Science & Global Security, 21:210–236, 2013

Copyright 
C Taylor & Francis Group, LLC
ISSN: 0892-9882 print / 1547-7800 online
DOI: 10.1080/08929882.2013.835592

A Revised Assessment of the

North Korean KN-08 ICBM
John Schilling
University of Southern California, Los Angeles, California, USA
The mobile ICBM displayed in Pyongyang last April is analyzed in light of North
Korea’s known and reasonably suspected capabilities in the field, particularly those
demonstrated in the recent and successful Unha-3 launch. Several possible configu-
rations for the “KN-08” missile are given, and performance estimates are presented.
While the displayed missiles are clearly mock-ups, they are consistent with an ongoing
development program for a missile with limited intercontinental capability using only
existing North Korean technology. The lack of flight testing strongly suggests that op-
erational deployment is still months or years in the future, though an initial test could
come at any time. Even with a successful test program, any such North Korean ICBM
would likely be unreliable, limited in mobility and performance, and available only in
small numbers.

INTRODUCTION
On 19 January 2013, the New York Times reported that North Korea had begun
initial deployment of a long-range mobile missile referred to as the “KN-08.”1
This missile had first been seen on 15 April 2012, when what appeared to be six
road-mobile intercontinental ballistic missiles were displayed on transporter-
erector-launcher (TEL) vehicles at a military parade in Pyongyang. It had been
reported by, among others, former U.S. Secretary of Defense Robert Gates that
North Korea was developing such a missile, but the reports seemed incredible
as such a weapon was widely seen as beyond North Korea’s capabilities.
A close analysis of the missiles, henceforth referred to as the “KN-08,” did
little to dispel that assessment. The missiles displayed in 2012 were simple
mock-ups.2
That the missiles displayed on parade were mock-ups, does not indicate
that they represented a hoax. Historically, missiles displayed in parades are
frequently mock-ups even when operational systems are available. Displaying

Received 2 June 2013; accepted 18 July 2013.

Address correspondence to Dr. John Schilling, 3111 Lemonwood Drive, Lancaster, CA
93536, USA. E-mail: [email protected]
210
 North Korean KN-08 ICBM 211
mock-ups is safer in several respects. And it allows the display of missiles
in advance of actual production. The former Soviet Union’s various military
parades through Red Square included mock-ups of operational missiles,
mock-ups of missiles under development, and mock-ups of missiles whose
development had been cancelled but which still served to impress crowds and
mislead western analysts. No firm conclusion can be made either way from
the fact that the missiles displayed in Pyongyang were mock-ups.
But even if only mock-ups were available a year ago, much can change in
a year. There have been many reports, though difficult to confirm, that North
Korea has been conducting an extensive ground-test program for engines and
other hardware associated with the KN-08.3,4 More obviously and spectacu-
larly, on 12 December 2012 North Korea successfully launched a satellite into
Low Earth Orbit using an Unha-3 rocket.5 Satellite launchers and long-range
missiles are fundamentally similar technologies—indeed, many satellites are
launched by repurposed military intercontinental ballistic missiles, and even
if the overall configuration is different it is common for a nation’s first satel-
lite launch vehicles and ICBMs to share common hardware. If nothing else,
the successful launch of the Unha-3 greatly weakens the argument that North
Korea is a nation of technological illiterates incapable of complex missile de-
velopment programs. Reports of shoddy manufacturing processes in North
Korea’s industries abound, but North Korea’s missile designers have clearly
learned to work within such limitations and can produce effective long-range
rockets—though perhaps not consistently or reliably.
It is still plausible that the KN-08 parade was nonetheless a hoax, a de-
ception, or a case of severe technological optimism such that no missile will be
built to match the appearance of the displayed KN-08. North Korea may be-
lieve that the illusion of an ICBM is as effective a deterrent as the real thing.
Possibly the intent is to build a completely different missile, with the KN-08
serving to focus attention elsewhere while the actual program proceeds un-
seen and unmolested. But, so long as this uncertainty is kept in mind, there is
value in attempting to reverse-engineer the design of an actual missile behind
the KN-08 mockups.
What follows is such an analysis, based on the key assumptions that the
actual missile will match those dimensions and features that the various dis-
played KN-08s have in common, and that to the extent possible it will use tech-
nologies North Korea has already demonstrated with at least partial success
in previous work. In particular, it will be assumed that the Unha-3 satellite
launcher serves also as a technology demonstrator for the KN-08 missile, with
a similar technological heritage and with common hardware used wherever
practical.
It is clear that North Korea does not have the ability to build a modern
road-mobile ICBM. In particular, most modern and virtually all mobile ICBMs
use solid-fuel rockets, and large solid rockets are a technology that North
212 Schilling

Figure 1: KN-08 mock-up on WS51200 TEL.

Korea has never demonstrated. But a missile does not need to be modern to
be effective. The question is, can North Korea build an “obsolete” but func-
tional ICBM matching the appearance of the KN-08, using technologies they
do possess?

DIMENSIONS
Figure 1 shows one of several KN-08 missiles displayed on what appear to be
Wanshan Special Vehicle Company WS51200 heavy-duty special off-road ve-
hicle. A wire-frame model is overlaid on the image for dimensional analysis,
along with critical dimensions from Wanshan marketing materials.6 There is
some uncertainty as to whether the 20.11 meter overall length specified for
the WS51200 is for the short-cab version seen in North Korea or the long-cab
version seen elsewhere; if the 20.11 meter figure is for the long-cab version,
the estimates of KN-08 longitudinal dimensions here will be 2–3% high. Lon-
gitudinal dimensions are in any event subject to small uncertainty due to fore-
shortening of the missile in the best available images, which can be estimated
but is difficult to calculate to high precision.
Figure 2 displays only the wire-frame model, in orthogonal projection and
with key dimensions of the missile itself. There is some degree of uncertainty
in these dimensions, and the diameter of the missile stages in particular may
be off by as much as 5%.

INTERIOR CONFIGURATION
Figures 3, Figure 4, and Figure 5 show the estimated internal layout, and key
dimensions, of three plausible models of the KN-08 missiles. These models,
 North Korean KN-08 ICBM 213

Figure 2: Geometry and Dimensions of the KN-08A.

Figure 3: Postulated “KN-08A” internal configuration.

referred to as “KN-08A,” “KN-08B,” and “KN-08C,” differ primarily in the

choice of engines for the various stages, as will be discussed later. Engine bays
are sized for the most probable engines for their respective stages, as will be
discussed later. Propellant tanks are sized to fill the available volume in their
stage at the appropriate propellant mixture ratio, maintaining clearance from
the engines and from such external access ports, ullage/separation rocket noz-
zles, and cable pass-throughs as are visible in images of the KN-08 mockups.
Guidance and control electronics are assumed to be located above the third-
stage propellant tank. It is possible that the short, slightly tapered section at
the front of the third stage is part of the warhead base. If so, it will nonethe-
less probably be hollow and available to house protruding elements of the third
stage; as will be discussed below the bulk of the warhead mass will be located
well forward of the separation plane.

Figure 4: Postulated “KN-08B” internal configuration.

214 Schilling

Figure 5: Postulated “KN-08C” internal configuration.

Figures 6 and 7 give details of two further configurations, “KN-08 Mini-

mum” and “KN-08 Maximum,” included as bounding cases. The KN-08 Min-
imum uses only engines that North Korea is known with high confidence to
have successfully tested, whereas the KN-08 Maximum assumes that North
Korea has successfully reverse-engineered the most sophisticated ex-Soviet
engines in its arsenal and can heavily modify these engines to meet the spe-
cific requirements of the KN-08 missile. Both of these are considered relatively
unlikely, the former on the grounds of limited utility and the latter as being
beyond North Korea’s technical ability, but they are included for the sake of
completeness and to establish upper and lower bounds on the possible KN-08
performance envelope.

ENGINES
The development and testing of a large liquid rocket engine is an extremely
demanding undertaking, which may be beyond North Korea’s technical abil-
ity. Past North Korean missile and space launch efforts have been based en-
tirely on heritage Soviet engines,7 perhaps reverse-engineered and modified by
North Korean engineers. If engines from such sources were at all suitable for
the KN-08 missile, it seems unlikely that the North Koreans would choose in-
stead to design a new engine—and if they did, the probability of success would
be greatly reduced. As the intent here is to analyze missiles North Korea might

Figure 6: Postulated “KN-08 Minimum” internal configuration.

 North Korean KN-08 ICBM 215

Figure 7: Postulated “KN-08 Maximum” internal configuration.

succeed in building in the near future, and as suitable engines are available,
hypothetical new engines will not be considered.
In particular, solid rocket motors will not be considered here. Large solid
rocket motors are unquestionably more durable than comparable liquid rocket
systems, and thus generally preferred for mobile missiles. They are also well
beyond North Korea’s technological capability. This is not a matter of solid
rockets being more “advanced” than liquid (or vice versa), simply different.
North Korea has chosen to specialize in liquid-propellant rockets, and it has
taken roughly a generation to reach the point where ICBM-class vehicles are
within reach. North Korea’s solid-fuel expertise is limited to much smaller bat-
tlefield systems, the KN-02 “Tochka” is less than one-tenth the size of the KN-
08 first stage, and the KN-02 is a direct copy of a Soviet missile. An indigenous
North Korean solid-fuel ICBM is not a credible technical possibility at this
time.
North Korea has demonstrated two engines reasonably well suited for use
in the KN-08 missile, and is believed to possess two others. The first of these is
the Scud engine derived from the ex-Soviet R-17 (aka “Hwasong-6” or “Scud-
B”) missiles, which has been in use by North Korea for over 30 years.8 The
second is the Nodong engine, which appears to be simply an enlarged Scud
engine. The basic Nodong engine was first demonstrated the late 1980s, and
an upgraded version appears to have been used in recent space launches by
Iran and North Korea.9 It should be noted that increasing the size of a rocket
engine is not in fact a simple matter, and it is not clear whether this upscaling
was performed by North Korean, Soviet, or ex-Soviet engineers.
Finally and most recently, North Korea is widely reported to have obtained
a supply ex-Soviet R-27 (aka SS-N-6 “Serb”) submarine-launched ballistic mis-
sile engines and airframes.10,11 Widespread speculation is that these engines
serve as the basis for North Korea’s Musudan missiles. The R-27 includes two
smaller vernier (control) rocket engines, which can be operated even when the
main engine is shut down. Iran’s Safir satellite launch vehicle uses what ap-
pear to be just the R-27 vernier rockets on its upper stage, indicating that the
verniers can (perhaps with some difficulty) be physically separated from the
216 Schilling

main engine and used as the basis for a smaller, independent propulsion sys-
tem.12 Close similarities between the observed dimensions and estimated per-
formance of the upper stages of the Safir and Unha launch vehicles, along with
the known collaboration between North Korea and Iran in liquid-propellant
rocketry, suggest that the North Korean Unha also uses engines derived from
the R-27 verniers as its primary propulsion system.
North Korea also possesses engines from smaller ex-Soviet surface-to-
surface and surface-to-air missiles, which would not normally be suited for use
in a large missile such as the KN-08. Only one such engine will be considered,
that of the S-75 (aka SA-2 “Guideline”) surface-to-air missile.13 This engine can
be adapted for use in ballistic missiles, as demonstrated by the use of two S-75
engines in the Indian Prithvi series of short-range nuclear-capable missiles.
It is considered here because it is the only engine confirmed to be available
and flight-tested in North Korea that would fit in the small third stage of the
KN-08 while leaving enough room for reasonable propellant tanks.
North Korea’s ability to manufacture these engines domestically is in some
doubt; certainly extensive Russian technical assistance was required in the
past, and it may be that North Korea is even now dependant on stockpiled ex-
Soviet hardware for the more complex and sophisticated components of these
engines. However, both the Hwasong-6 and Nodong missiles have been pro-
duced in substantial quantity and exported to Iran, Pakistan, and Syria, and
the missiles have evolved beyond direct copies of ex-Soviet systems.14 Depen-
dence on external assistance and/or legacy hardware does not appear to be a
major limitation on the use of these systems.
With regard to the R-27 missiles and engines the situation is less clear;
North Korea’s own use appears limited to a handful of Musudan missiles that
have probably never been tested along with the possible use of R-27 verniers
(but not main engines) in three unsuccessful and one successful space launch
attempts.15 It is unlikely that North Korea has a robust domestic manufactur-
ing capability for this system, and the best available estimate is that perhaps
150 such engines are unaccounted for from ex-Soviet production.16 Three suc-
cessful Iranian space launches have also been conducted using modified R-27
verniers, or small engines very similar to them. With Iranian assistance, North
Korea may be able to manufacture these smaller engines.
North Korean missile engines have traditionally incorporated jet vanes for
steering; a practice inherited from the former Soviet Union. This is an effec-
tive technique, but reduces the performance of the engine by several percent.
The R-27 missile instead used a pair of separate vernier engines on gimbaled
mounts for this purpose, a rather more efficient technique. Similar verniers
were found on the Nodong-powered first stage of the Unha-3 launch vehicle.
This represents an improvement over the original Nodong missile, which used
a nearly identical main engine with jet vanes rather than vernier engines. The
Unha first-stage verniers used propellants tapped from the main engine pumps
 North Korean KN-08 ICBM 217
rather than an independent propellant feed, but were otherwise of similar size
and appearance to the vernier engines of the Soviet R-27 missile and to the
main propulsion system of the Iranian Safir launch vehicle.17
Sources suggest thrust values of 15–30 kN for the Unha-3 first stage
vernier engine; the larger value is possible, but given the operating cycle would
most likely come at the cost of reduced thrust from the main engine. It will be
assumed here that any Nodong engines used on the KN-08 will incorporate
Unha-style verniers of 15 kN net thrust, similar to the R-27 but incapable of
independent operation. Scud and S-75 engines adapted for the KN-08 will be
assumed to have similar, but proportionately smaller, verniers. Where multi-
ple engines are used in a stage, each main engine is assumed to be coupled to
a single vernier, the same approach used on the Unha-3 first stage.
The North Koreans do not publish performance data on their engines, nor
would they be trusted if they did. Fortunately, the end of the Cold War has
resulted in an increase in the quantity and quality of technical data avail-
able from Russian sources, much of which is relevant to the North Korean
engines.18,19,20 It is also possible to estimate the performance of these engines
by observing the launch acceleration and burn time of North Korean missiles
during tests. Exports of Hwasong-6 and Nodong missiles to nations such
as Iran and Pakistan allow further opportunities for such observation. Esti-
mates of North Korean rocket engine performance vary by several percent,
which will have a small effect on estimated KN-08 performance. There is also
some uncertainty regarding the exact propellant combinations used, which
will affect propellant density and tank volume calculations.
For this analysis, engine performance is assumed as follows. Performance
of the Scud and Nodong engines will be somewhat higher than generally ac-
cepted values due to the presumed replacement of thrust vanes with vernier
rockets (see Table 1).

Table 1: Potential North Korean missile engines

R-27 (Vernier
Engine S-75 Scud Nodong R-27 Only)

Verniers 1 1 1 2 0
Thrust, Sea Level 30 kN 140 kN 300 kN 265 kN N/A
Thrust, Vacuum 34 kN 155 kN 330 kN 290 kN 32.5 Kn
Isp, Sea Level 224 s 230 s 235 s 265 s N/A
Isp, Vacuum 252.5 s 260 s 260 s 290 s 255 s1
Oxidizer AK-20I AK-27I AK-27I MON-10 MON-10
Fuel TG-02 TM-185 TM-185 UDMH UDMH
Mixture Ratio 3.2 : 1 3.5 : 1 3.5 : 1 2.0 : 1 2.0 : 1
Availability Confirmed Confirmed Confirmed Possible Probable

Note: 1280 seconds with nozzle extensions.

218 Schilling

The propellants used in these engines are believed to be as follows:

• AK-20I: 80% Nitric Acid, 20% Nitrogen Tetroxide, plus stabilizers

• AK-27I: 73% Nitric Acid, 27% Nitrogen Tetroxide, plus stabilizers
• MON-10: 90% Nitrogen Tetroxide, 10% Nitric Oxide
• TG-02: 50% Xylidine, 50% Triethylamine
• TM-185: 80% Kerosene, 20% Gasoline
• UDMH: Unsymmetrical Dimethylhydrazine

In addition, it is assumed that the S-75 engine requires a separate supply

of isopropyl nitrate estimated at 2.75% of the main propellant flow to drive the
turbopumps.

AIRFRAME
Virtually all long-range missiles and satellite launch vehicles use the propel-
lant tanks as the main structural elements of each stage, and examination
of the Unha-3 wreckage shows North Korea follows the same practice. While
the North Koreans have never demonstrated the ability or intent to develop
new rocket engines, they have repeatedly and successfully integrated existing
engines into new airframes of their own design. It is thus reasonable to as-
sume that the KN-08 will use new propellant tanks and structural elements
optimized for its requirements, built around existing engines.
Wreckage recovered from the most recent Unha-3 launch included a
largely intact first stage oxidizer tank with associated structure and plumb-
ing.21 The tank was of semi-monococque (structural sheet metal welded to an
underlying frame) construction, using aluminum-magnesium alloy. The tank’s
bulkheads were of conventional dome geometry with an aspect ratio of 0.25,
and it appears that separate bulkheads were used for the fuel and oxidizer
tanks. Common-bulkhead tanks would allow a somewhat greater propellant
load, and are part of the R-27 technology base, but have not traditionally been
used with North Korean Scud/Nodong-heritage missiles and were not seen on
the Unha-3.
This analysis will assume the same geometry, structure, and materials are
used in the KN-08. In all three stages, propellant tanks are assumed to fill all
space not required by engine bays or other equipment, with a minimum of 5 cm
of dynamic clearance between engines and tanks of the same or successive
stages. This also is consistent with the observed design of the Unha-3. The
performance of the KN-08 would be increased slightly if common-bulkhead
tanks were used on upper, R-27 heritage stages.
 North Korean KN-08 ICBM 219
For stages using the R-27 engine, it is assumed that the engine is sub-
merged in the rear of the fuel tank as was done in the original Soviet R-27
missile. In addition to providing a more efficient packaging, this is probably
necessary to provide adequate cooling and fuel feed to the engine, and to trans-
fer the thrust of the engine to the missile’s airframe.
In addition to the fuel and oxidizer tanks, each stage will require smaller
tanks of compressed gas (either helium or nitrogen) to pressurize the main
propellant tanks and possibly to drive pneumatic actuators. These tanks are
typically spherical, located around the periphery of the engine bays or inter-
stage regions, and are not depicted in the configuration drawings above. It is
also assumed that the S-75 engine requires tanks of isopropyl nitrate mono-
propellant to drive the propellant pumps; as these tanks do influence the size
and layout of the engine bay they are shown in the drawing.

GUIDANCE AND CONTROL

Long-range ballistic missiles almost invariably use inertial guidance; radio
command guidance is also possible and has been used on early cold-war era
weapons, but is subject to jamming. North Korea has demonstrated the ability
to use basic inertial guidance with its Nodong missiles and to some extent its
Unha space launch vehicles and it is assumed that inertial guidance is used
here. In its simplest form, this involves aligning the vehicle’s gyro axis with
the desired launch azimuth (either on the launch platform or via a roll maneu-
ver immediately after launch), and flying a precalculated pitch vs. time profile.
Third-stage engine cutoff is performed when an integrating accelerometer de-
termines that the appropriate velocity has been reached.22
Guidance commands are implemented using the vernier engines referred
to earlier, with two gimbaled engines per stage being sufficient for full three-
axis flight control. First- and second-stage engines will most likely run at full
throttle until the onset of propellant depletion is detected, at which point a se-
quencer will shut down the engine and activate explosive bolts and small solid
rockets as required for stage separation. For the third stage of the KN-08,
most likely powered by an R-27 engine, it will be assumed that the engine will
shift to vernier-only operation at least 10 seconds before command shutoff to
facilitate fine velocity control. A similar system was used on several early So-
viet missiles, including the R-27, and allows somewhat greater accuracy than
would be possible with the coarse thrust termination profile of the main en-
gine.23
The actual guidance system is almost certainly located in the third stage,
and from packaging considerations probably above the third-stage oxidizer
tank. The R-27 placed the guidance system in a sealed compartment within
the tank, and such a configuration cannot be ruled out here. Actual guidance
220 Schilling

hardware would consist of gyros (probably mechanical), accelerometers, and

either a basic computer or a set of sequencers and feedback controllers. The
details cannot be determined at this time, but will probably include common
heritage with North Korea’s previous missiles and/or with the R-27. Guidance
commands are sent to lower stages via the external cable guides clearly visible
on KN-08 imagery.

CONFIGURATION DETAILS
The first stage is in most cases assumed to use two Nodong engines. North
Korea has demonstrated a four-engine Nodong cluster on the first stage of
the Unha series of space launch vehicles, which have a base diameter of 2.4
meters.24 A two-engine cluster would fit a vehicle with a 2.0 meter base diam-
eter with almost exactly the same clearances, and would have sufficient thrust
to provide about 1.8 g launch acceleration. The bulky pump arrangement of
the Nodong engine, in a clustered installation, would require an engine bay
roughly 2.3 meters long.
A cluster of four Scud-type engines would also fit in the base of the KN-
08, and would allow for a somewhat shorter engine bay. This would have the
advantage of allowing an extra 3,600 kg of fuel to be carried. Unfortunately,
the lower performance of the Scud engine coupled with the increased weight
of the vehicle would reduce the launch acceleration to only 1.4 g. Most of the
additional propellant would be spent fighting gravity losses, and the overall
performance of the missile would be reduced.
The R-27 is a submerged engine, tightly integrated with the base of the
propellant tank.25 This allows a much greater packaging density, which is ad-
vantageous for a mobile missile, but it greatly complicates any attempt at
clustering the engine. The R-27 is intended and optimized for centerline use;
clustering it may be beyond North Korea’s ability and certainly has not been
demonstrated. Furthermore, a three-engine cluster would be required to pro-
vide sufficient thrust, and the R-27 is too bulky to fit three engines in a 2-meter
vehicle without substantial redesign.
Increased first-stage performance could also be achieved with a single,
larger engine using R-27 technology. The ex-Soviet R-29 missile engine would
be a good fit, but there is no evidence that this engine has reached North Ko-
rea. North Korea has never developed an engine in this class itself, and even
its more modest engine development programs have probably required foreign
assistance. These are all extremely remote possibilities. Nonetheless, in order
to place an upper bound on KN-08 performance, a “KN-08 Maximum” config-
uration is presented using a cluster of three heavily-modified R-27 engines. A
single R-29 or comparable North Korean engine would offer essentially identi-
cal performance.
 North Korean KN-08 ICBM 221
The second stage could use a single Nodong engine, a pair of Scud engines,
or a single R-27 engine. All three combinations will be considered, and will
be referred to as the “KN-08A” (Nodong 2nd stage), “KN-08B” (dual Scud 2nd
stage), and “KN-08C” (R-27 2nd stage). To be used as part of this arrangement,
the Scud engine would require an approximate 50% increase in operating life;
the R-27 would require an approximate 30% increase and would have to be in-
tegrated to a tank of substantially larger than normal diameter, as the engine
would remain on the centerline, the critical engine/tank interface could remain
unchanged.
For the upper stage, only the R-27 is considered for the baseline configu-
rations. Because of the short overall length of the upper stage, the engine bay
required for a Nodong or even Scud engine would leave little room for propel-
lant tanks. The stage length, along with the location of visible cable ducts, fill
ports, and access panels, strongly suggests a submerged-engine design, and
the R-27 is the only submerged engine North Korea is believed to possess.
In this application, an R-27 engine (including verniers) would be inte-
grated into a somewhat smaller tank and airframe than is the case on the
original R-27 missile. The appearance of the Musudan missile suggests that
North Korea might be comfortable with the idea of integrating R-27 main en-
gines with new tankage.26 As can be seen from the cross-section in Figure 3,
the third stage of the KN-08 is close to the minimum size needed to accom-
modate an R-27 engine without repositioning the turbopump and associated
plumbing. This is perhaps not a coincidence; if a three-stage missile of the KN-
08’s approximate scale were built without constraints, an optimized third stage
would be roughly two-thirds the size of that seen on the KN-08. A constraint
to use a stock R-27 engine would explain this discrepancy.
Another possible use of R-27 technology would be to use just the vernier en-
gines, as was done with the Iranian Safir and probably the North Korean Unha
satellite launchers. A single pair of verniers, while optimal for deployment of
small satellites, would provide inadequate thrust for delivering heavier war-
heads. There is, however, room for three pairs of verniers and their shared
turbomachinery around the base of a KN-08 third stage. The verniers are in
this case assumed to have nozzle extensions similar to those demonstrated by
Iran, increasing specific impulse to 280 seconds. The “KN-08BV” configuration
is identical to the “KN-08B” except for the use of a vernier-only upper stage;
this represents the most potent missile North Korea could build if it has access
to R-27 or similar verniers but not main engines.
If neither R-27 main engines nor their verniers are available to the North
Koreans, the most plausible alternative is a pair of S-75 / SA-2 engines, and
this forms the basis of the postulated “KN-08 Minimum” configuration. Other
engines from ex-Soviet surface-to-air or short-ranged ballistic missiles might
also be adapted, but would be unlikely to exceed the performance of the S-75
configuration considered here.
222 Schilling

WEIGHT AND VOLUME

The weight of the propellants can be determined from the estimated volume
of the propellant tanks and the density of the propellants. As this is a mobile
missile, caution is required—it will not be possible to completely load the tanks
to maximum density at short notice in a remote site. It will be assumed that
the propellant load is constrained to leave 2.5% ullage volume in the tanks
after allowing for thermal expansion to 30◦ C. For stages using the submerged-
engine R-27, this will be increased to 3.5% due to the possible difficulties of
dealing with the more complex tank geometry.
Weight of engines, airframes, and guidance hardware is much more dif-
ficult to estimate. Here the best that can likely be done is to make an esti-
mate by comparison to other known systems. We do not have reliable mea-
surements of the mass of the Unha-3 first stage, but as noted above it was of
semi-monococque construction using aluminum-magnesium alloy. The North
Korean Nodong missile has been exported to several nations and conducted
a number of highly visible test flights; from this it is possible to estimate
that the total dry mass (exclusive of warhead and guidance) is 11% of the
launch weight of the missile.27 The Nodong uses the same semi-monococque
construction of the Unha-3, but with heavier steel alloys. Russian data indi-
cates the ex-Soviet R-27 missile has a 9.5% stage mass fraction (again exclu-
sive of warhead and guidance), achieved with lightweight aluminum alloys,
a more sophisticated orthogrid construction, and a more tightly integrated
engine.28
The KN-08 is expected to fall between those values. It presumably draws
from the same technological toolkit as the Unha-3, which itself is based on
North Korea’s Nodong heritage as well as some imported R-27 components.
Stages using Scud/Nodong engines will thus be assumed to have a 10.5% mass
fraction, and stages with the more sophisticated and tightly integrated R-27
engines a 10.0% mass fraction. These masses include structure, propellant
tanks (which are themselves major structural elements), engines, plumbing,
and ancillary hardware such as stage separation mechanisms and control ac-
tuators.
Guidance systems of early U.S. ICBMs had weights of 100–300kg, using
technologies similar to those assumed for the KN-08.29 North Korea demon-
strated a certain degree of competence in torpedo and missile manufacture,
which indicates an ability to produce mechanical or electromechanical guid-
ance components of reasonable precision; the Unha wreckage showed a will-
ingness and ability to supplement domestic manufacture with imported mod-
ern consumer-grade electronics. It is also possible that North Korea’s access to
ex-Soviet R-27 hardware includes useful guidance components. It is therefore
reasonable to assume that North Korea can at least match the performance of
the better early U.S. ICBMs in guidance system mass, if not precision.
 North Korean KN-08 ICBM 223
As will be described, warhead weights of 500–1500 kg will be considered.
This includes the complete reentry vehicle, including aeroshell and thermal
protection system.
A final consideration is propellant residuals. It is impossible to completely
drain a missile of propellant. There is always some propellant trapped in dead
volumes of the plumbing. It is also impossible to precisely balance the fuel and
oxidizer load so that both are exhausted at the same time – and it is important
to shut down the engine before either propellant is completely exhausted, or
the turbopump will overspeed and destroy the engine. Standard practice is to
provide a slight excess of fuel, and shut down the engine at the first warning of
oxidizer depletion. In modern rockets and spacecraft, it is sometimes possible
to consume as much as 99.5% of the propellant.30 For North Korea’s more basic
designs, this analysis will assume 2% residual propellant in Scud- and Nodong-
based stages, with 3% residual propellant in stages based on the submerged-
engine R-27. Based on these assumptions, the mass breakdown of the notional
KN-08 configurations are shown in Table 2.

WARHEAD
The KN-08 uses a triconic reentry vehicle (RV) of slightly greater than 3 meters
overall length. Modern U.S., Russian, British, and French warheads use slen-
der single-cone RVs which offer better terminal performance – reduced drag
translates to greater accuracy and higher terminal velocity (i.e. less exposure
to terminal-phase missile defenses). However, the tip of a slender RV reenter-
ing the Earth’s atmosphere at ICBM velocities is subject to an extremely chal-
lenging thermal environment. Also, the single-cone design makes it difficult to
package a large-diameter nuclear device while preserving balance and aero-
dynamic stability. The triconic warhead is a reasonable compromise between
terminal performance, thermal design, and payload packaging.
It is important to note that in a triconic warhead, the rear conical region is
primarily a drag device to moderate terminal velocity and ensure aerodynamic
stability. Weight and balance considerations demand that the bulk of the pay-
load mass be in the central cylindrical region or, if possible, the forward tip.
Thus the rear cone can be largely empty space. So while it is unclear exactly
where the KN-08 third stage/warhead separation plane is located, there should
be no difficulty with the third-stage oxidizer tank or guidance package project-
ing forward into the conical base of the warhead.
The weight of North Korea’s nuclear weapons is unknown, and sub-
ject to a great deal of speculation. If indeed North Korea is unable to pro-
duce devices smaller than the Trinity Gadget or Hiroshima’s “Little Boy”,
there is no prospect of a nuclear-tipped Korean ICBM. However, this seems
unlikely—every technique ultimately used to produce lightweight nuclear
224
Table 2: KN-08 Missile Stage Mass Breakdown

Stage Mass KN-08 Min KN-08A KN-08B KN-08BV KN-08C KN-08 Max

1st Propellant, total 16,130 kg 16,130 kg 16,130 kg 16,130 kg 16,130 kg 16,640 kg

Propellant, usable 15,805 kg 15,805 kg 15,805 kg 15,805 kg 15,805 kg 16,140 kg
Total Dry 1,890 kg 1,890 kg 1,890 kg 1,890 kg 1,890 kg 1,850 kg
2nd Propellant, total 11,250 kg 8150 kg 11,250 kg 11,250 kg 10,080 kg 10,080 kg
Propellant, usable 11,025 kg 7985 kg 11,025 kg 11,025 kg 9,780 kg 9,780 kg
Total Dry 1,545 kg 950 kg 1,545 kg 1,545 kg 1,120 kg 1,120 kg
3rd Propellant, total 3330 kg 3,135 kg 3,135 kg 3700 kg 3,135 kg 3,135 kg
Propellant, usable 3260 kg 3,040 kg 3,040 kg 3630 kg 3,040 kg 3,040 kg
Total Dry 560 kg 450 kg 450 kg 535 kg 450 kg 450 kg
All Total, w/o warhead 34,705 kg 30,705 kg 34,400 kg 35,050 kg 32,805 kg 33,275 kg
 North Korean KN-08 ICBM 225

Figure 8: Possible KN-08 Warhead Packaging.

weapons in the early 1950s was known to the Manhattan project team in 1945,
was successful in its first postwar test, and is now in the public domain.31
The first generation of U.S. lightweight nuclear weapons included the Mark
5 (1400 kg total weight, 120 kiloton maximum yield) and Mark 7 (750 kg, 60
kt), both introduced in 1952, and 1954’s Mark 12 (450 kg, 15 kt). Compara-
ble devices were developed in Russia, Britain, France, and China shortly after
their first nuclear tests.32 It should be noted that these early weapons were
aircraft-delivered bombs and short-range missile warheads. An ICBM war-
head will necessarily be somewhat heavier due to the increased requirement
for thermal protection. However, as noted earlier, the triconic warhead design
reduces the peak thermal load. And even the Apollo spacecraft, with a vastly
more challenging thermal environment, had a head shield weighing no more
than 15% of the total reentry mass.
It is reasonable to assume that North Korea’s first missile-delivered war-
head will fall within the same approximate weight range, if not yield. Indeed,
the low yield of North Korea’s nuclear tests to date suggests they may be at-
tempting to produce as light and compact a weapon as their technology will
allow. A range of 500–1500 kg total warhead mass should cover any plausible
first-generation North Korean ICBM warhead. Warhead yield could plausibly
be anywhere from four to sixty kilotons for a pure fission device; the largest
North Korean test had a yield of perhaps ten kilotons, but a modest degree of
upscaling is possible without additional testing.
Warhead dimensions are also a concern. Figure 8 shows three early U.S.
nuclear warheads for which basic configuration data are publicly known, pack-
aged in the KN-08 reentry vehicle. This is not meant to suggest that North
Korea is actually copying old U.S. warhead designs; the limitations of nuclear
weapons technology will drive any emerging nuclear power to fairly similar
solutions. In particular, the U.S. Mark 7 represented an effort to develop a
warhead light enough for delivery by tactical aircraft or ballistic missiles, with
minimum technical risk and efficient use of scarce fissile materials. The KN-08
RV is clearly suitable for delivering a similar device.
Two other possibilities would be gun-assembly fission devices and first-
generation thermonuclear weapons with cylindrical primaries. The second
226 Schilling

example in Figure 8 is based on a U.S. Mark 10 gun-assembly device of 700 kg

total weight and 15 kiloton yield. This is a very simple weapon, easily within
the capability of any nuclear power, but is rather inefficient in its use of fis-
sile materials and consequently never entered U.S. service. The third example
shows a Mark 28 thermonuclear weapon, the first lightweight thermonuclear
device in the U.S. arsenal, yielding 1.4 megatons with a 900 kg total weight.
Lightweight thermonuclear weapons are almost certainly well beyond North
Korea’s present capabilities, but perhaps not beyond North Korean ambitions.

PERFORMANCE
With assumed stage mass and engine performance values, it is possible to esti-
mate the performance of the KN-08 missile. A high-fidelity performance model
would also require aerodynamic data, moments of inertia, control system au-
thority and response times, detailed operational constraints, etc, to support
a full six-degree-of-freedom model of the vehicle’s trajectory. While estimates
can be made for these parameters, they would be estimates only—and given
the uncertainties already accepted in weights and engine performance, exces-
sive effort along these lines is not warranted. At best only an approximate
prediction of range vs. performance can be made, with an accuracy of perhaps
±15% in range.
One operational constraint that must be considered even at this level of
analysis is 3rd stage burnout acceleration. If a stage powered by the R-27 en-
gine is flown to (near) propellant depletion with a light warhead, the acceler-
ation at burnout will be roughly 25 g. This is unprecedented in an ICBM or
space launch vehicle, and would require an unusually heavy structure to with-
stand the associated loads. Therefore, it may be necessary to shut down the
R-27 core engine early and complete the third-stage burn using vernier rock-
ets only if flying with a light warhead. As the nominal burnout acceleration
of the R-27 SLBM was approximately 13.5 g, it will be assumed that the KN-
08 third stage must shift to vernier-only operation when acceleration reaches
13.5 g.
Because the effect of the Earth’s rotation is not negligible for ICBM-type
trajectories, one cannot simply give a precise range—the location of the launch
site, and the direction of the launch, will affect the distance traveled. As the
KN-08 is apparently a mobile missile, the launch site could in principle be any-
where in North Korea. This analysis will assume a launch out of Musudan-ri
with a launch azimuth of 45◦ . Any launch from North Korea against a CONUS
target will approximate this trajectory; against targets in Europe or Asia the
range of the KN-08 would be somewhat reduced.
The resulting prediction of throw weight vs. range for the three postulated
KN-08 variants is shown in Figure 9. “Throw weight” refers in this case to the
 North Korean KN-08 ICBM 227

Figure 9: Predicted KN-08 Performance

complete mass of the separated warhead section; as noted earlier 500–1500 kg

covers the plausible range for first-generation fission warheads.
As can be seen, the KN-08 “Minimum” configuration, using only
Scud/Nodong technology, has a maximum range of just over 7,000 km with
the lightest plausible warhead. As the distance from Musudan-ri to Seattle is
approximately 7,950 km, such a missile would probably not pose a threat to
CONUS targets though it could reach almost any target of interest in East
Asia or the Pacific.
The KN-08A, B, BV, and C configurations, using mixed Scud/Nodong and
R-27 technology, would have a limited ability to reach CONUS targets with
light warheads. The KN-08A and B could perhaps deliver 500–600 kg war-
heads to Seattle or elsewhere in the Pacific Northwest, with the BV having
slightly greater performance with light warheads. The postulated KN-08C con-
figuration could reach such targets with 750 kg warheads or deliver lighter
warheads as far as Los Angeles or Denver. Alternately, a lightweight warhead
could be delivered to Seattle in the company of perhaps half a dozen credible
decoys. Central Asia and Europe would also possibly be within reach.
The KN-08 “Maximum” configuration would be a much more formidable
weapon, particularly with a lightweight warhead. Such a missile could reach
targets as distant as New York or Washington, DC, or could attack any west
coast target with either a heavier warhead or a light warhead plus decoys. As
228 Schilling

noted earlier, the KN-08 “Maximum” requires first-stage engines well beyond
any capability North Korea has demonstrated or is generally believed to pos-
sess.
While a rough quantitative prediction of range and payload can be made
from the external dimensions and geometry of the missile, accuracy will de-
pend on details of the guidance and control hardware that are harder to pre-
dict. The R-27 missile on which the upper stage(s) of the KN-08 are assumed to
be based, had a circular error probable (CEP) of 1.9 km at a range of 2400 km,
according to Russian sources.33 Ballistic-missile accuracy is largely a function
of velocity errors at burnout; if the KN-08 uses R-27 vernier rockets for fine
control and can at least match R-27 inertial measurement accuracy (perhaps
by using heritage R-27 accelerometers and gyros), it should have similar veloc-
ity errors.
As the KN-08 will have a substantially longer time of flight than the R-
27, the accuracy will be proportionately reduced velocity errors will have more
time to impact the final trajectory. At a range of 8000 km, a KN-08 which can
match the stated guidance performance of the R-27 would have an error ellipse
of approximately 4 km cross-range and 8 km downrange, half of all missiles
fired would be expected to fall within such an ellipse. If the North Koreans can
only match the accuracy demonstrated by their previous Nodong missiles, the
error ellipse of the KN-08 would be roughly 10 × 20 km at intercontinental
ranges.
North Korea’s most powerful nuclear test had a yield of approximately
10 kilotons, sufficient to destroy soft targets within a 1.5 km radius.34 The
U.S. Mark 7 bomb/warhead, a reasonable proxy for the largest first-generation
fission warhead that North Korea could plausibly deliver to CONUS targets by
a KN-08 missile, had a maximum yield of 60 kilotons, which still corresponds
to a soft-target destruction radius of only 2.8 km.35 A KN-08 missile cannot
reliably be used to destroy any specific target, and certainly not a hardened
military target. It would probably be sufficiently accurate to hit large urban
areas, but not to target specific neighborhoods or districts.

MOBILITY
The KN-08 mockups were displayed on what appear to be Transporter-Erector-
Launcher (TEL) vehicles with substantial cross-country performance. The
chassis, at least, is probably similar to that used on China’s DF-31 mobile
ICBMs. This gives the impression of a highly mobile field-deployed military
system, capable of being dispersed anywhere in North Korea and rapidly re-
located to evade attack. Such a capability is seen with smaller mobile mis-
siles such as the Scud/Hwasong series, and is of obvious value to a nation like
 North Korean KN-08 ICBM 229
North Korea whose limited strategic military forces are almost constantly in
the crosshairs of vastly more powerful nations.
Unfortunately for North Korea’s military planners, it is difficult to make
a large liquid-fueled missile highly mobile. Mobile IRBM and ICBM-class
weapons are almost invariably solid-fuel rockets. Liquid fuel tanks are inher-
ently less robust than solid motor casings, and at large scale the loads asso-
ciated with transporting a liquid-fuel rocket on an off-road vehicle would be-
come prohibitive. Even on-road travel, except at low speed on a flat paved road,
would risk damaging a fully fueled KN-08.
It probably would be possible to achieve a high degree of on-road mobility
with an empty KN-08; the fuel represents more than 80% of the weight and
thus load. This would then require that the missile be fuelled at the launch
site, after being transported and (probably) erected. Similar systems were used
in the early Cold War era; the U.S. Redstone and Jupiter missiles could be fu-
eled and launched within fifteen minutes.36 While these were of similar scale to
the KN-08, they were single-stage vehicles. Soviet R-9A (SS-8) missiles could
be fueled and launched in 20 minutes – this time included installation of a mis-
sile on a launch pad (it should be noted, though, that the launch sequence was
highly automated).37 Fueling a KN-08, with three stages and two propellant
combinations, would probably take half an hour or so. Other steps necessary
to prepare and fire the missile, not all of which can be conducted in parallel
with fueling, would probably increase the prelaunch time scale to at least an
hour.
It would also be necessary to accompany the KN-08 TEL with a small fleet
of support vehicles. At least four fuel tanker trucks would be required, and
probably a mobile launch control center. And it is unlikely that the North Kore-
ans would deploy such a valuable asset without robust security, maintenance,
and communications capabilities. Far from the image of a rugged cross-country
vehicle playing hide-and-seek in the North Korean mountains while prepared
to launch a nuclear strike on a minute’s notice, a KN-08 missile would probably
be the center of a significant military force, at least a dozen vehicles, tied to the
local road network, and would require an hour or more for launch operations.
As North Korea is a geographically small opportunity subject to intense
foreign scrutiny, this raises the possibility that a KN-08 could be destroyed
on the ground before it could launch. Such preemption would require a highly
responsive strike capability, preferably locally based (i.e., in South Korea or
on warships in the Yellow Sea or Sea of Japan), and a willingness to order
decisive military action in short notice. During an acute crisis justifying an
elevated alert posture, it could be an effective means of neutralizing a small
KN-08 force.
An alternate concept of operations would be to maintain the actual KN-08
missiles in hardened caves or tunnels, using the TELs only to roll the mis-
siles out to prepared launch sites immediately outside. China is known to have
230 Schilling

deployed some of its early ICBMs in this fashion, and it would probably be
possible to transport and erect a fully-fueled KN-08 if the trip were only a few
hundred meters along a paved road.38 An additional advantage with semi-fixed
basing is that the missiles could be fueled in a climate-controlled environment.
The analysis given previously assumed a requirement that the missile with-
stand temperatures from −25 to + 30◦ C; more propellant can be loaded if the
missile can be maintained at room temperature, corresponding to a range in-
crease of up to 250 km. Reliability would also be enhanced.
TELs could also be used to shuffle real and dummy missiles between a
large number of hardened sites in peacetime, to complicate any attempt at a
preemptive strike. While the North Koreans are certainly willing to show off
KN-08 mock-ups to confuse their adversaries, they appear to have purchased
no more than eight suitable TELs and it is unclear whether future procure-
ment from China will be possible.39

RELIABILITY AND TESTING

For a KN-08 to deliver a warhead, neglecting for the moment the unreliability
of North Korean nuclear warheads, three independent rocket stages must each
work properly. Two staging events must occur properly. And a high-energy at-
mospheric reentry must be successfully performed. North Korea has attempted
four launches of a rocket stage based on clustered Nodong technology, two of
which have failed. At most nine rocket stages using a single Nodong engine
have been flown by North Korea, with perhaps a single failure. The actual
number may be smaller, as the details of some Korean missile tests are un-
clear. There have been no known tests of stages using the R-27 main engine,
discounting unconfirmed reports of a test conducted in Iran using hardware
supplied by North Korea. North Korea has attempted six staging events with
one failure, discounting cases where booster failures rendered staging moot.
There have been no successful North Korean re-entry tests at anywhere close
to ICBM velocities.40
Considering only North Korea’s relevant test history, a simple statistical
analysis suggests the probability of the first KN-08 launch being fully suc-
cessful is roughly 6%. Extensive ground testing of the individual stages could
at most increase this to 22%, as staging and re-entry cannot be meaningfully
tested on the ground. And even a single successful test would give only an
estimated 35% reliability for the KN-08, given the past history of failures in
similar systems. In order to establish even a minimal degree of confidence in
the system, multiple flight tests would have to be conducted before the KN-08
could enter service as an operational weapon.
North Korea’s crude manufacturing methods might suggest that North
Korean missiles must be unreliable, but capable engineers can work around
 North Korean KN-08 ICBM 231
known manufacturing deficiencies. North Korean missile designers have no
doubt ample experience in such matters. There is no substitute for actual flight
testing in this process, however. Even the most aggressive ICBM development
programs elsewhere would not consider a weapon to be operational without
5–10 test flights, to validate both the design and the manufacturing process,
and North Korea’s Nodong MRBM conducted perhaps five flight tests before
operational deployment.41
It is conceivable that North Korea may attempt to deploy the KN-08 mis-
sile, and may consider it to be an operational weapon, without adequate flight
testing. No other nation has ever deployed an ICBM in this manner, and it
would be extremely unusual for North Korea to do so, but there is some prece-
dent with similar systems. Both Germany and the United States entered World
War II with torpedoes—the most complex weapons systems of their time—with
a single successful live-fire test in the U.S. case and none in the German. These
torpedoes suffered extremely high failure rates—up to 70%, in early opera-
tional service. New space launch vehicles, which are technically very similar
to ICBMs, frequently carry operational payloads on their first flights. Approx-
imately 50% of these launches fail, usually due to fundamental design faults
that must be corrected before a successful launch can occur. And North Korea’s
Musudan IRBM, if it not a hoax, appears to have entered service with zero or
one flight tests.
To deploy an untested strategic weapon may be foolish, but hubris is one
of the more persistent human follies. And if North Korea intends to use its
nuclear arsenal purely as a deterrent, it may not even be foolish—a deterrent
weapon needs to be credible, which is not quite the same thing as being reli-
able. Successful tests enhance deterrence, failed tests weaken it, and the North
Koreans may feel that an untested missile still poses enough of a threat to de-
ter foreign attack. They may be right—in which case, why risk a test failure?

OPERATIONAL AVAILABILITY
The KN-08 probably did not exist as even a flyable prototype in April of 2012.
While it is possible that the North Koreans would display mock-ups simply to
avoid risk of damage to real missiles, the mock-ups in that case would be built
to match the exterior design of the missiles—and as we have seen, the mock-
ups do not even match each other.42 It is possible that the parade mock-ups
were constructed specifically for the parade, or alternately that they were orig-
inally engineering models constructed at different phases of the development
process. In either case, the differences between the mock-ups may suggest that
the detailed design of the KN-08 had not been finalized at the time they were
built.
232 Schilling

Some shared heritage of the postulated KN-08 and the observed Unha-3
means that the North Koreans will not have to conduct an entire development
program from scratch to build and fly a KN-08. The Unha vehicle is the result
of at least eight and possibly as long as twenty-six years of effort, presumably
in parallel with North Korea’s current ICBM development program (whether
KN-08 or otherwise), and as the Unha has now been successfully tested, the
relevant technology and even specific hardware for the KN-08 may have been
validated.
It is unlikely that actual KN-08 missiles would have been built in advance
of the successful Unha-3 test, as it could not have been known in advance that
the test would have successfully validated shared elements of the design. If the
KN-08 does in fact represent North Korea’s planned ICBM, it would probably
still require at least several months and possibly more than a year to incorpo-
rate the lessons learned from the Unha-3 flight and construct a flightworthy
prototype missile. No KN-08 flight test has been observed as of this writing,
though at this point one could plausibly occur at any time.
As noted above, it is conceivable that North Korea may proceed to deploy
the KN-08 as an operational system without actual flight tests of that missile.
This could be driven by internal political and economic constraints, or by the
perception of an imminent external threat. If so, the North Koreans may very
well deploy a fundamentally flawed weapon that cannot possibly work, and
most likely will deploy an extremely unreliable weapon that works less than
half the time. If this is the case, the January 2013 New York Times report
may indicate the actual operational deployment of what North Korea believes
is an operational weapon, presumably based on overconfidence following the
successful flight of the related but not identical Unha-3.
More likely, any observed field operations represent pre-operational train-
ing activities. Before a weapon can be reliably operational, crews must train in
the relevant operating procedures—particularly for a mobile liquid-fuel mis-
sile, where a launch site must be established in an austere field environment
on short notice, and a complex fueling operation must be conducted with ex-
tremely toxic propellants. As North Korea clearly possesses real TELs and mis-
sile mockups, it is reasonable that launch crews may begin training some time
in advance of actual availability of operational KN-08 missiles.
In this case, initial operational capability for the KN-08 would follow a suc-
cessful flight test program, probably by several months. But, as noted above,
it is unlikely that the first test will be successful, and each failed test would
require at a minimum several months to investigate the cause and implement
corrective action. And more than a single success would be required to estab-
lish reasonable confidence in the system. Initial availability for the KN-08 is
therefore probably at least a year in the future, with highly visible flight test
activity as an indication to outside observers. Without observed flight tests
of the KN-08, we may assume that North Korea has foolishly deployed an
 North Korean KN-08 ICBM 233
untested and highly unreliable missile, or that the KN-08 is an outright hoax,
or that the missile is still under development and may be operationally de-
ployed sometime in the future.
If and when a KN-08 ICBM does become available, it will almost certainly
be in very limited numbers. A KN-08 with intercontinental range will require
at least one and probably two or more R-27 engines. Unless North Korea man-
ages to establish an independent ability to mass-produce this engine, they will
be dependent on ex-Soviet missiles. There are probably about 150 flightwor-
thy R-27 airframes not otherwise accounted for, to support all North Korean
Musudan and KN-08 missile production, Unha space launch attempts, ground
and flight testing, and export sales. Also, the supply of suitable TELs may be
limited now that it has been made clear to all concerned that any WS51200 in
North Korean hands is a missile transporter. The six to eight delivered TELs
with two to three missiles per TEL may represent the whole of North Korea’s
planned ICBM force.

CONCLUSION
While the “KN-08” missiles paraded through Pyongyang last April were almost
certainly non-functional mock-ups, it is quite possible that they represent a
missile presently under development. If this is the case, it is unlikely that the
missile is presently operational, and initial operational capability is probably
several months to years in the future and would almost certainly be preceded
by a series of flight tests.
Based on the exterior appearance of the KN-08 mockups, and on North Ko-
rea’s relevant experience in large rocket and missile development, it is possi-
ble to reconstruct a range of credible designs for a KN-08 ICBM within North
Korea’s assumed technological capabilities. If reports that North Korea has
obtained ex-Soviet R-27 missiles (or at least their engines) are true, the KN-
08 would likely be able to deliver first-generation fission warheads to at least
some CONUS targets. Without R-27 engines, it is highly unlikely that the KN-
08 would be able to reach any CONUS target, but it could threaten Alaska,
Hawaii, and most of East Asia.
If an operational KN-08 is deployed, it will be a weapon of extremely lim-
ited capability. It will not be a truly mobile missile, requiring substantial sup-
port infrastructure at the launch site. At best this support infrastructure could
be transported by truck and set up in the hours immediately prior to launch.
Based on previous North Korean experience with large multistage rockets, re-
liability will be poor. Accuracy also is likely to be poor, though sufficient to hit
large urban areas. No more than one to two dozen missiles are likely to be
available, and probably fewer launchers.
234 Schilling

Again, it must be noted that this is an analysis of North Korean capabili-

ties in light of their recent displays; capability is not intent. It is also plausible
that North Korea intends to build an entirely different long-range missile or
none at all. And it is possible that their intent is to build a missile that is in
fact beyond their technological capabilities, such that no operational system
will ever emerge from behind the KN-08 veil.
The message of the KN-08’s display, the Unha-3 satellite launch, and the
February 2013 nuclear test is that North Korea will soon have a nuclear deter-
rent capable of holding the United States at bay by threatening the destruction
of American cities. That message may be a bluff. It is, just barely, credible.

NOTES AND REFERENCES

1. T. Shanker and D. Sanger, “Movement of Missiles by North Korea worries U.S.”
New York Times, 17.January 2013.
2. M. Schiller and R. Schmucker, “A Dog and Pony Show: North Korea’s New
ICBM.” The Arms Control Wonk, 18 February 2012, http://lewis.armscontrolwonk.com/
files/2012/04/KN-08 Analysis Schiller Schmucker.pdf.
3. G. Oh, “N. Korea Tested Long-Range Missile Engine Before Nuke Blast:
Sources.” Yonhap News Agency, 17 February 2013, http://english.yonhapnews.co.kr/
national/2013/02/17/29/0301000000AEN20130217003300315F.HTML.
4. “North Korea Developing New Long-Range Missile: Report” Channel NewsAsia,
14 April 2012, http://www.channelnewsasia.com/stories/afp asiapacific/view/1195153/
1/.html.
5. J. Marcus, “Key Step in North Korea’s Missile Ambitions” BBC News, 12 December
2012, http://www.bbc.co.uk/news/world-asia-20694331 (25 Feb 2013).
6. Marketing brochure, Hubei Sanjiang Space Wanshan Special Vehicle Company,
Ltd., ww.wstech.com.cn. Note: all reference to the WS51200 was removed from the Wan-
shan corporate web site in May 2012.
7. T. Postol, A Technical Assessment of Iran’s Ballistic Missile Program, 2–8. East-
West Institute, 2009, http://docs.ewi.info/JTA TA Program.pdf.
8. Postol 2009, op. cit., 2
9. Postol 2009, op. cit., 3
10. S. Efron, “North Korea Working on Missile Accuracy.” Los Angeles Times, 12
September 2003, http://articles.latimes.com/2003/sep/12/world/fg-norkor12.
11. H. Lewis, “Origins of the Musudan IRBM.” The Arms Control Wonk, 11 June 2012,
http://lewis.armscontrolwonk.com/archive/5337/origins-of-the-musudan-irbm.
12. Postol 2009, op. cit., 8.
13. D. Lennox, ed., Jane’s Strategic Weapon Systems, Issue 44, Surrey, UK: Jane’s
Information Group, 51 (2006): 260–263.
14. D. Pinkston, The North Korean Ballistic Missile Program, Strategic Studies In-
stitute, (2008), http://www.strategicstudiesinstitute.army.mil/pubs/display.cfm?pubID=
842, 17–20.
15. M. Schiller and R. Schmucker, “Explaining the Musudan: New Insights on
the North Korean SS-N-6 Technology” The Arms Control Wonk, 31 May 2012,
 North Korean KN-08 ICBM 235
http: // lewis . armscontrolwonk . com / files / 2012 / 05 / Explaining the Musudan Schiller
Schmucker v1.2.pdf.
16. D. Wright, “SS-N-6 Missiles: How Many Could Have Been Transferred?” All
Things Nuclear, 8 September 2010, http://allthingsnuclear.org/ss-n-6-missiles-how-
many-could-have-been-transferred.
17. D. Wright, “South Korea’s Analysis of North Korea’s Rocket Debris.” All Things
Nuclear, 25 January 2013, http://allthingsnuclear.org/south-koreas-analysis-of-north-
koreas-rocket-debris/.
18. D. Wright, Personal email to the author, 2 May 2012.
19. Postol 2009, op. cit., 14.
20. N. Brugge, “Some Identified Engines from Isayev’s Design Bureau” Last
modified 21 December 2012, http://www.b14643.de/Spacerockets 1/Diverse/KB-
Isayev engines/index.htm.
21. D. Wright, trans, “North Korean Long-Range Missile Debris Survey,” ROK Min-
istry of Defense (2013), http://www.ucsusa.org/assets/documents/nwgs/SK-report-on-
NK-rocket-debris-analysis-translation-1-18-13.pdf.
22. M. Hobbs, Basics of Missile Guidance and Space Techniques, (New York: J.F.
Rider, 1959), 91–115.
23. P. Podvig, Russian Strategic Nuclear Forces, (Cambridge: MIT Press, 2004),
319–322.
24. Postol, 2009, op. cit., 38.
25. Podvig, 2004, op. cit., 319–320.
26. Brugge, Norbert, “The old Soviet SLBM ‘R-27’ and the re-use of the technology
for the North Korean and Iranian rocket development.” Last modified 26 Jan 2013,
http://www.b14643.de/Spacerockets 1/Diverse/R-27/index.htm.
27. Vick, Charles, “Nodong A, Ghauri II, & Shahab-3, Technical Data,” (2007),
Accessed 25 February 2013, http://www.globalsecurity.org/wmd/world/dprk/nd-a-
specs.htm.
28. Podvig, 2004, op. cit., 320–321.
29. S. Leslie, The Cold War and American Science: The Military-Industrial-Academic
Complex at MIT and Stanford, (New York: Columbia University Press, 1993), 92–93.
30. I. MacConochie and P. Klich, Techniques for the Determination of Mass Properties
of Earth to Orbit Transportation Systems, (1978), NASA Technical Memorandum 78661.
31. C. Hansen, Swords of Armageddon, Version 2: History of the US Development of
Nuclear Weapons, (Sunnyvale, CA: Chuckulea Productions, 2007), Volume I.
32. R. Norris et al. , Nuclear Weapons Databook, Vol. V: British, French and Chinese
Nuclear Weapons, (Boulder, CO: Westview Press, 1994).
33. Podvig, 2004, op. cit., 321.
34. J. Lewis, “Making Yield Estimates” The Arms Control Wonk, 15 Febru-
ary, http:// 32ad through (did title page, headers, refs, table/fig headerslewis.
armscontrolwonk.com/archive/6288/making-yield-estimates.
35. Hansen, 2007, op. cit., Volume V.
36. Grimwood, James and Strowd, Francis, History of the Jupiter Missile System,
(U.S. Army Ordnance Missile Command, 1962), 37.
236 Schilling

37. Podvig, 2004, op. cit., 194.

38. Norris et al. , op. cit., 1994.
39. W. Minnick, “Despite Evidence, China Denies Selling Launcher to N. Korea” Wen-
dell Minnick - Articles, 17 October 2012, http://minnickarticles.blogspot.com/2012/10/
despite-evidence-china-denies-selling.html.
40. Monterey Institute for International Studies, “North Korea Missile Chronology.”
Accessed 25 February 2012, http://www.nti.org/media/pdfs/north korea missile 2.pdf?
=1327534760.
41. Gerardi & Joseph, 1995
42. Schiller and Schmucker, op. cit.