Hi,

I am modeling a vertically oriented, vertically polarized loop for
160 meters. The loop will be supported at the top of a 130 foot tall
tower with a 6 foot standoff arm. The Bottom of the loop will be 10
feet from the ground. I was modeling the exact physical layout of the
loop with NEC2 without considering the tower structure in the model
at all.

The tower is Rohn 45G with non-conductive guy wires. Not wanting to
duplicate the exact tower model in NEC2 (by hand), I substituted in
essentially a wire with a diameter of 1 foot, using steel for the
conductor. I am modeling the antenna over 'real ground'. Putting in
this wire next to the antenna had a pretty big impact on the resonance
frequency that NEC2 reported. The pattern is slightly altered as
well.

My question is, 1) Am I taking too much of a shortcut to use a single
wire to represent the tower? I just wanted a *rough* idea if this
tower length (1/4 wavelength on 160) was going to have some impact on
the pattern in the real world. Resonance frequency shifted down in
frequency 35Khz in my model.

I've modeled and built loops that I've installed between trees and the
results have matched NEC2 very well, but I don't know what to make of
the support steel tower in this case - am I overstating its effect the
way I modeled it?

-Scott, WU2X

You've done the right thing. The tower features are a tiny fraction of a
wavelength, so the single wire substitution is fine.

To improve the accuracy of the model, use the equivalent radius of 0.421
foot for a triangular tower with one foot sides. This will likely make a
noticeable difference only if the tower is somewhere in the vicinity of
self resonance.

Roy Lewallen, W7EL

wrote:
Hi,

I am modeling a vertically oriented, vertically polarized loop for
160 meters. The loop will be supported at the top of a 130 foot tall
tower with a 6 foot standoff arm. The Bottom of the loop will be 10
feet from the ground. I was modeling the exact physical layout of the
loop with NEC2 without considering the tower structure in the model
at all.

The tower is Rohn 45G with non-conductive guy wires. Not wanting to
duplicate the exact tower model in NEC2 (by hand), I substituted in
essentially a wire with a diameter of 1 foot, using steel for the
conductor. I am modeling the antenna over 'real ground'. Putting in
this wire next to the antenna had a pretty big impact on the resonance
frequency that NEC2 reported. The pattern is slightly altered as
well.

My question is, 1) Am I taking too much of a shortcut to use a single
wire to represent the tower? I just wanted a *rough* idea if this
tower length (1/4 wavelength on 160) was going to have some impact on
the pattern in the real world. Resonance frequency shifted down in
frequency 35Khz in my model.

I've modeled and built loops that I've installed between trees and the
results have matched NEC2 very well, but I don't know what to make of
the support steel tower in this case - am I overstating its effect the
way I modeled it?

-Scott, WU2X

Thanks Roy!

I guess I was confused about the obvious connection the tower base has
to the ground. Will this have any addition affect?

-Scott, WU2X

On Dec 7, 3:34 pm, Roy Lewallen wrote:
You've done the right thing. The tower features are a tiny fraction of a
wavelength, so the single wire substitution is fine.

To improve the accuracy of the model, use the equivalent radius of 0.421
foot for a triangular tower with one foot sides. This will likely make a
noticeable difference only if the tower is somewhere in the vicinity of
self resonance.

Roy Lewallen, W7EL

wrote:
Hi,

I am modeling a vertically oriented, vertically polarized loop for
160 meters. The loop will be supported at the top of a 130 foot tall
tower with a 6 foot standoff arm. The Bottom of the loop will be 10
feet from the ground. I was modeling the exact physical layout of the
loop with NEC2 without considering the tower structure in the model
at all.

The tower is Rohn 45G with non-conductive guy wires. Not wanting to
duplicate the exact tower model in NEC2 (by hand), I substituted in
essentially a wire with a diameter of 1 foot, using steel for the
conductor. I am modeling the antenna over 'real ground'. Putting in
this wire next to the antenna had a pretty big impact on the resonance
frequency that NEC2 reported. The pattern is slightly altered as
well.

My question is, 1) Am I taking too much of a shortcut to use a single
wire to represent the tower? I just wanted a *rough* idea if this
tower length (1/4 wavelength on 160) was going to have some impact on
the pattern in the real world. Resonance frequency shifted down in
frequency 35Khz in my model.

I've modeled and built loops that I've installed between trees and the
results have matched NEC2 very well, but I don't know what to make of
the support steel tower in this case - am I overstating its effect the
way I modeled it?

-Scott, WU2X

wrote:
Thanks Roy!

I guess I was confused about the obvious connection the tower base has
to the ground. Will this have any addition affect?

Yes, the tower's connection to ground can have quite a profound effect
on how it interacts with the antenna. And unfortunately, that can be the
most challenging part to model.

If you put the bottom of the tower at Z = 0 when using any kind of
ground except perfect, you'll get the equivalent of a ground connection
with an unpredictable resistance (or an open circuit, depending on the
GE "card" parameter). About the best you can do is to use Sommerfeld
type ground and put some radials just above the ground. Of course, the
problem then becomes estimating the actual tower ground system
resistance and the resistance of the modeled radial system. You'll
probably end up trying a number of configurations which hopefully
include something representing reality, and assuming that the actual
system will behave somewhere within the range of the results you get.

There's a fair amount of information about ground system modeling in the
EZNEC manual, available in the free demo version. EZNEC's "High
accuracy" ground is the same as NEC Sommerfeld ground.

Roy Lewallen, W7EL