At 6:17 PM -0700 8/5/03, Michael Tope wrote:
>Yes, Chuck, I ran similar calculations and found that I needed to
>go with at least 7/8" heliax in order to be competitive with open
>wire line when trying to feed antennas with very high input
>impedances. If you have a sturdy tower and can find the line
>used for a good price, it does offer than advantage of shielding
>which alleviates the problem associated with routing open
>wire line.
What I did when I installed this antenna was to run 15 meters of
open-wire line to the end of my house nearest the antenna, then go
through the wall up near the peak of the roof with a pair of big
Steatite feedthroughs, into the attic. There I put a 4:1 balun,
followed by an L-C impedance-matching network, followed by a 50-ohm
coaxial common-mode choke; and then 21 meters of 5/8ths-inch Heliax
(LDF5-50) to my shack at the other end of the house.
Thanks to the 4:1 balun and L-C network, the SWR on the long Heliax
line is fairly low on all bands 80 through 10 meters.
>The other approach is try to find a quasi broadband
>antenna that offers a better than ~6:1 VSWR (ref 50 ohms)
>over the frequencies of interest. This also makes the tradeoff
>a little better.
That's a good approach, but my antenna had to be stealthy, so I stuck
with a single center-fed wire between two trees.
I designed and simulated with NEC-4 a bifilar log-spiral that could
be suspended from the same three points (two trees and the house),
presented a VSWR of 3:1 or less at all frequencies from 3.5 through
30 MHz, had a nearly omni-directional pattern in azimuth and the
elevation pattern of a horizontal doublet at the same height.
Unfortunately, it looked like a giant spider web, so I never erected
it. If someone out there has three high support points approximately
at the vertices of an equilateral triangle, would like to try the
idea, and isn't afraid of giant spiders, let me know.
-Chuck, W1HIS
|