Good to hear from you and get your comments.
Following my request for information and discussion, I got several
interesting e-mails, but few with hard data. There is a dearth of
information on skywave propagation in the medium-wave range. But I got some
good descriptions of experiments and experience, such as yours.
The effects of local ground conductivity are well-known on ground wave, but
not for launching a skywave. Your comment on larger ground planes resulting
in more energy radiated at the lowest angles matches the experience of
several hams who installed some very long (several wavelength) radials in a
favorite direction and saw enough improvement to verify the concept. Higher
ground conductivity should result in the same low angle improvement.
I spent 15 years in the broadcast business, and can verify the wild
variations in local ground conductivity. We often measured 2:1 to 4:1
variations. In the Midwest, anything from 4 mS/m to 20 mS/m could be found
where the map said "8" or "15", and often there was a discrete "step
function" to a higher or lower conductivity. Sometimes you could determine
the boundary (e.g. a flood plain or a ridge that suggested underlying rock),
but as often as not the abrupt change did not correspond to any surface
As to elevated vs. ground mounted radials -- W8JI and W4WA here in Georgia
have two of the strongest 160M signals from the Southeast U.S. One uses a
large on-ground radial system, the other 21 elevated radials (30 feet up).
It appears that both methods can be highly effective.
I have one more simple experiment to perform -- I have a rotatable 40M
dipole at 75 feet on my tower. I will install an elevated ground plane in
one of my trees and make a series of comparison measurements. I fully expect
the difference to be less than predicted by the uniform ground used in the
NEC-2 models, due to incomplete ground reflection caused by vegetation and
terrain roughness. The amount of difference and variations with wave angle
and direction will be interesting.
I have copied this to the Antennaware reflector as you suggested.
It is a plaesure to find someone interested in vertical ham antennas. For
years I have believed that in many cases they are the antenna of choice.
About twenty years ago I made a study of verticals for submarines. There
a vertical is an answer to some of the limited "real estate" problem. One
of the things I studied was the effect of ground plane size vs take-off
My ground plane consisted of a 1 meter diameter copper sheet. I
constructed an elevation range at a distance of abotu 8 meters (25 feet).
I measured received gain at 5 degree elevation increments from zero to 95
degrees every 50 MHz from 150 to 750 MHz. I also ran several azimuth
patterns to insure that symmetry existed and as a simple system check.
Data were plotted for variable frequency, and by scaling, assuming a
The latter proved to be interesting. As you might suspect, the larger the
radius of a circular ground plane, the lower the take-off angle. A ground
plane of greqter than one wavelength achieved a take-off angle in the two
During this time I realized that take-off angle must also be related to
conductivity. I used some of this info to design a vertical periscope
antenna to receive a polar orbit satellite. It worked very well.
In a related measurement, I made some ground conductivity measurements
locally. This is not quite as simple as one might think. The upshot of
the measurement set is that all the "conductivity" maps you might see are
BS. I found a four order of magnitude variation in a five mile diameter
The idea of a resonant, ground mounted radial is incorrect. A quarter
wavelength element half on an elevated antenna exists because of a
"boundary condition." At the end of the of the quarter wavelength,
current has no where to go. Therefore it must be zero. This fact alone
determines the impedance of a dipole antenna.
Ground mounted radials do not have such a sharply defined boundary.
Current flows at the end of this element. Therefore a "quarter
wavelength" radial doesn't really exist. In my experience, the longer the
radial, the better. But at some point it is better to invest in more
radials than longer radials. I think this point was well discussed in
Brown et al, back in the 1930s.
Ground vs tower mounted verticals.
I see no physical difference between a ground mounted and tower mounted
vertical, if both antennas have a substantial amount of radials. N1CQ and I
have agreed to disagreed on
this subject. Someday we might have the time to measure the difference.
After many years of thought on the subject, I am firmly convinced that a
ground mounted vertical, with many, many, long radials will outperform
most yagis at HF.
The reason for this is obvious. The most serious problem with most
horizontal ham antennas, including
yagis, is height. The low height of the antennas causes the beam to aimed
upward from horizontal. In some cases, most of the energy goes straight
up. As a rule, hams can not afford to provide the proper elevation for a
dipole or yagi.
well there is my two cents worth on my favorite HF antenna. Please post
this to your reflector site if you think it is worth it.
W1EQO / VE1EQO
National Wireless, Inc.
FAQ on WWW: http://www.contesting.com/antennaware-faq.html
Administrative requests: antennaware-REQUEST@contesting.com