From: Kellee Crisafulli
To: Steve and IBIS members
Re: Spice to IBIS
I think your progress is excellent!
It sounds like we need to help you get some spice models. I was hoping
that one of the semiconductor people would have given you their model
files by now. HyperLynx does not have any spice models that we can send
you without violating non-disclosure agreements.
(Note to Bob Ward): Is there any possiblity of getting models for Steve from TI?
Please.....
Relative to the problems with getting the models to converge. This is not
uncommon. I have used several vendors spice models and found numerous problems
with getting them to converge. If I use one of the commercial spice packages
as opposed to Berkley standard release I find the converge problems are reduced
significantly. Which spice package are you using? Also I have noticed that if
I stay close to the values that I receive the model set to that they work fine.
If I sweep the data too broadly it causes problems. Perhaps this should be
an input parameter to the converter.
Received on Thu Feb 10 09:37:58 1994
This archive was generated by hypermail 2.1.8 : Fri Jun 03 2011 - 09:52:28 PDT