I have a CW voltage multiplier that uses 30kV, 1nF capacitors and doubled 20kV, 5mA, 100ns diodes (for a total of 40kV). My ac source is a sine wave (viewed on the primary side of the transformer because the secondary is beyond the capacity of my oscilloscope) around 20-24kHz and produces 15kV DC when half-wave rectified (measured after the 150Mohm resistor.)


simulate this circuit – Schematic created using CircuitLab

The output voltage of the CW multiplier is only 1.5kV. I have tested the diodes and capacitors. 100ns is compatible with most frequencies below 1MHz, I believe.

I am thinking that it might be perhaps that the 12 \$\Omega\$ resistor is not allowing enough current to charge the capacitors, but maybe it is some thing else such as some reactance I am not taking into account?

  • \$\begingroup\$ Arcing (or almost invisible leaking) through the air comes to mind. I don't know the size or arrangement of your parts. Can you consider immersing this in pure mineral oil and also take a moment to make sure that the distances between nodes are appropriate? Have a look here for an example of what I'm thinking about. \$\endgroup\$ – jonk Apr 13 '19 at 20:56
  • 3
    \$\begingroup\$ How are you measuring voltage? \$\endgroup\$ – user_1818839 Apr 13 '19 at 21:00
  • 1
    \$\begingroup\$ Consider R2 and the input impedance of whatever you are using to measure voltage. \$\endgroup\$ – JRE Apr 13 '19 at 21:08
  • \$\begingroup\$ @jonk the multiplier is immersed in mineral oil and the 15kv half wave rectification of the input will charge the capacitor to 15 kilovolts. \$\endgroup\$ – user198606 Apr 13 '19 at 21:53
  • \$\begingroup\$ @Brian Drummond two ways. the first with a 100 volt voltmeter with the voltage split between the internal impotence of the voltmeter and an add-on resistor in the Mohms. The second, with a regular multimeter with a hundred times probe. The meters work just fine when measuring 15 kV when half wave rectified. \$\endgroup\$ – user198606 Apr 13 '19 at 21:58

Appearently the diodes aren't as good quality. I added a third diode to each pair and it works fine.

  • \$\begingroup\$ Do you need small capacitors across each diode, to encourage voltage splitting to be even? \$\endgroup\$ – analogsystemsrf Apr 14 '19 at 11:08

If it is only 1.5kV it will not spark over a 5mm gap. The general rule is 30kV per cm depending on humidity, electrode profile etc. That will tell you whether it is the circuit or your measurement technique that is wrong.


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.