Since I finally had some time to work on my fusor over winter break, I have made some performance upgrades. My shop is unheated, and I had to cut and weld stainless steel when it was 5 degrees ouside, but I've got to do these things when I can.
I have a new high voltage feedthrough installed which replaces the sparkplug I was using previously. I picked this large feedthrough up on ebay, and it required a bit of work to modify for use on the fusor. The stainless steel flange on the insulator was too large to fit into a 4.5" conflat flange, so I attached the feedthrough to the flange via a short piece of 3" tubing welded to a special connecting ring which was machined from 0.25" SS plate.
When I used a sparkplug as feedthrough I would hear occassional arcs which I assumed were flashovers at the sparkplug. I later found that it was actually the insulation in the RG-8 coaxial I was using to carry the high voltage which was breaking down. I have replaced the coaxial with some 60kv silicone wire made by Rowe Industries. I picked up about 12 feet of the wire at a hamfest last year for a couple of dollars. I have tested the new feedthrough and wire at up to 40kv so far, and everything seems to work great. Hopefully, this feedthrough should allow me to hit voltages of at least 60kv without any problems.
The other major change I made to the fusor was to convert it to a triple grid system. The ionization grid and filament are fed voltages through a dual pin feedthrough which I welded up using a couple of small weld lip feedthroughs I purchased from Kurt J. Lesker and a 2.75" conflat flange I already had.
The ionization grid is constructed from 0.031" SS wire and is 6.25" in diameter. It has a 2.5" opening at the end to allow me to insert my inner grid. With a chamber inner diameter of 7.75" I have a 0.75" spacing between the wall and ionization grid. The grid is supported by 6 pieces of alumina tubing which slide over a short length of wire which has been welded to the ionization grid.
The filaments are salvaged from 5U4GB vacuum tubes. Richard suggested using these tubes when he was visting a couple of months ago. I cut apart 3 tubes and spot welded the 6 filaments together to make a two foot filament. I then constructed a hoop from 0.060" SS wire with little feet sticking out radially. To this hoop I wired on ten short pieces of alumina tubing. I then threaded the filament through the pieces of alumunia rod and spot welded one side of the filament to the hoop. The entire filament assembly was then snapped into the copper gasket which seals the two hemispheres of my chamber together.
After connecting everything together I bolted up my chamber and pumped it down. I then ran a strong glow discharge in the chamber to clean the new parts I had installed. I left the system under vacuum for a day to look for leaks, none were found - I have now made over 30 welds which are exposed to vacuum in the system and have never had any leaks. Next the filaments and ionization grid were tested. At pressures of several millitorr a voltage of a few hundred volts on the ionization grid produces a strong glow throughout the chamber. As the pressure drops the intensity of the glow drops. An increase in gas pressure or ionization grid voltage increases the the current through the system. The acceleration voltage and current are basically independent in this mode of operation. As the acceleration voltage is increased at current actually begins to drop, and the ionization grid voltage has to be increased somewhat.
At pressures above 5 or 6 millitorr operation becomes unstable. In my system as I increased the current above about 5 ma the current jumped up to about 50 ma and a bright glow was visible in the entire chamber. If left at this setting it would continue to oscillate between these two modes of operation. As the pressure in the system was reduced this oscillation disapppeared.
Another observation is that a rotary pump is inadequate for the operation of a triple grid system. My pump oil had water in it from pumping my chamber down several times and the base pressure was only about 2 or 3 millitorr. When I was flowing deuterium through the system at a total pressure of 5 millitorr, the gas in the chamber was mostly air and water. This was easily seen by looking at the discharge in the chamber, at lower pressures the discharge looked like pure air, as the deuterium flow was increased the clean pink star began to emerge. It looks like a diffusion, turbo or molecular drag pump will be needed for operation in this mode.
With all the difficult modifications made to my system, I now just need to make some changes to my pumping system and I should be able to hit some pretty high neutron outputs.
Joe Zambelli
Created on Tuesday, January 16, 2001 5:48 PM EDT by Joe Zambelli