[TowerTalk] Off Center Fed Antennas

[Jim Lux]

On 6/5/13 1:50 PM, Jim Brown wrote:
> On 6/5/2013 1:33 PM, Larry Loen wrote:
>> Are you going to contend that dipoles don't work in city circumstances as
>> well?
>
> Gee, I thought I made my point quite clearly at the very beginning of
> this thread. Perhaps you should re-read it. Indeed, I observed that a
> true dipole (defined by most textbooks as a CENTER FED antenna -- that
> is, two equal lengths) is a far better choice, because it can be choked
> effectively.
>
> The problem is that the Windom is badly unbalanced, the feedline picks
> up noise, and it needs a serious common mode choke at the feedpoint to
> kill the noise. But because it is so badly unbalanced, there is a lot of
> common mode voltage across the choke, and a choke is likely to fry with
> high power.


I've always worked on the theory that you start with "close" to equal 
legs and then thrash from there. Sure, you're going to have asymmetries, 
but all the other stuff is less of an issue, the closer you are.

I think the gripe that Jim might have is with "deliberately" going OCF, 
when there's no good reason to.


About 10 years ago, I was looking to put up a new wire antenna and I ran 
across the whole OCF thing. I ran the models: hey, about the same 
several hundred ohm feedpoint impedance on all bands: all I need is a 
suitable transformer and I'm good to go.

Then I started doing some parametric perturbation (that is, I started 
tuning it a bit off, or throwing some other conductors in the model 
near, representing things like rain gutters, etc.)

All of a sudden, that "works on every band" went away. sometimes it was 
horrible, sometimes it was ok.  This, to me, indicates that the design 
is "over sensitive" to small things.

So I went back to a center fed multi wire dipole (like a DX-CC), and 
analyzed what the effect of a tuner at the feedpoint would be 
(calculating the losses, etc. using a version of AC6LA's spreadsheet 
turned into matlab code).  Hmm, not much loss for the "nominal" case. 
And what's better, when I perturbed the model with other things (other 
conductors, changing the droop, changing the length of one leg, etc.).

Not much difference. The pattern stayed reasonable, the loss in the 
tuner was small, etc.


So now, my "simple go-to" design is "tuner in the middle, wires close to 
the right length, don't hassle the last 6 inches of pruning"

And for those with a taste for heat.. a resistively loaded dipole with a 
big amp. That 3-4 dB loss makes everything a decent match.


>
> There's an extensive discussion of this in my Power Point on Coaxial
> Chokes. http://k9yc.com/publish.htm
>
> 73, Jim K9YC
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>