My spectrometry system

Scintillation crystals, PMTs, voltage dividers etc...
Post Reply
valentin
Posts: 22
Joined: 08 May 2015, 23:59
Location: Germany
Contact:

My spectrometry system

Post by valentin » 02 Aug 2015, 22:55

Let me share a few details about the spectrometry system I have at home.

The setup consists of a self-made probe, powered by a GS-1102-Pro, connected to an UR 22 external sound card (196 kHz, 24 bit) connected to my mac book pro. My first spectra where taken with PRA running in crossover, but since then I started writing my own software for the fun of it (in case you are a mac user, I will be happy to share this software after a few more improvements).

My probe is built from a XP1010 10 stage PMT attached to a 30mm x 70mm NaI(Tl) crystal. The voltage divider has a total resistance of 12MOhm, with 2MOhm between the photocathode and the first dynode, and 1MOhm each for the remaining stages to the cathode. I also added HV caps between GND and the last two dynodes for stabilisation. At 1250V, this gives a quite linear response (see below).

The first prototype was held together by duct tape, tin foil and cardboard, but otherwise it worked quite well. Over time, the improvised connection between crystal and PMT had loosened a bit, so I recently rebuild it using a 3D-printed crystal mount glued to the PMT for more stable connection. (I do not want to glue the crystal directly to the assembly, so it can be removed in case I want to replace the PMT one day).

This is what the interior looks like without housing (sorry for the blurry picture):
tube.JPG
tube.JPG (102.82 KiB) Viewed 4862 times
There are some "nubs" on the inside of the crystal mount, so the effective inner diameter is about 0.5mm smaller than the crystal diameter. The slits on the mount allow the plastic to bend a bit outside when the crystal is inserted, creating the pressure that keeps the crystal in place. Between PMT and crystal is a thin layer of silicone oil for adhesion and optical coupling.

The screenshot below shows the output signal, with 50x zoom on the y-axis (signal level). There is only little noise around the baseline, but heavy pulses cause some kind of rebound on the baseline. Not sure if this can be avoided by adding more capacity. For now, I just discard parts of the signal with too much slope on the baseline, which gives me less variation in the peaks of the recorded spectra.
Screen Shot 2015-08-01 at 4.42.21 PM.png
Screen Shot 2015-08-01 at 4.42.21 PM.png (144.82 KiB) Viewed 4862 times
Below is a recorded spectrum of thorium 232. The small utility window on the right side shows the relation between the recorded pulse heights and photopeak energy values cited by literature, as well as a regression line. It appears the response of the tube is quite linear below 1MeV, but higher energy pulses are slightly compressed (i.e., pulse amplitude is a bit lower than it should be).
Screen Shot 2015-08-01 at 4.52.12 PM.png
Screen Shot 2015-08-01 at 4.52.12 PM.png (133.07 KiB) Viewed 4862 times
Valentin

User avatar
Steven Sesselmann
Posts: 756
Joined: 27 Apr 2015, 11:40
Location: Sydney
Contact:

Re: My spectrometry system

Post by Steven Sesselmann » 03 Aug 2015, 08:20

valentin wrote:The setup consists of a self-made probe, powered by a GS-1102-Pro, connected to an UR 22 external sound card (196 kHz, 24 bit) connected to my mac book pro. My first spectra where taken with PRA running in crossover, but since then I started writing my own software for the fun of it (in case you are a mac user, I will be happy to share this software after a few more improvements).
I am a mac user and would love to evaluate your software.
valentin wrote:My probe is built from a XP1010 10 stage PMT attached to a 30mm x 70mm NaI(Tl) crystal. The voltage divider has a total resistance of 12MOhm, with 2MOhm between the photocathode and the first dynode, and 1MOhm each for the remaining stages to the cathode. I also added HV caps between GND and the last two dynodes for stabilisation. At 1250V, this gives a quite linear response (see below).
If it is one of those Russian 30x70 crystals, they were quite good, I bought quite a few of those in the past. 12M Ohm divider is a good choice for the GS-1100-PRO, just remember that the 1M load resistor is technically part of the divider, so the voltage on the dynodes will be lower by 1/13
valentin wrote:The first prototype was held together by duct tape, tin foil and cardboard, but otherwise it worked quite well. Over time, the improvised connection between crystal and PMT had loosened a bit, so I recently rebuild it using a 3D-printed crystal mount glued to the PMT for more stable connection. (I do not want to glue the crystal directly to the assembly, so it can be removed in case I want to replace the PMT one day).
In case you are interested, I sell blank aluminium housings that will fit your detector.

http://www.gammaspectacular.com/gamma-s ... part45x195

valentin
Posts: 22
Joined: 08 May 2015, 23:59
Location: Germany
Contact:

Re: My spectrometry system

Post by valentin » 04 Aug 2015, 05:44

sesselmann wrote: If it is one of those Russian 30x70 crystals, they were quite good, I bought quite a few of those in the past. 12M Ohm divider is a good choice for the GS-1100-PRO, just remember that the 1M load resistor is technically part of the divider, so the voltage on the dynodes will be lower by 1/13
Yes, it is one of those crystals from a well known russian seller. I had some fun at the german customs office when the official needed a "product category" for calculating the duty. So I tried to explain the purpose of this item while avoiding words like "radiation" etc… As a side note, I had a similar experience when your GS-PRO arrived. After a short lecture on the subject, the officer's face brightened up and he said "Ah, it is an external sound card" :-)
sesselmann wrote: In case you are interested, I sell blank aluminium housings that will fit your detector.
The spun metal looks pretty, in particular the end cap. However, with socket attached, my XP1010 is a bit too long. Even if I resoldered the capacitors, I would still need about 22-23 cm inside length. There are shorter PMTs on the market, maybe for a future update of my detector…

Valentin

Neil.
Posts: 9
Joined: 28 Apr 2015, 12:47
Contact:

Re: My spectrometry system

Post by Neil. » 07 Aug 2015, 04:53

Valentin,

Very nice work and thank you for sharing. I also use a home built system including an A/D converter rather than a sound card and have noticed the slight compression as you mentioned at higher energy levels. Perhaps this is due to baseline shift after a strong pulse? It is not a problem since I wrote the analysis software and use point to point linear interpolation to derived the energy histogram. Th232 makes a good calibration source due to the many known peaks for this element.

I like your crystal mounting method but there is one thing which you used which may have caused me trouble. I started doing spectroscopy when I obtained a small NaI crystal from a surplus seller. The crystal was old (circa 1950s) and was slightly yellow but it was clear otherwise. I coupled it using Dow 550 silicone oil and after a year or two I noticed that the crystal began to become opaque around the outside edge at the glass to aluminum seal. This was uniform so it didn't result from a chip or a crack. Over time this opaque area grew to where it now almost completely covers the crystal. The crystal is still usable but it looks bad.

I don't know for sure if the silicone oil leaked into the crystal but it is hard to imagine that after surviving for more than 60 years with grease coupling in the surplus unit the damage began when I used the oil. For my other probes I now use Dow Compound 4 grease and have seen no problems. Another advantage to the grease is it is not as messy as the oil.

Neil
Neil Edwards
Dublin, Ohio, USA

valentin
Posts: 22
Joined: 08 May 2015, 23:59
Location: Germany
Contact:

Re: My spectrometry system

Post by valentin » 08 Aug 2015, 06:48

Neil, thanks for pointing out the potential problems with silicone oil. When I disassembled the previous prototype (after about 10 month), the crystal still looked OK, possibly due to the fact that I used a special oil with a very high viscosity. If you turn the vial upside down, it takes about a minute until a first drop forms. Still, it behaves like a fluid and creeps slowly into cavities. Hopefully it does not leak into the crystal encasing…

ManuelDiaz
Posts: 24
Joined: 03 Sep 2016, 18:11
Location: Germany
Contact:

Re: My spectrometry system

Post by ManuelDiaz » 20 Oct 2016, 07:55

valentin wrote:
sesselmann wrote: If it is one of those Russian 30x70 crystals, they were quite good, I bought quite a few of those in the past. 12M Ohm divider is a good choice for the GS-1100-PRO, just remember that the 1M load resistor is technically part of the divider, so the voltage on the dynodes will be lower by 1/13
Yes, it is one of those crystals from a well known russian seller. I had some fun at the german customs office when the official needed a "product category" for calculating the duty. So I tried to explain the purpose of this item while avoiding words like "radiation" etc… As a side note, I had a similar experience when your GS-PRO arrived. After a short lecture on the subject, the officer's face brightened up and he said "Ah, it is an external sound card" :-)
sesselmann wrote: In case you are interested, I sell blank aluminium housings that will fit your detector.
The spun metal looks pretty, in particular the end cap. However, with socket attached, my XP1010 is a bit too long. Even if I resoldered the capacitors, I would still need about 22-23 cm inside length. There are shorter PMTs on the market, maybe for a future update of my detector…

Valentin

I was just going through the forum and saw your post valentin, I had the same experience at the customs office, and he ended up using computer software as the customs category hahaha

trigatronix
Posts: 4
Joined: 07 Jan 2017, 09:33
Contact:

Re: My spectrometry system

Post by trigatronix » 07 Jan 2017, 10:01

Hello Valentin,
I see this ElectronVolt software for the first time.
Where can i get it?
Greetings!

Alexander

valentin
Posts: 22
Joined: 08 May 2015, 23:59
Location: Germany
Contact:

Re: My spectrometry system

Post by valentin » 07 Jan 2017, 21:29

Hi Alexander,

it's some software I wrote on my own. It is just a prototype with most pulse detection parameters hardcoded to match my particular detector setup. Also, much functionality is only available through a command line interface.

Last year's new year's resolution was to evolve this to a more general, feature complete software to share with the community, but eventually I got stuck with other projects :/ Maybe in 2017?

Are you a mac user, too?

Valentin

Post Reply

Who is online

Users browsing this forum: No registered users and 3 guests