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

Protection networks (was RE: LOW VOLUME)

I have to disagree with the notion that an effective NST protection network
may be made without the use of bypass caps.  I maintain that bypass caps are
essential in any L-C or R-C protection networks.  Please use courier font to
view:

L-only circuit:
GAP---CHOKE---NSTsecondary---center-ground

L-C circuit:
GAP---CHOKE-+-NSTsecondary---center-ground
            |
        BypassCAP
            |
         ground

In the L-only circuit, the high frequencies present at the gap terminal sees
two inductors in series - the choke, and the inductance of the NST
secondary.  The RF voltage present at the NST secondary terminal is
determined by the simple ratio of the two inductances.  The protection choke
is typically a few mH, while the NST secondary is hundreds of Henries.
Since the inductance of the chokes is so drastically smaller, almost no RF
voltage is dropped across it, and the RF attenuation is negligible.

In the L-C circuit, the reactance of the NST secondary at RF frequencies is
high compared to that of the choke and bypass caps and may be ignored.
What's left is a simple L-C low-pass filter which is very effective at
attenuating RF frequencies.

The analysis of the L-only circuit was simplified, ignoring the capacitance
from the NST secondary winding to ground.  This capacitance does not count
towards making it an L-C network however, as it is a distributed
capacitance, not a lump at the terminal, capable of shunting the RF to
ground before it enters the NST case.

After having said all that, I must also suggest the use of R-C networks over
the use of L-C networks.  A pure L-C network, while very effective at
filtering the tank circuit's frequencies, will oscillate at it's own
resonant frequency at each gap firing, with equally destructive voltage
amplitudes.  Adding series resistors to de-Q the inductors will reduce the
duration, but not the amplitude of the oscillations.  An R-C network, while
somewhat less effective than an L-C network in terms of RF attenuation, has
no self-resonance oscillations.

Regards, Gary Lau
Waltham, MA USA


		-----Original Message-----
		From:	Tesla List [mailto:tesla-at-pupman-dot-com]
		Sent:	Tuesday, December 07, 1999 1:45 PM
		To:	tesla-at-pupman-dot-com
		Subject:	Re: LOW VOLUME

		Original Poster: Esondrmn-at-aol-dot-com 

		In a message dated 12/6/99 11:48:27 PM Pacific Standard
Time, 
		tesla-at-pupman-dot-com writes:

		> Hi Coilers:
		>  
		>  Is it just me, or has there been a marked decline in the
volume of
		>  messages here on the SIG lately?  Maybe everyone's too
busy with
		>  Christmas shopping!
		>  
		>  I'm well along with my second coil.  Four by sixteen
secondary with
		>  an 11" x 4" toroid top load, 15 degree 1/4" tube saucer
primary, 
		>  RQ multi-gap, mica cap, and two OBIT's for power.
		>  
		>  I managed to fry one of the OBIT's setting the safety
gaps, but I've
		>  got access to lots more.  Is 100 pF. across the safety
gaps OK when
		>  using a pair of pancake chokes wound of # 22 on a 2" OD
teflon form?
		>  
		J.B.

		There are may opinions regarding filter systems.  I would
suggest to use no 
		capacitors at all.  I use series inductors made with about
20 turns of #16 
		wire on a ferrite (donut) core in series with the H.V. feeds
and then 
		directly to the main spark gap.  I also use some series
resistance on the 
		smaller coil powered by neon sign transformers and none on
the pig powered 
		coil.  I believe it is Malcolm that suggests locating the
transformer 
		physically close to the spark gap and I agree this is the
best design.  After 
		blowing a few caps I also use safety gaps across my caps,
although after 
		fixing my 60 hz resonant problem I have not had a problem in
that area.

		Ed Sonderman