Progress in Electric Propulsion

There are three broad categories of electric propulsion for communications satellites, according to the mechanism transferring electric power to kinetic energy: electrothermal, electromagnetic and electrostatic. Electrothermal propulsion includes resistojets and arcjets (performance of the chemical propellant is augmented by electrical heating) and is used on operational satellites. Rf or microwave heated thrusters are in the research stage. Electromagnetic propulsion includes pulsed plasma thrusters (PPT) using Teflon as propellant (low thrust, LEO orbit and attitude trimming). Electrostatic propulsion includes stationary plasma thrusters (SPT) and ion thrusters (rf and electron bombardment). Some characteristics and development status are listed in Table 3.7.

Table 3.7
Categories of Electric Propulsion




Specific Impulse

Specific Impulse

Development Status


m/s (metric units)

s (U.S. units)



100 - 200

5,000 - 6,000

500 - 600



PPT (Teflon)

< 1






50 - 200

16,000 - 18,000

1,600 - 1,800



Ion engine

10 - 120

25,000 - 30,000

2,550 - 3,000


Despite a long history of development (NASA flew an ion thruster on ATS-6) and extensive use of stationary plasma thrusters (using the Hall effect) on satellites in the former Soviet union, electric propulsion had not yet found widespread adoption in commercial communications satellites at the time of the previous report (1992/1993).

In contrast, the last few years have seen a substantial change in the perception of electrical propulsion as several manufacturers have adopted some form of electrical propulsion system for north - south station keeping (NSSK) in GEO satellites and are considering it seriously for LEO application to raise the orbit after launch. Resistojets are used on the Iridium satellites, Arcjets have been used (Lockheed Martin satellites) and more recently the first operational ion propulsion subsystem is flying on a commercial satellite (Hughes HS 601 HP satellite). In addition, there is work in electric propulsion which is not specifically aimed at commercial communications satellites: a 26 kW arcjet and Hall thrusters (4.5 kW and 10 kW).

In today's GEO satellites, with bipropellant systems for apogee insertion and station keeping, the fuel amounts to about half the total mass in GTO. Electric propulsion can reduce the propellant mass needed for station keeping substantially in exchange for significant use of electrical power (the spacecraft battery may have to be used for several hours per day).

Another attractive application of electric propulsion is orbit raising for LEO and possibly GEO satellites. As the time from LEO to GEO may be substantial (several months) an operator may not want to wait that long without collecting revenue. On the other hand, electric propulsion is an attractive alternative for raising a LEO orbit as only a few weeks are necessary and the satellite can be used during this time. Finally, electric propulsion may be used for the final deorbiting of obsolete satellites.

Table 3.8 gives a summary of where electric propulsion may be used advantageously for communications satellites.

Table 3.8
Applications for Electric Propulsion


[delta]v needed


Satellite Mass





Power needed


GEO station keeping


2,000 - 4,000

20 - 200

2 - 6 h/day

1.2 - 2.5

HEO orbit control


2,000 - 3,000

10 - 250

3 - 6 h/day

2 - 2.5


5k - 15k

5,000 - 8,000

100 - 2,000

3 - 12 months

3 - 5

LEO/LEO orbit raising


300 - 1,500

10 - 100


.5 - 1

Recent Developments

Hughes has continued development of ion thrusters using xenon as propellant (the program started under NASA and INTELSAT sponsorship) and is life testing two models called XIPS (two models developed, 13 cm diameter and 25 cm diameter with approximately 110 mN thrust, a specific impulse of 2,675 s and 2.35 kW power consumption). Hughes is using ion thrusters on commercial satellites launched in the past year. The units will be used for station keeping and also during orbit transfer.

In the last few years electric propulsion technology developed in Russia has become attractive to Western commercial companies. Although many devices using this technology have extensive flight history in the former Soviet Union, additional work is necessary to make the hardware suitable for use on commercial satellites. Multinational efforts are under way to make use of the technology and produce hardware qualified to Western standards. One example is a joint venture, International Space Technology, Inc. (ISTI) (founded by SS/L, Fakel and RIAME and including SEP and Atlantic Research) that has integrated Russian and U.S. components into an electric propulsion subsystem and performed a complete qualification program. The subsystem consists of a Russian 100 mm stationary plasma thruster (SPT-100) and xenon flow controller, a U.S. power processing unit, tank and propellant management assembly. At least two satellite manufacturers have expressed interest in using such a subsystem. Currently, a disadvantage of this type of thruster is the wide divergence of the exhaust jet. Research in Russia (sponsored by SEP) is addressing this problem.

Updating the information in the 1992/1993 report on Japanese and European developments, the two European ion thrusters (developed by DASA and MMS) are scheduled to fly on the ARTEMIS satellite in 2000.

In Japan, ion engines developed by Toshiba have flown on the ETS-VI satellite. Despite the problems with orbit injection of this satellite, the units were tested in orbit. The same electric propulsion subsystem will also be used on COMETS.

In conclusion, much basic research on electric propulsion for communications satellites has been completed and the advantages are clear. What remains to be accomplished is to establish a solid track record of reliability in orbit. As this depends on many engineering and design details, as well as on parts reliability, further design iterations will be necessary in addition to more research and testing, to understand any potential life limiting factors. Nevertheless, it is expected that electric propulsion will see much wider use in the future of commercial communications satellites.

Thermal Control

Most of the prime dc power on a communications satellite is used by the transmitters in the payload. In today's satellites, the dc to rf efficiency for SSPAs is around 35% and for TWTAs around 50 to 60%. Therefore a substantial part of the dc power is dissipated and must be removed from the spacecraft. In a modern three axis stabilized GEO spacecraft (shaped approximately as a cube) only the north and south facing panels can be used to radiate heat to cold space (they are inclined with respect to the ecliptic at approximately 23 degrees and will receive some solar radiation); all other sides will be exposed directly to the sun at some time during the day, thus preventing effective heat rejection. The north and south facing panels carry usually a major part of the heat producing payload on the inside and the solar array on the outside. Heat pipes are used by most communications satellites to carry the heat from transmit amplifiers to radiating surfaces and equalize the temperature inside. A new, much lower cost, heat transport system to equalize temperatures between north and south panels has been developed in Europe. Conventional pipes connect two fluid loop exchangers (using a proprietary material) located on the north and south panels respectively. This design avoids bends in conventional grooved heat pipes.

As satellite prime power increases from 5 to 10 or more kW (see Figure 3.1), there is a consensus in the industry that thermal control is a major problem because the radiating surfaces are not increasing correspondingly in size. One solution envisaged by several manufacturers is deployable radiators. Further developments are needed to provide reliable solutions.

Another area of concern is thermal control in onboard processors. These have substantial dissipation; on the order of 500 to 1,500 W of heat must be carried away from a small box and radiated to space. Heat pipes can carry heat from one place to another, however, they are bulky. The main problem is to carry the heat from the semiconductors to the outside of the electronics enclosure.

The panel did not see any specific R&D work on thermal issues that would appear to provide long-term solutions to thermal control problems.

Attitude Control

In contrast to the 1992/1993 study, the panel did not see any specific R&D to advance the state of the art of attitude control systems in a major way. Conventional systems for three axis stabilized spacecraft with earth and Sun sensors, momentum wheels, jets and associated electronics are in production at many manufacturers all over the world. Star trackers (to improve pointing accuracy for large antennas with very narrow beams) are being developed in the United States and Japan, but are still far from operational use due to cost and operational complexity. There is engineering work on laser gyros in the United States, with improved versions eventually capable of replacing conventional gyros used for attitude sensing.

In the last five years, GPS receivers have been used on satellites to establish location and also attitude. The technology used for these receivers is conventional; only the application is novel.

Published: December 1998; WTEC Hyper-Librarian