Photo-multiplier/scintillation Detector

This area is for discussions involving any fusion related radiation metrology issues. Neutrons are the key signature of fusion, but other radiations are of interest to the amateur fusioneer as well.
Post Reply
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Photo-multiplier/scintillation Detector

Post by Dennis P Brown »

Well, after a number of confusing false starts, I finally got the scintillation detector that I assembled to work detecting fast neutrons from my fusor.

The key to getting the bugs worked out was that I bought a very nice Eberline bench unit (an old Model RM-21 but a steal at the price on ebay; this unit has adjustable voltage and user select Photo-multiplier (PM) or G.M. internally) and by using a properly made HV cable connection system (also, the detector is heavily shielded from x-rays/gamma rays*.) These improvements have enabled me to better experiment & determine the best operating voltage (650 VDC) for the PM while simultaneously better suppressing fusor noise - the PM/scint unit is fully enclosed in a metal shielded box w/coax connector (the later issue of noise I have now gotten to near zero.)

Currently, the fusor chamber is ultra clean and operates with only minor glitches and no other issues (I have left it under vacuum for a week to keep it clean.) Once I strike a plasma the unit runs extremely smoothly from near zero current to 40 ma and I can track it up/down in current by gas inlet flow control very precisely. Restarting the plasma is now as smooth** as turning on a diode light.

Background counts (@650 VDC) for the powered PM system were under 10/min (no fusor.) With the fusor running and under 10 ma at 28 kV (plasma glow), the PM read about 10 counts/min. As I raised the current up to 40 ma (holding 28 kV; around 17 - 20 microns D2) the counts climbed to 35 to 40 counts/min. The unit's neutron signal tracked the power supply current exactly over numerous runs from 0 to 40 ma (for fixed 28 kV.) The count level remained constant at the selected current level.

Apparently, the response to fast neutrons for my scintillation collector size isn't exactly overwhelming (plastic volume is about 14 ci). But certainly enough to use it to measure the gross response as well as track total flux as a function of current and/or voltage. Though it would be not very useful to bother to calibrate counts at this level of response.

I am extremely pleased with the unit I assembled. I simply could not have done this successfully without the extensive help here! Between diagrams for the PM tube (absolutely essential and thanks for people here confirming this) and advice on resisters & added caps***, as well as noise issues I would never have been able to make a successful PM unit, much less a fast neutron scintillation detector.

One of my recent goals with restarting my fusor efforts was to build a fast neutron PM/scintillation counter and I have finally managed to complete that task. Yes, this doesn't put someone in the higher levels (i.e. elite status) but designing, building and debugging a fast neutron detector is a skill I very much wanted to achieve (and ended up upgrading my fusor which was far more involved then I ever imagined.) Specifically getting a PM tube to work at all, much creating one for fast neutrons.

Setting goals for oneself is also an important component of progressing in this area of fusion research here at the forum - this enables one to learn the in's and out's of various detector systems rather then just buying off the self 'black boxes'. This - ironically - touches upon the recent update to the thread on "Financial Entropy". Building one's own support devices does add a depth of knowledge to doing fusor work that is the very heart of this forum and relevant to the the forum's primary purpose.

My next focus will be to hook this unit up to a computer via an interface system I've assembled in order to record counts vs. time for a more precise measurement system rather then reading the analog scale & listening to the audio of the Eberline.

Following this, I intend to get both my old BF3 tube working again (true to form, connection issues at the high voltage lead to noise issues) and my new 3He tube system. I was having signal noise issues with the BF3 due to interfacing the unit to a proper HV source - the Eberline should solve that issue with proper connectors & shielding. Then I still need to deal with the computer software issues for the 3He unit. These problems still need to be corrected but having a working fast neutron detector makes these tasks much easier simply by having a known counting system that is operational.

* I was running a GM detector system simultaneously to detect gamma/x-ray with no added shield system and the counts from the GM were rather low (also under 10 counts/min.) The counts did not change with fusor voltage indicating, I think, that my fusor shell is doing its job defeating the x-ray threat. The ceramic shield should suffice for any gamma (relative to the dynodes - the scintillation plastic doesn't respond to gamma or x-rays.)

** While I am not thrilled by this addition to my power supply, I decided to install a HV capacitor (0.01 uF) to help smooth the output to the fusor; the cap was supper cheap so I couldn't pass on it (and rather large in size ... and happens to be rated to 30 kV.) I installed a 1 Giga-ohm bleed resister to ground on the hot side and, of course, I have a grounding rod that I short the system/cap before touching any HV parts. I do feel the cap has improved the start up of the plasma and improved the operation but now I am restricted to operating under 29 kV due to the cap limit of 30 kV.

*** I use a set of 1.2 Megaohm resisters between dynodes and have 2 kV caps on the first three, and last dynodes on the PM tube. If interested in the circuit, see my earlier post on this very subject - link below. Note dynode pins 4, 12 & 13 are not used. The PM has a clear face that is 3 inch dia. as is the fast neutron sensitive plastic crystal - the 'crystal' is about 2 inches thick; these units are sealed together by special clear PM tube adhesive.

For details on the circuit see:
viewtopic.php?t=14239
Attachments
Open end of the 'Black Box'. Dynode connection end of PMT with resisters and caps
Open end of the 'Black Box'. Dynode connection end of PMT with resisters and caps
Enclosed PM with its scintillation Plastic detector
Enclosed PM with its scintillation Plastic detector
Eberline, fusor & PM/scintillation Detector; white block is a extra thick ceramic shield
Eberline, fusor & PM/scintillation Detector; white block is a extra thick ceramic shield
Last edited by Dennis P Brown on Sat Mar 12, 2022 12:53 pm, edited 5 times in total.
User avatar
Jim Kovalchick
Posts: 717
Joined: Wed Apr 13, 2011 8:00 pm
Real name:

Re: Photo-multiplier/scintillation Detector

Post by Jim Kovalchick »

Dennis,
Nice work on the fusor and the detector.

If you want to do a moderator check you may want to try slowing some of the fast neutrons down. Assuming your plastic scintillator isnt very good at detecting thermal neutrons, you could place slab of HDPE in between and expect your counts to go down.

Another way to make sure you aren't detecting xrays with your setup is to use another gas besides D, and get to the same voltage and current you know produces neutrons with D and verify no or low counts on your detector.

Jim K
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Photo-multiplier/scintillation Detector

Post by Dennis P Brown »

All good ideas. Thanks!
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Photo-multiplier/scintillation Detector

Post by Richard Hull »

You might have only x-rays and an astonishing background to worry about in your detection scheme with the PMT.

Some notes:

1. Record counts only head-on face of crystal to you fusor. (smallest cross section facing fusor)
2. Slab siding the PMT tube to the fusor, (as shown), might open you up to Compton and photo electrons from x-rays from a running fusor within the PMT dynodes as counts.
3. Your first duty is to get a stock background count. Take 3 or 4 ten minute counts due to NORM and cosmics. Derive by simple statistics an expected CPM rate for background.
4. Run the fusor with only air running at fusor vacuum pressures and currents to fusion levels with fusion voltages and again run a number of one minute counts to see if any x-radiation is to be added above the background. This is very important with a "so called" fast neutron detection crystal. If you get counts, place a 1/8-inch thick lead plate in front of the face of the crystal. Count again and add lead if needed until you reach background. fast neutrons will whistle right through the lead if any is needed. The moment you cross the 30kv threshold you will need to do all of the steps 1-4 over again as the fusor is then totally transparent to the high energy x-rays. This is mostly not the case at 20-25kv applied.
5. Only now will you be rather assured any count above background with deuterium in you fusor are indeed detected fast neutrons.

PMT neutron detection is at best a close tolerance battle against x-radiation and variable photonic NORM around high voltage vacuum systems. The best money is always spent on BF3 and 3He detection. BF3 is absolutely gamma/x-ray resistant to lethal levels of those pesky electromagnetic waves. The PMT is a gamma/x-ray detection system marvel with nothing better in the electronics kit.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Photo-multiplier/scintillation Detector

Post by Dennis P Brown »

With no running fusor the PM gave the typical BG counts of under 10/min. I also checked the PM system against the fusor running air (Plasma: 30 - 40 ma, 28 kV; 15 min. and 'head on' for the detector tube) and the counts were about 6/min. About where I expected and similar to BG. Since I was getting around 40 counts/min with deuterium and it tracked current well (up and down), I am fairly certain the PM/Scint. was working for fast neutrons. The PM system was heavily shielded (large area and thick -75 mm ceramic; besides the steel case of the fusor and 3.5 mm Al PM case.)

In any event, I started work on my old BF3 detector tube and discovered a short within the high voltage wiring buried where I didn't expect to go again; so I had to dig that out and redo the HV connection system (a bad HV cable that was rated for 25 kV!) I rebuilt a better connection system (open now) and the unit was calm (no clicks indicating a short) with the Eberline RM-21 set to 1850 volts (DC.) Did a test to 2000 DCV and still ok.

The fusor developed a nasty leak and I am now trying to figure that problem out; in a retest to high vac and seal off, the system only showed a 1 micron/min. rise for the entire assembly. Previously, it was showing a 20 micron per second leak rate with the turbo running! Lost on why the leak suddenly disappeared - that is not good. I did way over heat the cathode - accidently ran the fusor from 50 to 100 ma for a few minutes (ballast resisters heated way up, too.) So I am out-of-business until I resolve the leak issue - likely the cathode system; I have dissembled that and will do some checks on it.

Since the HV cathode is now out, I will see if I can improve the mounting system as well for future higher voltage.
Post Reply

Return to “Neutrons, Radiation, and Detection (& FAQs)”