Re: A TC Grounding and Filtering Scheme

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Gerry Reynolds" <gerryreynolds-at-earthlink-dot-net> 




 > Original poster: Dave Lewis <hvdave-at-earthlink-dot-net>


 >  > Yes there will be common mode stuff on the NST and the common mode
choke in
 >  > the line filter will turn that into differential stuff that can more
easily
 >  > be filtered.  The line filter should be grounded to line ground (like
you
 >  > show).  Just remove the RF block components.  I can't  comment on
reversing
 >  > the direction of the filter since I don't know if it was designed to
prevent
 >  > stuff from getting from the line to the load or stuff getting from the
load
 >  > to the line.  You might try it both ways if you have an interference
 >  > problem.
 >
 > The Corcom filter has a common mode choke on the line side which is why
 > I decided to reverse it.  If I turn it around, I figured the Y caps on
 > the load side would let alot of common mode through the NST primary and
 > into the line ground.

I was thinking that too..... But, the Y cap (the value of which is limited
by safety concerns) might be more effective before the common mode choke
since there would be more common mode voltage there and more common mode
energy might be shunted to ground.   I just don't know.  Experiment and tell
us the results.


 > I had pondered this one in some detail...   Here's a comparison between
 > the MOV's in Terry's design (P7215-ND) and the ones I picked
 > (P7261-ND).
 >
 > Digikey part number:     8x P7215-ND   14x P7261-ND
 > -----------------------  ----------    ----------
 > Max Allowed VAC:         1000V         625V
 > Max Allowed DC:          1485V         825V
 > Varistor Voltage (10%):  1800V         1000V
 > Min Varistor Voltage:    1620V         900V
 > Max Varistor Voltage:    1980V         1100V
 >
 > Number of MOV in series  8             14
 > Min String Clamp V       12,960V       12,600V
 > Max String Clamp V       15,840V       15,400V
 > Max Rated AC Voltage     8,000V        8,750V
 > Max Rated DC Voltage     11,880V       11,550V
 >
 >
 > At present, I already have the MOV's in hand from my original filter I
 > built and canabalized.   I just thought I would share my comparison that
 > led me to my choice.

Sorry, that's what I get for trying to do the math in my head.  14 MOVs
seems correct.



 >  > 3.  You are paralleling three 2.5 KV caps to get .0047 uf at 2.5KV and
then
 >  > using 8 groups of these in series.  Thats a lot of parts.  That should
work
 >  > but may be more expensive than needed.  I'm using eight .0047 uf at
1600V in
 >  > series for each bushing (One third your number of parts.  Terry
indicates
 >  > that breakdown voltage on the caps he recommends is not a problem.
However
 >  > I did increase his 7 series string to 8 in series and adjusted the
value
 >  > (like you did) to get the same effective capacitance.
 >
 > I got a big ol bag of 100 2.5kV 4700pF Centralab ceramic caps off Ebay
 > for $10 including shipping :-)!
 > As far as design goes, I wanted this filter to have more capacitance to
 > set the roll off frequency around 70kHz.  My coil should resonate just
 > above 100kHz which will give me 5 or 6 dB of attenuation back to my
 > NSTs.

Go for it!!!


 >  > 4. I'll let others comment on the two safety gaps.  I thinking that the
one
 >  > center grounded safety gap will do the entire job (protect both common
mode
 >  > and differential mode overvoltage) if set properly.
 >
 > Your probably right about the second gap not being necessary.  That one
 > is already built into my SRSG at present.   Also, my previous set up
 > never had the grounded gap anywhere.   That was one area I got a little
 > sloppy in.   I think the center ground gap will be a great improvement.

Certainly doesn't hurt.

Good Luck with your coil.

Gerry R
Ft Collins, CO