos, DVB’s sources probably didn’t take them but, instead, only later had access to them and made copies. They cover so much material—DVB’s source(s) simply grabbed whatever was available—that I expect I will have a number of future posts exploiting this information.

Internal Consistency

We spent a significant fraction of our time in Oslo trying to authenticate the information and judging its significance. Since very little is known about what’s going on inside Burma, most of this consisted of looking for internal consistency. This was fairly easy for the Big Odd Box(es), which aren’t really odd at all.

The image documentation show the Boxes at nearly all levels of construction; from clearing the forest and leveling the ground, to preparing the concrete pad and support beam holes, to stabilizing the surrounding banks with shotcrete, to finishing the interior, to installing the CNC machines. According to DVB’s source(s), both “Boxes” are essentially the same: loaded with sophisticated milling machine and other equipment for precision engineering. Some of these images show non-Asians (they actually look like Europeans to me, but I cannot say for sure) installing some of the sophisticated equipment.

The Burmese have filled this building with a wide range of numerically controlled milling machines, lathes, etc. Interestingly, they have laid out the machine shop by placing together those machines that are related. For instance, there is a hall with progressively larger milling machines, another for machines for cutting or welding, and another for precision 3-D measurements. The later, of course, could be used either for quality control or reverse engineering. I have not seen any evidence that the Burmese intend to reverse engineer missiles, which is probably a wise choice. However, what they are doing right now is not that much better.

The arrangement of equipment that I alluded to above makes sense for a general purpose machine shop, one that might get a wide variety of orders but always for one or two items. It might even be intended solely from prototyping, albeit some pretty massive prototypes, some weighing up to 20 tons! (When contacted by the producers of the DVB documentary, the companies exporting these sophisticated CNC machines claimed that both Boxes were set up as training centers for future machine operators and had nothing to do with missile or nuclear related production. Taking the big picture point of view, that, at best, just kicks the can down the road.) If, on the other hand, the shop was intended to produce thousands, or even hundreds, of copies of the same item—a centrifuge for instance—the layout would be, or should be, optimized for material flow with very different types of equipment positioned near each other. For instance, an electron beam welder might be positioned near a milling machine etc. So it seems unlikely that the shop is intended for producing centrifuges, which require thousands for any meaningful project. (And would not need the very large machines in any case.) It is, of course, conceivable that they might make missile parts since those are often done in onesies and twosies.

Evidence of a Desire to Make Missiles

According to the information gained by DVB, Burma is pursuing a least two different paths towards acquiring a missile production capability. One is a more or less indigenous path. The “less indigenous” comes from the fact that they have sent a number of Burmese military officers to Moscow for training in engineering related to missile design and production. The second in command of one of these “Boxes” received a degree in rocket engines. (He received a Master’s of Science in Engineering from the Bauman Moscow State Technical University in 2004. During his studies in Moscow, he specialized in Power Engineering for Rocket Engines, one of the specialties Bauman MSTU is known for. Here is a copy of his Diploma.) Here he is holding a test item manufactured at his “Box.” He identifies it as the impeller for a large kerosene/liquid oxygen engine intended for static testing.

From a purely evidentiary point of view, it is very significant that a different group than the Box designed the impeller. What took place at the Box was a conversion from the CAD files to machine instructions to make the impeller using the machine the gentleman is standing in front of. This two-group activity implies a significantly greater level of interest by the Burmese authorities than if the impeller had been designed in the same group as it was manufactured. In fact, it implies at least three organizational entities were involved: the design group, the manufacturing group at the Box, and a coordinating authority that approved the impeller being sent over to the Box for fabrication.

The engine that this impeller design—the item actually fabricated is simply a “proof of concept” item that lacks some significant features for an actual working impeller—is destined for is reported, in addition to burning liquid oxygen/kerosene, to have a combustion pressure of 25 mega-Pascals. That is about four times the combustion pressure of a SCUD engine. (My own calculations, based on assuming scaling from a SCUD-type engine, show that the impeller’s diameter is consistent with a large rocket engine, perhaps a Nodong. I did not try to estimate anything assuming it was for a liquid oxygen engine.) Such a large pressure—not to mention using a cryogenic propellant!—seems highly undesirable for the first engine produced by a country that has a serious plan for developing missiles or rockets on its own. A more realistic first attempt at designing an indigenous engine might have used a more conventional propellant combination and preferable a smaller engine with a lower combustion chamber pressure. There are simply too many hurdles for the novice to overcome on their first engine design without throwing in handling liquid oxygen. In fact, this example perfectly illustrates the risks involved in independent innovation: the personnel involved are simply too inexperienced to know when they are getting in trouble.

One is left with the impression that the higher-ups are interested in utilizing their foreign trained scientists and engineers for missile production but do not have a master plan for development. In stead, they are giving a green light to their workers to exercise their new-found skills. Perhaps they will get serious later but as of now we can definitely say that this indigenous path has a much, much greater risk of failure than the other path they seem to be pursuing.

Burma also appears to be following another acquisition path: purchasing missile production lines and know-how from the North Koreans. Here most of the evidence comes from a single source; a summary of a trip report describing the activities and accomplishments of a number of high-ranking Burmese officials made to North Korea. There is, however, considerable supporting evidence that the officials did actually make the trip. There are images of meetings of North Korean and Burmese officials and some photos that could be of sites mentioned in the trip report. The summary of the trip report is, however, the only evidence of the one of the results of the meeting: a Memorandum of Understanding where Burma gets assurances from North Korea that it will be able to purchase complete production lines for missiles with ranges up to 3500 km. A two stage U’nha-2 or a Simorgh come to mind. There is, unfortunately, no strategic reason given for why Burma would want such missiles.

There is, on the other hand, plenty of evidence in the DVB cache of information that Burma fears an attack by the United States and Diego Garcia—a major US air base—is almost exactly 3500 km away. So we can at least imagine a deterrent reason though that threat would be minimal without a nuclear warhead. That lack of a stated reason, and the lack of clear and independent confirmation of the trip report, makes me want to hold off on accepting that Burma is committed to purchasing a production line for a large missile from North Korea. However, I think we can be fairly confident that such an acquisition path would have a much, much higher chance of success than the indigenous path.

Signs of a Sea Change in the Proliferation Environment?

According to DVB’s sources, North Korea had nothing to do with setting up the two machine shops inside the Boxes. In fact, the Boxes seem to have been set up as general purpose machine shops and probably do not violate either the MTCR or even political sanctions imposed by Europe against the Junta (Europe’s sanctions against the Burmese Junta are considerably looser than those of the US and these exports were probably legal. Now that there is evidence of the production of missile related components those companies will probably want to rethink their future exports.) However, this whole episode is an indication of how proliferation might be changing.

Consider how India got started on its road to preeminence in solid propellant missile technology: it licensed the technology from France, received detailed written know-how on production (and training of technicians in France), and received a list of production equipment, which India purchased elsewhere. France was obviously capable of producing the needed equipment and chose—presumably for political reasons since the US was at the time trying to pressure other countries not to assist India’s rocket/missile program—not to sell them directly. North Korea is also at least claiming the ability to produce advanced production machines and probably did sell a certain level of technology to Iran for missile production. However, North Korea must wonder if it will always be able to ship large pieces of equipment out of its country or even if its clients would settle for DPRK’s finest. Instead, the spread of precision engineering worldwide— A. Q. Khan’s use of Malaysia’s SCOPE engineering is the clearest example of this—has opened up the possibility of proliferation networks more as consulting engineering firms rather than one-stop-shopping centers. After all, without the testimony of DVB’s sources, it would be impossible to tell the difference between the Boxes set up by Westerners with the equipment list coming from a North Korean consultant for WMD/delivery production and the Boxes set up by Westerners as general purpose machining.

A Special Thanks

DVB’s sources are brave people who have decided to smuggle out a variety of information about the Junta’s activities so that the world might know. Missile development is not causing as much harm to the Burmese people as many of the other activities of the Junta. Nevertheless, it is part of a military program that shows a remarkable disregard for the Burmese people. I have waited to publish this posting until being assured that any source who might be implicated by the information has been safely evacuated from Burma.


  1. Naing Ko Ko (History)

    Thanks for those who bring it to the international audience, esp to Ko Sai and the DVB.
    Bravo DVB and good on you.

  2. John Schilling (History)

    Based on the experience of amateur and small-business rocket engine and launch vehicle development in (mostly) the United States, I would suggest that liquid oxygen is actually a very good choice for an entry-level rocketry program. Yes, it is cryogenic and a severe fire hazard. All rockets need either an oxidizer or a monopropellant, and all oxidizers/monopropellants are severe fire hazards.

    All liquid oxidizers/monopropellants which are not LOX are at least one of: highly toxic, highly corrosive, or borderline explosive. Usually more than one of these will apply. LOX, being “merely” a cryogenic fire hazard, is arguably easier to work with than any of the others, certainly not greatly harder, and has been used successfully by amateurs from Goddard to the present day. Meanwhile, approximately no Western amateurs or small businesses are using IRFNA, nitrogen tetroxide, or the like.

    It is not the oxidizer of choice for modern missiles, but only because it cannot be readily stored in the field. This suggests the Burmese are A: planning to build very clumsy missiles, B: planning to build space launch vehicles but not missiles, C: only using LOX as a learning experience, or D: not being entirely open and honest.

  3. simorgh (History)

    Thanks a lot for those awesome descriptions, very interesting indeed!

    When talking about Burma there is one extremely important point to mention. The regime is both paranoid and irrational (astrologists have considerable influence on Than Shwe) So the line of argumentation “XY does not make sense” or “there is no possible use for xy” should be taken with a grain of salt. If Myanmar purchases a production line for 3500km rockets the reason might be because Than Shwe’s astrologist told him he should be able to strike targets in 3333km distance or because Burma’s generals simply fancy the missile. Seriously that’s what happens over there.

  4. Allen Thomson (History)
  5. Pedro

    I don’t know why this impeller has to be for a missile turbo pump, looks like a quite ordinary impeller. A nice item to test new machines. There isn’t even much skill needed for that, just certain software and some knowledge in turbo machinery (there are even user-friendly impeller generators in modern software).

  6. John F. Opie (History)

    The layout described could be used to justify the stated intent, to train operators. But I heartily agree that this is irrelevant for reverse engineering. Further, the training of operators would imply an industrialization policy that Burma simply doesn’t have in reality: further, given the extreme rural location (no urban centers near for a cohort of existing semi-skilled operators) for security, especially operational security so that no one knows what exactly is going on, and I think that you (and the DVB) have nailed this nicely.

    It is also disturbing: why would you invest significant assets to train a cohort of operators? There are a whole lot of things that Burma needs before they need highly trained machine tool operators, and given the fact that even in very high wage countries, personnel costs for operating such machinery are very low per unit produced (the whole point of introducing this kind of machinery into production processes), and the whole pretext tends to fall apart.

    If anything, this makes the Burmese junta look as if it is following the North Korean path towards the acquisition of military tools that have no feasible use except to initiate an arms race with neighbors that are vastly better equipped, economically, than Burma is: that way lies madness, but it is the same madness that drives the North Koreans.


    Oh, and re-organizing the layout, while not a trivial task, is not that difficult: you use airmats (which use compressed air to lift heavy machinery a fraction of an inch for short movements without having to disassemble said machinery) slid under the machinery to shift it around. Given the type of construction (self-supporting light metal building), internal walls are not load-bearing and can be shifted at whim.

  7. Jochen Schischka (History)

    I must say that the exact geometry of that impeller-mock-up does look somewhat peculiar to me.
    Normally, the blades are higher on the entry/suction side than on the exit/pressure side (this is true for radial compressors as well as turbines, although the pressure ratio is converse in this case, and is related to the respective entry/exit cross-sections aka the belonging volume-flow), but that particular article seems to handle this issue exactly vice versa.

    So what are the Burmese trying to accomplish with that design? A turbopump that doesn’t deliver as much pressure as would be possible (and can’t run with full speed because of cavitation-issues)? Or will this discrepancy be corrected in a futher processing step? Or is perhaps my impression of the impeller-geometry, based on that photo, wrong?

    (My guess is that, most likely, this is indeed only a ‘proof-of-concept’ article without any actual application other than testing the manufacturing equipment.)

  8. Martin Dirksen (History)

    See the BBC on Burma and this story at:
    Yours M.Dirksen-Fischer

  9. Allen Thomson (History)

    simorgh expresses something that’s been on my mind concerning the BOB discussions for a while:

    Maybe we’re seeing something that doesn’t really make a lot of sense when looked at from a rational point of view. I.e., the Burman/Myanmarian junta doesn’t seem to be composed of highly sophisticated folks, and they may have been listening to astrologers or con artists.

    Still, the intent seems to be ill, and a certain amount of continued attention seems in order.

    BTW, the new information doesn’t indicate DPRK involvement, which leaves in question a) US officials’ expressions of worry concerning such and b) the reported (and regrettably as-yet-unleaked) 47-page UN document that apparently does contain something on DPRK-Myanmar dealings.

  10. Pedro

    @Jochen Schischka

    To me it looks really ordinary radial pump for any purpose. Its high on the suction side, going close to axial for reduced cavitation (must be a very crude, maybe unfinished sample).

    I’m quite sure good CFD software has an impeller generator which gives you the same impeller, maybe with a little shorter blades on the pressure side.

    Generally I can only say that there is no reason why this alone should be connected to a missile program in any way.
    An impeller is a very good testing object for a modern CNC based workshop; I guess it would be also my first idea to test the new capabilities of my workshop. Complex geometry, many costumers in all industrial fields and a nice looking product to show to possible investors and costumers, everything purely civil.

  11. George William Herbert (History)

    Jochen –

    The impeller looks normal to me, it’s not uncommon for central-entrance, radial-exit impeller pumps to look like that. Plenty of similar examples in the literature. Often you blend a semi-axial or screw impeller into the front of it, and there’s some sign the taller 4 impeller vanes come up and start to do that, though the picture shows the piece not yet fully machined so the final shape is somewhat unclear.

  12. George William Herbert (History)

    It appears that at least much of this body of work predates the apparent reactor site discovery; Geoff, have you had a chance to review that?

  13. John Field (History)

    I think Jochen is correct, and besides this impeller also looks like it was machined with a 3-axis machine, not a 5-axis one. It’s not a work of art; any machine shop in the US could make it.

  14. Azr@el (History)

    Pintle injectors, ablative nozzles, Pistonless Pumps a la Flometrics, GPU’s under a grand, GPS, solid state sensors…I’m sure there is a safety margin inherent in reverse engineering the wheel absent in its reinvention…but as technomic levels continue to broaden out from the core states…said margin will seem smaller and smaller to project managers.

  15. Jochen Schischka (History)


    “…going close to axial for reduced cavitation”

    Only that it’s exactly the other way around! Check this out (or wikipedia, or, preferably, any better textbook on turbomachinery):

    The low pressure side is always axial, be that compressors/pumps or turbines. With this in mind, ‘going large to radial’ simply doesn’t make a lot of sense.

    But you’re certainly right with everything else. The whole item looks to me like a finger exercise in CAD/CAM/rapid prototyping – and without additional indicators (like e.g. the diploma of the guy holding that part?), it’s impossible to say if that impeller-mock-up belongs to a missile program of some kind or anything else, so we should be careful to not overrate that particular item.

  16. Jochen Schischka (History)

    George William Herbert:

    “The impeller looks normal to me, it’s not uncommon for central-entrance, radial-exit impeller pumps to look like that.”

    Look closer. Try to establish the axial and radial entry/exit cross sections based on that photo. Then compare this to “similar examples in the literature”. Dedicate a little bit more effort to the small, on first sight easily overlooked details. I’m sure then you will get my point.

    That said, you’re absolutely right about the axial inducer. Typical solution in case of turbopumps.

  17. Allen Thomson (History)
  18. Pedro

    @Jochen Schischka

    Ok I think I understand what you mean; the height of the blades on the pressure side is somewhat high.
    This is bad for the pressure it delivers but good for the volume flow. I think is still inside the range of what the costumer could want. But I don’t know why the cavitation risk should be higher, it looks ok to me.

  19. kme

    Pedro: I believe you may have missed the part of the story where the gentleman holding the part in question, the second-in-command of the facility that fabricated it, is the source for the identification of it as a rocket engine component.

  20. Allen Thomson (History)

    A question for those who know about such things: Did manufacturing any of the parts shown, with the possible exception of the impeller, require the use of the high-precision tools installed in Factories #1 and #2 (aka BOB1 and BOB2)?

  21. Allen Thomson (History)

    I’m watching the Al Jazeera special that includes Maj. Sai’s material and wonder how the “burning chamber” and “automatic autoclave sterilizer” fit into the nuclear narrative. Those (and maybe the glove box) sound more biological to me — or am I misunderstanding something?

  22. Bob Kelley (History)

    Thank you Allen. That is exactly what we are hoping to find from posting the raw material on Burma. We wondered the same thing but thought it was outside our main theme, but you spotted it!

  23. Allen Thomson (History)

    Well, another. The machine tools came from Germany and were inspected by German industrial and government (hopefully including BND) people after installation because they thought there was something a little odd going on. But, having been there, they are said to have bought the Burmese explanation that these two very large buildings were just for training machine operators.

    Now, one of the things that’s been keeping wonkers excited is the location/ siting of BOB1 and BOB2. They are in remote areas, dug into the side of modest mountains, and shielded from the surrounding terrain by either an artificial berm (BOB1) or existing topography (BOB2). So the Germans thought this was perfectly unexceptional for industrial training facilities?

  24. Bob Kelley (History)

    I’ll give the Germans some credit for insisting upon end-user inspections and allowing this material to come out. It is not the machines that are imortant because the stuff Burma is building is so crude. It is the access to the site and the pictures both with and without the Germans that catches your eye.

  25. Nick Black (History)

    wow, post of the year! thanks for a truly exceptional read!

  26. Greg Matteson (History)

    Fascinating and informative; however my historical training and interest in technological history informs me that the Schmucker-Shiller absolute declaration against Reverse Engineering is overstated. There are other examples, but in answer to the demand for a single example of successful reverse engineering, I cite the Tupolev TU4, and unambiguously successful R/E clone of the B-29. Given the many shortcomings of the B-29 (for example, prone to engine fires) the TU4 may have had significant improvements over the model cloned. While reverse engineering has been overrated by boosters of spy agencies, my example is proof positive it can be done.

    An addition peeve for me was the Schmucker-Shilling rehash of SpaceX’s early failures. They’re now up to 3 unambiguously successful launches for 6 tries, and 1 success out of 1 try for the Falcon 9, which is at least an order of magnitude more complex than the Falcon 1.

  27. John Schilling (History)

    As for reverse-engineering, the Soviet K-13/AA-2 air-to-air missile is an unambiguous case of successful reverse engineering of a missile (early Sidewinder), and the R-1/SS-1 is a ballistic missile reverse-engineered with limited and unwilling technical support from the original manufacturers.

    There seems, at this point, little reason to bother with “pure” reverse-engineering, when there are apparently veteran missile engineering teams willing to do rocket science for cash and without asking questions. But if someone has an ideological reason for absolutely eschewing foreign assistance, it can be done. As can pure independent development.

  28. Geoff Forden (History)

    The recent discussions on reverse engineering in this thread were probably meant to be posted under a different original posting. Yet I cannot let some common misconceptions about reverse engineering pass without some sort of attempt to set the record straight. In particular, it is irrelevant to point out that either Russia, China, or even the United States have had successful reverse engineering projects. In fact, as I pointed out (though perhaps not explicitly enough) in my post on the “How of Proliferation”, successful reverse engineering requires a great deal of preexisting knowledge, especially knowledge on manufacturing techniques. Countries, like Iraq, that do not have this preexisting knowledge are know to have failed many, many times in a variety fields. (I count Iraq’s attempt to produce VX as reverse engineering since they got the formula from abroad.)

  29. Jochen Schischka

    I can only agree with Geoff, all examples given here for ‘successful reveres engineering’ are definitely flawed.

    First of all, the technical and economic capabilities of the Soviet Union at its height can hardly be compared with those of North Korea today.

    Next, let’s take a closer look at those examples:

    The Tu-4/Bull hardly is an identical clone of the B-29 – 1t less max bomb load, 100km less max range, ~10kn less max speed, all this despite 800hp more installed engine power (BTW, the Russians did not clone the R3350-engine, but used another one already available to them – in contrast, the North Koreans allegedly also cloned the Scud’s engine to perfection…) and higher empty weight. Other technical details like slightly different dimensions, the missing crawling tunnel or no integral wing tanks make both aircraft easily distinguishable, although not neccessarily on cursory examination from a distance (although i can guarantee you that i always can tell the difference – look less cursory!). All this completely ignores the fact that in 1944, Tupolev already had copious first-hand experience, not only concerning general design of all-metal aircraft, but also in the design of 4-engined bombers (the U.S. back then tended to misinterpret the USSR, and some years earlier also Germany and Japan, as technologically underdeveloped – a grave mistake that lead to e.g. propeller-aircraft facing jet-fighters, or a kill ratio of ~15:1 for american Sherman- vs. german Panther-tanks).

    BTW, ‘Greg Matteson’: Please re-read Schmucker/Schiller’s paper – he explicitly links that statement to ballistic missiles, and the Tu-4 clearly does not fit into that category.

    The K-13/AA-2/Atoll respectively the Sidewinder is no large ballistic missile, but rather a small, solid-fueled unguided artillery rocket supplementary fitted with fins, air-vanes, a fragmentation warhead and an IR-seeker – fairly simple to copy, especially if you have access to a fully developed industrial base (like the Soviets had) and additional intelligence (which allowed access to the manufacturing process and original technical documentation), something the KGB excelled at. The most complicated part is the IR-seeker and respective electronics, and that was rather comprehensible in comparison to more recent, highly integrated digital electronics, too. Don’t know about any performance degradation, though, since i unfortunately don’t have access to the AA-2’s exact specifications.

    The case of the R-1/SS-1a/Scunner is particularly interesting: Here, the Soviets not only had access to the product to be copied (aka pieces of Aggregat-4/V2-missiles) plus technical experts on that system (see operation ‘Ossoaviachim’, the soviet ‘Paperclip’ – with the slight difference that the latter one was voluntary in nature) – they had access to the complete manufacturing line plus support infrastructure in and around Nordhausen! In fact, as soon as the USSR had captured the Mittelwerk, they re-started production with the original personnel and, as soon as they had learned enough, relocated that production line to Russia! Nonetheless, since the R-1 was forced to use russian materials and thus differs in certain, easily overlooked technical details from the original (again, although this is a quite close copy produced with the original manufacturing equipment, i’m rather confident that i can tell the difference in most cases), and thus offered less performance than the original. And the Russians extensively test-fired those missiles.

    The north korean ‘Scud-clones’, on the other hand, offer exactly the same nominal performance as the original – and this even without any serious testing!

    And, ‘John Schilling’, if you would have thoroughly read Schmucker/Schiller’s paper, then you’d know that in case of ‘reverse engineering’, functionality has to be the driving factor, not cosmetic appearance. Thus, it should be expected that real RevEng-missiles (like the chinese DF-3/CSS-2, which clearly was a RevEng-try at the soviet R-14/SS-5/Skean – and the result of that effort offers definitely less performance than the original) differ significantly in optical appearance from their respective originals – again, not the case with ‘wonderland’ North Korea!

    The DPRK allegedly even was able to recreate not only perfect clones of the missiles themselves, but also all neccessary technical documentation, the production line, all neccessary materials, specialized fuel refineries, the MAZ-543-TEL, and, last but not least, an independent production line for MAZ-543-clones based on their measurements of egyptian Scud-missiles alone, on a GNP less than that of Cameroon…this should trigger some thinking processes!

  30. Pedro

    @Jochen Schischka

    I don’t think that SCUD technology was impossible to RevEng for a country like North Korea back in the 80’s. It is a communist country that has/had a mechanical industry.

    I already had a discussion about this topic with you and I think you are overestimating the technological hurdles and underestimating the North Koreans or better said, what they are ready to do for such a strategic goal (own and independent missile industry).

    Furthermore you don’t take into account that testing could have been done in Iran, in a real war environment, as well as petro dollars and lather even technological support.

    Lets first realize what technology we are talking about.
    If it would be a high-performance turbofan jet engine I would agree with you that RevEng is likely to lead nowhere. Material and production tools needed to make the critical items are even today very high and well protected technology.

    The quality requirements are high and it has to work for a long time.

    I’m pretty sure that no part in a SCUD would be exposed to such high thermal and dynamic mechanical stress as a modern first stage turbine blade. Its operation time is also much shorter/quality requirements lower.
    Therefore there is no need for such exotic materials which a modern turbine blade uses.

    E.g A Nickel based cast alloy should be more than enough for all parts exposed to high thermal stress (expensive but not out of reach for a country like North Korea). Design elements for the film cooling can be simply copied.

    As we are also talking about the impeller for the turbo pump.
    Its a good example of what new machines are capable of, good luck trying to produce a batch of such impeller geometries with 1950’s vintage machines and see how easy it is today by CAD/CAM/CNC.
    Such an item saves be weight/space and gives me more pressure. But they don’t even try to improve, they have simply copied the design.

    However I admit that I have not a single good book on rockets.

  31. Jochen Schischka (History)


    Your last comment once more reveals how little you actually know about ‘rocket science’ (and this leads you to vastly underestimate the difficulty of reverse engineering missiles).
    Believe me, “a single good book on rockets” would be nowhere near enough. Add good books on fluidum dynamics, thermodynamics, turbomachinery, (subsonic/supersonic/hypersonic) aerodynamics, general flight dynamics, external ballistics, material sciences, lightweight construction, aeroelastics, production methods, fuel-related chemistry and probably a dozen more disciplines i forgot about at the moment to your list. And it’s by far not enough to only own these books. It’s even insufficient to have read all these books. It’s absolutely vital to understand the content of all of these books in detail and in context. Additional practical first-hand experience in any of these fields wouldn’t go amiss either (this is often vital to really understand what you’ve theoretically learned about).

    That said, let’s take a look at some of your statements:

    “I’m pretty sure that no part in a SCUD would be exposed to such high thermal and dynamic mechanical stress as a modern first stage turbine blade.”

    Once again, exactly the other way around!
    Typical turbine inlet temperatures in modern jet engines (doesn’t matter if tubojet, low- or high-bypass turbofan) don’t exceed ~1700°K – typical combustion chamber temperatures in rocket engines can reach temperatures of up to ~3500°K!
    Likewise, modern jet engines characteristically have maximum chamber pressures of up to ~30bar – even the ‘ancient’ Scud-engine reaches a pc of 69bar. In other, more ambitious pump-fed rocket engine designs, this can be as high as ~300bar!

    If you don’t believe me – do some research for yourself. It may be right that the life-time-requirements are much lower (otherwise, rocket technology simply wouldn’t be realizable at all!) – but in return, the requirements are that much higher per life-cycle plus, additionally, much higher considering lightweight construction and thrust/weight ratio.
    Consider e.g. the Scud-B engine: this ‘primitive’ engineering marvel generates over 13 tons (130.5 kN) of thrust at sea level at a dry weight of only 120 kilograms (for comparison: EJ200 (Eurofighter): 90kN:990kg, AL-31 (Su-27): 123kN:1570kg, F119 (F-22): 156kN:1770kg)! And more high-tech rocket engines easily eclipse this…

    “Lets first realize what technology we are talking about.”

    Indeed, let’s first realize what kind of technology we’re talking about!
    Especially rocket engines, as i just demonstrated, are a lot more demanding than comparably simple turbofans (don’t get me wrong, those are complicated enough!) – so if you don’t believe RevEnging a turbofan is possible (and this is indeed something the Chinese, the ‘masters of RevEng’, have difficulties with!), then your level of disbelief should logically be an order of magnitude higher in respect to missile technology, especially in respect to a country without former experience on that particular sector (or adequate industry or personnel or budget or…)!

    “you don’t take into account that testing could have been done in Iran, in a real war environment”

    That was actual wartime use – you’re suggesting something comparable to letting the customer do the car’s crash-testing himself…after buying that untested car for a lot of money. Or that airlines should do all airworthiness-testing of a newly designed airliner during regular passenger-transport. Not an option, period.

    Besides, those Scud-missiles used during GW1 by the Iranians worked perfectly and with nominal performance of the original soviet R-17 from the start – so this was obviously no untested, tinkered-with product (but definitely not tested in North Korea, either!)!

  32. Pedro

    I didn’t say that the North Koreans understood every element of rocket science when they first copied the Scud. It’s simply copying each element to the necessary quality level and start producing them in numbers. Of course basic knowledge on rockets is also necessary for that.

    Now after 30 years their engineers might have read enough of the things you mentioned to try to at least upgrade some missile elements.
    However, I’m just talking about copying a machine, not designing one.

    Well now to some of the points you mentioned.

    Yes you are right about the combustion chamber temperatures. However you missed my point; designing and producing a turbine blade which is exposed to 1500K is seen as more complicated by me because of the geometry, fine dimensions and dynamic stress it has to endure.
    A combustion chamber on the other hand is just a box exposed to vibrations, but more or less static (compared to supercritical turbine stages). It’s an item build to take those high pressures you mentioned, thus with rather massive walls while the geometry is almost simply enough to cast the main element in one run.
    Its massive walls bear enough space for cooling by liquid fluids (forced convection) and liquid film cooling.
    You can transport the heat energy amount which 3500K combustion chamber temperature produced if your coolant volume flow is enough.

    So excuse me but pure numbers of higher pressure and temperatures do not automatically mean to me that the technology is harder to master. A low bypass turbofan is extremely complex in geometry, fine in sizes and designed to work for hundreds of hours.
    But again you are slowly slipping into the design discussion; I’m just talking about the production means necessary to COPY the critical items of a Scud. So let’s talk about the technology we face here and let’s think about the capability of a nation in the class of North Korea in the 80’s.

    As for wartime testing of copied Scuds; there was no simple seller/costumer relationship between Iran and North Korea. Both nations had a common strategic goal on which they cooperated to reach it. I should also mention about western report of testing of a new unknown long range Scud inside Iran, during the early to mid 90’s period. I’m sure it was the Nodong and I already know about the flight test numbers comparison of the Nodong vs. other nations BM projects.
    I simply think you are vastly underestimating North Korea and its Industrial capabilities in the 80’s. I have my problems with communists but one thing some of them were good at is the education system, how many North Korean engineers learned their discipline in East Germany? One thing some here might agree on is that dictators do much to get new weapons, therefore its possible that all those NK scientists and engineers from East German universities were forced to go into the nations newly established missile industry.

  33. Jochen Schischka (History)


    “However, I’m just talking about copying a machine, not designing one.”

    And this is where you’re making a fundamental mistake. You assume that it’s less demanding to ‘just’ copy a highly complex technical item. Wrong! Additional to a full understanding of every little aspect of the to-be-copied item, you’ll need highly specialized analysts with extremely powerful analyzing techniques and -tools! So ‘copying’ is demonstrably even more demanding than just designing yourself!

    Let me again suggest that you make a little home ‘RevEng’ experiment: Since this is so easy, try to make some 100% copies of something as simple as a lowly 50 year old spark plug with the equipment found in your kitchen. Then screw this RevEng spark plug into the engine of an appropriate car. If that car still performs like with the original plugs (or even if you’d get the engine up and running), i’d be highly surprised.

    “A combustion chamber on the other hand is just a box exposed to vibrations, but more or less static (compared to supercritical turbine stages). It’s an item build to take those high pressures you mentioned, thus with rather massive walls while the geometry is almost simply enough to cast the main element in one run.”

    You’ve never seen a cutaway model of a real rocket thrust chamber in a museum, have you?
    “massive walls”??? In comparison, turbofan engines have rather massive walls! Even hollow turbine blades with cooling holes are massive in comparison to typical rocket engine thrust chamber walls!
    Why the **** do you think that rocket engines are over ten times lighter in comparison to air-breathing jet engines of comparable thrust, as i exposed in my last comment???
    “just a box”??? D’oh! Please, please, please with a lot of sugar on top, visit that museum!

    BTW, i hope you realize that pump-fed rocket-engines have lots of turbine-blades, too…

    “So excuse me but pure numbers of higher pressure and temperatures do not automatically mean to me that the technology is harder to master.”

    This statement in my eyes only once more reveals something about your level of knowledge considering technology in general.
    Double pressure at double temperature at less than one tenth the overall weight – and you don’t expect that to be more demanding?
    Can’t follow you here. Not at all.

    “A low bypass turbofan is extremely complex in geometry, fine in sizes”

    Exactly the same can rightfully be said even about ‘primitive’ rocket engines like the Scud’s Isayev 9D21. Just take a close look at cutaway-models of that thing in a museum!

    “and designed to work for hundreds of hours.”

    Well, this is basically why that type of engine is so much heavier than rocket engines. On the other hand, if gas turbines would be subject to such extreme conditions as rocket engines, i doubt that those a) would last anywhere near that long (BTW, make that thousands of hours, unless you refer to russian or chinese ones), b) would be anywhere near that ‘lightweight’ as they are today (still over ten times heavier than rocket engines of comparable thrust – didn’t you read that somewhere before?) and c) would be realizable at all.

    “I should also mention about western report of testing of a new unknown long range Scud inside Iran, during the early to mid 90’s period. I’m sure it was the Nodong…”

    Wrong again! That was the Shahab-2/SS-1d/Scud-C (as you can see by the existence of a DoD designator/NATO reporting name, that missile type already existed in the Soviet Union, which wartime-used that type in Afghanistan, and was no new design, not to mention not unknown).

    The first known Shahab-3/Nodong-A-test in Iran was on 22. July 1998.

    “I simply think you are vastly underestimating North Korea and its Industrial capabilities in the 80’s.”

    Au contraire. All evidence points at you vastly overestimating north korean capabilities!

    Look for example at the failure of Pratt & Whitney Rocketdyne’s license production (aka with a high level of support by the original’s manufacturer) of the russian rocket engine RD-180! You can say a lot about Rocketdyne, but not that they don’t have the documented capability to manufacture staged combustion rocket engines and a lot of experience in that field – and still, they fail at license production of russian rocket technology!
    Or what about the german-produced F-104G, which was in particular known as ‘the widow-maker’ because of its excessive accident rate – even in comparison to the american original?
    What about the teething problems the BMW X5 production in Spartanburg had?
    What about South Korea, which has considerable problems with its partially russian-manufactured KSLV/Naro-1 space launcher despite an unquestionably much broader industrial base than the north, an about 30 times higher GNP and massive russian help?

    And you expect a country which shows as ‘black hole’ on night-shots from space, whose population constantly faces starvation, which has a GDP less than Ethiopia and which is not known as an exporter of any type of consumer goods, be that high- or low-tech, to succeed blindfold where even highly industrialized, hundred times richer, fully-developed countries with vast experience on that particular field utterly fail?

    To be honest, i find only one description that fits your type of preconception in favor of the DPRK: EXTREMELY UNREALISTIC.

    “I have my problems with communists but one thing some of them were good at is the education system…”

    I have lots of personal experience particularly with East Germans – and believe me, they absolutely aren’t in any respect better educated than West Germans. Certainly better educated than the average Ethiopian or Uruguayan, but not better educated than any other western European. The only difference i’ve ever noticed is that they were better indoctrinated in their younger days. That doesn’t help you at all with technical things, though…

    Besides, i don’t see how studying in a country without any active missile industry (East Germany) in any way would help anybody to become a consummate ‘rocket scientist’, especially one comparable to the James-Bond-movie character Dr. No (works in absolute isolation and total secrecy on his remote island while performing incredible feats all superpowers find difficult, if not impossible to reproduce)…

  34. Pedro

    Claiming that higher pressures and temperatures are proportional to the necessary technology level is a big mistake.
    Nobody can change the laws of physics. If my metal industry can provide me with the best metal they have, my turbofan combustion chamber can be designed thinner than my Scud combustion chamber. This is a simple law and I know and you know that, that’s the easy thing about pressure.
    The motor block of my Diesel engine which has to endure twice the pressure than my petrol engine, MUST be heavier around the cylinders if I use the same steel.
    Now according to you a diesel engine block is something like twice as hard to master than the one of a petrol engine… and this in the age of helping tools like FEM.
    According to you the 1950’s metal industry capability of the Soviet Union is still more advanced than what North Koreans could get their hand on today.
    Let me take this further. Let’s assume that the North Koreans have got all the sizes and material properties of the combustion chamber, what if they now not just have better metals in their industry, but make also use of improved cast techniques, one of their engineers specialized on casting learned from the East Germans 1989? This means that some of the bottlenecks the Soviets encounter in the 50’s might be solved today.

    “So ‘copying’ is demonstrably even more demanding than just designing yourself!”

    I disagree; engineering knowledge is needed and with the right tools and methods copies of many mechanical items are much easier, especially if no one is ready to tech you about rocket science.
    “In comparison, turbofan engines have rather massive walls!”

    We are talking about combustion chambers walls right? Why should a gas turbine designer make a more massive combustion chamber if the pressures and temperatures are close to twice lower compared to a rocket engine combustion chamber? Might be because of lifetime and material factors, but there is no magical wall that divides this kind of technology from rocket technology.

    “Even hollow turbine blades with cooling holes are massive in comparison to typical rocket engine thrust chamber walls!
    Why the **** do you think that rocket engines are over ten times lighter in comparison to air-breathing jet engines of comparable thrust, as i exposed in my last comment???
    “just a box”??? D’oh! Please, please, please with a lot of sugar on top, visit that museum!
    BTW, I hope you realize that pump-fed rocket-engines have lots of turbine-blades, too…”

    First they are completely different as you know, but its surely not because turbofan designers are less capable engineers or have no access to the super materials, knowledge and tooling of the Soviet 1950’s rocket industry… it’s because of the different boundaries and goals.
    However the last part of the comment is interesting.

    If you have any data on the ingress temperature of the Scud turbines and if they are higher than even 1700K I will take my hat off, despite its very short lifetime.
    “Double pressure at double temperature at less than one tenth the overall weight – and you don’t expect that to be more demanding?”

    Excuse me but this is like comparing apples with oranges… at least take the complete compressor one of such a comparison and replace it with a small pump, taking out the energy wasted for the compressor and so on… As you know completely different machines.
    I’m sure no one thinks that anything of the stuff we are talking about is easy and everyone must realize that a country like NK would take the best it has to offer for such a project, not to mention financial and limited technological support by countries like Iran, simply a strategic project for a country.
    The license production of one of the world’s best rocket engines designed by a country with probably the most intense liquid engine experience on earth, to be licensed by a COMERCIAL company, is simply something completely different. We also know about Russians, especially those of that time and their support for any license production, just ask Indians, Chinese and Iranians about their experience with Russian expert knowledge with their license production deals.
    I recommend you to make less such comparisons; one can easily get a wrong impression.

  35. Jochen Schischka (History)


    I’m rapidly growing weary of unsuccessfully trying to teach you things you should have learned before presuming to have a mind of yourself in this regard, especially in such an unconvinceable manner.

    Don’t you notice that you are contradicting yourself? You argue “Diesel engines are heavier than petrol engines because of higher temperatures and pressures” (quite correctly, although that difference is almost neglectable in comparison to jet vs. rocket engines) – and in the same breath ignore the fact that it’s the much hotter and a lot higher pressurized rocket engines that are over ten times lighter, not the other way around! Shouldn’t some penny drop in the face of this almost unbelievable correlation?

    Take a good look at this (especially p.3):

    It may be right that this refers to LH2-pumps – but astonishingly enough, the 9D21’s turbopump operating parameters fall in a rather similar general envelope (e.g. nominal operating speed of 22815 R.P.M.). Also keep in mind that the oxidiser in this case is not LOX, but a mixture of NTO and nitric acid, which makes matters in no way less complicated, only less chilly.

    Do you still expect rocket technology (in this context, IT DOES NOT MATTER IF THIS IS 50 YEARS OLD OR RUSSIAN!!!) to be in any way less demanding than turbofans (for equitable comparability, this should be 50 years old russian ones, too – but you know what? THIS DOES NOT REALLY MATTER, SINCE IT’S APPLES AND ORANGES ANYWAY!)?

    Blatant lack of qualification is no excuse – a trivial five-minute internet survey could have disabused you on this matter.

    But instead, you prefer to stubbornly repeat the same baseless, unchecked, easily debunked cut and dried opinions over and over and over again!

    I’ll abandon this discussion at this point since i consider any continuation useless. Obviously, you’re intransigently preassigned to strictly ignore what i’m writing at all costs against any logic.

    Doesn’t matter whatever you’ll write next. I’ll let the readers ultimately decide who brought forward cogent arguments and persuasive examples and who didn’t.

    Besides, this frustratingly fruitless dispute doesn’t belong into this thread anyway, as Geoff pointed out some time ago.

  36. Pedro

    Abandon the discussion, it’s sad that you have the impression that I contradict myself; it proves that you have not understood what I trying to say. If a correlation appears so unbelievable to you, that your discussion partner must be retarded not to get it, you better check twice whether or not he has a good reason for that. That’s because one thing is sure, retards wouldn’t have a discussion like this…
    I can also just mention your ignorant repeating argument that a rocket engine has a 10 times better t/w ratio than a Turbofan… this alone means nothing, the matter is much more complex, the two machines are COMPLETLY different, just look at how both processes work thermodynamically, in a P/V diagram… But again, I’m sure you know it, but I don’t understand why you keep claiming this over and again to prove the technological complexity of rocket engines over jet engines.

    Thanks for the NASA pdf, I see my predictions confirmed, especially on page 3.

    Turbine inlet temperature rocket engine: 1000-1600F
    Turbofan 2400F (modern ones 2800F+).

    An operating speed of 22815RPM vs. 15000 from the chart is also not impressive, or better said is no proof for the complexity. Here an Iranian mini-jet-engine with a similar speed (RevEng from a French one).

    I admit, in this discussion I made no internet survey because we are talking about very a basic point.
    I don’t know how many people follow such discussions but well ok, everyone can make his own judgement.