[TenTec] OT 75 Ohm twin feeder

[Dr. Gerald N. Johnson]

On Wed, 2008-06-11 at 09:35 +0100, Steve Hunt wrote:
> I've been experimenting with ways of constructing 75 Ohm twin feeder for 
> a Hexbeam application - it's now very difficult to get the heavier duty 
> commercial cable - the only manufacturer in UK has ceased production.
> 
> On my website I recommend using 50 Ohm coax for the Hexbeam band 
> interconnects, but folk who opt to use the higher power handling RG213 
> sometimes have difficulty making off the ends. Twin feeder would be 
> easier to use, and as long as its characteristic impedance was  less 
> than about 80 Ohms it would work fine.

Two pieces of 75 ohm coax in parallel (corresponding conductors
connected together at each end) gives 37.5 ohms. Two sets of those in
push-pull gives 75 ohms shielded balanced. Or the same with 50 ohm coax
gives 50 ohms balanced.
> 
> It turns out that 2 lengths of RG58 taped side by side produce the right 
> value Zo, using the braids as  "fat" conductors spaced by the PVC 
> jacket. In a final version I would probably hold them together using 
> heatshrink tubing. I only need short lengths - maximum 24" - so losses 
> are not likely to be an issue.

The jacket is the least consistent part of the coax and I believe it to
be loaded with carbon to make it black and UV resistant. That would make
it lossy (and slightly improve the shielding properties) to some extent.
> 
> I wondered what's a good way of estimating the power handling capability 
> of this arrangement. What's likely to be the limiting factor - the 
> voltage breakdown through the PVC ?

I tested a shielded TV twinlead years ago with my link coupled antenna
tuner. I connected the high impedance side to be parallel tuned, and
connected a couple feet of twinlead to get the maximum voltage.
Initially I left the open end as cut. Running 200 watts to the tuner, I
had beautiful corona off the cut ends. So I stripped the ends back and
terminated the open ends with balls, then I had no visible corona. I
didn't detect corona in the capacitors either, but somewhere 200 watts
was being dissipated (surely not in the coil wound of 1/4" wide copper
strap on edge).

I think losses and voltage breakdown of the PVC will be the limit and
that will not be consistent (and so the Z wont be consistent) from one
maker to another.
> 
> Grateful for any views.
> 
> Steve, G3TXQ
> 
73, Jerry, K0CQ