[TowerTalk] One more ground radial question

Jim Lux jimlux at earthlink.net
Wed Dec 17 15:43:04 EST 2003


At 10:40 PM 12/17/2003 +0000, Rob Atkinson, K5UJ wrote:
>Y
>Fresh water (rain) does not have anywhere near the conductivity of salt 
>water.  It alone on your radial field should not make a drastic difference.

Indeed, the rain water may be fairly non-conductive, but once it hits the 
soil, everything's different. There are soluble salts in most 
soils.  There's also dielectric constant and loss tangent effects to 
consider. For instance, clay soil, with tiny particles, is very different 
than sand, in terms of what happens when it gets wet.

For instance, sandy soil at 1 MHz
Pct Water  Relative Epsilon  tan(d)
0       2.59    .0017
2.18%   2.5     .0025
3.88%   4.7     1.75
18.8%   20      4.00
Sandy at 10MHz
0       2.55    .0016
2.18    2.5     .0025
3.88    4.5     0.3
18.8    20      .35


Loamy soil @ 10 MHz (don't have data for all points at 1MHz
0       2.48    .0014
2.2%    4       .45
13.77%  14.5    1.3

Clay soil @ 10 MHz
0       2.44    .004
20.9%   21.6    1.7



tan d is the loss tangent, related to the lossiness (most damp soils have 
tan d increasing as frequency gets lower). It's also affects the skin depth 
(which is related to the square root of conductivity), which in turn, 
affects whether a simple uniform current approximation is going to work.

The thing to note here is that for clay, the loss gets bigger much faster 
as the water content increases.

More data at http://home.earthlink.net/~w6rmk/soildiel.htm
Some day, I'll get around to converting George Hagn's numbers into a nice 
consistent table.

Maybe the takehome message here is that ground systems, particularly for 
80m and higher frequencies are more complex than one might imagine, 
especially when the spacing between ground wires becomes a significant 
fraction of a wavelength.  In a lossy medium like earth, the wavelength is 
very different from free space.. epsilon = 20 means that the wavelength is 
1/4.5 that in free space... for a 40m system, a quarter wavelength, in the 
soil, is only about 6-8 feet, so even a foot or two, in the dirt, is 
significant.

All those nice "current sheet in a resistive medium" approximations so 
beloved of Brown and ham authors really don't hold up very well when you 
talk real soil at real frequencies.

Jim, W6RMK








More information about the TowerTalk mailing list