Re: min and max rising/falling waveforms in s2ibis2

From: Al Davis <aldavis@ieee.org>
Date: Sat Nov 18 2000 - 01:31:07 PST

On Fri, 17 Nov 2000, Betty Luk wrote:
> I am using s2ibis2 to generate my IBIS model. I have set the
> [Voltage Range] to be 3.3 3.0 3.6 (typ, min, max). The IBIS
> cookbook suggests that a 50 ohm resistive load Vdd be used for the
> rising and falling waveform VT data. In the s2ibis2-generated
> Spice files, the 50 Ohm load always connected to 3.3V, even for the
> min and max case.
>
> Is this the correct behaviour? And if so, why are 3.0 and 3.6V
> (for min and max respectively) not used?

The voltage the test load is connected to in the waveform tables has
nothing to do with the voltages the device operates on. You could
connect it to some middle voltage if you want, which would be a
better choice when the device is used differentially.

You should supply at least two waveforms of each type. Normally, one
would have a pullup load, and the other would have a pulldown load.
By supplying only one of each, you do not adequately characterize
both devices. A load to ground characterizes the pullup device
fairly well, but does nothing for the pulldown device. Likewise, a
load to Vcc characterizes the pulldown device fairly well, and
supplies almost no information about the pullup device.
 
Received on Sat Nov 18 01:33:47 2000

This archive was generated by hypermail 2.1.8 : Fri Jun 03 2011 - 09:53:47 PDT