[RFI] Earth isolation

[Ian White, G3SEK]

Jim Brown wrote:
>On Wed, 8 Sep 2004 22:39:22 +0100, Ian White, G3SEK wrote:
>
>>The cure was to place an RF block in the shack mains
>>feed, using a three-wire filter which includes a ground-line choke.
>
>What you really did was reduce the current flowing in those unintential 
>antennas by adding
>impedance to the return path. That was a band-aid made necessary by the 
>pin 1 problems
>in the gear. The more fundamental fix would have been to prevent those 
>currents from
>flowing into the gear by fixing the pin 1 problems.

I didn't have any RFI problems in my own shack - mostly because I'd 
already fixed my own pin 1 problems.

All the problems were with the neighbors' equipment, where the only 
practical solution is to reduce the RF current it has to contend with.

>
>Also, it is NOT a good or safe technique to place high impedances in 
>ground leads.
>Lightning wants to see a low impedance path to earth. The peak energy 
>of lightning is
>around 1 MHz, and it is a very broad peak (that is, lots of energy both 
>above and below 1
>MHz). When you place a choke in the ground lead, you essentially throw 
>away any lightning
>protection by defeating the required bond between safety grounds. Not a 
>good thing.
>Mother nature has a way of reminding us that such techniques are not a 
>good idea.

There is no conflict here. The ground path for all unwanted currents - 
be they lightning or RFI - should be through the ground bonding outside 
of the building. We specifically do NOT want any fraction of that 
current to be grounding out by passing through our equipment and the 
indoor mains ground.

For safety at 50/60Hz, I strongly recommend using a commercial mains 
filter that is liberally covered in approval stamps. If the ground 
connection through the filter satisfies all those paranoiacs at UL, TuV, 
BSI etc, then it's good enough for me too.


-- 
73 from Ian G3SEK         'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek