Scintillix gamma spectrometer

[Milos]

Hello,

Let me introduce my darling - Scintillix gamma spectrometer. As I mentioned in the "Introduction" section of this forum, it took me rather longer time to develop it (you know ... job, family, other hobbies ...). The first thoughts about gamma spectrometry came on my mind after Fukushima NPP accident and the final working prototype registered very first gamma quanta during late summer of 2016.







Spectrometer is based on a 9-stage high PHR 3" photomultiplier tube operated up to 1 kV and optically coupled to a 2.5" NaI(Tl) crystal. Pulses coming from the PMT are amplified and shaped directly in the probe housing, thus sound card unit is fed with amplified pulses, which are less prone to EMI. Only drawback of this solution is, that two cables are needed to interconnect a driver and the probe. In this setup, one cable is for HV and second one for signal and low power supply voltage needed for amplifier and shaper. Signal is processed with a sound card incorporated into the driver unit. With the acquisition card, there was little problem, I spent lot of time to find suitable one, which will not deteriorate resolution. Despite the fact, that at first glance the sound cards looked exactly the same (blue USB stick), external circuitry was fairly different and some of them was really sloppy. Difference between the best and the worst was up to 3% FWHM @662 keV of Cs-137. There was piece to piece variability as well.

When I checked my spectrometer in a laboratory with properly chosen and tweaked sound card, I obtained fairly good value of resolution 6.9% FWHM @662 keV of Cs-137, which is not far from physical limit of the used system.


Stability of the system is very good as well, if the spectrometer is left to warm up few minutes. In a following image is the result of a stability test done in course of 8 hours with uranium bearing mineral as a sample. Spectra were collected for 10 minutes every two hours. Upper spectrum is average of bottom four spectra.


Now I use my spectrometer for different kinds of measurements, covering mainly low level activities of naturally occurring radioactive materials (NORM) in the environment, everyday items, some special chemicals and in food.

Milos

[Conor Whyte]

Curious about the Scintillix. What is being used as an ADC in the device? What is the max sampling rate?

[Svilen]

A good looking system that also gives very good results - well done! Maybe you can write a bit about the shielding too? It looks interesting on the picture.

[Steven Sesselmann]

Milos,

Thanks for sharing. Nice professional looking unit and a good looking spectrum.

Well done..

Steven

[Milos]

Hi folks,

Regarding your questions:

1) As an ADC I implemented USB sound card which is hidden inside of metal box of a driver unit. This brand was used: https://www.konigelectronic.com/compute ... eo-1088473. Since, there is lot of Conig-like sound cards (fakes), it is quite hard to find one whith good parameters. There is also "analog output", so the pulses from the CSA-Shaper-amplifier block can be led to an alternative processing unit (or better soundcard). Sampling rate I usualy use is 192 kHz. One friend of mine promissed me to develop some kind of digital sampling unit with true sampling rate in order of MHz, but as all things it is matter of (free) time as well, so nobody knows when it will be or whether it ever will be.

2) Shielding is kind of "hybrid". Lower part originates from a professional shielding for gamma spectrometry, which was produced in our country in eighties of 20th century. Despite the fact that is quite old fashioned, it is very well designed. Among others it is equipped with the doors for sample placement so you can replace samples without need of probe removing. Interior offers several shelves with different distances from the probe for sample placement, so it simplifies situation when you need to maintain the same measurement geometry. Upper part of the shielding (the green one) is newly made, since original part was intended only for 2" probes. I needed one, with bigger hole for 3" probe. This part is composed from an iron container, which was filled with a molten lead. Lets say that my background without shielding is somewhere around 250 CPS and with shielding around 24 CPS, so decrease in background pulse count is approximately 10 times.

I have been overloaded with my job (profession) last couple of months, however I hope that I will be able to prepare and send some of my recent experiments (e.g. quantification of K-40, some new measurements of Cs-137 in a soil and some decay experiments with Rn daughters adsorbed on active charcoal.

[Conor Whyte]

Hi folks,

Regarding your questions:

1) As an ADC I implemented USB sound card which is hidden inside of metal box of a driver unit. This brand was used: https://www.konigelectronic.com/compute ... eo-1088473. Since, there is lot of Conig-like sound cards (fakes), it is quite hard to find one whith good parameters. There is also "analog output", so the pulses from the CSA-Shaper-amplifier block can be led to an alternative processing unit (or better soundcard). Sampling rate I usualy use is 192 kHz. One friend of mine promissed me to develop some kind of digital sampling unit with true sampling rate in order of MHz, but as all things it is matter of (free) time as well, so nobody knows when it will be or whether it ever will be.

2) Shielding is kind of "hybrid". Lower part originates from a professional shielding for gamma spectrometry, which was produced in our country in eighties of 20th century. Despite the fact that is quite old fashioned, it is very well designed. Among others it is equipped with the doors for sample placement so you can replace samples without need of probe removing. Interior offers several shelves with different distances from the probe for sample placement, so it simplifies situation when you need to maintain the same measurement geometry. Upper part of the shielding (the green one) is newly made, since original part was intended only for 2" probes. I needed one, with bigger hole for 3" probe. This part is composed from an iron container, which was filled with a molten lead. Lets say that my background without shielding is somewhere around 250 CPS and with shielding around 24 CPS, so decrease in background pulse count is approximately 10 times.

I have been overloaded with my job (profession) last couple of months, however I hope that I will be able to prepare and send some of my recent experiments (e.g. quantification of K-40, some new measurements of Cs-137 in a soil and some decay experiments with Rn daughters adsorbed on active charcoal.

For that high sample rate ADC you might want to have a look at the the newer RedPitaya boards. There are several projects utilizing them as Spectrum analyzers and gamma, beta and alpha spectrometers or possibly - the better option might be to use a Digilent Analog Discovery board which does utilize real time USB output. The Red Pitaya boards are not "real time" devices due to their streaming data over a network.