1
\$\begingroup\$

I am a physics student trying to build a proportional counter. I am having huge trouble dealing with possible ground loops in designing the system.

The current design is attached below. Could anyone point out where I got it wrong?

Current design

Schematic of the preamplifier: Schematic of the preamplifier

The capacitance of the proportional counter is around 400 pF.

Schematic of the isolated DC/DC converter: Schematic of the isolated DC/DC converter

Schematic of the high voltage DC/DC converter: Schematic of the high voltage DC/DC converter

The high voltage is intended to run at around 600 V.

Signal output from the preamplifier with everything connected except high voltage: Signal output from the preamplifier with everything connected except high voltage

Weird unwanted tail pulses appear in the output once the high voltage connected: Weird unwanted tail pulses appear in the output once the high voltage connected

Also appear as negative pulses: Also appear as negative pulses

Correct signal output I am looking for: Correct signal output I am looking for

I tested the preamplifier and the power supply (without high voltage) with a function generator and they behave fine. The problem arises when I connect everything to the detector with the high voltage enabled. The signal output from the preamplifier has a seemingly random tail pulse (~ 10 Hz, ~2 ms rise time, ~20 ms decay time) that completely masks the tail pulse signal from the detector (~ns rise time, ~200 μs decay time.) The output from the high voltage alone is stable with less than 10 mV ripple noise.

P.S. Further investigation indicates that the high voltage power supply might be the culprit. What is the proper way to monitor the behavior of the high voltage power supply under load using an oscilloscope? Could I just use a voltage divider to safely monitor its behavior?

\$\endgroup\$
17
  • \$\begingroup\$ Can you post the schematics of the sub assemblies? \$\endgroup\$
    – Lior Bilia
    Commented May 4, 2023 at 17:59
  • \$\begingroup\$ @LiorBilia Thanks for your reply, see the updated post for schematics. \$\endgroup\$
    – Martin
    Commented May 4, 2023 at 18:32
  • \$\begingroup\$ Welcome! Please post oscillograms of your mystery 10 Hz signal. \$\endgroup\$
    – winny
    Commented May 4, 2023 at 19:12
  • \$\begingroup\$ What do you mean by insulated and non-insulated connector? Is the latter a normal coax connector with the shell connected to a metal chassis? So that would make the former a coax connector with the shell insulated from the chassis? \$\endgroup\$
    – SteveSh
    Commented May 4, 2023 at 20:40
  • \$\begingroup\$ Does the high voltage go to the detector? \$\endgroup\$
    – SteveSh
    Commented May 4, 2023 at 20:42

1 Answer 1

1
\$\begingroup\$

Regarding your design: you might want to add a 10 kΩ resistor in series with your adjustment potentiometer. You don't want to short Vref to ground.

Monitoring high voltage with an oscilloscope: There are high voltage differential probes available from various vendors. They provide a safe method to probe such voltages.

EMC: Depending on the source of noise, few options exist to mitigate it in such a setup: ferrite beads, feed through capacitors or better PSU.

\$\endgroup\$
3
  • \$\begingroup\$ Thanks for your suggestions, I will look into these options. \$\endgroup\$
    – Martin
    Commented May 4, 2023 at 20:49
  • \$\begingroup\$ Besides the 10K resistor, you also might want to add a decoupling cap to the Vadj input of the high voltage chip. Any kind of noise on that input will disturb the high voltage output. \$\endgroup\$
    – SteveSh
    Commented May 4, 2023 at 21:55
  • \$\begingroup\$ @SteveSh Thanks for pointing it out, I will do so. \$\endgroup\$
    – Martin
    Commented May 4, 2023 at 23:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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