Oct4Run1023

[Doug Coulter]

The 1" eljen button is roughly twice as good (efficient) as the best I've made (or anyone else I know of, having tested some other attempts made and sold at HEAS). It's giving me about 980 cpm per 1m neuts/second right at the fusor shell - well, about 1" from it. At the same time, a long 3He tube is in pulse pileup with the tube about 14" away horizontally, and an old 18" GE B10 tube roughly 3' away is giving me about the same cpm as the hornyak. I've eliminated EMI issues on them all, but still trust the hornyak better, and due to proximity, it gets me a little more info on just where the fusion is occuring. The issue with using ZnS:Ag is that it's much more sensitive to charged particles than gammas, but the downside is the very high index of refraction of it (~2.45?), which tends to trap the light inside the phosphor (total internal reflection). There's nothing plastic or epoxy you can embed it in that lets the light out very well - those all tend toward indexes around 1.45 or thereabouts. You really do NOT need the super sensitivity of a 3He tube, and I had to in fact stop using mine for live data collection due to that pulse pileup issue (on audio though, you can hear the taller/wider pulses fine, it's just that counters only count one for that). Moving it further away to cut down the pulse rate means it will now detect motion of that other big moderator in the room - me. So I just hook it to the audio system, along with the 10B tube, for tuning things up. The ear is a pretty good pattern detector, and I don't have to look away from other controls or gauges to hear the results.

You could probably do about as well with a recoil gas detector. He has a "giant" resonance around 1 MeV, so big that in "Fast Neutron Physics" many experimenters were cleaning high energy neutrons out of their beams with a couple helium baloons. About a megavolt of energy in recoil is pretty easy to detect inside what is otherwise a geiger tube, or so I'd think, and it'd be a lot cheaper to make if you were actually making things, rather than just repackaging them. I have a book from 1945 showing that and a regular hydrogen recoil tube...so the pooh poohing I got on fusor.net when I mentioned that - even statements that it wouldn't work, rather than useful info on what moderator (probably thin carbon) would tend to work best to get 2.45 MeV neutrons down to around 1 MeV was pure bullshit - and part of the reason this board is here. I suppose I should have scanned in pages from books and papers they haven't read that show it DOES work from the 40's through the 60's and told them to eat dust, but I decided to simply stop posting over there - you can take them to school, but they just eat the books, mostly.

As John joked while here - we get hit by murphy's law more often because we actually DO things, not just sit in armchairs and make bad theory. In case no one else has noticed, fusors exhibit emergent behavior, and an armchair theorist thinking about individual particles knows about as much as someone thinking that because they study one ant, they understand an ant hill/colony. Good luck with that.

[Doug Coulter]

Yes, even the Eljen hornyak (which is better than any homebrew I've yet seen, including homebrews by others) is pretty "numb" to neutrons, low quantum efficiency. Unless you're in the survey business, that's actually a good thing. I completely saturate (pulse pileup) the 3He tube here, and almost do the B10 one - we get a lot of counts inside its "dead time", and it's far away from the action - a few feet. What I wanted was something I could put right up on the tank wall if I desired (well, there's about 1/4" lead in between to stop gammas from making it false-count) and have some "locality" of the information. This for example, is how I proved to myself we were getting fusion at the tank walls, where the beams "hit". I was able to just move this tiny thing around the circumference and see counts going up and down, up and down. Absolutely no hope of doing that with a 22" long 3He tube in a 7" diameter moderator, eh? And it would be utterly saturated and not counting at this flux level anyway, so no point even trying that - it's already saturated at a foot away.

I don't buy into the good ol' boy stuff over at fusor.net about "you have to have the detector one of we old guys has" to be sure you're getting neutrons. In fact, super sensitivity is a bug when you get really going hot - the main issue with say, silver, is that there's no real time feedback for tuning things - I use the B10 and the 3He for that, through a stereo audio amplifier. So far, I've not heard any interesting correlations in the audio, but you never know. They both count faster at the same time, but I'm not hearing things like phase coherence moving around, or audio rate pulsing at some fixed frequency or anything like that - they don't even both click at once during an HV arc, which tells me I've cured all the EMI issues; no small thing, that took "heroric effort" to overcome.

All the detectors are shielded, with the preamp inside the same shield. Not enough. All the HV is also separately shielded, with attention being paid to possible ground loops - it looks like coax of varying diameter all the way from inside the supply to inside the tank - that took some doing, but getting the noise at both ends is what it takes when you've got 50kv at maybe an amp (limited by the ballast resistor, we hope) over here, and millivolt signals only a couple feet away...

I suspect JohnF will report his works when he gets a chance to try it. Even a fairly bad one works if you've got the flux of fast neutrons. The really nice thing about them - zero drift over time and temperature. They just work. The pulse is reasonably wide, so they work well with audio-class stuff as well - the ZnS:Ag has a slow fall time all by itself, no shaping is really required. But, John, if you try this, remember ZnS has a looooonnnng persistance at the low photon level, so let it sit in the dark for a day or so before trying it...takes awhile for that stuff to get completely silent if it's been in room light for a bit. He should see a count every couple seconds or so from direct cosmic ray hits, if it's about the same flux in NZ as here.

[johnf]

Hey Doug
the honyack is still a work in progress
I have been given a set of old Bicron NAI detectors some leaking unmentionable bad coloured goop. The first one to open by cutting off the detector part came away with the photomultiplier still glued to the scintillation crystal breaking the leads to the tube socket --3 to go, I now need to find something to dissolve the glue --seems to be sikaflex or similar--might try diesel, it seems to fry most of these synthetic rtv type muck.

[Doug Coulter]

If you find something that eats silicone, please post it! It's a real pita otherwise. Lesker even says so - it's a real bear when changing diff pump fluid to really clean out the old stuff, for example, if it was silicone based.
I should have sent you back with a phototube, we have too many spare here. So far, all the homebrew hornyaks I've tried are at best 1/2 the sensitivity of the Eljen one, but very usable in the kinds of flux we produce.
I just happened to leave the best one in my rig, for some strange reason. Note, in my lashup, I'm using a rather wimpy tube - it's only a gain of 10e4 or so - I didn't need the 10e6 gain type for this, the light is fairly bright on the photomultiplier scale of things.

[Doug Coulter]

Yes, Steven, Hornyaks are in general pretty numb. But we're not prospecting for neutrons (not much Cf in nature, right?), we're actually producing plenty - enough to easily utterly saturate a He3 tube, and begin to see pulse pileup on an B10 tube (old type, not BF3) 3 feet from the action. So the Hornyak is what we use as a reference around here, it's by far more stable and useful for tuning things. Our other neutron detectors we simply put into an audio amplifier and listen to - nice real-time feedback you don't have to look at while doing tuning kinds of things with your hands and other senses.

In addition, I've experienced an effect on the moderated type 1/v sensitivity detectors. They are sensitive to where I am in the room - I'm a further moderator/neutron reflector, and that makes all the calibration of those rather suspect...when I move around, they count differently, while the hornyak stays perfectly stable (or as stable as the fusor, at any rate). Standing right behind the He3 tube at low neutron outputs (so you can even tell), my body adds about 30% to the count rate of that thing - and it's already in a 6" diameter 30" long piece of HDPE for a moderator. Now, I wouldn't normally stand there - it's hot in X rays there too - but...even squirming around in my operator chair, some feet away, seems to affect the moderated detectors.

This is a case of "what's the best tool for the job at hand". Hornyak wins hands-down in high flux situations.