LEVEL 47 BSIM3 Version 2 MOS Model

The BSIM3 version 2.0 MOS model from UC Berkeley is available as the LEVEL 47 Star-Hspice model.

LEVEL 47 Model Parameters

Name	Unit	Default	Description
VTH0	V	0.7	Threshold voltage of long channel at Vbs = 0 and small Vds (0.7 for n-channel, - 0.7 for p-channel)
K1	V 1/2	0.53	First-order body effect coefficient
K2		-0.0186	Second-order body effect coefficient
K3		80.0	Narrow width effect coefficient
K3B	1/V	0	Body width coefficient of narrow width effect
KT1	V	-0.11	Temperature coefficient for threshold voltage
KT2		0.022	Body bias coefficient of threshold temperature effect
GAMMA1	V 1/2	See "LEVEL 47 Model Equations"	Body effect coefficient, near interface
GAMMA2	V 1/2	See "LEVEL 47 Model Equations"	Body effect coefficient in the bulk
W0	m	2.5e-6	Narrow width effect coefficient
NLX	m	1.74e-7	Lateral nonuniform doping along channel
TOX	m	150e-10	Gate oxide thickness
XJ	m	0.15e-6	Junction depth
DL	m	0.0	Channel length reduction on one side (multiplied by SCALM)
DW	m	0.0	Channel width reduction on one side (multiplied by SCALM)
NPEAK	cm -3 (see The model parameters NPEAK and U0 can be entered in meters or centimeters. NPEAK is converted to cm-3 as follows: if NPEAK is greater than 1e20, it is multiplied by 1e-6. U0 is converted to m2/Vsec as follows: if U0 is greater than 1, it is multiplied by 1e-4. You must enter the parameter NSUB in cm-3 units.)	1.7e17	Peak doping concentration near interface
NSUB	cm -3	6.0e16	Substrate doping concentration
PHI	V	See "LEVEL 47 Model Equations"	Surface potential under strong inversion
XT	m	1.55e-7	Doping depth
VBM	V	-5.0	Maximum substrate bias
VBX	V	See "LEVEL 47 Model Equations"	Vbs at which the depletion width equals XT
DVT0		2.2	Short-channel effect coefficient 0
DVT1		0.53	Short-channel effect coefficient 1
DVT2	1/V	-0.032	Short-channel effect coefficient 2
U0	m 2 /Vsec (see The model parameters NPEAK and U0 can be entered in meters or centimeters. NPEAK is converted to cm-3 as follows: if NPEAK is greater than 1e20, it is multiplied by 1e-6. U0 is converted to m2/Vsec as follows: if U0 is greater than 1, it is multiplied by 1e-4. You must enter the parameter NSUB in cm-3 units.)	0.067	Low field mobility at T = TREF (0.067 for n-channel, 0.025 for p-channel)
UA	m/V	2.25e-9	First-order mobility degradation coefficient
UA1	m/V	4.31e-9	Temperature coefficient of UA
UB	m 2 /V 2	5.87e-19	Second-order mobility degradation coefficient
UB1	m 2 /V 2	-7.61e-18	Temperature coefficient of UB
UC	1/V	0.0465	Body bias sensitivity coefficient of mobility
UC1	1/V	-0.056	Temperature coefficient of UC
VSAT	cm/sec	8e6	Saturation velocity of carrier at T = TREF
AT	m/sec	3.3e4	Temperature coefficient of VSAT
RDSW	ohm `·` µm	0.0	Source drain resistance per unit width
RDS0	ohm	0.0	Source drain contact resistance
LDD	m	0.0	Total length of LDD region
ETA		0.3	Coefficient of drain voltage reduction
ETA0		0.08	Subthreshold region DIBL (Drain Induced Barrier Lowering) coefficient
ETAB	1/V	-0.07	Subthreshold region DIBL coefficient
EM	V/m	4.1e7	Electrical field in channel above which hot carrier effect dominates
NFACTOR		1.0	Subthreshold region swing
VOFF	V	-0.11	Offset voltage in subthreshold region
LITL	m		Characteristic length. The default is
VGLOW	V	-0.12	Lower bound of the weak-strong inversion transition region
VGHIGH	V	0.12	Upper bound of the weak-strong inversion transition region
CDSC	F/m 2	2.4e-4	Drain/source and channel coupling capacitance
CDSCB	F/Vm 2	0	Body coefficient for CDSC
CIT	F/m 2	0.0	Interface state capacitance
PCLM		1.3	Coefficient of channel length modulation
PDIBL1		0.39	DIBL (Drain Induced Barrier Lowering) effect coefficient 1
PDIBL2		0.0086	DIBL effect coefficient 2
DROUT		0.56	DIBL effect coefficient 3
DSUB		DROUT	DIBL coefficient in subthreshold region
PSCBE1	V/m	4.24e8	Substrate current induced body effect exponent 1
PSCBE2	m/V	1.0e-5	Substrate current induced body effect coefficient 2
A0		1	Bulk charge effect. The default is 4.4 for PMOS.
TNOM (TREF)	°C	25	Temperature at which parameters are extracted. This parameter defaults to the option TNOM, which defaults to 25 °C. See 4 and 5 in "Reminders for this Installation," below.
SUBTHMOD		2	Subthreshold model selector
SATMOD		2	Saturation model selector
KETA	1/V	-0.047	Body bias coefficient of the bulk charge effect
A1	1/V	0	First nonsaturation factor (0 for NMOS, 0.23 for PMOS)
A2		1.0	Second nonsaturation factor (1.0 for NMOS, 0.08 for PMOS)
UTE		-1.5	Mobility temperature exponent
KT1L	Vm	0	Channel length sensitivity of temperature coefficient for threshold voltage
UC0*	(V/m) 2		Temperature coefficient
BULKMOD		1	Bulk charge model selector
XPART		1	Charge partitioning flag
VFB	V		Flat-band voltage
PVAG		0	Gate dependence of output resistance
* UC0 has no effect on the model

Using the BSIM3 Version 2 MOS Model in Star-Hspice

The Star-Hspice LEVEL 47 model uses the same model parameters for source/drain diode current, capacitance, and resistance as do the other Star-Hspice MOS levels. The model parameter ACM controls the choice of source/drain equations.

The Star-Hspice LEVEL 47 model also uses the same noise equations as the other levels. The parameter NLEV controls the choice of noise equations.

This model, like all models in Star-Hspice, can be parametrized. This is useful for modeling process skew, either by worst-case corners or by Monte Carlo. For information on worst-case and Monte Carlo analysis, see Performing Worst Case Analysis and Performing Monte Carlo Analysis.

Notes:

1. Set LEVEL=47 to identify the model as a BSIM3 model.

2. This model is based on BSIM3 version 2.0 from UC Berkeley. Code was received from UC Berkeley in July 1994, in the form of SPICE3e2. Changes announced in a letter from UCB September 13, 1994, have been included. DC sweeps have been checked against SPICE3e2.

3. The default setting for CAPOP is CAPOP=13, which is the BSIM1 charge-conserving capacitance model. The BSIM3 capacitance model has not been installed.

4. The LEVEL 47 model supports the model parameter name TNOM as an alias for TREF. The conventional terminology in Star-Hspice is TREF, which is supported as a model parameter in all Star-Hspice MOS levels. The alternative name TNOM is supported for LEVEL 47, for compatibility with SPICE3.

5. The default room temperature is 25°C in Star-Hspice, but is 27°C in SPICE3. If the BSIM3 model parameters are specified at 27°C, TREF=27 should be added to the model, so that the model parameters is interpreted correctly. It is a matter of choice whether or not to set the nominal simulation temperature to 27, by adding .OPTION TNOM=27 to the netlist. This should be done when testing Star-Hspice versus SPICE3.

6. The default of DERIV is zero, the analytical method. DERIV can be set to 1 for the finite difference method. The analytic derivatives in the SPICE3e2 code are not exact in some regions. Setting DERIV=1 gives more accurate derivatives (GM, GDS, GMBS), but consumes more CPU time.

7. There are three ways for the BSIM3 model to calculate Vth:

Using K1 and K2 values that are user specified
Using GAMMA1, GAMMA2, VBM, and VBX values entered in the .MODEL statement
Using NPEAK, NSUB, XT, and VBM values that are user specified

8. The model parameters NPEAK and U0 can be entered in meters or centimeters. NPEAK is converted to cm-3 as follows: if NPEAK is greater than 1e20, it is multiplied by 1e-6. U0 is converted to m2/Vsec as follows: if U0 is greater than 1, it is multiplied by 1e-4. You must enter the parameter NSUB in cm-3 units.

9. The specified value of VTH0 for p-channel in the .MODEL statement should be negative.

10. The default value of KT1 is -0.11. The negative sign ensures that the absolute value of threshold decreases with increasing temperature for NMOS and PMOS.

11. Model parameter LITL is not allowed to go below a minimum value of 1.0e-9 m, to avoid a possible divide by zero error.

12. VSAT, after temperature adjustment, is not allowed to go below a minimum value of 1.0e4 m/sec, to assure that it is positive after temperature compensation.

13. There are seven model parameters for accommodating the temperature dependencies of six temperature dependent model variables. They are KT1 and KT2 for VTH, UTE for U0, AT for VSAT, UA1 for UA, UB1 for UB, and UC1 for UC.

14. Set up the conversion of temperature between Star-Hspice and SPICE3 as follows:

SPICE3: .OPTIONS TEMP=125

.MODEL NCH NMOS LEVEL=8

+ TNOM =27 ...

HSPICE: .TEMP 125

.MODEL NCH NMOS LEVEL=47

+ TREF =27 ...

15. The option SCALM does not affect the parameters unique to this model, but it does affect the common MOS parameters, such as XL, LD, XW, WD, CJ, CJSW, JS, and JSW.

16. LEVEL 47 uses the common Star-Hspice MOS parasitic models, specified by ACM.

17. LEVEL 47 uses the common Star-Hspice MOS noise models, specified by NLEV.

18. DELVTO and DTEMP on the element line can be used with LEVEL 47.

19. The impact ionization current determined by the model parameters PSCBE1 and PSCBE2 contributes to the drain-source current; it does not contribute to bulk current.

Leff and Weff Equations for BSIM3 Version 2.0

The standard Star-Hspice equations for Leff and Weff are:

The UCB SPICE3 equations used for BSIM3 are:

The units for these parameters are meters, with defaults of zero.

Star-Hspice uses the standard Star-Hspice equation for both cases, and accepting DL(DW) as the value for LD(WD). If both LD(WD) and DL(DW) are specified in an Star-Hspice .MODEL statement, Star-Hspice uses the LD(WD) value.

If LDAC and WDAC are included in the .MODEL statement,

Leff = L+XL-2·LDAC, Weff = W+XW-2·WDAC

The model uses the values of LD(DL) and WD(DW) to generate defaults for CGSO, CGDO, and CGBO. The values are also used with parameters RS and RD for ACM>0.