All,

I have been trying to determine the operation of a loaded vertical on 75 meters
and over a less then perfect ground. It is inconvenient to measure and change do
its location.

After many attempts to solve this I have struck upon a solution to 'get close'.
I have decided to build a model that works on 2 meters. This will be small
enough to construct in the shack and to experiment with various 'grounds'.

What are the pit falls in doing this?

Thanks Dan kb0qil

On Wed, 30 Nov 2005 21:46:22 -0800, dansawyeror
wrote:
After many attempts to solve this I have struck upon a solution to 'get close'.
I have decided to build a model that works on 2 meters. This will be small
enough to construct in the shack and to experiment with various 'grounds'.

Hi Dan,

One will be in the size of the components. If you use, say, 1/4 inch
tubing, then are you prepared to erect the full-scale with 10 inch
equivalents? Same goes for that coil load which is going to suffer
enormously with skin effect at 2M - unless you are going to build a
large coil at 75M.

Ground characteristics are not the same for 75M as they are at 2M.

Umm, let's see, the radiator and radials, the loading coil, ground -
have we left anything out?

73's
Richard Clark, KB7QHC

To make an exact scale model of an antenna, you not only have to scale
all physical dimensions, but also the conductivity - the model's
conductivity has to be freal/fmodel times the real conductivity.
(Dielectric constant and permeability remain the same in the model as in
the real antenna.) So in order to test the efficiency of various ground
systems, you'll have to bury radial wires in model "dirt" which is
several times more conductive than real dirt, while having the same
dielectric constant as real dirt. (For example, if the real antenna is
for 7 MHz, the model "dirt" will have to be about 20 times as conductive
as real dirt but with the same dielectric constant.) Technically, you
also have to do the same with the antenna conductor, but hopefully the
conductor loss will be small enough that you won't have to worry about
it. If the loading coil loss is significant, you'll have to construct a
model coil with a scaled inductance but the same resistance (that is,
the same Q) as the real one.

The pattern of a vertical antenna is heavily influenced by the
conductivity of the ground extending several wavelengths from the
antenna. So if you want to realistically evaluate the pattern, you'll
need to extend your highly conductive model "dirt" to at least that
distance.

Then there's the problem of measurement. Nearly everyone seriously
overestimates his ability to accurately measure RF quantities such as
impedance and field strength at HF. It's even more difficult at 2
meters. If you even want to attempt this, you should begin with models
of several antennas of known characteristics, such as a quarter
wavelength vertical and a resonant near-quarter wave vertical. See how
closely your measurements agree with theoretical results. If and when
you can get close agreement, then you have some chance on trusting
measurements of other antennas. But even then it's not a certainty.

Those are some of the pitfalls.

Roy Lewallen, W7EL

dansawyeror wrote:
All,

I have been trying to determine the operation of a loaded vertical on 75
meters and over a less then perfect ground. It is inconvenient to
measure and change do its location.

After many attempts to solve this I have struck upon a solution to 'get
close'. I have decided to build a model that works on 2 meters. This
will be small enough to construct in the shack and to experiment with
various 'grounds'.

What are the pit falls in doing this?

Thanks Dan kb0qil

Roy,

Thank you. My first objective is hopefully relatively simple. I would like to
determine resonance and measure Z. In order to do that I will use an HP 8405A
and directional couplers. Your suggestion is to practice that on known antennas.
That makes sense. Given that I have relatively good generators and can measure
forward and reflected signals at the antenna, is that sufficient to measure
model impedance?

Dan

Roy Lewallen wrote:
To make an exact scale model of an antenna, you not only have to scale
all physical dimensions, but also the conductivity - the model's
conductivity has to be freal/fmodel times the real conductivity.
(Dielectric constant and permeability remain the same in the model as in
the real antenna.) So in order to test the efficiency of various ground
systems, you'll have to bury radial wires in model "dirt" which is
several times more conductive than real dirt, while having the same
dielectric constant as real dirt. (For example, if the real antenna is
for 7 MHz, the model "dirt" will have to be about 20 times as conductive
as real dirt but with the same dielectric constant.) Technically, you
also have to do the same with the antenna conductor, but hopefully the
conductor loss will be small enough that you won't have to worry about
it. If the loading coil loss is significant, you'll have to construct a
model coil with a scaled inductance but the same resistance (that is,
the same Q) as the real one.

The pattern of a vertical antenna is heavily influenced by the
conductivity of the ground extending several wavelengths from the
antenna. So if you want to realistically evaluate the pattern, you'll
need to extend your highly conductive model "dirt" to at least that
distance.

Then there's the problem of measurement. Nearly everyone seriously
overestimates his ability to accurately measure RF quantities such as
impedance and field strength at HF. It's even more difficult at 2
meters. If you even want to attempt this, you should begin with models
of several antennas of known characteristics, such as a quarter
wavelength vertical and a resonant near-quarter wave vertical. See how
closely your measurements agree with theoretical results. If and when
you can get close agreement, then you have some chance on trusting
measurements of other antennas. But even then it's not a certainty.

Those are some of the pitfalls.

Roy Lewallen, W7EL

dansawyeror wrote:

All,

I have been trying to determine the operation of a loaded vertical on
75 meters and over a less then perfect ground. It is inconvenient to
measure and change do its location.

After many attempts to solve this I have struck upon a solution to
'get close'. I have decided to build a model that works on 2 meters.
This will be small enough to construct in the shack and to experiment
with various 'grounds'.

What are the pit falls in doing this?

Thanks Dan kb0qil

Thank you. I am not planning to scale the model, rather I am planning to learn
how to measure. My first goal is to build an antenna with a relatively low
radiation resistance and measure it.

Where can I learn about the difference in ground characteristics between 75m and
2m? William Orr wrote that a cross mesh under an HF vertical approximated a
perfect ground. It should be relatively easy to do the same under a 2m model. I
assume that metal reference ground would react the same.

Thanks - Dan

Richard Clark wrote:
On Wed, 30 Nov 2005 21:46:22 -0800, dansawyeror
wrote:

After many attempts to solve this I have struck upon a solution to 'get close'.
I have decided to build a model that works on 2 meters. This will be small
enough to construct in the shack and to experiment with various 'grounds'.


Hi Dan,

One will be in the size of the components. If you use, say, 1/4 inch
tubing, then are you prepared to erect the full-scale with 10 inch
equivalents? Same goes for that coil load which is going to suffer
enormously with skin effect at 2M - unless you are going to build a
large coil at 75M.

Ground characteristics are not the same for 75M as they are at 2M.

Umm, let's see, the radiator and radials, the loading coil, ground -
have we left anything out?

73's
Richard Clark, KB7QHC

dansawyeror wrote:
William Orr wrote that a cross mesh under an HF vertical
approximated a perfect ground.

That's only where the mesh is installed. Ground
reflections and attenuation extend beyond the
mesh area. How high above the mesh are you going
to position the 2m antenna?
--
73, Cecil, http://www.qsl.net/w5dxp

Cecil Moore wrote:
dansawyeror wrote:

William Orr wrote that a cross mesh under an HF vertical approximated
a perfect ground.



I've put up complete antenna systems that seem to approximate a perfect
ground .. some of them commercial products

On Thu, 01 Dec 2005 07:00:44 -0800, dansawyeror
wrote:

Where can I learn about the difference in ground characteristics between 75m and 2m?

Hi Dan,

Reggie's method of measuring earth conductivity in a coaxial tube
would work in these very short wavelengths (at HF it would be a joke
due to the necessity to survey to many meters of depth beneath the
system and out beyond it).

Aside from that, you would have to scrounge the web for references;
and they are actually quite numerous, but exceedingly tedious to sift
through. Basically, perform a search on soil characteristics for
ground penetrating RADAR. There is a lot of work that starts in VHF.
From this you might anticipate most references are very military in
nature. The problem of the sift is that many of these studies are
vertical niche research and may discard the paths of your interest.

Another method would be to reverse engineer it and find the soil
characteristics by discovery. That is model your construction, and
measure it both and compare results. The difference would probably be
the soil characteristic if you pay close attention to the details of
modeling and construction both. [By modeling, I mean in software.]

73's
Richard Clark, KB7QHC

dansawyeror wrote:
Roy,

Thank you. My first objective is hopefully relatively simple. I would
like to determine resonance and measure Z. In order to do that I will
use an HP 8405A and directional couplers. Your suggestion is to practice
that on known antennas. That makes sense. Given that I have relatively
good generators and can measure forward and reflected signals at the
antenna, is that sufficient to measure model impedance?

Provided that you can measure phase as well as amplitude, both with
adequate accuracy, and know how to establish a reference plane by either
calibration or calculation. It's not simple.

Roy Lewallen, W7EL

dansawyeror wrote:
Where can I learn about the difference in ground characteristics between
75m and 2m? . . .

If you want to model 75m ground, you need to begin with what you assume
the ground characteristics to be at 75m. Let's say you have average
ground, with conductivity of 5 mS/m and dielectric constant of 13. For
your model, you'll need something with conductivity of 5 * 146/3.8 ~ 192
mS/m conductivity and dielectric constant of 13. I have no idea where
you might find such a material.

William Orr wrote that a cross mesh under an HF vertical
approximated a perfect ground. It should be relatively easy to do the
same under a 2m model. I assume that metal reference ground would react
the same.

No problem approximating a perfect ground. If you have a perfect ground
under your 75 meter antenna, it won't be hard to make a decent model of
the perfect ground. A piece of screen or hardware cloth would do fine.
The trick will be making that perfect ground under your 75 meter
antenna. And unless you do, the model won't behave the same as the real
antenna.

Roy Lewallen, W7EL