The FREJA magnetospheric research satellite was launched on October 6, 1992 as a “piggyback” payload on a Long March 2C (CZ-2C) rocket from the Jiuquan Satellite Launch Center in China. The satellite was in an orbit between 601 and 1756 km at 63º inclination. FREJA was a sun-pointing spinner with a 2.2 m diameter and 214 kg mass. The Swedish Space Corporation was the Prime Contractor to the Swedish National Space Board. FREJA imaged the aurora and measured particles and fields in the upper ionosphere and lower magnetosphere.

Artist impression of FREJA in space

Swedish, German, Canadian and U.S. instruments were flown on the satellite (see table). The total cost of the FREJA program including satellite, launch, ground system and 2½ years of operation, but excl. experiments was $19 million (FY95). The satellite was in operation for about 1400 days.

Project origins

FREJA was initiated in 1987 when the CZ-2C launch reservation was “left over” from the cancelled Mailstar project, a store-and-forward low-orbit communications satellite. The launch reservation was the result of a competition for launching Mailstar, and not particularly adapted to an auroral mission. However, the 63o inclination provided by the CZ-2C was high enough to stimulate the interest of Swedish scientists into using it for an auroral research mission. The launch price was very attractive, making it feasible to build a satellite despite the acute lack of funds in the Swedish space budget. The FREJA project budget was less than half the cost of the low-cost Swedish VIKING project (not to be confused with the NASA Viking mission to Mars). The VIKING satellite [excl experiments] and launch cost $40 million. The FREJA and VIKING missions were quite comparable in scope and level of ambition. Even the launch costs were almost identical. Therefore FREJA represented a significant reduction in cost compared to VIKING. To achieve the cost goal the Swedish National Space Board and the Project Scientists agreed to let SSC use unconventional methods to cut costs. Whether or not the launch cost was low or high should be viewed against the fact that FREJA represented about 12% of the capability of the CZ-2C. The Chinese main satellite weighed about 1800 kg at launch and FREJA weighed 256 kg. So, FREJA was a companion payload requiring extensive modifications of the CZ-2C (See “Launching Arrangement”).

The financing of the development costs came from the Swedish taxpayers via the budget of the Swedish National Space Board, donations from Wallenberg Foundations, price reductions made by SSC and FFV Aerotech (the AIT contractor) and a generous contribution in cash, hardware and services from the German Ministry for Science and Technology amounted to approximately 25 % of the costs above. Canada contributed the tracking services of the Prince Albert tracking station. Experiment development costs were not included in the overall cost figure of $19 million, but they were covered by the Space Agencies of the respective country of origin of the Principal Investigators.

The last view of FREJA and crew before mounting of satellite

It is difficult to estimate experiment development costs, but they were estimated to amount to maybe 30 % of the satellite development costs. Operations planning and actual spacecraft operations for 30 months cost about $1 million. Thus, total project costs including everything was on the order of $23 million.


FREJA Experiment Summary

Nr	Experiment Name	Principal Investigator	Mass (kg)	Power (W)	Data (kbps)
F1	Electric Fields	Royal Inst. of Technology, Sweden	5.37	9.0	30.7
F2	Magnetic Fields	APL/Johns Hopkins Univ., USA	3.97	4.0	14.33
F3C	Cold Plasma	NRC, Canada	5.79	6.2	16.38
F3H	Particles; Hot Plasma	Swedish Institute of Space Physics, Kiruna	9.52	15.6	42.98
F4	Waves	Swedish Institute of Space Physics, Uppsala	8.40	11.6	42.98
F5	Auroral Imager	Univ. of Calgary, Canada	9.94	6.7	44.01
F6	Electron Beam	Max-Planck Inst., Garching, Germany
F7	Particle Correlator	Max-Planck Inst., Garching, Germany	8.49	13.0	26.61
	6 wire booms (l<15m) and 2 stiff (l=2m)		21.60	–	–
	Housekeeping		–	–	5.13
	Synchronization		–	–	6.15
	TOTAL		73.08	66.1	261.98

Project Schedule The FREJA project was started in August 1987 with feasibility studies concerning the use of the Long March 2C launch opportunity. The first half of 1988 was spent completing the system design, writing equipment specifications and collecting equipment bids. The protoflight satellite structure was completed in the fall of 1989. The structure qualification vibration test was performed in China in March 1990. The “System Unit” (see below) qualification model was delivered for integration on the satellite in February 1991. Satellite integration was finished in August 1991.

Launch of Long March 2C with FREJA and the main satellite

During the rest of 1991 the following tests were performed: System EMC, acoustic noise, satellite balancing, boom deployment and acceptance vibration tests. In the spring of 1992 the Solar Simulation Test, Magnetic survey and ground station compatibility tests were carried out. The satellite was shipped to China for launch in August 1992.

Launching Arrangement

The Long March 2C (CZ-2C), was a two-stage liquid fuel rocket with a lift-off mass of 191 tons and the lift-off thrust was 2747 kN. The rocket was 34 meters long and had a diameter of 3.35 meters. Both stages were propelled by N2O4 and UDMH.
For launching FREJA the CZ-2C was modified by the addition of a cylindrical transition bay (or “piggyback cabin”) between the Chinese FSW-1 satellite and the top of the second stage of the CZ-2C. The transition bay was split into two cylindrical sections. FREJA had an interface ring which was clamped between the outer rims of these two cylinders. 0 showed this arrangement. The main satellite was released first while the CZ-2C stage was under three-axes control. After a manoeuver to put the separation vector parallel to the orbital tangent at the orbital apexes, the top half of the “piggyback cabin” was separated by firing four explosive bolts. Then FREJA gently moved away from the CZ-2C stage which was braked by solid rockets. A few seconds after separation solid propellant spin rockets fired to give FREJA a 50 rpm spin in preparation for ignition of orbit adjust motors.

FREJA was put into a 213.7-317.4 km parking orbit at 63º inclination. This orbit was raised to reach scientifically interesting regions and to avoid drag decay. A Thiokol STAR 13 A fired at the southern apex of the parking orbit, 37 minutes after launch, to give FREJA an apogee of 1756 km. 53 minutes later a STAR 6B motor fired to raise perigee to 601 km. The satellite will decay in 250 years.

The unusual mechanical interface to the launch vehicle and the need to locate wire boom assemblies and scientific instruments on the largest possible radial distance from the spin axis made it impossible to locate solar panels on the outside perimeter of the satellite. Instead, the top surface was used to locate the solar panels. This mounting of the solar panels left the back of the panels free so that heat can be radiated. A sun-pointing attitude was selected to give maximum electrical power (130 W End-Of-Life) and acceptable, albeit not optimum, viewing angles for the scientific sensors. Despite these limitations, instrument viewing requirements were satisfied, but sometimes days or weeks of waiting for the correct attitude was needed. To have made a completely arbitrary orientation possible the power required by the science instruments would probably have had to be cut roughly in half. The spin vector was kept within 30º of the direction to the sun. Spin-stabilization was needed to keep the so-called wire booms for measuring electric fields extended. Also, it is suitable for a satellite making a two-impulse Hohmann transfer.

The two rocket motors were placed along the spin axis with nozzles pointing in opposite directions. The spin vector attitude was controlled by the well-proven quarter-orbit magnetic torquing method. This was used for FREJA to make the spin axis to approximately track the sun. The scientists planned the attitude maneuvers to provide appropriate pointing of particle sensors and imagers. The spin axis needed to be controlled to better than 5º, preferably, say, 2º. These were approximate requirements and they were mostly met during on-orbit operations. A three-axis magnetometer, redundant solar aspect sensors and an IR earth sensor were used for attitude determination.

Big MPEG movie of the launch

Attitude determination was done on the ground from sensor collected at the optimum location along the orbit and stored in an onboard memory. An attitude vector was determined at least once a day. The a posteriori attitude determination accuracy requirement was 0.5º (1σ). However, because of the long wire booms the satellite was subject to strong gravity-gradient torques, so the spin vector drifted a few degrees per day. Therefore, interpolation between consecutive attitude vectors was required.

Redundancy was used whenever possible. FREJA had redundant transmitters, receivers, batteries, solar array shunts, telemetry systems, tele-command decoders, computers, pyro circuits and attitude sensors and actuators.

For information concerning this page, please contact: Sven Grahn.