[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Necessity of NST Protection Filters and Safety Gaps



Original poster: "Lau, Gary" <gary.lau-at-hp-dot-com> 

Your calculations are mostly correct, although the actual current may in 
fact be far higher than 60mA (nothing is simple), and one's resistors may 
get hotter than expected.  There is also power dissipated in the resistors 
due to the bypass cap energy being dumped at each bang, more on this 
later.  In theory, performance should be "slightly" less with a filter, 
though I don't think I've heard of anyone performing a with/without 
performance comparison.

The R and C values needn't be precise any more than chicken wire needs to 
have its holes any particular diameter.  As long as the "stuff" you're 
trying to filter can't get through, you're fine.  An R-C filter does not 
have a sharp cutoff frequency, and contrary to intuition, the bad "stuff" 
is actually in the MHz range, not the KHz tank frequency.  Think about it - 
the tank frequency is only present when the gap is conducting, and when the 
gap conducts, it's shorting out the NST leads, so there's no tank RF being 
applied to the NST.  The MHz stuff is actually present at twice the peak 
cap voltage, and comes about when the gap briefly turns off at each zero 
current crossing.  The good news is that such high frequencies are very 
easily attenuated by an R-C filter.  See further explanations on my web 
site - www.laushaus-dot-com/tesla/protection.htm.

Note that if the bypass cap value is high, this causes power to be 
wasted.  Each time the main gap fires, the tank cap is discharged into the 
primary, and the bypass caps are discharged into the resistors.  If the two 
bypass caps in series have a value of 1/10th that of the tank cap, then 
about 1/11th of NST's power will be burned off as heat in the resistors, 
not even counting the IR loss in the rsistors as the tank cap is charged 
through them.  So more bypass cap is bad for performance, and unnecessary 
for filtering.

Gary Lau
MA, USA


Original poster: "Luke" <Bluu-at-cox-dot-net>

Thanx for the links
That answers my question about inductors in the filter circuit.
Using ohms law with a 60 ma NST the voltage drop across each resistor
would be about 60 volts totaling 120 volts.  Does this hurt performance?
Or is the 120 volts (0.8% of the NST output) small enough to be
considered negligible?

And why is the capacitor value across the NST not needed to be any
particular value?


Luke Galyan
Bluu-at-cox-dot-net