I have recently purchased an SBM-20 Soviet Geiger-Muller Tube, which I want to use in a Geiger Counter circuit. I powered it with a 12V to 400V DC-DC power supply, connected in series with a 5.6 megaohm anode resistor. However, when I connected the output of the Geiger tube to an LED, it stayed lit continuously. From what I understand, a Geiger-Muller tube should be off until an incident, in which case a short pulse is produced, then turns off again. I have barely used this tube, and always used a voltage within it's operating range, with a minimum 1MOhm resistor in series. I double-checked the polarity of the tube, and made sure the two terminals were not shorted. Is this tube defective, or am I doing something wrong?

  • \$\begingroup\$ There is an operational region (continuous discharge) that resides past the Geiger-Mueller region. Is it possible that your SBM-20 tube is operating there? Can you reduce the voltage a bit? (It might also be that the tube is malfunctioning, as well. Do you have several of them to try?) \$\endgroup\$ Jun 22, 2023 at 4:50
  • \$\begingroup\$ How are you stepping up the voltage? Might it be a high frequency inverter? The tube may be acting like a capacitor and leaking some high frequency noise into the LED. If you're using a white LED with an internal phosphorous layer there are a lot of things that can start that glowing, (such as nearby high voltages, nearby radio-active materials, and so on). Always use extra caution when working with high voltages and especially with Soviet surplus components , good luck comrade Leo... \$\endgroup\$
    – Nedd
    Jun 22, 2023 at 5:50
  • \$\begingroup\$ I found this that can help ... ebay.com/itm/… , see "initial" and "operating" voltages. \$\endgroup\$
    – Antonio51
    Jun 22, 2023 at 6:08
  • \$\begingroup\$ @periblepsis My power supply was tuned to just around 398 volts, and according to the datasheet, it should still be well within Geiger-Muller region (350-475). I only have one tube. I will try reducing the voltage next time I test it, to see if it makes a difference. If I can't figure this out, I will buy a second tube to compare it against \$\endgroup\$
    – leo
    Jun 22, 2023 at 13:26
  • \$\begingroup\$ @Nedd I am using this boost converter module amazon.com/gp/product/B07T6L61D9 from amazon. From what I understand, it's generating a square wave then stepping it up through a transformer. I used a green LED, not a white LED, and the only radioactive material present was an Americium-241 source. However, I also tried the circuit while the Am241 source was not present, and it continued to exhibit the same problem. The only high voltage present was the 400V power supply, and some very-very small EMF from AC city power. \$\endgroup\$
    – leo
    Jun 22, 2023 at 13:31

1 Answer 1


Answer: The Geiger-Muller tube I was using was not operational. After purchasing a new tube, I reassembled the circuit and tried it again. The LED remained normally off, and briefly flicked on for each incident captured by the new Geiger Tube. I then rewired the circuit to use a small piezo speaker instead of the LED, which produced a very faint click for each incident. In order to amplify the noise, I added two BJT transistors in a Darlington Pair amplifier configuration, which worked flawlessly. Now, each incident was accompanied by a loud audible click of the piezo element! I took this opportunity to try out a radioactive test source (Th-232), which as expected, produced very rapid successive clicks when brought close to the tube.

Thank you very much, everyone who helped! I think the LED was continuously on with the old tube due to some high frequency harmonics from the 400V boost converter circuit, as @Nedd mentioned. Hopefully this experience and the answers provided can help a future noob troubleshoot their Geiger. Time to graduate to the nuclear physics StackExchange..

  • \$\begingroup\$ To be perfectly clear, the new tube was an identical SBM-20 geiger tube. \$\endgroup\$
    – leo
    Jul 4, 2023 at 3:58

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.