The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Model Compilation error in BDASIM
https://designers-guide.org/forum/YaBB.pl?num=1277371423

Message started by Mayank on Jun 24^th, 2010, 2:23am

Title: Model Compilation error in BDASIM
Post by Mayank on Jun 24^th, 2010, 2:23am

Hi,

I am able to run successfully my circuit file with Spectre.
When i am trying to run the same circuit file with BDASIM, which reads spectre netlist directly as an input file, I get the following error during compilation of models.

The spectre element type 'cpolpnw_cpolynlp_model_gigb' used in instance 'gigb' is not a subcircuit, model or supported intrinsic element. gigb (plus minus ) cpolpnw_cpolynlp_MODEL_gigb (mult=mult capatype=capatype toxratio=toxratio a_dep=a_dep tox=tox b_dep=b_dep a_acc=a_acc b_acc=b_acc) Fatal Error from h2sparser

The statement in the model file reads like :

gigb (plus minus ) cpolpnw_cpolynlp_MODEL_gigb (mult=mult capatype=capatype toxratio=toxratio a_dep=a_dep tox=tox b_dep=b_dep a_acc=a_acc b_acc=b_acc) Fatal Error from h2sparser

But Spectre runs perfectly fine with this model file.

How to solve this issue ??

Title: Re: Model Compilation error in BDASIM
Post by pancho_hideboo on Jun 24^th, 2010, 5:57am

Mayank wrote on Jun 24^th, 2010, 2:23am:

The spectre element type 'cpolpnw_cpolynlp_model_gigb' used in instance 'gigb' is not a subcircuit, model or supported intrinsic element.
gigb (plus minus ) cpolpnw_cpolynlp_MODEL_gigb (mult=mult capatype=capatype toxratio=toxratio a_dep=a_dep tox=tox b_dep=b_dep a_acc=a_acc b_acc=b_acc)

Show me content of 'cpolpnw_cpolynlp_model_gigb'.

Do you have "simulator lang=spectre insensitive=yes" in model file ?

Title: Re: Model Compilation error in BDASIM
Post by Mayank on Jun 24^th, 2010, 9:55am

Quote:

Do you have "simulator lang=spectre insensitive=yes" in model file ?

Affirmitive. both in model file as well as circuit netlist.

Quote:

Show me content of 'cpolpnw_cpolynlp_model_gigb'.

There is no content on cpolpnw_cpolynlp_model_gigb.
There's just one line in the complete model file for this block & i quoted that line.

Spectre is running the simulation but BDASIM is not able to find the model for it in the model file. Neither am I.

Title: Re: Model Compilation error in BDASIM
Post by pancho_hideboo on Jun 24^th, 2010, 10:10am

Don't you use encrypted model for Cadence Spectre ?
http://www.designers-guide.org/Forum/YaBB.pl?num=1205223090

Title: Re: Model Compilation error in BDASIM
Post by Mayank on Jun 24^th, 2010, 9:03pm

Read the posts from the link you posted.
I am not using any encrypted files. My model files are normal model.scs files.
like the one you were using for BJT models rfModels.scs

Quote:

simulator lang=spectre
model NPNlower bjt type=npn is=73.6e-18 bf=121 nf=1 vaf=35 ikf=96e-3 \
ise=2543e-18 ne=1.9 br=6.76 nr=1 var=3.2 ikr=3.56e-3 \
isc=2543e-18 nc=1.9 \
rb=33.3 irb=240e-6 rbm=1.67 re=1.83 rc=12 \
xtb=0.5 eg=1.2 xti=3 cje=179e-15 vje=0.895 mje=0.41 \
tf=11e-12 xtf=15 vtf=1.5 itf=120e-3 ptf=10 \
cjc=120e-15 vjc=0.642 mjc=0.33 xcjc=0.3 \
tr=5e-9 cjs=138e-15 vjs=0.65 mjs=0.33 fc=0.875 \
trb1=1.05e-3 trb2=4.4e-6 trc1=1e-3 trc2=3.1e-6 tre1=0 tre2=0 \
bvbe=3 bvbc=14 bvsub=14 kf=0.55e-10 af=1.8 iss=1e-16

model NPNupper bjt type=npn is=24.5e-18 bf=121 nf=1 vaf=35 ikf=32e-3 \
ise=847e-18 ne=1.9 br=3.81 nr=1 var=3.2 ikr=2e-3 \
isc=847e-18 nc=1.9 \
rb=100 irb=80e-6 rbm=5 re=3.5 rc=25 \
xtb=0.5 eg=1.2 xti=3 cje=59.6e-15 vje=0.895 mje=0.41 \
tf=11e-12 xtf=15 vtf=1.5 itf=40e-3 ptf=10 \
cjc=45.8e-15 vjc=0.642 mjc=0.33 xcjc=0.3 \
tr=5e-9 cjs=75.5e-15 vjs=0.65 mjs=0.33 fc=0.875 \
trb1=1.05e-3 trb2=4.4e-6 trc1=1e-3 trc2=3.1e-6 tre1=0 tre2=0 \
bvbe=3 bvbc=14 bvsub=14 kf=1.32e-10 af=1.8 iss=1e-16

This isn't encrypted, right.

Title: Re: Model Compilation error in BDASIM
Post by pancho_hideboo on Jun 25^th, 2010, 3:52am

Mayank wrote on Jun 24^th, 2010, 9:03pm:

I am not using any encrypted files. My model files are normal model.scs files.

I think there are encrypted model files.
Some model files are included in other model files, so you might not notice encrypted model files.

Check all model files you are including.

Title: Re: Model Compilation error in BDASIM
Post by Geoffrey_Coram on Jun 28^th, 2010, 4:21am

I think there are three possibilities:
1) cpolpnw_cpolynlp_model_gigb is defined in some model file that you don't realize Spectre is using
2) the instance of cpolpnw_cpolynlp_model_gigb is in a subckt that Spectre is not using, and so Spectre doesn't care that the model is not defined.
3) Fatal Error from h2sparser means that bdasim is treating your input as hspice and this is causing problems.

You can check (2) by changing cpolpnw_cpolynlp_model_gigb to cpolpnw_cpolynlp_model_gigb_FAKE and seeing if Spectre still runs: if so, then the subckt isn't used.

You can also try setting a bogus parameter
(bogus=1 mult=mult ...
and then seeing if Spectre tells you more about how "bogus" isn't a parameter for ... defined in ...

Title: Re: Model Compilation error in BDASIM
Post by [MM] on Jul 7^th, 2010, 12:54pm

It would be great for you to contact BDA Technical Support so that we can help you with this testcase. (techsupport@berkeley-da.com). From the information that has been posted, here are some possibilities:

The error message indicates that AFS is unable to find the following model/subcircuit in any of the input files: cpolpnw_cpolynlp_model_gigb.

A quick scan of all the input files and libraries with:
grep -riI "cpolpnw_cpolynlp_model_gigb" *

would identify the part regardless of case. Start at the folder where the input file is defined. If this doesn't work, try the full user's home folder or start at the root file /.

If it is a case issue, then insensitive=yes would need to be used.

If the model is encrypted, then grep won't find it. Since it is Spectre proprietary, AFS will not read it.

Finally, there may be default include paths or libraries set up in the environment which would need to be passed to AFS (and can be done with certain commands).

Title: Re: Model Compilation error in BDASIM
Post by Mayank on Jul 22^nd, 2010, 11:48pm

Hello MM & all,

cpolpnw_cpolynlp_model_gigb has not been defined as a subckt in any of the model files.
But it has been instantiated time & again in the model files.
& it's not just cpolpnw_cpolynlp_model_gigb.
There are many other such undefined subckts in other model files as well which spectre reads perfectly but BDA gives a parsing error.
Tried this in many technologies.
What's the real issue ?

BDA is working fine with spectre models of a few technologies. But it just refuses to parse for others.

i checked all the files. i am not including any encrypted model file. Had there been some encrypted file, i would have seen sm incomprehensible ascii text which i am not able to see in any of these model files.
Maybe spectre ignores these names/so called subckts.
Are the encrypted models something that comes default with Spectre ?? ...coz i am not including any.

Quote:

If it is a case issue, then insensitive=yes would need to be used.

I am already using this statement.

--
M

Title: Re: Model Compilation error in BDASIM
Post by Geoffrey_Coram on Jul 26^th, 2010, 4:32am

Mayank -
Did you try my suggestion:

Quote:

You can check (2) by changing cpolpnw_cpolynlp_model_gigb to cpolpnw_cpolynlp_model_gigb_FAKE and seeing if Spectre still runs: if so, then the subckt isn't used.

What I mean is, if the model is instantiated in the model file -- but only in subckts that themselves are not used! -- then this may be OK for Spectre, but not for BDASim.

I don't really know what you expect the forum members to do for you. We've given you some suggestions of where to look; we all know that models need to be defined somewhere or be built-in (and we all agree that this cpolpnw_cpolynlp_model_gigb is not a Spectre built-in).