[TowerTalk] Wire grounding question

[Jim Lux]

On 11/21/13 6:25 AM, TexasRF at aol.com wrote:
> John, what is the problem with using multiple power strips?
>
> 73,
> Gerald K5GW
>

there are potential problems where you can actually make the transient 
worse when you have a long wire (which can store transient energy) with 
fast operating shunt switches at both ends.   You're basically making a 
Blumlein pulse generator.

The scenario is that you have a transient come in on the power line 
through shunt switch that triggers at some voltage (say 200V) and it 
doesn't fire.  It travels down the line to the next shunt switch which 
is lower voltage, and fires. Now you've got a short across the line, so 
two things happen:
the line discharging it's capacitance into the shunt has a high di/dt, 
which can couple the transient to the line on the load side of the shunt 
switch; and

the "short circuit" causes the remainder of the transient (assuming the 
switch fires fast) to reflect back towards the source.  This creates a 
double voltage pulse, which fires the first shunt when it gets there, 
creating another high di/dt transient, etc.


There's also the problem of leakage currents from "partly failed" MOVs. 
  MOVs don't fail abrubtly, they gradually degrade. Each hit consumes 
some of the insides, and their leakage current increases. So if someone 
designs very aggressively (low threshold voltage), ordinary line voltage 
"swells" with noise on them will trigger the MOV, consuming a bit of life.


So what you want is
a) lowest voltage at the service entrance
b) significant series L, to limit the di/dt of any transient.


Standler's book is worth getting
http://books.google.com/books/about/Protection_of_Electronic_Circuits_from_O.html?id=5-NEmE1EhYQC


http://www.rbs2.com/pq.htm