[TowerTalk] CAT5E based RX SWITCH for remote RX ANT distribution-BALUN

[Jim Brown]

On 2/14/2025 5:01 PM, Jim Lux wrote:
> I'd definitely terminate the shield at one end. You don't want it floating. And terminating at both ends might be a way to get an inadvertent loop.  Jim will probably weigh in - shielded twisted pair is his jam.

FWIW, I would NOT use shielded CAT cable for this system. Rather, I 
would use either a transformer or a common mode choke to transition from 
the coax to the twisted pair at both ends. Twisted pair inherently 
resists crosstalk at frequencies where the twist ratio in turns/unit 
length (called the "lay") is high as a fraction of the wavelength, which 
is certainly the case for the lower ham bands. And the transformer would 
need to have VERY low capacitance between turns. I gave a talk on this 
about five years ago, the material was added to the ARRL Antenna Book or 
Handbook (don't recall which). http://k9yc.com/RXChokesTransformers.pdf 
I achieved the low capacitance with low loss by placing the two windings 
on opposing sides of a suitably selected small toroid.

Almost 20 years ago, as part of an EMC workshop for audio professionals, 
I set up a tutorial demonstration to show the relative importance of 
twisting and shielding. The signal source was two identical electret 
microphone known to have RF susceptibility to VHF and UHF signals. One 
of these mics fed the mix console with a very high quality shielded 
twisted pair. The other used one pair of high quality CAT5 cable for the 
signal pair, and both conductors of another pair in that cable in place 
of the shield. That connection was required, because the FET follower is 
powered by "phantom power," 12-48VDC, with V+ applied through high value 
low tolerance resistors to both signal pairs, and V- to the shield (or 
in this experiment, the pair replacing the shield.

My signal injector was a Kenwood TH-F6A talkie, which I continuously 
keyed on and off as I moved it along each cable. The observed result was 
that the two cables were equally good, equally susceptible, the the 
talkie. That is, both heard it, but it wasn't overwhelming.

And it should be noted that the dynamic range of audio is in the range 
of 80-100 dB for normal program material, and that the peak output of 
these mics is on the order of 0.5V - 1V. Another observation is that the 
turns/inch of that very good mic cable was probably 1/20 that of CAT5.

In the world of pro audio, we learned long ago that twisting is at least 
as important as shielding.

Neil Muncy had a great demo on this for his tutorial workshops. He 
connected a dozen or so mic cables in series, running them around the 
lecture room, and moved a tape demagnetizer along the line as it passed 
through mating XL-connectors. Within the connectors, the pair was not 
twisted, and at that point, a very loud hum was heard. Anywhere else 
along the cable it was not.

73, Jim K9YC