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

Re: PSpice modeling of spark gaps

Hi Malcolm,

	The linear decrement was easy.  Once the losses were added to the models,
the decrement fell off as a nice linear ramp just like in the real Tesla
coils.  I have posted a picture that shows and actual primary current
waveform and a modeled waveform at:

http://www.peakpeak-dot-com/~terryf/tesla/misc/malcolm1.jpg

Sorry, the model waveform was a little dim.  This is a big file and you may
have to print it to see all the detail.

	You can see that the timing, decrement, and most of the important
parameters match.  The zero current crossing spikes are missing from the
model waveform and thus Gary Lau is concerned with adding them to the
model.  The resistance I assume accounts for this loss but it would be nice
if the zero crossing spikes showed up in the models too.  The initial spike
needs work as well.  When the gap first fires, there appears to be
capacitances being shorted and such.  I will have to zoom in and find the
ringing frequencies and add parasitic components to reproduce this in the
model.
	I have some tricks I will try to add in the crossing spikes.  Gary is
right in wanting these in the models since they are so pronounced in the
actual waveform measurements.  As time goes on, These details will become
more important now that the models are basically working well.  In fact, it
is scary how well they work!

	Terry



At 02:48 PM 11/9/98 +1200, you wrote:
>Hi Terry,
>
>
snip>> 
>> Aaah!  Malcolm too is playing with models! :-)
>> 
snip>
>Actually I'm not (at the moment anyway). Just asking a question. If 
>a tuned circuit with a gap doesn't have a linear decrement it isn't 
>being modelled accurately, at least for the L/R ratios one typcally 
>finds in a good tuned circuit. Will check for outlandish fixed 
>resistance values.
>
>Malcolm
>
>