The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Modeling >> Semiconductor Devices >> NOIMOD in weak inversion
https://designers-guide.org/forum/YaBB.pl?num=1159564297

Message started by Sarig on Sep 29^th, 2006, 2:11pm

Title: NOIMOD in weak inversion
Post by Sarig on Sep 29^th, 2006, 2:11pm

Hello,
What is more accurate for noise in weak inversion :
NOIMOD=2 or NOIMOD=3

Title: Re: NOIMOD in weak inversion
Post by Geoffrey_Coram on Oct 2^nd, 2006, 5:30am

NOIMOD=3 uses the "SPICE2" thermal noise model (8/3 kT (gm + gds + gmbs) and NOIMOD=2 uses the "BSIM3" thermal noise model (based on Qinv). Both use the same "unified" flicker noise model.

I would expect that the BSIM3 model is "better" than the Spice2 model, simply because it was developed later. I will also point out that NOIMOD=6 updates the SPICE2 noise model with an extra factor in BSIM3v3.3 which was released last summer. The extra factor is 1.0 in saturation, but can be up to 1.5 in the linear region.

I don't have any comments specific to weak inversion. Sorry.

Title: Re: NOIMOD in weak inversion
Post by David Lee on Oct 14^th, 2006, 3:06pm

The inversion charge-based thermal noise model (NOIMOD=2) should be more accurate.
Have a look at the excellent book by Tsividis, "Operation and Modeling of the MOS Transistor".
Chapter 8.5.3 is a section on thermal noise modeling, and discusses weak inversion operation in particular.