"Denny" wrote
You need a flat top to pull the current node higher from the ground...
Our NDB at KHYX is less than 100 feet tall, has a series fed vertical
wire with a long, multiwire flat top and is easily copied from 80
miles away at night and 40 miles in the day...
______________

Your NDB might have been in a location with much better ground conductivity
than this one.

I was attempting to analyze the hardware described in the OP, and used
rather pessimistic assumptions in doing so.

But even then, the FCC MW propagation curves for this power, radiator
efficiency, frequency and assumed earth conductivity (2 mS/m) show a
groundwave field of about 35 µV/m at 40 miles.

I don't know if that would be called "easy copy" in the daytime using the
receivers intended for this application. Does anyone know?

I assumed an r-f ground loss of 25 ohms. That loss for an AM broadcast
station is around 2 ohms. Reducing the loss in the r-f ground would help
here, at the penalty of reducing the r-f bandwidth.

RF

"Frank" wrote
At 1000 m, between a height of zero meters, and 1000 m
the field strength is in the range of 5 mV/m peak. (i.e. including
ground wave). Not sure why the calculation does not agree
with Richard Fry's analysis, but may be due to the fact that
NEC computes ground losses for the surface wave.
____________

At the bottom of this post is a link to another analysis, this time using
NEC-2 with the input assumptions of my first "spreadsheet" approach.

It shows a field of 84.12 mV/m at 1 km for 1 kW of radiated power.
Adjusting that 1 km field that for the power reduction to 50 watts brings it
to 18.8 mV/m -- which is in close agreement with my spreadsheet value of
18.5 mV/m.

NEC-2 cannot deal with buried radials, but the 1 km NEC-2 field as
calculated here for a perfect ground can be plugged into the applicable FCC
propagation curves to show the groundwave field for a given distance,
frequency and conductivity, as I did in earlier post. Repeating those:

Field Strength Radius
0.500 mV/m 10.3 miles
0.250 mV/m 15.5 miles

In another post, Frank, you wrote "With 50 W input the peak E-field at 1000
m is 62.9 mV/m (44.5 mV/m RMS). At 24 km the E-field is 2.2 mV/m (1.5 mV/m
RMS), at ground level, and 2.0 mV/m (1.4 mV/m RMS) at 10,000 m elevation.
These results appear to be very close
to Richard Fry's analysis, though not sure why there is a 6 dB difference."

I think you were looking at my first post, where I guesstimated 6 dB ground
loss for a 24 km path, and showed 0.773 mV/m there.

The (much) more accurate FCC approach shows only 0.25 mV/m for a 24 km path
with 2 mS/m conductivity, and the difference between that and your 1.5 mV/m
is 15.6 dB -- rather significant.

I don't know for sure what explains all this, but it is interesting to
consider.

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

RF

Correcting myself in this clip::

I think you were looking at my first post, where I guesstimated 6 dB
ground loss for a 24 km path, and showed 0.773 mV/m there.

The number from my post for those conditions was around 0.35 mV/m.

The 0.773 mV/m value was the inverse distance field at 24 km.

Sorry.

RF

This is a 125-foot tower, fed at 90 feet, by a cable 2 feet from a face
of the tower. This makes a loop about 184 feet in perimeter. This
perimeter is less than 0.1 wavelength. Thus, we have almost equal
currents flowing in opposite directions at any two points in the loop
and they are diametrically opposite each other. If there were no
separation between the cable and the tower, there would be no radiation
resistance, As it is there is very little radiation resistance.

Radiation resistance can be increased by lengthening the loop or
widening the loop. As long as the loop`s perimeter is less than
1/2-wavelength, it will have an inductive reactance which may be tuned
out by the vacuum variable capacitor. Were the tower a 1/4-wavelength,
it would be self-resonant and require no tuning. It is much shorter than
that.

To boost radiation, I would suggest extending the feed-cable to the top
of the tower to get current up there and to raise the feed-loop
radiation resistance. 1/4-wave folded unipoles radiate almost the same
as open-circuit unipoles. Shorter antennas will suffer by comparison.

I`ve installed aircraft beacons at my company`s airstrips around the
world but they were series-fed and used the 3-strand top loading
referred to in this thread. They all worked fine on original fire-up.

Best regards, Richard Harrison, KB5WZI

"Richard Fry" wrote in message
...
"Frank" wrote
At 1000 m, between a height of zero meters, and 1000 m
the field strength is in the range of 5 mV/m peak. (i.e. including
ground wave). Not sure why the calculation does not agree
with Richard Fry's analysis, but may be due to the fact that
NEC computes ground losses for the surface wave.
____________

At the bottom of this post is a link to another analysis, this time using
NEC-2 with the input assumptions of my first "spreadsheet" approach.

It shows a field of 84.12 mV/m at 1 km for 1 kW of radiated power.
Adjusting that 1 km field that for the power reduction to 50 watts brings
it to 18.8 mV/m -- which is in close agreement with my spreadsheet value
of 18.5 mV/m.

NEC-2 cannot deal with buried radials, but the 1 km NEC-2 field as
calculated here for a perfect ground can be plugged into the applicable
FCC propagation curves to show the groundwave field for a given distance,
frequency and conductivity, as I did in earlier post. Repeating those:

Field Strength Radius
0.500 mV/m 10.3 miles
0.250 mV/m 15.5 miles

In another post, Frank, you wrote "With 50 W input the peak E-field at
1000 m is 62.9 mV/m (44.5 mV/m RMS). At 24 km the E-field is 2.2 mV/m
(1.5 mV/m RMS), at ground level, and 2.0 mV/m (1.4 mV/m RMS) at 10,000 m
elevation. These results appear to be very close
to Richard Fry's analysis, though not sure why there is a 6 dB
difference."

I think you were looking at my first post, where I guesstimated 6 dB
ground loss for a 24 km path, and showed 0.773 mV/m there.

The (much) more accurate FCC approach shows only 0.25 mV/m for a 24 km
path with 2 mS/m conductivity, and the difference between that and your
1.5 mV/m is 15.6 dB -- rather significant.

I don't know for sure what explains all this, but it is interesting to
consider.

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

RF

Calculating space wave plus surface wave at 24 km, at ground
level, NEC shows 0.79 mV/m peak (0.55 mV/m RMS). I used
a ground conductivity of 2 mS/m and relative permittivity 4.
The vertical tower was modelled with 3/8" dia. aluminum, neglecting
the actual lattice structure. I have not added the capacity hat.
I used eight 200 ft radials 3" below ground, also 3/8" aluminum
(6063-T832 alloy).

The input impedance is 6 - j 1008, and I am driving it with
4.1 kV peak (50 W). Don't understand why I am getting
different results than the FCC method at 0.25 mV/m. This
analysis does assume a lossless matching network, where
practical systems would show 3 or 4 dB of additional loss.

Frank

WA4SZE wrote:
"I have shunt fed a 120 foot high Rohn 25G tower."

The Rohn has a 1-ft face, so the h/d is about 120. That`s OK. What`s
wacko is a 23-degree tower over (8) 200-ft radials at 529 KHz. Ground
connection resistance is high and eating up all the signal.

Shunt - feeding is OK. Bill Orr and Stu Cowan give feed capacitors for
scalimg in "All About Vertical Antennas".

Brown, Lewis and Epstein would be disappointed with your radials. Shoot
for the broadcast practice of (120) evenly distributed around from the
tower base.

A short tower radiates almost as well as a 1/4-wave. but it has a very
low radiation resistance so can`t tolerate any loss resistance.

Kraus gives advice for Electrically Small Antennas in the 3rd edition of
"Antennas". Page 710 says:
"To increase the radiation efficiency requires an increase in the
radiation resistance Rr or a decrease in the loss resistance Rl or both.

The SWR of a dummy load usually looks fine, but radiation is just
incidental.

Best regards, Richard Harrison, KB5WZI

"Richard Harrison" wrote
Brown, Lewis and Epstein would be disappointed with your radials. Shoot
for the broadcast practice of (120) evenly distributed around from the
tower base.
_________

BL&E's 1937 measurements show (Fig 30) that a vertical monopole of 25 to 90
electrical degrees used with 113 buried radials each of 0.412 wavelength
produced a measured groundwave field within a few percent of the theoretical
maximum for such radiators over a perfect ground (notwithstanding that the
conductivity at their test site was around 4 mS/m).

In Fig 32 of that paper it can be seen that if the 113 radials are only
0.274-wavelengths long, then at the 25-degree electrical height of this Rohn
tower, the measured field was about 79% of theoretical field over a perfect
ground.
..
So it's not just the number of radials that is important, but also their
length.

The referenced figures are linked below, under the "fair use" provisions of
copyright law.

http://s62.photobucket.com/albums/h8...BLERadials.gif

RF

Here is the correct link to BL&E Figs 30 and 32.

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

On Sat, 7 Jul 2007 15:53:36 -0500, "Richard Fry" wrote:

The referenced figures are linked below, under the "fair use" provisions of
copyright law.

http://s62.photobucket.com/albums/h8...BLERadials.gif

RF

Richard, I just now tried to access your link above, but it says the file is no longer available. Do you have
any other source of this data? I worked with BL&E, so I'm kinda partial to having all the data from their 1936
experiment that I can find. I have their 1937 IRE paper.

Walt, W2DU

On Sat, 7 Jul 2007 16:02:11 -0500, "Richard Fry" wrote:

Here is the correct link to BL&E Figs 30 and 32.

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

Got it, Richard, but I see the two figs are simply from their 1937 IRE paper.

Thanks anyway.

Walt