Subject: [IBIS-Users] IBIS Model quality observation
From: Lieby David (DLieby@acuson.com)
Date: Fri Oct 18 2002 - 10:00:37 PDT
Dear IBIS user colleagues,
I have been looking closely at some IBIS models for ECL
with an eye to cooking up a model for a chip that has no
model available from the manufacturer. For this I am
making a perhaps erroneous assumption that the IO cells
for portions of the ECL families are designed somewhat
the same. I looked at the models for several devices in
each group, like EL and LVEL, and found that there were
small changes but not too much more than a few percent.
Some of the models had differences between the pins of
differential drivers and receivers. I was sort of puzzled
over this until I looked closely at the [Pin] listing and
noted the differences in the contribution of the connecting
wire/pin combination. The differences has some correlation.
When I reviewed what appeared to be the pin location VS the
numbers in the curve it looks like they measured one pin
and then another (not the associated diff pin) and used
them for the model. So when the IBIS file is used the
data for the driver is somewhat erroneous to start with
and then the pin data is folded in again. Each pin has
a different set of pin impedance values though there is
some symmetry. What I am wondering is if there is a way
to use negative values for the pin impedance to adjust
for the way the tables were generated. (I have not tried
this yet.)
While I realize that most of the investigations I do are
going to only be close in the real world because of
manufacturing tolerances, I would like to not have things
skewed to begin with.
Does anyone have a rule of thumb they use that could make
things like this a little less messy?
Best Regards,
dav0
David Lieby, SI Engineering
Acuson Corporation, a Siemens Company
1230 Shorebird Way
Mountain View, CA 94043
Building E
P.O. Box 7393
Mountain View, Ca 94039-7393
(650)694-5030
|------------------------------------------------------------------
|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 : Fri Oct 18 2002 - 10:11:58 PDT