Download this article in .PDF format
This file type includes high resolution graphics and schematics when applicapable.

Two 12-in. sections of copper-plated waveguide (1/8-in. thick) terminated by 50-ohm coaxial connectors were built to determine operating transmission characteristics. Connectors were matched to the waveguide and positioned approximately λg/4 from the ends. One waveguide was built of silicone-fiberglass and the other of Teflon-fiberglass. Insertion loss in the Teflon-fiberglass section was slightly under 0.3 dB over a 300-Mc bandwidth, whereas that in the silicone-fiberglass was less than 0.5 dB over the same frequency range (see Fig 3). The input VSWR for the Teflon-fiberglass guide was less than 1.20 over a 400-Mc bandwidth and averaged about 1.10 between the limits of ±100 Mc from f0.

The center specimen in Fig. 4 is a bare silicone-fiberglass substrate section of one antenna array. The closely spaced holes forming the edges of the waveguide and the smooth surface finish are evident. The complete copper-plated array with radiating slots is also illustrated. In most cases only 0.003 to 0.005 in. of copper is plated on the surface. Note that the edges of the guide are cut along each side and through the center of the pleated holes; this was done to determine if any of the energy propagating in the waveguide radiated from the edges. Only a negligible amount of rf leakage was noted.

The array at the top of Fig. 4 is completely encased by copper-plating. Its radiation pattern is shown in Fig. 5.

Radant system performance

Insertion loss was measured of four waveguide test sections built into a radome before the slots were cut in the waveguide. The results (see Fig. 6) indicate insertion loss variation from 0.4 to 1.25 dB over a 400-Mc band. Normally, the insertion loss in a 12-in. length section averages 0.4 dB; however, because of the plated-through holes and epoxy coating used on the surface to minimize irregularities, as much as 1.25-dB loss was measured for the 14-in.-long section. With perfect samples and improved assembly techniques, the measurements over the same bandwidth indicated less than 1-dB loss. Insertion loss measurements taken on these antennas with the slots radiating varied from 8 to 12 dB. One-way elevation patterns of these antennas (measured in the radome) showed the average gain to be slightly above 7.0 dB with side lobes in each case suppressed 15 dB below the peak of the main lobe.

A two-way (two transmitting antennas, two receiving) azimuthal pattern taken on the antenna system is shown in Fig. 7. This is the radiation pattern generated around the body of the cone at an angle of peak intensity, or 72 degrees off the longitudinal centerline axis of the radome. The two transmitting antennas are diametrically opposite each other; the receiving antennas are positioned likewise. The peak gain of the (2 x 2) antenna arrangement as seen in Fig. 7 is approximately 9 dB. The symmetry is quite good, indicating good power balance in both the feed network and antennas.

Figure 8 shows a partially completed radant. Three of the antennas are covered with a thin spray coat of epoxy-base paint. The fourth antenna remains exposed to illustrate the slotted antenna and part of the silicone plastic cone.