Water Cooling the Cube Fusor

[Jon Rosenstiel]

I tried forced air cooling and it did help to stabilize neutron output, but it was not stable enough for what I wanted to do. What follows is a brief report on my water-cooling journey.

First order of business was to fill the original threaded mounting holes in the bottom of the cube with aluminum epoxy. As can be seen in the below image, the filling didn’t go as smooth as I had hoped because the epoxy was too viscous. (Some of the commenters on Amazon had complained that it was too runny) Luckily, it turned out that the cavities were in areas that didn’t really affect anything.
Next came milling a 5/16” wide by 3/8” deep groove into the bottom of the cube followed by the drilling and tapping of four new mounting holes at each corner. The 3/8” thick aluminum baseplate was drilled and tapped for two 3/8” NPT aluminum barbed hose fittings, water inlet at the center of the maze and water outlet at the end. Tapping was a slow trial and error process as I needed the ends of the hose fittings to be flush at the baseplate/cube interface when the fittings were tight. In an effort to prevent corrosion in the cooling loop everything (fittings, radiators, etc.) was either aluminum or plastic.

I purchased two 240 mm computer radiators from an eBay seller. On each radiator I mounted two 120 mm high-flow, 120 Vac muffin fans. Coolant pump was a Little Giant submersible PE-1 series. Baseplate/cube interface gasket was cut from 1/16” thick rubber-cork gasket material.

To test out the H20 cooling system I ran the system for a total of 40-minutes, 20-minutes at an input power of 400 W and 20-minutes at 900 W. At the 20-minute mark the cube’s temperature measured 41-degrees C, and at the 40-minute mark it was just under 61 C. (Thermocouple attached to the cube’s vertical midpoint near one end)

A few days later I ran a series of 0-degree/90-degree inverse-square measurements at 352 W input power. (44 kV, 8 mA) Twenty-six measurements were taken over a period of 38-minutes. Beginning temperature was 38.5 C and ending temperature was 40.8 C. To monitor the fusor’s NPR (Neutron Production Rate) a moderated 1” x 22” He3 detector was positioned at a 30-degree angle to the cube’s beam-line approximately 36-inches away. Average NPR for the 26 measurements was 393 cps, high was 403 cps and low was 384 cps. I’m happy with that.

Jon Rosenstiel

The machined maze. Those owning or having access to a CNC mill could machine a cool looking spiral.

Underside of the baseplate showing cooling water connections.

Radiators mounted to the fusor bench end panel.

Fans, plumbing, and 2.5-gallon coolant reservoir. Pink color is from Water-Wetter corrosion inhibiter additive.

[Dennis P Brown]

Nice machine work and plumbing; hope it works well.

[Mark Rowley]

Jon,
That is phenomenal! I’ve been dreading the idea of any more milling or machine work on my fusor but your arrangement is fantastic. I’ve got ample room and girth to put a milled waterway on each side of my fusor.

I’ll begin practicing on some spare aluminum plate and move on to the chamber in a week or so.

Many thanks for posting this!

Mark Rowley

[Jon Rosenstiel]

Thanks Dennis, and you’re welcome Mark. And yes, practice is a really good idea. I did three practice runs before attempting the real deal.

JonR

[Jim Kovalchick]

Nice! This is the water jacket I envisioned..

[Richard Hull]

Really great work, Jon!

I have been seriously thinking about this effort if fusor VI is ever bought into being. I have the body here for VI, but will it be a 10-inch sphere or an 8" cylinder? Both are in hand.

I'll stay with fusor V a bit, but not as long or even close to as long as fusor IV. Water cooling is a must for long term stable fusion at the high end. This is especially true for the longer half-lived isotopes that are sought to be activated. Without cooling most fusors are doomed to activate the short lived, quickly activated isotopes of Indium, Silver, and Rhodium. I could kick myself for passing on Rhodium when it was a steal at $750 per troy ounce a few years back. As noted Tony De Angelis brought his newly purchased Rhodium bar to HEAS 2017 and I was super impressed as fusor IV really slapped it way up in activation very quickly. (very large cross section of its 100% single natural isotope to its neutron activated 54 second half life.

Indium is but 13 seconds and Silver is 24.5 seconds and both cross sections are merely nice, not anything to write home to mom about.

Again, superb machine work, as always.

Richard Hull