RE: MMC resister problem

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

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

Hi Gary,

Lead inductance and the high frequency resonances could have a lot to do
with it.  This would be especially true when the gap first fires and we get
those giant tens of MHz spikes.  One would have to use a higher order model
for the caps and then put in stray inductances and capacitances and then
simulate the true currents going into it.  I would guess it is "messy"
then.  Leakage currents would also be a problems especially if the higher
frequencies are considered.  

The caps may see the same current but they certainly do not see the same
voltage with respect to ground.  The center caps tend to see near ground
level voltages while the outside caps see voltages like 10kV above ground.
It is possible that corona (a diode affect to a point) is unbalancing the
end caps through corona leakage.  It all can really get very complex.  The
simple resistors seems to fix everything.
 
Cheers,

	Terry


At 08:07 AM 6/5/2002 -0400, you wrote:
>
>A series-string of caps will all experience the same current flow and 
>(assuming they are all roughly the same capacitance value) will all gain the 
>same incremental voltage as they charge.  If one cap starting at zero volts 
>charges up to 2000V, another cap starting out at 500V will charge up to 
>2500V.  The voltage on one cap can't affect the charging of others.
>
>In a perfectly linear world (i.e. circuit simulators), all caps starting out 
>at zero volts WOULD guarantee that they charge and discharge uniformly and 
>all end up at zero volts.  But it appears that some non-linear effects are 
>complicating things.
>
>Gary Lau
>MA, USA
>