RE: NST Transformer ThougHts

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

Original poster: "Malcolm Watts" <m.j.watts-at-massey.ac.nz> 

Hi Terry,
          A couple of comments:

On 7 Sep 2004, at 21:47, Tesla list wrote:

 > Original poster: "rb" <randyaz-at-cableone-dot-net>
 >
 > All I know is that without your filter my NST's cook...
 >
 > -----Original Message-----
 > From: Tesla list [mailto:tesla-at-pupman-dot-com]
 > Sent: Tuesday, September 07, 2004 11:56 AM
 > To: tesla-at-pupman-dot-com
 > Subject: NST Transformer ThougHts
 >
 >
 > Original poster: Terry Fritz <teslalist-at-twfpowerelectronics-dot-com>
 >
 > Hi all,
 >
 > With Teri Mekenney's ID of the transformer and D.C.'s info on
 > windings*, I was able to "very roughly" work out the fluxes in the
 > NST.
 >
 > The transformer is an Allanson 15kV 30mA:
 >
 > http://hot-streamer-dot-com/temp/TRN06.jpg
 >
 > The dimensions are:
 >
 > http://hot-streamer-dot-com/temp/TRN-00.gif
 >
 > After the usual messy calculations:
 >
 > http://hot-streamer-dot-com/temp/TRN-01.gif
 >
 > One can see that the reluctances (sort of like resistance) in the
 > outer path and inner path are fairly equal (given the air gaps) at
 > ~~280 A-t/Wb:
 >
 > http://hot-streamer-dot-com/temp/TRN-02.gif
 >
 > When the transformer is not loaded, the flux is probably close to the
 > beginnings of saturation at 120 VAC input. I would imagine that a
 > scope look at the primary current of an unloaded NST shows this
 > saturation beginning as peak current distortions.
 >
 > http://hot-streamer-dot-com/temp/TRN-06.gif
 >
 > Interestingly, the saturation seems to start right at 125VAC!  the
 > waveform is always distorted like it shows though at any voltage.
 >
 > The center area of the core where the primary is wound has about twice
 > the flux of the rest of the core, so that is where saturation will be
 > most likely to occur.

Not surprising.

  The shunts are thinner but they are somewhat
 > helped by the air gaps (very small). However, they are certainly next
 > in line for saturation.

Are they? Even a tiny airgap has a pretty high reluctance. If the
shunts actually did saturate, the transformer's current limiting
ability is reduced to dependence on the winding resistances. Is the
total shunt x-sectional area comparable with that of the primary leg
or is it smaller? That would give some indication whether the shunts
are likely to saturate.

 > http://hot-streamer-dot-com/temp/TRN-04.gif
 >
 > If the secondaries are loaded, their flux opposes that of the primary,
 > so secondary load will counteract the saturation.

I can't agree with that view. The nett flux in the core has to
support a transfer of power from pri to sec doesn't it?

 > http://hot-streamer-dot-com/temp/TRN-03.gif
 >
 > This probably explains why NSTs that are firing do not go into this
 > ferrorresonance mode while NSTs run just with a cap on them do.

My spin on it is that a firing NST prevents winding currents from
building up to saturation levels rather than having any direct effect
on flux opposition (Y-N)?

  I
 > would guess that an unloaded transformer that will go into saturation
 > at say 140VAC will need much higher input voltage to saturate with a
 > load on the secondary (I am not an expert in transformer things...).
 > Of course, with a cap on the output as a load, things must get very
 > odd. I would guess that the secondary current would see substantial
 > phase shift so that the secondary counter flux may tend to further
 > increase saturation? At some point, the capacitance on the secondary
 > and the effective inductance in the transformer hit a 60Hz resonance
 > and the thing goes into this loud ferroresonant humming mode.
 >
 > Not sure what it all means. I am certainly not an expert in this stuff
 > so beware of newbi errors...
 >
 > Perhaps we can sort of play with the counter flux and ferroresonance
 > to get super powers out of NSTs. There is already evidence this can be
 > done as shown by Gary Lau's and Gerry's coils that seem to work better
 > than they "should". But right now there is no control or science
 > behind it aside from good luck.
 >
 > Much to ponder...
 >
 > Cheers,
 >
 >  Terry

I'm sure the problem isn't intractable. A few turns around each leg
and shunt and monitored under various loading conditions should bring
out the answers.

Malcolm