NEC in its various forms is an excellent means to evaluate the
radiation from various structures.

But without due care it still can lead to incorrect conclusions about
system performance when propagation effects are present -- as in the
recent thread here about the elevation patterns of MW monopoles.

Here is another situation...

http://i62.photobucket.com/albums/h8...PatCompare.jpg

RF

"Richard Fry" wrote in message
...
NEC in its various forms is an excellent means to evaluate the
radiation from various structures.

But without due care it still can lead to incorrect conclusions about
system performance when propagation effects are present -- as in the
recent thread here about the elevation patterns of MW monopoles.

Here is another situation...

http://i62.photobucket.com/albums/h8...PatCompare.jpg

RF

The example above does not compute the surface wave.
The following card computes "Space Wave plus SurfaceWave":

CM 144 MHz Vertical Dipole
CE
GW 1 13 0 0 33.15 0 0 30 0.010417
GS 0 0 0.304800
GE 1 -1 0
GN 2 0 0 0 13.0000 0.0050
FR 0 1 0 0 147.3 0.1
EX 0 1 7 0 65.66981572 0
LD 5 0 0 0 3.801E7
RP 1 11 1 0000 0 90 1.00000 1.00000 50000
EN

In this example the vertical half wave dipole, with the base 30 ft above
an average ground, on 147.3 MHz, shows a field strength at ground
level of: 0.418 uV/m from 30 W into the antenna.

Frank

In this example the vertical half wave dipole, with the base 30 ft above
an average ground, on 147.3 MHz, shows a field strength at ground
level of: 0.418 uV/m from 30 W into the antenna.

Frank

And, obviously, at 50 km.

Richard Fry wrote:
NEC in its various forms is an excellent means to evaluate the
radiation from various structures.

But without due care it still can lead to incorrect conclusions about
system performance when propagation effects are present -- as in the
recent thread here about the elevation patterns of MW monopoles.

Here is another situation...

http://i62.photobucket.com/albums/h8...PatCompare.jpg

RF

Your statement is one of the most correct I have ever seen offered in
this group.

However, it, NEC, still remains one of the best available "antenna
prediction" offering of software obtainable to "normal" people ... and
provides some sort of an income for those who write the GUIs' who feed
the text file to it ...

Regards,
JS

On Dec 22, 11:13*am, "Frank" wrote:
In this example the vertical half wave dipole, with the base 30 ft above
an average ground, on 147.3 MHz, shows a field strength at ground
level of: 0.418 uV/m from 30 W into the antenna.

And, obviously, at 50 km.
________________

Here is another method (Longley-Rice) for calculating the field
intensity produced at the receive site by your model. But the NEC
approach is less accurate than L-R for long path lengths (due to earth
curvature), and for specific terrain contours.

In your model the path loss calculated using L-R is about 68.8 dB more
than the free space loss. The peak, free space field produced by a
1/2-wave, linear dipole radiating 30 watts over a 50 km path is about
770 uV/m. This voltage reduction of 68.8 dB is a field multiplier of
about 0.00036, so the 770 uV/m field is reduced to about 0.28 uV/m --
a bit less than your NEC model predicts. Agreement probably would be
better over shorter paths (as long as no specific terrain profile
needed to be applied), and worse for longer paths.

In the L-R example I set the path over the middle of Lake Michigan in
order to get a smooth earth contour, such as used in NEC models.

This all just illustrates that analyses made using NEC and any other
method need to consider the limits inherent in their algorithms with
respect to the physical reality being analyzed.

http://i62.photobucket.com/albums/h8...strialPath.gif

RF

Richard Fry wrote:
On Dec 22, 11:13 am, "Frank" wrote:
In this example the vertical half wave dipole, with the base 30 ft above
an average ground, on 147.3 MHz, shows a field strength at ground
level of: 0.418 uV/m from 30 W into the antenna.
And, obviously, at 50 km.
________________

Here is another method (Longley-Rice) for calculating the field
intensity produced at the receive site by your model. But the NEC
approach is less accurate than L-R for long path lengths (due to earth
curvature), and for specific terrain contours.



Once you get away from the near field, there's tons of models and
modeling approaches available, depending on the kind of path you're
interested in, and what you're looking to find out. For instance,
ioncap and its ilk (VOACAP,etc.) model skywave paths in a statistical
sense. Other models do raytracing for a more "point solution" type
model. Yet others are good for things like forests or terrain.

Since nobody has a full up computed electromagnetics finite model of
everything at fine resolution, all those models basically trade off
computational resources against some approximations. Whether it's
approximating the earth as flat surface of a uniform dielectric (NEC)
or whatever..

Richard Fry wrote:

This all just illustrates that analyses made using NEC and any other
method need to consider the limits inherent in their algorithms with
respect to the physical reality being analyzed.

Absolutely true. http://radiomagonline.com/fcc/radio_fcc_clamps_down/.

Roy Lewallen, W7EL

"Richard Fry" wrote in message
...
On Dec 22, 11:13 am, "Frank" wrote:
In this example the vertical half wave dipole, with the base 30 ft above
an average ground, on 147.3 MHz, shows a field strength at ground
level of: 0.418 uV/m from 30 W into the antenna.

And, obviously, at 50 km.
________________

Here is another method (Longley-Rice) for calculating the field
intensity produced at the receive site by your model. But the NEC
approach is less accurate than L-R for long path lengths (due to earth
curvature), and for specific terrain contours.

In your model the path loss calculated using L-R is about 68.8 dB more
than the free space loss. The peak, free space field produced by a
1/2-wave, linear dipole radiating 30 watts over a 50 km path is about
770 uV/m. This voltage reduction of 68.8 dB is a field multiplier of
about 0.00036, so the 770 uV/m field is reduced to about 0.28 uV/m --
a bit less than your NEC model predicts. Agreement probably would be
better over shorter paths (as long as no specific terrain profile
needed to be applied), and worse for longer paths.

In the L-R example I set the path over the middle of Lake Michigan in
order to get a smooth earth contour, such as used in NEC models.

This all just illustrates that analyses made using NEC and any other
method need to consider the limits inherent in their algorithms with
respect to the physical reality being analyzed.

http://i62.photobucket.com/albums/h8...strialPath.gif

RF

Interesting comparison between methods at VHF frequencies. For curiosity
I had done a comparison between the FCC predicted curves, for an AM
broadcast station on 1655 kHz, and NEC. It seems that at the lower
frequencies NEC has greater accuracy. Of course NEC was never intended
as a propagation tool, but still appears to be reasonably useful. I had cut
and pasted an Excel spread sheet below, so not sure if it will retain the
formatting when posted.

Frank


Field Strength Comparison at 1655 kHz..

Antenna Description: 45.3 m ground mounted monopole. 120 X 45.3 m
radials, 15 cm below ground. All conductors copper.
Input power 100 W

Source: http://www.fcc.gov/mb/audio/73184/index.html per 47 CFR
Sections 73.183 and 73.184
Nittany Scientific GNEC Version 1.1a. Ground parameters: Conductivity
5 mS/m, permittivity 13 (Average ground)
Field strength RMS V/m.

Distance FCC GNEC Difference Difference
(km) (mV/m) (mV/m) (%) (db)

0.10 950.000 960.000 1.0 -0.09
0.50 170.000 168.000 1.2 0.10
1.00 77.000 75.000 2.6 0.23
5.00 8.500 8.110 4.7 0.41
10.00 2.400 2.270 5.6 0.48
50.00 0.068 0.067 2.1 0.18
100.00 0.014 0.015 7.0 -0.61
200.00 0.002 0.004 62.1 -5.58

"Frank" wrote in message
news:EcU3l.65$z%.25@edtnps82...

"Richard Fry" wrote in message
...
On Dec 22, 11:13 am, "Frank" wrote:
In this example the vertical half wave dipole, with the base 30 ft
above
an average ground, on 147.3 MHz, shows a field strength at ground
level of: 0.418 uV/m from 30 W into the antenna.

And, obviously, at 50 km.
________________

Here is another method (Longley-Rice) for calculating the field
intensity produced at the receive site by your model. But the NEC
approach is less accurate than L-R for long path lengths (due to earth
curvature), and for specific terrain contours.

In your model the path loss calculated using L-R is about 68.8 dB more
than the free space loss. The peak, free space field produced by a
1/2-wave, linear dipole radiating 30 watts over a 50 km path is about
770 uV/m. This voltage reduction of 68.8 dB is a field multiplier of
about 0.00036, so the 770 uV/m field is reduced to about 0.28 uV/m --
a bit less than your NEC model predicts. Agreement probably would be
better over shorter paths (as long as no specific terrain profile
needed to be applied), and worse for longer paths.

In the L-R example I set the path over the middle of Lake Michigan in
order to get a smooth earth contour, such as used in NEC models.

This all just illustrates that analyses made using NEC and any other
method need to consider the limits inherent in their algorithms with
respect to the physical reality being analyzed.

http://i62.photobucket.com/albums/h8...strialPath.gif

RF

Interesting comparison between methods at VHF frequencies. For curiosity
I had done a comparison between the FCC predicted curves, for an AM
broadcast station on 1655 kHz, and NEC. It seems that at the lower
frequencies NEC has greater accuracy. Of course NEC was never intended
as a propagation tool, but still appears to be reasonably useful. I had
cut
and pasted an Excel spread sheet below, so not sure if it will retain the
formatting when posted.

Frank


Field Strength Comparison at 1655 kHz..

Antenna Description: 45.3 m ground mounted monopole. 120 X 45.3 m
radials, 15 cm below ground. All conductors copper.
Input power 100 W

Source: http://www.fcc.gov/mb/audio/73184/index.html per 47 CFR
Sections 73.183 and 73.184
Nittany Scientific GNEC Version 1.1a. Ground parameters:
Conductivity 5 mS/m, permittivity 13 (Average ground)
Field strength RMS V/m.

Distance FCC GNEC Difference Difference
(km) (mV/m) (mV/m) (%) (db)

0.10 950.000 960.000 1.0 -0.09
0.50 170.000 168.000 1.2 0.10
1.00 77.000 75.000 2.6 0.23
5.00 8.500 8.110 4.7 0.41
10.00 2.400 2.270 5.6 0.48
50.00 0.068 0.067 2.1 0.18
100.00 0.014 0.015 7.0 -0.61
200.00 0.002 0.004 62.1 -5.58

Rats, loused up the formatting. Here is a 2nd attempt.

Distance FCC GNEC Difference Difference
(km) (mV/m) (mV/m) (%) (db)

0.10 950.000 960.000 1.0 -0.09
0.50 170.000 168.000 1.2 0.10
1.00 77.000 75.000 2.6 0.23
5.00 8.500 8.110 4.7 0.41
10.00 2.400 2.270 5.6 0.48
50.00 0.068 0.067 2.1 0.18
100.00 0.014 0.015 7.0 -0.61
200.00 0.002 0.004 62.1 -5.58

Dear Group: What a delight it is to see a computer doing the calculations
for VHF propagation.

Almost fifty years ago, I led a team who measured field strengths in the 100
to 250 MHz range (FM and TV broadcast transmitters) to verify (qualify) the
propagation model. Of course, I used a slide rule and log tables to perform
the calculations and manually extracted path profiles from topo. maps. The
goal was to place confidence in the model for estimating expected
interference levels at a radio-astronomy site located in a valley. The
result from extensive filed measurements and data reduction was that we
could be confident in the model.

I recall also doing some comparisons of predicted and measured strengths
involving scattering (over quite long distances) in the VHF range with good
correlation.

IONCAP, and its predecessors and successors, I have used to good effect
for almost as many years.

In short, the developed propagation methods have been proven by me, and
many others, to provide reasonably small uncertainties. Of course, the
critical element is knowing which tool to use. That, I believe, is part of
the point brought forward by Richard Fry and others. But put yet another
way, any dam fool can (now) put numbers into a computer and get numbers back
out of the computer - experience and judgment is needed to have significance
accrue to the results of such calculations.

Central to all of the propagation models is the need to understand what
the antenna and its environment actually does. I am also delighted that
several of you are providing the education to the silent so that they do not
fall into the traps that are always present.

Warm regards and season's greetings, Mac N8TT
--
J. McLaughlin; Michigan, USA
Home:
"Richard Fry" wrote in message
...
On Dec 22, 11:13 am, "Frank" wrote:
In this example the vertical half wave dipole, with the base 30 ft above
an average ground, on 147.3 MHz, shows a field strength at ground
level of: 0.418 uV/m from 30 W into the antenna.

And, obviously, at 50 km.
________________

Here is another method (Longley-Rice) for calculating the field
intensity produced at the receive site by your model. But the NEC
approach is less accurate than L-R for long path lengths (due to earth
curvature), and for specific terrain contours.

In your model the path loss calculated using L-R is about 68.8 dB more
than the free space loss. The peak, free space field produced by a
1/2-wave, linear dipole radiating 30 watts over a 50 km path is about
770 uV/m. This voltage reduction of 68.8 dB is a field multiplier of
about 0.00036, so the 770 uV/m field is reduced to about 0.28 uV/m --
a bit less than your NEC model predicts. Agreement probably would be
better over shorter paths (as long as no specific terrain profile
needed to be applied), and worse for longer paths.

In the L-R example I set the path over the middle of Lake Michigan in
order to get a smooth earth contour, such as used in NEC models.

This all just illustrates that analyses made using NEC and any other
method need to consider the limits inherent in their algorithms with
respect to the physical reality being analyzed.

http://i62.photobucket.com/albums/h8...strialPath.gif

RF