[RFI] Question re: Whole House Surge Protection

[Jim Brown]

On Wed, 21 Jul 2004 20:32:35 -0400, Tom Rauch wrote:

>While I not saying surge protection should be ignored,  how
>you connect things provides nearly all of the energy
>diversion. If you don't follow proper lead routing and
>grounding, there isn't a suppressor in the world that will
>protect your gear.

Yes. To understand that better, consider this. 

Virtually all "whole house" surge protectors, and nearly all branch circuit surge protectors, 
use shunt mode devices (most commonly, MOV's) to short the transient (lightning hit) to 
ground. The current in the surge will then flow through the impedance to ground, AND 
THERE WILL BE A VOLTAGE DROP ACROSS THAT IMPEDANCE!  It's essentially a 
voltage divider, with one leg being the impedance of the wire carrying the strike, and the 
other leg being the impedance to ground. The impedance that counts is at the frequency of 
the strike -- IEEE says that the energy in lightning has a broad peak around 1 MHz. 

That drop raises the voltage on the ground to some level that, depending on the divider ratio, 
may be a significant fraction of the full voltage of the strike. Any equipment connected to that 
ground will see that potential at one end and the voltage induced by the strike on whatever 
else the equipment it is connected to as the other potential. That can put some pretty high 
voltages across the equipment.  

One of the problems with the use of MOV's as surge protectors is that they may protect 
equipment on the line to which they are connected, but blow up equipment connected to the 
same ground because they raise the voltage on the "ground." 

Jim Brown  K9YC