get monte carlo transient result

Hi,

I need to perform a monte carlo simulation and capture value(VT("/
node"), 10n) - or the transient voltage at some *node* at some *time*.
I can't seem to figure out how to do so in the Analog Statistical
Analysis GUI - I know I can probably do it with ocean scripts, but it
would be nice if I can get this to work.
I tried these two expression :
1. value(VT("/node") 1e-08), type scalar
2. value(v("/node" ?result 'tran) 1e-08), type scalar
both does not work - gives expression evaluation error on the output
log. I know I'm supposed to use OCEAN-type syntax - what's wrong with
#2.

Thank's a lot in advance

 

Certainly the latter should work - and in fact in IC5141 ISRs from the last few
months, the ADE calculator function VT should be equivalent (previously it
wasn't equivalent, and so did not work reliably in monte carlo).

I'm assuming you're trying to do this as an expression that you'll get a
histogram for - in other words, spectre will evaluate it at each point?

Take a look in the psf directory for a file called "artil.log" - this normally
displays any error messages during the evaluation (the evaluation of the
expression is done by a sub-process invoked from spectre called "artil").

Also, what subversions of DFII (icfb -W) and spectre (spectre -W) are you
using? I can look for any known problems that way.

Regards,

Andrew.

 

I'm using the following versions
ICFB : sub-version 5.10.41_USR3.27.30
Spectre : sub-version 5.10.41_USR3.102405

This is the artil.log file :
Loading auCore.cxt
Loading awv.cxt
"mps2472"

Results "subckts-info" are not available for
"../psf/logFile". Use results() for a list of available
results.
Loading analog.cxt
Loading asimenv.cxt
Loading spectrei.cxt

You're right, I'm trying to get the histogram, I can post-process it
but it would just take more time. I'm simulating offset of a dynamic
comparator. I have a verilog-A file that does a binary search by
sweeping the input voltage. I just need to sample that input voltage
at the end of the sweep - thus the expression.

Thank's for the response !

Renaldi

 

Renaldi,

There were a bunch of problems with IC5141 USR3 related to monte-carlo, so I
suspect this is what you're seeing.

I think you'll need to move to a newer build (USR5?). Also, I'd recommend using
spectre from an MMSIM stream (e.g. MMSIM61 or the new MMSIM62 release).

Regards,

Andrew.