How to fix this: { *Error* eval: undefined function } on cadence IC5.1.4.1

[KimPhan]

Hi everybody,
I do simulation on cadence IC5.1.4.1 with PDK BiCMOS 180nm but when I want to see the operating point of bjt,
it shows me like this:



Can anybody help me to solve this error?
Thank you very much.

 

[pancho_hideboo]

Show me CDF parameters.

 

[KimPhan]

Thank for your attention,
This is the CDF parameter



Is it OK with your requirement?

 

[pancho_hideboo]

Show me CDF of Cell not Lib.

Confirm model description of BJT.

Is BJT a subckt ?
If so, it has to be inline sub circuit.

inline subckt gpdk090_npn (C B E)
parameters area=1 
gpdk090_npn (C B E)  gpdk090_npn_x area=area
model gpdk090_npn_x bjt type=npn is=cis*(1-2m/26m*pis_mc)*3.26E-16 va=cvaf*60 bf=cbeta*100 ise=cis*3.2E-16
+                      br=cbeta*6 nc=2 ikr=1m rc=10 vje=0.7  rb=1000 re=1 var=cvaf*5
+                      cjc=ccj*0.05e-12 fc=0.5 cje=ccj*0.07e-12 kf=0.1f ikf=0.5m
+                      tr=ctf*200e-12 tf=ctf*10e-12 itf=0.03 vtf=1 xtf=2 mvt0=1.41*2m
ends gpdk090_npn

What parameters do you want to show ?

 

[KimPhan]

Show me CDF of Cell not Lib.

Confirm model description of BJT.

Is BJT a subckt ?
If so, it has to be inline sub circuit.

inline subckt gpdk090_npn (C B E)
parameters area=1 
gpdk090_npn (C B E)  gpdk090_npn_x area=area
model gpdk090_npn_x bjt type=npn is=cis*(1-2m/26m*pis_mc)*3.26E-16 va=cvaf*60 bf=cbeta*100 ise=cis*3.2E-16
+                      br=cbeta*6 nc=2 ikr=1m rc=10 vje=0.7  rb=1000 re=1 var=cvaf*5
+                      cjc=ccj*0.05e-12 fc=0.5 cje=ccj*0.07e-12 kf=0.1f ikf=0.5m
+                      tr=ctf*200e-12 tf=ctf*10e-12 itf=0.03 vtf=1 xtf=2 mvt0=1.41*2m
ends gpdk090_npn

What parameters do you want to show ?

Here is the CDF of cell in 3 mode: effective, user, base


I tried to change CDF of cell or lib many times, but it still didn't work.

Actually I want to know everything about BJT like ic, ie, vbe, vce, ft,... through the print command in
analog environment.
But just show some typical specifications like ic, vce, vcesat, ft on the schematic.
In the PDK, just bjt npn have this problem.
I dont know how to fix this problem with your code.
Please give me some details to fix it?
Thank you.

 

[pancho_hideboo]

Confirm model description of BJT.

Is BJT a subckt ?
If so, it has to be inline sub circuit.

And try "Schematic>Edit>Component Display".

 

[KimPhan]

Confirm model description of BJT.

Is BJT a subckt ?
If so, it has to be inline sub circuit.

And try "Schematic>Edit>Component Display".

Sorry for that.
It is a component in the PDK BiCMOS 180nm, not a subckt.

 

[pancho_hideboo]

Sorry for that.
It is a component in the PDK BiCMOS 180nm, not a subckt.

No.
Generally models of PDK are described as subcircuit especially for RF device.
Confirm model description.

And try "Schematic>Edit>Component Display".

 

[KimPhan]

No.
Generally models of PDK are described as subcircuit especially for RF device.
Confirm model description.

And try "Schematic>Edit>Component Display".

This is about the security problem. So I can just show you some lines of the model description below:

---

Code dot - [expand]
1
2
3
4
5
6
7
8
9
10
11
12
13
subckt ln122_hicum (3 2 1 9)
parameters temprise=0 l=10.16e-6 postlayout=0 rth_flag=0 
*.PARAM
+ rth_flag_final = min( rth_flag+rth_flag_global, 1)
+ eml = l*1e6 
+ el = l*1e6
+ ef = 1
+ emw = 1.5e-07*1e6
+ sqrt_el = sqrt(m*ef*el)
+ sqrt_ew = sqrt(emw)
+ sqrt_em_area = sqrt(m*ef*emw*el)
.........
ends ln122_hicum

---

I think I can not edit the model description.
And I try to schematic-edit-Component Display and I set like this but it didn't work.


So what shoud I do now to fix this?

 

[pancho_hideboo]

As I suspect, it is not inline subcircuit.
You can not annotate anything.

Can you understand "inline subcircuit" ?

Change it to inline subcircuit.

I think I can not edit the model description.

Copy model files.

And I try to schematic-edit-Component Display and I set like this but it didn't work.

You don't select any instance.

 

[KimPhan]

As I suspect, it is not inline subcircuit.
You can not annotate anything.

Can you understand "inline subcircuit" ?

Change it to inline subcircuit.

Copy model files.

You don't select any instance.

I changed all subckt to inline subckt in the model description as you said and
then I tried to edit_component display and i choose the component on the schematic
like the picture I attached. As result, it still didn't work as expected.


In the PDK, only this npn device have this problem. I did with fet in PDK and it was OK.

 

[pancho_hideboo]

Show me your inline subcircuit description.
Can you understand inline subcircuit correctly ?

 

[KimPhan]

Show me your inline subcircuit description.
Can you understand inline subcircuit correctly ?

I just replaced all subckt to inline subckt like this:

inline subckt ln122_hicum (3 2 1 9)
parameters temprise=0 l=10.16e-6 postlayout=0 rth_flag=0
*.PARAM
+ rth_flag_final = min( rth_flag+rth_flag_global, 1)
+ eml = l*1e6
+ el = l*1e6
+ ef = 1
+ emw = 1.5e-07*1e6
+ sqrt_el = sqrt(m*ef*el)
+ sqrt_ew = sqrt(emw)
+ sqrt_em_area = sqrt(m*ef*emw*el)

ends ln122_hicum

I am not specialize in this. Can you tell me some details.
Thank you.

 

[pancho_hideboo]

You show only header.
Show me all contents.
There is no BJT device description.

 

[KimPhan]

You show only header.
Show me all contents.
There is no BJT device description.

I am so sorry to rely your comment slowly.
Please send me your email and I will send you
the model description file.
Thank you so much.

 

[pancho_hideboo]

Please send me your email

Reject.
I don't open e-mail.

 

[dick_freebird]

I don't see anything to say what the "annotation texts"
are. You can dig into the symbol view and see their
literal form. Compare those texts' construction to ones
that -do- annotate properly. Is there a difference?
That might lead you somewhere.

Now you can look for (I forget which branch of the
menu tree) Annotate>Design Defaults and this ought
to change those texts to the geometric properties
of the transistor. Does this work? It had better. If it
fails similarly then I'd guess that the CDF stuff that
maps those texts' functions' arguments is busted.

I don't see any mention that you tried to avail
yourself of the sourcing foundry's applications
engineers, who ought to know the library contents
better than we do (esp. since you do not mention
the foundry). But then I see "gpdk..." which may
mean this PDK is really just an illusion from a pool
of contributors, none of whom have any responsibility
to help you. Still, gpdk affiliated discussion boards
may be a more likely source of help (or, at least,
confirmation that you're not alone in this problem).

The last possibility that comes to mind is that the
function is, as it says, simply undefined (or, you are
missing some -other- definition / declaration). Look
to your install procedure and any special instructions
about where to put things and environment variables
that have to be set consistently with that. Look also
to your CDS.log file starting from the top for errors
and warnings about stuff-not-found.

 

[FvM]

Reject.
I don't open e-mail.

Right. Requesting communication via private Email is also against forum rules.

 

[KimPhan]

Reject.
I don't open e-mail.

I am sorry about that. I don't know this rule.
This is the file of model description.
I hope you can detect this problem exactly.
thank you so much.
View attachment npn_model.txt

 

[pancho_hideboo]

Surely see https://www.edaboard.com/showthread.php?t=371576#4

inline subckt ln121_hicum (3 2 1 9)
//q 3 2 1 9 mdn_hicum trise=temprise
ln121_hicum (3 2 1 9) mdn_hicum trise=temprise
ends ln121_hicum

inline subckt ln122_hicum (3 2 1 9)
//q 3 2 1 9 mdn_hicum trise=temprise
ln122_hicum (3 2 1 9) mdn_hicum trise=temprise
ends ln122_hicum

inline subckt ln232_hicum (3 2 1 9)
//q 3 2 1 9 mdn_hicum trise=temprise
ln232_hicum (3 2 1 9) mdn_hicum trise=temprise
ends ln232_hicum

inline subckt mn121_hicum (3 2 1 9 )
//q 3 2 1 9 mdn_hicum trise=temprise
mn121_hicum (3 2 1 9) mdn_hicum trise=temprise
ends mn121_hicum 

inline subckt mn122_hicum (3 2 1 9 )
//q 3 2 1 9 mdn_hicum trise=temprise
mn122_hicum (3 2 1 9) mdn_hicum trise=temprise
ends mn122_hicum 

inline subckt mn232_hicum (3 2 1 9 )
//q 3 2 1 9 mdn_hicum trise=temprise
mn232_hicum (3 2 1 9) mdn_hicum trise=temprise
ends mn232_hicum