The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Temp in Corners analysis
https://designers-guide.org/forum/YaBB.pl?num=1179368294

Message started by ywguo on May 16^th, 2007, 7:18pm

Title: Temp in Corners analysis
Post by ywguo on May 16^th, 2007, 7:18pm

Hi, Guys,

I use Corners Analysis in ADE. I fill in 25, -40, and 125 for typ, fast, and slow corners respectively. One more parameter named supply is defined for supply voltage variation. That is 1.8, 1.9, 1.7 for typ, fast, and slow corners respectively. The corners analysis works fine.

However, I do not find temp parameter in altergroup when I take a look at input.scs. In my input.scs file, the altergroup is shown like the following.

...
fast altergroup {
parameters supply=1.9
include \
".............model.scs" \
...

Which file is the parameter temp defined in when I run a corners analysis in ADE?

Thank you

Yawei

Title: Re: Temp in Corners analysis
Post by kiran123 on May 18^th, 2007, 6:53am

hi yawei,

If your question is on how to include temperature variations in the input.scs file please have a look at my posting subject name : Need help on PTV mebedded single deck

-

Title: Re: Temp in Corners analysis
Post by Andrew Beckett on Jun 4^th, 2007, 1:17pm

Yawei,

The input.scs file that you get when you use the corners tool is not exactly how it would be done if you run spectre standalone. This is because the corners tool talks to spectre interactively, and tells it to:

use a named altergroup
set the temperature
run the named analyses
use the next names altergroup
set the next temperature
run the named analyses again

Normally you'd repeat the analyses between each altergroup, and set the temperature directly - but the interactive MPS (Message Passing Subsystem) (effectively a form of Remote Procedure Call) allows the corners tool (or ADE) to control spectre directly.

It's somewhat unfortunate that it's done this way, because it makes life a little hard for us application engineers to debug corners problems in the simulator without a fairly complete testcase - although there are ways and means ;) . However, despite the fact that the netlist doesn't show the temperature changes, it should actually change in practice!

Best Regards,

Andrew.

Title: Re: Temp in Corners analysis
Post by ywguo on Jun 4^th, 2007, 8:18pm

Hi, Andrew,

That is life. :) Thank you.

Yawei