Simple, Low Loss Capacitor / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Simple, Low Loss Capacitor

Move Thread

LAN_403

ThereIsNoSky

Sat Dec 04 2010, 04:51AM

Registered Member #2423 Joined: Tue Oct 06 2009, 02:49AM
Location: Oklahoma City
Posts: 10

I have been thinking about Low Loss Capacitors a lot lately. I've been thinking that a concentric sphere capacitor would be ideal, so I came up with this crazy idea to use a metal sphere as the internal "plate". Next I would coat it with a healthy layer of wax acting as the dielectric. Then I would coat it with conductive paint, and then I could coat it with some more wax. Bam, low loss capacitor relatively easily. A couple of points:

- There are no plate "edges" which eliminates the "sharp" points on the plate where most corona leakage occurs.
- Using the Wax as my dielectric will cause variations in the distance between the two plates. This doesn't matter because I'm not striving for a precise capacitance. I just want low loss.
- The Max Voltage will be determined by the thinnest area of wax. Considering the properties of wax, I don't think it should have to be too thick.

Next, I started thinking about how much of a pain in the butt it would be for me to get conductive paint, and the expense, and the hassle of applying it ( Remember, I'm the guy who spilled wax ALL OVER my kitchen a few days ago...). So, I started considering alternatives. What if I make the internal sphere the HV plate. Next, instead of conductive paint I will just wrap the dielectric wax layer with aluminum foil. Once again, this will decrease the final capacitance by creating more space and greater variances; however, I believe that it will have a negligible effect on charge loss due to corona leakage. Right now I'm trying to talk myself out of that last phrase, but that's why I'm posting it. What do you all think?

-ThereIsNoSky

Proud Mary

Sat Dec 04 2010, 08:52AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

ThereIsNoSky wrote ...

I've been thinking that a concentric sphere capacitor would be ideal, so I came up with this crazy idea to use a metal sphere as the internal "plate".

What about Gauss's Law?

Ash Small

Sat Dec 04 2010, 09:04AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

From Wikipedia (Gaussian Surface):

"for a spherical Gaussian surface of radius r < R the enclosed charge is zero: hence the net flux is zero and the magnitude of the electric field on the Gaussian surface is also 0 (by letting QA = 0 in Gauss's law, where QA is the charge enclosed by the Gaussian surface)."

Does the outer sphere form a Gaussian surface? It can't be a complete sphere as the connection to the inner sphere has to pass through it at some point.

ThereIsNoSky

Sat Dec 04 2010, 09:53AM

Registered Member #2423 Joined: Tue Oct 06 2009, 02:49AM
Location: Oklahoma City
Posts: 10

Now I understand your reference to Gauss's Law. The internal sphere or plate should be as close to a perfect sphere as possible, but it will have a small hole through which it can be charged perhaps by a diode string rendering neither the inner nor the outter plate a perfect gaussian sphere. This will be similar to a VDG sphere, allowing us to hold as much charge as possible without corona leakage. The external plate could even be an actual square plate sitting on top of the insulated sphere.

I guess my point is that the position / orientation of the external, low voltage plate is more or less inconsequential when attempting to reduce charge leakage. The position / orienation will affect the amount of capacitance produced, but as long as the HV element is properly shaped and insulated by dielectric it has no effect on charge leakage.

Now, my understanding of non-plate capacitors is a little rusty, (non-existant) but it seems like this should work.

Proud Mary

Sat Dec 04 2010, 10:09AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

ThereIsNoSky wrote ...

Now I understand your reference to Gauss's Law. The internal sphere or plate should be as close to a perfect sphere as possible, but it will have a small hole through which it can be charged perhaps by a diode string rendering neither the inner nor the outter plate a perfect gaussian sphere. This will be similar to a VDG sphere, allowing us to hold as much charge as possible without corona leakage. The external plate could even be an actual square plate sitting on top of the insulated sphere.

I guess my point is that the position / orientation of the external, low voltage plate is more or less inconsequential when attempting to reduce charge leakage. The position / orienation will affect the amount of capacitance produced, but as long as the HV element is properly shaped and insulated by dielectric it has no effect on charge leakage.

Now, my understanding of non-plate capacitors is a little rusty, (non-existant) but it seems like this should work.

Charge Density in a System of Concentric Spheres

IntraWinding

Mon Dec 06 2010, 01:16AM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

Does thinking about the possibility of locating the circuit that uses this spherical capacitor within one or other spheres help? Just lateral thinking there... The main thing with high speed performance capacitors seems to be reducing series inductance. Next best to a sphere is a toroid or cylinder. You can eliminate inductance with a "transmission line': how about a cylindrical capacitor with coax through the middle for minimum connection inductance at each end?

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob