# Re: spark gap voltages (Secondary capacitance)

```Original poster: "Dave Larkin by way of Terry Fritz <twftesla-at-qwest-dot-net>" <teslaman15-at-hotmail-dot-com>

> > A small DC supply is used to charge the primary cap, until the spark gap
> > breaks down.  The output spark length under these conditions _is_
>(almost) a
> > direct relation of voltage.  So if a ground terminal with a large ROC
>(to
> > try and make the field a bit more uniform) is used and the single shot
>spark
> > length measured, one can determine the approximate output voltage, using
>the
> > fact that air breaks down at ~1MV/meter for large gaps.
>
>The correct figure is 30 kV/cm for parallel planes, or terminals where
>the radius of curvature (R) is much larger than the distance between the

I am well aware that the textbook breakdown for air is 3MV/meter.
1MV/meter is a crude approximation to account for the fact that the field is
seriously non-uniform.  The equation you posted below seems a much better
way of doing things!  Is there another equation which accounts for inequal

The really accurate method for the test I describe is not to rely on a
calculated voltage, but to hook an EHT power supply up to the (formerly)
grounded electrode, and actually measure the voltage taken to break down the
gap.  However the requirement for a multi hundred kilovolt test supply means
that for most people it'd probably be easier to simply build the fiber optic
probes!

-Dave-

>terminals (D). When D>>R the voltage tends to be determinated by R only,
>as V=60000/R (R in cm, assuming 2 identical balls). An approximate
>expression for the voltage between 2 balls with radius R and distance D
>is:
>V=30000*R*D/(0.9*(R+D/2)), R and D in cm.
>A spark with 20 cm of length between two balls with 2 cm of radius