RE: [IBIS-Users] damped oscillations in V-T curve


Subject: RE: [IBIS-Users] damped oscillations in V-T curve
From: erik.van.der.ven@philips.com
Date: Mon Jun 30 2003 - 00:37:31 PDT


__________________

Tom,

I am just using an ideal resistive load and I have no package model.
Herbert suggested that HSPICE does not accept two waveform tables because
this is not needed for ECL models according to the IBIS specification.
The ringing is likely caused by the internal feedback mechanism in the
PECL buffer.
Perhaps it is best just to truncate the falling waveform when it has
reached the zero level. But I don't know SI-engineers will miss relevant
information for their simulations. I wonder if simulators can handle
ringing caused by internal feedback anyway, since simulator have not
enough information to estimate the loop behaviour from just a data table.

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

"Tom Dagostino" <tom@teraspeed.com>
06/27/03 06:22 PM
Please respond to tom

 
        To: Erik van der Ven/NYM/SC/PHILIPS@EMEA2
        cc:
        Subject: RE: [IBIS-Users] damped oscillations in V-T curve
        Classification:

Eric
 
For the rest of us to help you you will need to be more specific about
what you are simulating. Is this a circuit with a transmission line in it
or just a buffer into a load R? What is your package model? What else
can you tell us?
 
Tom Dagostino
Teraspeed Consulting Group LLC Teraspeed Consulting Group LLC
2926 SE Yamhill St. Device Modeling Division
Portland, OR 97214 13610 SW Harness Lane
                                   Beaverton, OR 97008
http://www.teraspeed.com 503-430-1065
tom@teraspeed.com
-----Original Message-----
From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org]On Behalf Of erik.van.der.ven@philips.com
Sent: Friday, June 27, 2003 6:13 AM
To: ibis-users@eda.org
Subject: RE: [IBIS-Users] damped oscillations in V-T curve

__________________

Hello Herbert,

Thanks for your reaction. Iam using HSPICE2002.2.2. Indeed when I use 50
Ohm load (equal to Rfix) everything is OK. With 70 ohm load this peaking
occurs. I tried to use two wave forms in the IBIS model, one with Rfix=50
and one with Rfix=70. But SPICE gave an waring that only one curve is
used, I don't know why. Could this method help to get better results? I
don't know what a simulator exactly does with two waveform tables.

Are there more users who can comment on this issue?

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

herbert_lage@agilent.com
Sent by:
owner-ibis-users@eda.org
06/26/03 12:23 PM
        
        To: Erik van der Ven/NYM/SC/PHILIPS@EMEA2
ibis-users@eda.org
        cc:
        Subject: RE: [IBIS-Users] damped oscillations in V-T curve
        Classification:

Hi Eric,

I am not sure whether this will address your issue, but I noticed a while
back that HSpice's B-element when used for PECL buffers simulates best
when the actual load (in a system) does not deviate too much from the load
that was used to record the waveforms. This was certainly true for older
versions (2001.2), but may have been rectified in newer releases. It has
probably to do with the way HSpice interpolates between the "high" and
"low" state when switching PECL buffers. This is only an issue for PECL
buffers and not that much for other buffer types.

Another issue with buffer simulations is that HSpice does not like
"overclocking", that means a Vt transition needs to be finished before the
buffer is switched again.

As an aside, I tend to use "effective" loads so that eg. ac-coupled PECL
buffers are well behaved in the intended application. Please see example
below. This has worked well so far.

I hope this gives you some ideas where to look,

Best Regards,

                Herbert.

| **************************************************************
|
| The V_fixture and R_fixture values for the waveform tables
| are equivalent values which produce the right loadline for
| a PECL output with a 130 Ohm dc-bias resistor and an ac-coupled
| (50+15) Ohm load. The bit pattern is assumed to be balanced so
| that the ac-coupling capacitor is charged to (Vol + Voh)/2.
|
|
| 15 Ohms 50 Ohms
| | |
| PECL-Output -----------/\/\/\----| |----/\/\/\------
| | | | |
| | |
| \ |
| / |
| \ 130 Ohms |
| / |
| | |
| | |
| ----- -----
| --- ---
| - -
|
| **************************************************************
|
[Falling Waveform]
|
R_fixture = 43.33
V_fixture = 1.300
V_fixture_min = 1.147
V_fixture_max = 1.487
|
| time V(typ) V(min) V(max)
|
  0.000E+00s 2.316E+00V 1.985E+00V 2.633E+00V
  1.000E-11s 2.316E+00V 1.985E+00V 2.633E+00V
  2.000E-11s 2.316E+00V 1.986E+00V 2.633E+00V
  3.000E-11s 2.316E+00V 1.986E+00V 2.633E+00V
     . . . .
     . . . .
     . . . .

-----Original Message-----
From: erik.van.der.ven@philips.com [mailto:erik.van.der.ven@philips.com]
Sent: 26 June 2003 11:44
To: ibis-users@eda.org
Subject: [IBIS-Users] damped oscillations in V-T curve

Hello,

In my V-T curve I see a damped oscillation in the falling waveform (diff
PECL).
When I do a HSPICE simulation with different load resistances I observe
that the amplitude of the first peak increases dramatically while this is
not the case in a simulation of the original buffer in Spectre.
What should I do to model this behaviour correctly. And what causes this
increase in amplitude?

With kind regards,

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

|------------------------------------------------------------------
|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 : Mon Jun 30 2003 - 00:42:57 PDT