[SI-LIST] : Re: How to do correlation with two extra Rising/Falling waveforms?

Stephen Peters (sjpeters@ichips.intel.com)
Fri, 10 Apr 1998 14:25:53 -0700

Hello John:

My comments are below:

Regards,
Stephen Peters
Intel Corp.

> Hi ,
>
> In the IBIS Forum's I/O BUFFER MODELING COOKBOOK, page 12, there is a
> table - "Recommended Load Circuits and Waveforms for V/T Data
> Extraction."
>
> In the table, there are 4 waveforms to extract V/T data for [Rising
> Waveform] and [Falling Waveform] for a standard push/pull CMOS type
> buffer.
>
> I/O buffer+ R-fixture+V_fixture.
>
> For falling edge, two waveforms are obtained based on
> (1) V_fixture=0, and
> (2) V_fixture=VCC.
>
> For rising edge, two waveforms are gathered based on
> (3) V_fixture=0, and
> (4) V_fixture=VCC.
>
>
> The cookbook uses (2) and (3) to extract data in [Ramp] and says (1)
> and (4) are for correlation.

Yes, the 50 ohms to VCC and 50 ohms to gnd fixtures are used to extract
the data for the [Ramp] table, but the other two conditions are used to
provides further detail about the switching characteristics of the
buffer, so the simulator can build a better model. I don't belive the
cookbook says these waveforms are used only for corelation.

>
> How is the correlation done?
> What information can be extracted from (1) and (4)?
>
> Inside the Quad Design XTK manual, the (1) and (4) can be used to
> generate Ground
> Bounce Model. How?
>
> Any inputs from you are appreciated.
>
>
> Thanks,

>
> JOHNLIN
> CAE Engineer of EDA Department
> Digital Equipment Corp. Taiwan Branch
> Email: Linjohn@mail.dec.com
> TEL: 1-886-3-3900000 ext. 2152
>