Orbital Focus - International Spaceflight Facts and Figures
carousel image
North Korea

Tyneside, UK
2024 Jun 15
Saturday, Day 167

Curated by:


Common Threads

Mission 5

Mission 4

Mission 3

Rocket Trajectories

Kwangmyongsong 4 - Not So Dead?

KMS-4One way of monitoring activity of Kwangmyongsong 4 is to record the way in which the orbit changes or to look at the way rates of change vary.

This page looks at variations in the way that the orbit responds to air drag on the satellite, present even the 500 km altitude at which it orbits the Earth.

It also looks at the way variations have shown up during the first few weeks on orbit and asks what clues they give as to what is happening overhead.


In the published Twoline Orbital Element Sets is a value called Mean Motion - an alternative way of measuring the orbital period. It is expressed as 'Revolutions per Day' (Ndot). Over the lifetime of a satellite in an orbit similar to that of Kwangmyongsong 4, the value of Mean Motion rises because the satellite completes more circuits of the Earth per day as the orbit height and period reduce through the effect of air drag. The rate at which Mean Motion changes, expressed as 'Revolutions per day per day', is Ndot2.

Ndot2 Values for Kwangmyongsong 4

This plot shows the changing value of Ndot2 as successive Element Sets were released by SpaceTrack after launch. The very early sets are not included because thay have large inconsistencies including some negative values. It is quite normal for the first few hours or days after a launch. JSpOC's sensors take time to get a consistent 'lock' on the new object and sort out observations between satellite, rocket body and other items such as parts of the satellite release mechanism.

In the case of Kwangmyongsong 4, only the satellite itself and the rocket body had been catalogued up to the end of February. Its predecessor, Kwangmyongsong 3, was accompanied by two debris objects that might have been covers for small rocket motors used to push the Unha 3 upper stage down into a lower orbit.


The red points on the plot are raw values of Ndot2 taken directly from the Twoline Orbital Element Sets. They seem to break down into four distinct groups. Initially, there is a near straight, downward sloping line with a degree of scatter that is probably due to the JSpOC sensor 'early days' syndrome. It starts at about 0.00008 and drifts down towards 0.00006. Then, February 13-14, there is a sudden increase to 0.00010 before it settles down to the downward-sloping trend again.

On February 17, the value starts to rise and looks as though it goes into a cyclic trend whilst still reducing in overall value with time.

The blue line on the plot is a Solar Weather measurement. It is the intensity of the Sun's radio emission at 10.7 cm wavelength. There is quite a good correlation between the 10.7 cm radiation and the density of the Earth's upper atmosphere. If the 10.7 cm flux increases then, about four days later, satellites start to experience higher values of air drag.... and things go the other way if the flux decreases. In the plot, radiation readings have been shifted by four days to take the lag into account.

The 10.7 cm flux was falling slowly in the period on view so it had the effect of gradually reducing the actual drag experienced by Kwangmyongsong 4. It is the reason for the general drift downwards in the value of Ndot2 across the plot.

Looking From Another Angle

Because the 10.7 cm emission data are available, it is possible to build in a compensation factor to deal with changes in the upper atmosphere. In the plot below, the JSpOC values of Ndot2 have been corrected to show what they would have been if he 10.7 cm flux had remained constant at the value it held on February 7. It produces a slightly different picture.

ndot2 corrected

The two downward-descending groups of points are now horizontal, the sudden and rapid change is very obvious and so is the cyclic section to the right. A similar analysis of Kwangmyongsong 3's first few weeks in orbit shows nothing resembling the pattern here. It shows a near-horizontal line indicating that the satellite reached orbit and was inert once it got there.

Before looking at an analysis and trying to draw some conclusions from the plot, it is worthwhile re-capping the news items that seemed to emerge during the few days following launch.

US Statements

At various times following the launches of both Kwangmyong 3 and Kwangmyongsong 4, un-named US Government spokesmen gave out information on mission progress.

Kwangmyonsong 4 was launched February 7. Within hours the US Government announced that it was tumbling, it was unstable, out of control and was dead.

On February 8, the same or another spokesman announced "They've got the tumbling under control", only to be contradicted by another official comment a few hours later to say it was tumbling again. Since then the US has been quiet on the matter. No examples were ever given of how the satellite was tumbling, or how fast it was rotating so there is no information of value here at all.

In real life, many satellites develop a rotation after leaving the rocket upper stage. It is often by design. An onboard stabilisation mechanism then takes over to eliminate the spin. It can be the use of thrusters, though this is not the case with Kwangmyongsong 4. Alternatively there are methods using a combination of a gravity gradient stabilisation boom and magnets that align the satellite with the Earth's magnetic field, or there could be gyros. The jury seems to be out on what equipment Kwangmyongsong 4 actually carries.

There is quite a good ongoing narrative covering Kwangmyongsong 4 events, including some of the US comments, at the North Korea Tech website.

Russia Wades In

An incoming e-mail from James Oberg, February 22, contained a pointer to a story issued by Russian News Agency Itar-TASS the previous day. The item quoted a Russian military spokesman. He said that from the orbital data available, the satellite was intended for remote sensing - a reasonable conclusion from the fact the orbit is near-circular and close to being sun-synchronous.

He also said that the satellite had been "activated" but gave no further clue as to what it actually meant.

"The National Space Monitoring Center under the air and space forces of the Russian Defense Ministry confirmed that the DPRK's earth observation satellite Kwangmyongsong-4 is now normally operating.

"Andrei Kallyuta, chief of the center, said this in a program of Ruskaya Sluzhuba Novostei Broadcasting Service on Feb. 22.

"He added that he could draw a conclusion that Kwangmyongsong-4 is a remote controlled earth observation satellite on the basis of the information about its orbit such as the analysis of revolution period, angle of inclination at which it was put into its orbit and altitude and in other words it is capable of performing its observation function and its devices are working."

KCNA made no comment of its own, neither confirming nor denying the "...its devices are working" statement.

So - What Has Happened?

It is possible to look at the compensated/corrected plot and draw some tentative conclusions as to what the distribution of readings is telling us. This is one view and there will be others with equal validity. However, the overall conclusion stays the same - there was change in the way Kwangmyongsong 4 was oriented in space and it probably managed to get rid of any significant spinning or tumbling.

ndot2 corrected with comments

Initally the satellite was probably rotating slowly (a number of seconds per revolution) rather than tumbling rapidly, or even spinning as some would have seen it. There is a lot of scatter in the Ndot2 values but the trend is to stay roughly constant, just below 0.00008. it implies that Kwangmyongsong's average cross sectional area stayed constant. Around February 13 there is a relatively fast transition to 0.00010 where it stays constant for a while.

The implication of the change is that air drag increased. Air drag is determined by the cross sectional area presented towards the direction of travel. Kwangmyongsong is a cuboid about 1.3 times as long as it is wide. It will experience a little more drag if one of the larger surfaces is facing the forward direction than it will if one of the square-shaped ends is leading. With a tumble, drag will appear to be constant based on an average of the large and small faces of the shape as they are presented alternately to the oncoming air.

KMS-4 solar panelsThe February 13 change could be related to deployment of the solar panel. At launch, Kwangmyonsong 4 had solar panels covering three of the longer sides. Two of them were hinged so that they could swing out either side of the third to form a single flat surface covered in solar cells. Opening the panel itself would have reduced the satellite's rotation speed through conservation of angular momentum. The average cross sectional area would have increased, leading to higher drag, hence the exhibited rise in value of Ndot2.

Kwangmyongsong 4's design is similar to Kwangmyongsong 3. There is a discussion of the solar panel deployment mechanism in Zarya.info's pages covering the earlier launch. The picture here depicts the fully-opened panel on Kwangmyongsong 4.

The change around February 17 is distinct and suggests that the stabilisation mechanism kicked in. The two possibilities are that it extended a gravity-gradient stabilisation boom or that gyros were spun-up. Whichever, it seems to have settled into a relatively stable regime with a fluctuation that has a period of 7-8 days. The fluctuation seems to be damped as the second peak is significantly less than the first. It may eventiually settle down to a constant value when all rotation is damped out.

It may explain the Russian comment on February 22. Russia has its own tracking network that produces observations similar to those that feed the Twoline Orbital Elements. Its analysts will have seen this pattern and could justifiably have concluded that Kwangmyongsong 4 had been "activated".

In a separate development, Dr Marco Langbroek, an amateur observer of satellites in the Netherlands, saw Kwangmyongsong 4 in the night sky on February 28. His photograph of the satellite trail and his comments confirmed that it was not tumbling, or at least not tumbling at a rate fast enough to show up as flashes or short term variations in brightness.

Is Kwangmyongsong 4 Alive?

That is a big question. So far, there is no evidence that it is communicating with Earth stations. There are no positive reports from amateur trackers and North Korea has said nothing.

But it does seem to have performed some activities on orbit.

It could have acted autonomously with events triggered by the solar panel electrical output reaching a certain level. Alternatively, it might have responded to anonboard timer or to a radio command from the ground. We may yet see the start of transmissions but, at the time of writing, nearly four weeks have elapsed since launch.

What Next?

As usual, it's 'wait and see'. If the satellite is alive, North Korea will eventually tell us. Alternatively, the silence will continue. There may also be an alternative explanation of what is presented here. Whatever, only time will tell. In the meantime, the plot and notes below will be updated from time to time.

ndot2 current

added March 10 - the air drag remains remains relatively steady with an oscillation showing a period of about 1 week. It could be down to a very slowly changing attitude resulting in a variation of the cross section presented along the direction of travel, implying a period of about 14 days.

added March 12 - a sudden increase in the decay rate March 8/9 peaks on March 10 and then starts to decline. This is not natural and suggests the vehicle is changing its attitude under control.

added March 17 - as of March 12 Ndot2 is a near-straight line that could actually be horizontal given the uncertainty in the estimate for air drag.

Page Date: 2016 Mar 1
Updated: 2016 Mar 2 (KCNA & Radio)
Updated: 2016 Mar 10 (ongoing plot added)

Copyright © Robert Christy, all rights reserved
Reproduction of this web page or any of its content without permission from the website owner is prohibited