Spark Gaps.

    A few years back I was doing an experiment at home, using a spark gap,
and a multimeter to measure the current.  I was getting some random
readings, as I sent one pulse through the meter at a time, when I realized
something.  Solids, like the copper the wire is made of have a small amount
of stray capacitance, and a low voltages it low, and at higher voltages it's
higher.  This is a result of the number of electrons present, and the
potetial of ionization.  So, the more protons, and nutrons that are present
in a substance the more potential to store more electrons is there.  So, a
block of metal has a value of capacitance, and the electrons will be stored
as a result of sharing the same orbits of other electrons, and fill some of
the spaces inbetween molecular bonds.  This explained the random readings of
my multimeter, one the sample rate, two the charging upto the gap, and the
high current associated with the release of current when the arc actually
took place.  Stray capacitance places a charge time on any give piece of
wire, and until that charge has built up to the voltage required for the
spark to jump the gap current cannot flow, but only build up.  So, the
shorter the time period, and the measurments taken from the moment closer to
the discharge of the voltage, the higher the current reading, and the
farther from it the lower.  The sample rate of the DIMM showed a current
that was much higher than I had applied.  Copper is also a good capacitor at
a high voltage, but the capacitance has to do with how much voltage is
required to pack the electrons into a solid.  A block of metal that is
electrically isolated by itself on one side of a spark gap will act as a
capacitor without a second plate.  Or a large plastic bucket filled with
salty water, because the electrons will have to fill the vacant space
associated with it's volume before an arc can cross the gap.

    I tested 20 ft. long sections of copper wire, and the capacitance was
expressed in a static change in length, before the arc.  I was using a 45 KV
CTR DC supply to test it, and the wire shrank when the arc was discharged.