RE: [IBIS-Users] ODT topics

From: Todd Westerhoff \(twesterh\) <twesterh_at_.....>
Date: Mon Apr 04 2005 - 08:47:50 PDT
Radovan,

I don't remember that I've seen many replies from vendors in response to
your original question.

So ... I offer these words of advice.

Many models end clamp curves at 0V, because they assume clamp current will
be zero at that point.  As you have already pointed out, input termination
doesn't work that way, so the current at 0V is non-zero.

Because 0V in the clamp curves corresponds to the supply voltage, and
positive voltage extends out beyond the supply - a clamp curve that ends at
0V will have to extrapolate inside of the device's actual operating region.
If the current ans slope at 0V are zero, the extrapolation is trivial.
That's not the case for input termination, and we've found that different
simulators use different techniques for extrapolating data off end of the
curve.  Some tools extarpolate using the slope at the endpoint; others
simply extend the last value of the curve out flat (i.e. slope = 0).  

I've never seen a thorough listing of which tool does what - our knowledge
is based on empirical observation.

My recommendation - define your clamp curves completely - from -2*VDDQ to
VDDQ - so the simulator never needs to extrapolate the values.

Todd.

Todd Westerhoff
High Speed Design Group Manager
Cisco Systems
1414 Massachusetts Ave - Boxboro, MA - 01719
email:twesterh@cisco.com
ph: 978-936-2149
============================================

"Always do right.
 This will gratify some people and astonish the rest."

- Mark Twain


-----Original Message-----
From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf
Of Radovan.Vuletic@infineon.com
Sent: Friday, April 01, 2005 9:50 AM
To: ibis-users@eda.org
Subject: [IBIS-Users] ODT topics

Hi,

recently I have took responsibility for IBIS models in Memory Products of
Infineon and soon I have realised that it is very tricky. 

There were generated some ODT IBIS models (for DDR2  with thevenin
impedance, not possible to separate Pull-up and Pull -down impedance) with
s2ibis2 as Inputs with Power and Ground Clamps curves. Everything was OK
until first customer were complaining that during his simulations our ODT
behave other than comparing to Samsung and Micron ODT's. After some 5-6
hours of frustrating analysis we have discovered that the "only" problem is
that neither the clamp curves are cut where they reach zero nor they are
extrapolated  horizontally to 2*Vcc. As soon as we done that (cutting at
zero current and extrapolating to 2*Vcc) everything was OK. By the way, Mr.
Muranyi has nicely explained this procedure in his paper "Modeling on-die
terminations in IBIS (without double counting)". The problem that we have
experienced is obviously because of double counting of some data.

My question would be: what are simulators (e.g. HSPICE) really doing with
IBIS models (specially with Input models), how are data treated, what is the
algorithm of calculation (if designer was aware of this algorithm double
counting could be perhaps avoided)? Is it possible to get this things
written in electronic form or anywhere on Internet? 

By the way, on IBIS summit Meeting in Munich in March (DATE 2005) somebody
mentioned something about some Cookbook for ODT's. Is there any estimation
when will this Cookbook appear?

Many thanks in advance for any help or advice, Radovan Vuletic

Infineon Technologies AG
MP PD PDE
Room 03.911
Balanstraße 73
D-81541 München

Phone:		+49 (0)89 234 20108
Fax:		+49 (0)89 234 27705
Fax (PC):	+49 (0)89 234 955 5305 

E-mail: radovan.vuletic@infineon.com

|------------------------------------------------------------------
|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
Received on Mon Apr 4 08:48:19 2005

This archive was generated by hypermail 2.1.8 : Mon Apr 04 2005 - 08:52:37 PDT