Subject: RE: [IBIS-Users] load capacitor
From: Angulo, John (john_angulo@mentorg.com)
Date: Wed Aug 27 2003 - 11:29:25 PDT
Erik,
The reasons for a difference in LVDS results fall into two general categories. First, because IBIS was not originally intended for LVDS technology (LVDS wasn't yet in general use), IBIS models of LVDS devices usually are accurate only for the specific load used in model extraction. Second, there are several ways in which simulations in the two tools may be set up differently. The Hyperlynx simulation may include some parasitic characteristics for the capacitor leads. Or, there could be a difference in the device package modeling between the two tools. Or, a difference in the simulator time step settings may influence the effects of parasitics. The only way to validate the IBIS model and have confidence in the simulation is to see the model, the exact load, and to run a set of test simulations. Customer support will contact you offline to do this.
Best regards,
John Angulo
Hyperlynx Products
Mentor Graphics Corp.
john_angulo@mentor.com
425-869-2320
-----Original Message-----
From: erik.van.der.ven@philips.com [mailto:erik.van.der.ven@philips.com]
Sent: Tuesday, August 26, 2003 11:48 PM
To: ibis-users@eda.org
Subject: [IBIS-Users] load capacitor
Hello IBIS users,
I have made a model of a differential LVDS output buffer. When I simulate my IBIS model with a 100 Ohm load resistor, differentially connected, the results are very close to my circuit simulations with Spectre. The V-T curves and I-V match very well, I do not get any warnings from the IBIS parser. When I connect 2pF load capacitors between the outputs and ground the differnce with the Spectre simulation becomes very big. The output signals in Spectre look very well and smooth. THe edges are less steep and look perfect. When I simulate the IBIS model I see significant ringing. The falling edge drop till 2/3 of the amplitude and then there is horizontal step (looks a bit like an overshoot or ringing) and then it continues falling again. The relative maximum error (calculated with Ack) is about 30% with respect to the amplitude.
How can this big difference be explained? What can I do to improve my model, if possible?
I don't know if it is relevant but I did my IBIS simulation with Hyperlynx.
With kind regards,
Erik.
Ir. Erik van der Ven
Room DB1032
Business Line Networking Infrastructure
Philips Semiconductors BV
Gerstweg 2
6534 AE Nijmegen
The Netherlands
Phone: +31-24-3534334
|------------------------------------------------------------------
|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 : Wed Aug 27 2003 - 11:32:19 PDT