Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
6, MEASUREMENTS From the time in June l978 that a few subarrays were operating on the south face of the PAVE PAWS antenna, measurements of radiation in- tensity have been made at points on the ground at distances from l,600 feet to 2 miles from the antenna face. The contractor and government agencies have conducted test measurement programs. The PAVE PAWS Program Office has issued reports of the measurements made by government agencies. Reference 3 summarizes those obtained through August l978. These, as well as measurements made by the contractor, were discussed with the panel on September 7, l978. Table IV summarizes the measurements of average power density reported in Reference 3 by the Air Force team. The report noted in Reference 5, subsequent to Reference 3, includes measurements at many more locations. In general the levels reported are very much lower than Table III would predict, presumably because few of the locations have a clear line of sight to the radar. There is only one measurement of maximum power in Reference 5 that exceeds (slightly) what Table III predicts. Table IV lists the four nearby locations .where a clear line of sight exists. The measurements in Table IV were made with the radar operating in the enhanced search mode at a fixed carrier frequency and at a duty cycle of 0.20,. The Table also shows each measurement corrected, first by R^ to an equivalent measurement at lkmr and, second, by a factor l.25 (= .25/20) to a duty cycle of 0.25. The last column of Table IV therefore presents four long-term average power measurements for com- parison with the bottom line of Table III. The comparison verifies, that indeed the average power in Table III is conservative on the average by about lOdB, The model in Section 5 suggested 6.6dB. On page 7 of Reference 3 there are caveats about the measurements in Table IV--that these in fact are for conditions under which one would expect higher levels than Table III projects for the measuring stations. Hence measurements in Reference 3 confirm the conservative values of Table III at nearby points on the ground. They also are consistent with the model of the spiky sidelobe pattern used in the qualitative discussion in Section 5. This also suggests that the model of the modulation of the envelope of radiation at a point on the ground, as developed in Section 5, is valid. The contractor's time 29
30 Table IV MEASUREMENTS FROM REFERENCE 3 AND REDUCED TO 1km AND DUTY CYCLE 0.25 Station Distance from radar (feet) Azimuth from Bores ight Ave. power density yW/cm2 Adjusts km and ⢠0.25 (y l 3l00 l2° 0.87 .98 2 3900 0° 0.38 .68 3 l600 9° 3.26 .98 4 l800 63° l.7l .64 Average for four stations .82
31 records tend to confirm this latter conclusion as do data in Reference 3 on the recurrence periods of peak measurements. In reviewing the measurement program the panel concludes that the instrumentation used and the procedures followed conform to good engineering practice. In particular these conclusions apply to the instrumentation and procedures used by the Air Force team, which has done the bulk of the measurements reported to date. What has been reported to date and the measurements planned for the future should therefore give a valid picture of the intensities experienced at near- by points and.at those points of public access where exposure will be greatest. As this picture now stands, the EPA analysis (Reference l) appears to be conservative by at least several dB--a conservatism that is, moreover, supported by design data about the structure of the side- lobe patterns. The four stations shown in Table IV exhibit average measurements that, reduced to a standard distance in the last column, vary among themselves over a range of nearly 2dB. Other measurements reported in Reference 3 and the contractor's measurements as reported to the panel cover a range wider than this. The variations are not surprising even in average power measurements. The presence or not of a clear line of sight is important. The possibility of sharp differences from one azimuth to another during enhanced search has already been suggested. Reference 3 does not claim a calibration of the measuring system to closer than ± 2dB for average power measurements. It is in the nature of the fields being measured that partially standing waves created by reflecting features in the local terrain can create local peaks and nulls. Even though the standard procedure is to seek a worst position (highest measured power) within the accessible area around each measuring station some variability is to be expected. The panel therefore does not consider that the variability among the measurements as reported in Reference 3 reflects in any negative way on their validity. On the contrary, much less variability could be a cause for concern. Neither does this variability cast doubt on the conservatism of Table III, within its limitations. There is another source of variability mentioned in Section 5--namely, variability over time. There is an ambiguity that is more than merely semantic in the uses of the terms "peak power" and "peak field." A pulse from the PAVE PAWS radar can be as long as l6 milliseconds or as short as _25 milliseconds. In Section 5 the panel noted that it is possible that a single pulse may be sampled at maximum sidelobe gain while all other pulses that occur at times nearby are sampled through a filter having, relatively, many dB less gain. What is measured and reported as a "peak power," then, can depend in a dramatic way on the averaging time intrinsic to the instrument as well as on the statis- tical sampling procedure by which readings of "peak power" are defined and reported. The panel recommends that measurements of power be made with an instrument of known averaging characteristics, preferably one with a time constant not longer than 0.2 seconds, and that rapid repeated sam- ples be taken on a known and accurately defined sampling cycle that is
32 not accidentally synchronous with a subharmonic of .054 see~* nor with any cycle rates intrinsic to the search patterns being measured. Such measurements taken over a period of several times l0 seconds would allow a statistical analysis of "peak power" measurements capable of a more refined check against Reference l and Table III than has been made here. Given the instrumentation as reported in Reference 3, it seems likely that data of the kind just suggested already exist in the time records. If so, the panel suggests that these data be given some sta- tistical analysis. Of interest, as suggested by the considerations of Section 5, would be recurrence times or recurrence rates of power peaks above threshold and of the average power delivered in peaks above threshold measured at several different threshold levels.
