DIY scintillation probe: the easy way

This guide is under continuative update. 
updated 9 June 2019
updated 12 June 2019


Many backers of my USB PMT adapter project on Kickstarter asked me to write a simple guide that explains how to make a scintillation probe to be used with the USB PMT adapter for gamma spectroscopy. This guide is focused on the “dirty cheap” way to construct one at home without special tools.

I know two ways of making a probe:

  • Alluminium turned enclosure with internal teflon spacers. Internal mu-metal shielding and internal voltage divider.
  • Black tape and scissors “dirty cheap” probe.

The first type of probe is rugged, reliable and hard to make at home if you don’t have access to a lathe, a mill and have the skills needed to precise machining alluminium and plastics. This type of construction will be discussed in a new guide that I’ll write in few weeks. The second type that is treated here is easy, quick to make. It could be easily improved adding a proper aluminium enclosure.

How is made a scintillation probe and what I need to make one?

Example drawing of a scintillation probe


A typical scintillation probe consists of a light tight enclosure with, inside of it, a photomultiplier tube (PMT) that is coupled to a scintillation crystal. The voltage divider needed to feed the tube and extract pulses usually is placed on the back of the PMT tube. The tube could have wires soldered directly to the voltage divider PCB or some kind of socket. In this case the divider is connected to the socket that is connected to the PMT.

To make one you need:

On the right I’ve added a direct link to my favourite suppliers.

  • A scintillation crystal
  • Some silicon optical coupling grease The Rad Lab
  • A photomultiplier tube The Rad Lab
  • A voltage divider for the PMT and it’s socket (if it needs one) The Rad Lab
  • A light tight enclosure that puts all pieces together. In this case… just some well tight black electrical tape.

PMT tubes example

The following pictures shows a typical photomultiplier tube usable for gamma-ray spectroscopy. I strongly recomand you you to buy tubes from a reliable source. My favourite tubes are from Hamamatsu and specifically the R6095

R6095 PMT from iRad – The rad lab
R6095 with it’s socket
Front optical window

Scintillation crystals

There are many type of scintillation crystals avaiable. The most common are:

  • LYSO    usable but have internal Lu176 gamma emission
  • NaI(Ti)    this is the type of crystal to buy
  • CsI / CsI(Ti) / CsI(Na)    low FWHM compared to NaI(Ti) hard to find
  • BGO    low FWHM compared to NaI(Ti) harder to find

For gamma-ray spectroscopy the preferable type is NaI(Ti). They can be found on eBay from cheap from ex CCCP countries.

The quality of your spectrometer expressed by the FWHM capacity of the system crystal-PMT-amplifier-ADC must be in the range of 8% or below. Small value of FWHM indicates better resolution capability of the spectrometer. Usually surplus soviet crystals have a variable FWHM because of they age and storing condition in the range 8-12%. A good crystal must be colorless with ultra -brite white internal walls. If not maybe they could be yellowish or stained and this means that they are trash.

Russians NaI(Ti) crystals. Usualy they are enclosed in a round machined steel enclosure.
On the left a CsI undoped crystal. On the right four LYSO crystals. They could be used but I prefer to use the LYSO as radioactive calibration source because it’s internal Lu176 gamma emission or for display in my collection :3
Bicron plastic scintillators. Because they low Z they are useless for gamma spectroscopy

Copling between the crystal and the PMT tube

The optical face of the PMT tube is where the scintillation crystal will be coupled. For this parpuse use some specific silicon optical coupling grease. It is very cheap and you’ll need a very modest quantity.

An important thing to note is the size of crystal and PMT. You must use a PMT with an optical windows bigger than the crystal. If you cannot you’ll lose some resolution. For example if the crystal is 43mm diameter and PMT optical window is 38mm you can couple they and expect some little resolution loss. If you try to couple to a 24mm PMT this coupling doesn’t work well and must be avoided.

A perfect coupling. PMT is 30mm diameter and crystal is 28mm internal diameter. It will works just great!
Another good coupling. A 10mm crystal on the optical window of a 30mm PMT
This coupling is BAD! A 30mm PMT on a 43mm crystal
An acceptable coupling. A 38mm PMT on a 43mm crystal. Is not a good coupling,I expect some 1% FWHM loss but it will work

Assemply the probe

So, you’ve find all the needed parts to assemble a probe? Good! Lets build it.

First, clean your desk, clean your PMT’s optical window and your crystal’s optical window too. You must remove all kind of grease from it like fingerprints with some alcohol. Put the crystal face up and spread over it’s optical end some silicon optical grease. Put the PMT otical window in contact to the crystal and spread the grease with little circular movements, this will remove bubbles of air. Don’t apply any pressure. The grease will make an optical coupling film between the crystal and the PMT. Now with some piece of black tape fix the PMT in position over the crystal.

Now countinue to wrap the black tape around the tube. No light must enter. Take care to tight the tape and to tight the coupling between the PMT and the crystal. I’ve closed the end of my tube with some black silicon glue. If your tube needs a socket you’ll need to enclose the complete assembly into a light tight enclosure. This because you cannot put black silicon glue under the socket.

Now, if you have some adesive mu-metal shielding sheet, you can cut it and wrap aroun your probe to add magnetic shielding to the tube. This is a good thing to do but, the tube will work also without it so, don’t worry about it at the moment, you can always add it! Connect your voltage divider circuit and your coax cable. Your probe is ready to be tested. My USB PMT adapter supply positive voltage from the central point of a BNC connector. I suggest to use a short (<=1mt) RG59 coaxial cable to connect the probe with the board. One end of the cable must have a BNC male connector, the other end, screen shield of the cable goest to the voltage divider – “minus” ground/cathode, the inner insulated conductor of the coaxial cable, feeds the voltage divider with high voltage. It goes to the + “plus” anode. If you use a mu-metal shield, connect it to – “minus”.

Notes about voltage divider

Photomultiplier dynodes need specific voltages and voltage ratio between each other. The voltage divider is specific for each photomultiplier. This guide explains how voltage divider is calculated. The Rad Lab sells PMT’s with voltage divider kit and instructions to build it. I strongly suggest to buy it from him and follow the assembly instruction. Here you can see the instructions provided for the R6095 and R9420


Author: l.i.papadopol