accelerator column

This post also appears on the Fusor.net site. I figure that I might as well spread the info around as the project grows.

The quest for an accelerating column for my Van de Graaff generator has made some progress in the last few days. In order not to give unfair advantage to the VDG over the micro-cyclotron project, I decided to level the playing field by making the column on the back patio using only simple hand tools. The only electrified tool was a hot air gun. Of course I could have done just as well with a torch of alcohol lamp, so the rules were bent only slightly. All the parts so far have been right off the shelf from Home Depot.

The parts list includes five 3/4" to 1/2" copper reducing couplings and one 1/2" x 12" piece of polyethylene tube. Total cost: ~ $10. This being a 'proof of concept' effort, I made no attempt at dimensional accuracy or surface preparation except for removing the scanner labels. I mused about using Scotch Brite on the edges of the copper parts, but they looked pretty good already and I was more interested in the assembly process than vacuum cleanliness.

The assembly proceeded by using the hot air gun to soften the end of the PE tubing to the point at which I could force it over the wooden handle of a screwdriver that was secured in a vice. This gave me the desired shape for fitting over the 1/2" end of the copper coupling. Using a hand cutter designed for cutting plastic irrigation tubing, I allowed a slight overhang of the PE before parting it. The PE started with an I.D. that was smaller than 1/2" before being shaped by the screwdriver handle and now had an O.D. slightly larger than the 3/4" I.D. side of the fitting. By lining up two couplings in the vice, with the section of PE nesting between them, I was able to extrude the couplings together. Some amount of force was required to produce this extrusion and the parallel faces of the vice kept the couplings in remarkable alignment throughout the process. After repeating this tactic on all five couplings I was rewarded with an assembly that was very stiff and rolled on a flat surface like a piece of drill rod. This outcome surprised me because I was also dreaming up an assortment of methods to keep the assembly straight when all the parts were together. Now it's like sighting through a gun barrel and with no remedial help from me.

An advantage to this type design is that the ion beam should see a very smooth voltage gradient because the electrodes overlap. This overlap should shield the beam from being influenced by charged insulators caused by rouge charged particles.

The next step was to test the electrical resistance between each coupling. My insulation breakdown tester is limited to 5 kV DC and each section passed with flying colors. This gives me a rating of 25 kV minimum for the column as it sits. As a non sequitur, I'd like to add that I think it's a good thing to make an x-ray image occasionally, so this project was chosen to scratch that itch. (Sorry, I can't seem to get an image to attach itself here. Help!)

In the next installment I'll try to describe making some simple corona barriers and then festoon the column with 200 M ohm resistors to act as equipotential bleeders. I'm aiming for at least 10 kV per section when the inside volume is under vacuum. As there is no real heat generated at the currents that I'm planning to use, the PE should provide good service under reasonable vacuum. We'll see.