DanStrassberg said:Sometimes, however, the position of the antenna with respect to the tower structure is counterintuitive. A now-deceased friend engineered a station in central New York and achieved a relatively high gain toward the CoL by placing the antenna NOT on the side of the tower closest to the CoL but on the side opposite--and facing it AWAY from the CoL. Meanwhile, the competition did it the obvious way and wound up with a much inferior signal in the CoL.
The only correct way to get the intended coverage is to have the antenna manufacturer model the antenna AND the tower together, even if it's not a directional antenna (in which case it MUST be modeled in order to meet the CP's specs for the nulls of the pattern). I had to have this done with a station I put on the air about 15 years ago, at the Shiveley factory in Maine. It was a fascinating process and it took us four or five tries, but we got exactly what the CP specified for the pattern and everyone was happy with the result.