BC-720 Revisited

[Richard Hester]

I've been thinking of a way to make a device similar to the BC-720, and from the description of the device, it sounds like one could be made by milling coencentric grooves in a disc of plexiglas, then pouring ZnS loaded molding compond on top, with perhaps a short stay in vacuum to get the bubbles out. I was thinking of trying something similar, but with a hexagonal matrix of 1/8" diameter holes instead. A piece of commercial perforated metal makes a dandy template to spot the holes. I have a big, fat, slow (but fast enough) 5" photomultiplier that would make a nice guinea pig, and Tap plastics down the way has a bunch of cheap 1" thick lucite scrap with my name written all over it...

[guest]

Hi Richard, Merry Cristmas everyone!, I've been pondering the BC720 copy idea a fair bit myself, You'd have to ask Richard Hull to confirm this but I think the ZnS:Ag layer may be just a thin painted on layer with a clear resin backfill, I copuld well be wrong but thats the impression I get. When it comes time to mill or route some plastic I was thinking a grid pattern may be easier to make than a circular pattern which would likely need a CNC or some specialised milling table to make. Also is there any chance of obtaining a schematic and board pattern file for your preamp/discriminator once you have it tested and running?. It certainly looks superior to anything I'd be designing in the near future.

regards
Mark Harriss

[guest]

Oops, I see you've already posted the schematics Richard

[guest]

Not quite - what I posted on Christmas Eve was a few variations on the theme of the original Hamamatsu preamp. I will probably use the second schematic in that series for any work I do with NaI:Tl or ZnS, but the signal could probably use a bit of stretching to make life easier for the discriminator. I just finished building and debugging my discriminator circuit last week and haven't had the chance to test it with actual signals yet due to all this Christmas nonsense. More on that subject later...

[jim-frank]

Concentric circles... how would a piece of Fresnel lens material work?
Jim Frank

[guest]

The grooves aren't deep enough to do any good with a fresnel lens.

[guest]

I tried to find a description of the BC-720, but couldn\\\\\\'t find anything. Do you have a diagram of how it\\\\\\'s constructed?

[guest]

Richard Hull posted a picture of his BC-720 a few months back. Looking at it end-on (business end), it looks like a bulls-eye, with alternating rings of phosphor-loaded (white) and clear plastic. It wasn't shown in the picture, but the opposite end of the device was described as being totally white. I also remember the phosphor-loaded rings as being surprisingly thick. This is what leads me to believe that the device is manufactured by machining coencentric grooves into a thick lucite disk, and then filling them with phosphor-loaded molding compound. This would account for the end oppsite the business end of the detector being totally white. A person with a router could duplicate this effort without too much trouble. Perhaps it's time for a re-posting of the original BC-720 picture. Eljen in Texas also makes what appears to be a BC-720 look-alike, from the description.

[Richard Hull]

I will endeavor to rephotograph the big 5" BC-720 and repost the image in image du jour forum this evening.

Basically, the way I figure that they make the thing is to make up several different sized plastic rings and coat the outside of them thoroughly all around with P-4 phosphor. Next, they may pour a layer of resin in a 5" mold to only about 1/16" thickness. Next, let it harden or as it hardens, they sprinkle or spray on a thin layer of P-4 over the entire surface to make the white back side. After this hardens, they lay on top of this thin layer, the bulls eye concentric coated rings. Next, they fill the entire mold with resin to about 1/16" over the level of the rings and it is done, following a polish of the front and back surfaces.

The front mating surface is probably planed flat before polishing to mate to the PMT face. The foregoing is pure supposition on the methodology of its manufacture. The finished piece is a marvel to study and look at. It evokes the expresion "how dey do dat"?

I think that 5" BC-720 was about $500.00 when I ordered it from Bicron a couple of years ago. Not chump change for sure, but neither was the Surface Barrier detector from EG&G/Ortec that I got for myself for Christmas, this year either. I have already made a nice Alpha particle spectrometer with it.

Richard Hull

[guest]

I spent this afternoon drilling a hexagonal matrix of 1/8" diameter, 1" deep holes (don't ask me how many) in a 5 1/4" diameter, 1 1/4" thick disk of plexiglas. I glued a sheet of perforated metal to one side of the disk in order to spot the holes. The disk will get backfilled with a suspension of zinc sulfide in polyester casting resin. No doubt I will have to use a vacuum to get the resin to penetrate all those small holes. What will result is an array of tiny Hornyak buttons, too small to respond in a big way to gamma radiation, but still large compared to the range of a recoil proton. The small dimensions of each scintillator cell will also help the light to escape and be detected. The disk will get slapped on to the business end of a 5" PMT I have lying about. This in turn will get fitted with shielding, preamp, HV supply, and discriminator. I plan to send this assembly, a plastic scintillator setup of similar configuration, and my "Snoopy" moderated detector to Ludlum for calibration. It should be interesting to see how they compare. The "Snoopy" has a direct scaler output, so it should be possible for the techs at Ludlum to come up with a direct scaling factor for neutron flux. Judging from Richard Hull's experience with Ludlum (see the old Intranets archive) this will take some explanation and persuasion.











'

[guest]

Did R.Hull [ some many Richards now ] post the pic's of his BC-720 yet -- or did I miss them?
tks Richard

[guest]

Last weekend I cast my "holey " version of the BC-720 using a 5.25" diameter, 1.25" thick lucite disk, with a hexagonal pattern of 1/8" diameter holes 1" deep drilled into one face of the disk. I attached a strip of 0.01" thick mylar coated with mold release around the periphery of the disk using electrical tape. This served as a dam to hold the casting resin in place. I poured about 120ml of casting resin loaded with ZnS:Ag (4:70 proportion) on the face of the disk and placed it in a small vacuum chamber to coax the air bubbles out of the holes. Wrong move. The resin went into the holes, but also boiled over under vacuum, making a mess and wasting some resin. I added a little more, and poured the rest into a small glass jar coated with mold release in order to cast a small Hornyak button for one of my 1.5" PMTs. I placed the disk in a cardboard box inside a plastic bag, so that the smell wouldn't overwhelm me, and put it up on the shelf in a closet to harden for a week.
This Saturday, I got it down from the shelf and had a look. The resin filled the holes in most places, but there were still a few bubbles, Next time I try this, I'll simply poke into each hole with a toothpick to release the air bubbles. It's tedious, but it works and doesn't leave a mess. The disk wasn't quite level in the box, so the resin at the back ofthe box was uneven. I added another ounce or so, and now it's back up on the shelf curing again, this time level.
What I learned from this little project is the following.
1) Drill the lucite at low speed with a gentle feed so that the chips don't melt. I was forever cleaning chips out of the holes...
2) When adding casting resin, release the air bubbles out of the holes with a toothpick.
3) Eleven drops of catalyst per fl ounce of resin works ok for the small thicknesses involved in casting the detector. However, the Hornyak button I cast with the remaining resin (1.5" dia by ~ 1" thick) had a slight hourglass profile due to shrinkage. It doesn't really affect the button's usefulness as a detector, but next time I specifically cast a Hornyak button, I'll use less catalyst as required for thicker cross sections.