Re: Weird safety gap behaviour

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

Original poster: "Dr. Resonance" <resonance-at-jvlnet-dot-com> 


We've done that while designing our new xmfr.  The current limiting feature
is very clever in the NSTs we dissassembled.  Most use a rectangular core so
one straight leg is removeable.  This straight section can be tapped in and
out of place with a wood mallet but it is a very tight fit.  This makes for
easy winding and placement of the primary and HV coils over the straight
core leg.  Most coils are wound on thick cardboard tubes covered with mylar
for some good HV and tracking resistance.  Most NST performance isn't
limited by the core/coil design but more by the tar-like insulation used.
Heating and cooling cycles can produce cracking over the years.

The current limiting is produced by spacing the HV coils further away from
the core thus creating a lossy xmfr.  With our xmfr we are presently testing
7 different HV coils.  Each coil has more insulation between coil and core.
Each coil will produce a slightly different max current value under short
circuit conditions.

Another interesting advantange to current limiting in this manner is that
you can leave an extra width of mylar with a reasonable heavy build (250 -
300 mils) between inside of coil and core.  This will allow xmfrs to be
connected in series with single end grounded if desireable.  Or, run as a
classic CT grounded xmfr.

This design idea may also be carried over the MOTs to make them current
limited as well or to use them as current limiting inductors in series with
potentials xmfrs, etc.  One would wind up a variety of test coils with
greater spacing between core and inside of coil and then do some current
measurements under short circuit conditions.

For our initial design the core was 3.5 x 3.5 inches cross section for 4 kVA
operation.  This is running the core close to its limits but should work
fine if not abused.  Length of core is 18 inches x 8.75 inches high (outside
of core dimensions).

Dr. Resonance

 > >
 > > > Original poster: "Gerry Reynolds" <gerryreynolds-at-earthlink-dot-net>
 > > >
 > > > Are you saying the equivalent inductance of the NST is decreasing as
 > > > the amplitude is increasing and thus at low voltage (Cp=2.5 Cres), the
 > > > Fres is 38 Hz and when Lnst decreases, the Fres moves up closer to
 > > > 60Hz   If so, is this due to shunts saturating?
 > > >
 > > > Gerry R.
 > >
 > >Not the shunts, the core itself. The shunts would be most unlikely to
 > >saturate given that they have a significant airgap in series with
 > >them.
 > >
 > >Malcolm
 >
 >
 > Someday....  Somebody, needs to measure out an NST's core dimensions,
 > number of turns, etc, so we can do the math on the core, shunts, air gaps
 > and all to try and figure out what is really going on magnetically inside
 > and NST...  Sort of reverse engineer the thing magnetically and then maybe
 > we would know a bunch more...
 >
 > I think Gerry has an NST he is depotting.  I have one but I don't know the
 > original specs at all...  We just need the basic data, and it is all
"easy"
 > first year EE stuff then...
 >
 > Cheers,
 >
 >          Terry
 >
 >
 >
 >