Jochen SchischkaHello world!

Hello out there! Since Geoffrey Forden stopped blogging, ArmsControlWonk has been a little bit light in the ‘rocket science’ department – so Jeffrey asked me to fill that gap (frequent readers of this blog’s comment section may already have stumbled over my name here and there). Well, i’ll do what i can to round ACW out a little bit and thus make it even more enjoyable than it already is!

For my debut, i’ve decided to share some of my thoughts on the north korean ‘Mirim’-device (which is also known as ‘Musudan’ or ‘Nodong-B’). First of all, let’s understand that the ‘missiles’ shown in October 2010 almost certainly are non-functional mock-ups, as is common practice in case of military parades. Not only would it be a bad idea to put the top-ranking politicians  attending that parade potentially at risk by driving by live ammunition, what is more, this would also expose a highly valuable asset (the missile itself) to unneccessary mechanical stress and danger of accident. A blank is more than good enough for this purpose, and, let’s be honest, the overwhelming majority of observers wouldn’t notice the difference anyway (think along the lines of the classical stage prop in movies).

Additionally, no flight tests of this missile have been reported up to date, so we’ll have to assume that that system hasn’t even reached initial operative capability – more or less a mere paper project, at least up to now.

Next, i couldn’t help but notice that there is no provision for any form of sealed canister for that missile on the TEL (which itself is rather overdimensioned for that payload). To me, that is a tell-tale sign of this missile likely not using NTO as an oxidiser – and thus, being a bad candidate for an R-27/SS-N-6/Serb engine block aka a stretched R-27. Why? Well, because of the chemical/physical properties of dinitrogen tetroxide and typical military requirements.

NATO ground equipment usually is certified to a temperature range of -30°C up to +50°C for good reasons, the warsaw pact armies even insisted in temperature-resilience down to -40°C (siberian winter nights seem to be somewhat colder). And if this is inevitable for military applications in moderate climate zones like middle europe, i see no reason why this should be any different for the north korean army. Dinitrogen tetroxide, on the other hand, freezes at -11 to -10°C, which will render an unprotected missile inoperable, this should be self-evident. No big deal if that missile is in a thermally conditioned environment, like in a silo, or inside of a submarine, or a heated canister, but out in the open that means loss of reliable around-the-clock operational capability, particularly during winter season. I’d say that that would be rather inacceptable from a military point of view, particularly if all earlier systems (Scud, NoDong etc.) don’t share this unfavorable characteristic.

Another benefit of a canister would be that that would mimic the R-27’s launching mode (hot-launched from a flooded submarine tube), yet, the ‘Mirim’-device instead closely follows the Scud/NoDong-line with a launching stool with integrated turntable and thrust deflector, which clearly points at a Scud-like guidance system (and general launch preperations), too, not an R-27-like, with the consequence of less accuracy and slower reaction time due to more time-consuming preparations. The ending of the cable duct at a different place than on the R-27 just adds to the impression that there is no original R-27-guidance system involved.

Not only that, the little two-chamber steering engine of the R-27 may be too weak to sufficiently deal with the requirements of a larger, heavier missile, so that may have to be replaced not only because of a change of propellants, too, probably also with a less-efficient solution available to the North Koreans like the ubiquitous jet vanes. The lower acceleration due to ~3-4t additional weight at liftoff wouldn’t be relevant, on the other hand.

Last but not least, the R-27 was a rather maxed-out lightweight design – so simply adding length and weight will likely overstrain an existing missile body without supplementary structural reinforcements, and this will inevitably add inert dry weight and thus diminish the achievable maximum range.

So, the big question is:

Is the ‘Mirim’ really some sort of simply stretched ex-russian super-R-27/SS-N-6/Serb with the same engine and propellant combination (as apparently widely accepted) with all the quite inacceptable drawbacks, or will the North Koreans be forced to replace the propellants, the engine and the guidance/steering system with e.g. that of the NoDong and strengthen the missile body, measures which will spoil the overall performance down to approximately that of the already available NoDong-2 (~1500km range)? If the latter is the case – why bother at all with a larger, heavier and more expensive missile without better performance?

I for my part am beginning to ask myself if the DPRK’s latest show of force was perhaps only a cunning ruse along the lines of Field Marshal Grigory Potyomkin’s famous villages.


  1. Allen Thomson (History)

    If I might humbly suggest a theme or two, I’d certainly like to see some analysis of the capabilities of the SM-3 Block IIA as an ABM. How does the Block IIA deployed in Europe compare to the now-defunct 2-stage GBI? And, of course, how does it do as an ASAT?

    Ditto the SM-3 Block IIB, as more details on what the revised version will be like become available.

    • Jochen (History)

      Unfortunately, i simply don’t have enough reliable data on those systems for a credible analysis. Plus, i haven’t spent too much time poring over them, either, since my primary interest at the moment is the opposite side: offensive systems like the NoDong-2 or the ‘Musudan’-device, or the iranian ‘Khalij Fars’, which apparently is some sort of Fateh-110A-variant. And i also expect a new iranian satellite launch attempt soon.

      But i appreciate the suggestion, maybe i’ll add something in this direction in the future. Especially the 2008-ASAT-shot and how this compares to what the Chinese did may be a worthwhile addition.

    • yousaf (History)

      the block II will be a potent ASAT as well as a more potent ABM — in theory, of course. Countermeasures will still plague its use as an ABM.

      The burnout velocity of the block I is about 3.3 km/s and the block II could be as high as 5.5km/sec

  2. John Schilling (History)

    It should be noted that in rocketry, “Nitrogen Tetroxide” or “NTO” is a shorthand that essentially never means 100% dinitrogen tetroxide. The actual propellant is a mixture of N2O4 and N2O, with the latter being (among other things) a potent antifreeze for N2O4.

    The precise terminology is “Mixed Oxides of Nitrogen”, or MON-(X) where X is the weight percentage of NO2. MON-3 is the norm when people don’t care about freezing point, but field-deployed missiles have typically used MON-10 to MON-30, with freezing points down to -50°C or so. And often still referred to by the family name “NTO”.

    There’s no particular reason to believe that the North Koreans are using NTO-based propellants, but the absence of heated or insulated cannisters is not evidence either way.

    • Jochen (History)

      You’re addressing an important point – we don’t know the exact composition of the russian ‘AT’ aka the R-27’s (and R-29’s and UR-100N’s and R-36M’s) oxidiser component, nor that of the accompaning ‘НДМГ’ aka UDMH. Maybe some reader knows more?
      In case of MON (particularly MON-3), i only know the variant with NO, not N2O (but of course you can also add other oxides, like N2O5). Even ‘pure’ NTO typically consists of a mixture of N2O4 and NO2 due to decomposition/recomposition plus some traces of (mostly hygroscopically accumulated) water/nitric acid.
      You’re certainly right, the absence of a canister is no evidence per se (who can say that he knows for certain what’s going on in the heads of some North Koreans?) – but i consider it a rather strong indicator that the ‘Mirim’-device differs significantly from the R-27, in more than only length. And i personally consider NTO (or MON) as inferior to IRFNA (which itself is in essence a mixture of nitric acid and NTO) in case of military application, despite the higher corrosiveness of the latter.
      A very interesting point is: can you name an example of an actually field-deployed liquid-fueled military missile system without any form of controlled environment (silos, canisters, submarines) using MON (or NTO) as main oxidiser (i consider the comparably small amounts used in post-boost-systems as irrelevant in this context)?

  3. Pedro (History)

    It’s unlikely that the North Koreans have kept the R27 engine, more likely would be a uprated Nodong engine maybe even with R27 veniers for additional thrust(I know that a ordinary Nodong turbopump could not supply the veniers).
    The Iranians claim to have developed a 37t likely UDMH variant of the Nodong engine, compared to the 22t R27 engine this would make up for the larger size and range of the missile.

    Furthermore the increased length is imo due to the conventional Scud style tank layout, avoiding the more complex space efficient R27 tank layout which they have never tested.

    As always I disagree on the operational capabilities of the NK missile. Like the Iranians they should have made enough static testing and simulation work to develop an almost certainly working design (plus their SLV experience).

    • Jochen (History)

      Well, i agree that a NoDong-engine is a more likely choice than the R-27’s closed-cycle high-pressure engine.
      Considering the tank layout, well, that would be similar to that of the Scud-C, which does make use of a common bulkhead – so that would not per se be unknown to the North Koreans. Also, the Ghadr-1, multiply tested in Iran, seems to employ an R-27-like tank layout, too. Of course, we don’t know to what extent both nations cooperate at the moment on that sector. And unfortunately, the graphic material from the DPRK’s October-parade isn’t non-ambiguous enough to clarify that point beyond any doubt.
      But i must add that i am highly sceptical in respect to the iranian claims (or, probably, what the media made out of official statements) about 37t thrust, or UDMH, in the lower stage engine.
      Please keep in mind, changing the propellants means changing everything else on that engine, too – and the rocket engine the Iranians presented on 10. February as ‘Safir engine’ is still exactly the same old NoDong engine as ever (except for the rust…)! In contrast, i’d expect such a radically different, new engine to likewise look radically different. At the very least, i’d expect this to differ in certain visible details, which apparently is not the case.
      I don’t know what exactly happened, but to me, the whole story sounds like somebody with limited insight in the subject (a journalist?) picked up several isolated bits of information from a longer (somniferous?) lecture (like “the vacuum thrust of all engines together equals ~37t” and “the upper stage uses UDMH as fuel” etc.) and later on terribly mixed those little bits up. As well, this may also be a deliberate act of disinformation, can’t say for sure. Only that 37t thrust and UDMH in the Safir’s lower stage makes no sense to me.

  4. Pedro (History)

    I don’t know how sure we can be that the NK and Iranian Scud variants as well as the Ghadr use advanced R27/Scud-C style tanks.

    As for the improved Nodong engine (maybe in fact the Ghadr engine). Iranian claims for the Shahab-3 engine are known, 32t. Now they claim to have improved that on the Safir-B SLV which is claimed to have twice the payload of the Safir-A together with the 37t claim. Therefore there should be no translation problem here.
    They have not claimed the use of UDMH but claimed to have changed the fuel. This would correspond most likely to some UDMH combination.
    If they manage to orbit a 50kg satellite at 400km the 37t claim should become more than just a claim.

    A agree that there would be massive changes on the engine necessary however I disagree that these changes would be necessarily clearly visible on the outside. But well, it’s likely that they simply put out the show-engine of the Shahab-3 on that occasion.

    • Jochen (History)

      ‘Claims’ are always problematic. Not only don’t we know how close to reality those claims are, in most of the cases we also don’t know to what exactly those claims refer. “32t” – is that sea level or vacuum thrust? Is that nominal performance or the upper limit at minimum safety? Or perhaps a (unrealistic, or possibly even failed) future enhancement goal? Or simply disinformation? Or a misunderstanding by the reporter?
      So far, i haven’t seen any substantive photographic evidence of a sea-level thrust other than ~27-28 tons in case of neither the Shahab-3 nor the Ghadr-1 nor the Safir IRILV (which, unfortunately, is problematic because of the night launches), so i hope you understand my scepticism.
      On the other hand, i also don’t understand why the Iranians generally would need ~9 tons more liftoff thrust for a missile which, even with 20kg more payload, would still weigh about 22 tons. From an engineering point of view, 27-28 tons and an unmodified NoDong-engine would be sufficient and even consistent with the already observed performance.
      According to the NSSDC, the Omid (2) satellite at a weight of ~30kg reached an apogee of 364km – round that up a little bit and there you already are with the 400km and 50kg figure. My guess is that perhaps the Safir, even without any changes, may even offer this little bit of reserve, but not much more.
      We’ll see if the ‘Safir Rasad’ will show a higher net liftoff acceleration of ~0.7g (opposed to the ~0.25g of the ‘Safir Omid’ and ‘Safir Omid (2)’), but i’d suggest that we should postpone any further discussion on this subject to a future post on the occasion of the next launch attempt.
      BTW, in this context, please let me preemptively warn you and everybody else to not get fooled by a potentially slower recording/higher playback rate (a little disinformation trick that has already been used in published material from time to time by various sources in the past – if done correctly, this is very difficult to spot and even harder to debunk)!

  5. Hairs (History)

    Hello Jochen – I just wanted to say thank you for agreeing to provide rocketry coverage on ACW.

    In the past I always read both Geoff’s posts and your comments with great interest, and I’m looking forward now to seeing your thoughts on the latest in the proliferation of rocket technology.

    • Jochen (History)

      You’re welcome – i hope i’ll succeed in providing a little food for thought from time to time…

  6. George William Herbert (History)

    It’s not always true that propellant changes require significant changes in the engine justifying a new designation. The LR-87 on the Titan went from LOX/Kerosene (Titan I) to MON/Aerozene-50 (Titan II and beyond). A number of more recent aerospace startups have run the same engine on kerosene and isopropyl alcohol and other fuels; one of them was converted with intermediate efforts to liquid methane (I can’t quote the precise changes on that one).

    In particular, MON and IRFNA (or for that matter, IWFNA) are mostly a materials compatibility issue. The acids are denser, slightly.

    Changes that would matter most would be in the behavior of the cooling propellant, for cooled chamber motors (most, but not all – ablatives were out there until recently on Delta’s upper stage, are still on Kestrel, etc). Heat capacity going down or boiling behavior changing a lot would be potentially catastrophic, obviously, though there are tricks to handle that to some degree without complete redesign.

    • Jochen (History)

      You mention an interesting example.
      If i take a closer look at the LR-87-3 (Titan-I, LOX/RP-1) and the LR-87-5 (Titan-II, NTO/Aerozine-50), then i find it difficult to not see all the obvious differences – exactly confirming my argument. Compare for example the turbopump exhaust lines (long vs. short), or the exact installed location of the turbopumps (higher up vs. lower down), or some piping details (e.g. slightly slant vs. level lines to the chamber heads), or the nozzle extensions, etc., etc., etc. (BTW, irritatingly, there seem to exist some not-too-expertly improvised museum exhibits which apparently got cobbled together from LR-87-5 turbopump assemblies and LR-87-3 thrust chambers, inconclusively marked ‘LR-87’…).
      Oh, and please don’t overrate the experiments of some ‘alternative space’ activists in this regard – at least it would be nice to include a mention of the results of those experiments, which typically aren’t that favorable (yeah, i know, that’s usually omitted by the guys who want to sell their stuff)…
      Nonetheless, exchanging propellants normally requires at a very minimum a different turbopump and gas generator and a new injector plate (usually different optimum mixture rate, different densities, different viscosities etc. etc. etc.), even in case of ablatively cooled nozzles (and we’re talking about reactive cooling in this context, which makes matters even more complicated, as you’ve noted absolutely correctly yourself). Mark that i don’t say that this is impossible – only that this would with considerable certainty produce visible differences compared to the original (or, alternatively, results in vast performance losses and unforseen problems within other unforeseen problems).

  7. Amir (History)

    I am glad that the rocketry section is back. I was one of the fans of the rocketry articles.

    On another note, you mentioned a 22 tons weight for Safir SLV. Is this weight your estimation or it is from a different source? All sources I remember mentioned 26 tons weight for Safir. 32t thrust would be consistent with 26t weight and the photographic evidence. What do you think?

    • Jochen (History)

      First of all, welcome to this thread.
      Back to business: The ~22t-, maybe even 23t-figure is my own estimate, based on photo-measurements (which tell me that at an l/d of ~17-17.5 and an assumed diameter of 1.25m, the official figure of 22m length is approximately correct, maybe even lightly rounded up). This is what i get with an average density of ~0.85kg/dm³ (since the payload section will be rather lightweight, plus aluminum instead of steel for the tanks plus smaller fins plus an additional mostly empty interstage section, i expect Safir to have similar overall density as Scud-B despite common-bulkheads aka no intertank sections on both stages) and a volume of approximately 26-27m³ (simple mathematics). 26t would be too dense (at an average of 0.96-1.00kg/dm³!) for a liquid-propelled missile in my eyes (even highly compacted submarine systems which e.g. mount the main engines inside of the tanks typically don’t exceed 0.91-0.94kg/dm³ – less dense than the surrounding sea water).
      Additionally, the longer i’m trying to reconcile 32 tons sea level thrust with the given Safir/NoDong-engine geometry, the less consistent this looks to me (not to speak of 37 tons!). I’m beginning to suspect if this figure perhaps referred to the combined thrusts of lower and upper stage engines together (~27-28t s.l. + ~4t vac.). See my comments in answer to ‘Pedro’ – ‘claims’ are always problematic and thus should be handled with a not too small grain of salt. It’s always better to retrace the official figures based on photographic or other evidence (a classical example: if i’d blindly believe most t.v.-‘documentaries’, then the Columbia broke apart at a height of 60 kilometers – a look into the CAIB Final Report tells me that these were actually 60000 feet, not meters – just as i had suspected from the start).
      Let’s wait and see what ‘Safir Rasad’ will actually look like in the end.

    • George William Herbert (History)

      Jochen, the Columbia breakup was at 200,000 ft / Mach 19 (60 km) not 60,000 ft.

      CAIB Diag. 2.6.1, pp 40-41 (position 1 second prior to loss of signal was 200,861 ft altitude, 12,384 mph)
      CAIB Ch. 3.7, pp 73 and elsewhere “200,000 ft and mach 19”

    • Jochen (History)

      Hey, you’re right, thanks for the correction!
      I somehow had mixed that up with the height at foam-impact; Let’s assess this as a valuable lession in another vital discipline in addition to healthy skepticism: fact-checking.

      Although, this makes understanding the Columbia break-up rather less trivial, since the ram air pressure at that height (air density: 0.0003kg/m³) and speed (5933m/sec) would roughly correspond to only 334km/h aka 208mph aka 180kts, and even the ubiquitous Cessna typically won’t come apart at comparable speeds, either; Yet, i’ve omitted stagnation temperature, or that this is hyper-, not subsonic (i hope you understand that i’d like to avoid the associated five-fold-integral exercise in this context – after all, the original statement was only meant as an illustrative subordinate clause)…

    • Markus Schiller (History)

      With superheated air beyond 5,000 °F / 3,000 °C entering the wing, even a Cessna might be in trouble…

      See CAIB Report Volume I August 2003, p.69 and p.71

  8. Amir (History)

    Thanks Jochen,

    “Let’s wait and see what ‘Safir Rasad’ will actually look like in the end.”

    BTW, the interesting one should be Safir-e-Fajr (Safir-1B) as Fajr satellite is the heavy one. Rasad is about the same mass as Omid satellite, hence, Safir-1A will be used for its launch.