Xref: utzoo comp.lsi.cad:241 comp.lsi:806 Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!ames!vsi1!wyse!mips!mark From: mark@mips.COM (Mark G. Johnson) Newsgroups: comp.lsi.cad,comp.lsi Subject: BSIM model in SPICE Message-ID: <27668@obiwan.mips.COM> Date: 16 Sep 89 14:04:05 GMT References: <26135@iuvax.cs.indiana.edu> Reply-To: mark@mips.COM (Mark G. Johnson) Organization: MIPS Computer Systems, Sunnyvale, CA Lines: 62 In article <26135@iuvax.cs.indiana.edu> cdaf@iuvax.cs.indiana.edu (Charles Daffinger) writes: > >THis is in reference to Spice3c1 from Berkeley. > >I would like more information reguarding the bsim level of >simulation. Could someone point me to a reference? I >have mailed to the address supplied in the documention, >but is seems to be invalid now. > BSIM is an acronym for Berkeley Shortchannel Igfet Model. {and of course IGFET it itself an acronyn, for Insulated Gate Field Effect Transistor} It is a set of equations which attempt to predict the behavior of MOS transistors. There are other sets of equations (models), also present in SPICE, which attempt to do the same thing. These others go by the romantic names "Level=2" , Level=3" , etc. BSIM is merely a recent attempt at model formulation. It is a modification of the "CSIM" model from Bell Laboratories {ref 1}. Because it is relatively recent it applies a more newfangled approach to modeling some of the subtler, second-order effects. All three models (level=2, level=3, BSIM) incorporate code to model the subthreshold region; their differences are manifested in the Ids behavior in the region Vt<=Vgs<=(Vt+1.0); in the modeling of narrow- channel effects, in the conductance-slope of the saturation region, in temperature effects, and elsewhere. There's nothing magic about BSIM; it's yet-another-set-of-equations that try to fit the behavior of real devices. In fact one could argue that BSIM is _harder_ to use than the other models. There are 67 user-supplied parameters for BSIM; compare this with 23 parameters for level=2, 21 for level=1. {ref 3} As a side note, there are lots of circuit simulators that include among their various MOS transistor models, the BSIM model -- you don't need to get Berkeley-3c1 to use BSIM. The simulator that I use every day, HSPICE from Meta-Software, makes eleven different models available, one of which is BSIM. {Incidentally, among those model options, we didn't select BSIM. We chose to design and simulate the R3000 & R3010 microprocessor devices using the "level=3" model of MOS transistors}. I believe that several other commercially available simulators include BSIM in their set of MOS modeling choices. If you want to know the details, here are three papers of interest: [1] S. Liu & L. Nagel, "Small-signal MOSFET models for analog circuit design", IEEE Journal of Solid-State Circuits, Vol. SC-17, No. 6, December 1982, pp. 983-998. [2] B. J. Sheu et al, "BSIM: Berkeley Short-Channel IGFET Model for MOS transistors", IEEE Journal of Solid-State Circuits, Vol. SC-22, No. 4, August 1987, pp. 558-565. [3] C. P. Wang & B. J. Sheu, "Temperature Dependence Modeling for MOS VLSI Circuit Simulation", IEEE Transactions on Computer-Aided Design, Vol. 8, No. 10, October 1989, pp. 1065-1073. -- -- Mark Johnson MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086 (408) 991-0208 mark@mips.com {or ...!decwrl!mips!mark}