Re: Rolled foil caps

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Ed Phillips by way of Terry Fritz <twftesla-at-uswest-dot-net>" <evp-at-pacbell-dot-net>

Tesla list wrote:
> 
> Original poster: "Bert Hickman by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <bert.hickman-at-aquila-dot-net>
> 
> Chris,
> 
> Most likely, when you first put the caps into the oil, the oil wicks
> into the small air spaces between the foil and the polyethylene
> dielectric. Since the oil has a higher dielectric constant (k~2.2) than
> the air its displacing, the capacitance initially increases.
> 
> Then, over a longer period of time, some of the mineral oil begins to be
> absorbed by the polyethylene, causing it to swell, and the increased
> thickness causes a decrease in capacitance. Eventually, the poly may
> absorb 10% (or more) of the oil, by weight. This may occur somewhat
> unevenly, leading to the poly gaining a ruffled or scalloped appearance
> after it has been immersed in the oil for an extended period.
> 
> -- Bert --

	Note that a major source of capacitor failure is corona in the air
entrapped in the layer of foil and insulator.  Manufacturers of
oil-filled capacitors usually use vacuum thechniques to get rid of as
much the air as possible; typically, the capacitors are placed in a
chamger from which the air is evacuated, and the air readmitted while
the capacitors are submerged in oil.  The atmospheric pressure forces
the oil into the "empty" voids.  This same technique has been discussed
here in the past, and seems the best way to make a reliable TC
capacitor.
`
	The point I'm trying to make is that if your capacitor absorbs oil,
then it must be going into air-filled voids which could undergo corona
and lead to dielectric puncture.  For maximum reliability you need to
make sure there is no air before you run the thing.  Some have advocated
running at reduced voltage, which might result in pumping due to
vibration of the plates, but nor sure how effective than can be.

Ed