[TowerTalk] parasitic 4-square for 40 meters

[Jim Lux]

----- Original Message -----
From: "Chris Adams" <n4vi at arrl.net>
To: "'Ron Feutz'" <feutz at wctc.net>; <towertalk at contesting.com>
Sent: Monday, February 21, 2005 9:55 PM
Subject: Re: [TowerTalk] parasitic 4-square for 40 meters


> Ron,
>
> My suggestion would be to use the standard 4-square array layout and add a
> center element.  The center element would be the driven element and the 4
> parasitic elements would be directors/reflectors/no effect.  I suspect
that
> the two side elements would be non-contributors, leaving the "rear"
element
> as a reflector and the "forward" element as a director.

One should also consider other configurations of directors and reflectors...
3 els as reflector, 1 as director; 2 reflectors, 2 directors.  The work
involved in tuning and switch is essentially the same, the overall gain and
F/B might be different, but one of the configurations might prove to be less
sensitive to small changes in length, ground properties, etc.  A few hours
work with a modeling program will answer it, and will potentially save you
tens of hours standing outside trying to get the phasing line "just right".


 Two ways to tune the
> parasitic elements come to mind:
>
> 1.  Bring coax back to a center switch box.  I think the best idea
> would be that all coax length be cut equal to provide the correct
reactance
> to be either a reflector or director.  Additional reactance would then
have
> to be added to either the front or rear elements (via the coax) to get the
> desired effect.  Modeling would dictate what to do with the side elements.

Coax stubs aren't necessarily lower loss than a lumped L or C, right at the
base.  Unless you're centralizing the relays for some reason, it's probably
easier to put lumped components and relays at the parasitic element, and
then just run relay power (for which loss, RF handling, etc. aren't issues)
probably easier to decouple the relay control wire than some piece of coax.

>
>
> 2.  The other possibility is to have tunable/switchable reactance at
> the base of each parasitic array and change the reactance to change
> direction.
>
> I don't maintain that this solution is either easier to implement or
offers
> better performance then a standard 4 square array.

This is, in general, true. Assuming you have a reasonably flat, symmetrical
place to put the elements, and they're evenly spaced, you're going to need
some small number of components (Tline stubs, Ls, Cs) to do the tuning and
some small number of relays, regardless of the approach.

A simple approach, using the equipment you already have (i.e. the RCS-8V)
and coax lines is probably almost as good as any other exotic scheme.  The
classic statement about phased arrays is that they're easy to make work, and
tough to make work extremely well.

Jim, W6RMK