Orbit eccentricity will cause a cyclic fluctuation in satellite-to-rectenna distance. For an expected eccentricity of 0.04, this results in a daily power output variation of about 100 MW superimposed on the 6-month variation. Eclipses by the Earth (see sec. IV-A-3) will cause total shutdowns daily around local midnight for about 6 weeks in the spring and fall. Maximum duration is about 75 minutes. Eclipses by other satellites will cause shutdowns at about 6 a.m. and 6 p.m. local time for several days in the spring and fall, with a maximum duration of about 15 minutes. Total shutdowns will also be required at times for maintenance. Duration of shutdown is conjectured to be a few weeks, and the frequency is about once every 5 years. The primary operational problems will arise from the eclipses. At 0.5° spacing, as many as 38 SPS1s of a total of 112 will be in the Earth's shadow simultaneously, so that power sharing will be of limited usefulness. A shifting, nonequatorial orbit that is never eclipsed is possible but has not been examined in this study; it is expected that orbit maintenance propellant will be prohibitively heavy. No entirely satisfactory solution to the eclipse problem is apparent, and further study is required. Maintenance shutdowns will be less frequent, can be planned for off-peak conditions, and should consequently present less of an operational problem. The daily and 6-month cyclic variations in power output can be eliminated if necessary. A circular orbit avoids the daily fluctuation but at a cost in orbit maintenance propellant of some 200 metric tons/yr (I = 10 000 Ib-s/lb). The 6-month cycle can be eliminated by continuousssolar orientation instead of POP. Additional reaction control propellant on the order of a few hundred tons per year would be required or, for the column/cable configuration, counterweights of some 7000 metric tons could be used (sec. IV-B-4). The system trade-offs relating to the power station have not taken into account any adverse impact of these fluctuations on the distribution grid. It is possible that inclusion of these considerations may alter the orbit and attitude control considered in this study, and further iterations of the trade-offs should be made. F. Unit Costs Table IV-4 shows a list of cost estimating relationships (CER's) utilized for initial SPS costing. These CER's were produced using historically derived CER's for similar space equipment. Because of the very high volume production rates required for items such as solar cells, Schottky diodes (rectenna elements), and microwave generators, the CER's for these devices were substantially reduced below current values for space systems. An example of this expected cost reduction is illustrated in figure IV-18 for silicon solar cells. The cost reduction is projected by ERDA as a result of a major terrestrial photovoltaic research
RkJQdWJsaXNoZXIy MTU5NjU0Mg==