2
\$\begingroup\$

Imagine one wanted to measure the discharge from a Van De Graaff generator using an oscilloscope. The voltages are way beyond what a typical scope and probes can handle. I was thinking about how a voltage divider could be used to make this possible, but I am aware that the overall resistance of the voltage divider circuit would slow down the discharge from the Van De Graaf (since it's effectively a capacitor discharging).

What would be an effective way to measure the discharge from something like this?

PS: I am aware that you can calculate it, as you know the capacitance of the metal dome from its radius, but I'm curious how this is measured.

\$\endgroup\$
  • 1
    \$\begingroup\$ It's not clear what you mean or are asking for. \$\endgroup\$ – Andy aka Mar 31 '17 at 15:06
  • \$\begingroup\$ @Andyaka I believe the OP is asking how to measure the voltage/monitor the waveform of the output discharge of the VDG generator using an oscilloscope. \$\endgroup\$ – DerStrom8 Mar 31 '17 at 15:07
  • \$\begingroup\$ Yes, that is what I'm asking. \$\endgroup\$ – Kurt Newman Mar 31 '17 at 15:08
  • \$\begingroup\$ not really that much into HV electronics, but wouldn't these discharges be extremely transient, and the main problem being that introducing parasitic reactive components into the whole system would totally change the things seen on a 'scope? \$\endgroup\$ – Marcus Müller Mar 31 '17 at 15:22
  • 1
    \$\begingroup\$ If you wanted to measure the charged voltage of a van de graaf generator, you'd probably use an electroscope rather than a voltmeter or oscilloscope. \$\endgroup\$ – JRE Mar 31 '17 at 16:38
1
\$\begingroup\$

Usually for high voltage they use Capacitance dividers rather than resistive dividers.

schematic

The 1st stands about 1m tall and rated for 100kV, 2nd is rated for 400kV is 1~2m

and 2000kV not shown would be very tall, really!

Double donuts reduce e-field gradient which otherwise would lower breakdown voltage.

enter image description here

When testing HVDC, it becomes more of a challenge for testing as ANY insulator surface oxidation becomes a transport for charge flow and reduced breakdown threshold, starting with repetitive partial discharge or "corona" is visible before BDV is reached.

Testing a 5MVA transformer 40kV, a 200kV bushing (insulator) for me arc'd at 90kVdc. Since BIL200 only means for 200kV lightning impulse rating it failed below 100kV even after alcohol wipe, so I had to add 2nd bushing in series to test over 150kVdc for a contaminated oil insulation test I tried out. It ticked like a FET relaxation oscillator and an AM Radio could easily hear it. Even the epoxy paint got charged up over steel tank and gave little shocks touching it during testing.

enter image description here

\$\endgroup\$
  • \$\begingroup\$ What are those second F's? \$\endgroup\$ – Oskar Skog Mar 31 '17 at 15:38
  • \$\begingroup\$ I like that better.. though again I Might use a long chain of caps so you can stack them to the height of the VDG and minimize the effect of the wires on the E-Field. \$\endgroup\$ – Trevor_G Mar 31 '17 at 15:41
  • \$\begingroup\$ also. this is "fine" for measuring the voltage up there, but no so convinced it would accurately reflect the discharge curve. \$\endgroup\$ – Trevor_G Mar 31 '17 at 15:48
  • \$\begingroup\$ These are used to accurately measure Lightning transient tests in Transformer factories. Output goes to 50 Ohm coax with 1ns DSO. The stand includes both caps to some standard ratio. A long fiberglass pole and gnd wire strand discharges the divider for safe moving after tests. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Mar 31 '17 at 15:57
  • \$\begingroup\$ I have used these up to 200kV and they are limited by external line inductance only rise time. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Mar 31 '17 at 16:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.