M Nagatomo*
K Itoh**

The SPS Reference System was designed for the SPS Concept Development and Evaluation Program (CDEP) of the U.S.A., to study the wide range of problems anticipated when such a system is introduced as a national power system of the worlds largest industrialized country. As a result, many understandings have been obtained, and uncertain issues needing to be studied further have been indicated. The study concluded that the system was technically feasible and further study is required (1, 2).

The ISAS Solar Power Satellite Working Group is interested in a much smaller SPS as a strawman model study, which will be introduced and discussed in this paper. First, we will explain briefly about the working group as the background of the strawman model study,

ISAS SPS Working Group

ISAS, which is responsible for implementation of the Japanese space science program, provides researchers of universities and government agencies with convenience of nationwide research activities. A working group of the organization usually functions as a group to develop a concept of and conduct definition studies on a space mission. The purpose of the SPS working group established in 1987 was different from the other working groups, since it never intended to plan an SPS project for ISAS but, to remain as a research group to investigate the feasibility of "power from space" which was shown by the CDEP and the Reference system. The Scope of the SPS Working Group is described as follows:

"The solar power satellite was proposed (by Glaser) to solve future problems caused by activities of human beings on the global scale (3). The research areas of SPS are concerned with not only technology and engineering, but also big problems such as "large scale project", "global energy production" and "exploitation of extraterrestrial resources", as well as economical problems such as "large scale and long-term investment' and "risk analysis". The earth environment, which will be related to SPS, is already a societal issue of global scale.

Many research areas are beyond the capacity of the working group. The primary objective of the working group is not to plan an operational SPS, such as the DOE/NASA Reference System. As indicated by the US study, researches are most important at the present. In this respect, the members of the working group should find out a common interest in connection with the future of SPS, and make plans for space experiments along with a scenario of research and development for SPS. It should be emphasized that the development of high power technology for space use is not the main objective of the research".

The working group is divided .into thirteen subgroups by specialized research fields. Nine of them are concerned with studies on SPS subsystems and technologies. Other four subgroups are for studies on the effects of interaction of the SPS operation with the environment. The individual research fields of the thirteen subgroups are listed as follows:

Study on Subsystems and Technologies

Microwave transmission
Microwave reception
Large space structures
Guidance and control
Laser technology
Photovoltaic technology
Thermodynamic power generator
Propulsion
Space robotics

Study on Environmental Interaction Effects

Spacecraft environment
Space electromagnetic environment
Communication systems
Biology and ecology

It is noted that the working group is not a task force type of organization that should be functionally organized to accomplish a certain duty given from outside, but a group of researchers who are interested in various aspects of SPS. Thus, the subgroups have been voluntarily organized by researchers with common interests in each and are coordinated as a working group. One of the research areas is concerned with the strawman model concept.

The results of the CDEP give us a basic reference for interpolation of the growth of SPS from the present. The first problem the working group faced was that each member researcher could not find a definite image of his research related to the CDEP reference system. Then it was proposed to conceptualize what they would consider a more realistic SPS in terms of their research subjects. As a result of discussion, the size of the model should be as small as can be realized by the present space technology and space infrastructure, but should be large enough to be considered as an SPS.

A 10 MW model once studied by an industry study group was known and reexamined by the group. The original concept was substantially a smaller version of the Reference Model, which was built in a low earth orbit and transported to the geosynchronous orbit for operational tests. The mechnical configuration and power transmission method arei similar to those of the Reference System, which used solar panel and antenna structures connected by a gimbal mechanism with a slip ring (4).

In the guidelines for the new 10 MW model study, a low earth equatorial orbit is chosen for all the operations and the gimbal mechanism is eliminated. The primary reason to use the orbit is to make it possible to use launch systems available in or before 2000.

Consequently all the rectennas will be sited in developing countries located in the equatorial zone. This assumption implies the objective of this system is more than an experiment and could include evaluation of a pilot plant of SPS for power supply to the earth. Overall, the concept emphasizes demonstration of electric power supply to customers at the earliest opportunity rather than being designed first for research and development.

This model of SPS is intended to be realized as soon as possible. In this respect, most of the technologies employed are conventional or available in the near future. This means that the performance of the SPS is allowed to be relatively low, for example with high specific mass, low efficiencies of power conversion and transmission. The few exceptions are the significant amount of space assembly work with multiple launches, and high technology for beam control. The latter is concerned with the operational constraints due to low earth orbit that is the most crucial point of this system for final evaluation. Although the microwave frequency to transmit microwave power to a rectenna on the ground is the same as the Reference System, the beam has to sweep more widely to track ground rectennas while it passes over them.

Fig. 1. General configuration of SPS 2000

SPS 2000 system

There have been several general configurations proposed at the initial study phase. The following system designated as SPS 2000 is based on one which features the gravity gradient attitude stabilization of the entire system.

The general configuration of SPS 2000 has the shape of a triangular prism with a length of 800 m and sides of triangle of 100 m as shown by Fig. 1 The prism axis is in the north-south direction, perpendicular to the direction of orbital motion. The transmitting antenna is fixed to the bottom surface facing to the earth, and the other two side surfaces are used to deploy the solar arrays. No energy storage for power transmission during eclipse is provided.

The reason this configuration is adopted is easiness of construction of the system in orbit as well as of orbital operation under the gravity gradient force which is larger in a low earth orbit ( LEO) than in the geosynchronous earth orbit ( GEO). On the other hand, the requirements of high efficiencies in terms of power conversion and mass reduction, which have been generally accepted for space system design, are not considered to have high priority in this study.

A conventional rigid beam structure is assumed for the main structure of pyramid shape. The function of the structure is to deploy the solar array and the phased array antenna, and to provide the subsystems with references of geometrical locations. Since accuracy of shape is not required, the strength of the beam was determined considering the forces caused by off-center-of-gravity forces including gravity gradient torque and the attitude control force. A standard beam of the structure is 100m long and the cross section is triangular with sides of 3m. The fundamental element of the structure is aluminum pipe with an outer diameter of 12mm and 0.5mm wall thickness. The total mass of the structure is 4 tons.

The solar array consists of about 1500 rolls of 1m x 100m strips of solar panel, which will be stretched between the top longitudinal beam to the lower beams of the main structure. Conversion efficiency of the panel is assumed to be 14%. Then each strip of solar panel generates 1kV x 20A of DC electrical power at maximum. The panels are electrically connected to each other in parallel. The cable network of the system is used not only to conduct the electrical current to the transmitting antenna but also to serve as magnetic torquers for attitude control of the overall system. The power conducting cable is one of the heaviest elements of the system, especially in the case of the high power density antenna.

Transmission is possible only when a rectenna is in the field of view of the controllable microwave beam, which is assumed to be movable as much as 30 degree in any direction from the center position. Therefore, a rectenna located on the equator can receive power from a single satellite in a 1000 km equatorial orbit for 200 seconds in one orbit, and about 1600 seconds in a day. The relation of diameters of the transmitting antenna, Dt and rectenna, Dr and distance between them, d is given for transmitting frequency f (MHz) as:

f x Dt x Dr I d= 0.68 (length in km)

In this case where d is approximately 1100km, typical values of Dt and Dr are 100m and 3km respectively. If Dr is as large as that of the Reference System, Dt is only 30 m. One of the concepts of the antenna based on a current design (5) is shown by Fig. 2.

Fig. 2. A concept of transmitting antenna subarray for SPS 2000.

Reduction of the operating cost has been selected as the primary requirement of the guidance and control system design. From the standpoint of spacecraft design, easy orbit management and simple attitude and thermal control systems are desired to satisfy this requirement. To make this approach possible, the electrical power system should be operable over a wide range of conditions of the related factors. The standard orbit altitude is 1000km for this study, although it may be subject to change to 800 or 1200km to avoid the most crowded altitude of 1000km. The annual loss of altitude for these altitudes is several km, which will be acceptable for a decade or more period of operation without orbit maintenance. The gravity gradient force will be used to keep the system vertically in the orbital plane. Attitude keeping in the north-south direction can be achieved by the restoring magnetic force of the electric current loop in the geomagnetic field. Active control may be necessary if periodical forces caused by non-uniformity of the geomagnetic field could act resonantly on the structure.

Considering the complete assembly of SPS 2000 is much larger than the present launch systems, and the mass would be more than the payload of even the Energia launch vehicle, the SPS has to be divided into several flight units and deassembled to be packed in a payload envelope of a launch system. In the preliminary study, Arianne 5 launch vehicle has been assumed for the transportation, mainly because of its ease of access to an equatorial orbit. A launch capability of the vehicle into 1000km high altitude orbit has been predicted to be approximately 12 tons, based on an available standard data (6).

Table 1. Mass breakdown of integrated payload for the first flight unit of SPS 2000

According to the same data, the payload envelope is of cylindrical shape of 4.57m diameter and 12m length with a conical space on the top. An example of mass breakdown of a payload is shown in Table 1, which is a design target with a margin of 2 tons for one flight unit of SPS 2000. Each unit will consist of a modular portion of every subsystem of the completely assembled system, and can be operated as a small SPS even if the width of icrowave beams are not sharp enough due to miller antenna size. An integrated payload )nfiguration is shown by Fig. 3. Each unit is assembled in orbit and test operated and added to the unit(s) preceding in operation if it is a follow-on flight unit. Figure 4 shows a model of the flight unit assembled in orbit with the nose fairing on the same scale. There is a berthing space in the center of the first unit, where later flight units are unloaded. The construction work for additional modules will be made on both ends of the prism to maintain the position of the center of gravity.

Fig. 3. A configuration of integrated payload in Ariane V nose fairing.

As a result of the system analysis, several points requiring to be scrutinized for further study have been identified as follows.

As for energy conversion, solar cells are better fitted to this system than a solar dynamic system, considering the attitude control requirement. However, the assumed amorphous silicon solar cell has to be compared with others in terms of cost and mass properties and radiation degradation. The mass of the electrical cable which is larger than the structure mass is a critical problem for the sytem design. Further study on the possibility of using higher voltage, and analysis to shorten the length of the current path will be required.

Power transmission employing retrodirective as well as computer control is a key technology of this system in a LEO. The technology for this system is being developed with an engineering model as an experiment onboard the space station. The engineering data on the mass properties, thermal design, consistency of the high voltage of the power supply and other problems are not yet fully obtained. Selection of the type of semiconductor amplifier for the antenna elements will be most important from the standpoint of cost reduction and robustness of the system.

Many options can be considered for flight unit modularization. If the launch cost is higher than the payload system cost the size and mass of a flight unit will be affected significantly by selection of the transportation system. This problem is related to the payload integration and orbital construction work. Automated deployment and assembly with robotics will be used for the orbital construction work. The optimum combination of these two functions should be pursued for the purpose of reducing the related cost.

Rectennas must be located in the relatively limited zone where the SPS is visible at an elevation angle of at least 60 degrees. This is a zone between latitudes of 5.5 degrees north and south, or a 1200 km wide equatorial zone for an orbital altitude of 1000 km. Another constraint on site allocation is the minimum distance between two neighboring rectenna sites, which is required to make the power receiving time for each site as long as the beam angle can be controlled. Accordingly, the minimum distance between two rectennas should be 1200 km as well, and the reception time duration is approximately 3 minutes. Since the orbit altitude is nearly proportinal to the serviceable zone width of the latitude and the minimum distance between neighboring rectennas, a higher altitude covers more regions. On the other hand, the mass which can be carried by launch vehicle to a higher orbit will be reduced significantly. It will be necessary to compromise these two conflicting factors for orbit selection in addition to considering the debris condition.

Considering that the SPS 2000 system will be constructed modularly, the first few flight units will necessarily be used to test and verify the technologies of the system and operational functions. One of the rectennas will be designed as an experimental rectenna station, where the characteristics of the power transmission system are measured and newly developed equipment and facilities will be tested. Some scientific observation facility such as a radar for high altitude atmospheric research will be useful for observation of the phenomena caused by the interaction of the microwave beam and the ionosphere below 1000 km.

The ground system can be used for small local electric power stations, which are now powered by small hydraulic generators, photovoltaic generators or diesel generators. In this case, the power system need not compete with a large electric system for economical superiority. High reliability and quality of electricity are not mandatory. In this respect, this model can be used for isolated customers at many different sites in developing nations in the equatorial zone.

To prepare for evolution of the system, some of the rectennas will be designed as a future key rectenna sites. It is desirable for such a station to be provided with growth capability of the rectenna size and a longer distance from neighbouring rectennas for higher orbit operation. Such a rectenna will be useful for testing the first flight unit, whose microwave beam width is three times larger than that of a complete unit, so that power transmtssion tests will be performed more effectively with a larger rectenna.

Various types of rectenna can be designed for SPS 2000 independently from the orbital segment. More detailed discussions by other authors are expected (7 and 8).

This system will be flexible and evolutionary in several ways. Firstly, it can be operated while incomplete during construction as described. It will also be possible to increase the power output by adding extra flight modules. The technologies for this capability will be common to those required to maintain the orbital system. Accumulated experiences of the orbital operation will be valuable.

Secondly, multiple orbital power stations can be built and operated. A series of units in the same orbit could be coordinated to supply power for longer periods. This is the simplest way to increase electrical energy received at a rectenna without additional investment in the ground segment. If the orbital altitude is 1000 km, the maximum number of SPSs is thirty three and then all the rectennas can receive the nominal power from space almost constantly from early morning to early evening every day.

In the future, if advanced space technology such as electrodynamic plasma motors is available, it will be possible to move the SPS up to a higher orbit, which allows a rectenna to receive a larger portion of the electricity generated in an orbital period due to the longer visible time. Some coordination between rectennas will be necessary due to overlapping of the expanded operational area of each rectenna. Increase of the sizes of the transmitting antenna or rectennas will be required. If a rectenna diameter is 10 km as in the case of the Reference System, the present SPS 2000 can be operated approximately three times as high as the original orbital height.

The key technology developed for this system can be applied later to such larger systems as the Reference System. The rectennas for this early model can be designed to take more service from the more advanced system evolving from the smallest model. The quality of the electricity will be improved with adding energy storage systems to rectennas.

The CDEP has indicated that even if an SPS is technically feasible, it is difficult to realize it as an actual project or program because of other reasons. To look into this aspect, I would like to list and briefly discuss the important features required for the 10 MW class SPS which is much smaller than the Reference System, and the possibility for SPS 2000 to be realized from these viewpoints.

A 10MW class SPS discussed here is equivalent to a 300 kW terrestrial electrical power plant in terms of the average daytime power supply. Although just an experimental facility in industrialized nations, it could be useful in remote land. We need to find out the demand to use such a system as an operational power generating system, to justify the investment. Without such a demand, the SPS will not be successfully demonstrated as a power system for terrestrial application. This approach and the simple system will make SPS 2000 significantly realistic, compared with the Reference System whose concept has been developed as a hypothetical national electrical power system.

The first commercial electrical power station made by Edison started service with a single

200 HP engine in 1882 (9). The present terrestrial electrical power systems have evolved gradually from the early age models, responding to the demands and incorporating progress of technologies. The situation of SPS 2000 should be similar to Edison's first power station. It will not be a planned step of technology development of a future system in a framework of a huge single program. Space projects well defined in advance according to phased project planning. often give new technologies little chance to be employed when available. Since the mission of SPS 2000 is to simply supply electrical power from space and the system is modularized, technology updates which allows technologies used for major subsystems to be changed to avoid outdating of the system after it is planned can be applied more easily than the traditional sophisticated spacecraft. To ensure this, SPS 2000 definition of detail will be made only for each flight unit, but not for the overall system.

The cost estimation of the Reference System suggested that the SPS electrical power would be economically competitive with conventional power systems. However, the estimation is based on a low cost transportation system specially developed for this purpose. In the space industrialization era, the transportation service should be provided by the space infrastructure. Therefore, the SPS 2000 project will use available commercial transportation systems. The present transportation price is very expensive. The approach taken by the communications industry to make space business pay off the high price is to increase the performance of a space system per flight by employing expensive high technologies. This approach can not be applied for a power system which can not enjoy the value added economy. The only way to reduce the cost is to reduce the cost of the orbital and ground facilities. One definite cost target of the orbital system of SPS 2000 is terrestrial solar power systems. Assuming the cost is 10 US$/W, the approximate figure for a 10 MW model is 100M US$. In this case, the launch cost dominates the payload cost, so that future reduction of the transportation cost will directly enhance the merit of the SPS system. In other words, success of the first SPS will benefit the future of space transportation in return.

The radiation hazard caused by the transmitting microwave is a major public concern about SPS. The hazard is not known well but exaggerated and often misunderstood. A low power model

like SPS 2000 would be comparable with a conventional radar system in terms of the radio power level. A rough estimation of the power level is given for SPS 2000 here. When the diameter of the tran smitting antenna is 100 m and its distance from a rectenna is 1100 km the rectenna diameter is 3 km. The radiation hazard of microwaves of from an SPS 2000 which generates 10 MW will be roughly evaluated by comparison with the 5 GW Reference System having a 10 km diameter rectenna: the microwave power density for SPS 2000 is calculated to be forty five times smaller than that of the Reference System. The actual power level is about 0.5 mW per square centimeter at the beam center reaching to a rectenna. Considering also the exposure time of a few minutes in an orbital period of about 100 minutes, the hazard can be evaluated in advance by accelerated experiments. Therefore, the radiation hazard will not be a major obstacle to the start of this project. On-site experiments on hazards evaluation can therefore be conducted safely, and experimental data will be obtained for later projects.

An SPS to be placed in a low earth orbit inevitably provides many nations with an opportunity to participate in the project, while an SPS in the geostationary orbit can be used exclusively by the owner country. Especially, in the case of SPS 2000, the benefits of the SPS system can be shared by the countries located in the equatorial zone. Thus the project can be planned in the context of international cooperation between industrialized and developing nations. The international community has the opportunity to do something constructive for the future, based on the technologies developed in the twentieth century, which were mostly used for military purposes. SPS 2000 is not large enough for such a purpose but may work as the initiator of this kind of cooperative enterprise.

The ISAS solar power satellite working group has developed a preliminary concept of a 10 MW class SPS strawman model. In terms of the principle and key technologies, the model is in accordance with Glaser's original concept and the Reference System of the CDEP. However, the size and the purpose are different from theirs, since the study aims at demonstration of electric power supply to customers at earliest opportunity, while the Reference System was designed as a future national electric power system of the industrialized country.

To solve the problem of extremely high cost of space systems which is considered as the main obstacle against this purpose, the following points have been emphasized for the conceptual design.

1. To simplify the orbital segment, the solar array and antenna are fixed on the same structure to be stabilized by gravity gradient force.

2. To make the concept realistic, only existing technologies, including those y~et to be qualified, will be used, although renewal of technologies will be considered positively.

3. To reduce the cost for transportation to space which is dominant portion of the total project cost, the equatorial low earth orbit has been chosen rather than the geosynchronous orbit.

As a result, a concept of a model designated as SPS 2000 is being developed. SPS 2000 is modularized by flight units to be carried by a commercial launch vehicle. Every subsystem is divided into equal portions and installed evenly in each module of flight units. Thus, each unit can be operated and tested as a small SPS, although the performance is not satisfactory for customer service. SPS 2000 will transmit the microwave power to rectennas located along the equator separated from each other by distance of 1200 km for about three minutes in every orbit path.

Unique features of an SPS project indicated by this conceptual study can be summarized in comparison with the Reference System, as follows:

1. SPS 2000 is an evolutionary system which can be started from a size of a payload of a launch vehicle. It can be a first milestone even for the Reference System

2. SPS 2000 is substantially international system. International aspects of SPS will be discussed more seriously for SPS 2000 than the Reference System.

3. SPS 2000 can serve exclusively the equatorial zone, especially benefiting geographically isolated lands. This will be a aspect of societal issues which was not discussed in the CDEP.

These features suggest that the potential customers as well as electric utilities will play an important role in the next stage of the study. Their participation in the further study on the aspects of technology options, ground segment design and possibility of future evolution will be an essential step for SPS to be evaluated correctly without prejudice.

This paper is based on the works of ISAS working group. We would like to thank many people for valuable advices and suggestions on this paper. The opinions stated here are those of the authors and are not necessarily those of the working group.

1. US DOE and NASA, October 1978, " Solar Power Satellite, Concept Development and Evaluation Program", DOE/ER-0023
2. US DOE and NASA, November 1980, " Program Assessment Report Statement of Findings, SPS CDEP", DOE/ER-0085
3. P F Glaser, November 1968, " Power from the Sun: Its Future", Science vol 162, pp. 867-886
4. M Nagatomo, 1986, " 10MW Satellite Power System: A Space Station Mission beyond 2000", Space Power, Vol. 6, pp. 299-304.
5. N Kaya et al, A report on ISAS grant research in the Proc. of 7th Space Utilization Symposium, Tokyo, July 1990
6. J-L Claudon, private communication, 1990.
7. P Q Collins, 1991, "Design considerations for SPS 2000 ground segment", SPS 91
8. K Itoh et al, 1991, " An Inland Rectenna using Reflector and Circular Microstrip Antenna", SPS91
9. M Josephson, 'Edison', Japanese translation by T. Yano et al., Shincho-sha, 1962 (originally, Harold Ober Associates Inc., New York, 1959).