Re: supercaps

[Leigh Palmer <palmer@sfu.ca>]

> > You can demonstrate it with a Leiden jar as well. In fact I
> > do it with a disassemblable Leiden jar. I can charge the jar and remove
> > the aluminum electrodes. After touching them together, I replace them and
> > a healthy arc can be drawn off the jar. After a minute or so I can draw
> > another spark.
>
> Very interesting; I did not know about multiple sparks in jars. What does
> the removal of electrodes have to do with the showing of these sparks?

The removal has nothing to do with the multiple sparks. It was confusing
to add that, but I was sleepy. Sorry.

The removal merely shows that the energy is stored in the polarized glass.
The multiple sparks demonstrate that the polarization relaxation time is
more than a second (you can firmly short the Leiden jar or picture tube
for a second and it will still rebound. In the case of the picture tube
I think this rebound grows over a time period of hours.

> How do manufacturers determine C; from the discharge rate or from the
> defining equation, Q=C*V? For your Leyden jar both of these approaches
> are likely to yield practically the same C. I found the same to be true
> for the non-electorolytic and for the "ordinary" electrolytic capacitors
> (C<2200 microfarads) which were tested. Supercaps seem to be unique in
> that respect.

It would be my guess that the capacitance is related to the capacitive
reactance at some appropriate application frequency rather than to either
of the alternatives you suggest. That is the way capacitance would be
measured in a quality control situation, with a reactance meter or bridge.
As I said, real capacitors are not precisely linear devices, so all three
"definitions" are appropriate to characterize a real capacitor.

Leigh