RE: [SI-LIST] : Problem in IBIS simulation

About this list Date view Thread view Subject view Author view

From: Ingraham, Andrew (Andrew.Ingraham@compaq.com)
Date: Sat Apr 21 2001 - 19:22:09 PDT


> I am simulating an ibis file using HSPICE. I will be thankful to you if
> anyone of you can help me in the following queries:
>
> (a) While simulating an ibis buffer model of I/O type in the input mode,
> i.e. when the signal from external world appears on the pad, the high
> level output of input buffer is always 1V regardless of any other factor.
> Why is it so ?
 
Short answer: Because that's how they (Avant!) made their IBIS models work,
in Star-HSPICE.

As Michael explained, IBIS models don't represent the entire buffer. For
inputs, they represent only what the input looks like as seen from the
outside world: its nonlinear input impedance. Nothing beyond that is
represented. The purpose is to let you simulate the interconnections UP TO
the input pin, but not through the input buffer.

IBIS "models" (data sheets) also include information about the required
logic levels at that input pin, which the simulator can use to tell you
whether inputs would be interpreted as "low" or "high". HSPICE uses this
information to give you a binary signal that tells you whether the input
voltage is low (0) or high (1). This binary signal is either 0V or 1V.

> (b) Why if while using the I/O buffer in input mode I have to specify the
> input for the output buffer, which is of no use to me, otherwise it gives
> an error that no voltage sources are defined.
 
Again, that's how they (Avant!) made their IBIS models work; although in
general, all SPICE programs have a requirement that nothing should be left
floating. The input for the output buffer is a binary input signal (0V or
1V). If you don't connect it to something, HSPICE has no idea what you
wanted it to be. It probably does this checking once when the program is
starting, and it doesn't know at that time that you will only use the buffer
in input mode.

Regards,
Andy

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Jun 21 2001 - 10:11:41 PDT