[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: RF breakdown II



Original poster: "David Sharpe by way of Terry Fritz <teslalist-at-qwest-dot-net>" <sccr4us-at-erols-dot-com>

Jim

Below is some additional information and dialog on this subject.  This
info is based on PCB GMP guidelines, and is based on 60Hz application
and 2.45GHz in a domestic microwave oven.  It may be comparing
apples to oranges and getting watermelons, but there appears to
be some consistency in the correlation observed here, regardless of
frequency.

An industry standard for PCB trace spacing for HV with traces on
outer layers of PCB with conformal coat (I used acrylic) is 0.00012" / V.
These dimensions are based on DC or PEAK voltage applied differentially
between adjacent traces, and sea level application.  These PCB guidelines
are widely available on the WWW (use Google).

The magnetron ceramic insulator for 5kV peak e-field for antenna is
0.38" long (0.000076" / V or ~0.63X recommended PCB clearance
distance for 5kV).  I can understand this to some degree, ceramic is a
much better HF / HV dielectric then FR4/G10 substrate.  Also antenna
structure of magnetron has been through FEA at least a bizillion times,
so e-field stress (with magnetic interaction) for a mature technology
should hold few surprises.

My traces on my OHVM divider (HV input to ground plane) is ~ 1.2".
Based on PCB design rule, PCB with 1.2" should be good for ~10kV
peak each side to ground (20kV peak or ~14kV rms across divider).
Based on magnetron design standard of 0.38" / 5kV, 1.2" clearance
should be good for ~15.8kV peak.  My board started to exhibit corona
-at- ~16kV rms (22.6kVpk).  So in air, the board would work fine with
my present design (14.4kV rms input), by mounting under mineral oil,
assembly should be good for intermittent duty to 40kV peak with no
problems.

Notice that design values above are significantly greater then corona
inception lengths, which would eventually lead to dielectric failure.

FWIW
Regards
Dave Sharpe, TCBOR
Chesterfield, VA. USA

Tesla list wrote:

 > Original poster: "Jim Lux by way of Terry Fritz <teslalist-at-qwest-dot-net>" 
<jimlux-at-earthlink-dot-net>
 >
 > I'm looking for published data or papers on RF breakdown of micro
 > striplines (or, for that matter, any transmission line which uses a flat
 > conductor stuck down on a flat dielectric). Creeping discharges are of
 > particular interest.
 >
 > I've got some empirical data from folks doing magnifiers, but I'm looking
 > for other frequencies and geometries.  Multiple frequencies on the same
 > setup is nice so that I can extrapolate with some semblance of validity.