Hi, Everybody,

In the process of modeling a vertical antenna (specifically, I am using
EZNEC 5.0) I am noticing an effect I did not expect which could be the
result of a modeling error on my part.

The antenna is a 34-foot vertical above (12) 34-foot radials, making it a
1/4 wave on 40 and a 1/2 wave on 20.

On 40, the antenna works as I expected; as the ground conductivity goes up,
the gain and efficiency of the antenna both increase, too.

But on 20, if I increase the ground conductivity from, say, 0.005 to 0.008
S/m, the max gain and efficiency *decrease*! This is counter-intuitive to
me.

Can anyone point to something I'm doing wrong?

Thanks,

Al W6LX

Al Lorona wrote:
Hi, Everybody,

In the process of modeling a vertical antenna (specifically, I am using
EZNEC 5.0) I am noticing an effect I did not expect which could be the
result of a modeling error on my part.

The antenna is a 34-foot vertical above (12) 34-foot radials, making it a
1/4 wave on 40 and a 1/2 wave on 20.

On 40, the antenna works as I expected; as the ground conductivity goes up,
the gain and efficiency of the antenna both increase, too.

But on 20, if I increase the ground conductivity from, say, 0.005 to 0.008
S/m, the max gain and efficiency *decrease*! This is counter-intuitive to
me.

Can anyone point to something I'm doing wrong?

Thanks,

Al W6LX

Ground loss is a sort of impedance matching problem. If you have
perfectly conducting ground, there is no ground loss. If you have
perfectly insulating ground, there is no ground loss. There's always
some ground conductivity in between those extremes at which the loss is
maximum. This value depends on the frequency among other things. Try a
wider range of conductivities and you'll find this point.

You should also be aware that if you have radials which are above but
close to the ground, half wavelength ones can be considerably less
efficient than quarter wavelength ones. One reason is that the points of
maximum current are out near the centers of the radials, where they
induce current into the lossy ground. When the radials are a quarter
wavelength or shorter, the current maxima are near the center, so their
fields nearly cancel. Another often-overlooked fact is that radials very
close to the ground are electrically considerably longer than when more
elevated. So radials which are a quarter wavelength in free space can
have their current maxima well out from the center which results in
lower efficiency.

Roy Lewallen, W7EL

Very cool, Roy. Thank you for this, this is good information that makes
sense to me. There is a lot in there to think about and apply to the
application for which I'm analyzing this antenna.

When you talk about the current maximum and loss of a radial, you bring up
another question. My 1/2 wave vertical is going to have a much higher
feedpoint impedance than a standard 1/4 wave vertical. Let's say it's
somewhere around 2000 ohms as compared to 40 or so ohms for a 1/4 wave. What
effect does that higher feedpoint impedance have on the radial system? Does
it relax the requirements for the number or radials or the length of them?
Does it mask the loss that you'd have normally? Or do the "rules" of adding
radials apply no matter what the vertical's input impedance?

Thanks very much.

Al W6LX







"Roy Lewallen" wrote in message
news4CdnVoHG_g30pXVnZ2dnUVZ_vudnZ2d@easystreeton line...
Al Lorona wrote:
Hi, Everybody,

In the process of modeling a vertical antenna (specifically, I am using
EZNEC 5.0) I am noticing an effect I did not expect which could be the
result of a modeling error on my part.

The antenna is a 34-foot vertical above (12) 34-foot radials, making it a
1/4 wave on 40 and a 1/2 wave on 20.

On 40, the antenna works as I expected; as the ground conductivity goes
up, the gain and efficiency of the antenna both increase, too.

But on 20, if I increase the ground conductivity from, say, 0.005 to
0.008 S/m, the max gain and efficiency *decrease*! This is
counter-intuitive to me.

Can anyone point to something I'm doing wrong?

Thanks,

Al W6LX

Ground loss is a sort of impedance matching problem. If you have perfectly
conducting ground, there is no ground loss. If you have perfectly
insulating ground, there is no ground loss. There's always some ground
conductivity in between those extremes at which the loss is maximum. This
value depends on the frequency among other things. Try a wider range of
conductivities and you'll find this point.

You should also be aware that if you have radials which are above but
close to the ground, half wavelength ones can be considerably less
efficient than quarter wavelength ones. One reason is that the points of
maximum current are out near the centers of the radials, where they induce
current into the lossy ground. When the radials are a quarter wavelength
or shorter, the current maxima are near the center, so their fields nearly
cancel. Another often-overlooked fact is that radials very close to the
ground are electrically considerably longer than when more elevated. So
radials which are a quarter wavelength in free space can have their
current maxima well out from the center which results in lower efficiency.

Roy Lewallen, W7EL

Roy,

This is blowing my mind. What you said about ground loss peaking at some
certain conductivity makes perfect sense but I never thought about it before
as applied to a vertical antenna system like this.

Sure enough, I found a maximum. This is wild. We always think, "The better
the ground, the better the antenna system," but it's not that simple. It's
funny to think that really terrible ground can have an advantage over pretty
good ground.

Next I'm going to play with the radial length so I can see the other effect
you described for me.

Regards,

Al W6LX









"Roy Lewallen" wrote in message
news4CdnVoHG_g30pXVnZ2dnUVZ_vudnZ2d@easystreeton line...
Al Lorona wrote:
Hi, Everybody,

In the process of modeling a vertical antenna (specifically, I am using
EZNEC 5.0) I am noticing an effect I did not expect which could be the
result of a modeling error on my part.

The antenna is a 34-foot vertical above (12) 34-foot radials, making it a
1/4 wave on 40 and a 1/2 wave on 20.

On 40, the antenna works as I expected; as the ground conductivity goes
up, the gain and efficiency of the antenna both increase, too.

But on 20, if I increase the ground conductivity from, say, 0.005 to
0.008 S/m, the max gain and efficiency *decrease*! This is
counter-intuitive to me.

Can anyone point to something I'm doing wrong?

Thanks,

Al W6LX

Ground loss is a sort of impedance matching problem. If you have perfectly
conducting ground, there is no ground loss. If you have perfectly
insulating ground, there is no ground loss. There's always some ground
conductivity in between those extremes at which the loss is maximum. This
value depends on the frequency among other things. Try a wider range of
conductivities and you'll find this point.

You should also be aware that if you have radials which are above but
close to the ground, half wavelength ones can be considerably less
efficient than quarter wavelength ones. One reason is that the points of
maximum current are out near the centers of the radials, where they induce
current into the lossy ground. When the radials are a quarter wavelength
or shorter, the current maxima are near the center, so their fields nearly
cancel. Another often-overlooked fact is that radials very close to the
ground are electrically considerably longer than when more elevated. So
radials which are a quarter wavelength in free space can have their
current maxima well out from the center which results in lower efficiency.

Roy Lewallen, W7EL

On Fri, 18 Apr 2008 10:26:50 -0700, "Al Lorona"
wrote:

This is blowing my mind. What you said about ground loss peaking at some
certain conductivity makes perfect sense but I never thought about it before
as applied to a vertical antenna system like this.

Sure enough, I found a maximum. This is wild. We always think, "The better
the ground, the better the antenna system," but it's not that simple. It's
funny to think that really terrible ground can have an advantage over pretty
good ground.

Hi Al,

There are two sides to this coin and you are only looking at one.

Let's take for example the vaunted superlatives that are often tied to
swamp land, or even planting your antenna farm offshore (ah that
seawater!). There is more loss there than if you planted your antenna
farm on a 100 foot sand dune (think glass).

So, why do so many eyes mist up with the opportunity of being in a
swap or in the ocean? Launch angle.

That same high (lossy) conductivity is also responsible for what is
called the Brewster angle when it comes to vertical antennas. The
higher the conductivity, the lower the Brewster angle, the lower the
launch angle possibilities, the longer out to the first skip - DX!

Longer radials will not lower that Brewster angle unless you extend
them out very far with a lot of them (think of carpeting the soil out
with a copper mesh for several 10s of wavelengths).

So, returning to your comment:
"The better
the ground, the better the antenna system,"
Better ground may subtract power in the near proximity, but better
ground may also boost relative power at low angles. So, you need to
define what is meant by "better" because this is a choice of opposite
alternatives.

73's
Richard Clark, KB7QHC

Al Lorona wrote:
It's funny to think that really terrible ground can
have an advantage over pretty good ground.

Free space is just about the most terrible "ground"
that one can imagine. :-)
--
73, Cecil http://www.w5dxp.com

On 18 abr, 19:02, "Al Lorona" wrote:
Very cool, Roy. Thank you for this, this is good information that makes
sense to me. There is a lot in there to think about and apply to the
application for which I'm analyzing this antenna.

When you talk about the current maximum and loss of a radial, you bring up
another question. My 1/2 wave vertical is going to have a much higher
feedpoint impedance than a standard 1/4 wave vertical. Let's say it's
somewhere around 2000 ohms as compared to 40 or so ohms for a 1/4 wave. What
effect does that higher feedpoint impedance have on the radial system? Does
it relax the requirements for the number or radials or the length of them?
Does it mask the loss that you'd have normally? Or do the "rules" of adding
radials apply no matter what the vertical's input impedance?

Thanks very much.

Al W6LX

"Roy Lewallen" wrote in message

news4CdnVoHG_g30pXVnZ2dnUVZ_vudnZ2d@easystreeton line...

Al Lorona wrote:
Hi, Everybody,

In the process of modeling a vertical antenna (specifically, I am using
EZNEC 5.0) I am noticing an effect I did not expect which could be the
result of a modeling error on my part.

The antenna is a 34-foot vertical above (12) 34-foot radials, making it a
1/4 wave on 40 and a 1/2 wave on 20.

On 40, the antenna works as I expected; as the ground conductivity goes
up, the gain and efficiency of the antenna both increase, too.

But on 20, if I increase the ground conductivity from, say, 0.005 to
0.008 S/m, the max gain and efficiency *decrease*! This is
counter-intuitive to me.

Can anyone point to something I'm doing wrong?

Thanks,

Al W6LX

Ground loss is a sort of impedance matching problem. If you have perfectly
conducting ground, there is no ground loss. If you have perfectly
insulating ground, there is no ground loss. There's always some ground
conductivity in between those extremes at which the loss is maximum. This
value depends on the frequency among other things. Try a wider range of
conductivities and you'll find this point.

You should also be aware that if you have radials which are above but
close to the ground, half wavelength ones can be considerably less
efficient than quarter wavelength ones. One reason is that the points of
maximum current are out near the centers of the radials, where they induce
current into the lossy ground. When the radials are a quarter wavelength
or shorter, the current maxima are near the center, so their fields nearly
cancel. Another often-overlooked fact is that radials very close to the
ground are electrically considerably longer than when more elevated. So
radials which are a quarter wavelength in free space can have their
current maxima well out from the center which results in lower efficiency.

Roy Lewallen, W7EL

Hi Al,

From my experience, and simulation, a halve wave vertical has less
stringent radial requirements. The current reduces with a factor of
about sqrt(2000/50)= 6 (16 dB).

The result is less loss in the vicinity (near field) zone of the
antenna. Far field maximum of radiation (elevation) will virtually not
change because of canceling effect of reradiated field from ground
below the (pseudo) Brewster angle. Because of the somewhat higher
radiation center with respect to a 1/4 wave, the elevation of maximum
radiation will be somewhat less. Off course when you have high ground
conductivity (for example fertilized wet soil), you will have maximum
radiation under lower elevation.

It is likely that in the end the produced field under relative low
elevation will be higher than that from a horizontal dipole.

Elevated radials give less ground loss. At high frequency, the
relative epsilon becomes important. A High dielectric constant and low
conductivity may result in less loss. At high frequency a larger part
of the current goes to the capacitance of the ground instead of the
loss resistance (in a parallel equivalent circuit). See it as having a
low loss dielectric under your antenna.

Hope this help you a bit.

Best regards,

Wim
PA3DJS
www.tetech.nl
please remove abc when replying.

"Cecil Moore" wrote in message
...
Al Lorona wrote:
It's funny to think that really terrible ground can have an advantage
over pretty good ground.

Free space is just about the most terrible "ground"
that one can imagine. :-)
--
73, Cecil http://www.w5dxp.com

So much disinformation by W8JI School of DC circuitry :-)

Modeling various configurations shows benefits of good ground, especially
for taller than 1/4 wave radiators.
Myth that half wave radiators do not need ground is spreading like snake oil
wild fire. They need it but "looking" for it further out, not just at the
base.
I will anytime trade good ground (mirror) for lossy (RF sponge) ground. Its
just where the radiator is "looking" for the mirror, taller one - further
out, enhancing signals at lower angles.
3/8 vertical with some 3/8 physical length radials start morphing into far
field.

Yuri, K3BU.us

There are two quite separate ways which ground affects a vertical
antenna's performance.

The first is loss due to current returning to the antenna base when the
antenna is grounded, or induced in the ground under an elevated radial
system. To minimize loss, you want as much of the current to flow
through radial wires as you can. The power loss is I^2 * R. For a given
power input, I is much lower for a half wave bottom fed vertical than a
quarter wave bottom fed vertical. So the loss due to the conducted or
induced current is much less, and you can get by with a much simpler
ground system with the half wave vertical and still have low loss.

This ground loss is usually the chief determining factor of a vertical's
efficiency.

The other effect of ground is that the field from the antenna reflects
from it some distance from the antenna. The reflected field adds to the
directly radiated field to form a net field which is different at each
elevation angle. This is a major factor in determining the antenna's
elevation pattern. The conductivity and permittivity (dielectric
constant) of the ground affect the magnitude and phase of the the
reflected field, so the pattern changes with ground quality. In general,
the more conductive the ground the better the low angle radiation.
However, you can't compensate for this factor when the ground is poor by
improving the ground system. The reason is that the reflection takes
place much farther from the antenna than nearly any ground system
extends. And low angle radiation, where the improvement is most needed,
reflects the greatest distance away. The only way to improve the
situation is to move the antenna to a location where the ground is
better, which usually isn't possible or practical.

Because of the two separate effects, the overall field strength might be
better or worse as the ground conductivity improves, and it might even
be better at some elevation angles and worse at others.

Roy Lewallen, W7EL

Yuri Blanarovich wrote:
"Cecil Moore" wrote in message
...
Al Lorona wrote:
It's funny to think that really terrible ground can have an advantage
over pretty good ground.
Free space is just about the most terrible "ground"
that one can imagine. :-)
--
73, Cecil http://www.w5dxp.com

So much disinformation by W8JI School of DC circuitry :-)

Modeling various configurations shows benefits of good ground, especially
for taller than 1/4 wave radiators.
Myth that half wave radiators do not need ground is spreading like snake oil
wild fire. They need it but "looking" for it further out, not just at the
base.
I will anytime trade good ground (mirror) for lossy (RF sponge) ground. Its
just where the radiator is "looking" for the mirror, taller one - further
out, enhancing signals at lower angles.
3/8 vertical with some 3/8 physical length radials start morphing into far
field.

Yuri, K3BU.us

Very nicely put, Roy. Although I "knew" this in the recesses of
memory, the refresher will stick with my memory more, now. Thanks.

In my case, I am considering the use of a vertical at a new residence
built on sand. Since I am not concerned about low angle radiation
characteristics, the Half Wave may be something to consider..... giving me
a fairly efficient vertical operation with some NVIS characteristics.



Ed K7AAT