It seems like so long ago in the 1960s when I heard my first front-to-back demonstration of a two-element, 40 meter horizontal phased array by Van, W5ZED.  I was astonished by the 30+ dB of front-to-back from Van’s antenna, and I knew from that moment on that I could not rest until I put up a similar array.

I listened intently as Van described his set of two dipole elements, 30 feet high, spaced at 28 feet.  (For the old timers reading this, Van used the Kirk Helical-Wound elements for his dipoles.  These elements used copper tape that was helically wound on a 22 foot tapering fiberglass tube.  The 40 meter element was 44 feet tip-to-tip, about two-thirds the size of a full size 66 foot element.) The dipoles were fed with electrically equal lengths of coax, terminated into a switching arrangement that would allow him to choose different lengths of delay line between the elements.  The switching arrangement would also allow him to swap the delay line between the elements to reverse the direction of the array.  His dipoles ran north and south, producing a pattern that fired east and west.  Van lived in San Antonio, Texas, and I lived in Crowley, Louisiana, due east of Van and right down the center of his antenna pattern.  In addition to an obscene amount of front-to-back, Van consistently had one of the strongest signals on the 40 meter band.  I don’t know who invented this particular configuration of the antenna, but Van certainly was one of the early proponents of the two-element phased array.  Other notable phased array experimenters that I remember from this era were Merle Saxton, W5CNM, a broadcast antenna engineer by trade, and Jack, W5RGP.  While Van used coaxial delay lines, Merle and Jack used variable L/C phasing networks.  It did not take me long to put away my coax delay lines and become a proponent of the variable LC phasing network.

Van felt that he could make significant changes in the frontal pattern take-off angle by changing the phase relationship between the elements.  I think he believed this because of the changes in signal strength that he could detect off the front of the antenna as he selected different lengths of delay line.  Unfortunately, I have been unable to verify this with my modeling of the antenna with various NEC2 antenna programs.  To the best of my ability to determine with software and on-air tests, the take-off angle and frontal lobe of the phased array is not altered significantly by the phase relationship between the elements.  The biggest change in take off angle I could detect, from a wide range of phase angles, was one-and-a-half degrees.  As a result, I must conclude that the take-off angle on the front-side of the phased array is almost totally determined by the array’s height above ground.  The angle at which maximum front-to-back occurs, however, is drastically affected by the phase and current relationship of the two elements.  To summarize this point, as the phase angle between the two elements varies, the angle at which maximum front-to-back occurs on the rear of the pattern changes dramatically, while the frontal lobe changes very little. 

Van also described his phased array as having a cardioid pattern in the azimuth plane, but software generated patterns and on-air tests with rotatable two element phased arrays are in close agreement to a pattern virtually identical to a parasitic yagi azimuth pattern.  My guess is that Van looked at the azimuth patterns generated by vertical arrays and assumed that the horizontal array would be the same.  For those who are only able to put up fixed dipoles for the phased array, don’t be too discouraged.  I’m sure Van felt the azimuth pattern was a cardioid because the antenna is a very good performer, even 45 degrees off the main axis.

.