[TowerTalk] SteppIR Vert

[Jim Lux]

----- Original Message -----
From: "skip" <k3cc at fast.net>
To: <towertalk at contesting.com>
Sent: Saturday, September 25, 2004 9:17 AM
Subject: [TowerTalk] SteppIR Vert


> I've just ordered a SteppIR Big Vertical 40 thru 6 meters.
>
> My question is would it be feasible to make a 4 square out of 4 of them
> using a 1/4 wavelength spacing on 40 meters ???
>
> This would be 1/2 wave on 20  and full wave on 10
>
> If their were separate switch phasing lines, relay switched, and a phasing
> box to switch the direction.
>
> I have not yet calculated the lead/lag as yet for phase angles.
>
> Has anyone built or thought about a multiband 4 square ???
>
> Or is this just wishful thinking ????
>
> de Skip    K3CC

I've thought quite a lot about multiband phased arrays using tunable
elements such as the SteppIR. Here's the tricky parts:

1) 1/4 wave spacing on 40 might not be what you want. You need to choose a
spacing that works out for all bands, because spacings like 1 wavelength
tend to be less useful than you might think (you get lots of lobes and
nulls, and it's difficult to control exactly where they'll wind up).  But,
see later...

2) The mutual impedance will be different on the different bands (generally,
speaking, less on the higher bands, because the elements are farther apart
(in terms of wavelength), which means that your phasing networks will be
quite a bit different for each band.

3) You can use the variable length feature to help with the phasing (because
you can make the element a bit longer or shorter to add L or C to the
feedpoint impedance)

4) A square arrangement, while easy to do calculations on paper for, may not
be the optimum arrangement for you. You need to get a map out and figure out
where you want to form beams and/or nulls.  It might well be that a straight
line arrangement might work out better.

5) I think the true value of a phased array is not in forming the beam, but
in forming a null.  An array with 4 elements spaced a reasonable distance
apart isn't going to have a huge amount of gain, but can potentially form
very useful nulls and "low response" areas to cut down on signals coming
from areas where you don't want to talk to.  This, however, requires a
substantially more sophisticated approach to the phasing network control
(the networks are the same, but figuring out the Ls and Cs  is non-trivial)

If you want to fool with phase calculations with various arrangements of
elements, I've got an excel spreadsheet I can send you that allows you to
specify the x,y position of your elements, as well as the phase and
magnitude of the current in the element, and it calculates the horizontal
pattern. It can also use Excel's optimizer to try and find the currents that
produce a pattern as close as possible to what you desire.  (Of course, that
leaves the hard part of the feed networks).

Jim, W6RMK