the first 150 hours there were three failures. There were no more failures. If we regard these early failures as of an infant mortality type, a total of 2763 hours without a failure or noticeable degradation in efficiency performance had been achieved. Thus, 549,83 7 diode hours have been accumulated without a failure or noticeable efficiency degradation. If we apply MTBF tables to this life test data, and request a 90 percent confidence factor, we find that the MTBF is 23 9,059 hours or 27.3 years. If we relax the confidence factor to 50 percent, an MTBF of 785,481 hours or 91 years is obtained. The efficiency of the diodes in a set was monitored by periodically checking the current which flows into the common load resistor for each set, when the RF power which flows into the illumination horn is precisely set at a pre-established value. The data in Table 4-2 indicated negligible drift in efficiency performance after 1800 hours of life test operation. The average ratio of the last measured currents to the initial measured currents is 0.9982, representing a very small change and one which is easily attributable to measurement error. The average deviation of the ratios from unity is 0.0123. Of special interest is the fact that no degradation was noticed for the rectenna elements operating at high power levels. These power levels are several times what would be antifipa- ted in the SSPS rectenna. Figure 4-3 provides a summation of the life test data taken up to a total of slightly over 800, 000 diode hours. After the 2913 hours of life test just reviewed an operator accidentally applied much higher microwave power to the test set and four diodes failed in Group F. The first life test failure not co be confused with operator error occurred at 3078 hours. It occurred in Group F. The life test was continued without additional failure to 4178 hours when it was shut down for two weeks. In restarting the life test the single center element failed. It is believed that the applied microwave power was at an excessive level. The life test was discontinued on 24 March 1976, a little more than a year after it had started. In summary, a great deal was learned from the life test. The test was not a continuous life test and there were substantial periods of time in which the set up was not operating. Hence there was also a kind of shelf life evaluation to help rule out a "sleeping sickness" type of syndrome. Many of the groups of elements were run at power levels several times that anticipated in the rectenna for the SSPS and no degradation in efficiency was noted for these elements after a running time of 1800 hours. The life test results were marred by operator error but there was a continuous period of time with no failures of any kind which indicated a mean time between failure of 27 years with a confidence factor of 90%
RkJQdWJsaXNoZXIy MTU5NjU0Mg==