process is performed on the lunar surface, then chemical propulsion using terrestrial fuel and lunar oxygen becomes competitive with electromagnetic propulsion. These conclusions are based on the assumption that productivity in space will eventually achieve the same order of magnitude as on earth. Acknowledgements — This work was made possible by the H. N. Slater Flight Transportation Development Fund. The editor wishes to thank Gordon R. Woodcock and Gerald W. Driggers for their assistance in reviewing this paper. REFERENCES 1. P.E. Glaser, Solar Power via Satellite, Astronaut. Aeronaut.. August, 1973. 2. G.K. O’Neill, Space Colonies and Energy Supply to the Earth, Science, December 5, 1975. 3. G.K. O’Neill, The Low (Profile) Road to Space Manufacturing, Astronaut. & Aeronaut., March, 1978. 4. M.S. Tomusiak, Underwater Simulation of Zero-Gee Dynamics and Assembly, M.I.T. Space Systems Lab Progress Report on NASA HQ Contract NSG 7396, August, 1978. 5. Space Manufacturing Facilities, M.I.T. Space Systems Lab Progress Report (No. 2) on NASA MSFC Contract NAS 8-32935, August, 1978. 6. Megawatt Solar Array Study, Electro-Optical Systems Final Report for NASA MSFC Contract NASA MSFC Contract NASA-21396, January 7, 1970. 7. J.B. Kendrick, Ed., TRW Space Data, TRW Systems Group, 1967. 8. Livingston, Baker, Poon, and Kuo, Satellite Power System (SPS) Preliminary Resource Assessment, Jet Propulsion Laboratory, August 7, 1978. Fig. AL Silicon solar cell cost — actual and projected. APPENDIX A. SOLAR CELL COSTING One of the most difficult quantities to predict with any degree of certainty is the labor required to produce solar cells in the very large quantities required for SPS production. Present demand for solar cells does not as yet warrant mass production techniques and as a result the processes used are highly labor intensive. Nevertheless, the cost of solar cells has been decreasing along a well defined experience curve which, as shown in Figure Al, clearly indicates that during the next two decades a reduction in cost of two orders of magnitude to about $1000/kW may be reasonably anticipated. This experience curve is based on the current limited production rates and represents primarily improvements in the manufacturing processes. If, starting from this expected level of experience, a learning curve is adjoined to allow for the very considerable expansion in solar cell production quantity represented by the SPS program discussed in these pages, then a further reduction in cost may be anticipated. This would lead to a result similar to that given by equation (1) for the average cost of production of TV units, or In a single current production process, 214 people manufacture 3 MW of solar cells each year (Ref. 6). Allowing for system efficiencies, a 10 GW SPS will need 18.3 GW of on board electrical power. For 2(
RkJQdWJsaXNoZXIy MTU5NjU0Mg==