Hello Steve

About 20m band antenna height.

The height of your antenna depends on what you want to do.

For "local" daily traffic (ground wave propagation [not confusing
surface wave propagation]) you want as much as possible signal strength
in the horizontal plane (low elevation). In that case height gives
stronger signal. No matter the polarization, over land (for very small
elevation angles) the ground reflected wave is opposite to the direct
wave (elevation (pseudo) Brewster Angle). More height gives some
more phase difference between direct ray and the ray that arrives after
ground reflection. The "two ray propagation model" is based on the
direct wave and ground reflected wave (for very small elevation). So
the CB people aren't that crazy.

When the height equals many wavelengths, you get many lobes in the
elevation radiation pattern, but for local traffic this is not
important.

For sky wave propagation the situation is different (as also mentioned
by Greg). If you don't want lobes in the elevation radiation pattern,
your height is limited to about halve a wavelength. Further raising the
antenna gives lobes, but the elevation of the first lobe decreases.

About losses. You have near field and far field losses. If you place
your antenna about a wavelength (or more) above ground, you have far
field loss only. However placing your quarter wave with some radials
close to ground will bring the ground within the (strong) reactive
field zone. This increases the ground loss (you will notice this by an
increase in the real part of the input impedance). You can reduce the
near field at the ground by adding more radials (as suggested
indirectly by Greg).

In my previous posting I suggested to higher the base of the antenna.
This is also to reduce ground losses. A quarter wave vertical has low
input impedance, causing relative high current in the radials. When
the radials are close to (or in ) the ground, losses do occur.

I did some simulation on a quarter wave vertical (20m band) fed over 4
radials (horizontal), placed at various heights above ground (15m thick
dielectric layer representing average ground, very long run times, IE3D
3D planar simulator). The results were bad (Radiation efficiencies
below 10%). More height gives lower ground dissipation. I hoped to get
better results with a halve wave radiator (high impedance, hence
reducing the current in the radials significantly). The simulation
showed a significant increase in radiation efficiency (above 54%).

A halve wave radiator with 4 horizontal quarter wave radials (radials 2
m above average ground) has a first lobe at 17 degrees and about 4 dBi
in that direction).

So steve, I would recommend you to extend (if possible) the radiator to
a halve wave (you need impedance match at the base), you can run your
radials in the horizontal plane, radials should not touch ground.
Extending a 5/8 wave does decrease the input impedance (hence
increasing the currents in the radials) and thereby increasing the
ground loss.

If you cannot extend the radiator, then increase the base significantly
so that the ends of the radials are several meters above ground, and /
or use more radials.

I hope this will not give confusion.

Best regards,


Wim
PA3DJS