The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> modeling a NMOS in NWELL capacitor?
https://designers-guide.org/forum/YaBB.pl?num=1166692713

Message started by bluestatic on Dec 21^st, 2006, 1:18am

Title: modeling a NMOS in NWELL capacitor?
Post by bluestatic on Dec 21^st, 2006, 1:18am

The NMOS in NWELL cap, from what I know is also called varactor by someone. I'm trying to utilize such kind of capacitor in a 0.13um logic process while the model file is not provided by the foundry. So DIY the model (if possible) is necessary.

By checking the varactor model from other process/fab, it seems a "NMOS in NWELL" cap can be modeled by using P-type BSIM model, as long as the following param is revised , for example:

model PVAR bsim3v3 type=p
+version = 3.2 tnom = 25
+capmod = 3 voffcv = -0.015 k1 = 0.482
+vth0 = -1.33+dPVAR2 acde = 1.0 tox = 2.77E-9+dPVAR1
+toxm = 2.77E-9 nch = 2.77E+17 dskip = no
+binunit = 2 k2=0 kt1=0.08
ends PVAR

If the fab provide normal nmos/pmos BSIM model only(and without further modeling support from them) , is there a safe way to "convert to" a varactor model? How should I modify the BSIM model parameter , ie, which are the vital parameter that I have to know to specify a varactor model?

Thanks for all the help

Title: Re: modeling a NMOS in NWELL capacitor?
Post by ywguo on Dec 21^st, 2006, 5:46pm

Hi,

If you DIY the model, how do you get the right number of those parameters, like vth0?

Thanks
Yawei

Title: Re: modeling a NMOS in NWELL capacitor?
Post by greatqs on Dec 25^th, 2006, 3:12am

I think Bsim3v3 model is not as simple as you think.
And it is of no sense if the model itself is not correct, especially for RF

bluestatic wrote on Dec 21^st, 2006, 1:18am:

Title: Re: modeling a NMOS in NWELL capacitor?
Post by bluestatic on Dec 27^th, 2006, 2:05am

guys, thanks for the reply.

As you can see that I'm trying to find a internal-relationship between normal PMOS transistor model and the Nmos in NWELL capacitor(which of course kind of risky) . and someone told me this kind model could be accurate enough if vth0 and tox is set correctly. However, I'm not sure about it.

My application is not RF, and in terms of design itself I can tolerate 20% variation. I'm wondering if someone has experience of handling it with succeful approximation model.