Standard counters

This is bound to get mixed up with things in Electronics, check both. Physics-specific stuff here, mostly.

Re: Standard counters

Postby Doug Coulter » Tue Jan 11, 2011 4:41 pm

We are using activation of things that are easy to activate and short lived, eg silver and indium, to measure the neutron outputs from our fusors. Medium slow (not quite thermal) neutrons do that best. HDPE is a far better moderator than wax (or any other substance we know of) for this. The proposed counter is to be a standard so we can compare this indirect measurement of neutron flux. The "oven" we suggest is the most efficient at activating materials that have resonances just off thermal, like the above-mentioned substances (and several others that aren't as sensitive). There might be some people in a more advanced state making radioisotopes (some are, but they tend to have to own cyrogenic super sensitive gear to even be sure they did it, fusors make many orders magnitude lower neutron flux than a fission reactor -- 106 total neutrons/second *total* vs a fission reactor at 1016 neutrons per second per CC), but we're just doing it to measure neutrons, because every other way in existence suffers from being too expensive, too error prone, or too hard to do, and not repeatable run to run or across labs, all of which this solves neatly. This is described in the other posts above at length, please do read the other posts first.

Turns out that about 1.5" thick HDPE is just right for going from 2.5 mev neutrons to few-eV neutrons, which is where a resonance exists for activating these easy to activate, short half life materials, which makes the sample reusable in short order, and hot enough with relatively few neutrons to measure at all, without having to have to make so much radiation in the lab you put yourself in serious danger doing it. So, that's the design linked above. We put some more HDPE behind the sample to reflect some of the neutrons that missed back into the sample, it adds a little sensitivity to the process, not a heck of a lot, but some. You use short half live things, because due to fast decay, they count fast after activation with minimal neutrons needed to get a usable reading. Long half life things, like say gold, aren't so good as you don't get many counts back for a given neutron flux on them -- they barely come out of the background cosmic rays unless you really pushed a lot of neutrons into them. And you then have to wait weeks before you can do another experiment. Where say, silver is nice and loud. A good 5 min fusor run here gets it to about 1000cpm on a counter such as the one proposed, completely swamping the lab background count rate of maybe 90 cpm max, while not exposing me to lethal levels of neutrons and other radiation just to take the measurement.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Standard counters

Postby chrismb » Mon Jan 31, 2011 7:19 am

I am now gearing up to monitor for things that may or may not be emanating from my new setup. I let the ebay ratemeter go (viewtopic.php?f=11&t=280), which I am really quite regretting now in hindsight because I will have to figure out my own ratemeter if I want to get a PMT/scintillator set up.

Not sure it's a big deal because I guess the big issue is getting something that scintillates with fast neutrons. 'Fraid it is all out of my depth and a little beyond my interest level, really. I'm more interested in particle physics than radiation dosimetry. But, of course, one needs to have some degree of knowing what is happening if there are fast ions flapping around inside your apparatus. I bought one of those little scintilating pager-dosimeters of Geo some time back. I have to trust it will flag up if neutrons are arriving, I've never seen it count anything yet [and why would it?].

Sorry, digressing the thread here, I'm meaning to discuss the issue of a standard way of counting up silver activation, with a view to comparisons, and was wondering if, perhaps, it would be better to equate silver activation in terms of the reading that a 'standard' dosimeter would give, in uSv/h or whathaveyou.

You mention that activating the silver lifts it up 10x the background, so presumably if a typical personal dosimeter were to be used (mine reads 9uR/h for background where I live) then it'd go to 90uR/h?

So, concluding; is there some sort of doserate-from-activated-silver measurement that would be more consistent to cross-check with?
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Standard counters

Postby chrismb » Mon Jan 31, 2011 7:31 am

Other thing to ask - I have one of those cheap little Ukranian Geiger 'counters' that go click once every several seconds (and if it clicks more than once in 3 seconds then it is higher than some recommended level they suggest). So if such a counter were to be monitored for its response to background, then placed behind a silver screen (and behind moderator) then could a minimum statistical deviation for its counts be determined so as to demonstrate activation of the silver?

Again, from Doug's observations, I might casually presume that a x10 increase in the count rate would indicate a flux of 'n' neutrons/m^2 [sorry, Doug, you didn't say how far the silver was from the source].
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Standard counters

Postby Doug Coulter » Mon Jan 31, 2011 9:59 am

In our case, the fusor is in a 6" pipe, on which the oven rests, the first layer between the fusor and silver being 1.5" thick and 4" diameter. We set another piece of hdpe on top of the samples that is 4" by 4" or so. Putting a counter right there would simply ruin the counter from the X rays we get -- it's quite loud at that point (and the counter would spoil the back reflection of any neutrons from the back moderator). We use a 2" sq sample and count the whole thing, which means a detector with 2" area is needed. We lay the silver right on the window of the geiger tube to count it. So, depending on where the action in there actually is, we're 4.5" from grid focus region with the sample. Whether that's where the neutrons are produced is anybodies guess at this point.

Scintillators will count the silver (and everything else), but only if the proper window is on them that will let in betas. Then they tend to count about 10x as many counts as a geiger tube would. A smaller counter will count background and the silver lower simply due to not being hit by all the output of the silver.

Any plastic scintillator will count fast neutrons, and everything else. There is no sure way to make one cross calibrated, due to high variation of gain of the phototube with voltage, type, type of radiation, variable relative response to alpha, beta, gamma variations, along with where you set the threshold to deal with "dark noise". They are just too sensitive in general for this work, and far too variable to use as a standard. They work nice for finding hot rocks in the environment, mostly. Due to a different sensitivity ratio to the big 3 types of radiation, a relative count from background won't mean the same thing as with a standard geiger tube, which we can see here in tests.

To be clear, we only count the silver after the run, not during -- it's a slow, integrating neutron measure, not something you can really use for real time feedback -- a specialized tube is better for realtime (any of 3He, BF3, B10). You could maybe tease absolute neutron numbers out of this, but even RH (and CW) on fusor.net has a >2::1 variance in what he thinks is an absolute number depending on who does the math and "other things". For now, no worries on that one here -- relative is good enough when you're in the microwatts of fusion -- you mainly need to know did you do better this run or not. RH's numbers for about 500 cpm out of silver vary from nearly a million fusions a second to several million neutrons per second, but he doesn't repeat this for us or run short, measured runtimes either. Gets you to order of magnitude or so, but not "accurate". Once you get out of the microwatts of fusion range, you'd better be thinking real hard about doing it from a remote position unless you're a true believer in hormesis.

We set up a NaI:Tl head to see if we can see the extra large pulses from the 16 mev gammas from that rare fusion pathway, results inconclusive (we think we saw some maybe), but its first real run was during the Tech guys visit, and no one watched it real closely (I couldn't because there were people between me and the scope). I'll report on that after the next run.

We like the silver best here for any sort of really sure measurement because unlike every other way to do it, silver isn't fooled by anything whatever. No neutrons, no hot silver, period. Everything else we've used (pretty much everything there is) has other issues.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Standard counters - tubes arrived

Postby Doug Coulter » Mon Jan 31, 2011 1:14 pm

They just got here from Ukraine. See picture.
GeigerTubes.jpg
Russian tubes


They were more than slightly optimistic about the sensitive area these have, which of course means they'll count less for the same silver activation than one that covers the entire 2" piece of silver.

But of course, if we all use the same setup, it's a scaling factor that only matters if you're trying to get to absolute accurate numbers of neutrons, which is nearly impossible with any technology.
Truly, that's only going to matter when we can use a calorimeter to measure net heat gain, other than for safety. For safety, I'd say that if you're getting to over 500 cpm silver in 2" square on a five minute activation, you should stop about there and find a way to get yourself farther away, run less frequently, or scale things down for now. That will be a lower number on these smaller tubes and a smaller piece of silver -- hey, it's cheaper.

This is why we suggest a standard neutron oven as well. All the neutrons in a certain angular area from the fusor hit it, but far from all are taken into the silver -- some scatter back out, some are captured in the moderator, and this is material and geometry dependent. You could cross calibrate one to a BTI detector, if you trust those, then all you'd have to do is explain why putting the BTI's in slightly different places makes one read 3x different than one right next to it -- which to believe, or do you average them? Both Tyler C and I have seen this effect and we are not yet sure if it's due to where the neutrons are coming from (eg not the grid center as one might expect) or some sort of beaming. The 4" neutron oven face spans enough angle to hopefully average any of that out.

Our standard oven was tuned for maximum sensitivity so small neutron productions could be measured more easily and accurately. Silver has a pretty huge capture resonance well above thermal, and the initial thickness is calculated (and tested) to get the most neutrons into this resonance. Silver will also pick up some neutrons of lower or higher energy, of course, but this is the main activation effect.
Cross_Ag.gif
Note x100 part of chart on this one

Cross_Ag_In.gif
CarlW's charts for silver and indium


As you can see in these charts (note log scales) it's only at the resonances that any major capture of neutrons occurs, and that thermal neutrons really don't count much here. Sadly, the moderation is pretty random, so even if we have mono-energetic neutrons to begin with, the moderator spreads the spectrum out quite a lot, there is currently no way I know of to just slow them to the resonance at a few eV (compared to thermal at about .025 eV or so).

Of course, background is highly variable everywhere and everywhen -- here where the lab is pretty clean, cosmic ray variations take a 2" counter all the way from about 6 to about 190 cpm at the extremes in ten second counts (scaled up by factor 6 to get cpm). So basing anything off that is not a good idea. An anti coincidence counter setup would be about the only way to get truly accurate counts at low levels, as this range can be seen in ten seconds. Normally it's more like 2::1 range, the above is the extremes we see here but averaging in the 1 cps (60-70 cpm) range.
A scintillator, being fast enough to resolve individual particles in a burst will see far higher variations and does here -- if you have electronics fast enough to count say, 5ns pulses 5-6 ns apart, as phototubes produce those frequently with a good scintillator even with low (say, ~1k cps depending on variables) count rates on background. The first rule of radiation -- it's the most random thing known to man, and it's common at even 2 counts/second to have both within a very tiny time of one another. This points out another superiority of a geiger tube for this work - it will see a whole burst as just one count, having lousy time resolution and some dead time.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Standard counters - data sheets

Postby Doug Coulter » Mon Jan 31, 2011 2:00 pm

Anybody speak Russian well enough to translate the important parts of these?
russian1.gif
One side of tiny folded data sheet

russian2.gif
The other side


Some I can guess, but I'm not supposed to know Cyrillic, else I'd know too many secrets from my old job and they'd have to shoot me (no kidding). I was only allowed to be in the room with some stuff because I supposedly (and mostly actually) can't read it. Of course, with a geiger tube, it's not hardly rocket science to figure it out on the bench, the good old "test it" way. Just use a big series R and find the pedestal with a known source (we have plenty), and there you are.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Standard counters

Postby chrismb » Tue Feb 01, 2011 7:07 am

Doug, In all the data you have accumulated, is there a beta penetration depth graph for silver?

It is often said on fusor.net that silver foil is best, but if silver has 2MeV betas, then wouldn't that be able to get through a full millimetre of silver? I presume that is the max depth (to how many sigmas??) but there must be some optimum upper thickness of silver such that, after a given activation flux, external beta emissions from that piece would be a max. If the max penetration depth is 1mm (?) then I'd have thought there would be an optimum around 0.3 to 0.5mm so as to maximise activation detection.

Advice welcome... because I am thinking I might try to make a scintillation silver sandwich, and silver price will be the most costly single component so I need to maximise it.

Also, the other question that goes with that, what is the penetration depth of a 2MeV beta through a scintillation crystal?
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Standard counters

Postby Doug Coulter » Tue Feb 01, 2011 9:59 am

I've not seen one for betas in silver, but it's fairly easy to calculate with the stuff in say, Haliday's book. You have to leave some for penetrating the detector window and depositing some in the detector of course. Since electrons lose some fraction of energy per interaction, it's a bit of math as the likelihood of energy loss is not a constant but varies with (remaining) energy.

We are using silver that is .01" (ten mil) thick and it works great. It was the cheapest part from a jewelry supply house, I think about 30-40 dollars for a 2x6" piece (eg three samples at our standard size). As you seem to know already, it's getting the radiation back out, not the neutrons in, that is the limit there. We are careful to use the side pointing at the fusor to lay on our detectors to get consistent (relative) readings there, as that side sees more resonance region neutrons than the other side, and activates better for whatever reason, probably the lousy capture and low numbers of reflected neutrons (at resonance, or any energy) in the moderator stack. We've not tried thicker than that here, yet, but Bill brought a coin the other day, pretty thick, we might try that if he brings it back next time.

I can't see exposing my scintillators to full-snot radiation from a fusor, in your case it might be different. Fusor.net has some info on silver-painted geiger tubes and it was kind of disappointing.
Further, many many things get activated other than silver so getting good data out of that is going to be quite a challenge -- some of them have long half lives and will constitute an ever increasing background. The I in NaI is especially a problem -- it ruins them for a long time. But the sheer amount of X rays means that during a run, you see nothing but them (and the radiation damage to the scintillator). It takes so much lead to stop those that you scatter neutrons in (and out of) the lead too...there's no easy answer to that one.

A test coming here soon is an NaI head with a very high threshold to only see the really hot gammas from the rare fusion event that goes straight to gammas. We think we saw some of that during the last demo, but I was kinda busy with the full house, and need to repeat that test. Of course, no one has been looking at the relative percentages of the three possible DD reaction paths so far, and that's something we'd really like to know about if fusion conditions are other than thermal where we know the numbers. Could be that with magnetic and electrostatic fields one or the other gets more probable than in a pure-thermal situation? Which might mean that tuning for max neutrons isn't tuning for max fusion!

One issue with fast phototubes and scintillators is dropped counts due to either sheer speed (easily quicker than any TTL you can get) or pileups leading to only one pulse being counted when really there were 2 or three in that threshold crossing. They really don't do well in hot environments, they are more for detecting tiny amounts of radiation. Something more "numb" seems to be more applicable to high radiation work. At some point, you are better off just integrating or averaging the DC current out of a phototube, which of course is far less accurate than counting.

The radiation depth, X0, of a crystal depends on both crystal (or plastic) type, density, Z, and the type of radiation involved. Since you've not responded to my question of what type you're going to use, I can't look it up in the books (basically, the one in the background of a picture just above this). It varies all over the map. Of course, you don't have to catch it all unless you're going for energy spectrometry, just enough to make a count. An inch thick will definitely get you to "counts", often much less. In general, something dense like BGO has the shortest radiation length, followed by NaI, followed by plastic.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Standard counters

Postby chrismb » Tue Feb 01, 2011 11:11 am

PVT, I expect

- but depends on what I can get, really
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Standard counters

Postby Doug Coulter » Tue Feb 01, 2011 12:18 pm

Geo was selling some really nice PVT stuff, we use it here. Of course, that's just the base plastic, not the fluor, of which there are zillions of types optimized for this and that. (Well, maybe 20 or so listed on Eljen or Bicron) Of course, the word crystal is not appropriate for an amorphous plastic, you had me thinking of actual crystals like NaI or BGO, BaF, and so on.

The radiation depth of that stuff is looonnnnggg, but if you just want to see if a particle hit it, an inch square is plenty, it's good and sensitive. It is also (most of the fluors are) very fast stuff, about 2-3ns risetime, so coupled to a good bialkali (blue sensitive) tube, you'll see 5ns pulses out of it, which can be pretty challenging to count with normal gear, especially when two hit inside 12 ns or so, an unfortunately very common occurrence even at low count rates. The plastic will also see fast neutrons, which tend to give wider pulses due to multiple proton knock-on events per neutron, but of course those will be lost in the X rays in most cases (definitely here). There are some old papers in the library about using fast shape correlation to separate the neutrons from the other stuff. Fairly non trivial job of signal processing.

Which of course is why our standard is going to be something else far simpler. It's hard to think of something less appropriate for a fusor than a scint for counting either neutrons or silver activations. We've tried this here but don't use it. It is good for personal safety, as the "Henny Penny" nature of it will make you scared enough to be really careful.

Too many counts (pile up), too much variability on cosmics (resolves early and late particles from some but not all showers) and a host of noise issues you just don't have with a simple, dumb, geiger tube that's already plenty sensitive enough to get good data with a tithe of the complexity and interpretation issues.

Also, the geiger tube can see alphas when you want to look at natural sources, where the PMT/Scint can't unless you can arrange a window so thin as to let them in but not light -- more than a little bit challenging, in my experience. Very thin Be foil might do there.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

PreviousNext

Return to Metrology

Who is online

Users browsing this forum: No registered users and 3 guests

cron