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

Re: MMC advice



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

I'm thinking that the safety gap firing all the time at 30% variac might be
a clue when the safety gap with no load was set at 105% variac.  Sounds to
me like the system would have to be runing at (or very close to) resonance
for this to be the case.  With 4 strings of 6 caps (please verify that this
wasn't 6 strings of 4 caps), the Vdc rating would  be 9000V - very close to
the peak voltage of a 6000Vrms transformer.  I would bet that this 6000V
tranny could put out more that 6000V (I've measured a 12KV NST at no load to
put out 12.6KV.  Opening up the safety gap, I think, was the wrong thing to
do.  Rechecking the safety gap and looking else where for the cause for
small streamers might have been more prudent.

One suggestion might be to input the design to JAVATC version10.  The tool
will calculate the peak current for the physical geometry of the TC primary
and Cp and give a cross check to the tuning.  Another idea if everything
else checks out, is to scope the safety gap voltage (with cap removed) and
then run the coil for a very short time, turn power OFF,  and then see how
warm the caps are getting.  Peak current will be a function of the firing
voltage, Cp, and Lp.  RMS current will be a function of the peak current
parameters as well as the BPS (and maybe k).  It is difficult to determine
unless you have a picture of the current waveform and SW to calculate it.
Measureing the temp rise of the caps might be easier.  I would also recheck
the construction of the MMC array to ensure proper clearences are given
(between caps and between a cap and it's discharge resister).  Also make
sure good high voltage techniques are given to solder joints and lead
trimming.

Gerry R
Ft Collins, CO


 > Original poster: Mark Broker <mbroker-at-thegeekgroup-dot-org>
 >
 > 1200A across 5 strings of caps is only 240A per string - just over half
 > their rating of 432A peak.  Even if using an "unrecommended" design of 9x3
 > .150uF caps for .050uF 20kV, the current rating is still within
 > manufacturer's specified limits....
 >
 > FWIW, I have yet to come across a functional TC design that exceeds either
 > the peak or RMS current rating of the caps without also exceeding the
 > voltage rating by a good amount.  The only feasible designs I've been able
 > to come up with (using an Excel MMC sheet based off Terry's MMC_Calc2
BASIC
 > app) that would cause problems use extremely high break rates (well over
 > 500BPS), which cause excessive cap heating.  I could see big coils run
with
 > very low surge impedances causing peak current problems, but only if the
 > secondary has, say, 400 turns and micro-small toroid.
 >
 > That brings up a question with all the magnifier discussions going on: how
 > do the peak and RMS currents look in the primary circuit of a magnifier -
 > both optimally built using Antonio's research or just thrown together a la
 > Dave's Labor Day Coil?
 >
 > Regards,
 >
 > Mark Broker
 > Chief Engineer, The Geek Group
 >