All,

This question is: How can the ground system for a vertical be improved?

The RF ground calculates to about 52 Ohms. It consists of 16 radials
varying in length from 24 to 45 feet. Four are buried and 12 are on
ground. The location is the Pacific Northwest part way up a hillside.
The surface is grass, the subsurface is clay. The value varies a little
with rain, it has measured as low as 46 to 48 Ohms.

Just as additional information, the antenna is a loaded vertical,
overall about 4 meters. The coil is about 35% from the base. The antenna
is raised about 1 meter. I am aware the configuration is not optimum and
a top hat would help, those changes will be made independently.

Several articles indicate this radial system should yield well under 20
Ohms, some estimates are under 8 Ohms. At this point my first target is
under 20 Ohms. Little that I do seems to affect it. My question is:

What am I missing?

The problem is the value does not appear to change very much. I started
with 4 radials and have been adding them periodically. Those changes do
not seem to have a measurable effect. (However I must admit that earlier
numbers not as good as the current measurement. I believe the current
measurements are better then 10%.)

Thanks - Dan

dansawyeror wrote in
:

Dan,

You refer to calculation of ground resistance and later to measurements.
It would help if you explained what you measured, the frequency of
measurement, and how you measured it, then how you calculated your
results.

Perhaps the regimen of that might help you to an answer.

Owen


All,

This question is: How can the ground system for a vertical be improved?

The RF ground calculates to about 52 Ohms. It consists of 16 radials
varying in length from 24 to 45 feet. Four are buried and 12 are on
ground. The location is the Pacific Northwest part way up a hillside.
The surface is grass, the subsurface is clay. The value varies a little
with rain, it has measured as low as 46 to 48 Ohms.

Just as additional information, the antenna is a loaded vertical,
overall about 4 meters. The coil is about 35% from the base. The
antenna
is raised about 1 meter. I am aware the configuration is not optimum
and
a top hat would help, those changes will be made independently.

Several articles indicate this radial system should yield well under 20
Ohms, some estimates are under 8 Ohms. At this point my first target is
under 20 Ohms. Little that I do seems to affect it. My question is:

What am I missing?

The problem is the value does not appear to change very much. I started
with 4 radials and have been adding them periodically. Those changes do
not seem to have a measurable effect. (However I must admit that
earlier
numbers not as good as the current measurement. I believe the current
measurements are better then 10%.)

Thanks - Dan

Owen,

I did not want to concentrate on that; however it is a fair question.
The frequency is 3970 kc and the methodology is based on use of an hp
network analyzer. The analyzer has both polar and Cartesian displays.
The outputs were cross checked.f
The feed line was calibrated using a 25 Ohm termination to 180
degrees. The return measured about -9.4db (close to what is predicted
with line loss). This was checked against an open circuit confirmed the
calibration by a reading 1 degree and 1db. (very close to what is
predicted)

For the actual measurement the return loss was measured at just over
-26db at 0 degrees. (no feed-point network) Three models were used to
calculate the input resistance of the antenna. They all predicted an
antenna resistance of about 3 to 4 Ohms. (the suggested input network is
120 pf)

To focus on the question is: Why is the ground resistance so high? It is
not important at this stage to determine its precise value. The point is
it is high enough to cause a return of 0 degrees. This puts the 'system'
at over 50 degrees. Even if the antenna were 6 to 8 Ohms the ground loss
would be at least 42 to 44 Ohms.

- Dan

Owen Duffy wrote:
dansawyeror wrote in
:

Dan,

You refer to calculation of ground resistance and later to measurements.
It would help if you explained what you measured, the frequency of
measurement, and how you measured it, then how you calculated your
results.

Perhaps the regimen of that might help you to an answer.

Owen


All,

This question is: How can the ground system for a vertical be improved?

The RF ground calculates to about 52 Ohms. It consists of 16 radials
varying in length from 24 to 45 feet. Four are buried and 12 are on
ground. The location is the Pacific Northwest part way up a hillside.
The surface is grass, the subsurface is clay. The value varies a little
with rain, it has measured as low as 46 to 48 Ohms.

Just as additional information, the antenna is a loaded vertical,
overall about 4 meters. The coil is about 35% from the base. The
antenna
is raised about 1 meter. I am aware the configuration is not optimum
and
a top hat would help, those changes will be made independently.

Several articles indicate this radial system should yield well under 20
Ohms, some estimates are under 8 Ohms. At this point my first target is
under 20 Ohms. Little that I do seems to affect it. My question is:

What am I missing?

The problem is the value does not appear to change very much. I started
with 4 radials and have been adding them periodically. Those changes do
not seem to have a measurable effect. (However I must admit that
earlier
numbers not as good as the current measurement. I believe the current
measurements are better then 10%.)

Thanks - Dan

dansawyeror wrote in
:

....

I think a summary is that at 3970KHz, the feedpoint Z looks like about 45
+j0 and you reckon the radiation resistance should be around 4+j0,
suggesting the earth system contributes around 40 ohms of resistance.
Observations at a single frequency provide a limited view of what might
be happening.

Elevated radials should exhibit a clear resonance, and will offer the
lowest impedance at resonance.

Buried radials will not exhibit such a clear resonance in lossy soil, and
using your wire to form more short radials might give better performance
than few longer radials.

Radials lying on the ground are likely to be somewhere in between. You
may observe resonance, and in that case the ground system impedance will
be optimised by "tuning" those radials (which will probably be a good bit
shorter than formula length for free space radials).

Owen

Owen Duffy wrote in
:

dansawyeror wrote in
:

...

I think a summary is that at 3970KHz, the feedpoint Z looks like about
45 +j0 and you reckon the radiation resistance should be around 4+j0,
suggesting the earth system contributes around 40 ohms of resistance.
Observations at a single frequency provide a limited view of what
might be happening.

Dan,

I asked your for the details of your antenna and measurements, and how
you did your calculations, but I am still left wondering how you have
what appears to be a purely resistive feedpoint impedance and a radiation
resistance of 4 ohms. The second implies a short vertical, and if that is
the case, the first implies some form of loading... but you didn't
mention loading of any kind. Loading, if you have used it, may introduce
an equivalent series resistance at the feedpoint.

Once again, a dansawyer problems leaves us guessing to fill in the
missing dots before attempting to joint them up to make a picture.

Often, solving a problem is about being able to draw the picture, once
the picture is draw, the answer becomes trivial.

Owen

I asked your for the details of your antenna and measurements, and how
you did your calculations, but I am still left wondering how you have
what appears to be a purely resistive feedpoint impedance and a radiation
resistance of 4 ohms. The second implies a short vertical, and if that is
the case, the first implies some form of loading... but you didn't
mention loading of any kind. Loading, if you have used it, may introduce
an equivalent series resistance at the feedpoint.

Once again, a dansawyer problems leaves us guessing to fill in the
missing dots before attempting to joint them up to make a picture.

Often, solving a problem is about being able to draw the picture, once
the picture is draw, the answer becomes trivial.

Owen

I was also confused by the base of the antenna being 1 meter above
ground, and the radials lying on the ground, or buried. If the base
of the antenna is 1 meter high, then any connection to the radials
is part of the radiation system. Why would you feed the antenna
1 meter up, and not at the base? The antenna is therefore
a ground mounted 5 meter vertical.

NEC predicts an input impedance of 4 - j 1300 with 36 ten meter
radials 1" below an average ground. Loading coils will of course
add to the input impedance.

The measured data are suspect. It would be interesting to know
the length, and type, of coax connecting to the network analyzer.
The return loss of 25 ohms at the end of a piece of coax will be
9.5 dB. Unless the coax is cut to a precise known length it
is unlikely that the phase angle of the return loss will be zero.


Frank.

Owen,

The antenna is a short loaded vertical. The base is about 1.1 meters
long and 90mm in diameter, the coil is about 160mm long and about 80mm
in diameter, the top is 3 meters. The coil wire is 12 gage and the
spacing is about .5. As a model cross check, the impedance of the coil
measures about 60 uH.

- Dan

Owen Duffy wrote:
Owen Duffy wrote in
:

dansawyeror wrote in
:

...

I think a summary is that at 3970KHz, the feedpoint Z looks like about
45 +j0 and you reckon the radiation resistance should be around 4+j0,
suggesting the earth system contributes around 40 ohms of resistance.
Observations at a single frequency provide a limited view of what
might be happening.

Dan,

I asked your for the details of your antenna and measurements, and how
you did your calculations, but I am still left wondering how you have
what appears to be a purely resistive feedpoint impedance and a radiation
resistance of 4 ohms. The second implies a short vertical, and if that is
the case, the first implies some form of loading... but you didn't
mention loading of any kind. Loading, if you have used it, may introduce
an equivalent series resistance at the feedpoint.

Once again, a dansawyer problems leaves us guessing to fill in the
missing dots before attempting to joint them up to make a picture.

Often, solving a problem is about being able to draw the picture, once
the picture is draw, the answer becomes trivial.

Owen

On Sun, 08 Apr 2007 20:20:36 -0700, dansawyeror
wrote:

To focus on the question is: Why is the ground resistance so high?

Hi Dan,

Focus on Why? Why? This is something that you have absolutely no
control over without a huge investment in new dirt several 10s of
meter deep out to at least as far as the antenna is tall.

You want less resistance? then dig a hole and fill it with quartz
sand. Build a dune to drive the loss down further.

A suitable substitute is to lay out a ground field (radial wires). The
more the better. Here is the Why? you should be asking (Why more
wires?).

The path through the earth is shortened between adjacent ground wires.
Less path, less loss. The earth current travels not IN toward the
center as the radials do. The earth current travels ACROSS or
circumferentially towards the radial wires. This is due to the phase
lag between the induced earth current and the radial current. The
greater the distance between radials, the more path loss from the
average distance between the radials, to the radials.

As this is very difficult to treat in words alone, it is undoubtedly
confusing in the description above.

73's
Richard Clark, KB7QHC

Richard Clark wrote:

The path through the earth is shortened between adjacent ground wires.
Less path, less loss. The earth current travels not IN toward the
center as the radials do. The earth current travels ACROSS or
circumferentially towards the radial wires. This is due to the phase
lag between the induced earth current and the radial current. The
greater the distance between radials, the more path loss from the
average distance between the radials, to the radials.

As this is very difficult to treat in words alone, it is undoubtedly
confusing in the description above.

Indeed it is. Can you point me to a reference where I can get a more
detailed explanation of this circumferential current and its cause?

Roy Lewallen, W7EL

Well, I can't explain why lagging or leading current and voltage would
change the physical direction of propagation of the EM wave front
through the ground... But I do know that Dan does not have enough
radials.. He needs at least another 16 and better yet would be in the
range of 50 total...

denny / k8do