This is still a work in progress. Making this thing in such a way that it meets the specifications necessary to test the theory is very difficult. For example, I spent a few weeks in April working on finding a good enough way to make a barium titanate dielectric.

Even then I may have to build a new one since I realized that the air gap I left between the two output cylinders is too small and will arc when the voltage that I calculated is necessary is applied.

At this point I am able to demonstrate that the output cylinders embedded in the capacitor hold their charge, even when they are "shorted". This is crucial since that charge has to be there while the voltage is being fluctuated in order to create particles.

A demonstration that charging works

The following is a video demonstrating the charging and that the charge stays even then a load is applied. Note that after making the video I redid the test using a piece of copper (i.e. low resistance) as the load and the output cylinders still held their charge.

The power supply used in the test in the above video was one with three outputs: HV+, ground, HV-. Click here for more details about this power supply.

The following is the sequence of steps demonstrated in the above video.

These are the last steps shown in the video. Their purpose was to demonstrate that the copper mesh output cylinders really were charged. This is done by first neutralizing the capacitor plates (the outer aluminium perforated cylinder and the inner copper rod.) After step 5 below, there is no more capacitor charge to hold the charge on the output cylinders but their charge has nowhere to go. Then, in step 6, a path is given. Notice that the current flows in the opposite direction that it did during charging in step 3 above. This is shown in the video on the scope when the polarities in step 6 are the opposite of those in step 3.

How it should work when finished (if it will actually work)

This is what everything should look like once the output cylinders within the capacitor are charged and the variable capacitor rotating and adding voltage fluctuations.

Construction

The capacitances.

The Barium Titanate and epoxy dielectric straight out of the mold.	The first output cylinder is attached.
Two layers of acrylic sheet spacers are added, including in the middle for where the second cylinder's ends will meet. Notice that this will leave an air gap between the cylinders.	The second output cylinder is attached - view showing the place where the ends meet.
The second output cylinder again - showing the side where the wire is attached.	Two layers of acrylic are wrapped around - notice the cylinder is completely covered.
The perforated aluminium outer cylinder is added. Notice the nylon fishing line wrapped around it holding the cylinder in place and holding the connecting wire firmly against the cylinder.	Spacers are added to the end and epoxied in place. This was important to inslute this outer cylinder from the two copper mesh output cylinders inside. Without doing this, ions would likely flow between them.
The whole finished thing with arcylic wrapped around it. Notice all the epoxy added to the ends (and where two ends of the acrylic overlap - not shown) and where the wire would otherwise be exposed.	This shows an earlier version and shows how I coat all edges with corona dope.