Re: Minimum or Maximum package for SI Worst Case simulation?

Hi John,
At 08:28 AM 8/7/98 +0800, John Lin - TAO wrote:
>When doing the worst case simulation of signal integrated for a design,
>how shall I use the IBIS models? Here, the worst case means the maximum
>amount for overshoot, undershoot and ringback.
>I know that I shall use the IBIS models with Max V-I table and Minimum
>C_comp.
Actually this is a valid assumption only for outputs, for inputs you
should use max C_comp and minimum V-I for the clamps to get the
biggest reflection.

>I am not sure if use minimum package(R,L,C) or maximum package(R,L,C)
>information.
Min/max pin information should only be used if the pin specific information
is not provided. The pin specific informaiton has not min/max data.
If no pin specific information is provided than use the R,L,C min to get the
largest signal launched (i.e. most energy for reflections). Use the R,L,C
maximum for the worst step into the line and most local ringing at the device.

>One article says the minimum package information shall be used to do the
>worst case analysis. (the article is on
>http://www.eia.org/eig/ibis/intel.htm
><http://www.eia.org/eig/ibis/intel.htm> )
>
>Can somebody explain why?
>
>
>
> Thanks,
>
>JOHNLIN
>
>
>

-------------------------------------------------------------------------
Have a great day...
Kellee Crisafulli at HyperLynx
kellee@hyperlynx.com http://www.hyperlynx.com
-------------------------------------------------------------------------
Received on Thu Aug 6 19:21:19 1998