BetterGeiger S-1 radiation detector

[RobertD]

Santa got me a BetterGeiger S-1 radiation detector, which is a rather inexpensive compact and rugged scintillator based instrument. I do not have any details about the detector like crystal size and type, though.
I could only resist for a few days before taking the unit apart and analyzing the raw signals on the circuit board.
Turns out that the analog pulse signal on the board can be used for some basic gamma spectrometry, despite the low pulse rate (around 4 cps measuring Lu2O3, as shown below).
Here is a 30 minute spectrum of Lu-176, taken with the RedPitaya and imported into NuclearPhoenix's web app. Longer measurements would surely make it look better, but the two characteristic peaks at 202 and 307 keV are recognizable.

[NuclearPhoenix]

Robert,

interesting findings. I wondered what would keep someone from modifying this device and try things out. Turns out that having a RedPitaya at hand can be quite useful :)

The signal doesn't look too pretty but you can definitely see the peaks, nice. How large was your Lu2O3 sample? 4 cps is really low, that definitely explains the bad SNR.

Btw, you can also directly export your plot by clicking the camera icon, if you didn't know that. See the screenshot below.

Matthias

[RobertD]

Hi Matthias,
A 5 gram glass container of Lu2O3 powder was set on top of the BetterGeiger, which gives about 1 cps with background only. We have to keep in mind that the BetterGeiger is not intended for gamma spectrometry. It was just a quick test out of my curiosity and there is lots of room for improvement in this setup, for example I did not match the signal amplitude to the input range of the RedPitaya and I did not play around with different sampling rates either.
Thanks for the tip regarding the camera function, will use it next time.
Btw, my RedPitayas are normally being used for ham radio stuff (WSPR receiver).

Robert

[NuclearPhoenix]

I see. Well, thanks for trying it out. Love to see "hacks" like this!

It would be interesting if you try to fine tune your settings more and record for a day or two and see what you get then.

[RobertD]

Will do and post the results in a couple of days :-)

[RobertD]

Ok, recording for 18 hours did in fact reduce the noise considerably. The peaks got a lot wider though, so I assume there is some thermal drift involved. But let's not forget that the instrument was never designed for spectrometry so being able to clearly identify the Lu-176 peaks was impressive for me.

[RobertD]

On a side note, there are two more raw signals accessible on the pcb of the BetterGeiger:
- a digital 5V pulse indicating a count
- a analog signal 0-5V depending on dose rate. I have not yet looked into the scaling of this signal.

[kotarak]

I dont have one but I do suspect that the detector is some sort of a single semiconductor component (PIN diode?) - would be interesting if you can confirm this? I dont think it is a "traditional" scintillator with scintillating crystal coupled to optical detector - could be wrong.
Somebody reported that the reported dose is energy compensated - I wonder if the MCU used has a built-in ADC or there is an ADC chip on board? It must be doing something more than just counting.

[RobertD]

Sorry, I can't confirm the technology of the detector without risking to damage my Christmas gift. The part of the pcb where the detector (and a number of other components) is located is conformally coated.

I dont think it is a "traditional" scintillator with scintillating crystal coupled to optical detector - could be wrong.

Would it be possible to get an energy spectrum like the above without using a scintillator? I honestly don't know, I'm still struggling with the basics of gamma spectrometry.

The energy correction feature is mentioned on their website and they are using an ATmega 328 as MCU.

[Madmax]

It's not a big secret, crystal size seems to be 4*4*8 mm...