Jeffrey LewisDPRK Road Mobile ICBM?

At the Shangri-La Dialogue in Singapore in June, outgoing-Secretary of Defense Robert Gates dropped a little bombshell:

With the continued development of long-range missiles and potentially a road-mobile intercontinental ballistic missile and their continued development of nuclear weapons, North Korea is in the process of becoming a direct threat to the United States.

I thought road-mobile ICBM?  That must be some serious jet-lag.  Now I realize that, two weeks later, Gates said it again and in a way that leaves no doubt he means it:

North Korea now constitutes a direct threat to the United States. The president told [China’s] President Hu that last year. They are developing a road-mobile ICBM. I never would have dreamed they would go to a road-mobile before testing a static ICBM. It’s a huge problem. As we’ve found out in a lot of places, finding mobile missiles is very tough.

Gates didn’t say anything about the precise type of missile, but my initial assumption is that it is a modified Musudan.  US participants in the December 2009 Joint Threat Assessment meeting with the Russians suggested that North Korea might “further develop the technology for an IRBM based on their new MRBM, in the same way the No Dong was a path to the Taepo Dong.”  (Warning: Wikileaks)  Given another leaked cable that suggests a 500 kg/4,000 km range/payload estimate for the Musudan, perhaps 5,500 km ICBM-class missile isn’t such a stretch if the North Koreans can work out the staging.

Gosh, I wish we knew where Jochen had went to.  Someone should model this sucker!


  1. MarkoB (History)

    You go for the B83 or some other earth penetrator and they go for the road mobile ICBM. Can’t complain, now can you?

    • Jeffrey (History)

      Ah, but you see, the US is not building a B83 Mod X, while North Korea is building a road-mobile ICBM. I am sure, though, you can find some US action to blame for the road-mobile ICBM, and vice-versa. That’s what makes this discussion so fruitless. Each side has a very long bill of particulars.

      The problem is that neither party seems to have the stomach for another round of negotiations right now. When that changes, I’ll be cheerleading about the benefits of striking a reasonable deal. Until then, let’s do the sled test.

    • Tom Nichols (History)

      Since the development of a road-mobile Nork ICBM would have had to long predate anything written about a modified B83, one thing clearly doesn’t have anything to do with the other. If anything, the collapse of the whole RNEP business should have assured NK (as if such a thing were possible) that the Americans would not try to dig things out of the ground with nukes.

      But if this report is true (and since we’ve never seen them test anything near that range, I wonder what’s going on), then it suggests two things.

      1. U.S. choices in nuclear deterrence have no effect on Nork choices of nuclear systems. Their motives are the independent variable here, not ours.

      2. If they go ahead with a road-mobile, this will obviate the B83 issue; the nuclear hawks will say that this makes missile defenses imperative (as it removes most of the preemption options) and that only quick-reacting strategic nukes, set for airbursts, will be able to to stop a first, and especially a second, strike. There won’t be time for mobile-hunting, and blanketing a suspected deployment area with nuclear force will be the obvious answer to any use, or threat of use, of a road-mobile ICBM.

      I always thought the Norks were crazy, but not stupid. Now I think they’re just plain stupid.

  2. Bruno (History)

    Modelling the Musudan should not be “too” difficult but obtaining critical info on key features for the missile(such as material, propellant (mass and type), dimensions..) would be the Devil. And that’s where you will determined whether that’s a medium to intermediate range or an ICBM.
    In any case, Gates really dropped a bomb there.

  3. Bruno (History)

    I am aware of most possibilities/assumptions on the Musudan’s features. What I am saying is that depending on the assumptions you make you could actually have a very wide modification of the range (notwithstanding any staging or other modification the Koreans could have made to the original design).
    I am not trying to be “Russian” on that, only to point out that a lot is possible but not everything is feasible.

    • Jeffrey (History)

      I see, yes, of course you are right that the assumptions will matter a great deal.

      On the other hand, we can probably recreate the extant model used by the US intelligence community, since we know their assumptions about fuel, performance and so on. That allows us to ask whether such assumptions seem likely or reasonable.

      I recall that a colleague of mine did this sort of modeling for the DF-31A and, as a result, had some concerns about the IC estimate of how energetic Chinese solid propellants might be. This seems, to me, to be a useful focusing of the debate to a question that we might plausibly answer.

  4. kme (History)

    It is interesting that in the space of a fortnight, the weasel-word “potentially” disappeared from SECDEF’s patter. I wonder if that is significant – perhaps indicating something about the IC’s official confidence in the assessment versus Gates’ personal opinion on the evidence he’s seen.

  5. John Schilling (History)

    Alas, I can’t read wikileaks from my work computers, so this will be a day later than it should have been. But, turns out my launch vehicle performance estimation tool does a pretty good job extrapolating down to IRBM/ICBM class suborbital trajectories, giving 2400 +/- 350 km for the stock R-27 aka SS-N-6 with 650 kg payload. That’s using stage mass and engine data from, and matches quite nicely with other published performance numbers. So let’s try it on a Musudan:

    The Musudan, from available drawings and photos, appears to be an R-27 stretched by 2.7 meters. Assume all of that stretch is extra fuel tanks, with a conservative 5% tank mass ratio at the margin. And, per wikileaks, upgrade the oxidizer from IRFNA to NTO (really, MON-3). I get:

    3600km range with 500kg warhead
    3300km range with 650kg warhead
    3100km range with 750kg warhead
    2750km range with 1000kg warhead

    Those are quite close to David Wright’s numbers, and they match exactly if I stick with the old IRFNA/UDMH propellant combination. 500 kg/4000 km is pretty close to the upper edge of the 95% confidence interval, so somebody may be giving Pyongyang the benefit of quite a few doubts there.

    Alternately, the extra length could be hiding a stubby second stage. Same propulsion technology but an extra 10s Isp for high-altitude nozzles, and the overall dry mass fraction will be 15%. We’ll also need an interstage, call it 1.0 meters and 5% of the second-stage wet mass. Conceivably the Norks could use the later Russian trick of immersing the second-stage engine in the first-stage fuel tank, but that’s tricky and it actually doesn’t buy them much in this case. So,

    5450km range with 500kg warhead
    4650km range with 650kg warhead
    4250km range with 750kg warhead
    3450km range with 1000kg warhead

    Not quite a nominal ICBM, even with the light warhead, but within the margin of error. Staging is tricky business, nobody gets it right the first time, but it is essential to ICBM-class performance in an operational system.

    • Bruno (History)

      Could you please elaborate on the structural mass you have choosen in your simulation. Again, I am not trying to be Byzantine about it but I am surprise that by only staging (so basically adding structures at least at launch) you could actually preserve the same Structural Coefficient without modifying the material (which would be tricky).
      I think it is safe in all circumstances to assume that the Koreans will not go below 0.10 (but if I get you right hence my question you put it at 0.15).

    • John Schilling (History)

      Sure. uses a ~10% dry mass fraction for the stock R-27. That’s structure, tanks, engines, plumbing, controls, basically everything that isn’t propellant or payload, and a reasonable figure for a missile of that type and vintage.

      For the case where North Korea hypothetically stretches the R-27, I have assumed that cylindrical tankage and structure alone has a dry mass fraction of 5%. Adding 950kg of propellant requires an additional 50kg of metal. The stretched R-27 comes in at a 9.1% dry mass fraction – better than the stock R-27, because the Norks are using the same engines, controls, etc on a larger vehicle. The performance model accounts for the reduced acceleration, and perhaps charitably assumes that the R-27 had enough control margin and so forth to accomodate the stretch. We’ll find out when they deign to test the thing.

      For the second-stage hypothesis, I assume that the first stage is identical to the R-27 except for the change to NTO oxidizer (which will require tweaking the tanks and plumbing). Same 10% dry mass fraction. The second stage is assumed to have a 15% dry mass fraction – 150 kg of metal per 850 kg of propellant. Reduced economies of scale, and an indigenous design without full benefit of Russian engineering expertise. I have also assumed that there is an interstage between the first and second stage, to better distribute the launch loads and handle the mechanical separation requirements. This is assumed to have a mass equal to 5% of the second-stage mass.

      For the specific assumed masses
      Stretched R-27 case:
      1st stage dry mass, 1585 kg
      1st stage propellant mass, 15800 kg
      total launch mass, 17385 kg w/o payload

      Two-stage case:
      1st stage dry mass, 1350 kg
      1st stage propellant mass, 11125 kg
      interstage mass, 220 kg
      2nd stage dry mass, 660 kg
      2nd stage propellant mass, 3740 kg
      total launch mass, 17095 kg w/o payload

      Note that propellant masses have changed slightly from the stock R-27 because of the different density and mixture ratio of the NTO oxidizer.

    • Allen Thomson (History)

      > 5450km range with 500kg warhead

      > Not quite a nominal ICBM, even with the light warhead, but within the margin of error.

      Yes, this is still looking like a regional missile with only a handful of plausible(*) targets that the Norks couldn’t reach with other existing rockets.

      (*) For plausible values of “plausible.”

  6. ArkadyRenko (History)

    Though the numbers may be marginal for an ICBM, it depends on the range from North Korea to Seattle / Hawaii, the effect of a road mobile MRBM will still be quite crucial.

    First, North Korea has to start somewhere. Building a road-mobile ICBM is probably quite different from messing around with a SCUD missile. This will give them a place to learn before probably radically redesigning the Musudan to give it true ICBM range.

    Second, the road Mobile Musudan gives them a chance to build a deterrent weapon to aim at Guam, for example, if not Okinawa. This means that North Korea can put US bases at risk, if it wants to enforce a “keep out” against the US. Imagine the DPRK saying: “This is purely an internal incident, don’t get involved. And if you do, we’ll nuke Guam.”

    Finally, its interesting that Gates is continuing his truth telling tour. I wonder what his motivation is for saying this; when its clear that this fact is getting very little press. Maybe he fears some radical change in the North / South Korea dynamic?

    • Mark Gubrud (History)

      If North Korea nuked Guam, the U.S. would nuke North Korea. Game over.

      I don’t mean to be flippant. I just can’t imagine what kind of “internal incident,” that the U.S. might otherwise intervene in, in connection with you think the threat to launch a (still hypothetical) nuclear-armed Musudan at Guam would be credible, or would deter the U.S. from intervention.

      Clearly, a North Korean threat to use its nuclear weapons is credible only in the event of a wholesale attack, aimed at regime change, by the U.S. or its allies. And for that, the threat against South Korea, Japan or U.S. bases therein is more effective. North Korean WMD also serve to deter unprovoked military aggression against the regime, but not on the basis of a threat that they would respond to a minor action by going nuclear, but just because nobody wants to find out.

    • ArkadyRenko (History)

      When I said ‘internal incident,’ what I meant was that North Korea would claim that some crises (serious but short of a full out war), extended shelling across the DMZ for example, was something which should be left solely to the Koreans to settle. A road mobile MRBM would allow North Korea some means of threatening the US / Japan. Here, the situation would be short of an all out war yet large enough that the US may want to intervene but doesn’t feel absolutely obligated to. Thus, North Korea’s nuclear weapon could reach its maximum deterrent effect.

      Next, I don’t think that a nuclear attack on Anderson AFB would actually spark a full fledged retaliation by the US. I base this off the idea that nuking a civilian target, Pyongyang, has some ‘difference’ from a military target, Anderson AFB, thus the US may not have the political will to escalate strikes to the next level. Also, remember that the US won’t be able to eliminate all of North Korea’s nuclear arsenal, even with atomic strikes; thus there remains the threat of an attack against civilian targets in South Korea and Japan.

      But, the Musudan road mobile weapon probably is in all likelihood a prelude to a full sized road mobile ICBM project. That is the challenge for the US deterrent posture in Asia.

    • Mark Gubrud (History)

      If North Korea actually used a nuclear weapon against a major U.S. military base, which American president would not use nuclear or whatever other weapons were needed to deal with the problem once and for all?

      If this could be done while minimizing civilian casualties in North Korea and the risk of attacks on civilian targets elsewhere, it would be. Otherwise, it would still be done.

    • ArkadyRenko (History)

      I’d disagree in two points:

      1) The threat of medium range nuclear strikes against military targets will deter the US from intervening. Its a new level of MAD; right now, North Korea only has MAD in the sense that it can threaten civilian targets in Japan and South Korea. With a road mobile MRBM, North Korea can deter US military targets and force the US to choose between a nuclear strike, which it won’t be able to adequately respond to (see below), or avoiding the situation. Remember, I said this only applies to medium sized crises, not all out war. Whether or not that’ll work falls back into deterrence theory…

      2) The problem for US counter strikes is that we can’t hit all the North Korean nuclear missiles (have fun killing every SCUD in North Korea…) without running the risk that North Korea will nuke / chem bomb Seoul or Tokyo. In that situation, threats of US retaliation ring hollow, further increasing the power of this new deterrence.

  7. Ben (History)

    It seems strange that NK is still developing/testing liquid fueled missiles, given Iran’s success with the Sejil program, you would expect them to have already developed a solid fueled version of the No-Dong and to be in the process of developing a road mobile solid fueled IRBM.

    • John Schilling (History)

      North Korea is not Iran. North Korea, as a result of missile-building decisions over the past few decades, has a good many engineers who understand liquid-fuel rocketry, and none with any experience building large solid-fuel rockets. Iran made different decisions, and has more engineers to begin with, and thus also has engineers who know how to build solid-fuel rockets.

      Thus Iran, and only Iran of these two, can build solid-fuel IRBMs. It isn’t a matter of solid fuel being more advanced, more difficult, or even necessarily better, it’s different. Even if Iran were to send a complete set of Seljil blueprints to Pyongyang, North Korean engineers wouldn’t entirely understand them – every blueprint includes thousands of implied pages of “everybody who is in this line of work already knows this stuff” detail that doesn’t get written down, and what every solid-fuel rocket scientist knows is not quite the same as what every liquid-fuel rocket scientist knows.

      For North Korea to build solid-fuel missiles, there would need to be an exchange program where several hundred North Korean engineers work alongside their Iranian colleagues for many years learning the tricks of the trade. Or, spend even longer trying to figure it out on their own. The North Koreans seem to have not done this yet, chosing instead to focus their limited resources in the areas where they already have significant expertise.

  8. CaptainCanuck (History)

    The thought of North Korea investing in the development of another advanced weapon system while its own children starve to death fills me with rage.

    – anemia
    – beriberi
    – pellagra
    – scurvy
    – diarrhea
    – skin rashes
    – edema
    – lethargy
    – irritability
    – fatigue
    – apathy
    – atrophy
    – disease
    – stunted growth
    – heart failure

    Children. Thousands upon thousands of them. Experiencing these things. Every single day.

    And their government expends its resources on artillery and submarines and missiles and nuclear bombs and F*****G ROAD MOBILE ICBMS.

    “Supreme Leader”, indeed.

    • Mark Gubrud (History)

      I’ll see your rage and raise you a thousandfold, Captain.

      Children all over the world, millions and millions and hundreds of millions of them, suffer the maladies on your list, and live and die in poverty, thanks, at least in part, to the priorities and policies of the United States and the global capitalist order, its military violence and private greed, the market fundamentalism imposed by the IMF, World Bank, and local tyrants armed by Uncle Sam, resource and labor extraction under exploitative terms, pollution and environmental destruction, and the wholesale destruction of societies in our wars. Turn your gaze to Afghanistan, Iraq, Gaza, and then look across Africa, South America, Southeast Asia, Indonesia and the Philippines. Look at the sadness of the new Russia, the pollution of the new China, and the hopelessness of deindustrialized, jobless America. Think of the world our own children will inherit.

      I make no apology for Kim Jong Il. No doubt the people of North Korea would be much better off if that regime collapsed and Korea was reunited. But it’s hard to see the evidence, after six decades of war and confrontation, that perpetuating that confrontation will hasten positive change of any kind.

      The problem with little exercises in rage like yours, apart from utter hypocrisy (think of what we could do for kids with the Pentagon’s budget), is that they provide support for the perpetuation of military confrontation, and ultimately make war more likely, which is no way to rescue children.

    • CaptainCanuck (History)


      When I read about NK developing a road-mobile ICBM, though, my thoughts were of the many children in North Korea, and of the insanity of that nation expending its resources on building advanced weapon systems. That was the topic of my comment.

      Is it your view that any expression of horror at the suffering inflicted on the North Korean people, is hypocritical unless it also expresses appropriate amounts of revulsion for similar suffering elsewhere in the world?

      Further, you instruct me to “direct my gaze” to other miseries elsewhere in the world. Was that because you believe I’m unfamiliar with them? If so, thank you, but please rest assured I am not. I have seen many of them first-hand, and they make me angry too.

      So to clarify, Mark, my expression of outrage at the behaviour of the DPRK government was in no way meant to…

      a) minimize other very real problems elsewhere in the world,
      b) give tacit support to continued military confrontation in the Korean peninsula, or
      c) made “war more likely” (!)

      No, I just commented on the insanity of a government developing a road mobile ICBM when so many of its children are starving to death.

      If you’d like to make a post about the US defence budget, IMF policies, the World Bank, pollution, or any other injustice, by all means go ahead. But it may be a little off-topic on a post about DPRK ICBM developments.

    • Mark Gubrud (History)

      Fair enough, Captain, but I wasn’t just pointing out “other miseries elsewhere in the world” but the fact that such “miseries” result in part from the policies and priorities of the U.S. and its allies, and that if one were to weigh somehow the total amount of misery attributable to our policies and priorities and compare it with that attributable to the North Korean regime, one would find the balance on our side by a very large factor.

      As for all this being off-topic, you raised it. This blog is trying to be part of a discussion that has some effect on the military and diplomatic policies of the U.S. and its allies (I expect there are few readers in North Korea). In that context, raging against the priorities of the North Korean regime, and blaming those for the unacceptable and unconscionable suffering of thousands of children, tends to feed the war machine.

  9. Chris (History)

    OK, I know this is kinda irrelevant, BUT…. ICBM is synonymous with LONG RANGE BALLISTIC MISSILE. Something we ALL can agree the Nutty Norks DON’T HAVE.

  10. Andrew Tubbiolo (History)

    Just to take a step back and look at what we know. Have we seen a successful test of any notional NORK ICBM? I’m running through my pea brain and the much more substantial memories of the google corporation, and I can’t find an instance of known NORK ICBM success. Can anyone say with confidence if their underground test was a dud, or simply a low yield test? My point being, Gates is describing an entirely new vehicle, with new modes of operation. Given the record, is there any reason we can expect the North Koreans to start a whole new design, and expect it to work any time soon?

    • John Schilling (History)

      W/re the North Korean nuclear tests, it has been reported that the Norks pre-announced to China that they were planning a 4 kiloton test on both occasions. That they got ~4 kT on the second test, suggests that this was a successful test of a low-yield nuclear warhead. As minimizing collateral damage in tactical nuclear warfighting is unlikely to be a major concern for Pyongyang, that means either economizing on fissile use or a lightweight warhead. Or some combination thereof.

      As far as “whole new design” is concern, the assumption most of us are making is that the Norks are modifying an existing design, not starting from scratch. The R-27 is a design they seem to have access to, intended for mobile deployment, and capable of minimal ICBM-class performance with a modest second stage, a lightweight warhead, and a bit of optimism.

      Previous attempts at staging have not gone well for North Korea, but it is a capability they will want eventually. That they are attempting it now with the Musudan is speculation, but well-founded speculation consistent with Gates’s remarks. The only other options I can see are, A: Gates is wrong or B: the Norks are trying something completely new.

    • Mark Gubrud (History)

      John, I don’t know your source for the reported pre-announcement, but I’m willing to bet it was a post-report, which leaves me to doubt its veracity.

      Other obvious possibilities to explain the low North Korean yields include that they either failed to achieve high compression or that their plutonium is heavily contaminated with 241 and higher isotopes, which might be due to pushing production rates.

      Your hypothesis assumes a relatively high level of competence on the part of the North Koreans, to achieve high compression hence low critical mass in a miniature warhead for their first-ever design. On the other hand, if one assumes that the North Korean program is being conducted under low-budget conditions and heavy political pressure, other explanations seem more natural.

  11. Mike Elleman (History)

    My two cents worth on the Musudan. Firstly, the ‘missile’ paraded by North Korea in October 2010 was very likely a ‘mock-up’ so extrapolations of its performance based on its geometry must contain caveats. Secondly, this missile has not been flight-tested; deploying it without the necessary development tests would be quite risky. A while there is good reason to believe the core R-27 was used as the second stage of the Unha-2 satellite launcher, this would not constitute a flight test of the military version of the Musudan. Because the missile has not been flown, estimates of the Musudan’s performance characteristics are purely speculative.

    Thirdly,despite the aforementioned caveats, it is possible to speculate about the Musudan and its performance, afterall this is what we do at ACW.

    Photographs of the Musudan reveal a missile that retains many of the visible features of the original R-27, except that it appears to have been lengthened by roughly 2 metres. The original R-27, according to several Russian sources, has a range of 2400km when carrying a 650kg warhead. These same sources indicate that the R-27 weighs 14200kg at lift-off, and carries 12,300kg of propellant, consisting of UDMH and NTO. The main engine produces about 23 tons of thrust, with the two steering engines generating about 3 tons. The engine is submerged into the propellant tank to reduce the missile’s overall length, necessary for submarine deployment.

    We do not know if North Korea received R-27 missiles or R-27 components (i.e. engine, guidance, etc without the airframe). If NK received whole missiles, and wanted to deploy them on road-mobile launchers, two significant modifications would be necessary. Engineers would have to incorporate structural reinforcements (the original R-27 was deployed in a very stable vertical launch tube), adding some 200-300kg of inert mass to the Musudan. Moreover, NTO oxidizer is very sensitive to temperature changes; NK would either have to deploy the missile in a launch canister to protect against the ambient environment, or switch to the more stable IRFNA.

    Range partials can be used to approximate the loss of range due to the two modifications. The added inert mass would rob the missile of 100-300km, or more, of range. The lower specific impulse (IRFNA vs. NTO) would further reduce the range by another 100-200km. These reductions could be countered by adding propellant to increase the engine burn-time. The stretched version (i.e. the additional 2 metres length) could carry enough additional propellant to off-set the losses described above, thus resulting in a Musudan with a maximum range of about 2400km when armed with a 650kg warhead. It is very unlikely that the Musudan featured in the 2010 parade has a range approaching 3000km; it is even less likely to fly more than 3000km.

    Finally, most have presupposed that the missile was stretched by 2 metres so that it could carry additional propellant. This may indeed be the case. However, we cannot dismiss the possibility that NK received only the engine, and not the airframe of the R-27. As such, NK engineers may not have had the ability to ‘submerge’ the engine into the propellant tank, as is done on the R-27. The added length might be the result of a longer engine bay. The fact is, we just do not yet know.

    Speculation about the Musudan is fun, and helpful as we try to understand what North Korea might be able to achieve in the future. However, when speculating, we must be careful to consider the engineering realities that North Korea’s specialists will face when attempting to operationalise the Musudan.

    Comments and criticisms are of course welcomed.

    • Greg (History)

      Many excellent points. On the submerged engine issue, it may possible that this is outside north Korean technical expertise… Particularly if they received r-27 technology in piecemeal fashion.

      But would it even be desirable to have a submerged engine on a mobile platform? I thought (perhaps mistakenly) that fueling a r-27 would essentially render the missile unfit for re-use. The propellants are sufficiently corrosive that not firing the missile just turns the engine into a paperweight. This doesn’t seem to jive with a weapon that’s supposed to be versatile. And its not like the north koreans can afford to be ruining their missile engines. Also, A fueled r-27 would also be pretty immobile.

      Please correct me if i am mistaken on these issues. I’m not a full time analyst. But given these constraints,,your point about a longer engine bay makes a lot of sense. It would solve fueling and mobility issues… And if the norks only got the engine technology, they’d be re-engineering the airframe anyway. Are these other potential advantages supportive of the view that the Musadan does NOT actually use it’s increased size to expand the range?

    • Bruno (History)

      Nice Mike,
      I think I am in perfect agreement with that. I did not check on John’s calculations but they give a good idea of what is technologically feasible, so let’s take them for a useful illustration.
      Now, if DPRK do have actually engineered a (roadmobile) 5000km+/500kg missile they will have to test it at one point or another since there is a lot of nagging issues to be verified before proclaming it operational : not least of them having a vehicle to carry the thing and launch it, ensure stage (and warhead) separation, etc.
      Anyway, I think we now have to wait for further info while keeping in mind that Gates’ announcement are worrying.

    • George William Herbert (History)

      I’m not so sure about this structural reinforcement. The liquid fueled R-27 and on were filled with propellant at the factory and shipped loaded from there up to the sub bases, via rail or road (not clear to me which).

      That’s not as extended as the handling associated with “road mobile” but it’s of statistical increase in spectrum duration rather than expected peak values etc. Essentially everyone overdesigns for handling, as having to junk a missile that got jostled a bit too much in transit is too expensive (though, strangely, aircraft are routinely retired if overstressed in flight…). Once you do that, within aluminum fatigue limits, …

  12. Distiller (History)

    If the road mobile IRBM/ICBM thing is true some serious words with China are needed. And coordination with Moscow in that matter. And also for the Air Force the question of a new strategic nuclear LO cruise missile arises, which is the only way of increasing the U.S. offensive potential without violating the treaty with the Russian Federation.

    NKorea is one country I wouldn’t bet on deterrence to work. Which raises the spectre of a first strike in certain situations …

    The other aspect is that NKorea has so far sold everything it developed. The prospect of road-mobile IRBM/ICBM proliferation is less than amusing.

  13. Doug Richardson (History)

    While I can sympathise with Jeffrey’s comment that “Someone should model this sucker!”, the problem with modelling the Musudan (or any other IRBM/ICBM-class weapon) is that the devil is in the details. It’s the classic case of ‘garbage in = garbage out’.

    As a contact (much more knowledgeable than I in such matters) has several times pointed out to me, a relatively small change in the assumptions made about the upper stage(s) can have a major effect on the range prediction, so unless you have reliable data on upper stage weights, propellant loadings, and Isp, little faith can be placed in the range predictions that emerge from modelling.

    My good friend Mercurius Cantabrigiensis believes that some good can result from modelling based on estimates, arguing that using by the most optimistic estimated data and the most pessimistic values we can establish upper and lower bounds for the likely performance of a missile. However, he admits that in some cases the spread of these values can be so wide as to be of little use.

    There is also the question of how accurate the results from modelling software are, even when high-quality data is used.

    Inevitably, the best of such software is classified, but I have been told that when the same data is run in several classified models, the results agree to within about 2 per cent.

    The only package offered commercially that I am aware of is subject to ITAR export restrictions, so is not available outside of the defence community.

    A couple of years ago Mercurius and I spent an entire holiday weekend running data on about eight obsolete ballistic missiles on all the unclassified range-prediction software we could find. These programs were of varying sophistication, and all seemed to ignore (or at best crudely allow for) some of the parameters that influence missile performance.

    We established that while most of the resulting range-prediction estimates were within 10% of the real-world value, and sometimes within 5%, the occasional instance might be 20% or even 30% from the real-world value.

    Earlier in this thread, John Schilling referred to “my launch vehicle performance estimation tool “, but gave no details of this. The only unclassified one I know is the ‘Rocket Ballistics 3D Simulator’ that was downloadable from the Bharat Rashak website, but this is not being maintained. Running this software in its high-speed mode requires a temporary licence key from the author. Mine has expired, and I have not been able to find new contact details for the author.

    I am not aware of any freely-available modelling software that takes into account or fully reproduces factors such as optimal flight-path shaping, changes in Isp from sea level to vacuum, or the amount of propellant remaining in tanks and propellant lines after engine cutoff.

    David Wright has apparently developed his own software, but once told me (if my memory is correct) that this is very user-unfriendly, and would not be useable by an outsider. He was thinking about doing a major rewrite of the code to take account of more factors (such as unburned residue propellant), but I don’t know if this was ever done.

    Theodore Postol uses some form of modelling software. I once emailed him to ask what it was, and how accurate he thought it was, but never received a reply.

    • Jeffrey (History)
    • John Schilling (History)

      If it’s details you want, my tool is described and implemented, respectively, at:

      The implementation is specific to space launch vehicles; among other things it has not been generally validated for ICBM trajectories. But a long-range ICBM is sufficiently similar to a launch vehicle that the extrapolation should give good results, and as previously noted seems to be dead-on for the R-27.

      It is an estimator, not a model. It uses an updated empirical relationship that Martin-Marietta developed in the 1960s for dealing with “factors such as optimal flight-path shaping, changes in Isp from sea level to vacuum”, among other things, by noting that whenever rocket designers go through the trouble of dealing with all those things in detail the results pretty much always fall in a narrow band defined by suborbital burn time. So if you’re interested in a +/10% estimate, you don’t need to run a full model every time – and you often don’t have the additional data you’d need for a full model anyhow.

      Which is not to say the full model wouldn’t also a useful thing to have, for cases where you have the full data and want the full precision. It would also be enough work that if I have to do it all myself I probably won’t be hosting it for free on my web site.

  14. Doug Richardson (History)

    I was aware of Forden’s software. It does not currently model the effects of altitude on Isp. Development of this program may have ceased. Although an improved version was apparently contemplated, to the best of my knowledge it never materialised.

    We can also regard as moribund the Rocket Ballistics 3D Simulator (RB-3D-S) that was downloadable from the Bharat Rashak website. Running this software in other than a slow real-time mode requires a temporary licence key from the author. Mine has expired, and I have not been able to find new contact details from the author.

    He is no longer developing or maintaining this software, but has not posted a version that eliminates the licensing system (the original rationale for which was that it would force users to regularly update to the latest version.). Although all rights to the software were passed to an Indian academic institution, my inquiry there as to its current status has not been answered.


    • Jeffrey (History)

      I have Geoff’s program. I have encouraged him to pursue it on an open-source basis so that a community might continue its development. Would others be interested in contributing to open-source ballistic missile modeling software?

    • John Schilling (History)

      I have some expertise in the area, and I’d certainly be interested. There is, of course, the question of whether an ballistic missile modeling tool might be seen as a ballistic-missile design tool, such that making it open-source might be construed as an ITAR violation in the United States.