[TowerTalk] K8UR wire 4-SQR characteristics vs 1/4 vertical array design from N0AH

[Dick Green]

Excellent write-up, Paul.

A few years ago I built a 40M 4-square with full-size vertical elements and
240 ground-mounted radials. I can tell you that the construction took much
more time and effort than was required for your dipole version. I built the
elements out of telescoping aluminum tubing from Texas Towers. That only
took 1-2 hours per element, including carefully removing burrs, cleaning the
tubing, cutting slots, drilling a hole for a feedpoint screw, assembly, base
construction, mounting, etc. But it took months to construct and install the
radial system. I measured and pre-cut each radial, colied it, soldered one
end to a metal ring at the base of an element, uncoiled it, and pinned it
down with bobby pins. I also soldered dozens of radials to a common bus
where they met inside the square. It took forever!

Results were excellent, however. I tuned the array for the CW portion of the
band because that's where I do more than 90% of my operating on 40M (I only
work 40M SSB when forced to in a Phone contest!) The designed center
frequency is 7.025 MHz -- i.e., the element spacing, phasing lines and
radials are all cut to 1/4 of that wavelength. The idea was to optimize for
40M CW contests and DXing, most of which takes place from 7.000 MHz to 7.050
MHz. I also wanted to hedge for the possibility that someday WARC will give
us a CW allocation that extends 100-300 Hz below 7.000 MHz. My dumped power
at resonance is 2%, pretty close to ideal, and it's under 3% from 6.900 MHz
to 7.150 MHz. F/B is the advertised 20-25dB. I imagine the 1% reading on
your dipoles is due to the better match they provide to the feedline/coupler
at resonance than is possible with a 1/4-wave vertical with 60 radials. The
downside is that dumped power creeps up to about 7% at 7.300 MHz.

Had I centered the array at 7.150, the dumped power would have been 3-4% at
the band edges, with under 3% across a 250KHz bandwidth. Does your array
maintain 1% dumped power across the band, or does it rise higher at the band
edges? If it stays under 2% across the 40M band, it's far superior to what
can be accomplished even with full-size vertical elements.

I happened to have had an old Colatchco hybrid coupler, the predecessor to
the Comtek box. Jim acquired Colatchco and improved the coupler with a
double-hybrid design that eliminates two short external coaxial
phase-adjustment lines. Even though I didn't buy the coupler from Jim, I
acquired other parts from him including the phasing lines and current
baluns. He was incredibly helpful during design and construction -- he knows
his stuff about these arrays! When I build an array for 80M, there's no
question that I'll use a Comtek coupler.

For the benefit of those contemplating a 4-square, it's a good idea to
permanently install a power meter on the dummy load to monitor dumped power.
Not only does this make the initial testing easier, it will also immediately
reveal if there's something wrong with the array -- which you can't tell
from looking at the SWR in the shack because it's always the perfect SWR of
the dummy load. Problems have been revealed a couple of times. Once critters
chewed through the control line leading to the coupler. Each element has
slightly different dumped power, so an identical dumped power reading on all
elements alerted me to the problem. Another time a massive wind storm made a
giant tree branch fly through the air and crush one of the elements (dumped
power soared!) Some people run a coax line from the dummy load to the shack
and install the meter there. My array is over 300 feet from the shack, so
losses in the coax run would introduce inaccuracy. Instead, I installed a
remote directional coupler at the dummy load and ran a DC line to the meter
in the shack.

An exciting aspect of the dipole array is that, according to the diagrams in
ON4UN's book, it has about the same radiation angle as the 1/4 vertical
design. The low angle is one of the great strength of these arrays, and it's
nice to see that the dipole array has it, too.

I suppose one slight advantage of the vertical array is that it will
probably require less maintenance in the long run. Ropes, wires and cords
are more subject to breakage over time, especially here in New England.

All in all, I'd say that if you will have an easier time putting up 65' of
tower than constructing radials, and don't mind climbing, the dipole array
is the way to go. No doubt, it can be constructed *much* faster and there
appears to be no performance penalty.

73, Dick WC1M


--
FAQ on WWW:               http://www.contesting.com/FAQ/towertalk
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com