> IMHO, the best length for triband (160, 80, 40) use
> is 500 to 600 ft. I gave up trying to maintain terminating
I use my phased 800 foot long Beverages (three wide with 350 foot
spacing) all the way up to 40 meters and higher.
The reason they work when they are so long is the far ends don't
carry any current anyway because of losses.
> I realize that unterminated and / or Bidirectional
> receiving antennas are not quite as good as a properly
> terminated and properly fed Beverage. That said,
> my unterminated and bi-directional wires DO hear
You can obtain valid numbers by plotting the pattern, determining
average gain, and subtracting actual gain in the desired direction
from average gain. My Web page describes that, and gives some
values for various antennas.
> For 160M, I found a 1000 ft Beverage sometimes
> provided as much improvement over my 500 footer
> as the 500 footer provided over my vertical.
In hundreds of A-B tests over the past few years, where both
antennas were available at the same time, I've almost never found a
case where a properly terminated 1000 foot Beverage beats a
shorter Beverage...say one around 600-700 feet long.
Because there is so much loss along the length of the wire, all the
extra length does is add minor lobes and extra unwanted nulls. I
measure only about 50-70% current remaining after 600 feet. With
70% current remaining, it is like feeding that last area of the
antenna through a 3dB pad. This loss causes poor null formation,
and the result is little improvement in directivity factor. It's like
stacking two beam antennas with one robbed of 3dB power!
Not only that, a second effect comes because the phase of arriving
signals is very unstable when the distance between two points is 2
wavelengths or more. It is impossible to keep the ends of the array
in-phase. This effect increases fading, and decreases reliablility of
the antenna.
That is why all my arrays are limited to about 800 feet long or
shorter, and 700 feet wide. That is also why I run diversity receivers
in stereo on wide-spaced arrays, because my "brain" can ignore
the phase errors and add the signals.
The more compact the array for a given directivity or pattern width,
the more reliable the array will be in daily use. A compact array will
also have less QSB than a similar pattern but physically more
"spread out" array.
As a matter of fact, I soon hope to shorten my last and final
attempt at using long phased Beverages. A pair of 1000-foot west
antennas will be shortened to 780 feet or so, because an existing
pair 780 feet long are better than the 1000 footers about 95% of the
time.
> 80M, the 1000 footer seemed 'dead' in the early
> evening but 'came to life' several hours later. My
> guess is that waveangles at / after sunset are
> higher and the long Beverage was not hearing
> those higher angles.
It doesn't hear low angles either, because of phase errors caused
by the velocity factor from earth losses below the antenna. If you
want to hear low angles, you need phased verticals like those on
my web page.
One thing that saves the antenna on 80 meters is the high loss-per-
wavelength along the antenna. The loss effectively divorces the far
end of the antenna from the system, so phase errors inherent on
ionospheric paths with a spatial separation of a several
wavelengths do not increase fading or produce significant deep
nulls.
Another common case where long Beverages work better than
shorter counterparts is when the shorter Beverages are improperly
terminated. Longer antennas self-terminate because loss along the
antenna helps correct termination errors. As a matter of fact the
unterminated F/B is a bit more than half of the unterminated return
loss of the antenna. The very fact you see F/B ratio in an
unterminated wire clearly illustrates how the far end is severely
restricted in power from loss along the antenna!
When there is more than a few dB of loss, the extra length can't
help the signal. It can only make things worse. John Kuecken
discusses this in his book "Antennas and Transmission Lines.
While his subject is transmitting longwires, it obviously applies to
receiving systems as well. Kuecken, in actual antenna
measurements for commercial communication systems, found that
long wires 1/3 of their height above ground reach a point of
maximum gain when just over 3 wavelengths long, and gain
decreases beyond that length. Of course that point would occur
much sooner with a wire close to earth because losses would be
much much higher.
Another factor in observed improvements could be the user's luck
in just by chance placing a null in the direction of a dominant noise
source.
By the way, remember whatever we do the electrical length of the
wire is always longer than the physical length might lead us to
believe. You can measure the electrical length with something like
a MFJ259, using the "distance-to-fault" mode.
Anyway, enough of this. Most of it is detailed on www.w8ji.com
anyway, complete with patterns and other visual aids.
73, Tom W8JI
W8JI@contesting.com
________________________________________________________________________
Where do you get ICE bandpass filters & beverage matching boxes? The
same place that pays for the hosting of this list: The eHam Store.
Order online at http://store.eham.net.
-----------------------------------------------------------------------
-----
FAQ on WWW: http://www.contesting.com/FAQ/towertalk
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
|