[TowerTalk] Impact of high water table levels on antennas

[Jim Lux]

On 11/3/13 5:25 PM, Rudy Bakalov wrote:
> So, what is a desirable skin depth- low or high value?

It depends on the loss.  A shallow skin depth is usually better, since 
you get shallow depths with high conductivity (e.g. copper is better 
than brass, and has shallower skin depth).

The real issue is that if you have something like dry soil over a 
shallow water table, there's not much loss in the dry soil, and the 
dominant effect is that water table, even if it's a couple meters down.

This also is where blindly using numbers from MF broadcast stations 
(e.g. the FCC charts of soil properties) may be misleading for HF.  The 
wavelength in the soil goes as 1/square root of epsilon, so if the 
epsilon is 13 (the classic "good soil" number), the wavelength "in the 
soil" is about a quarter what it is in air. So a layer that's 1-2 meters 
down is actually about 1/4 wavelength for 14 MHz.

When you're a BC station at 1 MHz in the 300 meter band, this is less of 
an issue.  Skin depth is deeper (goes as 1/sqrt(frequency)), but the 
wavelength is much longer.

For H-pol a grazing angles, the difference isn't much (the air/surface 
reflects most of the power), but if you're looking at verticals, or a 
low horizontal antenna, this kind of thing can account for a lot of the 
difference between "observed" and "expected" performance.


I'll also point out that soil properties are not particularly 
homogenous, at least in places where I've measured them.  Maybe if 
you're out in a plain with hundreds of feet of soil deposited long ago, 
and no local water variations, it's consistent, but in any sort of 
suburban environment, there's huge differences in soil properties over 
distances of a few meters.



>
> Rudy N2WQ
>
> Sent using a tiny keyboard.  Please excuse brevity, typos, or inappropriate autocorrect.
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>