2
$\begingroup$

I have a Photonis XP5301 PMT to use for fast neutron detection (detector optically glued to the PMT window). I can't quite figure out why its not producing a spectrum. First of all, I have multiple different values of operating voltage that I found on the internet, but I think it's either -800 V or - 1000V. I used the same setup with a NaITl scintillation detector + attached PMT, and it gave me proper results, but not mine. What could I do to locate the issue? The Voltage divider I'm using is a Photonis VD202K/40. ALso, while the NaITl detector's PMT has a power cable that connects to the amplifier, my PMT doesnt have a power cable. What could be the issue here? What am I not addressing? Thank you!

$\endgroup$
  • 2
    $\begingroup$ Can you add a block diagram showing the hardware involved in your working and non-working setups? If you're using an multichannel analyzer or something to make a spectrum with your sodium scintillator and that's not magically producing a spectrum, the first step is to look at the PMT output/analyzer input with an oscilloscope to see how your two PMT signals are different. $\endgroup$ – rob May 19 '17 at 20:59
  • 1
    $\begingroup$ I'm with @rob on this. Step one is to learn how far I to the chain (if at all) you do have signal. For one thing it is possible to kill PMTs several ways. $\endgroup$ – dmckee --- ex-moderator kitten May 19 '17 at 21:14
  • 3
    $\begingroup$ "My PMT doesn't have a power cable" is definitely suspicious. You need to understand the signal path. The PMT is current output until it goes through an amplifier; in some cases just a load resistor is sufficient to see a signal. But you can't just "plug it in". $\endgroup$ – Floris May 19 '17 at 21:14
3
$\begingroup$

You say you are using the setup for a neutron detector.

What is the material you are using? What is the expected light output (in photons / event) and how fast would they appear? When you use the NaI based detector, what is your source? What is the activity?

Here are the troubleshooting steps I would take:

  1. To make sure the PMT is working, I would hook the output (straight from the PMT anode) up to a scope, into a low-ish impedance (1 k to ground). Turn the gain of the scope to maximum (or 10 mV / cm). Slowly increase the operating voltage (while PMT is in darkness); eventually, you should see "grass" on the baseline of the scope: this is the signal from individual (thermally induced) photoelectrons.
  2. Turn the voltage down about 100 V from the point where noise started appearing, and introduce a very weak flashing LED into the enclosure of the PMT (if you can see the flash with the naked eye it's too bright for the PMT). You should now see a massive spike of signal for every flash of the LED. Turn the HV off and demonstrate to yourself that the signal is gone (so there is no cross talk)
  3. Now connect the PMT output to a shaping amplifier (it's not clear whether you have one - if you don't, then that might be your problem). The shaping amplifier should give a "slow" output at every pulse coming in - this is usually needed as the input to a spectrum analyzer. Check the output of the amplifier on your scope (you will need a much less sensitive setting). Again, using the LED could be very helpful here. If you don't get a signal from the amplifier, check its power etc.
  4. Now that you have a PMT with roughly appropriate voltage and a working amplifier, connect to the spectrum analyzer. Introduce a robust source (so you get good signal) and slowly sweep the HV in the range you found above, until you see triggers on the spectrum

When you have done all the above, and you found a place where things don't work as expected, you've likely narrowed your problem significantly. If you still have trouble, leave a comment.

$\endgroup$
  • $\begingroup$ Thanks for all the help, everyone. I figured out the issue. It's a very silly one. After banging my head at it for hours, I realized that my PMT does not have an in-built preamp, so I had to get one (PMTs I usually use all had in-built preamps). Thanks again everyone, I appreciate the support and insight. You troubleshooters are the best! $\endgroup$ – Betsy May 22 '17 at 20:22
  • 1
    $\begingroup$ Glad you figured it out. In the future, if you add a block diagram of your setup we will be able to give more pointed assistance... with the limited information given the advice was necessarily broad (but maybe also more useful for future visitors who have a "slightly different" setup). $\endgroup$ – Floris May 22 '17 at 20:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.