[antennaware] Radial Study (long)

[Dave Bowker]

Reference Gary Breeds note of 23 September.

The following is actual data which may be of interest to the discussion
relating to 1/2 or 5/8 wavelength radials versus 1/4-wavelength raidators.

I recently completed Phase-1 of a 2-phase project to build a dual-band
4-element cardioid array (similar to the classic 4-square except for the
phasing and direction of firing).  Phase-1 provides basic 40M operation and
phase-2 will add the 80M capabilities next spring.  The primary objective
of this project is to collect actual data on the typical 1/4-wave spaced
array and on 1/8-wave spaced arrays (which this array will become on 80M).

The array presently consists of four self-supporting 1/4-wave elements of
identical physical length and cut to be as close to self-resonant at 7.010
as possible and spaced 1/4-wave per side on 40M.  The radial system
consists of 120 1/2-wave radials, each 70-feet long under under each
element.  The radials are layed on top of the sod by cutting grass short,
laying in the wire, and fertilizing and watering to restore the grass.  The
radials extending outward from each element and from the geometrical center
of the array are full-length 1/2-wave long.  Where radials extend inwards
towards the geometrical center of the array and where radials of one
element would "cross" those of another element, they were terminated in a
common bus which crosses through the geometrical center of the array
(typical of broadcast directional array installations...see Jasik chapter
20 and papers by Smith Electronics, circa '70's). 

Estimated soil constants are for very poor ground, as soil content consists
of approximately 2 feet of sandy loam over gravel and shale base river bed.
 The array is positioned on 1-acre of clear land with tall popular and
spruce trees on one side (the side adjacent to elements 2 and 4) of the
array and spaced approximately 3/4-wavelength from it.  Nearest structure
is 1-wavelength distant (at 40M).

Measurements of self-impedance and coupled impedances were made using a 
GR-916A RF impedance bridge with calibration traceable to NIST.  The
self-impedances measured at 7.010 MHz were as follows:  Z11 = 45 + j4.64, 
Z22 = 43 + j5.0,  Z33 = 45 + j6,  Z44 = 43 + j3.3.  

Note that these self-impedances are not the classical 36 +j0  (38 + j0 if
one were to include the typically accepted loss resistance of 2 Ohms for
such a radial system over average ground...see Brown, Lewis, Epstein).  I
have not thoroughly investigated and analyzed all the factors which may be
causing these self-impedances to be considerably higher, however, I believe
the ground characteristics are certainly a factor.  At the time this data
was taken, the soil was extremely dry and the grass had not yet over grown
the radials.  I will obtain another data point in next few weeks as the
radials are now completely overgrown and soil is moist down through the
loam top layer.

As a point of interest, I extended each of the self-supporting elements to
be resonant at 4.6 MHz (approximately 53 feet) and took self-impedance data
as follows (same radial system, same soil conditions).  At 4.6 MHz the
radials are approximately 0.31 wavelengths.  The following self-impedances
were measured.  Z11 = 43 + j1,  Z22 = 46 + j0,  Z33 = 44 + j0,  Z44 = 45 +
j1.5.

Comparing the self-impedance's of the 1/4-wave radiators over 1/2-wave
radials at 7 MHz to the 1/4-wave radiators over 0.31-wave radials at 4.6
MHz shows little difference in the measured data.  

Measured performance data of the array on 40M is 21 dB F/B and 3.7 dB gain
(as compared to a single resonant element), based upon actual FS data taken
at ~1/2-mile distance from one face of the array (using calibrated HP
attenuators - not S-meters!).  Front/side is 4.9 dB down from main lobe
(obtained by switching direction of array with measurement equipment at
same location as for F/B and gain measurements)

Photos of the array, radial system installation, element mechanical design
and construction, phasing network, measurement setup, and data will be
available to subscribers via email in a month or so (I'll post to reflector
when available). 

73, Dave, K1FK
Fort Kent, ME

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