Subject: RE: [IBIS-Users] IBIS issue at high frequencies
From: Todd Westerhoff (twesterh@cisco.com)
Date: Thu Aug 29 2002 - 14:02:43 PDT
If you attempt to switch the buffer before the end of the V-T curve sweep,
then you have an abrupt change in the buffer's output state. The buffer
jumps from the point on the transient curve that was switching to the start
point on the curve you are transitioning to.
You'll see the same problem at any frequency if the DC settling points of
the rise and fall curves are different. If your rise curve settles at 2.5V
(as an example), and the corresponding fall curve starts at 2.3V, then there
will be an abrupt change (and "spike")at the start of the falling edge.
Bottom line: Herbert is right - attempting to switch an IBIS buffer before
it finishes the V-T sweep can cause weird problems. Because each
simulator's IBIS implementation is a bit different, the problems you will
see can vary.
IBIS models weren't meant to model inertial delays, and they don't ...
Todd.
Todd Westerhoff
Signal Integrity Engineer
Cisco Systems
250 Apollo Drive - Chelmsford, MA - 01824
email:twesterh@cisco.com
ph: 978-497-0272
============================================
"When did the choices get so hard, with so much more at stake?
Life gets mighty precious when there's less of it to waste"
- Bonnie Raitt, "Nick of Time"
-----Original Message-----
From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org]On
Behalf Of herbert_lage@agilent.com
Sent: Thursday, August 29, 2002 6:47 AM
To: ibis-users@eda.org; ibis@eda.org
Subject: RE: [IBIS-Users] IBIS issue at high frequencies
Hi Nirav,
I am not quite sure that you have the same problem as I had with Avanti
HSpice a little while ago. I encountered funny effects when a buffer was
switched before the end of a waveform was reached, e.g. if a waveform was
1.2ns long but I swiched/toggled the buffer again after less than 1.2ns. I
got spikes in the transient analysis. These went away when I toggled the
buffer at (1.2+ x)ns.
When I create IBIS models I now make sure that the waveforms are shorter
than the fastest bit period the part is intended to run at. I do that by
cutting the lead-in time (in a consistent manner for all corners). Below is
an excerpt of a conversation that I had with an IBIS model user:
> I believe that the downward spike has to do with the way
> HSpice interprets the IBIS model.
>
> The downspike only occurs when the bit pattern is very short
> and the simulator jumps straight from the falling waveform of
> the IBIS file to the rising waveform. Why this creates a spike
> is a mystery to me at the moment, as the curves nicely match up.
> It does not occur in the Ones either, only in the Zeros.
>
> I was able to fix it by shortening the falling and rising
> waveforms so that the simulator is forced to use the static
> pullup/pulldown curves before it invokes another waveform.
This effect may well depend on the version of HSpice.
I hope this helps,
Best Regards,
Herbert.
-----Original Message-----
From: Nirav Patel [mailto:nirav@agere.com]
Sent: 29 August 2002 07:05
To: ibis-users@eda.org; ibis@eda.org
Subject: [IBIS-Users] IBIS issue at high frequencies
IBIS gurus,
We are using Avanti HSPICE to correlate the IBIS
model performance to spice. We notice that the model
starts misbehaving abruptly as we increase the toggle
rate.
Our observation has been that the onset of this
misbehavior is a function of the delay through the
buffer. Buffers that have long propagation delays
tend to misbehave at lower toggle rates.
Does anybody have a clue on what may be the cause
of this issue? Any help will be very much appreciated.
Thanks in advance.
--Nirav
|------------------------------------------------------------------
|For help or to subscribe/unsubscribe, email majordomo@eda.org
|with just the appropriate command message(s) in the body:
|
| help
| subscribe ibis <optional e-mail address, if different>
| subscribe ibis-users <optional e-mail address, if different>
| unsubscribe ibis <optional e-mail address, if different>
| unsubscribe ibis-users <optional e-mail address, if different>
|
|or email a written request to ibis-request@eda.org.
|
|IBIS reflector archives exist under:
|
| http://www.eda.org/pub/ibis/email_archive/ Recent
| http://www.eda.org/pub/ibis/users_archive/ Recent
| http://www.eda.org/pub/ibis/email/ E-mail since 1993
|------------------------------------------------------------------
|For help or to subscribe/unsubscribe, email majordomo@eda.org
|with just the appropriate command message(s) in the body:
|
| help
| subscribe ibis <optional e-mail address, if different>
| subscribe ibis-users <optional e-mail address, if different>
| unsubscribe ibis <optional e-mail address, if different>
| unsubscribe ibis-users <optional e-mail address, if different>
|
|or email a written request to ibis-request@eda.org.
|
|IBIS reflector archives exist under:
|
| http://www.eda.org/pub/ibis/email_archive/ Recent
| http://www.eda.org/pub/ibis/users_archive/ Recent
| http://www.eda.org/pub/ibis/email/ E-mail since 1993
|------------------------------------------------------------------
|For help or to subscribe/unsubscribe, email majordomo@eda.org
|with just the appropriate command message(s) in the body:
|
| help
| subscribe ibis <optional e-mail address, if different>
| subscribe ibis-users <optional e-mail address, if different>
| unsubscribe ibis <optional e-mail address, if different>
| unsubscribe ibis-users <optional e-mail address, if different>
|
|or email a written request to ibis-request@eda.org.
|
|IBIS reflector archives exist under:
|
| http://www.eda.org/pub/ibis/email_archive/ Recent
| http://www.eda.org/pub/ibis/users_archive/ Recent
| http://www.eda.org/pub/ibis/email/ E-mail since 1993
This archive was generated by hypermail 2b28 : Thu Aug 29 2002 - 14:21:24 PDT