SNM-11/Ludlum 2000 Characterization and Troubleshooting

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
Nick Babusis
Posts: 20
Joined: Thu Dec 08, 2022 12:11 pm
Real name: Nick Babusis

SNM-11/Ludlum 2000 Characterization and Troubleshooting

Post by Nick Babusis »

I have acquired a Ludlum model 2000 scaler and a SNM-11 boron-lined neutron tube for my neutron detector. The SNM-11 is housed inside a ø5” HDPE cylinder and is connected to the model 2000 via a series bias resistor and a BNC cable. The BNC cable and moderator were removed for testing. Important specs are listed below:

SNM-11 specs:
Ignites at 700V, working voltage of 1500-3000V (recommended: 1600V)
15% thermal neutron efficiency
0.8 “relative neutron sensitivity”
1.02 CPM background
27.5uA maximum background corona current
ø18.5 x 336mm

Ludlum 2000 specs:
Supports 60M ohm scintillation loads up to 1500V
Maximum applied voltage: 2500V
Proportional tube recommended threshold: 2mV
GM tube recommended threshold: 35mV


The bias resistor and tube connections:
IMG-6027.jpg

Tube connected directly to the scaler:
IMG-6029.jpg

Testing
I expected to be able to detect gamma rays from my test sources (18000 CPM Co-60 and 24000 CPM unknown material, likely I-129), but could not get any reading above background at any voltage or discrimination setting. I need to determine if the tube needs to be replaced or if something else is wrong in my setup. I tried varying the bias resistance and measuring the current at the maximum attainable applied voltage and found that the current is above the maximum rated background current:
Bias resistance -> max applied voltage, current
0 ohm -> 820V, 73uA
10M ohm -> 1600V, 54uA
100M ohm -> 2400V, 18uA

I also found that varying the applied voltage could not raise the voltage across the tube, which plateaued around the ignition voltage of 700V (and increased slightly as bias resistance decreased).

I tried measuring the background count rate while varying the applied voltage or the discriminator setting:
Applied V, background CPM: (minimum discriminator, 100M ohm bias)
1600V, 61
1500V, 46
1400V, 32
1300V, 11
1200V, 8
1100V, 0.5

Discriminator potentiometer setting, background CPM: (1600V applied, 100M ohm bias)
0 turn, 61
1/8 turn, 62
2/8 turn, 59
3/8 turn, 22
3.5/8 turn, 6
3.75/8 turn, 1
4/8 turn, 1
5/8 turn, 0
6/8 turn, 0
7/8 turn, 0

I am using an older model which does not have a numbered discriminator dial, so I am assuming the 0-7/8 turn range corresponds linearly to the 0-100mV threshold range listed in the manual.

Finally, I tried sweeping the applied voltage with the discriminator set to 4/8 turn (assumed 57mV pulse height discrimination), since this setting gave me comparable values to the expected background count rate:

Applied V, background CPM: (57mV discriminator, 100M ohm bias)
2000V, 16
1900V, 10
1800V, 0
1700V, 0
1600V, 0

These readings were highly variable, and could be changed significantly just by cycling power to the detector.

What to do now?
Does anyone have any insight as to what is causing the voltage across the tube to plateau at the ignition voltage and why this tube has no sensitivity to gamma rays? Could the tube be damaged? I doubt the model 2000 is damaged since it accurately detects gamma rays and rejects alpha particles using a Ludlum model 44-3 scintillation probe.

Also, am I correct using a 100M ohm bias resistor for this tube? I do not entirely understand the counter schematics below, since it shows the op-amp connected *after* a 500M resistor, and then sending the signal through the discriminator potentiometer and into the comparator. My understanding was that the bias resistor was needed to allow the charge pulse from the tube to temporarily create a voltage potential across it which the op-amp could detect, as shown in the image from Bob Higgins’ paper below.

Model 2000 main board:
Model_2000_main_board.png

Model 2000 comparator circuit:
model_2000_comparator.png

Bob Higgins' schematic:
Higgins_schematic.png

If I cannot fix this problem and confidently detect neutrons (and nothing else), I am considering buying a much better characterized 3He tube for $350 (ouch!).
William Turner
Posts: 31
Joined: Sun Nov 27, 2022 10:09 am
Real name: William Turner

Re: SNM-11/Ludlum 2000 Characterization and Troubleshooting

Post by William Turner »

There are several things to unpack here...

Your implementation of the bias resistor is problematic (see 3rd paragraph below). R4 (100k) is at the output of the second stage of the RC filter and R6 (1Meg) is for control loop stability (C11 is probably ceramic or film and its low ESR would cause a considerable phase shift in the feedback loop) and both assist with current limiting. R2 (500M) is for voltage feedback of the output for the high-voltage power supply control pin (U10, pin8). R1 (500M) is also part of a voltage divider that is a reference level for the comparator U3A. The NET "SIGNAL" is what is fed to the amplifier. The amplifier is on sheet 5 of 5 and gains the AC coupled (C27) portion of the signal and is also fed to comparator U3A.

I think Bob Higgins' paper is a good starting point but he failed to heed the advice in the actual Russian datasheet for the corona tubes. (Explained here: https://gigabecquerel.wordpress.com/201 ... snm-tubes/). They were meant to be used with a high pass filter (100 pF / 16k ~ 100kHz) to remove the corona noise. The corona noise also disappears in the "proportional" region but gain is lost and gamma discrimination is challenging.

A gas detector should be configured in an "T" arrangement. The detector needs to be DC-biased through a high-value resistor, but the detector also needs to be AC-coupled. In order to accomplish this, R6 (1M) could be replaced with the high-value resistor (probably should be 47Meg). Additionally, the high pass filter element should be considered. I think getting an oscilloscope involved is a must. The output of C27 is biased at 1.25V by R39/R8 through 100k (R38) which would complicate the filter addition. But no doubt the placement you have for the bias resistor is a problem.

How are you measuring the voltage across the tube? What is the input impedance of the voltmeter? Depending on the measurement technique the reading could be different than when the voltmeter isn't connected.
JoeBallantyne
Posts: 327
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: SNM-11/Ludlum 2000 Characterization and Troubleshooting

Post by JoeBallantyne »

Per this post: viewtopic.php?p=82392#p82392 1600V is the maximum do not exceed voltage.

The thread that is part of, indicates at the start (Joe Gayo post) that the tube works better at ~700V for him, than at 1600.

This is in agreement with the posts about using the CHM/SNM 11 with a Ludlum 3 at 750V bias, from Mark Rowley.

It looks like you have large amounts of bare copper wire wound around various parts of the tube. Why?

Are the wires actually soldered to the tube?

If I were you, I would use insulated wires, and strip off just enough so that you can solder the wire to the tube directly. (Without damaging the tube in the process of course.) Furthermore for the HV wire, you should use wire with insulation that exceeds the max voltage you are going to supply to the tube.

A soldered connection is going to be far superior to a wrapped wire. Plus those wrapped wires can short to something else.

Are you using an electrostatic voltmeter to measure your voltages? Most multimeters CANNOT be used to measure bias voltages because they load the circuit too much.

You keep trying to test your neutron tube with gamma rays.

Get a neutron source, and test it with that.

Joe.
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: SNM-11/Ludlum 2000 Characterization and Troubleshooting

Post by Dennis P Brown »

As someone ignorant on these subjects, I see you are using a 100 Meg resister with the Ludlum 2000. Richard never indicated such was required. So, am I missing something? The Ludlum circuit does show an existing 500 Meg resister already.

Certainly I used a 100 Meg on my simple detector system with its BF3 tube (the pre-amp/amp unit was not designed for neutron detectors) but no idea if that is an issue for the Ludlum 2000.

I do agree that without a neutron source, you are operating blind with these types of detectors. One can (as I recall, you did) use background but that is so low a rate one can't be certain the settings are valid. As I posted, my Ludlum & BF3 detector was following an expected BG signal but until I run my fusor, I really have no idea if that is real or not. (A project further delayed.)

I do understand only too well that this type of work can be really frustrating! People like William and Joe are such a great help and resource.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: SNM-11/Ludlum 2000 Characterization and Troubleshooting

Post by Richard Hull »

Indeed, with the 2200 Ludlum and apparently with the 2000, the circuit already has a current limiting protective HV output resistor internally. All tubes of the normal type can be tested and set up without any resistor in series with the tube. As I am unfamiliar with the Russian corona tubes, listen to the people who have actually got them biased and working properly.

As noted by Dennis in an earlier post, the 2000 has a "secret" internal properly dialed internal discriminator pot that controls the "level set span" for the front panel pot. Thus, the user (prior owner) may have set this internal pot for perhaps GM tube only levels or PMT only levels for use with the front panel pot. This is especially true if the 2000 was part of a maintenance facility covering only GM tube testing where limits were established for the front panel pot in a go, no-go range of acceptable GM tube spans. This was why Dennis moved this pot to the front panel.

I assume the voltage measured and reported by Rich were taken off the 2000's panel meter. These are properly set up, obviously, for their low current bias supply voltages. You can trust the Ludlum's front panel analog meter voltages to +/- 50 volts, I would think. (tricky for boron lined tubes, I would think)

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: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: SNM-11/Ludlum 2000 Characterization and Troubleshooting

Post by Dennis P Brown »

I assume you have read my previous post in this forum (viewtopic.php?t=14874).

Since you have a gamma source, use a Geiger detector tube (null out the DISC) and set 400 volts (I did this for my as a first test) and see what occurs. I used a simple Uranium source and it matched by normal Geiger counter device fairly accurately. In this manner I checked the Ludlum 2000 to be certain it wasn't inoperative or had other issues.

Then, all I did was place the Ludlum 2000 discriminator to mid level, ran my BF3 tube up to 1600 volts (slowly and with long delays at voltage steps staring at 700 v) and waited for a count to occur. Once a count showed, I used longer wait times (10 min.) as I added 100 v increments (I think my first count occurred at 1200 v). Once I reach 1600 I got an expected count rate (1.7 counts/min.) So I stopped.

This result was that at this DISC setting and voltage enabled the unit to achieve the expected count rate for cosmic rays back ground (BG) over a 10 minute time period. This did not prove very much except that IF the tube & system were likely set correctly, and appeared to work. From what I understand, this is a first test performed for neutron detection.

If that count rate is not occurring for you with BG, something is likely amiss. I'd be careful with voltage on the Boron tube and I'd not bother with a gamma ray source for settings other than for a Geiger tube. Rather the std BG cosmic ray secondary's method is a reliable first test (through not real proof) and will provide a 1 to 2 count per minute rate, on average and valid your settings on the unit.

Of course, once I have a working fusor again, I'll report back on the results of my Ludlum 2000 unit and BF3 tube (hopefully, positive results.)

Aside: my BF3 tube is in a copper tube with metal end caps. I, like you, then use a BNC. I soldered the signal wire to the end lug and the ground wire to the outer metal body. It wasn't easy but can be done. I did not use a 100 Meg-ohm resister, though I can't see any harm in that at all.

No idea if my soldering and metal case enhances or even has any benefit at all but just the way I did it. I was worried about outside electrical noise - likely a needless exercise in complexity on my part.
Post Reply

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