2" Hornyak build

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

2" Hornyak build

Postby Doug Coulter » Sun Jul 17, 2016 3:35 pm

I'm finding out yet again that any system of categorization fails here. This could be in fabrication just as well, but since most of the fab is "not real hard" perhaps it's better here, since it's for measuring. Going forward I will put build details at this link, and keep this thread to the metrology details.

Some time(s) ago, BillF had brought me a hornyak button made I think by Jon Rosenstiel (fusor.org), and a couple nice 2" phototubes of unknown provenance. One was kind of wired up - with 40 megs per resistor in the dynode chain(!), and I decided to use the other one for this job - and I will probably use much lower values (like 100-200k) for this, since I have them. Even one meg seems too much per dynode for an 11 stage tube. If it's like the one it is supposed to be like....big if, they look similar but the one I have almost-specs on has a real base - gain of up to 107 is possible with these. Since my other hornyak is 1" (button from Eljen) and uses a 5 stage tube (surplus from a mass spec) that has 105 gain, I think I won't be pushing it too hard here.

I believe this tube is bi-alkali, which is good because IR is darned hard to stop. This sort is mostly blue sensitive, and the ZnS:Ag in a hornyak is...blue. Still, from prior experience, I'm going fairly hard at making a light tight enclosure. As you can see in the pic, I've already glued the button onto the tube (using Elgen's good optical epoxy, but really, only because I've got some) and covered it with copper tape (still around, search amazon, but be sitting down when you check the price). I used a piece of super thin PCB material (like they use for inner laminations in multilayer boards) for the end.
I will probably put Cu tape on the tube one more 2" extended from what's on there and connect them together. Not only for light, but because I plan to run the cathode negative of ground, so the floating anode will be at ground, and I've had leakage noise issues with other phototubes doing this. So the copper around the tube will be at the negative supply potential. I will wrap the PVC pipe with metallic tape also, but it'll just be grounded.
Here's what I have so far:
HornyakParts.JPG
Most of the pieces


The Hornyak button can be lit up by gammas too, so I made a lead end cap for this similar to my other one, so as to at least slow them down a bit. As it worked out, the 2" pvc pipe is 2.4" OD, so I used a hole saw to put a sloppy 2.5" hole in some hardwood scrap I had. I cut off a little piece of 2" pvc to use as a punch and chamfered one end, and used a bit of plywood to adapt it to the pusher on my hydraulic press. Using a little rule of thumb and eyeball, I pushed the pipe through the wood block and formed the lead into a cap I'll be able to solder to the Cu tape on the outside of the housing later on. Did a nice job (don't try this at work, but for a one-shot job at home, fine). Not shown is the 1/4" long piece of 1.5" pvc I cut to space the end of the Hornyak back from the cap, as they will be 1000 or 2000 volts different in potential.

The HV board (the one shown will be modified a good bit, it was originally for some other use) will be on the aluminum bracket, and wrapped in EMI shield, so as not to put that nasty 50khz noise into the signal coax from the tube, which will pass underneath the aluminum bracket. I'll probably dedicate some sort of wall wart to this, and have a bnc output, but if I can manage it, I'll build in some audio too. Work in progress...you know how it is.

Oh, that aluminum end cap for the power input and signal output end came out nice on my lathe but didn't want to cut off clean with my tooling, so I brute forced it on the metal bandsaw. It left some real pretty artsy-fartsy marks on the end...so I left it that way. It's a tight press fit into the PVC.
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: 2964
Joined: Wed Jul 14, 2010 8:05 pm
Location: Floyd county, VA, USA

Re: 2" Hornyak build

Postby Doug Coulter » Thu Aug 25, 2016 6:33 pm

So, why does a guy who already has all the neutron detectors you could imagine want yet another?

Well, all the other ones are moderated slow-neutron detectors. This makes them very broad in their spatial response (as well as time smeared). The one other hornyak is our standard, kind of affixed to the fusor, and never moved, as I agree with Richard Hull on that - once you have a calibrated standard, you don't mess with it.

Yet I wanted a portable one with some spatial resolution to track down something a bit odd. As most scientists know, "hmmm, that's funny" is where most of the good science comes from. Well, we had one of those. There is a ring of super hot gammas coming out of the front of my tank in about a 20" diameter circle. And it's just a ring, not a filled circle. These are very loud - drive most geiger counters and other detectors into blanking or extreme "get out of here" mode. What's really odd is that when we check the spectrum with a GR 130 - it's 84kEv and a fairly narrow line.

There is nothing at that voltage in our system. Further, this is on the "user side" of plenty of lead thick enough to easily stop 84kv X rays. In the vernacular "WTF?".

The working hypothesis is that there is a heck of a lot of neutron generation around that ring, for whatever reason. That almost makes sense with our geometry, maybe. Nothing else really does. At any energy we think we have on the deuterium ions, there is no such thing as neutron beaming (that takes mega-volts). So let us make the assumption that it's local - perhaps some D's are getting going fast enough and hitting other D buried in the tank walls around the front where it's thermally cool enough for the embedded D to not just be driven out (the place we thought most of the "action" was gets very hot very fast, and yes, we see neutron production go down as that happens).

But a gamma spectrum is just that - one piece of info. Doesn't prove the hypothesis. If it's fast neutrons somehow exciting the K line of lead (which just happens to be right there and is where lead absorbs *the best*) near the user side, we'd kind of like to know that, it could lead to some interesting new things to try, and would also indicate we've been doing a pretty big factor better than we thought all along. To test that idea, we also took a gamma spectrum back near "the action" and yes, there's a bit of 84k, but nowhere near as much...that's becoming extremely interesting! Both areas are similarly shielded with sheet lead, overlapped and soldered.
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: 2964
Joined: Wed Jul 14, 2010 8:05 pm
Location: Floyd county, VA, USA


Return to Metrology

Who is online

Users browsing this forum: No registered users and 1 guest