[Elecraft] Re: LONG: Background info on Re: WTB: CPK for ButternutHF-6V

[Bob Tellefsen]

Hello Niel
I'd like to talgate Don's comments with a small one of my own.
If you should use just a single radial, you may or not be able to
see the little bit of directivity the antenna will develope in the
direction of the single radial away from the vertical.

I did a quick check with EZNEC on a 40m vertical with one
radial.  There is a 5 dB front-to-back ratio, with the best signal
in the direction of the radial.  The flip side of this is that the
antenna gain is only -1.6dB in the favored direction, and even
less in other directions.

You might be able to take advantage of this, if your property
is aligned in desireable directions.

Good luck and 73
Bob N6WG

----- Original Message ----- 
From: "Don Wilhelm" <w3fpr at embarqmail.com>
To: "Niel Skousen" <nskousen at talisman-intl.com>
Cc: <w3fpr at arrl.net>; "Elecraft Reflector" <elecraft at mailman.qth.net>
Sent: Wednesday, August 20, 2008 1:29 PM
Subject: [Elecraft] Re: LONG: Background info on Re: WTB: CPK for
ButternutHF-6V


> Niel,
>
> Fine on your constraints.
> Remember that a vertical with raised radials is just a dipole 'on
its
> side' with one leg (a single radial) bent at 90 degrees to the
radiating
> element.
>
> So how does only the vertical part radiate you may ask.  That
happens
> because you add another radial physically opposite the first one,
and
> the radiation of one radial is 180 degrees out of phase with the
> radiation from the other, so they cancel leaving only the vertical
> portion to radiate a signal.
>
> If the pair of radials are not symmetrical, then you may be just as
well
> off with only a single radial per band.
>
> The radials must each be tuned (just like one end of a dipole) with
the
> vertical radiator.  The radials must be an *electrical* quarter wave
> long at the band of operation - they can be loaded with inductors
near
> the junction with the coax shield, or they can be end loaded with a
"T"
> wire, or some combination of both, but they must be resonant.
> If they are placed close to the ground (4 feet is close for the HF
> bands), there will be some effect from ground coupling that will
de-tune
> them, so the easiest is to cut them long and do the trimming with
the
> aid of an antenna analyzer (if there are several radials for each
band,
> they must be tuned one at a time - connect only one and tune it,
then
> connect only the next one and tune, etc. until done.
>
> Remember that the radials will not be 'radials' unless they are
> symmetrically placed around the vertical, it is of no use to run two
> radials for any one band out in the same direction for they will
produce
> the same radiation pattern as a single radial.
>
> Another 'solution' is to trade that antenna for one that is an
> electrical half wave - which does not need radials being an antenna
> complete within itself.
>
>
> 73,
> Don W3FPR
>
> Niel Skousen wrote:
> > Hi Don (and all who have offered advice & help)
> >
> > Thanks to all for the input that has been offered.  Based on some
of the
> > comments, I thought I might share the 'rest of the story', both to
seek
> > additional input, and to show the technical trade off's associated
with
> > a compromise antenna environment.
> >
> > I'm installing a used HF-6v in a CCR environment, so I have some
> > environment limitations.  The CCR 'enforcement' committee is not
> > horribly aggressive, but if I don't aggravate the situation, I'm
much
> > better off.
> >
> >
> > So, the HF6 will be mounted about 4' off the ground, a full
elevated
> > radial system is not possible.  Choices seem to be the CPK
solution from
> > Butternut or a modified radial system.  Two options are being
considered
> > on the modified radials.
> >
> > First, 3 tuned radials (either individual wires for 40/15, 30, 20,
15,10
> > or the Butternut tuned twinlead variety) could be connected at the
base
> > of the antenna, then run down (hidden in the bushes) at a 60
degree
> > angle to ground level, then run under/around the rocks.
Performance is
> > expected to be better than the CPK, but still a compromise.
Routing is
> >  not linear.  Tuning of this radial environment is expected to be
a bear !
> >
> > The second, the CPK, is expected to be enhanced by the low
elevation,
> > but NOT as effective as a good ground radial system.  I might be
able to
> > enhance this configuration by laying out a psuedo radial set of
6-10 8'
> > wires at the base of the 4' mast, which would more accurately be a
> > surface capacitive element to enhance the CPK's effect.
> >
> > In neither case will the near field ground losses be impacted.
There
> > are some spiral wound counterpoise idea's out on the web, as well
as
> > some thoughts on constant angle spirals, which are tempting, but
at this
> > juncture, I expect to lean to the CPK to get it up and running,
with
> > spirals and experiments to follow...
> >
> > So any additional input, idea's, are solicited and welcome
> >
> > Thanks again all,
> >
> > Niel
> >
> >
> >
> > Don Wilhelm wrote:
> >> Remember (or understand if you did not before) that elevated
radials
> >> should be tuned (pruned for length) just like a part of the
antenna
> >> (they *are* part of the antenna).
> >>
> >> So get out your antenna analyzer and cut one radial a bit on the
long
> >> side - attach only that one radial to the vertical and shorten it
until
> >> you have the correct readings on the antenna analyzer.  Cut the
second
> >> radial to the same length - move to the next band and do the same
thing
> >> until you are done.  Note that the radials can interact on a
multiband
> >> setup (just like parallel dipoles), so tune the lowest bands
first and
> >> move upward in frequency one band at a time.
> >>
> >> The twinlead solution does work, but suffers badly from
interaction.
> >>
> >> 73,
> >> Don W3FPR
> _______________________________________________
> Elecraft mailing list
> Post to: Elecraft at mailman.qth.net
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
>  http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com