Won't the fields tend to cancel out near the center of the spiral?

 

Won't the fields tend to cancel out near the center of the spiral?

When you say center, do you mean looking at the top/bottom or at its side? There should be a little dead zone towards the center if looking at the top or bottom of it. I just want to make sure I have the proper dielectric and thickness for what I want to do. I am going to use a 20KHz, fixed frequency, 12,000V, 30 mA neon light power supply. That will be another learning experience because I have never had the need to calculate something like that before. Anyone know anything about dielectric constants of materials and how they are affected by frequency, voltage, and current?

 

RTV silicon runs around 500V per mil, so you would need at least 1/32" for a safety margin above breakdown.

Epoxy depends on specific type. Can be as low as 50V per mil. Epoxies made specifically for HV electronics can be above 400V per mil.

Dielectrics don't conduct current - until breakdown voltage is exceeded, that is.

 

Dielectrics don't conduct current - until breakdown voltage is exceeded, that is.

So you can run any amount of current through a capacitor as long as the voltage remains below breakdown? It seems there must be some upper limit per unit volume of a given material that a capacitor can handle. Or, can a capacitor take all the current you give it as long as you provide an adequate heatsink? The "current" would be proportional to the amount of molecular or atomic electron cloud deformation per unit time and per unit volume and that effect would propogate to the electrons on the other plate as if the dielectric was a wire. Is that right?

 

That's right. Limiting factor is the leads, not the dielectric. Current does not flow "through" a capacitor. Current flows to one plate and away from the other plate - charge builds up on either side of the dielectric without flowing through it.

High ESR could also play a factor, of course. There is no such thing as a perfect dielectric.

Side note: other than inclusions and impurities, there are no molecules in metal.

 

Side note: other than inclusions and impurities, there are no molecules in metal.

I was actually referring to the dielectric and not the plates when I said, "The current would be proportional to the amount of molecular or atomic electron cloud deformation per unit time and per unit volume and that effect would propogate to the electrons on the other plate as if the dielectric was a wire."

Do you think that there may be a frequency and/or specially shaped waveform for each dielectric where it would cause them to literally breakdown well below the DC breakdown voltage...............such as the video of that bridge that collapsed from the wind due to the harmonics it created within the system as it blew by?

 

"Galloping Gurdy," aka the Tacoma Narrows Bridge? Maybee. Perhaps you can get the state of Washington to give you a study grant.

 

Well, I decided not to use the Neon transformer because I didn't think the insulation on the wires would hold up after I saw what happened to some test wires I tried it on. I put a thin coat of silicone on the coil and then put aluminum foil tightly pressed on either side of it for use as a conductor. I ran 4 amps of 60Hz AC through the coil and didn't even get a single milliamp out. I also tried the frequency setting which was accurate with the wall socket to 0.001 Hz........nothing again. So, now that the reflected impedence of a capacitor is ruled out, I can start development on the magnetic transistor.