On Dec 26, 2005, at 11:49 PM, Peter Dougherty (W2IRT) wrote:
> The tower is behind the house, so no possibility of putting radials
> out to the east or southeast, and to the other directions, I won't be
> able to put out more than a couple of full-size 240-foot radials bent
> all over creation, another 4 or 5 128 footers and a bunch of 33's and
> 66's. Will this be enough, do you think?
AM broadcasters usually spec out 120 1/2 wavelength radials. That's
considered an excellent radial system. In general, so long as the
ends of the radials are within 0.025 to 0.050 wavelength, a set of
radials of a given length virtually indistinguishable from a solid
sheet.
That said, few amateurs have room for such a radial system. A
compromise is in order. If you look at how much "work" each radial
does, the most important part is the part nearest the radiating
element. A shorter radial field will work nearly as well as a longer
radials.
I have a similar situation to you, that my tower is located at a
corner of my house. Therefore, there's 90 degrees of ground that I
can't lay radials.
My advice is to put down as many radials as long as you can. Mine are
only 60 feet long (.22 wavelength on 80m). A couple are shorter. One
is only 41 feet. With 25 of them, that's not a bad radial field for 80m.
The key is to put down a number of radials according to the rule of
thumb above. I'd suggest the minimum should be about 30 .20
wavelength radials in a full circle. Put down whatever part of this
isn't blocked by the house. If you have to shorten some to fit, then
do so. If others can be longer, lengthen them to about .40 wavelength
or so, max. (Remembering that the far ends don't do as much work --
that wire may be better spent putting shorter radials close to the
antenna)
> How exactly does one shunt-feed a tower? I'm reading the ARRL antenna
> book and I'm afraid I don't really understand what they're describing
> fully.
A shunt-feed is like a gamma match to the tower. Like a gamma match,
the difference in diameter between the shunt wire and the tower, as
well as the length and spacing of the arm determines an impedance
transformation.
Now, if you could compute all the parameters of your tower and shunt
arm, you could figure out the exact length, and spacing to make a
perfect match (or one that you'd only need a little series
capacitance). Doing so is quite difficult. Figuring this out by trial
would mean a lot of tower climbs.
The advise that K9AY gave me when I put up my tower was to run a
shunt arm as high in the tower as I could, spaced a reasonable
distance (1-3 feet). Then, one could measure the impedance at the end
of the shunt arm, and design a matching network to suit.
For me, this was 12.5m up my 15m tower, spaced about 12 inches. I
used two 12 gauge wires spaced about 10.5 inches apart. (This forms a
skeletal ribbon about the same width as the tower diameter) Since I
didn't have access to impedance-measuring equipment, I designed my
networks by trial and error. For 80m, I use an omega match. For 160m,
I have an L network.
> Vacuum-variable caps at $600 a pop are a little scary to the
> budget of this new home owner, too!
How much power do you run? If you only run 100 watts, you don't need
the vacuum-variable caps. Even if you are QRO, you might be able to
use caps with a reasonable spacing. It all depends on the voltages
and currents that are present in the matching network.
> If I go with an inverted-L suspended from the tower, how is it fed,
> and how far from the tower should it be run?
It's fed from the bottom, just above the ground. You should space it
3-5 feet away.
> Will it interact with the tower itself and the yagis on top?
Yes.
> If so, favourably or unfavourably?
Hard to say.
> I have a C31XR and Force12 2/2/1 up top so I'm hoping
> this could help things, i.e. capacity hat.
If you shunt feed the tower, yes, they will act like a capacity hat.
> What size wire should be used for radials? I've got insulated #14
> from Home Despot (about $25 for 500'), but I think that may be
> overkill. I suspect i'll have to bury them, as much as I realize that
> elevated radials work better -- but this is a suburban house first
> and a ham-shack second <grin>.
Any size will do. 16 gauge is cheaper, and will do great. You could
probably go down as far as 20 gauge. There's no much current to
carry, especially with more radials. The main problem is getting
something strong enough so it won't break when placed on the ground.
You don't have the bury the radials. I clip them to the ground on top
of the grass. After a season, the grass grows over the wire and you
can't even FIND the radials. I make little U-shaped clips out of 5"
of 12 gauge copper wire. Making them is work. You can also buy little
clips for holding down sod at Home Depot. I use one about every 3 or
so feet of wire, so we're talking hundreds of clips.
Contrary to popular myth, elevated radials do NOT work better than a
good ground-mounted radial field. You're better off.
> Am I correct in believing that two separate feedlines are needed for
> this, one for 80 and one for 160?
For my shunt feed, I use a relay to switch matching networks.
I've seen others use two shunt arms, one for each band.
> Are there any known RFI pitfalls to either of these methods?
Good news and bad news. The bad news is, if you shunt-feed the tower,
your radiating element is right next to the house. The good news is,
on 160m and 80m, the capture area of the house wiring is small, so
the RFI effect will be small as well. I don't have any trouble with
100 watts.
Bill Coleman, AA4LR, PP-ASEL Mail: aa4lr at arrl.net
Quote: "Not within a thousand years will man ever fly!"
-- Wilbur Wright, 1901