Re: [ng-spice-devel] Another bug in the source 5_2


To ng-spice-devel@ieee.ing.uniroma1.it
From Anthony Wilson <awilson@mrcmicroe.com>
Date Mon, 24 Jan 2000 09:40:19 -0700
Delivered-To mailing list ng-spice-devel@ieee.ing.uniroma1.it
In-Reply-To <Pine.LNX.3.96.1000119082039.18465A-100000@ieee.ing.uniroma1.it >
Mailing-List contact ng-spice-devel-help@ieee.ing.uniroma1.it; run by ezmlm
Reply-To ng-spice-devel@ieee.ing.uniroma1.it

At 08:24 AM 1/19/2000 +0100, you wrote:
>Hi,
>
>There is another or probably more than one bug in the source 5_2. If you
>try to use level 49 model (BSIM3 3v3.1) you will soon discover that test
>file from berkeley (http://www-devices.eecs.berkeley.edu/~bsim3) bsim 3
>site fails. One of the test I run did not showed i(vds), filling that
>vector with zeroes. I am going to investigate, if level 50 suffers of this
>too.
>
>Wilson, as you ported devices on macspice, do you have some hints or
>ideas ?

No, I have not experienced this before.  but you might want to make sure 
that the device data structure is getting initialized correctly.  When I 
was trying to get the various bsimsoi models implemented, there was a 
problem in defining the device structure and having the appropriate model 
call the correct soi structure(pd vs dd vs fd).  this may just be my 
ineptness as a programmer.

Couple of things to look at:
1. are you sure that in implementing the level 49 model that it is being 
called when running the spice deck.
2. In setting the up the device and the inp2m.c file are you sure that you 
are accessing the level 49 model or is it seeing the default?

you may have to step into the device structure to make sure or put a debug 
statement in the device structure to print out that it has been initialized 
correctly.

Have you run any other spice decks?  What was the outcome?  Have you ran 
this spice deck using the level 8 model, what happened?

let me know.
-anthony.

Partial thread listing: