Henrik wrote
>
> Thanks for your input.
> This is a 10GHz circuit and the RF guys have been pretty particular about
> the bevels. Now it would be interesting to know if beveling is just
another
> superstition. If rounded corners will work just as well or better, not
> introducing impedance transitions in the line, that would indeed make this
> task easier. Could your field solver answer that question?
>
Well that's not quite all of it. Because gerbers dint draw round corners,
your output becomes a series of little crummy looking bitmapps. The
superstition you are really dealing with is a wider corner trace measured
from the inside corner to outside corner as a trace makes a bend. An
example: a 5 mil trace with a 90 degree angle bend has about 7 mils, or
roughly a 40 percent increase in capacitance at the corner. The same 5 mil
trace at 45 degrees has a corner bend or aprrox 5.4 mils. So you have
to ask your self, I have .5 oz copper, what is my etchback, what does it
do at the corners? ( they tend to etch back even more at the corners )
Is the .4 mils critical? Should not be. If your design relies on less
than .4 mil of corner sticking out at a few corners, I would consider
another occupation, Thats the reallity of it .0004 inch sticking out for
aprox 1 mil. Scale to a 50 mil trace and you are only talking about 4 mils,
again less than the width of a hair, well I guess that might start to get a
little squirrelly at super high frequencies.
Mike Reagan
EDSI
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/subscrib.html
* - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/[email protected]
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
