TSSP: List Archives

From: "B2"
Date: Mon, 9 Jul 2001 15:11:06 -0400
Subject: [TSSP] Measurement of Topload Voltage

Hi Terry, All,                                     -----e-mail 686- posted previously -----
    I have used a similar Ross probe in the lightning lab.  It was rather weakly put together and the connector broke off after a week of use.  It required the use of a 75 Ohm to 50 Ohm balun.  I found that two big toroids, spaced at tire thickness / 2 and coaxial, provided a field of sufficient uniformity to give a reproducable measurement of high voltages up to several hundred thousand volts.  A bnc connector is mounted coaxially in the bottom toroid with a tiny disk soldered to the center pin.  The size of the disk determines the voltage division ratio.  The capacitance between the disk and the lower toroid should yield a reactance of about 50 Ohms at the frequency of interest.  A neon bulb can be placed across the cable connection before it reaches the oscillscope along with the transzorb.

    Calibration was done by attaching the top toroid to ground through a water resistor of about 100 Ohms (low L/R) with a current probe.  A Marx charged to a known voltage is fired into the toroid.  The charging voltage times the number of stages in the Marx give the voltage reading within a few percent if the Marx erected capacity is at least ten times the probe capacity.  Voltage can also be back figured from the current probe reading by multiplying it by the resistance.  Any inductance of the resistor will tend to filter out the high frequencies.  If the calculated and measured Marx waveforms are close enough then the division ratio may be determined by comparing with the signal from the lower toroid bnc connection.  The high voltage will cause a great force between the toroids.  This will lead to a mechanical oscillation if not damped.  Any movement of the toroids will add an additional waveform to the measured waveform.  This may have to be accounted for if great accuracy is desired over any perion of time.

    I have used a similar calibration procedure to measure the output of the 2 MV Marx with just a simple wire antenna placed about 30 feet from the output toroid.  I think that the same procedure could be applied to the Tesla coil itself.  The Marx would be on the opposite side of the Tesla coil from the oscilloscope and antenna probe.  In this case the resistor would be applied across the Marx directly.  The Marx would be attached to the Tesla coil with a piece of copper tubing about five feet long.  After the voltage is back figured then the Marx is removed.  The antenna probe and Tesla coil must not be moved after calibration.  If the secondary goes into corona/streamers then a lot of filtering may be required.

    All of the foregoing assumes that a peak measurement is most desired.  A pure waveform requires a bit more work.

    If you still want to use the oil method then use copper sheets for the cylinders.  Solder copper tubing rings to the overlapping ends.  The outer cylinder has the copper tubing ring tangentialy soldered to the outside edge.  The inner cylinder has the copper tubing ring tangentially soldered to the inner side.  Sulfur hexafluoride gas should be available by ordering through any welding supply place.

Cheers,

Barry



Maintainer Paul Nicholson, paul@abelian.demon.co.uk.