This is experimental expansion of a solid instead of a gas to see if electricity can be generated directly instead of generating mechanical energy first, like most Stirling engines.

I am heating and cooling the dielectric inside two charged capacitors to push electrons back and forth between them.

I think it is not carnot cycle since it is not a gas expansion. However, I suspect efficiency is directly related to the absolute Kelvin temperature ratios, like carnot. This idea eliminates friction. So it could be relatively efficient. However the device might need to be physically huge. I have not done the math yet.

My next experiment will be with two open homemade flat plate electrolytic capacitors at higher voltages and faster heat transfers. Another idea is two 12000 volt tesla capacitors. Energy density is the square of voltage so higher voltage might enable smaller devices. However capacitance suffers and dielectric properties like thermal expansion coefficients matter. Plus thermal loss of the heating and cooling of plates and electrolyte is a complete loss, since the dielectric is all we really want to change. So keep them very thin and add a regenerator equivalent.

MIT demo of changing dielectric thickness:

Loading more stuff…

Hmm…it looks like things are taking a while to load. Try again?

Loading videos…