Topband: Ground Conductivity

Wed Mar 25 15:07:12 EDT 2020

Rudy is not the only one taking measurements.

I've been doing EZNEC since 1996 (v2) and NEC4/EZNEC Pro since 2002. The
latter cost me well over a thousand bucks, but has been well worth it in
the long run and then some. NEC4 does on the ground and under the ground in
a reliable way, but still does not match some aspects of reality. Rudy got
it to track in very confined circumstances, which by the way is a huge
clue. Rudy's confines are not available to most folks.

Back before we published on the FCP, we were trying to make sense of how
BOGs behaved, and decided on a plan to measure VF to either confirm or
blast a theory that the electrical length of a BOG varied enormously and
simply had to be taken into account if modeled dimensions were to be of any
use at all.

A lot of us roughly in the Raleigh area were putting a 151 foot (46m)
dipole on the ground (DOG) and taking measurements, frequency and R where
X=0 and ditto for X = +/- 25 ohms. About ten of us were taking measurements
in a roughly 100 mile radius around Raleigh. The 151 feet was to give an
even metric equivalent, and to get X=0 roughly in the 160 meter region.

The **measured**, not modeled, velocity factor (VF) of a DOG was anywhere
between 45 and 85 percent. The R at X=0 was anywhere from 80 to 200 ohms.
Very frequently various placements of the DOG on the same property made
maximal changes in the results. Rotating a DOG 90 degrees around its center
would sometimes make huge differences. A permanently set, frequently
measured DOG would have large variations with dampness of the general
weather pattern and under snow, etc, and exhibited the gradual change
"settling in" measured by Rudy.

This resulted in a realization that a BOG would have to be modeled for it's
electrical length, and adjusted somehow at the exact placement. E.g. if a
DOG is trimmed to get it's X=0 at 1142 kHz, and then soldered back together
and insulated at the center and grounds, terminations and cable only placed
when the end points that matched the model's electrical length were known.

The composition and behavior of ground varies wildly from place to place. A
very dense, uniform around the compass, and decently large ground radial
system will tame that. Elevated counterpoises will tame that to a lesser
degree, depending on as much independence from inducing ground as can be
managed. Or you can go horizontal and get into a different and more
friendly paradigm. Not so easy for that on 160.

I myself, and perhaps ten other NEC4 users I know of have tried to get NEC4
ground characteristics to match, with little or only isolated successes.
This is known well enough by the commercial AM BC antenna folk. In filings
the FCC will accept the NEC4 ground constant which generates the field
strength which matches the actual measured field strength after antenna
construction. Only the hams seem to think that the program has mastered the
phenomena. It has not.

It probably could be reprogrammed to deal with data from about a hundred
ground measurements in the antenna near field. But then the square and
cubic law aspects of the program run time would get in the way again, as
was critically still the case when those programs were developed.

Further there is an aspect to the skin effect which NONE of the programs
takes into effect.

At N4AF, where we were doing the NY4A multi contests, there was an
interesting SWR anomaly that would sometimes creep in. Howie had a monster
5 element 40m quad fixed on Europe, and hung across a 200 foot catenary
between two towers. The quad element centers were at 84 feet, so ground was
in play. Some early mornings, the SWR would rise to 1.5:1 from it's normal
1:1. It coincided with a still night and dew on the grass. That was when
the layer of air above the ground was at 100 percent humidity, which meant
that moisture from the ground was not percolating upward and being
evaporated into the air. With any wind, the ground close to surface would
have less moisture the closer to surface, and if it had been dry, a good
deal less. Less moisture, less conductivity.

Effectively, the height above actual RF ground had changed, or its
resistance had changed or some ghastly mathematical formula relating the
two had changed.

All the formulas about skin effect ASSUME uniform conductivity to a surface
which is bounded by an insulator. NOT a layer of a more resistive
conductor, and certainly not a conductor that gradually gets more resistive
as it gets drier as it approaches the surface, and also varying in a daily
or even hourly sense with the weather.

The problem for ground radials is that frustrating most-varying medium is
what the wire is on/in, and on/in at the point of most variance.

I'm to the place where I view those problems as something needing avoidance
altogether, IF I don't have the land and the copper to put down the
commercial standard. Sparse, too-small, irregular radials will NOT avoid
the near certain losses from that medium. Either go commercial or get
everything up in the air and monkey with the counterpoise to minimize the
RF fields at ground. Do it right or get up in the air with some trickery to
avoid the loss, as simple as that.

73, and everyone please keep safe.
Guy K2AV

On Wed, Mar 25, 2020 at 12:17 PM Grant Saviers <grants2 at pacbell.net> wrote:

> "If NEC2 is wrong, so is NEC4."  - I don't think so since a different
> ground modeling approach is available in NEC4.2.  Running EZNEC Pro4
> antennas both in 2 and 4.2 sometimes shows significant differences.
>
> My .02.  Exactly where is ground?  It sure isn't the flat smooth plane
> in NEC.  And where does "real" ground start considering the huge variety
> of stuff going on in the first few inches from 100% air to solid dirt?
>
> I used 4.2 to model the inductive loaded BOG idea and found very small
> (0.2") differences in BOG height (0.2 to 2") made large differences in
> pattern and gain, even reversing the pattern.  Conclusion - not
> practical, even though the "best" model was neat.
>
> Given the posts re NEC doesn't work at small above ground distances: the
> N6LF data in his "lost" BOG performance article and some modeling I did
> at small distances above and below ground convinced me that 4.2 is very
> good at the small distances.  Of course the conductor can't be in the
> ground plane.  The mathematics may blow up then but mother nature is
> continuous.
>
> Then I was surprised by the difference a mild freeze made in my DC
> ground field resistance (the freeze was just a few inches).  The change
> had to be in the earth resistance of the rod interconnections (bare #2 &
> #6), not the 8 & 10ft long rods.  Ice has much lower conductivity than
> water so even a frost, freezing the grass, might have a noticeable
> effect on a BOG.
>
> As the N6LF work shows and perhaps this is another example of "you get
> what you pay for" - ie BOGs are small, easy, and useful, but much more
> sensitive to the environment than elevated Beverages.
>
> Bing found this cool essentially "Engineer's Handbook for AM Antennas".
> http://crawfordbroadcasting.com/Eng_Files/AM%20Antenna%20Systems.pdf
>
> Included is this:
> "14.2     Antenna Resistance
> Periodic checks should be made of the resistance at the common point or
> non-directional antenna base. The frequency of this resistance check
> will vary from station to station, and will probably be determined by
> the nature of the environment around the antenna. Areas with poor ground
> conductivity or sites with poor or deteriorated ground systems may see a
> considerable shift in base or common point resistance with changes in
> the amount of water in the soil. Similar (and sometimes more dramatic)
> changes can occur when the ground freezes."
>
> Grant KZ1W
>
> On 3/24/2020 19:46, Michael Tope wrote:
> > Dan,
> >
> > Reading that QRZ.com thread you linked to and reflecting on K4SAV's
> > measurements and how much they appeared to vary over a fairly short
> > period of time, got me wondering about what could be responsible. I am
> > fairly certain that at least part of the radial system for my 160 meter
> > vertical is over the drain field of the septic system for my cabin. Who
> > knows, maybe a properly timed toilet flush could make the difference
> > between working a new one and coming up empty :-)
> >
> > Okay, the toilet aided pileup busting is a stretch, but seriously, I am
> > not sure if I can think of any better explanations than septic discharge
> > for why ground characteristics could change that quickly. I do know in
> > some cases there can be water table very close to the surface, but I
> > don't know how much that typically varies over the short term. My only
> > other thought is that maybe in K4SAV's case the lay of the grass under
> > the wire is changing slightly over time and the impedance is
> > super-sensitive to the gap between the antenna and the ground (although
> > I think K4SAV said that his grass was all dead).
> >
> > 73, Mike W4EF..........
> >
> > On 3/23/2020 10:18 PM, Dan Maguire via Topband wrote:
> >> Grant wrote:
> >>>>> If you really want to know the parameters, see antennasbyn6lf.com
> >>>>> as Rudy describes techniques for ground RF properties measuring.
> >> Turns out that very subject was being kicked around on a recent
> >> qrz.com thread:
> >>
> >>
> https://forums.qrz.com/index.php?threads/indirect-measurement-of-ground-constants-with-a-dipole.696955/
> >>
> >>
> >> Dan, AC6LA
> >> _________________
> >> Searchable Archives: http://www.contesting.com/_topband - Topband
> >> Reflector
> >
> > _________________
> > Searchable Archives: http://www.contesting.com/_topband - Topband
> Reflector
> _________________
> Searchable Archives: http://www.contesting.com/_topband - Topband
> Reflector
>