[Elecraft] Coax Cable Interconnections

[Jim Brown]

On Sat,2/20/2016 8:19 PM, Ron D'Eau Claire wrote:
> I use right-angle adapters to minimize the radius of the turns if I'm restricted to RG-8 size coax, but almost all of my inter-equipment connections are RG-58. It's FB all the way above 1 KW as long as it's not subjected to very high SWR's at high power. (Actually RG-58 can handle 2 or 3 KW at an SWR of 1:1).

Power handling is only one property of coax. Another is SHIELDING, and 
one measure of shielding is the Transfer Impedance, which is the ratio 
of the voltage between center and shield as a result of shield current. 
The lower the transfer impedance, the better the shielding. Transfer 
impedance is the result of shield quality -- density and uniformity -- 
and its resistance at the frequency of interest. The lower limit on 
transfer impedance is the RESISTANCE, so a shield with lower resistance 
will provide better shielding. Since shield resistance is largely a 
function of skin effect, larger diameter coax generally provides better 
shielding than a smaller one. THAT'S why all the coax in my station is 
RG8, specifically Davis Buryflex.

What's the big deal about shielding? Well, for several years, I've been 
working on my station to minimize issues with harmonics when I'm running 
SO2R on harmonically related bands. I have very good bandpass filters, 
and I've added double stub networks to kill harmonics on several bands. 
But I wasn't getting as much suppression as I expected (and measured on 
the networks on the bench), so I suspected issues with coax. Last 
summer, I replaced all the inside cable with hand-soldered Amphenol 
83-1SP connectors on cables cut to length. The result was an improvement 
of about 10 dB in harmonic suppression! The lesson -- when you're 
looking more than about 50 dB down, little stuff really matters.

73, Jim K9YC