Hello all you antenna experts,
I have been a ham and an engineer for a long time, but I have never
delved into antenna theory. So, consider me a newbie for this
question.

I recently was helping a friend set up a crude antenna to connect to
his DTV converter box following the June 12th conversion to DTV. I
explained that a dipole was a simple antenna that could be used. He
was very interested in the subject so the conversation soon turned to
theory. I explained the 468/freq formula and eventually mentioned
EZNEC. I have never used EZNEC myself so I downloaded the demo
version and now have all of 24 hours experience with it. I started
with the included backyard 20 meter dipole. I was surprised that
there was no radiation toward the horizon. I figured that I was too
close to the ground so I changed the frequency to 491 MHz (RF TV
channel 17) and shortened the dipole accordingly. Still no zero
degree radiation. I raised the dipole to 100' - then 1000' - then
10000' - still no radiation at zero degree elevation. I then found
this in the help section:

-----------------------
Because the far field sky wave from a horizontally polarized source is
zero at a zero elevation angle for any ground type, and a vertically
polarized source produces zero sky wave for any finite-conductivity
ground, attempts to calculate a 2D pattern without the ground wave
component under these conditions will result in an error message.
-----------------------

Sure enough, changing to free space instead of a real ground changed
the pattern to what I would have expected. I would have thought being
many wavelengths above ground would be just as good as free space, but
EZNEC doesn't think so. Am I missing something? Does a horizontal
dipole really have a problem seeing a broadcast TV transmitter out on
the horizon? Thanks. ...Pat

On Sat, 20 Jun 2009 09:44:14 -0400, wrote:

Sure enough, changing to free space instead of a real ground changed
the pattern to what I would have expected. I would have thought being
many wavelengths above ground would be just as good as free space, but
EZNEC doesn't think so. Am I missing something? Does a horizontal
dipole really have a problem seeing a broadcast TV transmitter out on
the horizon? Thanks. ...Pat

Hi Pat,

EZNEC is presenting you with lobe characteristics that are at a very
great distance (and, yet, it has no sense of skip as that would be
tested in the domain of a propagation modeler). Zero degrees up to 10
degrees are minutely examined by those interested in skip, but this is
not a TV phenomenon (not to be confused with ducting which can offer
similar DX from great distances).

The horizontally polarized transmission has its E-Field parallel to
earth. Earth is a conductor (albeit a poor one, but in comparison to
free space, it is quite a short circuit). That E-Field's two
potentials are being laid across that conductor during the wave
propagation to that far point where EZNEC then sums up all field
contributions to present you with the lobe characteristic. It stands
to reason that at that great distance, the wave will have attenuated
considerably - hence the low value.

Removing the short circuit (going to free space) removes this
attenuation.

73's
Richard Clark, KB7QHC

The horizontally polarized transmission has its E-Field parallel to
earth. Earth is a conductor (albeit a poor one, but in comparison to
free space, it is quite a short circuit). That E-Field's two
potentials are being laid across that conductor during the wave
propagation to that far point where EZNEC then sums up all field
contributions to present you with the lobe characteristic. It stands
to reason that at that great distance, the wave will have attenuated
considerably - hence the low value.

Removing the short circuit (going to free space) removes this
attenuation.

73's
Richard Clark, KB7QHC

NEC will calculate "Space wave plus surface wave" if required.

Frank

Frank wrote:
The horizontally polarized transmission has its E-Field parallel to
earth. Earth is a conductor (albeit a poor one, but in comparison to
free space, it is quite a short circuit). That E-Field's two
potentials are being laid across that conductor during the wave
propagation to that far point where EZNEC then sums up all field
contributions to present you with the lobe characteristic. It stands
to reason that at that great distance, the wave will have attenuated
considerably - hence the low value.

Removing the short circuit (going to free space) removes this
attenuation.

73's
Richard Clark, KB7QHC

NEC will calculate "Space wave plus surface wave" if required.

Frank

Have you tried doing this calculation with a horizontally polarized VHF
antenna? What did you find?

Roy Lewallen, W7EL

On Sat, 20 Jun 2009 12:28:30 -0700, Roy Lewallen
wrote:

Frank wrote:
The horizontally polarized transmission has its E-Field parallel to
earth. Earth is a conductor (albeit a poor one, but in comparison to
free space, it is quite a short circuit). That E-Field's two
potentials are being laid across that conductor during the wave
propagation to that far point where EZNEC then sums up all field
contributions to present you with the lobe characteristic. It stands
to reason that at that great distance, the wave will have attenuated
considerably - hence the low value.

Removing the short circuit (going to free space) removes this
attenuation.

73's
Richard Clark, KB7QHC

NEC will calculate "Space wave plus surface wave" if required.

Frank

Have you tried doing this calculation with a horizontally polarized VHF
antenna? What did you find?

Roy Lewallen, W7EL
I'm not sure what you mean. EZNEC seems to say that a horizontally
polarized dipole seems to have zero gain (-99.99DBi) at zero degrees
elevation regardless of the frequency. So far, I have only tried 14
(the 20 meter example that came with EZNEC) Mhz, 491 Mhz (TV channel
17 center), and 527 MHz (TV channel 23 center). I switched to 527
because I can actually see a channel 23 transmitting antenna from my
window. For those who may not missed my original post, I find it hard
to believe a horizontal dipole tuned to the right frequency (near 1:1
SWR with 75 ohm source) would not be able to hear a signal coming from
zero degrees elevation. In the real world, there are all sorts of
reflections off of all sorts of things that will make it work, but is
it true that there should be no signal if everything was ideal?

Richard explained the attenuation of the E-field. That makes sense to
me, but doesn't really explain the other nulls at 6 degrees elevation
and every 6 degrees above that. There are strong positive lobes at 3
degrees and every 6 above that. The plot looks like a nice flower :-)
I would think that attentuation of the E-Field would explain zero
degrees, but as elevation increased, the attenuation would decrease.
The EZNEC plot looks more like it is showing additive and subtractive
combining of the signal. Another reply mentioned a different program
that calculated ground wave in addition to skywave. Maybe that is
what I am missing. I normally think of ground wave as why VLF, LF,
and MF signals travel further than line of sight, though. Does ground
wave have a significant effect at VHF/UHF?

I'm still confused,
Pat, N8CQV

wrote in message
...
On Sat, 20 Jun 2009 12:28:30 -0700, Roy Lewallen
wrote:

Frank wrote:
The horizontally polarized transmission has its E-Field parallel to
earth. Earth is a conductor (albeit a poor one, but in comparison to
free space, it is quite a short circuit). That E-Field's two
potentials are being laid across that conductor during the wave
propagation to that far point where EZNEC then sums up all field
contributions to present you with the lobe characteristic. It stands
to reason that at that great distance, the wave will have attenuated
considerably - hence the low value.

Removing the short circuit (going to free space) removes this
attenuation.

73's
Richard Clark, KB7QHC

NEC will calculate "Space wave plus surface wave" if required.

Frank

Have you tried doing this calculation with a horizontally polarized VHF
antenna? What did you find?

Roy Lewallen, W7EL
I'm not sure what you mean. EZNEC seems to say that a horizontally
polarized dipole seems to have zero gain (-99.99DBi) at zero degrees
elevation regardless of the frequency. So far, I have only tried 14
(the 20 meter example that came with EZNEC) Mhz, 491 Mhz (TV channel
17 center), and 527 MHz (TV channel 23 center). I switched to 527
because I can actually see a channel 23 transmitting antenna from my
window. For those who may not missed my original post, I find it hard
to believe a horizontal dipole tuned to the right frequency (near 1:1
SWR with 75 ohm source) would not be able to hear a signal coming from
zero degrees elevation. In the real world, there are all sorts of
reflections off of all sorts of things that will make it work, but is
it true that there should be no signal if everything was ideal?

Richard explained the attenuation of the E-field. That makes sense to
me, but doesn't really explain the other nulls at 6 degrees elevation
and every 6 degrees above that. There are strong positive lobes at 3
degrees and every 6 above that. The plot looks like a nice flower :-)
I would think that attentuation of the E-Field would explain zero
degrees, but as elevation increased, the attenuation would decrease.
The EZNEC plot looks more like it is showing additive and subtractive
combining of the signal. Another reply mentioned a different program
that calculated ground wave in addition to skywave. Maybe that is
what I am missing. I normally think of ground wave as why VLF, LF,
and MF signals travel further than line of sight, though. Does ground
wave have a significant effect at VHF/UHF?

I'm still confused,
Pat, N8CQV

what you are missing is the 'real world'. eznec is probably modeling over a
perfectly flat infinite surface. In the far field in a perfect world the
signal along the surface is a combination of ground wave and sky wave, the
ground wave decays rapidly with distance leaving the sky wave which will
always be very small along the surface. now remember, the frame of
reference is at ground level, not the antenna height, so zero degrees is
along the infinite flat surface. And there is nothing in there that models
where the other antenna is... it just creates a picture of how the strength
of the fields are at a given elevation/azimuth angle from the reference
point.

Now, in the real world... the ground is never level, even on the ocean where
it may look flat it curves down in every direction... so the horizon is not
at zero degrees for any antenna above ground level, over flat ground its
always below horizontal, on a hill or from a tower its even more negative,
and in a valley it can be way above horizontal. both of those cause those
predicted patterns to be changed a bit. Also, if the other antenna is not
at ground level then it is at some positive angle above horizontal... or if
its far enough away maybe a negative angle. You should also note that many
broadcast antennas for TV and FM are designed with a tilt to send the signal
down toward the ground, especially if they are on high hills or big towers,
they would have weak signals near the tower if they didn't tilt it down...
of course they don't care about beyond the horizon stuff anyway.

to confuse things even more on vhf/uhf frequencies signals are easily
reflected from hills, buildings, and other objects... they are also bent by
changes in air temperature and humidity. So in many cases a vhf/uhf signal
sent out toward the horizon may get bent down toward the ground and go
beyond the horizon by quite a distance, see 'tropospheric ducting' for more
info. hf signals of course get refracted back down toward the ground by the
ionosphere, so for very distant stations the arrival angle can still be
quite high, and almost never straight from the horizon.

wrote in message
...
On Sat, 20 Jun 2009 12:28:30 -0700, Roy Lewallen
wrote:

Frank wrote:
The horizontally polarized transmission has its E-Field parallel to
earth. Earth is a conductor (albeit a poor one, but in comparison to
free space, it is quite a short circuit). That E-Field's two
potentials are being laid across that conductor during the wave
propagation to that far point where EZNEC then sums up all field
contributions to present you with the lobe characteristic. It stands
to reason that at that great distance, the wave will have attenuated
considerably - hence the low value.

Removing the short circuit (going to free space) removes this
attenuation.

73's
Richard Clark, KB7QHC

NEC will calculate "Space wave plus surface wave" if required.

Frank

Have you tried doing this calculation with a horizontally polarized VHF
antenna? What did you find?

Roy Lewallen, W7EL
I'm not sure what you mean. EZNEC seems to say that a horizontally
polarized dipole seems to have zero gain (-99.99DBi) at zero degrees
elevation regardless of the frequency. So far, I have only tried 14
(the 20 meter example that came with EZNEC) Mhz, 491 Mhz (TV channel
17 center), and 527 MHz (TV channel 23 center). I switched to 527
because I can actually see a channel 23 transmitting antenna from my
window. For those who may not missed my original post, I find it hard
to believe a horizontal dipole tuned to the right frequency (near 1:1
SWR with 75 ohm source) would not be able to hear a signal coming from
zero degrees elevation. In the real world, there are all sorts of
reflections off of all sorts of things that will make it work, but is
it true that there should be no signal if everything was ideal?

Richard explained the attenuation of the E-field. That makes sense to
me, but doesn't really explain the other nulls at 6 degrees elevation
and every 6 degrees above that. There are strong positive lobes at 3
degrees and every 6 above that. The plot looks like a nice flower :-)
I would think that attentuation of the E-Field would explain zero
degrees, but as elevation increased, the attenuation would decrease.
The EZNEC plot looks more like it is showing additive and subtractive
combining of the signal. Another reply mentioned a different program
that calculated ground wave in addition to skywave. Maybe that is
what I am missing. I normally think of ground wave as why VLF, LF,
and MF signals travel further than line of sight, though. Does ground
wave have a significant effect at VHF/UHF?

I'm still confused,
Pat, N8CQV


p.s. you want to see more real world get one of the terrain analysis
programs that lets you place real antennas over real non-flat ground and see
where the signals really go.

On Sat, 20 Jun 2009 16:13:07 -0400, wrote:

I switched to 527
because I can actually see a channel 23 transmitting antenna from my
window.

Hi Pat,

If you think this experience contradicts EZNEC (or conventional
teachings which it concurs with), then the confusion comes from the
sense of "very far away." Truly, if you can see the Channel 23
transmitting antenna, then it is not that far away in the scheme of
things. It merely points out that you have not correctly modeled your
experience.

For those who may not missed my original post, I find it hard
to believe a horizontal dipole tuned to the right frequency (near 1:1
SWR with 75 ohm source) would not be able to hear a signal coming from
zero degrees elevation. In the real world, there are all sorts of
reflections off of all sorts of things that will make it work, but is
it true that there should be no signal if everything was ideal?

What is going to be a reflector to a source that is bore-sight with
the horizon?

Richard explained the attenuation of the E-field. That makes sense to
me, but doesn't really explain the other nulls at 6 degrees elevation
and every 6 degrees above that.

You didn't ask about that.

There are strong positive lobes at 3
degrees and every 6 above that. The plot looks like a nice flower :-)
I would think that attentuation of the E-Field would explain zero
degrees, but as elevation increased, the attenuation would decrease.

You have, again, lost sight of the meaning of "very far away."

The EZNEC plot looks more like it is showing additive and subtractive
combining of the signal.

It is. What you see is called the Fresnel Zone if you were line of
sight.

Here, even if you can "see" the Channel 23 transmitting antenna, then
its various reflections could add up to ZERO. This, again, confounds
expectation, but it is the experience of every mobile operator who
encounters "picket fencing."

Another reply mentioned a different program
that calculated ground wave in addition to skywave. Maybe that is
what I am missing. I normally think of ground wave as why VLF, LF,
and MF signals travel further than line of sight, though. Does ground
wave have a significant effect at VHF/UHF?

Yes, it is dead within a mile for Channel 23.

I'm still confused,

and so are a number of your respondents.

73's
Richard Clark, KB7QHC

wrote:
Frank wrote:

NEC will calculate "Space wave plus surface wave" if required.

Frank
Have you tried doing this calculation with a horizontally polarized VHF
antenna? What did you find?

Roy Lewallen, W7EL

I'm not sure what you mean.

My comment was directed to Frank. I'm hoping he'll try running the
calculation, or has already. He'll discover that:

1. A horizontally polarized wave does not propagate via a surface wave, and
2. The attenuation of surface waves increases with frequency. At VHF,
it's essentially zero at any useful distance.

Either of these is adequate to explain the result he'll see, or has
seen, that the result with surface wave is the same as the result
without surface wave, for this situation.

Richard and Dave have answered your question -- the reason for the zero
field at zero elevation angle EZNEC result is that the model ground is
perfectly flat and infinite in extent, and the observation point is
essentially at an infinite distance. This is useful for evaluating sky
wave propagation at long distances, but not for line-of-sight
propagation where Earth curvature can be a factor. Free space analysis
is better for this. Programs which calculate reflection coefficients for
various ground shapes aren't very useful at VHF and above, because local
features, especially in an urban environment, cause reflections that are
often much more significant than the ones calculated by the program. All
you can do with these programs in a situation like yours is to get a
general idea of what's happening.

EZNEC seems to say that a horizontally
polarized dipole seems to have zero gain (-99.99DBi) at zero degrees
elevation regardless of the frequency. So far, I have only tried 14
(the 20 meter example that came with EZNEC) Mhz, 491 Mhz (TV channel
17 center), and 527 MHz (TV channel 23 center). I switched to 527
because I can actually see a channel 23 transmitting antenna from my
window. For those who may not missed my original post, I find it hard
to believe a horizontal dipole tuned to the right frequency (near 1:1
SWR with 75 ohm source) would not be able to hear a signal coming from
zero degrees elevation. In the real world, there are all sorts of
reflections off of all sorts of things that will make it work, but is
it true that there should be no signal if everything was ideal?

If by ideal you mean that the ground is perfectly flat and infinite in
extent, yes.

Richard explained the attenuation of the E-field. That makes sense to
me, but doesn't really explain the other nulls at 6 degrees elevation
and every 6 degrees above that. There are strong positive lobes at 3
degrees and every 6 above that. The plot looks like a nice flower :-)
I would think that attentuation of the E-Field would explain zero
degrees, but as elevation increased, the attenuation would decrease.
The EZNEC plot looks more like it is showing additive and subtractive
combining of the signal.

That's exactly what EZNEC is doing. It calculates the sum of two "rays"
-- idealized straight line radiation from the antenna. One is the
"direct ray", which goes directly from the antenna to the observation
point. (Actually, the fields from all segments are individually
calculated and summed.) The second is the "reflected ray", which
reflects from the ground plane to arrive at the same distant point. Some
geometric investigation will show that the reflected ray at an elevation
angle of -x degrees will add to the direct ray at an elevation angle of
+x degrees at the distant point. The reflection coefficient of the
ground, which is determined by the frequency, wave polarization, and
ground conductivity and permittivity, is used to determine the amplitude
and phase of the reflected ray. The reflection coefficient of low angle
horizontally polarized waves is nearly -1 for any reasonable ground
characteristics, so the low angle pattern is nearly the same for a
horizontally polarized antenna over the Earth as for one over a
perfectly conducting ground. If you look the the distance from the
antenna to a distant point via the direct route and compare it to the
distance to the same point via the reflected route, you'll see that the
difference between the two routes (that is, the distances traversed by
the two "rays") monotonically increases with increasing elevation angle.
At zero elevation angle, the two distances are the same. Since the
reflected ray undergoes a phase reversal (as Richard explained),
expressed as a -1 reflection coefficient, the direct and reflected rays
cancel at the distant point, resulting in zero field strength. As the
elevation angle increases, the difference between the two two ray paths
increases, resulting in imperfect cancellation. At some elevation angle,
assuming the antenna is high enough off the ground, the difference in
path lengths will be exactly a half wavelength. At this angle, the two
rays will reinforce at the distant point. At a higher angle, they'll be
exactly one wavelength different, and the rays will cancel again. And so
forth. The path distance difference goes from zero at zero elevation
angle to a maximum straight up of twice the antenna height above ground,
so the number of "flower petals" you see depends on the height, in
wavelengths, of the antenna above ground.

Another reply mentioned a different program
that calculated ground wave in addition to skywave. Maybe that is
what I am missing. I normally think of ground wave as why VLF, LF,
and MF signals travel further than line of sight, though. Does ground
wave have a significant effect at VHF/UHF?

No, it doesn't. Which is why I posted the query to Frank.

I'm still confused,
Pat, N8CQV

Hope this helps. There's additional information in the _ARRL Antenna
Book_ and other sources which explains how the patterns of antennas are
affected by ground.

Roy Lewallen, W7EL

Roy Lewallen wrote:

Have you tried doing this comre it before.

I ran this with a 15' 9" boom 6m beam at 20 feet over real high accuracy
ground, which is at a typical height for a height for a home TV antenna.

The vertically polarized version had a peak of 11.5 dBi at 9 degrees
while the horizontal had 16.0 at 13 degrees. At 9 degrees the
horizontal still had 15.0 dBi, clearly the winner.

Given that this was very close to the ground in terms of wavelength, it
may not be a good example.

tom
K0TAR