International Network of Engineers and Scientists Against Proliferation

An Analysis of the

North Korean Missile Launch of 31 August 1998

David Wright

North Korea launched a missile around noon local time (0300 GMT) on August 31, 1998. It was originally believed to be a two-stage Taepodong-1 missile, which North Korea was known to be developing. The test caused alarm in Japan since it overflew the northern tip of the main Japanese island of Honshu. On September 4, North Korea announced that the missile actually had three stages and had placed a small satellite into orbit. After ten days of silence, US intelligence announced that it had determined that the missile did indeed have a third stage, but that it had failed to place a satellite in orbit.

In this article, I briefly describe what information is available about the launch and what the implications are.

The launch is reported to have taken place from the Hwadaegun Missile Test Facility, Musundanri, Hwadae County, North Hamgyong Province, about 100 km south of the city of Ch'ongjin.

The missile appears to use North Korea's 1,000-1,300 km-range Nodong missile as its first stage and modified Scud as the second stage. Both use liquid fuel. Adding a second stage is the obvious next step in missile development once North Korea had a working Nodong. Modeling of this missile gives a range of 1,500-2,000 km with a one-tonne payload, depending on details like how heavy the missile casing is, what fuel it uses, etc.1

Instead of a warhead, the missile carried a small third-stage booster and a small satellite. The third-stage booster was reportedly solid fueled. The mass of the satellite is probably a couple of tens of kilograms. An expert from the Russian Space Agency reportedly stated that the satellite mass was 13 kg.2 The mass of the first US satellite, Explorer 1, was 14 kg.

Press reports state that the first stage of the missile landed in the Sea of Japan, 400-500 km from North Korea and 300 km southeast of Vladivostok. The rest of the missile apparently passed over the northern tip of the main Japanese island of Honshu. The second stage splashed down in the Pacific Ocean at a range of 1,400 to 1,650 km from the launch site, according to various reports.3

North Korea claimed the third stage burned for 27 seconds and placed a satellite in an orbit with a perogee of 218.82 km, an apogee of 6978.2 km, and a period of 165 minutes and 6 seconds. It said that the satellite, called the Bright Star 1 (Kwangmyongsong 1), was broadcasting patriotic songs at a frequency of 27 megahertz.4 The stated launch azimuth of 86 degrees would correspond to an orbital inclination of about 41 degrees.5

From the photograph of the missile launched released by the Korean Central News Agency, the payload section appears to be roughly 4 meters long and 1 meter in diameter, based on the dimensions of the other stages. This gives a rough estimate of the size of the third stage. In addition, assuming the first two stages were a Nodong and a Scud, from the range at which the second stage is reported to have traveled one can estimate that the mass of the third stage must be in the range of a tonne.

US Space Command has been unable to detect a satellite, and it is generally believed now that the attempt failed.



However, the launch showed that North Korea has now successfully demonstrated staging technology, since the staging of the first two stages appears to have worked successfully. This is not particularly surprising since staging can be done in ways that is not technically demanding, although it increases the complexity of the missile and may reduce the reliability. The photograph of the launch appears to show a space between the first and second stages, which suggests that North Korea used a similar technology to Chinese and early Russian missiles, in which the upper stage engine is ignited while the lower stage is still burning.

The demonstration of staging is important because it is a necessary step for North Korea to develop longer-range missiles. In addition, one clear, verifiable possibility for implementing a limit on North Korea's missile development was to ban the flight testing of multi-stage missiles before North Korea demonstrated staging capability. The August test removed that as a possibility.

The launch also raises a number of questions, which we address below.

Is it credible that North Korea could have put a satellite in orbit?

Putting a satellite into a circular, low-earth orbit requires reaching a speed of 8 km/s. The highly eccentric orbit that North Korea claimed it achieved would require accelerating the satellite to 9 km/s. By firing the missile nearly due east over Japan, North Korea gained roughly 0.3 km/s from the rotation of the earth. From the range at which the missile's second stage fell into the Pacific Ocean, one can estimate that the missile's speed was about 3.5 km/s when it released the third stage. Thus, the increase in speed required from the third stage in order to reach orbit would have been roughly 4.2 km/s for low-earth orbit and 5.2 km/s for the orbit claimed by North Korea.

Knowing this, one can use the rocket equation to calculate what combination of specific impulse and fuel fraction (propellant mass divided by total mass) would be required for the third stage plus payload in order to achieve these increases in speed (see Table 1). If the actual system has a fuel fraction less than that in Table 1 for a given specific impulse, it will not be able to achieve the required velocity change.


Velocity change


Specific impulse

(sec) (vacuum)

Fuel fraction

(stage + satellite)






  300 0.76


240 0.89
  270 0.86
  300 0.83

Table 1. Specific impulse and fuel fraction required to achieve a given velocity change.

One can then look at the parameters for solid-fuel upper stages for existing space launch vehicles (considering those with a mass of roughly a tonne or less, a length of roughly 4 meters or less, and a diameter of roughly a meter or less) and ask what the corresponding figures are, assuming a satellite mass of 15 kg (Table 2).



Specific impulse

(s) (in vacuum)

Fuel fraction

(stage + satellite)

US Taurus 3rd Stage



US Scout 3rd stage



US Scout 4th stage



Chinese CZ-1D 3rd stage



Indian SLV-3 3rd stage



Indian SLV-3 4th stage



Table 2. Specific impulse and fuel fraction for some existing small solid engines.

Comparing the numbers in the two tables shows that a third stage with the capabilities of the US Taurus or Scout engines would be able to place a small satellite in low-earth orbit, but only the Taurus engine (which was developed in the 1990s) would be able to place a satellite in the orbit claimed by North Korea. Presumably North Korea did not have access to an engine of this sophistication. The other motors shown in the table would not be able to place a satellite into either orbit.

Thus, unless the estimates above of the burnout speed of the lower two stages of the Taepodong 1 are significantly low, it appears that North Korea would not have been able to put a satellite into the highly-eccentric orbit it claimed, and very likely could not have even achieved a circular low-earth orbit.

It is worth noting that there are other reasons to suspect the credibility of the North Korean claims about the launch. For example, the period of the satellite reported by KCNA during its 100th orbit was 8 minutes longer than the original period.7 However, as satellite orbits decay, their periods decrease rather than increase.

Why did the launch fail?

How the missile failed is unclear from press reports. Several press reports quote officials saying that the third stage broke up somewhere over the Pacific Ocean,8 and one states that it broke up "several seconds before reaching orbit."9 Other reports have stated that debris from the missile fell in the ocean some 6000 km from North Korea, although the Pentagon has stated that it cannot confirm those reports.10 A Pentagon spokesperson stated that "the only way we have to trace the debris of this missile is through radar tapes, and there is considerable disagreement throughout our own intelligence community as to how to interpret those tapes."11

It is perhaps interesting that the ground track of the first orbit that the North Korean satellite would have had (apparently calculated based on the information released by North Korea) shows that the track passes near Hawaii but not Alaska.12 Even if the satellite did not reach orbit, it might still have followed the early portion of this track. A malfunction of the third stage could, of course, have steered it off this track.

Another possibility for the failure of the third stage is that it simply did not ignite. This possibility is interesting because it could explain why the United States originally thought the missile had only two-stages, as discussed below.

Why did the United States not initially realize the test was three stages?

The reports that debris was carried far into the Pacific Ocean, if true, would imply that the third stage burned for some time after the second stage burned out, which should have been observable to US intelligence. After the North Korean announcement of its launch attempt, a US official was reported as stating that the United States detected something separating from the third stage that appeared "to have some thrust behind it."13 That same official reportedly said "If there were a device and it was very small, it wouldn't be easily detectable." However, while the satellite may have been small, the third stage may have been 3 to 4 meters long and perhaps a meter in diameter, as estimated above, which should have been easily observable.

The United States presumably got a very good look at the missile launch. US intelligence detected launch preparations two weeks in advance and reportedly monitored the launch with two spy planes and a reconnaissance ship, the "Observation Island," which has Cobra Judy radars on board. Why the intelligence community appears to have initially missed the third stage is therefore puzzling.

As noted above, one possible explanation, which is supported by some off-the-record conversations, is that the third stage failed to ignite. Intelligence assets monitoring the flight may have seen an object fall into the ocean from the second stage and assumed it was a dummy warhead. Subsequent US announcements that the launch had been an attempt to launch a satellite may have come from recognizing the unusual flight path of the missile during boost phase.

What does the launch attempt imply about North Korean missile capability?

With the launch, North Korea successfully demonstrated staging of the first two stages. This is probably the most important aspect of the test, since this technology could be used in a more powerful missile.

The test also appeared to show that North Korea has some access to the technology for solid-fuel engines, although it is unclear what the basis is of the claim that the stage uses solid fuel. The source of this technology is not known. North Korea may have begun an effort to develop small solid-fuel rocket motors several years ago and the engine used may have been built in North Korea. It may have received assistance in this effort, or may have purchased an engine, from several sources. Iran is believed to be working on solid propellant stages, reportedly based on assistance from China, and it is possible it could have assisted North Korean development or transferred relevant technology. Some reports state that Iranian observers were present for the August missile test. North Korea is believed to have transferred Nodong technology to Pakistan for its Ghauri missile test earlier this year, and it is conceivable that it could have received some technology in return. Pakistan produced small solid-fueled rocket motors for its Hatf 1 and 2 missiles in the late 1980s, and these appear to have roughly the right size and mass for the August launch, given the above estimates. It could have supplied technical assistance or possibly motors to North Korea.14

If it is able to correct the problems with the third stage, North Korea would likely be able to deliver small payloads of several tens of kilograms to intercontinental distances. If the missile could achieve the speeds needed to put a satellite into orbit, it could in principle reach any point on earth with a payload of that mass.

One can also estimate the ability of such a missile to carry heavier payloads. To do this, I developed a computer model of a missile that was able to boost a small satellite to the speed required for orbit, and then calculated what range that missile would have if the payload were increased to 500 and 1000 kg, which is the range of masses one would expect for a first generation nuclear weapon. I looked at the same two cases considered above: missile burnout speeds of 8 and 9 km/s (taking into account the rotation of the earth).

These calculations show that for the 8 km/s case, the missile would have a range of roughly 4100 km for 500 kg, and roughly 2500 km for 1000 kg. For the 9 km/s case, the missile would have a range of roughly 4600 km for 500 kg, and roughly 2600 km for 1000 kg. (As noted above, it seems unlikely that the missile North Korea fired would have been able to have achieved 9 km/s.)

This missile configuration, if it becomes operational, would therefore appear to be able to deliver a small payload of chemical or biological weapons to intercontinental range, but could deliver a nuclear warhead only to much shorter ranges.


Photo of the supposed North-Korean satellite (source:



1 Such a model was discussed in David Wright and Timur Kadyshev, An Analysis of the North Korean Nodong Missile, Science and Global Security, Vol. 4, 1994, pp. 129-160.

2 N. Korea's launch may have been a satellite-Russian official, Associated Press (Moscow), 9/15/98.

3 J. Anselmo, R. Wall, E. Sekigawa, Missile Test Extends North Korea's Reach, Aviation Week and Space Technology, 9/7/98, p. 56-7; Korean Central News Agency, Successful launch of first satellite in DPRK, 9/4/98 (available at in the archives).

4 Korean Central News Agency, Successful launch.

5 David Fulghum, US Doubts Korean Space Launch Claim, Aviation Week and Space Technology, 9/14/98, p. 58.

6 Steven Isakowitz, International Reference Guide to Space Launch Systems (Second Edition), (Washington, DC: American Institute of Aeronautics and Astronautics, 1994).

7 Korean Central News Agency, Artificial satellite `Kwangmyongsong No. 1' makes 100 rounds of the earth, 9/14/98 (available at in the archives).

8 See, for example, Steve Lee Meyers, US Call North Korean Rocket a Failed Satellite, New York Times, 9/15/98, p. A6.

9 Bill Gertz, N. Korean missile seen posing risk to US, Washinton Times, 9/16/98, p. 1.

10 Gertz, N. Korean missile seen posing risk to US; Pyongyang Rocket `Crashed Off Alaska Coast', South China Morning Post, 9/17/98, p. 1; US is unable to confirm if North Korea rocket hit Alaskan Seas, Chosun Ilbo (Seoul), 9/21/98.

11 Kenneth Bacon, Regular Pentagon Briefing, Sept. 17, 1998, US Information Agency Transcript.

12 Sven Grahn, Ground track of North Korean satellite launch-if there was one!, www.users/, 9/98.

13 Dana Priest, N. Korea May Have Launched Satellite, Washington Post, 9/5/98, p. A21

14 Some reports state that the solid-fuel Chinese M11 missile has two stages, although many reports dispute this. If true, the upper stage might be roughly the right size for the North Korean missile. Pakistan is believed to have some tens of unassembled M11 missiles that it purchased from China.

David Wright is a Senior Staff Scientist at the Union of Concerned Scientists and a visiting scholar at the MIT Security Studies Program. Address: UCS, Two Brattle Square, Cambridge, MA 02238, USA; tel +1-617-547-5552, fax -864-9405, email