I have a pulsed power supply that provides 250v pulses with a pulse width of 2 uS and a frequency of 10 Hz. However, I'm unaware of the polarity of these pulses.

In order to detect the polarity of the pulses, I'm planning to place a diode in series with the output of the pulsed power supply (which only has a single output lead). Depending on the polarity of the voltage, the pulse will either pass through the diode or be defected.

The diode is followed by a charging resistor (20 ohms) in series with a 4 nF capacitor. A voltmeter probe makes contact with the other pole of the capacitor, with the opposite voltmeter probe being grounded. The set up may be shown below:


simulate this circuit – Schematic created using CircuitLab

Based on my specs, I'm expecting to receive some sort of a signal/blip on my multimeter (Fluke 87 III True RMS Meter) to provide some evidence of the presence of a pulse.

Please note that I am not looking to measure the pulse width, voltage, or frequency of the pulses. I'm only looking to see if they are present after the diode.

EDIT 1: The question is whether or not this set up will work to detect a measurable signal.

EDIT 2: The diode shown is a 300v diode, the label shown was only auto-generated by the system.

  • 1
    \$\begingroup\$ Is there a question? 1N4148 will block 75V. \$\endgroup\$ – Spehro Pefhany Feb 19 at 16:10
  • \$\begingroup\$ @SpehroPefhany The question is whether or not my set up works, I'll adjust the question. I'll also fix the diode type, it was auto-generated. \$\endgroup\$ – James Li Feb 19 at 16:11
  • \$\begingroup\$ A single lead output? Pulse power must return to the power supply. Have you determined with certainty where/how pulse current is returned to the power supply? Your circuit shows a ground symbol, which is an assumed return path. \$\endgroup\$ – glen_geek Feb 19 at 16:16
  • \$\begingroup\$ @glen_geek The output of the power supply is also grounded. \$\endgroup\$ – James Li Feb 19 at 16:23
  • \$\begingroup\$ Will the RC discharge ever? \$\endgroup\$ – User323693 Feb 19 at 16:30

As shown it's pretty dubious. You'll get a big pulse the first time the diode conducts and not much afterwards after some pulses in that direction, and then reverse pulses due to the diode capacitance, which may be larger than the forward ones.

If you just put an RC (eg. 1M/1uF) in front of the meter on DC volts you should read the voltage of 5mV average without any more parts.

You can model the meter by something like 10M resistor across the leads, followed by a low pass filter like 1M/0.1uF. But I don't think I'd feel comfortable feeding 250V pulses directly into the meter on the 200mV range.

| improve this answer | |
  • \$\begingroup\$ Theoretically in my schematic, if the voltage of the source turned out the be positive (for all intents and purposes) and the diode is not forward conducting, would any signal be sent to the voltmeter or would it be completely reflected? \$\endgroup\$ – James Li Feb 19 at 17:28
  • \$\begingroup\$ Some would get through, depending a lot on the diode. \$\endgroup\$ – Spehro Pefhany Feb 19 at 18:52
  • \$\begingroup\$ What should I look for on the diode to see if it will work? \$\endgroup\$ – James Li Feb 19 at 18:57

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.