[Prev][Next][Index][Thread]
Re: Results of Paper & Poly & Oil
---------- Forwarded message ----------
Date: Mon, 29 Sep 1997 14:22:04 -0400
From: Thomas McGahee <tom_mcgahee-at-sigmais-dot-com>
To: Tesla List <tesla-at-pupman-dot-com>
Subject: Re: Results of Paper & Poly & Oil
----------
> From: Tesla List <tesla-at-pupman-dot-com>
> To: tesla-at-pupman-dot-com
> Subject: Re: Results of Paper & Poly & Oil
> Date: Monday, September 29, 1997 1:12 AM
>
> ---------- Forwarded message ----------
> Date: Sun, 28 Sep 1997 23:21:35 +0500
> From: "Alfred A. Skrocki" <alfred.skrocki-at-cybernetworking-dot-com>
> To: Tesla List <tesla-at-pupman-dot-com>
> Subject: Re: Results of Paper & Poly & Oil
>
> On Sun, 28 Sep 1997 11:57:22 -0400 Thomas McGahee
> tom_mcgahee-at-sigmais-dot-com> wrote;
>
<SNIP>
> degrading at the edges of the plates is somewhat expected, I would
> think increasing the overall size of the dielectric in relation to
> the conducting plates would tend to eliminate this. I'm not sure of
> what you mean by "ends of the plates" though, I assume the
> discoloration is occurring all along the peripheral edges of the
> conductor plates. Could you be a bit more descriptive as to what
you
> are calling the ends though?
>
> Sincerely
>
> \\\|///
> \\ ~ ~ //
> ( -at- -at- )
> -----o00o-(_)-o00o-----
> Alfred A. Skrocki
> Alfred.Skrocki-at-CyberNetworking-dot-com
> .ooo0 0ooo.
> -----( )---( )-----
> \ ( ) /
> \_) (_/
>
Alfred,
My flat plate capacitor "ends" refer to the section of the capacitor
plate that is furthest away from the plate connection point. A
diagram may help clarify. Here is a side view of the plate positions:
______________________
| ______________________
|_____________________ |
| _____________________|
|_____________________ |
_____________________|
^ ^
| |
| |
These are the "end"
points mentioned in
my original post.
The corona problem here is much greater than anywhere else along the
periphery of the plates. What surprised me was how MUCH greater it
was! I believe that because the opposite plate extends underneath the
poly and past these "ends" that the corona is thereby extended much
more than it is around the outside edges. The corona is sort of drawn
across the poly surface because of the proximity of the opposite
plate.
By the way, the *corners* did not show any appreciable increase in
corona over the ends. That initially surprised me, as I thought the
sharp edge there would have created an even greater e-field. But it
seems that it is not so much the sharp edge that is creating the
corona, but the extension of the opposite plate beneath the poly.
This has implications for the way we build our caps, whether they are
flat or rolled. The "ends" need to have MUCH more linear insulation
distance than the outer edges. My guess is that something like three
times the distance required on the outer edges.
Please note that the failure mechanism here can take place over an
extended period of time. I would be interested in knowing if anyone
else who has disassembled their HV caps has noticed this increased
tendency to have corona along the "ends" as opposed to the sides.
Hope this helps.
Fr. Tom McGahee