Device models are defined with the .MODEL statement and assign electrical parameters to device instances.
.MODEL name type (param=val ...)
Parameters not specified use their default values. All models support temperature scaling -- the simulator adjusts parameters from the nominal temperature (TNOM, default 27C) to the circuit temperature (TEMP).
Model type: D
.MODEL DMOD D (IS=1e-14 N=1.05 RS=10 CJO=2P VJ=0.7 BV=100)
| Parameter |
Default |
Unit |
Description |
| IS |
1e-14 |
A |
Saturation current |
| N |
1 |
-- |
Emission coefficient |
| RS |
0 |
ohm |
Series resistance |
| BV |
inf |
V |
Reverse breakdown voltage |
| IBV |
1e-3 |
A |
Current at reverse breakdown voltage |
| Parameter |
Default |
Unit |
Description |
| CJO |
0 |
F |
Zero-bias junction capacitance |
| VJ |
1 |
V |
Junction potential |
| M |
0.5 |
-- |
Grading coefficient |
| TT |
0 |
s |
Transit time (diffusion capacitance) |
| FC |
0.5 |
-- |
Forward-bias depletion capacitance coefficient |
| Parameter |
Default |
Unit |
Description |
| EG |
1.11 |
eV |
Bandgap energy (silicon) |
| XTI |
3 |
-- |
Saturation current temperature exponent |
| Parameter |
Default |
Unit |
Description |
| KF |
0 |
-- |
Flicker noise coefficient |
| AF |
1 |
-- |
Flicker noise exponent |
The diode current is:
where is the thermal voltage (approximately 26mV at 27C).
Model type: NMOS or PMOS with LEVEL=1
.MODEL NMOS1 NMOS (LEVEL=1 VTO=0.7 KP=110U GAMMA=0.4 LAMBDA=0.04)
| Parameter |
Default |
Unit |
Description |
| VTO |
0 |
V |
Zero-bias threshold voltage |
| KP |
2e-5 |
A/V^2 |
Transconductance parameter |
| GAMMA |
0 |
V^0.5 |
Body-effect parameter |
| PHI |
0.6 |
V |
Surface potential |
| LAMBDA |
0 |
1/V |
Channel-length modulation |
| Parameter |
Default |
Unit |
Description |
| RD |
0 |
ohm |
Drain ohmic resistance |
| RS |
0 |
ohm |
Source ohmic resistance |
| Parameter |
Default |
Unit |
Description |
| CBD |
0 |
F |
Zero-bias bulk-drain junction capacitance |
| CBS |
0 |
F |
Zero-bias bulk-source junction capacitance |
| IS |
1e-14 |
A |
Bulk junction saturation current |
| PB |
0.8 |
V |
Bulk junction potential |
| CGSO |
0 |
F/m |
Gate-source overlap capacitance per unit channel width |
| CGDO |
0 |
F/m |
Gate-drain overlap capacitance per unit channel width |
| CGBO |
0 |
F/m |
Gate-bulk overlap capacitance per unit channel length |
| CJ |
0 |
F/m^2 |
Bottom junction capacitance per unit area |
| CJSW |
0 |
F/m |
Sidewall junction capacitance per unit periphery |
| MJ |
0.5 |
-- |
Bottom grading coefficient |
| MJSW |
0.33 |
-- |
Sidewall grading coefficient |
| FC |
0.5 |
-- |
Forward-bias capacitance coefficient |
| Parameter |
Default |
Unit |
Description |
| TOX |
1e-7 |
m |
Oxide thickness |
| LD |
0 |
m |
Lateral diffusion |
| U0 |
600 |
cm^2/V-s |
Low-field surface mobility |
| NSUB |
0 |
1/cm^3 |
Substrate doping |
| TPG |
1 |
-- |
Gate material type (+1 opposite, -1 same, 0 Al gate) |
| NSS |
0 |
1/cm^2 |
Surface state density |
Threshold voltage with body effect:
Drain current ():
Model type: NMOS or PMOS with LEVEL=2
.MODEL NMOS2 NMOS (LEVEL=2 VTO=0.7 KP=110U UCRIT=1e4 UEXP=0.1)
Level 2 extends Level 1 with analytical models for short-channel and narrow-channel effects. It inherits all Level 1 parameters and adds:
| Parameter |
Default |
Unit |
Description |
| NFS |
0 |
1/cm^2 |
Fast surface state density |
| UCRIT |
1e4 |
V/cm |
Critical field for mobility degradation |
| UEXP |
0 |
-- |
Critical field exponent |
| UTRA |
0 |
-- |
Transverse field coefficient |
| VMAX |
0 |
m/s |
Maximum carrier drift velocity |
| NEFF |
1 |
-- |
Total channel charge coefficient |
| XJ |
0 |
m |
Metallurgical junction depth |
| DELTA |
0 |
-- |
Width effect on threshold voltage |
Key differences from Level 1: mobility degradation via UCRIT/UEXP, velocity saturation via VMAX, narrow-channel effect via DELTA, subthreshold conduction via NFS, and analytical channel-length modulation.
Model type: NMOS or PMOS with LEVEL=3
.MODEL NMOS3 NMOS (LEVEL=3 VTO=0.7 THETA=0.1 ETA=0.1 KAPPA=0.5)
Level 3 uses a semi-empirical approach to model short-channel effects. It inherits all Level 1 parameters and adds:
| Parameter |
Default |
Unit |
Description |
| THETA |
0 |
1/V |
Mobility modulation |
| ETA |
0 |
-- |
Static feedback (DIBL effect) |
| KAPPA |
0.2 |
-- |
Saturation field factor |
| VMAX |
0 |
m/s |
Maximum carrier drift velocity |
| NFS |
0 |
1/cm^2 |
Fast surface state density |
| XJ |
0 |
m |
Metallurgical junction depth |
| DELTA |
0 |
-- |
Width effect on threshold voltage |
Level 3 is a reasonable choice for technology nodes from about 1um down to 0.35um. For geometries below 0.25um, BSIM3v3 (Level 8) is recommended.
Model type: NMOS or PMOS with LEVEL=8
.MODEL NMOS8 NMOS (LEVEL=8 VERSION=3.3 TNOM=27
+ VTH0=0.5 K1=0.6 K2=-0.1
+ TOX=9e-9 U0=300 VSAT=1.5e5
+ CGSO=2.5e-10 CGDO=2.5e-10)
BSIM3v3 is the industry-standard model for sub-micron MOSFETs. It has 150+ parameters, typically extracted by foundries and provided in process design kits.
| Parameter |
Default |
Unit |
Description |
| VERSION |
3.3 |
-- |
Model version |
| TNOM |
27 |
C |
Parameter extraction temperature |
| TOX |
1.5e-8 |
m |
Gate oxide thickness |
| TOXE |
1.5e-8 |
m |
Electrical oxide thickness |
| DTOX |
0 |
m |
TOX - TOXE |
| EPSROX |
3.9 |
-- |
Gate oxide dielectric constant |
| WINT |
0 |
m |
Channel width offset |
| LINT |
0 |
m |
Channel length offset |
| Parameter |
Default |
Unit |
Description |
| VTH0 (VTHO) |
0.7 |
V |
Long-channel threshold voltage at Vbs=0 |
| K1 |
0.5 |
V^0.5 |
First-order body effect coefficient |
| K2 |
0 |
-- |
Second-order body effect coefficient |
| K3 |
80 |
-- |
Narrow width effect coefficient |
| K3B |
0 |
1/V |
Body effect of narrow width |
| DVT0 |
2.2 |
-- |
Short-channel effect coefficient 0 |
| DVT1 |
0.53 |
-- |
Short-channel effect coefficient 1 |
| DVT2 |
-0.032 |
1/V |
Short-channel effect coefficient 2 |
| NLXL |
1.74e-7 |
m |
Lateral non-uniform doping length |
| W0 |
0 |
m |
Narrow width effect parameter |
| VFB |
-1 |
V |
Flat-band voltage |
| Parameter |
Default |
Unit |
Description |
| U0 |
670/250 |
cm^2/V-s |
Low-field mobility (NMOS/PMOS) |
| UA |
2.25e-9 |
m/V |
First-order mobility degradation coefficient |
| UB |
5.87e-19 |
(m/V)^2 |
Second-order mobility degradation coefficient |
| UC |
-4.65e-11 |
1/V |
Body-bias sensitivity of mobility degradation |
| Parameter |
Default |
Unit |
Description |
| VSAT |
8e4 |
m/s |
Saturation velocity |
| A0 |
1 |
-- |
Non-uniform depletion width effect coefficient |
| AGS |
0.2 |
1/V |
Gate bias coefficient of Abulk |
| Parameter |
Default |
Unit |
Description |
| VOFF |
-0.11 |
V |
Subthreshold offset voltage |
| NFACTOR |
1 |
-- |
Subthreshold swing factor |
| CIT |
0 |
F/m^2 |
Interface trap capacitance |
| CDSC |
2.4e-4 |
F/m^2 |
Drain/source to channel coupling capacitance |
| Parameter |
Default |
Unit |
Description |
| ETA0 |
0.08 |
-- |
DIBL coefficient in subthreshold |
| ETAB |
-0.07 |
1/V |
Body-bias coefficient for DIBL |
| DSUB |
0.56 |
-- |
DIBL coefficient in strong inversion |
| Parameter |
Default |
Unit |
Description |
| PCLM |
1.3 |
-- |
Channel-length modulation parameter |
| PDIBLC1 |
0.39 |
-- |
First DIBL correction parameter |
| PDIBLC2 |
0.0086 |
-- |
Second DIBL correction parameter |
| PSCBE1 |
4.24e8 |
V/m |
Substrate current body effect parameter 1 |
| PSCBE2 |
1e-5 |
V/m |
Substrate current body effect parameter 2 |
| PVAG |
0 |
-- |
Gate dependence of output resistance |
| Parameter |
Default |
Unit |
Description |
| CGSO |
0 |
F/m |
Gate-source overlap capacitance per width |
| CGDO |
0 |
F/m |
Gate-drain overlap capacitance per width |
| CGBO |
0 |
F/m |
Gate-bulk overlap capacitance per length |
| CLC |
1e-7 |
m |
Constant term for the short-channel model |
| CLE |
0.6 |
-- |
Exponential term for the short-channel model |
| CF |
0 |
F/m |
Fringing field capacitance |
| DLC |
0 |
m |
Length offset for capacitance |
| DWC |
0 |
m |
Width offset for capacitance |
| Parameter |
Default |
Unit |
Description |
| CJ |
5e-4 |
F/m^2 |
Bottom junction capacitance per area |
| CJSW |
5e-10 |
F/m |
Sidewall junction capacitance per perimeter |
| CJSWG |
0 |
F/m |
Gate-side sidewall junction capacitance |
| MJ |
0.5 |
-- |
Bottom junction grading coefficient |
| MJSW |
0.33 |
-- |
Sidewall junction grading coefficient |
| PB |
1 |
V |
Bottom junction built-in potential |
| JS |
1e-4 |
A/m^2 |
Bulk junction saturation current density |
BSIM3v3 supports automatic length/width scaling. Each parameter P can have associated LP, WP, and PP variants:
Model type: NMOS or PMOS with LEVEL=14
.MODEL NMOS14 NMOS (LEVEL=14 VERSION=4.5 TNOM=27
+ VTH0=0.4 TOX=1.8e-9 U0=300 VSAT=1.2e5
+ RDSW=200 RDSWMIN=0)
BSIM4 extends BSIM3v3 with 200+ additional parameters for deep sub-micron and nanoscale MOSFETs. It is the standard model for technology nodes from 130nm down to 22nm.
| Parameter |
Default |
Unit |
Description |
| AIGBACC |
1.36e-2 |
-- |
Accumulation gate current parameter A |
| BIGBACC |
1.71e-3 |
-- |
Accumulation gate current parameter B |
| CIGBACC |
0.075 |
-- |
Accumulation gate current parameter C |
| AIGC |
1.36e-2 |
-- |
Gate-to-channel tunneling current parameter A |
| BIGC |
1.71e-3 |
-- |
Gate-to-channel tunneling current parameter B |
| CIGC |
0.075 |
-- |
Gate-to-channel tunneling current parameter C |
| TOXREF |
3e-9 |
m |
Nominal gate oxide thickness for tunneling |
| Parameter |
Default |
Unit |
Description |
| RDSW |
200 |
ohm-um |
Source/drain resistance per unit width |
| RDSWMIN |
0 |
ohm-um |
Minimum RDSW |
| PRWB |
0 |
1/V^0.5 |
Body effect on RDSW |
| PRWGS |
0 |
1/V |
Gate bias effect on RDSW |
| WR |
1 |
-- |
Width offset from Weff for Rds |
| Parameter |
Default |
Unit |
Description |
| AGIDL |
0 |
A/V |
GIDL pre-exponential coefficient |
| BGIDL |
2.3e9 |
V/m |
GIDL exponential coefficient |
| CGIDL |
0.5 |
V |
GIDL reference voltage |
| EGIDL |
0.8 |
V |
GIDL activation energy |
- BSIM4 model cards from foundries are typically 200-500 lines. Do not hand-edit them.
- The
VERSION parameter should match the version used during parameter extraction (4.0 through 4.8).
- All BSIM3v3 geometry-scaling conventions are retained.
- spice-rs supports BSIM4 versions 4.0 through 4.8.
Model type: NPN or PNP
.MODEL NPN1 NPN (IS=1e-15 BF=200 NF=1 VAF=100 IKF=0.04
+ ISE=1e-13 NE=1.5 BR=5 NR=1 VAR=20
+ RB=100 RE=1 RC=10
+ CJE=2P VJE=0.7 MJE=0.33
+ CJC=1P VJC=0.75 MJC=0.33
+ TF=0.3N TR=6N)
| Parameter |
Default |
Unit |
Description |
| IS |
1e-16 |
A |
Transport saturation current |
| BF |
100 |
-- |
Ideal maximum forward beta |
| NF |
1 |
-- |
Forward current emission coefficient |
| VAF |
inf |
V |
Forward Early voltage |
| IKF |
inf |
A |
Corner for forward beta high-current roll-off |
| ISE |
0 |
A |
Base-emitter leakage saturation current |
| NE |
1.5 |
-- |
Base-emitter leakage emission coefficient |
| Parameter |
Default |
Unit |
Description |
| BR |
1 |
-- |
Ideal maximum reverse beta |
| NR |
1 |
-- |
Reverse current emission coefficient |
| VAR |
inf |
V |
Reverse Early voltage |
| IKR |
inf |
A |
Corner for reverse beta high-current roll-off |
| ISC |
0 |
A |
Base-collector leakage saturation current |
| NC |
2 |
-- |
Base-collector leakage emission coefficient |
| Parameter |
Default |
Unit |
Description |
| RB |
0 |
ohm |
Zero-bias base resistance |
| RBM |
RB |
ohm |
Minimum base resistance at high currents |
| IRB |
inf |
A |
Current where base resistance falls halfway to RBM |
| RE |
0 |
ohm |
Emitter resistance |
| RC |
0 |
ohm |
Collector resistance |
| Parameter |
Default |
Unit |
Description |
| CJE |
0 |
F |
Zero-bias base-emitter depletion capacitance |
| VJE |
0.75 |
V |
Base-emitter built-in potential |
| MJE |
0.33 |
-- |
Base-emitter grading coefficient |
| TF |
0 |
s |
Ideal forward transit time |
| XTF |
0 |
-- |
Transit time bias dependence coefficient |
| VTF |
inf |
V |
Transit time dependency on Vbc |
| ITF |
0 |
A |
Transit time dependency on Ic |
| Parameter |
Default |
Unit |
Description |
| CJC |
0 |
F |
Zero-bias base-collector depletion capacitance |
| VJC |
0.75 |
V |
Base-collector built-in potential |
| MJC |
0.33 |
-- |
Base-collector grading coefficient |
| XCJC |
1 |
-- |
Fraction of Cbc connected to internal base |
| TR |
0 |
s |
Ideal reverse transit time |
The Gummel-Poon model extends the Ebers-Moll model with base-width modulation (Early effect) via VAF/VAR, high-injection roll-off via IKF/IKR, and low-current non-ideal base current via ISE/NE and ISC/NC.
Collector current:
where is the normalized base charge:
Model type: NJF (N-channel) or PJF (P-channel)
.MODEL JMOD NJF (VTO=-2 BETA=1e-4 LAMBDA=2e-4 RD=10 RS=10 CGS=5P CGD=1P)
| Parameter |
Default |
Unit |
Description |
| VTO |
-2 |
V |
Pinch-off voltage (negative for NJF, positive for PJF) |
| BETA |
1e-4 |
A/V^2 |
Transconductance coefficient |
| LAMBDA |
0 |
1/V |
Channel-length modulation |
| IS |
1e-14 |
A |
Gate junction saturation current |
| N |
1 |
-- |
Gate junction emission coefficient |
| B |
1 |
-- |
Doping tail parameter |
| Parameter |
Default |
Unit |
Description |
| RD |
0 |
ohm |
Drain ohmic resistance |
| RS |
0 |
ohm |
Source ohmic resistance |
| Parameter |
Default |
Unit |
Description |
| CGS |
0 |
F |
Zero-bias gate-source junction capacitance |
| CGD |
0 |
F |
Zero-bias gate-drain junction capacitance |
| PB |
1 |
V |
Gate junction potential |
| FC |
0.5 |
-- |
Forward-bias depletion capacitance coefficient |
Drain current (NJF, ):