McMaster Carr Mica Sheets

From:  Bert Hickman [SMTP:bert.hickman-at-aquila-dot-com]
Sent:  Thursday, April 02, 1998 7:08 AM
To:  Tesla List
Subject:  Re: McMaster Carr Mica Sheets

Tesla List wrote:
> ----------
> From:  Hollmike [SMTP:Hollmike-at-aol-dot-com]
> Sent:  Wednesday, April 01, 1998 12:32 PM
> To:  Tesla-2; tesla-at-pupman-dot-com
> Subject:  McMaster Carr Mica Sheets
> Hi coilers,
>     I have recently discussed the use of mica sheets obtained from McMaster-
> Carr as a dielectric in a plate cap.  The catalog claims 20kV per mil
> dielectric strength.  I, however, just proved otherwise by accident.  I set up
> a full-wave bridge rectifier using 10kV diodes from my 12kV neon.  As I raised
> the voltage to the neon,  everthing was fine until I reached full voltage.
> The peak voltage was at 19.7kV(perhaps a wee bit more).  I heard the fizz of
> the dielectric puncturing and shorting two of the plates.  No damage occurred
> to anything(except of course the sheet of mica),  but I thought those
> interested should be aware that the 4 mil thick sheets of this mica should be
> considered to have a maximum of 20kV per SHEET, not per mil thickness.  I had
> intended to use 3 layers between each of the plates, but must have suffered a
> brain fart at one point while assembling my cap.  It has been fixed now and
> checked for the right number of layers.  I will be testing it again later(
> after allowing it to sit for awhile as the capacitance tends to rise with
> time).
>    Incidentally,  5kV per mil dielectric strength is consistant with values I
> have seen for mica,  so it makes sense not to assume the value given by the
> catalog.
>    I believe that three layers of this mica will prove to be adequate for 15kV
> neons, but I will post results when I confirm this.  I hope it works well as
> this makes for a nice compact cap.  It also is easier to put together than
> rolled caps and can be assembled under mineral oil to eliminate air bubbles.
> A 0.01uF cap can be made with 8 plates of 4 X 6 inches using three layers of
> this mica.  I cut 1 inch radius curves on the corners of the plates to reduce
> stresses so the area is 23.1416 sq. inches per plate.
> Party on!!!!
> Mike Hollingsworth


Based on the physical size of these plates, it sounds as though the
material is reconstituted mica - fragmented mica that's held together
with a resin binder. Although the mica itself may have relatively low
loss, the binder may introduce some dielectric heating. Pure mica plates
will look very shiny and very flat, and can be easily peeled apart to
form identical, thinner, sheets. Reconstituted mica has a very
polycrystalline appearance.

Good luck, and please keep us informed on how well these work!

-- Bert --