Hi Carl...
It's only a gut feeling, but I'd prefer a single thicker piece of Teflon to
multiple thin ones, because the latter creates more interstices (!?!)
between layers - where air can lie in wait for the chance to getcha.
Think John mentioned rounding/radiusing the edges of he electrodes to avoid
sharp edges AND maybe using some sort of (low-loss, obviously) silicone
paste at all interfaces and especially in the tapered/radiused areas around
the perimeter of electrodes. Guess National & Millen had the righ idea when
they made all those old neutralizing caps with nicely machined or stamped,
rounded plates.
73, Dick
-----Original Message-----
From: km1h@juno.com [SMTP:km1h@juno.com]
Sent: Wednesday, February 17, 1999 9:26 AM
To: amps@contesting.com
Subject: Re: [AMPS] isolation question
On Tue, 16 Feb 1999 12:10:15 -0700 "Richard W. Ehrhorn" <w4eto@rmii.com>
writes:
>
>In real life the actual rf breakdown voltage of a capacitor commonly
>is
>determined by unintended/unrecognized/unknown regions of unexpectedly
>high
>voltage GRADIENT - often many times higher volts per inch or mm than a
>
>simple calculation of (Etotal)/(dielectric thickness) would indicate.
>
>Such regions can and often do occur (a) WITHIN the dielectric, due to
>
>inhomogeneities such as air bubbles or other contaminants, and (b) at
>the
>dielectric surface, typically due to sharp (small radius) edges of
>electrodes OR air gaps between the dielectric and the electrodes. The
>
>latter is almost inevitable unless electrode edges are of large radius
>and
>very smoothly polished, and the electrode-to-dielectric interface is
>filled
>with a material soft or fluid enough to eliminate air pockets and of
>dielectric constant roughly comparable to or greater than that of the
>main
>dielectric.
>
>This problem of dielectric inhomogenieties which create locally large
>
>voltage gradients, which in turn lead to corona discharge, which
>"eats"
>holes in and eventually breaks down the insulation, is at least one
>reason
>why HV caps are often potted and HV transformers are either vacuum
>impregnated and potted or filled with oil. (Of course cooling also is
>often
>a consideration for filling transformers and even caps with oil.) I
>believe
>that relatively soft dielectrics like teflon and nylon generally are
>more
>susceptible to erosion by corona than are harder ones like mica,
>glass, and
>high quality ceramic.
>
>Despite design precautions, the end result is that rf dielectrics
>generally
>must be many times thicker than their "dielectric strength" spec would
>
>indicate. It's also why the insulation in critical power transformers
>-
>e.g., utility distribution transformers and ALPHA power
>transformers(!) -
>is applied on the basis of ~50V/mil rather than the several hundred or
>even
>1000 volts/mil that the insulation's ratings suggest.
>
>73, Dick W0ID
Nice and concise Dick, thanks for that info.
I have, and sure others also, blown the Teflon sheet as used in many ham
type low pass filters. The failure points always seem to be around the
edges of the aluminum or brass disc; corona effect as you have suggested.
Repair attempts have not been successful and it has been about 10 years
since the last try.
Do you have any suggestions? Would making the dielectric from 2 or more
thin sheets with a film of silicone be an improvement?
BTW, I have only had this problem with HF and 6M LP and BP filters; not
in VHF/UHF amps when used as bypass/feedthru caps at voltages up to 4
kV.
___________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
or call Juno at (800) 654-JUNO [654-5866]
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|