Subject: [IBIS-Users] diode modeling in ibis vs other modeling issues
From: Scott McMorrow (scott@teraspeed.com)
Date: Wed Jul 10 2002 - 19:06:24 PDT
All,
About once every 3 months someone comes up and complains that in many
spice decks the clamp diodes are not correctly modeled. At that point,
others jump onto the pile and add their comments which generally amount
to "how could anyone put out such an obiviously bad model? Look, even
the ibis checker can find it."
The sad truth is that about 70% of all IBIS models available on the web
have serious flaws and about half of those are so serious as to render
the models worthless. However, this is not the total story. Those are
the models which could be checked by the ibis checker and validated as
having errors. IBIS models suffer from an extreme flaw in the modeling
of packages. There is only one method to model package coupling. No
one that I know of has created a model that uses it. 95% or more of all
IBIS models have fatal flaws in the modeling of the package. Many are
only a guess and not supported by any measurement or simulation data
whatsoever.
Unfortunately, no "checker" in the world can check whether a package has
been correctly modeled. We in the signal integrity world go to great
lengths to characterize the crosstalk and coupling that occurs on a
printed circuit board. Yet no one stands up to complain that there is
virtually no data on this available within the ibis model, and the
package has the greatest routing density of any object in an electronic
system (except for the silicon itself ... which is yet another issue.)
Those of us that have looked at this extensively have concluded that the
greatest single impediment to good signal integrity on a board is, in
fact, the package. Trust me, there is a legacy of failed designs out
there where the package is the culprit. Designs fail because they were
simulated with IBIS simulators, using a perfect ground reference and
flawed package modeling. The modeling of clamp diodes is the least of
your concerns ... but it is an easy target.
regards,
Scott
-- Scott McMorrow Teraspeed Consulting Group LLC 2926 SE Yamhill St. Portland, OR 97214 (503) 239-5536 http://www.teraspeed.com|------------------------------------------------------------------ |For help or to subscribe/unsubscribe, email majordomo@eda.org |with just the appropriate command message(s) in the body: | | help | subscribe ibis <optional e-mail address, if different> | subscribe ibis-users <optional e-mail address, if different> | unsubscribe ibis <optional e-mail address, if different> | unsubscribe ibis-users <optional e-mail address, if different> | |or email a written request to ibis-request@eda.org. | |IBIS reflector archives exist under: | | http://www.eda.org/pub/ibis/email_archive/ Recent | http://www.eda.org/pub/ibis/users_archive/ Recent | http://www.eda.org/pub/ibis/email/ E-mail since 1993
This archive was generated by hypermail 2b28 : Wed Jul 10 2002 - 19:20:32 PDT