Re: IBIS MOdel Question

tomda wrote:
>
> There is one sanity check that I would like to see done on all models
> produced from SPICE data. I have seen too many models that are constructed
> correctly fail this check.
>
> Check to make sure the IV characteristics derived from SPICE match the IV
> characteristics of real parts. The easiest was to do this is to put one of
> the buffers from each strength into both its high and low states. Us a DC
> current meter and measure the output current to ground for the high state
> and to Vcc for the low state. Then check to see if the measured current
> falls between the min and max values at the corresponding points on the IV
> curves.

Maybe I'm just a hard case, but why should this be a one-time exercise?
IBIS is perfectly suited to the job of specifying I/O characteristics to
manufacturing test. They have to do drive-strength tests anyway to make
sure that there aren't any open output device legs, so instead of silly
stuff like "8.0 mA at 330 mV" or whatever just give them the IBIS model
and let them pick the best test points.

> -----Original Message-----
> From: Kim Helliwell [SMTP:khelliwe@acuson.com]
> Sent: Wednesday, October 20, 1999 11:20 AM
> To: Tony Palmieri
> Cc: ibis-users@eda.org
> Subject: Re: IBIS MOdel Question
>
> Make sure that:
>
> 1. The curves are continuous
> 2. The ramps (if any) cover the whole switching range
> 3. All pins are assigned to models
>
> These are the three worst problems I've run into trying
> to use IBIS models supplied by well-known parts vendors.
>
> If you have access to one of the commercially available
> signal integrity tools, try actually using the model
> with it. Your customers will thank you if you actually
> make your model work in the simulators they are going
> to use!
>
> Kim Helliwell
>
> Tony Palmieri wrote:
> >
> > Hello,
> >
> > I have a question about an IBIS model. I have created a chip level model
> > that is composed numerous IBIS models. After I completed the model I
> > ran the ibischk program and it passed. I will be providing this model
> > to our customers, but want to make sure that it is filly tested.
> >
> > Are there any other tests that I should run to insure that it is
> > correct. I have reviewed the pin assignments, and that everything is
> > properly referenced.
> >
> > Please reply to apalmieri@amcc.com
> >
> > Thanks for your help,
> >
> > Tony Palmieri
> >
> > AMCC
> > Applications, IC Test and Verification Manager
> >
> > --------------------------------------------------------------------
> ----
> >
> > Applications Manager
> > AMCC
> >
> > Applications Manager
> <apalmieri@amcc.com>
> > AMCC
> > 200 Brickstone Square ;Andover;Massachusetts;01810; Fax: 978-623-0024
> > Work: 978-623-0009
> x3063
> > Additional Information:
> > Last Name Palmieri
> > First Name Anthony
> > Version 2.1
>
> --
> Kim Helliwell
> Senior CAE Engineer
> Acuson Corporation
> Phone: 650 694 5030 FAX: 650 943 7260

-- 
D. C. Sessions
dc.sessions@vlsi.com