Modern neutron detection - Ludlum's new scintillator

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.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Modern neutron detection - Ludlum's new scintillator

Post by Richard Hull »

I have always felt that software, in the many variants, revisions, updates, and failed corporate structures that no longer support it coupled with platform specific issues and hardware issues coupled to software controlled systems would, at some point, be a causative agent in a doomsday event. We already see hacking creating major disruptions in modified software updates from outside. we are too connected in critical infrastructures that run 100% on software internally. While we here are just annoyed by the above and would never create the doomsday event, we will be forced to barely survive of die due to its result once it is set in motion. Too much easily transportable malware on a thumb drive brought in from outside. Think about it. Thus far, in this posting, most seem to have software that works and Ludlum is still very much alive and well, (One of the few old line radiation measuring firms that were not gobbled up by a rather uncaring Thermo.)

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
Bruce Meagher
Posts: 148
Joined: Thu May 12, 2011 11:25 pm
Real name: Bruce Meagher
Location: San Diego

Re: Modern neutron detection - Ludlum's new scintillator

Post by Bruce Meagher »

Finn,

Would it not be a trivial exercise to take the ascii log file produced by the meter and process it with any modern language you're comfortable using (Matlab, Python, C, Java, Excel, etc.)? I'm not sure why Matlab would be more ideal except Andrew provided the solution for you.

Bruce
Andrew Seltzman
Posts: 815
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Andrew Seltzman »

I compiled a updated standalone program that will run without Matlab, I believe you will have to install the matlab runtime.
Attachments
Ludlum2363_Neutron_background.zip
(1.5 MiB) Downloaded 263 times
Andrew Seltzman
www.rtftechnologies.org
User avatar
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Finn Hammer »

First of all, a warm and heartfelt thank you to you, Andrew, for the compiled special, which enabled even me to report fast and efficient, in a timely manner.
You are a true friend, Andrew.
Hope is such a great thing, when reports of a less than accurate item description in a group buy are starting to emerge, and ironically enough, near the end hope was all there was to cling to.
But then, I was the fortunate one, and received a near mint item, with only minor scratches from 8 years of shelf life, the original readout protective sticker in place, the cable laced to the handle in a fasion I imagine only a Ludlum employee would device, and hardly any dust at all.

Near mint Ludlum 2362
Near mint Ludlum 2362
It powered up and with the dips set as advised, very soon the numbers started to roll in.
Around 3UTC, 29/12/2021 in Skoerping, Denmark, 80 meters above sea level, it recorded:
Average CPM= 7.0869 over 29.35 min

1st smoothed.JPG
1st smoothed.JPG
1-st smoothed.JPG




A stable gamma source is not at hand at the moment, so this will have to wait till I get one.

Eager to see the fusor raise the numbers above background, the probe was put next to the fusor with 10mm of lead in between, and let it rip.

20211229_163906.jpg


The following graph shows the raw numbers out of the ascii file, and is more for entertainment than anything else, as I can't say how they translate to neutron output, but I hope this thread will evolve into presenting a recipy on how to calibrate this counter.

FirstRawNumbers.JPG


Cheers, Finn Hammer
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Modern neutron detection - Ludlum's new scintillator

Post by Richard Hull »

Nice report Finn! You background is exactly what mine hovers around using the big 3He 4ATM detector. You got a good one for sure. Thanks for the high quality report, as usual.

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
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Finn Hammer »

This is a report of the run this morning, Skoerping, Denmark, 11:00 - 12:00 UTC
I placed the probe so that it is 80mm from the center of the fusor, see attached photo, and ignore the parallax, I did not want to back off and use a tele.

Prescila probe jammed right close up to the fusor.
Prescila probe jammed right close up to the fusor.


Publishing the number from center of fusor, to face of the Prescila probe, may already be controversial, since there doesn't seem to be an agreement about exactly where the fusion takes place in a cube fusor, with a tube kathode. Is it in the center, perhaps mainly on the internal surface of the endplugs, or is it just randomly distributed allover the goddamn place?
Well, hopefully, this is something we are about to find out, by strategically placing the probe, extracting the numbers and subjecting these numbers to math.
I ran the fusor for 10 minutes at 43kV, 30mA, 32Microns and activated silver to 20kCPM.

Remote during the run. Notice that the 901P which delivers data to the oled readout, is 6 microns off the reading from the calibrated baratron.
Remote during the run. Notice that the 901P which delivers data to the oled readout, is 6 microns off the reading from the calibrated baratron.


After this run, the 2363 Interface crashed, so I did not get the numbers from that run.
Instead I ramped the fusor up to the same level again, over a minute, to grab the numbers from the probe.


Still a bit of water in the chamber, there is a leak that needs to be fixed.
Still a bit of water in the chamber, there is a leak that needs to be fixed.


This time I got the numbers, allright, and here they are. The output in mrem/hr are translated to cpm via the LANL-endorsed 350 conversion factor


rampup.JPG

The counts top out at 49000cpm at time 51.5 secs.
In a previous post, Andrew writes:

-----------------------------------------------------------------------------(block quote substitute, hehe)

A fusor producing 1e5n/s will generate 0.27mrem/min at 8cm from the center (equivalent to a BTI bubble detector placed on the surface of a 6" diameter sphere), or ~16.2mrem/h. At this dose rate the PRESCILA probe would be indicating 5670cpm.

------------------------------------------------------------------------------(EBQ)

Since my numbers are close to an order of magnitude higher, this should work out to a neutron output around 8.6e5n/s, in this run.


But something is bothering me, and that is the decay factor of 0.6 which creates a tail in the time domain, and what I presume is similar to the pile up seen in spectography from the tail from the shaping amplifiers, try to look at the numbers from one of the background runs:

hits and mis.JPG

The red dots indicate a hit by a neutron, creating an increase in reading by 0.08mrem/hr.
As an isolated event, this 0.08mrem/hr should decay to 0 in 4 seconds. But here during a fusion run, we are hammering the probe continously with hits and this must leave an elevated level of reading which should be compensated for.

The question is how?

I am probably messing things up here, from lack of understanding of the math involved. It sure would be easy if it is possible to take the reading at face value and conclude the 8.6e5n/s .


Edit:
Having an electronically recorded neutron rate available makes it much more efficient to create runs for higher numbers. This is the best up to now, with a calculated 1.54e6n/s. Frankly, I thought I was doing better than that and I guess a set of bubble detectors has to be brought in to deliver the last word.

Pouring 2kW+ in to it,  for hardly over the megamark makes this fusor the most inefficient model in history.
Pouring 2kW+ in to it, for hardly over the megamark makes this fusor the most inefficient model in history.
Second edit:

I do have some heavy stuff, and pulled out the 3phase variac, to get a higher voltage out, in this case, I got to 75kV.
The fusor gets rather temperamental at this point, and can shoot up above 50mA in a glimps of the eye, and you will see this behaviour reflected in the graph, where I have to turn the voltage down to abruptly, after the psu has tripped the over current protection. There is still plenty of headroom in the PSU, but for now, I think it is better to tend to the rather bad leak I have, which limits the bottom pressure to 4e-4torr
75kV run.JPG
I don't quite understand these numbers, I think others have reached 2ish meganeuts/s at much lower currents,

Cheers, Finn Hammer
Last edited by Finn Hammer on Thu Dec 30, 2021 10:02 am, edited 1 time in total.
Andrew Seltzman
Posts: 815
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Andrew Seltzman »

Hi Finn,

The tail in time domain is a function of the time constant filtering in the meter and is not a problem; this is exactly how it is designed to work and does not need to be compensated for. The filter takes the discrete pulses and filters it to an average value displayed on the meter face and in the output removing the "noise" of the individual pulses. When the count rate is sufficiently large, the filtered value appears as a smooth constant dose rate. At very low dose rates you see a "blip" in the dose rate after each individual pulse followed by an exponential decay with the filtering time constant. For count rates high enough that the effects of individual pulses are not visible in the output, the calibration should be taken at the displayed value in mRem/h, at very low (background) dose rates the dose due to an "individual" neutron count does not have much meaning. Unfortunately this counter does not have a scalar function as it is designed to only be used as a rate meter / integrated dose meter making background counting over long intervals difficult.

The code I wrote takes advantage to the blips in dose rate after discrete neutron counts to provide a scalar function for background dose rate. It first takes the derivative of the dose rate with respect to time and divides by the peak increase in rate per pulse (0.08). This provides a spike of ~1 at the leading edge of each pulse. Then it thresholds any values <0 due to the decay tail and sets them back to 0. Then it rounds counts to the nearest integer value, and adds the tally of counts over the logging period and divides by time.

Question on your 901p: is the calibration factor set to hydrogen?
Andrew Seltzman
www.rtftechnologies.org
JoeBallantyne
Posts: 327
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: Modern neutron detection - Ludlum's new scintillator

Post by JoeBallantyne »

Finn, have you checked to see if the orientation of the probe makes any difference in your numbers.

In the only picture you have that shows the probe orientation, you have it with the end opposite the handle facing the fusor.

I'm not sure if there is a Hornyak button on that face of the probe.

If I were you, I would try putting one of the other 4 useable faces of the probe (each of which has a button in it for sure) up against the fusor. (ie: rotate your probe 90 degrees in the horizontal plane, or sit it on its end so the handle is vertical, and then jam one side up against the fusor)

See if your numbers change - and if so how.

Joe.
User avatar
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Finn Hammer »

Thanks, Andrew,

Answer is: no, nitrogen. Setting it to hydrogen returned the 901P to track the baratron with remarkable accuracy.

20211230_161833.jpg


Edit:

Joe, sounds like a must do, although the manual shows this graph to verify the dependence on rotation. "They" have taken every measure to make this probe omnidirectionally sensitive.

orientation.JPG
Cheers, Finn Hammer
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: Modern neutron detection - Ludlum's new scintillator

Post by Liam David »

Not to interrupt Finn's work, I just finished a 3-day background count with the Ludlum. Total counts were 39736, yielding a mean of 9.198 CPM.

Matlab code used:

s = serial('COM3','BaudRate',38400,'Terminator','CR/LF','Timeout',1);
fopen(s);

neutronData = zeros(86400*3*2,1);

for i=1:86400*3*2
t0 = tic;
fprintf(s,'RR');

try
data = fscanf(s);
nums = regexp(data,'[0-9.]+','match');
neutronRate = str2double(nums{1})/100;
neutronData(i) = neutronRate;
catch
disp('Connection Error');
fclose(s);
fopen(s);
end

dt = toc(t0);
pause(max(0,0.5-dt));
end

mRem_per_hour_count = 0.08;
counts = [0;diff(neutronData)]./mRem_per_hour_count;
counts(counts<0) = 0;
counts = round(counts);
counts(counts>10) = 0;
Andrew Seltzman
Posts: 815
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Andrew Seltzman »

Do you have a plot of counts vs time? It would be interesting to see if there is a day/night diffrence
Andrew Seltzman
www.rtftechnologies.org
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Modern neutron detection - Ludlum's new scintillator

Post by Richard Hull »

I have recorded minor diurnal differences of what I feel are real to me, but are down in the scud of what I term statistical indifference for my money. I found a slight increase by one or two counts per minute. However, the solar activity will always drive such things. If I count a significant difference over my usual 6-8 cpm background, I race for the "space weather" spot on the internet to see what old Sol is up to. Due to the Earth's magnetic field, day and night values tend to merge.

Finn I assure you you are doing 2e6 n/s at the voltages and currents you record. You are driving the hell out of your system!!

Of course I have a 6" spherical fusor and yesterday drove it to 6e5 n/sec at 43kv, 11 ma, 9microns. It gets better each day I run it due to wall loading of the huge inner surface area for targeted implantation.
The very small cube fusors seem to rely on direct forceful focused drive on the target ends.

My calibration, ( 3He count multiplication factor) helped by calibrated Rem ball readings and reduction of related mr/hr rates as well as my recent volumetric close agreement gives me confidence in my numbers.


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
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: Modern neutron detection - Ludlum's new scintillator

Post by Liam David »

Looking for day/night differences was one reason for doing this. The main reason is I noticed the count rate trending upwards over the course of a day from ~8.5 to ~9.5 CPM.

Here is a 1-hour moving-mean plot of all the data. There are no discernable day/night effects although the count rate does decrease toward the middle, coincident with less rain.


background.png
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Modern neutron detection - Ludlum's new scintillator

Post by Matt_Gibson »

I think I have neutrons!

Before adding deuterium, my background count was around 5.5cpm over a 10 minute run. After adding deuterium via my MKS MFC, I have a 46cpm result over 10 minute run. 37kV, 7.8mA, 15.4microns, probe placed about 4in away from chamber.

If this sounds good, I’ll document a little better and then submit for the neutron club?
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Modern neutron detection - Ludlum's new scintillator

Post by Dennis P Brown »

While getting background is important, that is due - for a running fusor - to possible electrical noise. As such, adding deuterium isn't the criteria to determine neutron counting but rather, readings with the moderator and then without (the later which accounts for BG counts) for say a minute each run (ten minutes is fine, too.) If the difference is statistically significant then that is a valid indication of neutron detection.

Aside: maybe this is a good time to start a new thread on your experiment rather then this thread.
Andrew Seltzman
Posts: 815
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Andrew Seltzman »

The moderator trick will not work with this type of detector (proton recoil) though since it's energy response is almost flat from thermal to 100MeV. Unlike a He3 of BF3 detector which can be removed from it's moderator, in the PRESCILA probe, the detecting element and moderator is the same element.
Andrew Seltzman
www.rtftechnologies.org
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Modern neutron detection - Ludlum's new scintillator

Post by Matt_Gibson »

I’ll update my thread for my next run :-)

I just wanted to add some raw/crude results to this one since this detector is “new” and I have little idea how to interpret the data other than seeing a large jump in background count.

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

Re: Modern neutron detection - Ludlum's new scintillator

Post by Dennis P Brown »

Couldn't one still could add an additional moderator and see the change in counts? That would be rather straight forward method to confirm a real neutron signal.
Andrew Seltzman
Posts: 815
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Andrew Seltzman »

An additional moderator shouldn't make much difference on a PRESCILA probe since it is sensitive to both fast and thermal neutrons. A moderator with a neutron absorber should make a difference though by reducing neutron rate, however it would physically be a huge moderator compared to that used in a HE3 or BF3 tube since the detector head is much larger in diameter. Running with deuterium and hydrogen separately should produce similar electrical noise output and is likely a simpler test.
Andrew Seltzman
www.rtftechnologies.org
User avatar
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Modern neutron detection - Ludlum's new scintillator

Post by Finn Hammer »

All,

Bruce was right, after all (never doubted that, of course, just too befuddled to respond to it at the time :-) ) there are other ways to communicate with the 2363, than via mathcad.
Of course, getting the first numbers across for initial assessment of the functionality of the items, the matcad code was essential, and this rush was just another aspect of the Ludlum 2363 frenzy, which some of us experienced first hand.

But for the average Joe, like myself, to whom mathcad language looks unreasonably gibberishly, communicating with something like C+ would be preferable.
I have just tested the RS232 interface with a terminal, and it works well, (of course it does, it is a Ludlum!) this means that it will indeed be trivial to build a small microprocessor based querying adapter, with a readout displaying whatever parameter is desired, probably to present neutron rate and averaged neutron rate, even save these data to a SDcard or something.
Here is a little snip of the first attempts, soon crowned by luck:

The learning curve was not that steep this time. RTFM. Case sensitive for the most.
The learning curve was not that steep this time. RTFM. Case sensitive for the most.

To arrive at an accumulated, or averaged count number can also be aquired by the Ludlum interface. It has, apart from the 1/2 secont dump facility, also the option to set an arbitrary count interval, with averaging over the interval, as seen here, where I set the interval to 10sec. But it could be 600Sec. It even sports a countdown to the next count!, right handy in the case of long count intervals.

Ludlum interface showing the averaged arbitrary count interval.
Ludlum interface showing the averaged arbitrary count interval.

My fondness of this instrument is increasing by the day, it's versatility is steadily increasing!

Cheers, Finn hammer
JoeBallantyne
Posts: 327
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: Modern neutron detection - Ludlum's new scintillator

Post by JoeBallantyne »

I bought this instrument as well, and the shipment finally arrived this week after getting stuck in the bowels of FedEx for a couple of weeks.

I don't yet have a neutron producing fusor to test it with, but I did a background run, and most of the time it spits out about 6 counts per minute. Sometimes up to 8, sometimes as low as 3 or 4, but typically 6.

I didn't yet try running Andrew's script, or dumping out anything over the RS232 port, simply used a timer on my phone to countdown a minute and manually counted the higher toned neutron output clicks from the speaker. Did that about 10 times in a row, and the most frequent result was 6 cpm.

Joe.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Modern neutron detection - Ludlum's new scintillator

Post by Matt_Gibson »

Joe,

How’s the condition of the Ludlum? Finn and myself got ours in good/great shape while others got some in rough shape.

6cpm is around what I get.

-Matt
JoeBallantyne
Posts: 327
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: Modern neutron detection - Ludlum's new scintillator

Post by JoeBallantyne »

I would say the original condition when shipped was decent, but certainly not new. There was significant damage incurred during shipment due mostly to failure of the seller to appropriately package and mark the shipment. Needless to say, I was not happy when I saw the box, or when I opened it up. I am of course following up with the seller.

Joe.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Modern neutron detection - Ludlum's new scintillator

Post by Matt_Gibson »

Any new progress with this? I’m curious as to how these are comparing to known detectors.

I had a confusing experience with mine:

As I was pushing my fusor voltage up close to 60kV (pulls around 7-8mA at 13 microns or so) I had my readings skyrocket to over 1600mR/hr. I was able to repeat this over and over again. Once I realized that X rays are now a problem again (I was previously limited to 40kV) I stopped and added lead sheeting. After adding the lead, I was never able to get anywhere near the results.

To date, my best results after adding the lead, is 60mR/hr. This was done at 60kV 7mA 12microns, hammer about 5cm from chamber center.

I’m stumped as to what is going on, but guess that the X rays were interfering.
JoeBallantyne
Posts: 327
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: Modern neutron detection - Ludlum's new scintillator

Post by JoeBallantyne »

There are two independent detectors. One for neutrons, one for gamma. I believe the bottom end of the gamma response is about 60keV.

I expect you were simply measuring xrays at 60keV. Which you would have been getting once you pushed the voltage up to 60KV. Clearly you were getting a lot of them.

You should have been able to switch counting the xrays on the meter on and off by simply toggling from N+G to N. N+G will count the xray exposure, and N only should not.

Joe.
Post Reply

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