starman wrote in message ...


I agree that the feedline of an elevated ground plane can be effectively
decoupled using radials but it's not clear to me how you would use
radials with the typical inverted-L. Where would you locate the radials
in that case?




On the ground. And in that case, I almost always do ground the radials
to earth at the feedpoint. Radials on the ground are detuned, and are
not required to be cut to any certain length. So they are not really
resonant at any given freq like elevated radials, or if they are, it's
likely not where you would expect. They normally need to be resonant
to decouple the feedline, so it's best to also ground them if they are
on the ground and detuned. If you had an elevated GP with 1/4 wave
radials for a certain freq, they will not work at say twice that freq.
"1/2 waves" They would show a high impedance, and the radial system
would not function properly, and the decoupling would be poor. But 3/4
wave radials can work ok, as they show a low Z. The best ground
planes, verticals , etc use twin decoupling sections. IE: a 1/4 wave
ground plane with a set of 1/4 wave radials, would have a second set
of 1/4 wave radials, 1/4 wave below the main set. Or if a sleeve
vertical, two sets of cones, tubes, etc. You can also use chokes,
ferrite beads as extra decoupling with any coax system. I'm not
against grounding in some cases. I just wanted to clarify that it's
really the improved decoupling, rather than the addition of ground per
say, that reduces the noise ingress. Grounding is just one method used
to improve decoupling of the line in some cases. Usually with radials
on the ground, or when using no radials at all. If the antenna is
already balanced and properly decoupled, adding an "rf ground" will
not do a thing as far as noise ingress. Might even make things worse
being ground is usually a noise source. MK