# High voltage horizontal rotating test platform

For my testatika and my vacuum energy tapping experiments I needed a rotating arm or disk with the following features:

• adjustablity to get rid of wobble
• able to distribute high voltage to segments and parts mounted to the arm or disk
• very adjustable way of having parts that faced the arm or disk but that did not rotate

Basically, since I can't make precision parts I had to make something with the correct adjustablity to make up for the lack of precision. This is what I came up with.

## The pulley belt

The pulley belt is a long length of heatshrink tube normally used for adding electrical insulation to wires. The following video shows how it was made.

## Towers for parts facing the rotor

A problem I've had in the past was how to have infinite adjustability in the position of objects facing the rotor. This horizontal layout was chosen to solve that problem. The towers shown below support parts facing the rotor. Using all the various nuts and bolts, the tilt and height can be adjusted and by sliding the towers around horizontally on the wooden base, their location with respect to each other and the parts on the rotor can easily be changed.

## A sample rotor

Using the above basic structure, pretty much any rotor can be used. I even left room for one that is a complete disk. As an example, the following subsections contain details of the first one I made for this structure, illustrating what I hope are some useful ideas for adjustablity and charging. The original purpose for this rotor was to search for the principle behind testatika principle experiment though I've used it for other things as well.

### How it's charged

As shown below, the rotor carries its own high voltage in the form of a homemade high voltage capacitor. A charging stick was made that has two balls on it, each of which is wired to my positive/negative high voltage power supply. Before starting rotation, the the two balls on the charging stick in contact with two smooth surfaced balls mounted on either side of the capacitor, each of which is wired to one plate of the capacitor.

The HV capacitor consists of an inner copper cylinder cut from a half inch hollow pipe, around which is a sheet of polyethylene wrapped around a few times, followed by a thin sheet of copper. All of this is wrapped with acrylic sheets and further encased in epoxy and plastic endcaps. From the web, the breakdown voltage of polyethylene is 25kV/mm so I wrapped enough polyethylene to make up just over 1mm thick. The capacitance is 39pF. Click on the photos below for larger versions.

The capacitor is held onto the rotor with rubber bands and so that it can easily be replaced with another.

### Selecting the polarities for the rotor

I wanted to be able to have the following options for the polarities of the parts on the two ends of the rotor arm:

• both HV positive,
• both HV negative, or
• one HV positive and the other HV negative.

The following photos illustrate how that's done. The first two are actually redundant since the cylinders of the capacitor can be charged either way: outer cylinder HV+ and inner cylinder HV-, or outer cylinder HV- and inner cylinder HV+.

rimstar.org
Contact:
Liked this? Share it with: