Re: [IBIS-Users] PullUp/PullDown toggle?

From: al davis <ad105_at_.....>
Date: Tue Dec 15 2009 - 20:41:19 PST
On Tuesday 15 December 2009, Lars Snith wrote:
>    2.40    -12.7710mA        -7.5000mA           -21.3261mA
>    2.46    -12.8090mA        -8.0000mA           -21.3770mA
>    2.52    -12.8400mA        -7.0000mA           -21.4280mA
>    2.58    -12.9000mA        -7.0000mA           -21.4810mA
>    2.64    -13.0000mA        -7.0000mA           -21.5310mA
>    2.70    -13.0000mA        -8.0000mA           -21.5800mA
>    2.76    -13.0000mA        -8.0000mA           -21.6300mA
>    2.82    -13.0000mA        -10.0000mA          -22.0000mA
>    2.88    -13.0000mA        0.0A                -22.0000mA
>    2.94    -10.0000mA        -10.0000mA          -22.0000mA
>    3.00    -10.0000mA        -10.0000mA          -22.0000mA
>    3.06    -20.0000mA        -10.0000mA          -20.0000mA
>    3.12    -20.0000mA        0.0A                -20.0000mA
>    3.18    -10.0000mA        -10.0000mA          -30.0000mA
>    3.24    -10.0000mA        -10.0000mA          -20.0000mA
>    3.30    -10.0000mA        -10.0000mA          -20.0000mA
>    3.36    -10.0000mA        -10.0000mA          -30.0000mA
>    3.42    -20.0000mA        0.0A                -20.0000mA
>    3.48    -20.0000mA        -10.0000mA          -20.0000mA
>    3.54    -10.0000mA        -10.0000mA          -30.0000mA
>    3.60    -10.0000mA        -10.0000mA          -20.0000mA

It's not toggle.  It's numerical noise.  It's not just noise 
buried under the numbers, those numbers are nothing but 
numerical noise.

You get things like this from the subtraction that separates the 
clamps from the composite.  The clamp current in this region is 
usually huge, and exists only as a figment of the simulator.  A 
real device would let the smoke out and not work any more.

You need more precision in currents, and less precision in 
voltage. 

It may cause convergence problems seemingly at random.  Some 
have argued that only the sum of the clamps and pullup/down 
matters, but you can still get glitches that cause trouble.  
Usually, it doesn't matter because it never operates in this 
region, where clamp diodes are forward biased.  Once in a while, 
a non-converged iteration will land there and throw the process 
off, leading to non-convergence, or more likely just a few more 
iterations.

This problem has existed for years.  Most dedicated SI 
simulators will filter out bad data like this, without even a 
warning.



-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

--------------------------------------------------------------------
|For help or to subscribe/unsubscribe, e-mail majordomo@eda-stds.org
|with 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 e-mail a request to ibis-request@eda-stds.org.
|
|IBIS reflector archives exist under:
|
|  http://www.eda-stds.org/pub/ibis/email_archive/ Recent
|  http://www.eda-stds.org/pub/ibis/users_archive/ Recent
|  http://www.eda-stds.org/pub/ibis/email/         E-mail since 1993
Received on Tue Dec 15 20:39:46 2009

This archive was generated by hypermail 2.1.8 : Tue Dec 15 2009 - 20:40:45 PST