[Elecraft] Twisted Pair Wiring and RFI Rejection

[Jim Brown]

On Mon, 30 Aug 2010 10:14:14 +0100, David Woolley (E.L) wrote:

>That's another of Jim's suggestions that the UK amateur radio trade 
>and RSGB shop have failed to take note of. 

I don't want to take credit for repeating fundamental principles that 
have been well understood for at least a century. The great telephone 
companies began using twisted pairs for both short and long lines 
early in the 20th century. In the early days, the only source of 
interference was 60 Hz power, so wires were run as parallel 
conductors, with the twist in the form of a crossover every other 
pole. This worked fine, because it was much much shorter than the 
wavelength of the 60 Hz power, and telephone lines were able to run 
directly under power lines without interference! 

Today, CAT5/6/7 cable rejects noise SOLELY by virtue of its high 
quality twisted pair construction. Crosstalk is further reduced by 
the fact that the pairs are twisted at different rates. 

These fundamental principles are WELL understood in the EMC world, 
and those of us working in EMC in pro audio have had to learn them to 
keep hum, buzz, and RFI out of our systems. 

I find it ironic that RF folks look down on audio engineering, 
because "it's only 20-20,000 Hz." In fact, audio systems are far more 
complex than RF systems -- they span 3 decades of frequency, require 
transducers that can produce high power over this wide frequency 
range with controlled dispersion far more complex than most antenna 
systems, must work in an acoustic environment that is at least as 
complex as the ionosphere, have dynamic range requirements of 100 dB 
or more, and must regularly achieve that with signal levels in the 
millivolt range in an environment that is full of all sorts of 
electrical noise. 

73, Jim Brown K9YC