[Prev][Next][Index][Thread]
Ideas for measuring voltages of Tesla coils
----------
From: Jim Lux [SMTP:jimlux-at-earthlink-dot-net]
Sent: Wednesday, March 11, 1998 9:46 AM
To: Tesla List
Subject: Re: Ideas for measuring voltages of Tesla coils
>
>
> ----------
> From: John H. Couture [SMTP:couturejh-at-worldnet.att-dot-net]
> Sent: Monday, March 09, 1998 10:45 PM
> To: Tesla List
> Subject: Re: Ideas for measuring voltages of Tesla coils
>
>
> Jim -
>
> The methods mentioned below sound good but have the same problem as
many
> other methods. How do you calibrate the devices?
The Kerr and Pockels effect are fundamental properties of the material
dependent on the E field. Generating a precise E field for testing is quite
easy: 2 large plates a known distance apart with a known voltage on them
will create a fairly uniform field. For an even better field either a
Rogowski or Bruce profile should work. The tricky part is going to be
integrating the field gradient over some path from the electrode to ground.
You'd have to carefully measure the distance you traverse between each
measurement. You'd also have to measure in three axes to make sure that you
didn't wind up with a cos(theta) error (i.e. make sure the cell is
perpendicular to the local E-field)
A further complication that would need some attention is the perturbation
of the E-field because of the different dielectic constant of the cell
material, which would change the local field in the neighborhood of the
cell. Although, the effect would certainly be a whole lot less than that of
any metallic probe.
I note that the electrooptical properties of fiber optic fibers has been
used as a sensor in EMP testing where you don't want the sensor to perturb
the measurement. Sensitivity might be an issue, although it would be easy
to get, say, 1000 meters of fiber as your "cell" in a small volume. I don't
recall the details, but they may have been doing B-field measurements here
with the Faraday effect. But, for AC signals, B & E fields are related
(e.g. B field is 1/(3E8^2) * dE/dt)