Page 1 of 2

Collection of information about russian SNM tubes

Posted: Tue Oct 16, 2018 8:32 am
by Lukas Springer
Some information was brought together in a trading post, this is that information condensed down before the thread in trading gets deleted.

As many of you are aware some russian SNM tubes are floating around and there seems to be much confusion on how to use them,
This problem is not made easyer by the fact that there are many different types of detectors under the same name.

SNM tubes can be:
-Boron 10 as cathode coating in corona tubes (SNM9-14)
-10BF3 as fill gas in proportional tubes (SNM3, 5, 8 & 20)
-Helium 3 corona (see link below)
-Helium 3 proportional (see link below)

It's hard to get datasheets for most of them, but a russian manufacturer is still supplying some of the He3 tubes, with datasheets: ... n-counters

I have only worked with my SNM-18-1 so far, which is a 4.6 bar 3He corona tube, if you have a different detector please post your results here :)

We all know the paper by Bob Higgins, which I would like to add to.

This is Higgins' circuit for driving 3He corona tubes:
Which gives me the following trace for a neutron event:
DS1Z_QuickPrint5.png (36.29 KiB) Viewed 4442 times
This is the circuit described by Consensus-group in their SNM-18-1 Datasheet:
corona tube.PNG
Which gives me the following trace for a neutron event:
DS1Z_QuickPrint4.png (36.54 KiB) Viewed 4442 times
As you can see, using 16k to "terminate" the signal all of the high impedance corona noise is removed and all you get is a clean pulse, which can easily be amplified, needing almost no noise rejection.

The current through my tube is 8 µA with 100 MOhm and 14 µA with 47 MOhm (both at 2400V), allthough the output signal dosen't seem to change in amplitude with different currents, which might be explained by the very constant voltage across the tube.
I'd assume the same circuit works for 10B corona tubes aswell.

I can't say anything about my tubes sensitivity, but 4.6 Bar 3He are 4.6 Bar 3He.

If you have any further information about any of the SNM tubes feel free to add it here!


Re: Collection of information about russian SNM tubes

Posted: Mon Feb 18, 2019 9:41 am
by Lukas Springer
Next thursday I'm getting access to a decent neutron source, to test the detectors properly.
Right now I have 1 SNM-18-1 He3 and 2 SNM-11 B10 tubes.

I will try to measure their efficiency, gamma rejection, if they can be used in the proportional region, etc.

Is there anything you want to be tested?
Anything specific you want me to focus on?

Re: Collection of information about russian SNM tubes

Posted: Wed Feb 27, 2019 11:44 am
by Lukas Springer
I finally got around collecting the data, here are my results:

My setup was:
Canberra 3106D HVPS
HP 5554A charge sensitive preamp (47M Bias Resistor)
Canberra 2031 SCA
NM 172 Ratemeter
Rigol DS1054z Oscilloscope

Amersham Buchler 3.7 GBq AmBe source, ~2.2E5 n/s

One SNM-18-1 4.6 Bar He3 Corona tube
Two SNM-11 10Boron coated Corona tubes
A cylindrical paraffin moderator, 6 cm of paraffin around the detector tube

Settings of the preamp were 300 mV/pC * 4 for the He3 tube and 30 mV/pC * 8 for the B10 tubes.
The SCA was set to a 1 to 4V Window.
Since I have two SNM-11 and they behaved somewhat differently I'll be talking about SNM-11(1) and SNM-11(2) from now on.
All tests on both SNM-11 were done at 1600V, the SNM-18-1 was tested at 2500V.

Currents were:
SNM-11(1): 18 µA
SNM-11(2): 19µA
SNM-18-1: 22µA

Neutron events were one to four volts out of the SNM-18-1 and one to two volts for both SNM-11, giving 0.8 to 3.3 pC in He3 and 4.2 to 8.3 pC in B10.
Corona noise was 300 mVpp in the SNM-18-1 and 900 mVpp in both SNM-11, making noise rejection somewhat harder in the B10 tubes.

With these values and SCA settings gamma rejection was very strong, as you can see in the videos below: [SNM-18-1 without moderator] [SNM-18-1 with moderator] [Gamma only dose rate of the neutron source]

I measured the absolute efficiency of my detectors by measuring count rate with the source 1 meter away from the detector, resulting in:
SNM-18-1: 60 cps
SNM-11(1): 6 cps
SNM-11(2): 4 cps

Giving me an absolute efficiency of 6.5%, 0.65% and 0.43%; as you can see the He3 detector is way more sensitive than both B10, which can be attributed to a higher active volume.

I also played around with the SNM-11 tubes connected to my ludlum model 3, and it is possible to use them at 1500V, if you put an external 100 Mohm with 100 pF in parallel in series to the tube. [Ludlum model 3 with SNM-11] [Ludlum model 3 with SNM-11 beta rejection, ~2 MBq Sr90] [Ludlum model 3 with SNM-11 gamma rejection, ~50 µSv/h pitchblend]

At ~2/3 of the corona ignition voltage all three tubes have a "proportional" region, but gamma rejection is extremly hard, so I did not persue this any further.

The geiger counter in those videos (Graetz X50zs) is energy compensated and goes from "ticking" to a solid beep above 25 µSv/h.

As you can see those russian corona tubes are very good neutron detectors and can definitely be used to detect fusion.
I hope they'll get more love and appreciation in the future!

If you have any further questions I'd be happy to help!


Re: Collection of information about russian SNM tubes

Posted: Wed Feb 27, 2019 4:43 pm
by Joe Gayo
Can you try running the CHM-11 just below the corona starting voltage (ie 670V)?

Re: Collection of information about russian SNM tubes

Posted: Wed Feb 27, 2019 4:49 pm
by Lukas Springer
From ~2/3 of the corona ignition voltage to the ignition voltage it is in that "prop mode" region, being very responsive to gamma.
What do you want to achive with running it below its nominal voltage?
Maybe I can test for that.

Re: Collection of information about russian SNM tubes

Posted: Wed Feb 27, 2019 6:29 pm
by Joe Gayo
The SNR improved dramatically, so gammas can be discriminated

Re: Collection of information about russian SNM tubes

Posted: Wed Feb 27, 2019 7:01 pm
by Lukas Springer
There is no corona noise, that is true, but gamma peaks get huge.
With some corona current gamma rejection becomes really easy, since all gamma events are drowned out in the noise and only neutron events get above it. Setting your disc to ~1.5 times the voltage of your noise pretty much garantees you to discriminate against all non-neutron events.

Re: Collection of information about russian SNM tubes

Posted: Wed Feb 27, 2019 8:30 pm
by Richard Hull
The take-away, from my view point, over all these years of hoopla, is that the Russian neutron tubes are tricky to use and adjust if you do not have co-existing a decent neutron source to compare against a decent Gamma source. Neutron sources are very tough to come by, while decent gamma sources very easy to obtain. Radium is easy to get in old war bird instrumentation found at hamfests. Many fleamarkets have old large wesclox "Big Ben" alarm clocks. 10 uCi Cs137 sources in the U.S. can be ordered through the mail. Such items blast out rather significant amounts of easily detectable gammas up to over 1mev with most in the 200-600kev range. For really hot gammas a Thorium chemical compound is what you want as thorium spits out a copious quantity of gamma over 2 mev.

To really tune and see what is counted and what is truly rejected, a decent video needs to be made with a nice setup just like Lucas and others here have at their disposal. A really hot gamma source needs to be placed next to the tube in question, (no moderator). Using an oscilloscope, the bias needs to be adjusted to show that as the bias is raised the gammas at various levels of increasing bias start to, effectively, disappear. At this point in the effort, insert the tube into a moderator and place the neutron source next to the gamma source, still showing the oscilloscope. Ideally, the neutron source should be suddenly obvious, maybe at a low level, but plainly there. If not, you haven't shown anything related to neutron counting or gamma rejection.

If it turns out there is a real hyper-narrow bias range where gamma rejection is acceptable and neutron counting will rise out of the noise using windowing techniques, then you have something that might be worth using. While not a great system due to the criticality of adjustment, you are assured that as long as all things remain equal, the tube is good and counting neutrons and neutrons only.

Naturally, there is a local and cosmic background, but that is normal. A good system should not see more than 8 to 15cpm of real background.

I am sorry I do not have a Russian neutron tube to monkey with.

Richard Hull

Re: Collection of information about russian SNM tubes

Posted: Wed Feb 27, 2019 10:21 pm
by Lukas Springer
I'm sorry, maybe I'm bad at explaining, but I wanted to make the point that people don't have to be "scared" by those russian detectors, but that they're instead pretty easy to use once one got the hang of it. They reject the 59 keV of Am-241, tested with a source giving off > 3mSv/h at point blank, and the U-238 / Ra226 spectrum, tested with a piece of pitchblend giving off >50 µSv/h, perfectly well.

Instead of tuning an operating point they're quite comfortable at their rated voltage with a 47 Mohm bias resistor, gamma rejection is done by rejecting everything below ~1.5x the corona noise peak to peak voltage.

I have compared the counting rates with and without neutrons / gamma / moderator, the setup described above is the result of that

My He3 tube gives ~12 cpm background, the SNM-11 <5 cpm.


Re: Collection of information about russian SNM tubes

Posted: Thu Feb 28, 2019 12:09 am
by Steven Sesselmann
Hi Guys,

Just a reminder that David Housley and myself developed this little board for the Russian Corona counters, specifically the SNM-12 (CHM-12) and similar tubes. It's a schmidt trigger circuit and turns the noisy signal into a nice flat baseline with strong pulses.

It can operate on a 9V battery for several days, but requires an external HV source.

GS-PREAMP ... eutron-pcb