On 6/1/16 5:07 PM, Jim Brown wrote:
On Wed,6/1/2016 1:45 PM, Hans Hammarquist via TowerTalk wrote:
Anybody with a "simple" explanation?
It can be tough to provide a simple explanation of a complicated issue,
but I tried about a week ago in this thread. :) It is FAR more than
capacitance to ground, although that's a component.
The complex part of the interaction of a horizontal dipole with the
earth below it is that the earth is somewhat conductive, current flows
in it, and it couples back into the dipole like any other passive
element. The amplitude and phase of the reflection depend on height and
on the current in the ground, which in turn depends on soil
conductivity. At some heights, the reflection adds more nearly in phase,
at others more nearly 180 out of phase, and at a different phase
relationship at every point in between. THAT'S why the feedpoint Z
changes in the complex way it does.
I've seen plots of this in one of the ARRL antenna publications,
probably the ON4UN book. For an antenna that's "high" as a fraction of a
wavelength, Z varies around 70 ohms, for a low one, around 50 ohms.
Combine that with the mutual impedances of 2 other elements in a 3
element yagi, and it gets real interesting. This is why, back in the
day, before modern modeling codes were available, there was a lot of
"we'll calculate for free space and hope for the best".
Modeling is your friend.
You may not have any clue what your local soil properties are (and in
any case, they're not all that uniform), but you can easily run the
model with a range of them and see, in general, what's going to happen.
The impedance discontinuity at the air/soil interface is big, so that's
actually the dominant effect (that is, if you modeled the Z over a
perfect ground, it's probably pretty close to what the real Z is, as
long as you're more than, say, 1/10th wavelength away)
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|