Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Matlab/Simulink Cadence Cosimulation Problem

223 views
Skip to first unread message

Debjit

unread,
Mar 24, 2010, 1:41:35 PM3/24/10
to
Dear All,

I am trying to do some Matlab Cadence co simulation following the AVUM
Workshop included in Cadence Distribution.
My circuit is a very simple CMOS inverter. I am providing the Pulse
Input to the inverter from Matlab/Simulink though the coupler and
feeding back the output of the inverter to Matlab/Simulink by the
coupler.
I am noticing that whwever the square pulse is going through the
coupler from Matlab to Cadence, it is getting converted into
Triangular Pulse. And hence my circuit is not functioning properly.
How can I make it work properly?

I am using the following:

Matlab Version: R2009a
icfb version: 5.10.41.500.6.14
spectre version: 7.2.0.174.isr3
ncsim version: 08.20-s022
I am using Fixed Cell Coupler. I also tried Simulink Coupler from
Cadence analogLib. But failed in bot the cases.
I am using the option "start Matlab before AMS starts."

i suppose I am making a mistake in the Configuration Parameters of
Simulink. But I am clueless how to deal with it.

Please suggest.

Thanks and Regards,

Cheerss!!!

Debjit.

Debjit

unread,
Mar 24, 2010, 3:25:04 PM3/24/10
to

Dear All,

Please reply some one on the above issue... I am really in a fix and
stalled completely.

Another thing, is it possible to make some Real Time Analysis with the
spectre Toolbox provided with MMSIM distribution?

Please suggest..


Thanks and Regrads !!!

Cheerssss !!!!

Debjit

Debjit

unread,
Mar 24, 2010, 3:31:46 PM3/24/10
to

Hi All,

Can I read the data of AMSD simulation with the scripts of spectre
toolbox?

Please suggest.


Tanks and Regards,

Cheers!!!!

Debjit.

Andrew Beckett

unread,
Mar 25, 2010, 12:34:53 PM3/25/10
to
Debjit wrote, on 03/24/10 19:25:

If you're really in a fix, you should contact Cadence Customer Support (which is
the best way to deal with urgent issues) rather than relying on the generosity
and free time of the community - it's rather hard to debug this without having
access to your data.

The spectre Toolbox (as I think I said before) allows you to read the simulation
results from spectre or AMS, but won't be part of the simulation. It's purely a
post-processing tool.

Regards,

Andrew.

kon.vasil...@gmail.com

unread,
Mar 14, 2017, 6:37:34 PM3/14/17
to
Hi Debjit,

I ran into the same problem. It has to do with the sampling step you are using for the pulse in Simulink. If you use the clock block, it generates just 2 points per period and the scope plots it as a pulse. However, Cadence interpolates linearly between these points plotting a triangle. The actual data is the same it's just drawn differently.

Try using rectangular pulse block instead of clock in Simulink and give more than 1 point for the high and low levels (at least 4 points per period in total)
0 new messages