Community Custom IC Design Thermal noise modeling

Stats

  • Locked Locked
  • Replies 5
  • Subscribers 125
  • Views 11192
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Thermal noise modeling

mahdiee86

Hi. 

I am interested in calculating the thermal noise power ("id.src" value in noise section of the result browser) of a MOSFET modeled by the BSIM4.5 model in Matlab. During my dc as well as noise simulation of a single transistor, I utilized all the operating point information found in the "dcOpInfo" section and the model parameters found in the "model" section of the result browser. In order to calculate thermal noise, I used the "holistic model". However, the results were different from the simulation. I would appreciate any assistance you can provide. Thanks 

Parents Reply Children
  • in reply to Andrew Beckett

    Hi, thanks for your reply. in "Spectre Circuit Simulator Components and Device Models Reference" version 20.1. section "Channel Thermal Noise" page 1412, there are two models for thermal noise which can be selected by "tnoiMod" parameter. The first model is charge-based, while the second is holistic (tnoiMod=1). The model parameters and operating point values were extracted from the simulation and these values were used in a simple MATLAB code to calculate the equations (equation number 22-149 , page 1413). However, the results are very different from what Spectre (result browser, noise results, "id.src") produced.

  • in reply to mahdiee86

    I think that's the problem - the "id" output you are seeing from the simulator (id.src) is the noise source for the shot noise arising from the drain current. It's not the thermal noise in that equation.

    Unfortunately there are a number of output noise sources within the bsim4 model which aren't documented and don't directly match the equations shown in the documentation. For example, there is: fn (flicker noise), id, rs, rd, igb, igd, igds, rbsd, rbdb, rbpb, rbps, rbpd, rgbi. I know I've requested in the past that we actually document all these output sources but it's not happened (I think my request has got lost along the way because it was a very long time ago).

    So you probably should contact customer support and request these are properly documented. I don't think you'll be able to quite tie up the equations you see with what is output by the simulator because they are partitioned in some way and some of the noise sources aren't in the equations anyway.

    Regards,

    Andrew

  • in reply to Andrew Beckett

    thanks for your reply. but based on the below link content, it seems that "id" output is the thermal noise and not shot noise. 

    support.cadence.com/.../ArticleAttachmentPortal

  • in reply to mahdiee86

    I've been having a discussion with Spectre R&D, and the BSIM4 part of the documentation is rather confusing. It's really not clear which noise contributor matches which equation (they suggested the BSIM3v3 documentation is clearer - but even that isn't that clear either). I suggest you contact customer support as there's a good argument that we should clear this up in the documentation.

    The article you reference is rather old - from 17 years ago (it was written by a former colleague of mine in the UK) and I don't know how accurate it is for current MOS models.

    Regards,

    Andrew

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.