RE: [SI-LIST] : **error**: internal timestep too small

About this list Date view Thread view Subject view Author view

From: Chan, Michael ([email protected])
Date: Fri Jan 21 2000 - 08:15:02 PST


I agree to Arpad in this case especially when you try to run
simulation using W-element for memory bus. Some trace lengths
on the memory card can be quite short and if you use W-element
to represent these traces, the simulation will run very long.
In several cases spurious oscillation shows up on the waveforms
as well. When those short traces are replaced with lumped elements,
then the simulation time is much shorter. These simulations are
performed at least two years ago when W-element just came out. I
have so fare haven't re-run under newer version of Hspice. The
problem might have been solved. Any new ??

Regards,
Michael Chan

-----Original Message-----
From: Muranyi, Arpad [mailto:[email protected]]
Sent: Friday, January 21, 2000 9:57 AM
To: '[email protected]'
Subject: RE: [SI-LIST] : **error**: internal timestep too small

Chris,

There may be situations when you may want to use U-elements.
If the length of the W-element is very short, simulation
time step can decrease, and simulation time can go up
significantly. This will not happen with the U-elements.

Using U-elements in such cases is equivalent to using lumped
circuits instead of transmission lines, and that is perfectly
legitimate if your rise times are slow compared with the time
delay across the transmission line.

It would be nice to have a feature (which I saw in some tools)
that would automatically switch to lumped (or U-elements) when
the transmission line length is electrically short.

Arpad Muranyi
Intel Corporation
==============================================================

-----Original Message-----
From: [email protected] [mailto:[email protected]]
Sent: Friday, January 21, 2000 7:00 AM
To: [email protected]
Subject: Re: [SI-LIST] : **error**: internal timestep too small

Kim,

I'm curious why Avanti's abandonment of the U model would bother you. Is
there a situation where the use of the U element is an advantage over the W
element (in terms of accuracy, runtime, ???) I guess I hadn't thought
about it too much but the W element seems to give accurate results with
reasonable runtimes for all of the cases I've investigated.

Chris Simon
General Dynamics Information Systems

**** To unsubscribe from si-list: send e-mail to
[email protected]. In the BODY of message put: UNSUBSCRIBE
si-list, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****

**** To unsubscribe from si-list: send e-mail to [email protected]. In the BODY of message put: UNSUBSCRIBE si-list, 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 Apr 20 2000 - 11:34:47 PDT