Help to Troubleshoot a Canberra GR1218 (HPGE)

[Magnetite]

I am troubleshooting a Canberra GR1218.

After working perfectly for years, the FWHM suddenly rose, there was tailing on the leading edge of peaks, and the count rate also dropped.

After thermally cycling the detector everything appeared normal, however, within a few days the problem returned. The detector was thermally cycled again, which, only temporarily fixed the problem for a few hours. Each time, the detector was allowed to warm for ~24 hours, followed by a ~18 to 24 hour cooldown period, before applying high voltage.

To keep the detector going, I lowered the voltage (from -3500 to -2000). At the lower voltage, the FWHM, peak shape, and count rate are fine.

I'm looking for any insight and/or clues to determine what the issue is. Could there be an accumulation of gases in the cryostat (that have saturated the mol sieve)? Or maybe it's a problem with the FET?

I think I heard a clicking noise coming from the detector when I applied high voltage, so perhaps there's a short circuit? Maybe, there are multiple problems?

Any ideas/guesses/knowledge would be helpful.

Thank you :)

[Sesselmann]

Ruwan,

I am not familiar with the Canberra GR1218, but I do have experience with clicking noises from HV supplies. What you describe sounds like a short across the board or across a capacitor, humid weather can make it worse and if it is allowed to short for a while, the track can become conductive, which will make it worse. Sometimes cleaning the HV board with some isopropyl alcohol and drying it thoroughly could help (if this is the problem).

Steven

[GigaBecquerel]

Hello Ruwan,

By what you're describing it sounds a lot like you need to pump your cryostat.
Trailing and bad rez is caused by leakage currents, from ice that has built up, the clicking might even be HV sparks, do this any longer and your detector WILL die!
Let me guess, when the detector is thermal cycling, and perhaps already when it's "steady state cold" it's "sweating" with a lot of condensation?
That's your cryostat no longer having enough vacuum to properly isolate the LN2.
Pump and bake your detector and it will work just fine again.

Lukas

[Magnetite]

Thanks for the advice :)

Pictures below. I haven't pulled vacuum/heated a detector before, but understand the concept. I'm not sure what adapter I'd need to the vacuum pump line. Is the port on the detector a vacuum line or is it a pressure relief valve?

What temperature do I heat to and what vacuum pressure?

If it's vacuum and heating with a heating tape, I think I can accomplish this.

The click when first applying HV didn't happen again in a test today.

[GigaBecquerel]

It's not just sweating, it's ICED UP!
It definitely needs to get pumped right now, and stop applying high voltage!

In the second pic is your vacuum port, you need to seal around that and then pull the "cork" out under vacuum, pump, bake, close, then release the vacuum around it.
In terms of pressure, the lower the better but a good rotary vane definitely does the job, going below E-3 mBar would definitely help of course, but after all it does a lot of cryopumping when cold.
You can heat the stalk with heater tape to ~80 °C, going above that can damage stuff and needs some more preparations. Pumping and baking should be done for at least two to three days after having that much gas in it.

[Magnetite]

Thanks for the advice/details. I found a great tutorial here: https://gigabecquerel.wordpress.com/202 ... -part-1-n/

I think the hardest part will be the vacuum adapter, so I may buy one from Mirion.

The iced up picture was taken when the detector was cooling down in LN2. It isn't like that once cooled . but, I've taken your advice and will perform the baked out/pump down before adding HV again.