Re: MOVs and NSTs

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

Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>

Hi Peter,

"I" think 90% of NSTs die (used to) because people used resonant cap sizes
and "somehow" they let the voltage on the NST resonate up far too high.
The MOVs and safety gaps hit that problem twice over.  The safety gaps take
most of the problem (and scare the user to fix what is wrong) but the MOVs
are there if the user decides to start widening the gaps.  Many NST deaths
were caused by people trying to get better output by widening gaps :-(
MOVs don't let them do that :-))  

The other problem is what happens on the NST windings if 300KHz at 20kV
hits them?  The first outer windings are going to see the vast majority of
the current due to the giant inductance of the windings.  If you put 20kV
at 300kHz across the first 100 turns of the NSTs secondary, it's gonna blow
up.  Thus the RC filtering section to attenuate any RF.  That is a
"theoretical" effect and not a proven one, but the research there is real
though :-))  I thought of getting blown NSTs shipped to me for autopsy, but
that was not practical...

If one simply had well adjusted safety gaps, that would protect 90% of the
NSTs out there.  However, as I looked over NST failures and reports of what
went wrong, the "user" was a big problem :-))  So I built a lot of stuff in
like MOVs to take user error out of the picture.  I also over rated the
components like the resistors so super hot resistors (300 degrees C is
normal for power resistors) would not get people "bent out of joint" over
that :-))

So it has a lot of "human factors" and mistake buffers built into it :-))
I decided if I was going to design a filter for NSTs, it should not only
protect against over voltage and RF but might as will protect against user
error too.  I was thinking far beyond the electrical theory :-))

If someone wants to spend $50 on an NST filter, they deserve to get the
best known protection out there.  Would less work?  Sure, sometimes...

The input fuse protects against a later discovered effect of NST core
saturation.  With a large primary cap, the NST current can overcome the
limits of the NST shunt cores and the current will become unlimited to a
degree.  Sort of cool that a small NST transformer can put out pig
currents, but deadly to NSTs...  A simple input fuse stops core saturation
problems right quick.  Maybe that effect could be controlled to great
advantage, but too scary to play with for even me :-))

So I did the best I knew how to save NSTs with all the over kill and over
design I could...  There is a chance that the RC filter part is not
terribly needed, but I put a lot into "my" NSTs and I was not feeling lucky
:-))

NST used to blow like popcorn, but since there has been so much effort into
protecting them and LTR stuff, NST death is relatively rare.  So we must
have done something right :-))  I note that this has become "Terry's"
filter.  I really didn't want to take a chance of screwing it up :-)))

Cheers,

	Terry


At 05:18 PM 7/24/2001 -0700, you wrote:
>Terry,
>      I've seen your simulation outputs of the difference between connecting
>the NST power supply across the tank cap verses across the spark gap, and
>I am wondering why keep the RC filter in the second case with power across
>the SG.
>
>It still seems to make sense to me to have safety gaps and/or MOVs, but if
>there is no RF ringing to the NST when it is connected across the SG why
>have the RC filter.
>
>-Peter Lawrence.
>
>(ps, I have not blown an NST since I started using safety gaps and started
>connecting my NST across the SG. I sometimes include an RC filter, but not
>always, and not having the RC filter does not seem to have cost me anything
>yet..... knock on wood.....)
>
>
>
>>
>>Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>
>>
>>Hi All,
>>
>>Using MOVs with NSTs seems to be catching on :-))
>>
>
>