All the prior posts are true. Some perspective sounds in
order. Perhaps I can help.
FR4 is a compromise material, and frankly the resin con-
tent varies within a single sheet, never mind across an
There is also variation in process methods; different
etching technologies produce different cross-sectional
shapes - from trapeziod to (yes) rectangular.
I've lost the beginnings of this thread but it sounded
like the problem was one that I call "controlled stackup"
used primarily to be certain the impedance doesn't vary
within a board. Moving this level of specification from
vendor to vendor will produce surprisingly different im-
pedance results. Some design folks will try to use this
technique to buy a "cheaper" impedance controlled board.
More correct to call it "impedance contained".
Next striation I make is between controlled impedance and
etched components. For digital domain (up to a point)
simply targeting the impedance and specifying the test
method is adequate. In this realm it is only appropriate
to verify that your intended stackup is robust and provides
a reasonable window for any fabricator to reach intended
performance. Up to and including differential impedance.
When reaching the limits of that technique, or when you
seek explicit sub-parameters, you must specify those
parameters - impedance alone is not enough. The fabrica-
tion community will reveal its capable vendors when you
reach into the specifics. I'd rather guess that we're
not hearing from the SI engineers on-staff at those ven-
dors in this thread because of current trade shows.
Correction of pad geometries for etch factors versus art-
work usually falls in the manufacturing domain. Done all
the time, at least at a certain tier of supplier.
Obviously, your mileage has varied and will undoubtedly
continue to do so.
Hoping the above improves it an MPG or so,
-- Jeff Seeger Applied CAD Knowledge Inc Chief Technical Officer Tyngsboro, MA 01879 jseeger "at" appliedcad "dot" com 978 649 9800
**** To unsubscribe from si-list: send e-mail to email@example.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****