A sun-power satellite? Not likely in this century

For 13 years, Peter E. Glaser has promoted a mind-boggling dream. An array of satellites equipped with solar batteries could transform solar energy into radio waves, beam that power to Earth, and meet much, if not all, of the world's energy needs.

But while the idea has captured the imagination of many people and while the US government has spent $20 million studying it, the concept seems too far beyond our reach to be useful in this century. Indeed, a committee of the National Research Council (NRC), the operating arm of the US National Academy of Sciences, has recommended no further research funds be spent on it in this decade.

Yet the NRC is reluctant to dismiss the idea out of hand. What seems hopelessly ambitious today could be a valuable energy option 50 to 100 years from now, the NRC notes. Thus it urges the Department of Energy (DOE) and National Aeronautics and Space Administration (NASA) to monitor development of relevant technology and periodically to report to Congress on the Feasibility of a Solar Power Satellite (SPS) system.

The breathtaking scale of what Glaser, a vice-president of the research company Arthur D. Little Inc., has in mind is reflected in the so-called SPS reference system. This is a design worked up during that $20 million DOE/NASA study, which the two agencies and the National Science Foundation asked NRC to evaluate.

Sixty satellites, each as large as Manhattan Island and of 56,000 tons mass, would be built in the geosynchronous orbit 36,000 kilometers high where they would remain over a given spot on the surface. This array, to be completed by 2030, would beam down 300,000 megawatts of power, about half the 1980 US generating capacity.

The DOE/NASA study came up with a cost of $1.3 trillion. NRC, calling that study's final report "optimistic rather than pragmatic," says the cost would be more like 2.5 times as much. Funding such a system would mean spending something like 12 times the current NASA budget annually for the next 50 years. Ferrying the ment and materials into orbit to build the satellites requires spacecraft with payload capacities 13 times that of the shuttle to be launched at a rate of more than one a day for 30 years. Moreover, 1,500 workers would have to be in space at any one time.

These are just a few of the needs that make the project at present infeasible. But NRC says it is "too early to attempt to pick winners and losers" and would keep an open mind on SPS for the next century.

A study by the congressional Office of Technology Assessment, not released at this writing, draws a similar conclusion. However, it finds that spending $5 million to $30 million a year on SPS research now could be useful in defining needed technologies and in encouraging development that could cut future SPS costs.

Perhaps a century hence Peter Glaser will be remembered with gratitude as a farsighted prophet.