[Elecraft] K3 TVI on 40 meters only

[Jim Brown]

On 8/26/2017 9:07 AM, Lynn W. Taylor, WB6UUT wrote:
> That said, the short xDSL part isn't supposed to radiate. 

"Supposed" is the key word here. Radiation occurs due to several 
mechanisms.

1) Poor circuit layout inside equipment that is also poorly shielded. 
That wiring radiates.

2) Poor common mode isolation of wiring connected to equipment. On 
paired cable, a low quality output transformer or poorly balanced output 
stage puts common mode on the cable. On coax, a Pin One Problem puts 
common mode on the cable. The cable radiates, just like any other 
antenna. Power supply wiring is also a potential radiator, and the power 
supply itself is often a noise source (if it's switch-mode).

3) Poor quality paired cable. Structured cable (CAT5/6/7) is, in 
general, pretty high quality twisted pair, but quality also varies with 
manufacturer. Belden, for example, builds its structured cable from 
molded pairs, which makes their construction more uniform, which 
minimizes radiation/reception of differential mode signals. By contrast, 
parallel wire cables have lousy rejection (20-40 dB worse) and are prone 
to radiation and reception of ANY interfering signal, from audio to VHF.

As an example of #3, zip cord, glorified or otherwise, is TERRIBLE 
speaker cable, because it is a sitting duck for any audio or radio 
frequency noise. When I lived in Chicago, solved a lot of RFI from 
broadcast TV to hi-fi systems by replacing zip cord speaker cables with 
twisted pair.

Sadly, most modern local telco wiring is either poor quality twisted 
pair or not twisted at all. Using this wiring for DSL is an open door to 
RFI.

These, and other issues related to RFI, including CATV and DSL systems, 
are addressed in detail in several app notes and tutorials on my 
website. k9yc.com/publish.htm

73, Jim K9YC

73, Jim K9YC