A lot of folks have done the "iterate the soil properties in the model to match
measured performance" (all the way back to the 1960s, at least). Some antennas
(aka probes) work better than others for this (in terms of
accuracy/uncertainty/frequency range). The problem with antennas, in general,
is that they're influenced by more than the soil underneath them. Hagn was
trying to develop modeling parameters for various jungles using a dipole at
varying heights, for instance. And subsequently, there's been a fair amount of
work on developing "slab models" for the vegetation and other stuff above
"ground level" - although most of that focuses on VHF and up.
The OWL probe and similar were invented to be "better" (smaller, not as
influenced by what's above the soil surface, etc.)
There are even commercial probes that you can bury - they're used in
agriculture.
I think this works at 50 MHz, but might be adjustable.
https://stevenswater.com/resources/datasheets/HydraProbe%202020_draft3.pdf
Here's some actual data from Owens Valley Radio Observatory
https://github.com/ovro-lwa/leda_docs/wiki/Soil-Permittivity-Sensors
And here's a guy who studies this
https://www.umass.edu/earth-geography-climate/about/directory/william-p-clement ;
although the link in the OVRO memo is broken.
On Wed, 4 Mar 2026 11:56:13 -0700, David Gilbert <ab7echo@gmail.com> wrote:
I recently did an EZNEC analysis of a vertical antenna fed against a
square wire mesh instead of radials. Like all radials, the mesh is just
a counterpoise for tuning to resonance except that the resultant
feedpoint impedance is lower than with longer radial(s). I previously
posted the link to a YouTube video I did of it here.
The mesh is HEAVILY influenced by height above ground because basically
the mesh acts like top hat capacity loading does for a vertical, or "top
hat" capacity loading does for the ends of a dipole. The EZNEC model I
created simply used the default ground conditions, but it seems that an
actual test of a vertical element fed against a defined mesh various
distances above the surface might generate a plot of resonance that good
be curve-shaped against EZNEC plots that used different soil constants
as a way of working backwards to the actual soil constants.
In my case, I'd be most interested in trying to determine how deep the
equivalent RF ground is for future modeling (because it certainly isn't
at the surface of my dry Arizona hillside) but maybe the conductivity
and permittivity tell us the same thing. I've never known whether EZNEC
calculates an equivalent RF depth based upon the soil conditions or not.
73,
Dave AB7E
On 3/4/2026 9:05 AM, Jim Lux wrote:
> Yeah.. the geophysical prospecting literature is full of things like this.
> And as you say, tends to be low frequencies (for them, using an AM broadcast
> station at 1 MHz is "high frequency").
>
> I'm particularly interested in higher frequencies to do things like building
> a model of the subsoil structure.
>
> I might just have to take a VNA and hook it up to some coils and try it.
>
>
>
> On Wed, 4 Mar 2026 14:58:20 +0000 (UTC), RT Clay via TowerTalk wrote:
>
> Non-contact measurement of ground conductivity is a standard method in
> geophysics and archaeology. What is usually used is the "Slingram method". In
> that method, an AC signal is sent to a small coil, and the induced voltage is
> measured in a second sensing coil a short distance away. The effective depth
> of the measurement can be controlled by how far apart the coils are (typical
> distance between coils is 1-2 meters). In a typical measurement, the coils
> are placed on or just above the ground, and the device is moved to measure
> conductivity along a line or grid. A contour map of conductivity can then be
> generated to locate buried features. A long time ago I played around with one
> of these devices (my father is in archaeology).
> The problem with adapting this for use with antennas is that the analysis is
> done assuming the DC limit. Typical measurements are done at something like
> 10 KHz. A quick google search did find some work at getting this to work in
> the few-MHz region, but the analysis there is not simple.
> Tor N4OGW
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk@contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>
>
>
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk@contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|