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

Re: Spark Gap VI Scope Capture



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

Hi All,

I made some detail pictures of the scope captures.  No much time to comment 
on them now since I am making BIG deals on 15kVA varaics and pole 
transformers tonight :-))))  I seem to have some magnetic attraction to 
very large iron electrical objects tonight.  Better start lifting iron just 
to move this crap around!!!  But here they are for your pleasure.  Pretty 
darn interesting!!!

Just the whole waveform in MathCad.  Note the current is 10X and I have 
corrected for the 2.8 amp offset of the scope (was too lazy to recal 
it).  Note that the voltage dips for much of the cycle.

http://hot-streamer-dot-com/temp/Spark1.gif

This is the very start of the cycle.  Seems the gap starts cooking or 
something at first.  The big wave is about 2.7MHz.  Jim just wondered if 
these were "Trichel pulses"...  Looks like about 5 amps with 10 amp peaks 
(current x100 as seen by the multiplier on Bx).  It is interesting that the 
voltage goes above 3000 volts suggesting that capacitance may be collapsing 
or something odd like that.

http://hot-streamer-dot-com/temp/Spark2.gif

This is the gap firing.  The signals after the big dip in voltage are close 
to the bandwidth limits of the probes so phase may be screwed up.  Not the 
consistent ring in the current waveform.

http://hot-streamer-dot-com/temp/Spark3.gif

Just a closer look right after the gap fires.  This noise probably goes 
into the GHz...  But I must admit this is the cleanest signal I have ever 
seen of this stage of gap firing!!  Normally these signals swamp the test 
equipment with noise.

http://hot-streamer-dot-com/temp/Spark4.gif

Here you can see how the ring on the current dies down slowly during a 
cycle.  The loss must be extremely low in whatever causes this!!  You can 
easily see the zero current crossings where the gap "winks out".

http://hot-streamer-dot-com/temp/Spark5.gif

This is a zero current crossing where the gap winks out for ~50nS as noted 
by the width of the voltage spike.  This rings up the gap into the 
oscillation again.

http://hot-streamer-dot-com/temp/Spark6.gif

Just a look at the wave with the front cut off.

http://hot-streamer-dot-com/temp/Spark7.gif

Here is the last zero current crossing and the very quiet final quench of 
the gap.  The current is about 7.2 amps at the last hump and the tail 
voltage is 350 volts.  We can calculate the final energy left:

E = 1/2 x C x V^2 = 1/2 x 24e-9 x 350^2 = 1.47mJ

E = 1/2 x L x I^2  = 1/2 x 55e-6 x 7.2^2 = 1.43mJ

Pretty darn close and shows I must have done something right ;-))

http://hot-streamer-dot-com/temp/Spark8.gif

The quench and the 2.7MHz final voltage ring...

http://hot-streamer-dot-com/temp/Spark9.gif

This is interesting!  Note that as the gap voltage offset went to zero that 
the gap quenched:

http://hot-streamer-dot-com/temp/Spark7.gif

http://hot-streamer-dot-com/temp/Spark8.gif

I have no idea where this voltage offset comes from but it seems to 
certainly have an effect on quenching here!!

Much to ponder in this little spark ;-)))

Cheers,

	Terry