rm12 463 766 PDF

Chapter 25 Macros
What's in this chapter

This chapter describes some of the macro circuit files provided with the Micro-
Cap package. Most have numeric parameters that allow the macro function to be
adapted to particular circuit uses.
• Absolute value (full wave rectifier) function (ABS)

• Amplitude modulator (AM)
• Amplifier (AMP)
• Balun (BALUN)
• Center-tapped transformer (CENTAP)
• Limiter circuit (CLIP)
• Comb Filter (Comb Filter)
• Fast comparator using TANH functions (COMP)
• Comparator (COMPARATOR)
• Constant power load (Constant Power)
• Continuous Phase Frequency Shift Keyer (CPFSK)
• DC Motor (DC Motor)
• Delay (DELAY)
• Diac (DIAC)
• Differentiator (DIF)
• Digital potentiometer (DIGPOT)
• Divider (DIV)
• Linear transfer function (F)
• Frequency-shift keyer (FSK)
• F weighting function (FWeighting)
• Gyrator for impedance transformation (GYRATOR)
• Ideal 2-port transformer (IDEAL_TRANS2)
• Ideal 3-port transformer (IDEAL_TRANS3)
• Integrator (INT)
• Integrator with reset (INT_RESET)
• Memresistor using Biolek or Joglekar window (Memresistor)
• Monostable (MONOSTABLE)
• Monostable Retriggerable (MONOSTABLE_RT)
• Two-input weighted multiplier (MUL)
• Time domain noise source (NOISE)
463
• Peak detector (Peak Detector)
• Photodiode (PHOTODIODE)
• Photodiode_r (PHOTODIODE_R)
• Proportional integrator derivative controller (PID Controller)
• Potentiometer (POT)
• Phase-shift keyer (PSK)
• Programmable Unijunction Transistor (PUT)
• Pulse width modulator (PWM)
• Pulse width modulator with a T flip-flop(PWM_T)
• Pulse width modulator without a T flip-flop(PWM_NT)
• Simple relay model (RELAY1)
• State-variable relay model (RELAY2)
• Resonant circuit (RESONANT)
• Resistance temperature detector (RTD)
• Schmitt trigger circuit (SCHMITT)
• Silicon-controlled rectifier (SCR)
• Slip or hysteresis function (SLIP)
• Snubber (SNUBBER)
• Spark gap device (SPARKGAP)
• Two-input weighted subtracter (SUB)
• Two-input weighted summer (SUM)
• Three-input weighted summer (SUM3)
• Triac (TRIAC)
• Trigger (TRIGGER6)
• Triode vacuum tube model (TRIODE)
• Voltage-controlled oscillator (VCO)
• Wideband transformer (WIDEBAND)
• Crystal model (XTAL)
• 555 Timer (555)
464 Chapter 25: Macros

ABS
Sometimes referred to as a full-wave rectifier, the ABS block provides the abso-
lute value of the input signal. Its definition is:
VOut(t) = | VIn(t) |

The function is implemented with the ABS macro:
Figure 25-1 ABS macro circuit
There are no input parameters. This implementation uses an NFV Function

source to provide the absolute value function. The macro block mainly serves to
provide a more suitable symbol than the general source symbol of the NFV Func-
tion source.
See the circuit SYSTEM2 for an example of the use of this macro.
465
AM
This block provides an amplitude modulation function implemented as follows
Figure 25-2 AM macro circuit
The device consists of a single NFV source whose VALUE attribute is selected
by the TYPE parameter which can be one of two values, SM or CM. These two
symbolic variables SM and CM are defined as follows:
.define SM Offset*(1+ModIndex*V(In))*VPeak*Sin(2*PI*FS*t)
.define CM Offset*(1+ModIndex*V(In))*VPeak*Cos(2*PI*FS*t)
Parameter Definition
FS Frequency of the carrier signal
VPEAK Magnitude of the carrier signal
MODINDEX Modulation Index
OFFSET Offset added to the input signal
TYPE Modulation type. SM=Sine modulation.
CM=Cosine modulation
See the circuit AMTEST1.CIR for an example of how to use this macro.

AMP
This block provides a simple linear amplifier. Its definition is:
VOut(t) = gain VIn(t)

The function is implemented with the AMP macro:
Figure 25-3 AMP macro circuit
The single input parameter, GAIN, multiplies the input to produce an amplified
output. This implementation uses a simple linear dependent VOFV source. It
could have been done with a Function source or a Spice poly source. In general,
the simplest type of source that will perform the function is preferred.
467
BALUN
The BALUN macro is a macro which converts balanced signals to unbalanced
signals or vice versa.
Figure 25-4 BALUN macro circuit
See the circuit BALUNMIXER for an example of the use of this macro.

CENTAP
The CENTAP macro is a center-tapped five terminal transformer with parameters
defining the primary and two secondary inductances.
This circuit is implemented with three linear inductors and one K (coupling)
device.
Figure 25-5 CENTAP macro circuit
L1 Primary inductance
L2 Secondary inductance 1
L3 Secondary inductance 2
See the circuit TRANS for an example of the use of this macro.
469
CLIP
The clip macro can be used as a limiter, ideal OPAMP, or inverter. It provides an
output that is a scaled copy of the input, but limited to the specified maximum
and minimum levels.
This function is implemented with the CLIP macro:
Figure 25-6 CLIP macro circuit
A pair of input parameters, LOW and HIGH, define the lowest value and highest
value of the output. Between these limits, the output equals the input multiplied
by the GAIN parameter. The block is constructed of a NTVOFV Function table
source.
GAIN Linear gain between clip limits
LOW Lowest output value
HIGH Highest output value

Comb Filter
A comb filter adds a delayed version of a signal to itself in order to cause either
constructive or destructive interference to the signal. Comb filters are used in a
number of signal processing application such as CIC filters, television decoders,
and audio effects.
Here is the circuit that implements the macro.
Figure 25-6 Comb Filter macro circuit
SCALE Scaling factor applied to the delayed signal
DELAY Delay applied to the original signal in seconds
TYPE 1 = Feed-forward comb filter 2 = Feedback comb filter
See the circuit Comb Test for an example of the use of this macro. There is also
an article describing the macro at:
http://www.spectrum-soft.com/news/fall2012/comb.shtm
471
COMP
The COMP macro is a simple, fast, and convergence-friendly voltage comparator
implemented with a TANH function.
Figure 25-7 COMP macro circuit
A pair of input parameters, VOL and VOH, define the lowest value and highest
value of the output. Between these limits, the output equals the input multiplied
by the GAIN parameter. The block is constructed of a function source whose ex-
pression is DC+VA*TANH(GAIN*(V(INP,INM))), where
VA = (VOH-VOL)/2
DC = (VOH+VOL)/2
The use of the smooth TANH function with its continuous derivatives improves
convergence in critical circuits.
GAIN Linear gain between clip limits
VOL Lowest output value
VOH Highest output value

COMPARATOR
This circuit is a macro model for a comparator with hysteresis and is similar in
construction to the Schmitt macro. The macro circuit looks like this:
Figure 25-8 COMPARATOR macro equivalent circuit
The parameter definitions are as follows:
VIH Lower limit of input state change voltage
VIL Upper limit of input state change voltage
VOH Lower limit of output voltage
VOL Upper limit of output voltage
ROUT Comparator output resistance
SIGN -1 for inverting version, 1 for non-inverting version
See the circuit COMPDEMO for an example of the use of this macro.
473
Constant Power
The Constant Power macro is an implementation of a constant power load.
Figure 25-9 Constant Power macro circuit
It is implemented with a function current source whose expression is,
1/(Rlow/V(G1) + V(G1)/Power)
Rlow Minimum resistance of the load
Power The constant power in watts the load draws

CPFSK
The CPFSK is an implementation of a Continuous Phase Frequency Shift Keyer.
Figure 25-10 CPFSK macro circuit
AC Amplitude of the output waveform
FC Base carrier frequency
DF Frequency deviation of the modulated signal
VTHRESH Threshold voltage to differentiate between a binary one
and a binary zero
TYPE Type = 1 means high frequency occurs at a logic 1.
Type = 2 means high frequency occurs at a logic 0.
See the circuit CPFSK for an example of the use of this macro.
475
DC Motor
This is an electrical as well as mechanical model of a DC motor.
Figure 25-12 DC Motor macro circuit
LA Electrical inductance of the motor
RA Electrical resistance of the motor
KE Back EMF electromotive force constant
KT Torque electromotive force constant
JM Moment of inertia of the motor
BM Viscous friction constant
TL Load torque

DELAY
The delay macro provides a programmable signal delay.
VOut(t+delay) = VOut(t)
This function is implemented with the DELAY macro:
Figure 25-12 DELAY macro circuit
The single input parameter, DELAY, provides the specified delay through a trans-
mission line.
477
DIAC
The DIAC macro is based on an elaboration of the TRIAC macro and looks like
this:
Figure 25-13 DIAC macro circuit
A pair of input parameters, RS and VK define the series impedance and the volt-
age at which breakover occurs.
RS Series resistance
VK Breakover voltage
See the DIAC1 and DIAC2 circuits for an example of the use of this macro.

DIF
The differentiator is the inverse of the integrator. It provides an output which is a
scaled version of the time derivative of the input signal:
VOut(t) = scale d(VIn(t))/dt
This function is implemented with the DIF macro:
Figure 25-14 DIF macro circuit
The single input parameter, SCALE, multiplies or scales the derivative. This
particular implementation has a buffered output, allowing it to drive very low
impedance networks.
479
DIGPOT
The digital potentiometer macro is loosely based on the MAX5450-5455 series of
devices from Maxim. These devices are digital potentiometers with an up/down
interface consisting of two fixed resistors with 256 digitally controlled wiper con-
tacts (taps). There are three digital inputs that determine the tap point that will be
accessible to the wiper. These digital inputs are INC, CS, and U/D.
The INC input is the wiper increment control input. A high to low transition on
this input increments or decrements the wiper position depending on the states
of CS and U/D. The CS input is the chip select input. When this input is low,
the wiper position may be changed through INC and U/D. The U/D input is the
up/down control input. When this input is high, a high-to-low transition on INC
increments the wiper position, and when the input is low, a high-to-low transi-
tion on INC decrements the wiper position. If the wiper is at either the high or
low end of the resistor array, any increment or decrement in the direction of the
endpoint will not change the wiper position, so in other words, there is no wrap
around effect.
Finally, upon power-up the initial setting for the wiper is at midscale (tap 127).
This function is implemented with the DIGPOT macro:
Figure 25-15 DIGPOT macro circuit

DIV
Occasionally, system blocks require a function that divides two analog signals.
The desired function is:
VOut(t) = scale Va(t)/Vb(t)
This function is implemented with the DIV macro:
Figure 25-16 DIV macro circuit
The single input parameter passed to the macro by the calling circuit, SCALE,
multiplies or scales the ratio of the two input waveforms at the output.
481
F
This system block merely provides a convenient shape to house a general linear
transfer function, F(S). It is implemented with a Laplace LFVOFV source.
F(s) = VOut(s) /VIn(s)
This function is implemented with the F macro:
Figure 25-17 F macro circuit
The input parameter is an expression representing the complex frequency transfer

function.

FSK
This block provides a frequency-shift keyer encoder.
Figure 25-18 FSK macro circuit
The input parameters are as follows:
WMAG Magnitude of the output waveform
NC0 Number of cycles of the output waveform that will occur
in the duration of a single zero bit of the input waveform
NC1 Number of cycles of the output waveform that will occur
in the duration of a single one bit of the input waveform
TB Duration of a single bit in seconds
See the circuit FSK2 for an example of the use of this macro.
483
FWeighting
This is a frequency weighting filter macro
Figure 25-19 FWeighting macro circuit
TYPE Selects the frequency weighting scale.
Must be A, B, C, or D.
A-Weighting
.define Ka 7.39705e9
.define A (Ka*s^4)/((s+129.4)^2 * (s+676.7) * (s+4636) * (s+76655)^2)
B-Weighting
.define Kb 5.99185e9
.define B (Kb*s^3)/((s+129.4)^2 * (s+995.9) * (s+76655)^2)
C-Weighting
.define Kc 5.91797e9
.define C (Kc*s^2)/((s+129.4)^2 * (s+76655)^2)
D-Weighting
.define Kd 91204.32
.define D (Kd*s*(s^2+6532*s+4.0975e7))/((s+1776.3) * (s+7288.5) *
(s^2+21514*s+3.8836e8))
GYRATOR
The gyrator can be used to scale a resistive impedance. It can also transform an
inductive impedance to a capacitive impedance or a capacitive impedance to an
inductive impedance.
This function is implemented with two cross-referenced linear dependent VOFI

sources.
Figure 25-20 GYRATOR macro circuit
A single parameter, G, determines the impedance transformation as follows:
Req = 1/(R*G*G)
Leq = C/(G*G)
Ceq = L*G*G
See the circuit GYRTEST for an example of the use of this macro.
485
IDEAL_TRANS2
This is an ideal two port transformer. It provides a fixed voltage gain.
This function is implemented with two cross-referenced linear dependent sourc-

es. The gain is constant with frequency.
Figure 25-21 Ideal_Trans2 macro circuit
A single parameter, GAIN, determines the transformer voltage gain.
VOUT = VIN*GAIN
IOUT = IIN / GAIN
See the circuit IDEALTRANS for an example of the use of this macro.

IDEAL_TRANS3
This is an ideal three port transformer. It provides a fixed voltage gain from the
input port voltage to each of the two output ports. The gains are constant with
frequency.
This function is implemented with four cross-referenced linear dependent sourc-

es.
Figure 25-22 Ideal_Trans3 macro circuit
Two parameters, GAIN12 and GAIN13, determine the transformer port voltage
gains.
VOUT2 = VIN*GAIN12
IOUT2 = IIN / GAIN12
VOUT3 = VIN*GAIN13
IOUT3 = IIN / GAIN13
See the circuit IDEALTRANS for an example of the use of this macro.
487
INT
One of the most useful functions for system modeling is the integrator. Ideally
this function provides an output which is the integral of the input signal:
VOut(t) = vinit + scale

∫ V (t) dt
In
This function is implemented with the INT macro:
Figure 25-23 INT macro circuit
Two parameters are passed to the macro by the calling circuit: SCALE and VI-
NIT. SCALE multiplies the integral and VINIT provides its initial value. This
particular implementation has a buffered output, allowing it to drive very low
impedance networks. It also has a voltage limiting resistor. The resistor keeps the
output voltage finite when there is a DC voltage input. It should be large enough
to avoid placing any practical limit on the frequency response.
SCALE Scale factor to multiply the integral by
VINIT Initial value of the integral

INT_RESET
This macro is similar to the INT macro but includes a reset input. It is imple-
mented as follows:
Figure 25-24 INT_RESET macro circuit
The macro uses the same parameters as its parent, the INT macro.
SCALE Scale factor to multiply the integral by
VINIT Initial value of the integral
A 1 volt pulse on the Reset pin will reset the output voltage to zero. Reset is ac-
complished by the RON (1e-6) of the switch. The discharge time constant is
set by the 1e-6 resistance of the switch and the 10mF capacitor, requiring about
10*1e-6*1e-2 or 100nS for a complete discharge. Returning the Reset pin voltage
to 0 volts will then allow integration to restart.
489
Memristor
The memristor device is the fourth elementary passive element (in addition to
the resistor, capacitor, and inductor) that was first theorized about by Prof. Leon
Chua in 1971 and finally developed recently by HP Labs. It is based upon a
model developed by Zdenek Biolek, Dalibor Biolek, Vierka Biolek.
Figure 25-25 Memristor macro circuit
The Memristor can use either the Biolek or the Joglekar window function for
simulating the nonlinear dopant drift.

MONOSTABLE
This macro provides a monostable multivibrator function which creates a single
pulse of specified duration. It is useful in switch bounce applications.
Figure 25-27 MONOSTABLE macro circuit
This macro implements a monostable multivibrator using the Timer component

and several function sources. Its parameters are:
PWIDTH High state pulse width
VLOW Low state voltage level
VHIGH High state voltage level
THRESH Voltage threshold level at which the device will trigger
See the circuit MONOSTABLE_TEST.CIR for an example of the use of this

macro.
491
MONOSTABLE_RT
This macro is a retriggerable version of the monostable macro.
Figure 25-27 MONOSTABLE_RT macro circuit
This macro implements a retriggerable monostable multivibrator using the Timer

component and several function sources. Its parameters are:
PWIDTH High state pulse width
VLOW Low state voltage level
VHIGH High state voltage level
THRESH Voltage threshold level at which the device will trigger
There is an article describing the macro at:
http://www.spectrum-soft.com/news/summer2010/mono.shtm

MUL
Phase detectors and other system blocks require a function that multiplies two
analog signals. The desired function is:
VOut(t) = scale Va(t) Vb(t)
This function is implemented with the MUL macro:
Figure 25-28 MUL macro circuit
The single input parameter passed to the macro by the calling circuit, SCALE,
multiplies the product of the two input waveforms. The scaled product is provid-
ed at the output. This implementation is done with an NFV function source.
SCALE Scale factor to multiply the input by
493
NOISE
This macro implements a random noise generator to produce a noisy time-do-
main waveform.

Figure 25-29 NOISE macro circuit
TS Noise repetition rate. A new noise value is generated
every TS seconds.
VS The maximum value of the noise source. All values will
be in the range of 0.0 to VS volts.

Peak Detector
This behavioral model of an ideal peak detector eliminates any droop error that
might occur with diode detector implementations.
Figure 25-30 Peak Detector macro circuit
The implementation is done with a Sample and Hold whose expression is:
.if type==2
.define TypeExp V(In) < V(Out)
.else
.define TypeExp V(In) > V(Out)
.endif
Depending upon the type parameter specified, the conditional code (in the text
page of the macro) implements a positive or negative detector.
TYPE Type = 1 for a positive peak detector
Type = 2 for a negative peak detector
495
PHOTODIODE
The model consists of a diode paralleled by a voltage-controlled current source,
G1. The source creates the light current in the junction under control of the volt-
age at the IN pin. The voltage on this pin is an analog for the optical power in
watts incident on the detector.
Figure 25-31 PHOTODIODE macro circuit
RESPONSIVITY Sensitivity expressed in units of amps per watt. The
exact units do not matter so long as the control voltage
pin produces the expected current,
I = RESPONSIVITY*V(IN).
IDARK Current that flows across the junction when the diode is
reverse biased and there is no illumination. The diode IS
value parameter is set equal to IDARK.
RSERIES Value of a resistor in series with the diode
CJO Junction capacitance of the photo diode
BV Junction breakdown voltage
N Diode emission coefficient. It mainly affects the slope of
the dark current vs. temperature curve.

PHOTODIODE_R
This macro is similar to the PHOTODIODE macro except that RSHUNT is en-
tered instead of IDARK. Some manufacturers provide both values. If the photo
diode is to be operated in photoconductive mode, where the diode is reverse bi-
ased, then use the PHOTODIODE macro since accuracy of the IDARK value is
more important. If the photo diode is to be operated in photovoltaic mode, where
the diode operates near zero bias, then use this model since accuracy of the
RSHUNT value is more important.
Figure 25-32 PHOTODIODE_R macro circuit
RESPONSIVITY Sensitivity expressed in units of amps per watt. The
exact units do not matter so long as the control voltage
pin produces the expected current,
I = RESPONSIVITY*V(IN).
RSHUNT Incremental resistance of the diode near zero bias

RSERIES Value of a resistor in series with the diode
CJO Junction capacitance of the photo diode
BV Junction breakdown voltage
N Diode emission coefficient. It mainly affects the slope of
the dark current vs. temperature curve.
497
PID Controller
The PID (Proportional-Integral-Derivative) controller is a common feedback
mechanism used within closed loop control systems. The controller automatically
adjusts a variable to keep a specified measurement at a set point. It can be found
in temperature, flow, and position control applications. The macro employs other
math function macros, SUB, AMP, INT, DIF, and SUM.
Figure 25-33 PID Controller macro circuit
KP Proportional scale value
KI Integral scale value
KF Derivative scale value

POT
This macro implements a potentiometer using several resistor value expressions.
The two parameters determine the initial pot setting. Percent is in % so 70%
would be given as 70 and not .70.
To sweep the pot wiper arm, step the Value field of the macro resistor R1 from
the Stepping dialog box. The stepped value will override the default percent pa-
rameter. See POTDEMO.CIR for an example.

Figure 25-34 POT macro circuit
Potsize Maximum resistance value
Percent Percent of Potsize between PINA and PINB
See the circuit POTDEMO for an example of the use of this macro.
499
PSK
This is a phase-shift keyer.
Figure 25-35 PSK macro circuit
WMAG Magnitude of the output waveform
NC Number of cycles of the output waveform that will occur
in the duration of one bit of the input waveform
TB Duration of a single bit in seconds
See the circuit PSK2 for an example of the use of this macro.

PUT
This circuit is a macro model for a PUT, a Programmable Unijunction Transistor.
The macro equivalent circuit is as follows:
Figure 25-36 PUT macro circuit
IH DC holding current
IGT Gate trigger current
TON Turn-on time
VTMIN Minimum anode to cathode on-state voltage
VDRM Maximum repetitive peak off-state voltage
DVDT Critical rate of rise of off-state voltage
TQ Turn-off time
K1 Tweak factor for DVDT
K2 Tweak factor for TQ
See the circuit THY1 for an example of the use of this macro.
501
PWM
This macro models a pulse width modulator and is implemented with a ramp
voltage source and a function source with the expression:
Vo+Va*tanh(100k*(v(In,Ramp)))
Figure 25-37 PWM Controller macro circuit
MODHIGH High voltage value of the modulating waveform
MODLOW Low voltage value of the modulating waveform
MODFREQ Frequency of the modulating waveform
MODTYPE 1 = Trailing edge modulation
2 = Leading edge modulation
3 = Centered pulse modulation
PWMHIGH High voltage value of the PWM output
PWMLOW Low voltage value of the PWM output

PWM_T and PWM_NT
The macro circuit is based upon the March 7, 2002 EDN article, “Modular mac-
romodeling techniques for SPICE simulators”, which featured a very workable
model for the UC1845 pulse width modulator. We have implemented the model
in Micro-Cap and expanded it to model the full line of UCX84X PWMs. This
macro is the version with an integral T flip-flop. Its schematic looks like this:
Figure 25-38 PWM_T macro circuit
The PWM_NT macro is the same circuit without the T flip flop.
VOFF VDD voltage at which regulation ceases
VON VDD voltage at which regulation begins
See the circuit UC1845_BOOST circuit for an example of the use of this macro.
503
RELAY1
This relay model includes a user-specified coil resistance and inductance. The
coil current is sensed and converted to a voltage by H1 which drives a Schmitt
macro to provide hysteresis between the ION and IHOLD currents. The output of
the Schmitt drives a standard voltage controlled switch S1.
Figure 25-39 RELAY1 macro circuit
RCOIL Resistance of the relay coil
LCOIL Inductance of the relay coil
RON Closed resistance of the output switch
ROFF Open resistance of the output switch
ION Input current required to close the relay contacts
IOFF Input current required to hold the relay contacts closed
See the circuit RELAY for an example of the use of this macro.

RELAY2
This relay model includes a flux circuit and derives a magnetizing force from
the flux. It then algebraically sums the magnetizing, stop, friction and restoring
spring forces acting on the relay plunger to arrive at a net force which is inte-
grated once to get the plunger velocity and again to get the plunger position. This
plunger position directly controls the switch contacts.
Figure 25-40 RELAY2 macro circuit
RCOIL Resistance of the relay coil
LCOIL Inductance of the relay coil
RON Closed resistance of the output switch
ROFF Open resistance of the output switch
See the circuit RELAY for an example of the use of this macro.
505
RESONANT
This macro implements a resonant circuit with a resistor, capacitor, and inductor.
The L value is entered as an input parameter, LIN. The C value is computed from
the LIN and F0 input parameters. The resistor is computed from all three input
parameters. The implementation is as follows:

Figure 25-41 RESONANT macro circuit
LIN Value of inductance
F0 Center frequency
Q0 Quality factor

RTD
The RTD macro is a resistance temperature detector sensor commonly used for
temperature measurements. The response of the resistance versus temperature is
based upon the Callendar - Van Duesen equation which is as follows:
R(Temp) = Rnom * (1 + A * Temp + B * Temp^2 + C * Temp^3 * (100-Temp))
Figure 25-42 RTD macro circuit
RNOM Nominal resistance value at 0C
A Linear temperature coefficient
B Quadratic temperature coefficient
C Quartic temperature coefficient
507
SCHMITT
This circuit is a macro model for a Schmitt trigger, a circuit with a large number
of uses, including noise filtering, hysteresis, and level shifting. The circuit looks
like this:
Figure 25-43 SCHMITT macro circuit
X1 Lower limit of input state change voltage
X2 Upper limit of input state change voltage
Y1 Lower limit of output voltage
Y2 Upper limit of output voltage
ROUT Output Resistance of circuit
SIGN -1 for inverting version, 1 for non-inverting version
See the circuit OSC1 for an example of the use of this macro.

SCR
This circuit is a macro model for a silicon controlled rectifier or SCR. The macro
equivalent circuit is as follows:
Figure 25-44 SCR macro circuit
TON Turn-on time
TQ Turn-off time
See the circuit THY1 for an example of the use of this macro.
509
SLIP
The SLIP macro models hysteresis, or backlash. The output is zero within the slip
zone, -DX to +DX. Outside of the hysteresis zone, the output is proportional to
the input. The output is clipped to MAX.
This function is implemented with the SLIP macro:
Figure 25-45 SLIP macro circuit
A pair of input parameters, DX and MAX, define the slip zone and the maxi-
mum output level. The function is constructed with an NTVOFV Function table
source.
DX Slip value
MAX Maximum value

SNUBBER
The SNUBBER macro models a snubber diode.
This function is implemented with the SNUBBER macro:
Figure 25-46 SNUBBER macro circuit
A pair of input parameters, CSNUB and RSNUB, specify the parallel RC net-
work that provides the energy absorbing parasitics.
CSNUB Parallel capacitance value
RSNUB Parallel resistance value
511
SPARKGAP
This is a macro circuit model for a spark gap arrestor.
Figure 25-47 SPARKGAP macro circuit
VTHRES Voltage at which the spark-gap strikes
VARC Voltage across the spark-gap once struck
ISUS Sustaining current under which the arc is stopped
RNEG Negative resistance once struck
LPL Lead inductance
RPL Flux loss associated with LPL
CPAR Gap capacitance
CARC Arc capacitance
See the circuit SPARK for an example of the use of this macro.

SUB
A common requirement is to subtract two analog signals. The desired function is:
VOut(t) = ka Va(t) - kb Vb(t)
This function is implemented with the SUB macro:
Figure 25-48 SUB macro circuit
The two input parameters, KA and KB, scale each input. The scaled input signals
are then subtracted to produce the output. This implementation is done with an
NFV function source.
KA Multiplier of input A
KB Multiplier of input B

513
SUM
Many system simulations call for a function to perform the analog addition of
two signals. The desired function is:

VOut(t) = ka Va(t) + kb Vb(t)
This function is implemented with the SUM macro:
Figure 25-49 SUM macro circuit
The two input parameters, KA and KB, scale each input. The scaled input signals
are then added to produce the output.

SUM3
A triple summer is sometimes convenient. The desired function is:
VOut(t) = ka Va(t) + kb Vb(t)+ kc Vc(t)
This function is implemented with the SUM3 macro:
Figure 25-50 SUM3 macro circuit
The three input parameters, KA, KB, and KC, multiply each input. The three
scaled input signals are then added to produce the output. This implementation is
done with an NFV function source.
KC Multiplier of input C
See the circuit SYSTEM1 for an example of the use of the SUM3 macro.
515
TRIAC
This circuit is a macro model for a TRIAC. The macro circuit is as follows:
Figure 25-51 TRIAC macro circuit
TON Turn-on time
TQ Turn-off time
See the circuit THY1 for an example of the use of the TRIAC macro.
TRIGGER6
This circuit is a macro model for a thyristor gate trigger circuit. The macro equiv-
alent circuit is as follows:
Figure 25-52 TRIGGER6 macro circuit
There are no parameters for the macro.
See the circuit RECTIFIER_45.CIR for an example of the use of this macro.
517
TRIODE
The TRIODE is a macro model of a vacuum triode device. Its equivalent circuit
is as shown below.
Figure 25-53 TRIODE macro circuit
The Triode macro is implemented with a 3/2 power function voltage-controlled

current source. The K, MU, CGP, CGC, and CPC values are passed as parameters
when the macro is used in a circuit.
K Tube constant k
MU Tube constant mu
CGP Grid to plate capacitance
CGC Grid to cathode capacitance
CPC Plate to cathode capacitance
See the circuit F4 for an example of the use of the TRIODE macro.

VCO
A voltage-controlled oscillator, or VCO, is an oscillator whose instantaneous fre-
quency is dependent upon a time-varying voltage. The VCO macro has a voltage
whose time dependence is given by:
VOut(t) = vp cos (2�(f0t + kf

∫ V (t) dt))
in
In this form, f0 is the center frequency and kf is the frequency sensitivity in

Hz/volt. This form of linear VCO is easily implemented as a macro:
Figure 25-54 VCO macro circuit
The VCO uses a nonlinear function source, which uses the output of an integrator
stage to control the frequency. The input parameters specify the magnitude, cen-
ter frequency, and the frequency sensitivity.
VP Peak magnitude of the output signal
F0 Center frequency
KF Frequency sensitivity in Hz/Volt
See the circuit F1 for an example of the use of this macro.
519
WIDEBAND
This is a wideband model of a transformer.
Figure 25-55 WIDEBAND macro circuit
RS Primary series resistance
N Number of turns
FL Low frequency breakpoint
FH High frequency breakpoint

XTAL
XTAL is a macro model of a crystal. Its circuit is as shown below.
Figure 25-56 XTAL macro circuit
The XTAL macro is implemented with a standard tank circuit model for crystals.
F0 Center frequency
R Resistance
Q Crystal quality factor
For examples of how to use the macro, see the circuit XTAL1, which shows a
crystal oscillator application.
521
555
The 555 is a model of the ubiquitous 555 timer circuit. Its circuit is as follows:
Figure 25-57 555 macro circuit
The 555 uses several nonlinear function sources to monitor the THRES and
TRIG input voltage values. When they cross a certain threshold the sources
switch to a high or low level and charge capacitors which drive the R and S in-
puts of a RS flip-flop. The flip-flop drives the output and an NPN which provides
a discharge path. There are no input parameters.
For examples of how to use the macro, see the circuits 555MONO, which shows
a monostable application, and 555ASTAB, which demonstrates how to use the
555 in an astable application.
You can change the power supply by using a .PARAM statement as follows:
.PARAM V555_VDD=<VDD value desired>

.PARAM V555_VSS=<VSS value desired>
The statement may be placed in the text area, the grid text, the User Definitions,
or an appropriate text library file.
Chapter 26 Analog Devices
What's in this chapter

This chapter covers the parameter syntax, model statements, model parameters,
and model equations used by each of the Micro-Cap analog devices.
Model statements describe the model parameters for the more complex devices.
Model parameters are the numeric values to be used in the model equations.
They are obtained from model statements or binary model libraries (*.LBR).
Model equations use the numeric model parameter values in a set of mathemati-
cal equations that describe three aspects of a device's electrical behavior:
1. The static relationship between terminal currents and branch voltages.

2. Energy storage within the device.
3. Noise generation.
In the chapter that follows, each component is described in terms of:

1. SPICE parameters and / or schematic attribute formats.
2. Model statement format (if any).
3. Model parameters (if any).
4. The electrical model in terms of its schematic and model equations.
If the SPICE parameter format is not given, the component is available for use
only in schematic circuits and not in SPICE text file circuits.
Devices usually have PACKAGE, COST, SHAPEGROUP, and POWER attri-

butes. The PACKAGE attribute specifies the package to be used for PCB netlists.
SHAPEGROUP specifies the shape group to be searched first for the indicated
shape name. COST and POWER attributes specify the cost and power contribu-
tions for the Bill of Materials report.
523
References
This chapter provides a comprehensive guide to the device models used in Mi-
cro-Cap. It does not, however, teach the principles of analog or digital simulation,
or the operation of semiconductor devices. The following references provide a
good introduction to these and related topics:
Circuit design
1. Paul R. Gray, and Robert G. Meyer
Analysis and Design of Analog Integrated Circuits
John Wiley and Sons, 1977, 1984
2. David A. Hodges, and Horace G. Jackson.

Analysis and Design of Digital Integrated Circuits
McGraw-Hill, 1983
3. Richard S. Muller and Theodore I. Kamins

Device Electronics for Integrated Circuits
John Wiley and Sons, 1977, 1986
4. Adel Sedra, Kenneth Smith

Microelectronic Circuits
Oxford, 1998
5. John P. Uyemura
Introduction to VLSI Circuits and Systems
John Wiley & Sons Inc, 2002
6. Martin S. Roden and Gorden L. Carpenter

Electronic Design
Discovery Press, 2002
Analog simulation
7. A. Ruehli, Editor
Circuit Analysis, Simulation and Design
Advances in CAD for LSI, Part 3, Vol 1
North-Holland 1986
8. J. Vlach, K. Singhal
Computer Methods for Circuit Analysis and Design
Van Nostrand Reinhold 1994
524 Chapter 26: Analog Devices
9. Kenneth Kundert
Designers Guide to SPICE & Spectre
Kluwer Academic Publishers, 1995
10. William J. McCalla

Fundamentals of Computer-Aided Circuit Simulation
Kluwer Academic Publishers, 1988
12. Andrei Vladimiresescu and Sally Liu

The Simulation of MOS Integrated circuits using SPICE2
University of California, Berkeley
Memorandum No. UCB / ERL M80/7
12. Lawrence Nagel.

SPICE2: A Computer Program to Simulate Semiconductor Circuits
University of California, Berkeley,
Memorandum No. ERL - M520
13. Andrei Vladimirescu

The SPICE Book
John Wiley & Sons, Inc., First Edition, 1994. ISBN# 0-471-60926-9
Device modeling
14. H. Statz, P. Newman, I. W. Smith, R. A. Pucel, and H. A. Haus
GaAsFET Device and Circuit Simulation in SPICE
IEEE Transactions on Electron Devices, ED - 34, 160-169 (1987)
15. Graeme R. Boyle, Barry M. Cohn, Donald O. Petersen, and James E.

Solomon
Macromodeling of Integrated Circuit Operational Amplifiers
IEEE Journal of Solid-State Circuits, Vol. SCV-9 No. 6 (1974)
16. W.R Curtice

A MESFET model for use in the design of GaAs integrated circuits
IEEE Transactions on Microwave Theory and Techniques
MTT - 28,448-456 (1980)
17. Ian Getreu

Modeling the Bipolar Transistor
Tektronix, 1979. The original Tektronix version is out of print but it is
available at lulu.com/iangetreu.
525
18. Liu, William
MOSFET Models for SPICE Simulation Simulation including
BSIM3v3 and BSIM4
Wiley-Interscience ISBN: 0-471-39697-4
19. MOSPOWER Applications

copyright 1984, Signetics, Inc.
20. Bing Jay Sheu

MOS Transistor Modeling and characterization for Circuit Simulation
University of California, Berkeley Memo No. UCB / ERL M85/85
21. Yannis P. Tsividis

Operation and Modeling of the MOS Transistor
McGraw-Hill, 1987
22. Paolo Antognetti, and Giuseppe Massobrio

Semiconductor Device Modeling with SPICE.
McGraw-Hill, 1988
23. D. C. Jiles, and D. L. Atherton

Theory of Ferromagnetic Hysteresis
Journal of Magnetism and Magnetic Materials, 61, 48 (1986)
24. Sally Liu

A Unified CAD Model for MOSFETs
Memorandum No. UCB / ERL M81/31
25. Y. Cheng, M. Chan, K. Hui, M. Jeng, Z. Liu, J. Huang, K. Chen, J. Chen, R.

Tu, P. Ko, C. Hu
BSIM3v3.1 Manual
Department of Electrical Engineering and Computer Sciences
26. Daniel Foty

MOSFET Modeling with SPICE Principles and Practice
Prentice Hall, First Edition, 1997. ISBN# 0-13-227935-5

Filters
27. Lawrence P. Huelsman
Active and Passive Filter Design, An Introduction
McGraw-Hill, 1993
28. Kendall L. Su
Analog Filters
Chapman & Hall, 1996
29. Arthur B. Williams

Electronic Filter Design Handbook
McGraw-Hill, 1981
Switched-mode power supplies

30. Christophe Basso
Switch-Mode Power Supplies, SPICE Simulations and Practical Designs
McGraw-Hill, 2008
31. Steven M. Sandler

SMPS Simulation with SPICE 3
McGraw Hill, First Edition, 1997. ISBN# 0-07-913227-8
General SPICE modeling

32. Ron Kielkowski
SPICE - Practical Device Modeling
McGraw Hill 1995. ISBN# 0-07-912524-1
33. Ron Kielkowski

Inside SPICE - Overcoming the Obstacles of Circuit Simulation
McGraw Hill 1993. ISBN# 0-07-912525-X
34. Connelly and Choi

Macromodeling with SPICE
Prentice Hall 1992. ISBN# 0-13-544941-3
RF circuits
35. Vendelin, Pavio, and Rhoda
Microwave Circuit Design
Wiley-Interscience, 1990
527
Animated analog bar
Schematic format
PART attribute
<name>
Example
BAR1
LOW attribute
<low_value>
Example
1
HIGH attribute
<high_value>
Example
2
The animated analog bar produces a colored bar whose height tracks its input
voltage. When the voltage equals low_value the bar is at a minimum. When the
voltage equals high_value the bar is at a maximum.

Animated analog LED
Schematic format
PART attribute
<name>
Example
LED2
COLOR attribute
<color name> , <on voltage> , <on current> , [rs] , [cjo]
Example
Red,1.7,0.015,250m,30p
This device is a light emitting diode whose color appears when the diode voltage
equals or exceeds <on voltage>. It is modeled as a conventional diode with the
model parameters RS set to [rs] and CJO set to [cjo]. Other model parameters are
adjusted so that the diode voltage and current match the specified <on voltage>
and <on current>. [rs] defaults to .500. [cjo] defaults to 20pF.
The choice of color is controlled by the user-selected palette color associated

with color name. To change the color name or actual on-screen color, double
click on a LED, then click on the COLOR attribute, then on the LED Color but-
ton in the Attribute dialog box. This invokes the LED Color dialog box which lets
you edit the existing LED COLOR attributes or add additional ones. The follow-
ing are supplied with the original Micro-Cap library.
YELLOW, 2.0, 15m, 500m, 20pF

GREEN, 2.1, 15m, 500m, 20pF
BLUE, 3.4, 12m, 500m, 20pF
RED, 1.7, 15m, 500m, 20pF
529
Animated DC motor
Schematic format
PART attribute
<name>
Example
MOTOR1
RPSPV attribute
<revs_per_sec_per_volt>
Example
1
RMOTOR attribute
<motor_resistance>
Example
50
LMOTOR attribute
<motor_inductance>
Example
1
The animated DC motor rotates at an angular velocity controlled by the instanta-

neous input voltage. The rate of rotation is:
Revolutions per second = (input voltage ) * ( revs_per_sec_per_volt )
The motor is modeled electrically as a resistor of value motor_resistance in series

with an inductor of value motor_inductance.
The voltage on the Velocity pin is equal to the velocity in revolutions per second.
Warning: If the rotation rate is too high the display will appear to be still due to a
rapid rotation rate. A rate of 10 RPS is a good working maximum.

Animated DPST, SPDT, and SPST switches
Schematic format
PART attribute
<name>
SPDT switch
STATE attribute
<UP | DOWN>
SPST or DPST switch

STATE attribute
<OPEN | CLOSED>
RON attribute
<ron>
Example
.001
ROFF attribute
<roff>
Example
1E12
GROUP attribute
<group_name | NOT group_name>
Examples
Group1
NOT Group22
Double clicking in the switch area toggles the state of these switches between
open and closed (SPST and DPST), or between up and down (SPDT). The
switches are modeled with a simple resistor whose on and off values are set by
ron (default=.001) and roff (default=1E15). Double clicking just outside of the
switch area accesses the Attribute dialog box for the device. A click on a switch
with a GROUP attribute causes all others of the same attribute to behave the
same. If group_name is preceded with the reserved word NOT, the switch be-
haves oppositely to that of switches with group_name.
531
Animated meter
Schematic format
PART attribute
<name>
Example
METER1
LOW attribute
<low>
Example
-10
HIGH attribute
<high>
Example
10
SCALE attribute
< T | G | MEG | K | m | u | n | p | f | none >
Example
m
AUTOSCALE attribute
<ON | OFF>
Example
OFF
ANALOG OR DIGITAL attribute

<ANALOG | DIGITAL>
Example
ANALOG
AMPS OR VOLTS attribute

<AMPS | VOLTS>
Example
VOLTS
INPUT_RESISTANCE attribute
<rin>
Example
1E9
The animated meter is a voltage or current meter whose display can be either
analog or digital.
Click on the Digital or Analog text on the meter shape to select either an analog
or digital display.
Click on the Amps or Volts text on the meter shape to select either an ammeter or
a voltmeter.
Click on the Autoscale text on the analog meter shape to toggle between automat-
ic scaling and manual scales. In manual scaling the lower limit is set by low and
the higher limit is set by high. The digital meter mode always uses autoscaling.
The SCALE attribute is effective only when AUTOSCALE is OFF and the ANA-
LOG meter option is selected. With an input voltage of 8000, a scale of k would
read 8 when the analog range is 10 (LOW=0 and HIGH=10).
When the meter is in voltmeter mode, rin is added across the input terminals. If
unspecified, the value of rin is 1/GMIN.
When the meter is in ammeter mode, the input resistance is .001 ohms.
533
Animated relay
Schematic format
PART attribute
<name>
Example
RELAY1
LIN attribute
<input_inductance>
Example
1E-4
RIN attribute
<input_resistance>
Example
100
ION attribute
<input_on_current>
Example
50m
IOFF attribute
<input_off_current>
Example
15m
RON attribute
<on_resistance>
Example
1
ROFF attribute
<off_resistance>
Example
1E9
BIDIRECTIONAL attribute
< YES | NO >
Example
NO
The animated relay is an SPST relay that responds dynamically to changes in in-
put current. It behaves like a smooth transition, current-controlled W switch. The
relay input is modeled as an inductor of value input_inductance and a resistor of
value input_resistance.
Model Equations
If BIDIRECTIONAL attribute is set to NO

IC = Current flow into the Plus Input
else
IC = ABS(Current flow into the Plus Input )
LM = Log-mean of resistor values = ln((RON•ROFF)1/2)

LR = Log-ratio of resistor values = ln(RON/ROFF)
IM = Mean of control currents = (ION+IOFF)/2
ID = Difference of control currents = ION-IOFF
RS = Switch output resistance
If ION > IOFF

If IC >= ION
RS = RON
If IC <= IOFF
RS = ROFF
If IOFF < IC < ION
RS = exp(LM + 3•LR•(IC-IM)/(2•ID) - 2•LR•(IC-IM)3/ID3)
If ION < IOFF

If IC <= ION
RS = RON
If IC >= IOFF
RS = ROFF
If IOFF > IC > ION
RS = exp(LM - 3•LR•(IC-IM)/(2•ID) + 2•LR•(IC-IM)3/ID3)
535
Animated traffic light
Schematic format
PART attribute
<name>
Example
LIGHT1
TURN-ON VOLTAGES attribute

<on_voltage>
Example
2
The animated traffic light is designed to simulate a typical traffic light. There are
three lights, red, yellow, and green. Each of the three lights is illuminated when
its input control pin voltage exceeds on_voltage.

Animated digital switch
Schematic format
PART attribute
<name>
Example
U1
I/O MODEL attribute

<I/O model name>
Example
IO_STD
IO_LEVEL attribute
<interface subckt select value>
Example
1
POWER NODE attribute

<digital power node>
Example
$G_DPWR
GROUND NODE attribute

<digital ground node>
Example
$G_DGND
STATE attribute
<output_state>
Example
1
The animated digital switch generates a 1 or 0 at its digital output. To change the
state, double-click on the switch body.
537
Animated digital LED
Schematic format
PART attribute
<name>
Example
U1
I/O MODEL attribute

<I/O model name>
Example
IO_STD
IO_LEVEL attribute
Example
1

Example
$G_DPWR

Example
$G_DGND
It is designed to represent the display of a light emitting diode. It has a single in-
put pin. Depending on the digital state or the analog voltage at the input pin, the
LED will be lit with a different color on the schematic. The colors the LED uses
are defined in the Color/Font page of the circuit's Properties dialog box. In this
page, there is a list of digital states that have a corresponding color.

Animated seven segment display
Schematic format
PART attribute
<name>
Example
U1
I/O MODEL attribute

<I/O model name>
Example
IO_STD
IO_LEVEL attribute
Example
1

Example
$G_DPWR

Example
$G_DGND
ON STATE attribute
<HIGH | LOW>
Example
HIGH
Each segment in the display is lit if its control pin is at the ON STATE, otherwise
it is unlit. ON STATE can be specified as either HIGH or LOW.
539
Battery
Schematic format
PART attribute
<name>
Example
V1
VALUE attribute
<value>
<value>[LOT[/#][/GAUSS|/UNIFORM|WCASE]=<tolvalue>[%]]
Examples
10 LOT=10%
3 LOT/1/Gauss=10%
5.5
The battery produces a constant DC voltage. It is implemented internally as the

simplest form of SPICE independent voltage source. Its main virtue lies in its
simplicity.
The battery provides a constant voltage value. If you need a voltage source that is
dependent on other circuit variables or time, use one of the dependent sources or
the function source (NFV).
A battery can be toleranced by adding the Lot statement to the VALUE attribute.
For example, if a 10V battery was to have a 20% tolerance, the VALUE attribute
would be defined as:
10 Lot=20%

Bipolar transistor (Standard Gummel Poon Level=1)
SPICE format
Syntax
Q<name> <collector> <base> <emitter> [substrate]
+<model name> [area] [OFF] [IC=<vbe>[,vce]][M=<multiplier>]
Examples
Q1 5 7 9 2N3904 1 OFF IC=0.65,0.35 M=20
Q3 C 20 OUT [SUBS] 2N3904
Schematic format
PART attribute
<name>
Examples
Q1
BB1
VALUE attribute
[area] [OFF] [IC=<vbe>[,vce]][M=<multiplier>]
Example
1.5 OFF IC=0.65,0.35
MODEL attribute
<model name>
Example
2N2222A
The initialization, 'IC=<vbe>[,vce]', assigns initial voltages to the junctions in

transient analysis if no operating point is done (or if the UIC flag is set). Area
multiplies or divides parameters as shown in the model parameters table. The
OFF keyword forces the BJT off for the first iteration of the operating point.
<multiplier> is the number of parallel devices.
Model statement forms

.MODEL <model name> NPN ([model parameters])
.MODEL <model name> PNP ([model parameters])
.MODEL <model name> LPNP ([model parameters])
541
Examples
.MODEL Q1 NPN (IS=1E-15 BF=55 TR=.5N)
.MODEL Q2 PNP (BF=245 VAF=50 IS=1E-16)
.MODEL Q3 LPNP (BF=5 IS=1E-17)
Name Parameter Unit/s Default Area

IS Saturation current A 1E-16 *
BF Ideal maximum forward beta 100.0
NF Forward current emission coefficient 1.00
VAF Forward Early voltage V ∞
IKF High-current roll-off corner A ∞ *
ISE BE leakage saturation current A 0.00 *
NE BE leakage emission coefficient 1.50
BR Ideal maximum reverse beta 1.00
NR Reverse current emission coefficient 1.00
VAR Reverse Early voltage V ∞
IKR BR high-current roll-off corner A ∞ *
ISC BC leakage saturation current A 0.00 *
NC BC leakage emission coefficient 2.00
NK High current rolloff coefficient 0.50
ISS Substrate pn saturation current A 0.00 *
NS Substrate pn emission coefficient 1.00
GAMMA Epitaxial region doping factor 1e-12
QUASIMOD Quasi-saturation temperature flag 0.00
RC Epitaxial region resistance Ω 0.00
VO Carrier mobility knee voltage V 10.00
RCO Epitaxial region resistance Ω 0.00 /
RE Emitter resistance Ω 0.00 /
RB Zero-bias base resistance Ω 0.00 /
IRB Current where RB falls by half Α ∞ *
RBM Minimum RB at high currents Ω RB /
TF Ideal forward transit time S 0.00
TR Ideal reverse transit time S 0.00
XCJC Fraction of BC dep. cap. to internal base 1.00
MJC BC junction grading coefficient 0.33
VJC BC junction built-in potential V 0.75
CJC BC zero-bias depletion capacitance F 0.00 *
MJE BE junction grading coefficient 0.33
VJE BE junction built-in potential V 0.75
CJE BE zero-bias depletion capacitance F 0.00 *
MJS CS junction grading coefficient 0.00
VJS CS junction built-in potential V 0.75
CJS Junction zero-bias capacitance F 0.00 *
VTF Transit time dependence on VBC V ∞

Name Parameter Units Default Area
ITF Transit time dependence on IC A 0.00 *
XTF Transit time bias dependence coefficient 0.00
PTF Excess phase 0.00
XTB Temperature coefficient for betas 0.00
EG Energy gap eV 1.12
XTI Saturation current temperature exponent 3.00
KF Flicker-noise coefficient 0.00
AF Flicker-noise exponent 1.00
FC Forward-bias depletion coefficient 0.50
T_MEASURED Measured temperature °C
T_ABS Absolute temperature °C
T_REL_GLOBAL Relative to current temperature °C
T_REL_LOCAL Relative to AKO temperature °C
TRE1 RE linear temperature coefficient °C-1 0.00
TRE2 RE quadratic temperature coefficient °C-2 0.00
TRB1 RB linear temperature coefficient °C-1 0.00
TRB2 RB quadratic temperature coefficient °C-2 0.00
TRM1 RBM linear temperature coefficient °C-1 0.00
TRM2 RBM quadratic temperature coefficient °C-2 0.00
TRC1 RC linear temperature coefficient °C-1 0.00
TRC2 RC quadratic temperature coefficient °C-2 0.00

BJT model equations
Figure 26-1 Bipolar transistor model
543
Definitions
The model parameters IS, IKF, ISE, IKR, ISC, ISS, IRB, CJC, CJE, CJS, and ITF
are multiplied by [area] and the model parameters RC, RE, RB, and RBM are
divided by [area] prior to their use in the equations below.
T is the device operating temperature and Tnom is the temperature at which the
model parameters are measured. Both are expressed in degrees Kelvin. T is set
to the analysis temperature from the Analysis Limits dialog box. TNOM is de-
termined by the Global Settings TNOM value, which can be overridden with a
.OPTIONS statement. T and Tnom may both be customized for each model by
specifying the parameters T_MEASURED, T_ABS, T_REL_GLOBAL, and
T_REL_LOCAL. For more details on how device operating temperatures and
Tnom temperatures are calculated, see the .MODEL section of the Command
Statements chapter.
The substrate node is optional and, if not specified, is connected to ground. If, in
a SPICE file, the substrate node is specified and is an alphanumeric name, it must
be enclosed in square brackets.
The model types NPN and PNP are used for vertical transistor structures and the
LPNP is used for lateral PNP structures. The isolation diode DJ and capacitance
CJ are connected from the substrate node to the internal collector node for NPN
and PNP model types, and from the substrate node to the internal base node for
the LPNP model type.
When adding new four terminal BJT components to the Component library, use
NPN4 or PNP4 for the Definition field.
When a PNP4 component is placed in a schematic, the circuit is issued an LPNP

model statement. If you want a vertical PNP4, change the LPNP to PNP.
VT = k•T/q
VBE = Internal base to emitter voltage
VBC = Internal base to collector voltage
VCS = Internal collector to substrate voltage
In general, X(T) = Temperature adjusted value of parameter X
Temperature effects
EG(T) = EG - .000702•T2/(T+1208)
IS(T) = IS•e((T/Tnom-1)•EG/(VT))•(T/Tnom)(XTI)
ISE(T) = (ISE/(T/Tnom)XTB)•e((T/Tnom-1)•EG/(NE•VT))•(T/Tnom)(XTI/NE)
ISC(T) = (ISC/(T/Tnom)XTB)•e((T/Tnom-1)•EG/(NC•VT))•(T/Tnom)(XTI/NC)
BF(T) = BF•(T/Tnom)XTB
BR(T) = BR•(T/Tnom)XTB
VJE(T) = VJE•(T/Tnom)-3•VT•ln((T/Tnom))-EG(Tnom)•(T/Tnom)+EG(T)
VJC(T) = VJC•(T/Tnom)-3•VT•ln((T/Tnom))-EG(Tnom)•(T/Tnom)+EG(T)
VJS(T) = VJS•(T/Tnom)-3•VT•ln((T/Tnom))-EG(Tnom)•(T/Tnom)+EG(T)
CJE(T) = CJE•(1+MJE•(.0004•(T-Tnom) + (1 - VJE(T)/VJE)))
CJC(T) = CJC•(1+MJC•(.0004•(T-Tnom) + (1 -VJC(T)/VJC)))
CJS(T) = CJS•(1+MJS•(.0004•(T-Tnom) + (1 - VJS(T)/VJS)))
Current equations
Q1 = 1/ (1 - VBC/VAF - VBE/VAR)
Q2 = IS(T)•(e(VBE/(NF•VT))-1)/IKF + IS(T)•(e(VBC/(NR•VT))-1)/IKR
QB = Q1•(1+(1+4•Q2)NK ) / 2
Current source value
ICT = IS(T)•(e(VBE/(NF•VT))-1)/QB - IS(T)•(e(VBC/(NR•VT))-1)/QB
Base emitter diode current

IBE = ISE(T)•(e(VBE/(NE•VT))-1)+IS(T)•(e(VBE/(NF•VT))-1)/BF(T)
Base collector diode current

IBC = ISC(T)•(e(VBC/(NC•VT))-1)+IS(T)•(e(VBC/(NR•VT))-1)/QB/BR(T)
Base terminal current

IB = IBE + IBC
IB = IS(T)•(e(VBE/(NF•VT))-1)/BF(T)+ISE(T)•(e(VBE/(NE•VT))-1)+
IS(T)•(e(VBC/(NR•VT))-1)/BR(T)+ISC(T)•(e(VBC/(NC•VT))-1)
Collector terminal current

IC = IS(T)•(e(VBE/(NF•VT))- e(VBC/(NR•VT)))/QB
- IS(T)•(e(VBC/(NR•VT))-1)/BR(T)- ISC(T)•(e(VBC/(NC•VT)) -1)
Emitter terminal current

IE = IS(T)•(e(VBE/(NF•VT))-e(VBC/(NR•VT)))/QB
+IS(T)•(e(VBE/(NF•VT))-1)/BF(T)+ISE(T)•(e(VBE/(NE•VT))-1)
Capacitance equations
Base emitter capacitance
GBE = base emitter conductance = ð(IBE) / ð(VBE)
545
If VBE ≤ FC•VJE(T)
CBE1 =CJE(T)•(1 - VBE/VJE(T))-MJE
Else
CBE1 = CJE(T)•(1 - FC)-(1+MJE) • (1 - FC•(1+MJE)+MJE•VBE/VJE(T))
R = IS(T)•(e(VBE/(NF•VT))-1)/(IS(T)•(e(VBE/(NF•VT))-1)+ITF)
CBE2 = GBE•TF•(1+XTF•(3•R2-2•R3)•e(VBC/(1.44•VTF)))
CBE = CBE1+CBE2
Base collector capacitances

GBC = base collector conductance = ð(IBC) / ð(VBC)
If VBC ≤ FC•VJC(T)
C = CJC(T)•(1 - VBC/VJC(T))-MJC
Else
C = CJC(T)•(1 - FC)-(1+MJC) • (1 - FC•(1+MJC)+MJC•VBC/VJC(T))
CJX = C•(1- XCJC)

CBC = GBC•TR + XCJC•C
Collector substrate capacitance

If VCS ≤ 0
CJ = CJS(T)•(1 - VCS/VJS(T))-MJS
Else
CJ = CJS(T)•(1 - FC)-(1+MJS) • (1 - FC•(1+MJS)+MJS•VCS/VJS(T))
Noise
RE, RB, and RC generate thermal noise currents.
Ie2 = (4 • k • T) / RE
Ib2 = (4 • k • T) / RB
Ic2 = (4 • k • T) / RC
Both the collector and base currents generate frequency-dependent flicker and
shot noise currents.
Ic2 = 2 • q • Ic + KF • ICBAF / Frequency

Ib2 = 2 • q • Ib + KF • IBEAF / Frequency
where
KF is the flicker noise coefficient
AF is the flicker noise exponent
Bipolar transistor (Philips Mextram Level = 2 or 21)
SPICE format
Syntax
Q<name> <collector> <base> <emitter> [substrate] [thermal]
+<model name>
+ MULT=<no_parallel>
Examples
Q1 1 2 3 MM1
Q2 1 2 3 4 MM2
Q3 1 2 3 4 5 MM3
Q4 1 2 3 4 5 MM3 MULT=20
Schematic format
PART attribute
<name>
Examples
Q1
BB1
VALUE attribute
MULT=<no_parallel>
Example
MULT=100
MODEL attribute
<model name>
Example
MLX
Model statement form

.MODEL <model name> NPN (LEVEL=2 [model parameters])
.MODEL <model name> NPN (LEVEL=21 [model parameters])
Example of standard model

.MODEL MODN NPN (LEVEL=2 ...)
547
Example of thermal model
.MODEL MODP PNP (LEVEL=21 ...)
Mextram is an advanced compact model for bipolar transistors suitable for

double-poly, SiGe processes, high-voltage power devices, and lateral NPN-tran-
sistors in LDMOS technology. It provides an accurate description of both the
forward and reverse modes of operation for digital and analog circuits. It realisti-
cally handles quasi-saturation, avalanche multiplication, epitaxial collector layer
resistance and charge storage, Early voltage bias dependence, inactive region
modelling, current crowding, and conductivity modulation of base resistance.
The equivalent circuit of the Mextram model is as follows:
Figure 26-2 Mextram bipolar transistor model

Model parameters for Mextram bipolar transistor level 2/21
Name Default Description

LEVEL 2 Level = 2 for standard, or 21 for thermal model
AB 1 Temperature coefficient of the resistivity of the base
AC 2 Buried layer resistivity temperature coefficient
AE 0 Emitter resistivity temperature coefficient
AEPI 2.5 Epilayer resistivity temperature coefficient
AEX 620m Extrinsic base resistivity temperature coefficient
AF 2 Exponent of the flicker-noise
AQBO 300m Temperature coefficient of the zero-bias base charge
AS 1.58 For a closed buried layer: As=Ac. For an open buried
layer: As=Aepi
AXI 300m Smoothness parameter for the onset of quasi-saturation
BF 215 Ideal forward current gain
BRI 7 Ideal reverse current gain
CBCO 0 Collector-base overlap capacitance
CBEO 0 Emitter-base overlap capacitance
CJC 78f Zero-bias collector-base depletion capacitance
CJE 73f Zero-bias emitter-base depletion capacitance
CJS 315f Zero-bias collector-substrate depletion capacitance
DAIS 0 Parameter for fine tuning of temperature dependence of
collector-emitter saturation current
DEG 0 Bandgap difference over the base
DTA 0 Difference between the local ambient and global
ambient temperature
DVGBF 50m Bandgap voltage difference of forward current gain
DVGBR 45m Bandgap voltage difference of reverse current gain
DVGTE 50m Bandgap voltage difference of emitter stored charge
EXAVL 0 Flag for extended modelling of avalanche currents
EXMOD 1 Flag for extended modelling of reverse current gain
EXPHI 1 Flag for the distributed high-frequency effects in
transient analysis
IBF 2.7f Saturation current of the non-ideal forward base current
IBR 1f Saturation current of the non-ideal reverse base current
IHC 4m Critical current for velocity saturation in the epilayer
IK 100m Collector-emitter high injection knee current
IKS 250u Base-substrate high injection knee current
IS .022f Collector-emitter saturation current
ISS .048f Base-substrate saturation current
549

KAVL 0 Switch for white noise contribution due to avalanche
KF 20p Flicker-noise coefficient of the ideal base current
KFN 20p Flicker-noise coefficient of the non-ideal base current
MC 500m Coefficient for the current modulation of the collector
base depletion capacitance
MLF 2 Non-ideality factor of the non-ideal forward base current
MTAU 1 Non-ideality factor of the emitter stored charge
MULT 1 Number of devices in parallel
PC 500m Collector-base grading coefficient
PE 400m Emitter-base grading coefficient
PS 340m Collector-substrate grading coefficient
RBC 23 Constant part of the base resistance
RBV 18 Zero-bias value of the variable part of base resistance
RCC 12 Constant part of collector resistance
RCV 150 Resistance of the un-modulated epilayer
RE 5 Emitter resistance
SCRCV 1.25K Space charge resistance of the epilayer
SFH 300m Current spreading factor of avalanche model
(when EXAVL=1)
TAUB 4.2p Transit time of stored base charge
TAUE 2p Minimum transit time of stored emitter charge
TAUR 520p Transit time of reverse extrinsic base charge
TEPI 41p Transit time of stored epilayer charge
TREF 25 Reference temperature
VAVL 3 Voltage determining curvature of avalanche current
VDC 680m Collector-base diffusion voltage
VDE 950m Emitter-base diffusion voltage
VDS 620m Collector-substrate diffusion voltage
VEF 44 Forward Early voltage
VER 2.5 Reverse Early voltage
VGB 1.17 Bandgap voltage of the base
VGC 1.18 Bandgap voltage of the collector
VGJ 1.15 Bandgap voltage recombination emitter-base junction
VGS 1.2 Bandgap voltage of the substrate
VLR 200m Crossover voltage of the non-ideal reverse base current
WAVL 1.1u Epilayer thickness used in weak-avalanche model


XCJC 32m Fraction of the collector-base depletion capacitance
under the emitter
XCJE 400m Fraction of the emitter-base depletion capacitance that
belongs to the sidewall
XEXT 630m Part of Iex, Qtex, Qex and Isub that depends on Vbc1
instead of Vb1c1
XIBI 0 Part of ideal base current that belong to the sidewall
XP 350m Constant part of Cjc
XREC 0 Pre-factor of the recombination part of Ib1
For the thermal model (Level = 21) there are three additional parameters to han-
dle thermal effects.

ATH 0 Thermal resistance temperature coefficient
CTH 3n Thermal capacitance
RTH 300 Thermal resistance
551
Bipolar transistor (Philips Modella Level = 500 or 501)
SPICE format
Syntax
Q<name> <collector> <base> <emitter> [substrate] [thermal]
+<model name>
Examples
Q1 1 2 3 MM1
Q2 1 2 3 4 MM2
Q3 1 2 3 4 5 MM3
Q4 1 2 3 4 5 MM3 MULT=20
Schematic format
PART attribute
<name>
Example
Q1
VALUE attribute
MULT=<no_parallel>
Example
MULT=100
MODEL attribute
<model name>
Example
LATPNP

.MODEL <model name> LPNP (LEVEL=500 [model parameters])
.MODEL <model name> LPNP (LEVEL=501 [model parameters])
Example of standard model

.MODEL ML LPNP (LEVEL=500 ...)
Example of thermal model
.MODEL MLT LPNP (LEVEL=501 ...)
Modella (MODEL LAteral) models modern lateral PNP structures, account-
ing for the complex two-dimensional nature of the device. In this physics-based
model, the main currents and charges are independently related to bias-dependent
minority carrier concentrations, providing realistic modeling of current crowding,
high-level injection, and bias-dependent output impedance.
The equivalent circuit of the Modella model is as follows:
Figure 26-3 Modella lateral PNP transistor model
Model parameters for Modella transistor levels 500/501

LEVEL 500 Level = 500 for standard or 501 for thermal model
AE 4.48 Temperature coefficient of BF
AF 1 Flicker noise exponent
BF 131 Ideal forward common-emitter current gain
BR 25 Ideal reverse common-emitter current gain
CJC 390f Zero-bias collector-base depletion capacitance
CJE 61f Zero-bias emitter-base depletion capacitance
CJS 1.3p Zero-bias substrate-base depletion capacitance
DTA 0 Difference between the device temperature and the
ambient analysis temperature
553

EAFL 20.5 Early voltage of the lateral forward current component
EAFV 75 Early voltage of the vertical forward current component
EARL 13.1 Early voltage of the lateral reverse current component
EARV 104 Early voltage of the vertical reverse current component
EXPHI 0 Not used
IBF 26f Saturation current of non-ideal forward base current
IBR 120f Saturation current of non-ideal reverse base current
IK 120u High injection knee current
IS .18f Collector-emitter saturation current
ISS 400f Saturation current of substrate-base diode
KF 0 Flicker noise coefficient
MULT 1 Area factor
PC 360m Collector-base grading coefficient
PE 300m Emitter-base grading coefficient
PS 350m Substrate-base grading coefficient
RBCC 10 Constant part of the base resistance `rbc'
RBCV 10 Variable part of the base resistance `rbc'
RBEC 10 Constant part of the base resistance `rbe'
RBEV 50 Variable part of the base resistance `rbe'
RCEX 5 External part of the collector resistance
RCIN 47 Internal part of the collector resistance
REEX 27 External part of the emitter resistance
REIN 66 Internal part of the emitter resistance
RSB 1000T Substrate-base leakage resistance
SNB 2.6 Temperature coefficient of the epitaxial base electron
mobility
SNBN 300m Temperature coefficient of buried layer electron mobility
SPB 2.853 Temperature coefficient of epitaxial base hole mobility
SPC 730m Temperature coefficient of collector hole mobility
SPE 730m Temperature coefficient of emitter hole mobility
SX 1 Temperature coefficient of combined minority carrier
mobility in emitter and buried layer
TFN 200p Low injection forward transit time due to charge stored
in the emitter and the buried layer under the emitter
TFVR 30n Low injection forward transit time due to charge stored
in the epilayer under the emitter
TLAT 2.4n Low injection (forward and reverse) transit time of
charge stored in the epilayer between emitter and
collector

TREF 25 Reference temperature. Default set by option `tnom'
TRN 3n Low injection reverse transit time due to charge stored in
the collector and the buried layer under the collector
TRVR 1n Low injection reverse transit time due to charge stored in
the epilayer under the collector
VDC 570m Collector-base diffusion voltage
VDE 520m Emitter-base diffusion voltage
VDS 520m Substrate-base diffusion voltage
VGB 1.206 Bandgap voltage of the base between emitter and
collector
VGCB 1.206 Bandgap voltage of the collector-base depletion region
VGE 1.206 Bandgap voltage of the emitter
VGEB 1.206 Bandgap voltage of the emitter-base depletion region
VGJE 1.123 Bandgap voltage recombination emitter-base junction
VGSB 1.206 Bandgap voltage of the substrate-base depletion region
VLF 540m Crossover voltage of non-ideal forward base current
VLR 480m Crossover voltage of non-ideal reverse base current
XCS 3 Ratio between saturation current of c-b-s transistor and
c-b-e transistor
XES 2.7m Ratio between saturation current of e-b-s transistor and
e-b-c transistor
XHCS 1 Fraction of substrate current of c-b-s transistor subject to
high injection
XHES 700m Fraction of substrate current of e-b-s transistor subject to
high injection
XIFV 430m Vertical fraction of forward current
XIRV 430m Vertical fraction of reverse current
For the thermal model (Level = 501) there are three additional parameters to
handle thermal effects.

555
Capacitor
SPICE format
Syntax
C<name> <plus> <minus> [model name]
+ <capacitance> [IC=<initial voltage>]
Example
C2 7 8 120P IC=2

<plus> and <minus> are the positive and negative node numbers. The
polarity references are used to apply the initial condition.
Schematic format
PART attribute
<name>
Example
C5
CAPACITANCE attribute
<capacitance> [LOT[/#][/GAUSS|/UNIFORM|WCASE] = <tolerance>[%]]
[IC=<initial value>]
Examples
1U LOT=2% IC=1.0
120P IC=3
1N/(1+V(C1)^2)
CHARGE attribute
[charge]

Example
ATAN(V(C1))
FREQ attribute
[freq]
Example
1.2+10m*log(F)

MODEL attribute
[model_name]
Example
CMOD
CAPACITANCE attribute
[capacitance] may be a simple number or an expression involving time-
domain variables. It is evaluated in the time domain only. Consider the
following expression:
1N/SQRT(1+V(C1))
V(C1) refers to the value of the voltage across C1 during a transient analysis,
a DC operating point calculation prior to an AC analysis, or during a DC
analysis. It does not mean the AC small signal voltage across C1. If the DC
operating point value for V(C1) was 3, the capacitance would be evaluated
as 1N/SQRT(1+3)) =.5n. The constant value, .5n, would be used in AC
analysis.
CHARGE attribute
[charge], if used, must be an expression involving time-domain variables,
including the voltage across the capacitor and perhaps other symbolic
(.define or .param) variables.
Rules about using CHARGE and CAPACITANCE expressions

1) Either [capacitance] or [charge] must be given.
2) If both [capacitance] and [charge] are given, the user must ensure that
[capacitance] is the derivative of [charge] with respect to the capacitor
voltage. [capacitance] = d([charge])/dV
3) If [capacitance] is not given and [charge] is given, MC12 will create an

expression for capacitance by taking its derivative, C = dQ/dV.
4) If [capacitance] is given as an expression, C(V), and [charge] is not

given, MC12 creates an equivalent circuit consisting of a current source of
value C(V)*DDT(V). In this case, the charge variable for the capacitor will
always be zero. In all other cases the charge variable is available.
5) If [charge] is given, its expression must involve the voltage across the
capacitor. Even for a constant capacitance, Q(C) = C*V(C).
557
6) If [capacitance] or [charge] is given as a time-varying expression, the
MODEL attribute is ignored. Capacitance and charge values are determined
solely by the expressions and are unaffected by model parameters.
FREQ attribute
If <fexpr> is given, it replaces the capacitance value determined during the
operating point. <fexpr> may be a simple number or an expression involving
frequency domain variables. <fexpr> is evaluated during AC analysis as the
frequency changes. For example, suppose the <fexpr> attribute was this:
1n + 1E-9*V(1,2)*(1+10m*log(f))

In this expression F refers to the AC analysis frequency variable and V(1,2)
refers to the AC small signal voltage between nodes 1 and 2. Note that there
is no time-domain equivalent to <fexpr>. Even if <fexpr> is present,
[capacitance] and/or [charge] will be used in transient analysis.
MODEL attribute
If [model_name] is given then the model parameters specified in a library or
a model statement are employed. If either [capacitance] or [charge] are
given as time-varying expressions, then the MODEL attribute is ignored.
Initial conditions
IC=<initial voltage> assigns an initial voltage across the capacitor.
Stepping effects
The CAPACITANCE attribute and all of the model parameters may be stepped,
in which case, the stepped value replaces [capacitance], even if it is an expres-
sion. The stepped value may be further modified by the quadratic and tempera-
ture effects.
Quadratic effects
If [model_name] is used, [capacitance] is multiplied by a factor, QF, a quadratic
function of the time-domain voltage, V, across the capacitor.
QF = 1+ VC1•V + VC2•V2
This is intended to provide a subset of the old SPICE 2G POLY keyword, which
is no longer supported.

Temperature effects
If [model_name] is used, [capacitance] is multiplied by a temperature factor:
TF = 1+TC1•(T-Tnom)+TC2•(T-Tnom)2
TC1 is the linear temperature coefficient and is sometimes given in data sheets as
parts per million per degree C. To convert ppm specs to TC1 divide by 1E6. For
example, a spec of 1500 ppm/degree C would produce a TC1 value of 1.5E-3.
nominal capacitance was measured. T is set to the analysis temperature from the
Analysis Limits dialog box. TNOM is determined by the Global Settings TNOM
value, which can be overridden with a .OPTIONS statement. T and Tnom may
be changed for each model by specifying values for T_MEASURED, T_ABS,
T_REL_GLOBAL, and T_REL_LOCAL. See the .MODEL section of the Com-
mand Statements chapter, for more information on how device operating tem-
peratures and Tnom temperatures are calculated.
Monte Carlo effects

LOT and DEV Monte Carlo tolerances, available only when [model_name] is
used, are obtained from the model statement. They are expressed as either a per-
centage or as an absolute value and are available for all of the model parameters
except the T_ parameters. Both forms are converted to an equivalent tolerance
percentage and produce their effect by increasing or decreasing the Monte Carlo
factor, MF, which ultimately multiplies the value of the model parameter C.
MF = 1 ± tolerance percentage /100
If tolerance percentage is zero or Monte Carlo is not in use, then the MF factor is
set to 1.0 and has no effect on the final value.
The final capacitance used in the analysis, cvalue, is calculated as follows:
cvalue = [capacitance] * QF * TF * MF * C, where C is the model parameter

multiplier.

.MODEL <model name> CAP ([model parameters])
Examples
.MODEL CMOD CAP (C=2.0 LOT=10% VC1=2E-3 VC2=.0015)
.MODEL CEL CAP (C=1.0 LOT=5% DEV=.5% T_ABS=37)
559
Model parameters
Name Parameter Units Default
C Capacitance multiplier 1
LS Series inductance Henrys 0
RP Parallel resistance Ohms ∞
RS Series resistance Ohms 0
VC1 Linear voltage coefficient V-1 0
VC2 Quadratic voltage coefficient V-2 0
TC1 Linear temperature coefficient °C 0
-1
TC2 Quadratic temperature coefficient °C-2 0

The model used for the capacitor when the parasitic values of LS, RP, and RS are
specified is as follows:
Figure 26-4 Capacitor model
Noise effects
RS and RP, if specified, generate thermal noise currents in the usual manner.
IRS = (4•k•T / RS)0.5

IRP = (4•k•T / RP)0.5

Dependent sources (linear)
Schematic format
PART attribute
<name>
Example
DEP1
VALUE attribute
<value>
Example
10
These ideal, linear, two-port functions transform the input voltage or current to an
output voltage or current. Input voltage is the voltage between the plus and minus
input leads. Positive input current is defined as into the plus input lead.
Model equations
Part Equation
IOFI Iout(Iin) = value• Iin
IOFV Iout(Vin) = value• Vin
VOFV Vout(Vin) = value• Vin
VOFI Vout(Iin) = value• Iin
Figure 26-5 Dependent sources-linear
561
Dependent sources (SPICE E, F, G, H devices)
Standard SPICE formats:
Syntax of the voltage-controlled voltage source
E<name> <plusout> <minusout> [POLY(<k>)] n1p n1m
+ [n2p n2m...nkp nkm] p0 [p1...pk] [IC=c1[,c2[,c3...[,ck]]]]
Syntax of the current-controlled current source

F<name> <plusout> <minusout> [POLY(<k>)] v1 [v2...vk]
+ p0 [p1...pk] [IC=c1[,c2[,c3...[,ck]]]]
Syntax of the voltage-controlled current source

G<name> <plusout> <minusout> [POLY(<k>)]
+n1p n1m [n2p n2m...nkp nkm] p0 [p1...pk]
[IC=c1[,c2[,c3...[,ck]]]]
Syntax of the current-controlled voltage source

H<name> <plusout> <minusout> [POLY(<k>)] v1 [v2...vk]
+ p0 [p1...pk] [IC=c1[,c2[,c3...[,ck]]]]
Standard PSpiceTM supported formats:

Extended syntax of the voltage-controlled voltage source
[E | G]<name> <plusout> <minusout> VALUE = {<expression>}
[E | G]<name> <plusout> <minusout> TABLE{<expression>} =

+ <<input value>,<output value>>*
[E | G]<name> <plusout> <minusout> LAPLACE {<expression>} =

+ {<Laplace transfer function>}
[E | G]<name> <plusout> <minusout> FREQ

+ {<expression>} = [KEYWORD]
+<<frequency value>,<magnitude value>,<phase value>>*
n1p is the first positive controlling node.

n1m is the first negative controlling node.
nkp is the k'th positive controlling node.
nkm is the k'th negative controlling node.
p0 is the first polynomial coefficient.
pk is the k'th polynomial coefficient.
v1 is the voltage source whose current is the first controlling variable.
vk is the voltage source whose current is the k'th controlling variable.
c1 is the 1'st initial condition.
ck is the k'th initial condition.
SPICE Examples
E2 7 4 POLY(2) 10 15 20 25 1.0 2.0 10.0 20.0
G2 7 4 POLY(3) 10 15 20 25 30 35 1.0 2.0 3.0 10.0 20.0 30.0
H2 7 4 POLY(3) V1 V2 V3 1.0 2.0 3.0 10.0 20.0 30.0
E1 10 20 VALUE = {V(2)*V(3)}
E2 10 20 FREQ {V(1,2)} = (0,0,0) (1K,0,0) (10K,0.001,0)
G1 10 20 TABLE{V(5,6)*V(3)} = (0,0) (1,1) (2,3.5)
E2 10 20 LAPLACE {V(5,6)} = {1/(1+.001*S+1E-8*S*S)}
Schematic format
The schematic attributes are similar to the standard SPICE format without the
<plusout> and <minusout> node numbers. The TABLE, VALUE, LAPLACE,
and FREQ features are not supported in the schematic versions of the E, F, G,
and H devices. These features are supported in the Function and Laplace devices,
described later in this chapter.
PART attribute
<name>
Example
G1
VALUE attribute
[POLY(<k>)] n1p n1m [n2p n2m...nkp nkm] p0 [p1...pk]
+ [IC=c1[,c2[,c3...[,ck]]]]

[POLY(<k>)] n1p n1m [n2p n2m...nkp nkm] p0 [p1...pk]
+ [IC=c1[,c2[,c3...[,ck]]]]
[POLY(<k>)] v1 [v2...vk] p0 [p1...pk] [IC=c1[,c2[,c3...[,ck]]]]
[POLY(<k>)] v1 [v2...vk] p0 [p1...pk] [IC=c1[,c2[,c3...[,ck]]]]
Examples
POLY(2) 10 15 20 25 1.0 2.0 10.0 20.0
POLY(3) 10 15 20 25 30 35 1.0 2.0 3.0 10.0 20.0 30.0
POLY(2) V1 V2 1.0 2.0 10.0 20.0
POLY(3) V1 V2 V3 1.0 2.0 3.0 10.0 20.0 30.0
563
Model equations
Figure 26-6 Dependent sources-SPICE poly
When the POLY keyword is not used, the general equation for the dependent
source function is:
F = p0 + p1•V1 + p2•V12+ p3•V13+... pk•V1k

F is the output value of the dependent voltage or current source.

V1 is the k independent variable value.
p0, p1, p2...pk are the k +1 polynomial coefficients.
When the POLY keyword is used, the polynomial is computed as follows

The values of the exponents E1, E2, E3, ..., En are chosen by a procedure which is
best understood by referring to Table 22-1.
For a description of how the LAPLACE and FREQ versions work, see the La-
place sources section of this chapter.

Number of input variables (Order)
1 2 3 4
Coefficient E1 E1 E2 E1 E2 E3 E1 E2 E3 E4
P0 0 0 0 0 0 0 0 0 0 0
P1 1 1 0 1 0 0 1 0 0 0
P2 2 0 1 0 1 0 0 1 0 0
P3 3 2 0 0 0 1 0 0 1 0
P4 4 1 1 2 0 0 0 0 0 1
P5 5 0 2 1 1 0 2 0 0 0
P6 6 3 0 1 0 1 1 1 0 0
P7 7 2 1 0 2 0 1 0 1 0
P8 8 1 2 0 1 1 1 0 0 1
P9 9 0 3 0 0 2 0 2 0 0
P10 10 4 0 3 0 0 0 1 1 0
P12 12 3 1 2 1 0 0 1 0 1
P12 12 2 2 2 0 1 0 0 2 0
P13 13 1 3 1 2 0 0 0 1 1
P14 14 0 4 1 1 1 0 0 0 2
P15 15 5 0 1 0 2 3 0 0 0
P16 16 4 1 0 3 0 2 1 0 0
P17 17 3 2 0 2 1 2 0 1 0
P18 18 2 3 0 1 2 2 0 0 1
P19 19 1 4 0 0 3 1 2 0 0
Table 22-1 Polynomial exponents
565
The lightly shaded parts of the table mark the coefficients used for summing the
input variables. The heavily shaded portions of the table mark the coefficients
used for forming a product of the input variables. Other combinations of polyno-
mial products are shown in the rest of the table.
For example, to create a voltage source whose value equals the sum of three input
voltages, use this:
E1 4 0 POLY(3) 1 0 2 0 3 0 0 1 1 1
This creates a voltage source whose output is a third-order polynomial function

of three sets of differential voltages.
Input variable 1 = V(1) - V(0)= V(1)= voltage on node 1

The exponents E1=1, E2=1, and E3=1 are chosen from the p1, p2, and p3 rows, re-
spectively, of the 3'rd order column. The output of this source is:
V = p0•V10•V20•V30+ p1•V12•V20•V30+ p2•V10•V21•V30+ p3•V10•V20•V31

V = p0+ p1•V1+p2•V2+p3•V3
V = 0 + 1•V1+1•V2+1•V3
V = V1+V2+V3
To create a current source whose value equals the product of the current flowing
in two sources, use this:
F1 3 0 POLY(2) V1 V2 0 0 0 0 1
This creates a current source whose output is a second order polynomial function
of the current flowing in the sources V1 and V2.
Input variable 1 = I1= current flowing through V1

Input variable 2 = I2 = current flowing through V2
The exponents E1=1 and E2=1 are chosen from the p4 row and second order col-
umn. The output of this source is:
I = p0•I10•I20+ p1•I12•I20+ p2•I10•I21+ p3•I12•I20+ p4•I12•I21

I = 0•I10•I20+ 0•I12•I20+ 0•I10•I21+ 0•I12•I20+ 1•I12•I21
I = I1•I2
Diode
SPICE format
Syntax
D<name> <anode> <cathode> <model name> [area] [OFF]
+ [IC=<vd>][M=<multiplier>]
Example
D1 7 8 1N914 1.0 OFF IC=.001
Schematic format
PART attribute
<name>
Example
D1
VALUE attribute
[area] [OFF] [IC=<vd>][M=<multiplier>]
Example
10.0 OFF IC=0.65 M=2
MODEL attribute
<model name>
Example
1N914
Both formats
[area] multiplies or divides model parameters as shown in the model parameters
table. The presence of the OFF keyword forces the diode off during the first itera-
tion of the DC operating point. The initial condition, [IC=<vd>], assigns an initial
voltage to the junction in transient analysis if no operating point is done (or if the
UIC flag is set). <multiplier> is the number of parallel devices.

.MODEL <model name> D ([model parameters])
Example
.MODEL 1N4434 D (IS=1E-16 RS=0.55 TT=5N)
567
Diode model parameters
The diode model is the standard PSpiceTM diode model with an additional linear
parallel resistance added to account for leakage effects.
Name Parameter Units Def Area

Level Model level (1=SPICE2G, 2=PSpice) 1.0
IS Saturation current A 1E-14 *
N Emission coefficient 1.00
ISR Recombination current param. A 0.00 *
NR Emission coefficient for ISR 2.00
IKF High-injection knee current A ∞ *
BV Reverse breakdown knee voltage V ∞
IBV Reverse breakdown knee current A 1E-10 *
NBV Reverse breakdown ideality 1
IBVL Low-level reverse breakdown current A 0 *
NBVL Low-level reverse breakdown ideality 1
RS Parasitic series resistance Ω 0 /
TT Transit time S 0.00
CJO Zero-bias junction cap. F 0.00 *
VJ Built-in potential V 1.00
M Grading coefficient 0.50
EG Energy gap eV 1.12
XTI Temperature exponent for IS 3.00
TIKF IKF temperature coefficient(linear) °C-1 0.00
TBV1 BV temperature coefficient(linear) °C-1 0.00
TBV2 BV temperature coefficient(quadratic) °C-2 0.00
TRS1 RS temperature coefficient(linear) °C-1 0.00
TRS2 RS temperature coefficient(quadratic) °C-2 0.00
KF Flicker noise coefficient 0.00
AF Flicker noise exponent 1.00
RL Leakage resistance Ω ∞
The parameters ISR, NR, IKF, NBV, IBVL, NBVL, TIKF, TBV1, TBV2, TRS1,
and TRS2 are used only if Level is set to 2. The RL parameter is used only if
Level is set to 1.

Model equations
Figure 26-7 Diode model
Notes and Definitions

The model parameters IS, ISR, IKF, IBV, IBVL, and CJO are multiplied by
[area] and the model parameter RS is divided by [area] prior to their use in the
diode model equations below.
specifying the parameters T_MEASURED, T_ABS, T_REL_GLOBAL, and T_
REL_LOCAL. See the .MODEL section of the Command Statements chapter, for
more information on how device operating temperatures and Tnom temperatures
are calculated.
Temperature Effects
VT = k • T / q = 1.38E-23 • T / 1.602E-19
IS(T) = IS • e((T/Tnom - 1)•EG/(VT•N)) • (T/Tnom)(XTI/N)
ISR(T) = ISR • e((T/Tnom - 1)•EG/(VT•NR)) • (T/Tnom)(XTI/NR)
IKF(T) = IKF • (1+TIKF•(T - Tnom))
BV(T) = BV • (1+TBV1•(T-Tnom)+TBV2•(T-Tnom)2)
RS(T) = RS • (1+TRS1•(T-Tnom)+TRS2•(T-Tnom)2)
VJ(T) = VJ•(T/Tnom)- 3•VT•ln(T/Tnom)- EG(Tnom)•(T/Tnom)+EG(T)
EG(T) = 1.16-.000702•T2/(T+1208)
EG(Tnom) = 1.17-.000702•Tnom2/(Tnom+1208)
CJO(T) = CJO•(1+M•(.0004•(T-Tnom) + (1 - VJ(T)/VJ)))
569
Current source equations
I = Ifwd - Irev
Inrm = IS(T) • (e(V/(VT•N)) - 1)
If IKF > 0
Kinj = (IKF /(IKF +Inrm))1/2
Else
Kinj = 1
Irec = ISR(T) • (e(V/(VT•NR)) - 1)
Kgen = ((1-V/VJ(T))2 + 0.005)M/2
Irev = IBV • (e-(V+BV(T))/(VT•NBV) - 1) + IBVL • (e-(V+BV(T))/(VT•NBVL) - 1)
Ifwd = Kinj • Inrm + Kgen • Irec
Capacitance Equations
Transit Time capacitance
Gd = DC conductance of the diode
CT = TT•Gd
If V ≤ FC•VJ(T) then
CJ = CJO(T)•(1 - V/VJ(T))-M
Else
CJ = CJO(T)•(1 - FC)-(1+M)•(1 - FC•(1+M)+M•(V/VJ(T)))
C = CT + CJ
Noise Equations
Flicker and shot noise is generated by the diode current, I. The resistors RS and
RL generate thermal noise. The noise currents are computed as follows:
IRS = (4•k•T / RS)0.5

IRL = (4•k•T / RL)0.5
II = (2 • q • I + KF • IAF / Frequency)0.5

Diode (Philips JUNCAP and JUNCAP2)
SPICE format
Syntax
D<name> <anode> <cathode> <model name>
+ AB=<diffusion_area>
+ LS=<nongate_sw_length>
+ LG=<gate_sw_length>
Examples
D1 7 8 MM1 LS=.2u LG=.3u AB=2.4p MULT=50
D2 7 8 MDX LS=12u LG=12u AB=121p MULT=2
Schematic format
PART attribute
<name>
Example
D1
VALUE attribute
AB=<diffusion_area>
LS=<nongate_sw_length>
LG=<gate_sw_length>
MULT=<no_parallel>
Example
LS=.46u LG=.12u AB=1.5p MULT=5
MODEL attribute
<model name>
Example
MM2

.MODEL <model name> D (LEVEL=4 [model parameters])
.MODEL <model name> D (LEVEL=200 [model parameters])
571
Examples
.MODEL JUNCAP D (LEVEL=4 ...)
.MODEL JUNCAP2 D (LEVEL=200 ...)
<diffusion_area> is the diffusion area.
<nongate_sw_length> is the length of the sidewall of the diffusion area AB

which is under the gate.
<gate_sw_length> is the length of the sidewall of the diffusion area AB which is

under the gate.
<no_parallel> is the number of devices in parallel.
These device instance parameters can be specified as device attributes or as mod-

el parameters. Device attributes, if specified, supersede model parameters.
The JUNCAP models are intended to describe the behavior of the diodes that are
formed by the source, drain, or well-to-bulk junctions in MOS devices. The cur-
rent equations are formulated and AC effects are modeled via charge equations
using the quasi-static approximation. In order to include the effects of differences
in the sidewall, bottom, and gate edge-junction profiles, each is handled separate-
ly in the model. Both diffusion and generation currents are treated in the model,
each with its own temperature and voltage dependence.
The JUNCAP and JUNCAP2 models are provided for use with the Philips MOS-
FET models 12, 20, 31, and 40. A subcircuit is usually used to bundle one of
the diodeless Philips MOSFET models (12, 20, 31, 40) with one or more of the
JUNCAP models. The PSP MOSFET employs its own built-in JUNCAP2 model.

Model parameters for JUNCAP diode level 4

AB 1p Diffusion area
CJBR 1p Bottom junction capacitance at V=VR
CJGR 1p Gate edge junction capacitance at V=VR
CJSR 1p Sidewall junction capacitance at V=VR
DTA 0 Temperature offset of the junction with respect to TA
IMAX 1K Explosion current
JSDBR 1m Bottom saturation-current density due to diffusion from
the back contact
JSDGR 1m Gate edge saturation-current density due to diffusion
from the back contact
JSDSR 1m Sidewall saturation-current density due to diffusion from
the back contact
JSGBR 1m Bottom saturation-current density due to electron-hole
generation at V=VR
JSGGR 1m Gate edge saturation-current density due to diffusion
from the back contact
JSGSR 1m Sidewall saturation-current density due to electron-hole
generation at V=VR
LG 1u Length of the sidewall of the diffusion area AB which is
under the gate
LS 1u Length of the sidewall of the diffusion area AB which is
not under the gate
NB 1 Emission coefficient of the bottom forward current
NG 1 Emission coefficient of the gate edge forward current
NS 1 Emission coefficient of the sidewall forward current
PB 400m Bottom junction grading coefficient
PG 400m Gate edge junction grading coefficient
PS 400m Sidewall junction grading coefficient
TR 25 Temperature at which the parameters have been
determined
VB 900m Reverse breakdown voltage
VDBR 1 Diffusion voltage of the bottom junction at T=TR
VDGR 1 Diffusion voltage of the gate edge junction at T=TR
VDSR 1 Diffusion voltage of the sidewall junction at T=TR
VR 0 Voltage at which the parameters have been determined
573
Model parameters for JUNCAP2 diode level 200

AB 1p Junction area
CBBTBOT 1p Band-to-band tunneling prefactor of bottom
component
CBBTGAT .001f Band-to-band tunneling prefactor of
gate-edge component
CBBTSTI .001f Band-to-band tunneling prefactor of STI-edge
component
CJORBOT 1m Zero-bias capacitance per unit area of bottom
component
CJORGAT 1n Zero-bias capacitance per unit length of
gate-edge component
CJORSTI 1n Zero-bias capacitance per unit length of
STI-edge component
CSRHBOT 100 Shockley-Read-Hall prefactor of bottom
component
CSRHGAT 100u Shockley-Read-Hall prefactor of gate-edge
component
CSRHSTI 100u Shockley-Read-Hall prefactor of STI-edge
component
CTATBOT 100 Trap-assisted tunneling prefactor of bottom
component
CTATGAT 100u Trap-assisted tunneling prefactor of gate-edge
component
CTATSTI 100u Trap-assisted tunneling prefactor of STI-edge
component
DTA 0 Temperature offset with respect to ambient
temperature
FBBTRBOT 1G Normalization field at the reference temperature
for band-to-band tunneling of bottom component
FBBTRGAT 1G Normalization field at the reference temperature
for band-to-band tunneling of gate-edge
component
FBBTRSTI 1G Normalization field at the reference temperature
for band-to-band tunneling of STI-edge
component
IDSATRBOT 1p Saturation current density at the reference
temperature of bottom component


IDSATRGAT .001f Saturation current density at the reference
temperature of gate-edge component
IDSATRSTI .001f Saturation current density at the reference
temperature of STI-edge component
IMAX 1K Maximum current up to which forward current
behaves exponentially
LG 1u Gate-edge part of junction perimeter
LS 1u STI-edge part of junction perimeter
MEFFTATBOT 250m Effective mass (in units of m0) for trap-assisted
tunneling of bottom component
MEFFTATGAT 250m Effective mass (in units of m0) for trap-assisted
tunneling of gate-edge component
MEFFTATSTI 250m Effective mass (in units of m0) for trap-assisted
tunneling of STI-edge component
PBOT 500m Grading coefficient of bottom component
PBRBOT 4 Breakdown onset tuning parameter of bottom
component
PBRGAT 4 Breakdown onset tuning parameter of gate-edge
component
PBRSTI 4 Breakdown onset tuning parameter of STI-edge
component
PGAT 500m Grading coefficient of gate-edge component
PHIGBOT 1.16 Zero-temperature bandgap voltage of bottom
component
PHIGGAT 1.16 Zero-temperature bandgap voltage of gate-edge
component
PHIGSTI 1.16 Zero-temperature bandgap voltage of STI-edge
component
PSTI 500m Grading coefficient of STI-edge component
STFBBTBOT -1m Temperature scaling parameter for band-to-band
STFBBTGAT -1m Temperature scaling parameter for band-to-band
STFBBTSTI -1m Temperature scaling parameter for band-to-band
TRJ 21 Reference temperature
575

TYPE 1 Type parameter, 1=n-type, -1=p-type
VBIRBOT 1 Built-in voltage at the reference temperature of
bottom component
VBIRGAT 1 Built-in voltage at the reference temperature of
gate-edge component
VBIRSTI 1 Built-in voltage at the reference temperature of
STI-edge component
VBRBOT 10 Breakdown voltage of bottom component
VBRGAT 10 Breakdown voltage of gate-edge component
VBRSTI 10 Breakdown voltage of STI-edge component
XJUNGAT 100n Junction depth of gate-edge component
XJUNSTI 100n Junction depth of STI-edge component

Function sources
Schematic format
PART attribute
<name>
Example
F1
VALUE attribute for formula (NFV and NFI) type

<formula>
Example of formula type

10*Sin(2*PI*1E6*T)*V(3)*I(L1)*EXP(-V(IN)/100NS)
FREQ attribute
[<fexpr>]
Example
1200*(1+sqrt(F/1e6))
DERIVATIVE attribute
[<Algebraic> | <Numeric> | <Default>]
Example
Algebraic
NOISE EXPRESSION attribute for NFI only

[<noise_expr>]
Example
1200*(1+sqrt(F/1e6))
TABLE attribute for table (NTIOFI, NTIOFV, NTVOFV, NTVOFI) types

(<x1,y1>) (<x2,y2>) ... ([(xk,yk)])
Curly braces {} are required for expressions and optional for variables.
Examples of table type

(-1m,25) (1m,25) (2m,30)
({start - 1m}, {25*level} ) (end,level) ({end+3m}, level2)
577
Function sources provide the principle time domain analog behavioral modeling
capability. You can express a voltage or current source's time-domain dependence
on circuit variables as an algebraic formula or as a tabular function. The two ba-
sic types are distinguished by the way the value of the output current or voltage is
calculated.
Formula type
The Formula type, which is similar to the SPICE3 B device, uses an algebraic
formula, or expression, to compute the value of the output variable as a function
of any set of valid time-domain variables. There are two versions of this source:
NFI Function current source

NFV Function voltage source
Here is an example of an expression that models a vacuum triode:
K* pow((V(Plate)-V(Cathode)+Mu*(V(Grid)-V(Cathode))),1.5)
Table type
The Table type, which is similar to the SPICE3 A device, uses a table of ordered
data pairs which describe the output variable as a function solely of the input
variable. The table describes a time-domain transfer function.
The input variable for a Table source may be either:
Current flowing into the positive input lead.

Voltage between the positive input lead and the negative input lead.
There are four basic types of Table source:
Source type Input Output Definition

Current-controlled current source I I NTIOFI
Current-controlled voltage source I V NTVOFI
Voltage-controlled voltage source V V NTVOFV
Voltage-controlled current source V I NTIOFV
There are several rules for constructing the data pairs in the TABLE attribute.
1. The x,y values and the pairs may be separated by commas or spaces. The
x,y values may be replaced by expressions containing constants or symbolic
variables created with a .define statement.

Expressions are evaluated only once, in the setup phase of the analysis, so
they must not contain variables that vary during an analysis run, like V(1) or
T, or even simulation control variables like TSTART that are unknown when
the expressions are evaluated.
2. Data pairs must be arranged in input ascending order.
x1<x2<...xk
3. Expressions whose value stays constant within an analysis run may be

used within a table. These must be enclosed with curly braces {}. For
example;
0,1 1,2 2,{VAR1}
VAR1 is created with a .define or .param statement and may change

between runs.
Output is calculated from the input value as follows:
1. The output value is constant at y1 for input values below x1.
2. The output value is constant at yk for input values above xk.
3. Output values are interpolated for input values between table values.
For example:
(-.010,-10) (.010,10)
For an NTVOFV source, this describes an ideal amplifier having a gain of 1000
with the output clipped to +-10 volts. The output value when the input is greater
than .010 is limited to +10.0. Similarly, the output value when the input is less
than -.010 is limited to -10.0.
FREQ usage
If <fexpr> is present, it replaces the ordinary small-signal AC incremental value
determined during the operating point. <fexpr> may be a simple number or an
expression involving frequency domain variables. The expression is evaluated
during AC analysis as the frequency changes. For example, suppose the <fexpr>
attribute is this:
579
1+V(3)*(1+1e6/F)
In this expression, F refers to the AC analysis frequency variable and V(3) refers
to the AC small-signal voltage from node 3 to ground. There is no time-domain
equivalent to <fexpr>. Even if <fexpr> is present, only <value> will be used in
transient analysis.
NOISE_EXPRESSION usage
If noise_expr is present, it generates a noise current equal to the expression. For
example to simulate shot noise you might use an expression like this:
1E-16 * pow(6.5ma,1.1) / F
Note that the expression should contain only frequency (F) dependent variables.
The feature is available only in the NFI source.
DERIVATIVE attribute
Derivatives of the expressions are evaluated in several ways:
Algebraic:
Algebraic formulas are created for each derivative of each variable in the expres-
sion. This is generally the preferred way as it is usually more accurate. However,
complex expressions with many variables sometimes result in large unwieldy
derivative expressions which take much longer to evaluate than simple numerical
derivatives.
Numeric:
Derivatives are calculated numerically by simple perturbation. This method often
works the best when the formulas are complex but well behaved and have no
discontinuities.
Default:
In this case derivatives are evaluated according to the state of the Global Settings
NUMERIC_DERIVATIVE flag.
See the sample circuit T1 for an example of table sources, and the sample circuits
F1, F2, F3, and F4 for examples of formula sources.
SMOKE attribute
This is the SMOKE model name.

SMOOTHING attribute
This option chooses whether smoothing is to be employed when the VALUE at-
tribute is set to TABLE. Yes or No turns smoothing on or off. Inherit From Pref-
erences uses the settings from the Preferences (Ctrl+Shift+P) dialog box.
FRACTION attribute
For limit functions, this is the fraction of the interval between high and low val-
ues where smoothing begins. For TABLE functions, the fraction of the interval
between X,Y points where smoothing begins. The fraction can be entered as a
decimal(0.1) or a percent (10%). FRACTION is limited to the range 0.0 to 0.50.
INTERPOLATION attribute
This specifies the log or linear nature of the table entries. There are four choices:
Linear_X _Linear_Y
Linear_X _Log_Y
Log_X _Linear_Y
Log_X _Log_Y
581
GaAsFET
SPICE format
Syntax
B<name> <drain> <gate> <source> <model name>
+ [area] [OFF] [IC=<vds>[,vgs]][M=<multiplier>]
Example
B1 5 7 9 2N3531 1 OFF IC=1.0,2.5 M=10
Schematic format
PART attribute
<name>
Example
B1
VALUE attribute
[area] [OFF] [IC=<vds>[,vgs]][M=<multiplier>]
Example
1.5 OFF IC=0.05,1.00
MODEL attribute
<model name>
Example
GFX_01
The device is an n-channel device. There is no p-channel version. Level 1 = Cur-

tice model, level 2 = Raytheon or Statz model, level 3 = Triquint model, and level
4 = Parker-Skellern model. The [OFF] keyword forces the device off for the first
iteration of the operating point. The initial condition, [IC=vds[,vgs]], assigns ini-
tial voltages to the drain-source and gate-source terms. <multiplier> is the num-
ber of parallel devices.

.MODEL <model name> GASFET ([model parameters])
Example
.MODEL B1 GASFET (VTO=-2 ALPHA=2 BETA=1E-4 LAMBDA=1E-3)
Model Parameters
Name Parameter Units Def. Level Area
LEVEL Model level (1, 2, 3, or 4) 1 ALL
VTO Pinch-off voltage V -2.50 ALL
ALPHA Saturation voltage parameter V -1
2.00 ALL
BETA Transconductance coefficient A/V2 0.10 ALL *
B Doping tail extender V-1 0.30 2
LAMBDA Channel-length modulation V-1 0.00 ALL
GAMMA Static feedback parameter 0.00 3
DELTA Output feedback parameter (A-V)-1 0.00 3
Q Power law parameter 2.00 3
RG Gate ohmic resistance Ω 0.00 ALL /
RD Drain ohmic resistance Ω 0.00 ALL /
RS Source ohmic resistance Ω 0.00 ALL /
IS Gate pn saturation current A 1E-14 ALL
N Gate pn emission coefficient 1.00 ALL
M Gate pn grading coefficient 0.50 ALL
VBI Gate pn potential V 1.00 ALL
CGD Zero-bias gate-drain pn cap. F 0.00 ALL *
CGS Zero-bias gate-source pn cap. F 0.00 ALL *
CDS Fixed drain-source cap. F 0.00 ALL *
FC Forward-bias depletion coeff. 0.50 ALL
VDELTA Capacitance transition volt. V 0.20 2,3
VMAX Capacitance limiting voltage V 0.50 2,3
EG Bandgap voltage eV 1.12 ALL
XTI IS temperature coefficient 0.00 ALL
VTOTC VTO temperature coefficient V/°C 0.00 ALL
BETATCE BETA exp. temperature coeff. %/°C 0.00 ALL
TRG1 RG temperature coefficient °C-1 0.00 ALL
TRD1 RD temperature coefficient °C-1 0.00 ALL
TRS1 RS temperature coefficient °C-1 0.00 ALL
KF Flicker-noise coefficient 0.00 ALL
AF Flicker-noise exponent 1.00 ALL
T_MEASURED Measured temperature °C ALL
T_ABS Absolute temperature °C ALL
T_REL_GLOBAL Relative to current temp. °C ALL
T_REL_LOCAL Relative to AKO temperature °C ALL
583
GaAsFET model equations
Figure 26-8 The GaAsFET model

The model parameters BETA, CGS, CGD, and CDS are multiplied by [area] and
the model parameters RG, RD, and RS are divided by [area] prior to their use in
the equations below.
are calculated.
Vgs = Internal gate to source voltage

Vds = Internal drain to source voltage
Id = Drain current
VT= k • T / q = 1.38E-23 •T / 1.602E-19
In general, X(T) = Temperature adjusted value of parameter X
Temperature Dependence
BETA(T) = BETA•1.01BETATCE•(T- Tnom)
EG(T) = 1.16-.000702•T2/(T+1208)
EG(Tnom) = 1.17-.000702•Tnom2/(Tnom+1208)
VTO(T) = VTO + VTOTC•(T - Tnom)
IS(T) = IS(Tnom) • e((EG/(VT•N))(T/Tnom -1))
RG(T) = RG•(1 + TRG1•(T - Tnom))
RD(T) = RD•(1 + TRD1•(T - Tnom))
RS(T) = RS•(1 + TRS1•(T - Tnom))
VBI(T) = VBI•(T/Tnom)-3•VT•ln((T/Tnom))-EG(Tnom)•(T/Tnom)+EG(T)
CGS(T) = CGS•(1+M•(.0004•(T-Tnom) + (1 - VBI(T)/VBI)))
CGD(T) = CGD•(1+M•(.0004•(T-Tnom) + (1 - VBI(T)/VBI)))
Current equations level 1

Cutoff Region : Vgs ≤ VTO(T)
Id = 0

Linear and Saturation Region: Vgs > VTO(T)
Id=BETA(T)•(1+LAMBDA•Vds)•(Vgs- VTO(T))2•tanh(ALPHA•Vds)

Id = 0

Linear and Saturation Region : Vgs > VTO(T)
If 0< Vds< 3/ALPHA
Kt = 1 - (1 - Vds•ALPHA/3)3
Else
Kt = 1
Id=BETA(T)•(1+LAMBDA•Vds)•(Vgs-VTO(T))2•Kt/(1+B•(Vgs-VTO(T)))

Id = 0

Linear and Saturation Region : Vgs > VTO(T)
If 0< Vds< 3/ALPHA
Kt = 1 - (1 - Vds•ALPHA/3)3
Else
Kt = 1
585
Idso = BETA•(Vgs-(VTO-GAMMA•Vds)Q•Kt
Id = Idso/(1+DELTA•Vds•Idso)
Capacitance equations level 1

If Vgs ≤ FC•VBI(T)
Cgs = CGS/(1 - Vgs/VBI(T))M
Else
Cgs = CGS•(1 - FC)-(1+M)•(1 - FC•(1+M)+M•(Vgs/VBI(T)))
If Vds ≤ FC•VBI(T)
Cgd = CGD/(1 - Vgd/VBI(T))M
Else
Cgd = CGD•(1 - FC)-(1+M)•(1 - FC•(1+M)+M•(Vgd/VBI(T)))
Capacitance equations level 2 and level 3

Ve = (Vgs +Vgd + ((Vgs - Vgd)2 + ALPHA -2 )1/2 )/2
If (Ve +VTO(T) +((Ve - VTO(T))2+DELTA2)1/2)/2 < VMAX

Vn = (Ve +VTO(T) +((Ve - VTO(T))2+DELTA2)1/2)/2
Else
Vn = VMAX
K1 = (1 + Ve - VTO(T))/((Ve - VTO(T))2+DELTA2)1/2)/2
K2 = (1 + (Vgs - Vgd)/((Vgs - Vgd)2+ALPHA-2)1/2)/2
K3 = (1 - (Vgs - Vgd)/((Vgs - Vgd)2+ALPHA-2)1/2)/2
Cgs = CGS•K2•K1/(1 - Vn/VBI(T))1/2 +CGD•K3)
Cgd = CGS•K3•K1/(1 - Vn/VBI(T))1/2 +CGD•K2)
Noise
The parasitic lead resistances, RG, RD, and RS, generate thermal noise currents.
Ig2 = 4•k•T / RG
Id2 = 4•k•T / RD
Is2 = 4•k•T / RS
The drain current generates a noise current.
I2 = 4•k•T•gm•2/3 + KF•IdAF / Frequency

where gm = ∂Id / ∂Vgs (at operating point)

IBIS
Schematic format
These are the IBIS input and output models.
PART attribute
<name>
Example
X1
NAME attribute
<subckt_name>
Example
LVC_33_4_MIN
The IBIS primitive comes in four forms:
Type Purpose
IBIS_Input1 Input only with internal power supplies
IBIS_Input3 Input only with external power supplies
IBIS_Output3 Input, enable, and output pins. Internal power supplies
IBIS_Output5 Input, enable, and output pins. External power supplies
When you place the IBIS part on a schematic, Micro-Cap invokes the IBIS
Translator. It lets you select an input or output model from an IBIS source file.
Micro-Cap then creates an equivalent SPICE subcircuit model for the chosen part
and invokes the Attribute dialog box showing the subcircuit. Click the OK button
and the IBIS model is placed in the schematic.
All of this can be done manually by running the IBIS to SPICE Translator, plac-
ing the subckt generated by it in a circuit, and creating an IBIS component from
the Component editor. The IBIS component and the automatic routines that sup-
port it simply make it easier to create an IBIS part.
See the file IBIS3.CIR for an example of how the component is used.
587
IGBT
SPICE format
Syntax
Z<name> <gate> <collector> <emitter> <model name>
+ [AREA=<area>]
+ [WB=<base_width>]
+ [AGD=<agd>]
+ [KP=<kp>]
+ [TAU=<tau>]
+ [OFF]
+ [IC=[vgs[,vae[,vds[,veb]]]]]
Example
Z1 1 2 0 IXGH10N170
+ AREA=1E-4 WB=100U AGD=.1N KP=1E-5
+ TAU=100N OFF IC=1,2,3,4
Schematic format
PART attribute
<name>
Example
Z1
VALUE attribute
[AREA=<area>] [WB=<base_width>] [AGD=<agd>] [KP=<kp>]
[TAU=<tau>] [OFF] [IC=[vgs[,vae[,vds[,veb]]]]]
Example
AREA=2E-7 WB=10u
MODEL attribute
<model name>
Example
IXGH10N170
This device model is based upon the implementation by Gregory Oziemkiewicz

of the IGBT model developed by Allen R. Hefner at the National Institute of
Standards and Technology. The parameter names, default values and units
are compatible with most SPICE versions of the same model.
The <area>, <base_width>, <agd>, <kp>, and <tau> attributes, if specified,

supercede model parameters of the same name.
The [OFF] keyword forces the device off for the first iteration of the operating
point. The initial condition, IC=[vgs[,vae[,vds[,veb]]]], assigns initial voltages to
the model nodes. Additional model information can be found in these references:
Implementation and Development of the NIST IGBT Model in a SPICE-based

Commercial Circuit Simulator, Gregory Todd Oziemkiewicz, Engineer’s Thesis,
University of Florida, December 1995.
Modelling Buffer Layer IGBT’s for Circuit Simulation, Allen R. Hefner Jr, IEEE
Transactions on Power Electronics, Vol. 10, No. 2, March 1995
An Experimentally Verified IGBT Model Implemented in the Saber Circuit

Simulator, Allen R. Hefner, Jr., Daniel M. Diebolt, IEE Transactions on Power
Electronics, Vol. 9, No. 5, September 1994
Model parameters for IGBT model
Name Default Parameter

AGD 5u Gate-drain overlap area
AREA 10u Device area
BVF 1 Avalanche uniformity factor
BVN 4 Avalanche multiplication exponent
CGS 12.4n Gate-source capacitance per unit area
COXD 35n Gate-drain oxide capacitance per unit area
JSNE 650f Emitter saturation current density
KF 1 Triode region factor
KP 380m MOS transconductance
MUN 1.5K Electron mobility
MUP 450 Hole mobility
NB 200T Base doping
T_ABS undefined Absolute temperature
T_MEASURED undefined Parameter measured temperature
T_REL_GLOBAL undefined Relative to current temperature
T_REL_LOCAL undefined Relative to AKO model temperature
TAU 7.1u Ambipolar recombination lifetime
THETA 20m Transverse field factor
589
VT 4.7 Threshold voltage
VTD 1m Gate-drain overlap depletion threshold
WB 90u Metallurgical base width
IGBT model
Figure 26-9 The IGBT model
Definitions
IMOS = MOSFET channel current (amps)
IC = IT = Collector or anode current (amps)
ICSS = Steady-state (bipolar) collector current (amps)
IBSS = Steady-state base current (amps)

IMULT = Avalanche multiplication current (amps)
b = Ambipolar mobility ratio
Dp = Hole diffusivity (cm2/sec)
Qeb = Instantaneous excess carrier base charge (coulombs)
Qb = Background mobile carrier charge (coulombs)
NI = Intrinsic carrier concentration (cm-3)
M = Avalanche multiplication factor
Igen = Bipolar collector-base thermally generated current (amps)
Esi = Dielectric permittivity of silicon
q = Electronic charge =1.6e-19 (coulombs)
wb = 100•WB (cm)
wbcj = sqrt(2•Esi•(0.6+vbc)/(q•NB)) (cm)
W = Quasi-neutral base width = wb - wbcj (cm)
Rb = Conductivity-modulated base resistance (ohms)
Vgst = Vgs - VT (volts)
DC current source equations
IMOS
Vgst<0
IMOS = 0
Vds <= Vgst/KF

IMOS = KF•KP•(Vgst•Vds - 0.5•KF•Vds2)/(1+THETA•Vgst)
For Vds > Vgst/KF

IMOS = KP•Vgst2/(2•(1+THETA•Vgst))
IC (or IT)
Vgst<0
IC = IT = Vae/Rb
ICSS
Veb<= 0
ICSS = 0
Veb > 0
ICSS = IT/(1+b) + b•(4•Dp/W2)*Qeb/(1+b)
591
IBSS
Veb<= 0
IBSS = 0
Veb > 0
IBSS = Qeb/TAU + (Qeb2/QB)•(4•NB2/Ni2)•JSNE•AREA
IMULT
Imult = (M–1)•(IMOS + ICSS) + M•Igen
Charge / Capacitance equations
Gate-source
Cgs = CGS
Qgs = CGS•VGS
Drain-source
Wdsj = sqrt(2•Esi•(Vds+.6)/(q•NB))
Cds = (AREA-AGD)•Esi/ Wdsj
Qds = q•(AREA-AGD)•NB•Wdsj
Drain-gate
Vds = Vgs - VTD
Cdg = Cdgj•COXD/(Cdgj+COXD)
Wdgj = sqrt(2•Esi•(Vdg+.6)/(q•NB))
Cdgj = AGD•Esi/Wdgj
Cx = COXD•Wdgj/(Esi•AGD)
Qdg = (q•NB•Esi•AGD2/COXD)•Cx-log(1+Cx))-COXD•VTD
Ccer
Wbcj = sqrt(2•Esi•(Vbc+.6)/(q•NB))
Cbcj = AREA•Esi/Wbcj
Ccer = Qeb•Cbcj/(3•QB)
Cmult
Cmult = (M-1)•Ccer
Qmult = (M-1)•Qcer
Emitter-base
Ceb = d(Qeb)/d(Veb)
Qmult = (M-1)•Qcer
592 Chapter 26: Analog

Devices
Independent sources (Voltage Source and Current Source)
SPICE format
Syntax for the voltage source
Vname <plus> <minus> [[DC ] dcvalue]
+ [AC magval [phaseval]]
+ [PULSE v1 v2 [td [tr [tf [pw [per]]]]]]
OR [SIN vo va [f0 [td [df [ph]]]]]
OR [EXP v1 v2 [td1 [tc1 [td2 [tc2 ]]]]]
OR [PWL t1 v1 t2 v2 ...[tn , vn]]
OR [SFFM vo va f0 [mi [fm]]]
OR [NOISE interval [amplitude [start [end [seed]]]]]
OR [GAUSSIAN amp tpeak width [period]]
Syntax for the current source

The current source syntax is the same as the voltage source except for the
use of I for the first character of the name.
Examples
V3 2 0 DC 0 AC 1 0 SIN 0 1 1MEG 100NS 1E6 0 ;voltage-sin
V5 3 0 DC 0 AC 1 0 EXP 0 1 100N 100N 500N 100N ;voltage-exp
I3 4 0 DC 0 AC 1 0 SFFM 0 1 1E6 .5 1E7 ;current-sffm
V1 5 0 DC 1 AC 1 0 NOISE 10N 1 100N 700N 1 ; voltage-noise
Schematic format
These are the Voltage Source and Current Source components from
Component / Analog Primitives / Waveform Sources.
PART attribute
<name>
Example
V1
VALUE attribute
<value> where value is identical to the SPICE format without the name and
the plus and minus node numbers.
Examples
DC 1 PULSE 0 1MA 12ns 8ns 120ns 240ns 500ns
DC 0 AC 1 0 SFFM 0 2 2E6 .5 1E7
593
Equations
The equations and sample waveforms that follow are for transient analysis
only. AC analysis uses AC magval (volts) and AC phaseval (degrees) to set
the amplitude and phase of the small signal stimulus. TSTEP is the print
interval. TSTOP is the run time. These values are obtained from the Analysis
Limits dialog box. For SPICE files, Micro-Cap obtains these values from the
.TRAN statement and copies them to the Analysis Limits dialog box.
EXP type
Name Description Units Default
v1 Initial value V or A None
v2 Peak value V or A None
td1 Rise delay S 0
tc1 Rise time constant S TSTEP
td2 Fall delay S td1+TSTEP
tc2 Fall time constant S TSTEP
The waveform value generated by the EXP option is as follows:
Time interval Value

0 to td1 v1
td1 to td2 v1+(v2-v1)•(1-e-(TIME-td1)/tc1)
td2 to TSTOP v1+(v2-v1)•((1-e-(TIME-td1)/tc1)-(1-e-(TIME-td2)/tc2))
Figure 26-10 Waveform for "EXP 1 2 150n 50n 550n 100n"

PULSE type

v1 Initial value V or A None
v2 Pulse value V or A None
td Delay S 0
tr Rise time S TSTEP
tf Fall time S TSTEP
pw Pulse width S TSTOP
per Period S TSTOP
The waveform value generated by the PULSE option is as follows:
From To Value
0 td v1
td td+tr v1+((v2-v1)/tr)•(T-td)
td+tr td+tr+pw v2
td+tr+pw td+tr+pw+tf v2+((v1-v2)/tf)•(T-td-tr-pw)
td+tr+pw+tf per v1
where From and To are T values, and T=TIME mod per. The waveform
repeats every per seconds.
Figure 26-12 Waveform for "PULSE .4 1.6 .1u .1u .2u .1u .5u"
595
Smoothing
The normal, unsmoothed waveform has sharp angles. For a smoother
waveform use either the tanh or quadratic smoothing. The fraction parameter
specifies the fraction of the rise time or fall time over which smoothing
occurs. It applies only to quadratic smoothing. Here is an example of the
three waveform shapes.
When using the smoothed waveforms, keep this constraint in mind:
td + tr + pw + tf + fraction*(tr+tf) < per
This constraint helps to ensure that there is adequate time in which to include the
transition regions.
Note the fraction value is limited to the range 0 to 0.5. 0 means no smoothing and
.50 means smoothing out to the 50% point between the rise and fall edges.

SFFM type

vo Offset value V or A None
va Peak amplitude V or A None
f0 Carrier frequency Hz 1/TSTOP
mi Modulation index 0
fm Modulation freq. Hz 1/TSTOP
The waveform value generated by the SFFM option is as follows:

F = vo + va•sin(2•π•f0•T+mi•sin(2•π•fm•T))
where T = Transient analysis time
Figure 26-12 Waveform for "SFFM 2 1 8Meg 4 1Meg"
597
SIN type

vo Offset value V or A None
va Peak amplitude V or A None
f0 Frequency Hz 1/TSTOP
td Delay s 0
df Damping factor s-1 0
ph Phase degrees 0
The waveform value generated by the SIN option is as follows:
From To Value
0 td vo
td TSTOP vo+va•sin(2•π•(f0•(T-td)+ph/360))•e-(T-td)•df
where T = Transient analysis time
Figure 26-13 Waveform for "SIN 1 1 10Meg 100n 5E6"

PWL type
General Form:
PWL
+ [TRIGGER=<{trigger_expression}>]
+ [TIME_SCALE_FACTOR=<ts_value>]
+ [VALUE_SCALE_FACTOR=<vs_value>]
+(data_pairs OR FILE="filename")
where the syntax of data_pairs is:

<tin>,<in> Data pairs must be separated by spaces or commas.

ts_value, if present, multiplies all tin values and vs_value, if present,
multiplies all in values.
Syntax for a single point on the waveform:

(<tin>,<in>)

Syntax for m points on the waveform:
(<tin1>,<in1>) (<tin2>,<in2>) ... (<tinm>,<inm>)
Syntax to repeat (data_pairs)* n times:

REPEAT FOR <n> (data_pairs)* ENDREPEAT
Syntax to repeat (data_pairs)* forever:

REPEAT FOREVER (data_pairs)* ENDREPEAT
Syntax to repeat content of (filename)* n times:

REPEAT FOR <n> FILE="filename" ENDREPEAT
Syntax to repeat content of (filename)* forever:

REPEAT FOREVER FILE="filename" ENDREPEAT
The file "filename" must be a text file and its content must use the same
format as used above in data_pairs.
trigger_expression, if true, enables the waveform or disables it if false.
Each data pair specifies one point on the waveform curve. Intermediate
values are linearly interpolated from the table pairs. There is no specific limit
on the number of data pairs in the table. They may be added indefinitely until
system memory is exhausted.
599
Examples:
VALUE attribute for a single 5 volt or amp 5ns wide pulse:
PWL (0,0) (5n,0) (5n,5) (10n,5) (10n,0)
VALUE attribute for another single 5 volt or amp 5ns wide pulse:
PWL TIME_SCALE_FACTOR=1n (0,0) (5,0) (5,5) (10,5) (10,0)
VALUE attribute for a single 5000 volt or amp 5ns wide pulse:
PWL VALUE_SCALE_FACTOR=1E3 (0,0) (5n,0) (5n,5) (10n,5) (10n,0)
VALUE attribute for a 10ns square wave, repeated 20 times:

PWL REPEAT FOR 20 (0,0) (5n,0) (5n,5) (10n,5) (10n,0) ENDREPEAT
VALUE attribute for a 10ns square wave, repeated forever:

PWL REPEAT FOREVER (0,0) (5n,0) (5n,5) (10n,5) (10n,0) ENDREPEAT
VALUE attribute when the content, contained in a file, is repeated forever:

PWL REPEAT FOREVER FILE="my pwl file" ENDREPEAT
The VALUE attribute for a triggered waveform:

TRIGGER={(V(A1)>2) AND (V(A2)>2)} 0,0 10n,3 20n,2 30n,1.5 40n,0
Here is an example of a PWL waveform:
Figure 26-14 Waveform for

"PWL 0.05u,1 0.20u,1.5 0.20u,2 0.40u,1 0.51u,2 0.61u,1.5 0.80u,2"

Smoothing
The normal unsmoothed waveform has sharp angles. For a smoother
curve use quadratic smoothing. The fraction parameter specifies the fraction
of the time difference between adjacent points over which smoothing occurs.
Here is an example of a PWL curve with and without quadratic smoothing.
The blue curve is normal (no smoothing). The red curve is with smoothing
enabled. Note that you may need to use a small time step to be able to see the
smoothed regions.
Note the fraction value is limited to the range 0.0 to 0.50.
601
Noise Type

interval Interval between values Secs 0
amplitude Noise Amplitude V or A 0
start_time Start of random interval Secs 0
end _time End of random interval Secs 0
seed Random number seed None 0
The waveform starts at the dcvalue and stays there through start_time.
Interval seconds later a random value between + amplitude/2, and
- amplitude/2 is added to the baseline dcvalue. Interval seconds later another
random value is added to the dcvalue and the source is updated. This process
is repeated every interval seconds until end_time where the waveform value
returns to the dcvalue. The value at T = start_time and T = end_time is
dcvalue. The first random value occurs at T = start_time + interval. The last
random value occurs at T = end - interval.
If seed is >=1 the values are random but are the same on every run.
Otherwise the values are both random and different on every run. The seed is
initialized at the beginning of every temperature, Monte Carlo, or stepping
run.
Figure 26-15 NOISE Waveform for

"DC 1 AC 1 0 NOISE 2N 1 100N 700N 1"
This source type is an extension to the standard SPICE Voltage Source and
Current Source.
Gaussian Type

amp Amplitude V or A None
tpeak Time to reach A Secs None
width Width at 50% Secs None
period Repetition period Secs 0

The waveform, within each period is a Gaussian pulse calculated as follows,
where T is the elapsed time from the start of the simulation.
dcvalue + amp*exp(-pow(((T mod period) - tpeak)/(width/1.6652),2))
Figure 26-16 GAUSSIAN Waveform for

"DC 0 AC 1 0 GAUSSIAN 1 500N 200N 1U"
This source type is an extension to the standard SPICE voltage and current
sources.
603
Inductor
SPICE format
Syntax
L<name> <plus> <minus> [model name]
+ <inductance> [IC=<initial current>]

Examples
L1 2 3 1U
L2 7 8 120P IC=2

<plus> and <minus> are the positive and negative node numbers.
Positive current flows into the plus node and out of the minus node.
Schematic format
PART attribute
<name>
Example
L1
INDUCTANCE attribute
<inductance> [LOT[/#][/GAUSS|/UNIFORM|WCASE] = <tolerance>[%]]
[IC=<initial current>]

Examples
1U LOT=10% IC=.002
1U/(1+I(L2)^2)
FLUX attribute
[flux]

Example
1u*ATAN(I(L2))
FREQ attribute
[fexpr]
Example
1.2mh+5m*(1+log(F))

MODEL attribute
[model name]
Examples
LMOD
INDUCTANCE attribute
[inductance] may be either a simple number or an expression involving
time-varying variables. Consider the following expression:
1u/(1+I(L2)^2)
I(L2) refers to the value of the L2 current, during a transient analysis, a DC

operating point calculation prior to an AC analysis, or during a DC analysis.
It does not mean the AC small signal L2 current. If the operating point value
for the L2 current was 2 amps, the inductance would be evaluated as
1u/(1+2^2) = 0.2u. The constant value, 0.2u, is then used in AC analysis.
FLUX attribute
[flux], if used, must be an expression involving time-domain variables,
including the current through the inductor and possibly other symbolic
(.define or .param) variables.
Rules about using FLUX and INDUCTANCE expressions

1) Either [inductance] or [flux] must be supplied.
2) If both [inductance] and [flux] are time-varying expressions, the user

must ensure that [inductance] is the derivative of [flux] with respect to the
inductor current.
[inductance] = d([flux])/dI
3) If [inductance] is not given and [flux] is a time-varying expression, MC12

will create an expression for inductance by taking the derivative, L = dX/dI.
4) If [inductance] is a time-varying expression, and [flux] is not given, MC12

will create an equivalent circuit for the inductor consisting of a voltage
source of value L(I)*DDT(I).
5) If [flux] is a time-varying expression, it must involve the current through

the inductor. Even for a constant inductor, X(L) = L*I(L).
605
6) If [inductance] or [flux] is given as a time-varying expression, the
MODEL attribute is ignored and the inductor cannot be referenced by a K
device (mutual inductance). The inductance and flux values are determined
solely by the expressions and are unaffected by model parameters.
Time-varying expression means any expression that uses a variable that can
vary during a simulation run, such as V(L1) or I(L2).
FREQ attribute
If <fexpr> is used, it replaces the value determined during the operating
point. <fexpr> may be a simple number or an expression involving frequency
domain variables. The expression is evaluated during AC analysis as the
frequency changes. For example, suppose the <fexpr> attribute is this:
10mh+I(L1)*(1+1E-9*f)/5m

In this expression, F refers to the AC analysis frequency variable and I(L1)
refers to the AC small signal current through inductor L1. Note that there
[inductance] will be used in transient analysis.
Initial conditions
The initial condition assigns an initial current through the inductor in transient
analysis if no operating point is done (or if the UIC flag is set).
Stepping effects
Both the INDUCTANCE attribute and all of the model parameters may be
stepped. If the inductance is stepped, it replaces [inductance], even if it is an
expression. The stepped value may be further modified by the quadratic and tem-
perature effects.
Quadratic effects
If [model name] is used, [inductance] is multiplied by a factor, QF, which is a
quadratic function of the time-domain current, I, through the inductor.
QF = 1+ IL1•I + IL2•I2
This is intended to provide a subset of the old SPICE 2G POLY keyword, which
is no longer supported.
Temperature effects
The temperature factor is computed as follows:
If [model name] is used, [inductance] is multiplied by a temperature factor, TF.
example, a spec of 200 ppm/degree C would produce a TC1 value of 2E-4.
nominal inductance was measured. T is set to the analysis temperature from the
value, which can be overridden with a .OPTIONS statement. T and Tnom may
be changed for each model by specifying values for T_MEASURED, T_ABS,
T_REL_GLOBAL, and T_REL_LOCAL. See the .MODEL section of the Com-
mand Statements chapter, for more information on how device operating tem-
peratures and Tnom temperatures are calculated.
Monte Carlo effects

LOT and DEV Monte Carlo tolerances, available only when [model name] is
except the T_parameters. Both forms are converted to an equivalent tolerance
factor, MF, which ultimately multiplies the value of the model parameter L.
The final inductance, lvalue, is calculated as follows:
lvalue = [inductance] * QF * TF * MF* L, where L is the model parameter

multiplier.
Nonlinear inductor cores and mutual inductance

Use the coupling (K) device to specify a nonlinear magnetic material for the core
using the Jiles-Atherton model.

.MODEL <model name> IND ([model parameters])
607
Examples
.MODEL LMOD IND (L=2.0 LOT=10% RS=5m IL1=2E-3 IL2=.0015)
.MODEL L_W IND (L=1.0 LOT=5% DEV=.5% CP=1.8p T_ABS=37)
Model parameters
L Inductance multiplier 1
CP Parallel capacitance F 0
RS Series resistance Ohms 0
IL1 Linear current coefficient A-1 0
IL2 Quadratic current coefficient A-2 0
TC1 Linear temperature coefficient °C-1 0
TC2 Quadratic temperature coefficient °C-2 0
The model used for the inductor when the parasitic values of CP and RS are
specified is as follows:
Figure 26-17 Inductor model
Noise effects
The parasitic resistance RS, if specified, generates a thermal noise current in the
usual manner, IRS = (4•k•T / RS)0.5

Isource
Schematic format
PART attribute
<name>
Examples
I1
CURRENT_SOURCE
VALUE attribute
<value>[LOT[/#][/GAUSS|/UNIFORM|WCASE]=<tolvalue>[%]]

Examples
5
10u LOT/1/Gauss=10%
10m
609
JFET
SPICE format
Syntax
J<name> <drain> <gate> <source> <model name>
+ [area] [OFF] [IC=<vds>[,vgs]] [M=<multiplier>]
Example
J1 5 7 9 2N3531 1 OFF IC=1.0,2.5
Schematic format
PART attribute
<name>
Example
J1
VALUE attribute
[area] [OFF] [IC=<vds>[,vgs]] [M=<multiplier>]
Example
1.5 OFF IC=0.05,1.00
MODEL attribute
<model name>
Example
JFET_MOD
The value of [area], whose default value is 1, multiplies or divides parameters as

shown in the table. The [OFF] keyword turns the JFET off for the first operating
point iteration. The initial condition, [IC= <vds>[,vgs]], assigns initial drain-
source and gate-source voltages. <multiplier> is the number of parallel devices.
Negative VTO implies a depletion mode device and positive VTO implies an
enhancement mode device. This conforms to the SPICE 2G.6 model. Additional
information on the model can be found in reference (2).

.MODEL <model name> NJF ([model parameters])
.MODEL <model name> PJF ([model parameters])
Examples
.MODEL J1 NJF (VTO=-2 BETA=1E-4 LAMBDA=1E-3)
.MODEL J2 PJF (VTO= 2 BETA=.005 LAMBDA=.015)
Model Parameters
Name Parameter Units Def. Area
AF Flicker-noise exponent 1.00

ALPHA Ionization coefficient 0.00
BETA Transconductance parameter A/V2 1E-4 *
BETATCE BETA exp. temperature coefficient %/°C 0.00
CGD Zero-bias gate-drain junction cap. F 0.00 *
CGS Zero-bias gate-source junction cap. F 0.00 *
IS Gate junction saturation current A 1E-14 *
ISR Gate p-n recombination current A 0.00 *
KF Flicker-noise coefficient 0.00
LAMBDA Channel-length modulation V-1 0.00
M Gate junction grading coefficient 0.50
N Gate p-n emission coefficient 1.00
NR Emission coefficient for ISR 2.00
PB Gate junction potential V 1.00
RD Drain ohmic resistance Ω 0.00 /
RS Source ohmic resistance Ω 0.00 /
T_REL_LOCAL Relative to AKO temperature °
VK Ionization knee voltage V 0.00
VTO Threshold voltage V -2.00
VTOTC VTO temperature coefficient V/°C 0.00
XTI IS temperature coefficient 3.00
611
Model equations
Figure 26-18 JFET model

Parameters BETA, CGS, CGD, and IS are multiplied by [area] and parameters
RD and RS are divided by [area] prior to their use in the equations below.

Id = Drain current
Temperature Dependence
.OPTIONS statement. Both T and Tnom may be customized for each model by
are calculated.
VTO(T) = VTO + VTOTC•(T-Tnom)

BETA(T) = BETA•1.01BETACE•(T-Tnom)
IS(T) = IS•e1.12•(T/Tnom-1)/VT•(T/Tnom)XTI
EG(T) = 1.16 - .000702•T2/(T+1208)
PB(T) = PB•( T/Tnom)- 3•VT•ln((T/Tnom))-EG(Tnom)•(T/Tnom)+EG(T)
CGS(T) = CGS•(1+M•(.0004•(T-Tnom) + (1 - PB(T)/PB)))
CGD(T) = CGD•(1+M•(.0004•(T-Tnom) + (1 - PB(T)/PB)))

Current equations
Gate recombination and ionization current (IGD and IGS)

IG = IGS + IGD
IGS = gate-source leakage current = In + Ir·Kg

In = normal current = IS·(exp(Vgs/(N·Vt))-1)
Ir = recombination current = ISR·(exp(Vgs/(NR·Vt))-1)
Kg = generation factor = ((1-Vgs/PB)2+0.005)M/2
IGD = gate-drain leakage current = In + Ir·Kg + Ii

In = normal current = IS·(exp(Vgd/(N·Vt))-1)
Ir = recombination current = ISR·(exp(Vgd/(NR·Vt))-1)
Kg = generation factor = ((1-Vgd/PB)2+0.005)M/2
Ii = impact ionization current
if(0 < Vgs-VTO < Vds)

Forward saturation region:
vdif = Vds - (Vgs-VTO)
Ii = Idrain·ALPHA·vdif·exp(-VK/vdif)
else
Ii = 0
IDRAIN (current through the IDRAIN source)

IDRAIN = 0
Saturation Region : Vds > Vgs - VTO(T)
IDRAIN = BETA(T)•(Vgs - VTO(T))2•(1+LAMBDA•Vds)
Linear Region : Vds < Vgs - VTO(T)
IDRAIN = BETA(T)•Vds•(2•(Vgs - VTO(T))- Vds)•(1+LAMBDA•Vds)
Total non-capacitive current

Gate current = area·(IG)
Drain current = area·(IDRAIN - IGD)
Source current = area·(-IDRAIN - IGS)
613
If Vgs ≤ FC • PB(T) then
Cgs = area· CGS(T)/(1 - Vgs/PB(T))M
Else
Cgs = area· CGS(T)•(1 - FC•(1+M)+M•(Vgs/PB(T)))/ (1 - FC) (1+M)
If Vgd ≤ FC • PB(T) then

Cgd = area· CGD(T)/(1 - Vgd/PB(T))M
Else
Cgd = area· CGD(T)•(1 - FC•(1+M)+M•(Vgd/PB(T)))/ (1 - FC) (1+M)
Noise
The resistors RS and RD generate thermal noise currents.
Ird2 = 4•k•T / (RD / area)

Irs2 = 4•k•T / (RS / area)
The drain current generates a noise current.
I2 = 4•k•T•gm•2/3 + KF•IdAF / Frequency

where gm = ∂Id / ∂Vgs (at the operating point)

K (Mutual inductance / Nonlinear magnetics model)
SPICE formats
K<name> L<inductor name> <L<inductor name>>*
+ <coupling value>
K<name> L<inductor name>* <coupling value>

+ <model name>
Examples
K1 L1 L2 .98
K1 L1 L2 L3 L4 L5 L6 .98
Schematic format
PART attribute
<name>
Example
K1
INDUCTORS attribute
<inductor name> <inductor name>*
Example
L10 L20 L30
COUPLING attribute
<coupling value>
Example
0.95
MODEL attribute
[model name]
Example
K_3C8
If <model name> is used, there can be a single inductor name in the INDUC-
TORS attribute. If model name is not used, there must be at least two inductor
names in the INDUCTORS attribute.
615
The K device specifies a mutual inductance between two or more linear induc-
tors. You can specify a nonlinear magnetic core by using a MODEL attribute.
A resistive impedance of 1/GMIN is added between the positive pins of the cou-
pled inductors to avoid DC convergence problems.
Coupled linear inductors

In this mode, the K device provides a means to specify the magnetic coupling
between multiple inductors. The equations that define the coupling are:

Vi = Li Ii' + Mij Ij' + Mik Ik' + ...
where Ii is the current flowing into the plus lead of the i'th inductor and Ii' is its
time derivative. For linear inductors, <model name> is not used.
Nonlinear magnetic core(s)

If a <model name> is supplied, the following things change:
1. The linear K device becomes a nonlinear magnetic core. The model for the
core is a variation of the Jiles-Atherton model.
2. Inductors are interpreted as windings and each inductor <value> is interpreted

as the number of turns for the winding. In this case, <value> must be a constant
whole number. It may not be an expression.
3. The list of coupled inductors may contain just one inductor. Use this method to
create a single magnetic core device, not coupled to another inductor.
4. A model statement is required to define the model parameters or <model

name> must be in the model library referenced by .LIB statements.
The nonlinear magnetics model is based on the Jiles-Atherton model. This model
is based upon contemporary theories of domain wall bending and translation. The
anhysteretic magnetization curve is described using a mean field approach. All
magnetic domains are coupled to the bulk magnetization and magnetic fields. The
anhysteretic curve is regarded as the magnetization curve that would prevail if
there were no domain wall pinning. Of course, such pinning does occur, mainly
at defect sites. The hysteresis effect that results from this pinning is modeled as a
simple frictional force, characterized by a single constant, K. The resulting state
equation produces a realistic ferromagnetic model.

The Core is modeled as a state-variable nonlinear inductor. Micro-Cap solves a
differential equation for the B and H fields and derives the terminal current and
voltage from these values. The B field and H field values are available as output
variables in transient analysis.
Plotting B and H fields

B(L1) Plots the B field of inductor L1 in CGS units of Gauss.
H(L1) Plots the H field of inductor L1 in CGS units of Oersteds.
BSI(L1) Plots the B field of inductor L1 in SI units of Teslas.

HSI(L1) Plots the H field of inductor L1 in SI units of Amps/Meter.
To place a single core in a circuit, use this procedure:
1. Place an inductor in the circuit with these attributes:

PART L1
INDUCTANCE 1
The value 1 represents the number of turns.
2. Place a K device in the circuit with these attributes:

PART K1
INDUCTORS L1
COUPLING 1.0
MODEL KCORE
Step 2 changes the inductor L1 from a standard linear model to a nonlinear

core whose properties are controlled by the model statement. See the circuit file
CORE for an example of a single core device and how to do BH loop plots.
To place two magnetically coupled cores in a circuit, use this procedure:
1. Place the first inductor in the circuit with these attributes:

PART L1
INDUCTANCE Number of primary turns
2. Place the second inductor in the circuit with these attributes:

PART L2
INDUCTANCE Number of secondary turns
3. Place a K device in the circuit with these attributes:

PART K2
617
INDUCTORS L1 L2
COUPLING Coupling coefficient between L1 and L2 (0-1.0)
MODEL KCORE
This procedure creates two coupled cores whose magnetic properties are con-
trolled by the KCORE model statement. See the sample circuit file CORE3 for an
example of multiple, coupled core devices.

.MODEL <model name> CORE ([model parameters ])
Examples
.MODEL K1 CORE (Area=2.54 Path=.54 MS=2E5)
.MODEL K2 CORE (MS=2E5 LOT=25% GAP=.001)
Model parameters
Area Mean magnetic cross-section cm2 1.00
Path Mean magnetic path length cm 1.00
Gap Effective air gap length cm 0.00
MS Saturation magnetization a/m 4E5
A Shape parameter a/m 25
C Domain wall flexing constant .001
K Domain wall bending constant 25
Note that the model parameters are a mix of MKS or SI units (a/m) and CGS
units (cm and cm2).
Model Equations
Definitions and Equations
All calculations are done in MKS (SI) units
µ0 = magnetic permeability of free space = 4*PI*1e-7 Webers/Amp-meter
N= number of turns
Ma = Anhysteretic magnetization
H = Magnetic field intensity inside core
B = Magnetic flux density inside core
M = Magnetization due to domain alignment
I = Core current
V = Core voltage
Ma = MS • H/( |H| +A)
Sign = K if dH/dt > 0.0
Sign= - K if dH/dt <= 0.0
dM/dH = (Ma-M) / ((Sign)•(1+C)) + (C / (1+C))•dMa/DH
B = µ0 • (H +M)
L = µ0 • (1+dM/dH) • N2 • AREA / PATH
V = L • dI/dt
To derive model parameters from data sheet values, use the MODEL program.
Doing it manually requires this procedure:
1. Many data sheets provide the value of Bsat in Gauss. To calculate the required
value of MS in units of Amps/meter, multiply the Bsat value in Gauss by 79.577.
This yields the required model value for MS in Amps/meter.
2. Run the sample circuit CORE.CIR and adjust the values of A, K, and C, to fit
the data sheet BH curve. The effect of increasing each parameter is as follows:
Parameter µ HC BR
A - + +
K + +
C + - -
where µ is the slope or permeability, HC is the coercive force value, and BR is

the remanence.
Figure 26-19 Typical BH loop
619
Laplace sources
Schematic format
PART attribute
<name>
Examples
FIL1
LOW1
LAPLACE attribute of LFIOFI, LFIOFV, LFVOFV, LFVOFI

<expression>
Example
1/(1+.001*S+1E-8*S*S)
FREQ attribute of LTIOFI, LTIOFV, LTVOFV, LTVOFI

<(f1,m1,p1) (f2,m2,p2)...(fn,mn,pn)>
Example
(0.0,1.0,0.0) (1Meg,0.9,-10) (10Meg,0.2,-35)
KEYWORD attribute (for use with FREQ attribute)

[[DB | MAG] [DEG | RAD]] | [R_I]
Examples
DB RAD
MAG DEG
R_I
There is no SPICE version of this source. Use the Dependent source, E or G de-
vice.
The keywords DB, MAG, DEG, RAD, R_I are interpreted as follows:
DB: Magnitude value is expressed in decibels. (default)

MAG: Magnitude value is true magnitude.
DEG: Degrees value is expressed in degrees. (default)
RAD: Degrees value is expressed in radians.
R_I: The table contains real and imaginary parts.

Laplace sources are characterized by a linear transfer function. The two basic
types are distinguished by the way the transfer function is calculated. The For-
mula type uses an algebraic expression to describe the transfer function in terms
of the complex frequency variable, S. The Table type uses a table of ordered data
triplets which describe the transfer function. Each data triplet comprises the fre-
quency, magnitude, and phase of the transfer function.
In AC analysis, the value of the transfer function is computed from the algebraic
expression involving S, where S = 2•π•frequency•j, or obtained by interpolation
from the given table.
For DC analysis, the value of the transfer function is computed from the given
algebraic expression involving S, where S = 0, or obtained from the table, using
the lowest frequency data point supplied.
For transient analysis, it is necessary to first determine the impulse response of

the function. The impulse response is obtained by performing an inverse Fourier
transform on the transfer function. Then, during the transient run, the output of
the source is obtained from the convolution of the waveform at the source input
nodes and the impulse response waveform. This allows the source to accurately
respond to any input waveform, not just simple, predefined waveforms.
The accuracy of this procedure is limited by the number of time points in the
impulse response, or alternatively, by the bandwidth of the function. As a practi-
cal matter, no more than 8192 time points should be computed for the impulse
response, due to memory and time limitations. The actual number of time points,
N, is a logarithmic function of the value of RETOL.

N=2 6-log10(RELTOL)
For example, for RELTOL= .001, 512 time points are computed.
As a general rule, Laplace sources will give the best transient analysis results on
narrow band functions. Wide band functions, such as the differentiator, f(s)=s,
and the integrator, f(s)=1/s, are best modeled by using discrete components. See
the sample circuits INT (integrator macro) and DIF(differentiator macro).
621
Formula types
The input and output variables and definition names for the Laplace formula
sources are as follows:

Current-controlled current source I I LFIOFI
Current-controlled voltage source I V LFVOFI
Voltage-controlled voltage source V V LFVOFV
Voltage-controlled current source V I LFIOFV
Here are some examples:
1/(1+.001*S) a simple low pass filter.

1/(1+.001*s+1E-8*S*S) a second order filter.
exp(-pow((C*S*(R+S*L)),.5)) equation of a simple lossy, transmission
line. R, L, and C are .define constants.
For illustration, see the circuits L1, L2, and L3.
Table types
In a Table type, the transfer function is defined with a table. The table contains
ordered triplets of numbers listing the frequency, magnitude or real value, and
phase or imaginary value of the transfer function. The general form of the table
entries is:
(F1,X1,Y1) (F2,X2,Y2) ... (FN,XN,YN)
Fi is the i’th frequency value in hertz.

Xi is the i’th magnitude value or the real value.
Yi is the i’th phase value or the imaginary value.
There are six rules for forming the table:
1. Values are separated by commas, triplets are enclosed in parentheses and

are separated by spaces.
2. Data triplets must be arranged in order of ascending frequency.

3. The function is constant at X1,Y1 for inputs below F1.
4. The function is constant at XN,YN for inputs above FN.

5. The function is logarithmically interpolated for frequency values between
the table values.

6. The table should contain one data point at DC or zero frequency.
The table may be entered directly as the parameter string or indirectly using the
.define statement. For illustration, see the circuit P1.
The input variable and output variables and definition names are as follows:

Current-controlled current source I I LTIOFI
Current-controlled voltage source I V LTVOFI
Voltage-controlled voltage source V V LTVOFV
Voltage-controlled current source V I LTIOFV
623
Macro
Schematic format
PART attribute
<name>
Example
2N5168
FILE attribute
<macro circuit name>
Example
SCR
Macros are the schematic equivalents of subcircuits. They are circuit building
blocks that have been created and saved to disk for use in other circuits.
To create a macro:
1. Create a circuit. Place grid text on the nodes that you want to make available
as macro pins. If you want to pass numeric parameters to the macro, use sym-
bolic variables for attributes and/or model parameter values and declare these
names in a .parameters statement. Save the circuit to the LIBRARY folder using
the desired macro name with an extension of .MAC.
2. Enter a component in the Component library as follows:
• Enter the macro file name for the Name field.

• Choose a suitable shape.
• Choose Macro for the Definition field.
• Place pins on the shape by clicking in the Shape drawing area. Name the
pins with the same text names you placed on the nodes in the macro circuit.
• Add optional .MACRO statements to one of the *.LIB files to substitute
long parameter lists for shorter names.
Alternatively, create the circuit as in Step 1, then use the make Macro command
at Edit / Box / Make Macro. It automates Step 2.

To use a macro:
Select the macro from the Component library. Place it in the circuit that will
use it and edit its parameters, if it has any. You can also use an alias which,
using a .macro statement, substitutes a short name like 2N5168 for the macro
FILE name and a corresponding set of parameters.
The format of the macro command is:
.MACRO <alias> <macro circuit name(parameter list)>
This statement lets you store the parameters that turn a general macro for, say an
SCR, into a specific model for a specific part like the 2N5168 SCR, and to access
the part with a simple meaningful name, like 2N5168. For more information on
the .MACRO statement see the Command Statements chapter.
When a macro is placed in a circuit, the program reads the macro circuit file,
determines if it has parameters from the .PARAMETERS statement in the macro
circuit file and shows these parameters and their default values in the Attribute
dialog box. Edit the parameter values from their default values to those you want.
625
MOSFET
SPICE format
Syntax
M<name> <drain> <gate> <source> <bulk> <model name>
+ [M=<mval>]
+ [L=<length>] [W=<width>] [AD=<drainarea>] [AS=<sourcearea>]
+ [PD=<drainperiphery>] [PS=<sourceperiphery>]
+ [NRD=<drainsquares>] [NRS=<sourcesquares>]
+ [NRG=<gatesquares>] [NRB=<bulksquares>]
+ [OFF][IC=<vds>[,vgs[,vbs]]]
Example
M1 5 7 9 0 IRF350 L=1.5E-6 W=0.25 OFF IC=25.0,8.0
Schematic format
PART attribute
<name>
Example
M1
VALUE attribute
[M=<mval>]
Examples
M=20 NRD=10 NRS=25 NRG=5
L=.35u IC=.1, 2.00
L=.4u W=2u OFF IC=0.05,1.00
MODEL attribute
<model name>
Examples
IRF350
MM150
Supported Models
Level Model Name

1 Schichman-Hodges
2 MOS2 Grove-Frohman model (SPICE 3F5)
3 MOS3 empirical model (SPICE 3F5)
4 BSIM1 Original BSIM model (Berkeley Short Channel IGFET)
5 BSIM2 Second generation BSIM model
8, 49 or 53 BSIM3 Third generation BSIM3v3.3 (7/29/2005)
14 or 54 BSIM4 Fourth generation BSIM4.8.1 (2/15/2017)
44 EKV 2.6 Charge based-short channel model developed by the
Swiss Institute of Technology. See the MOSFET(EKV) section
following this MOSFET section
Various Philips MOSFET models. See the Philips MOSFET section
following the EKV section.
<width> and <length> are the drawn device dimensions, before side diffusion, in
meters. They can be specified as device attributes, model parameters, or by us-
ing the global default values, DEFW and DEFL. For Levels 1-3, and BSIM 1-3,
DEFNRD and DEFNRS define default drain and source squares, and DEFPD and
DEFPS default drain and source periphery. Device attributes supersede model
parameters, which supersede the global values from the Global Settings dialog
box or a local .OPTIONS statement.
The initialization [IC=<vds>[,vgs[,vbs]]] assigns initial voltages to the drain-

source, gate-source, and body-source terms in transient analysis if no operating
point is done (or if the UIC flag is set). The [OFF] keyword forces the device off
during the first iteration of the DC operating point.
<sourceperiphery> and <drainperiphery> are the diffusion peripheries (m).

<sourcearea> and <drainarea> are the diffusion areas (sq. m). Source and drain
junction capacitances may be specified directly by the model parameters CBS
and CBD. If absent, they are calculated from area and periphery terms.
The parasitic resistances may be specified directly with the model parameters
RS, RD, RG, and RB. If unspecified, they are calculated from the product of
the sheet resistivity, RSH, and the number of squares terms, <drainsquares>,
<sourcesquares>, <gatesquares>, and <bulksquares>. If these terms are absent,
or zero, and the model parameters RS, RD, RG, and RB are absent or zero,
then the parasitic resistances are not included in the model. <drainsquares> and
<sourcesquares> default to 0. The other parameter line values also default to
zero.
627
<mval> is a multiplier (default = 1) that provides a way to simulate the effect of
paralleling many devices. It multiplies the effective width, overlap, and junction
capacitances, and the junction currents. It multiplies the drain and source areas,
the device width, and the two peripheries, and divides the four resistances RS,
RD, RG, and RB.

.MODEL <model name> NMOS ([model parameters])
.MODEL <model name> PMOS ([model parameters])
Examples
.MODEL M1 NMOS (W=0.2 L=0.8U KP=1E-6 GAMMA=.65)
.MODEL M2 PMOS (W=0.1 L=0.9U KP=1.2E-6 LAMBDA=1E-3)
Common model parameters

These model parameters are common to levels 1, 2, 3, 4, 5, and 8: Levels 1-5
share common default values. Level 8 defaults are shown in the last column.
Name Parameter Units Default Values For

Lev 1-5 Lev 8
LEVEL Model level 1-5 8
L Channel length m DEFL DEFL
W Channel width m DEFW DEFW
RDS Drain-source shunt resistance Ω ∞ ∞
RD Drain ohmic resistance Ω 0.00 0.00
RS Source ohmic resistance Ω 0.00 0.00
RG Gate ohmic resistance Ω 0.00 0.00
RB Bulk ohmic resistance Ω 0.00 0.00
RSH Source and drain sheet res. Ω/sq 0.00 0.00
CGDO Gate-drain overlap cap. F/m 0.00 0.00
CGSO Gate-source overlap cap. F/m 0.00 0.00
CGBO Gate-bulk overlap cap. F/m 0.00 0.00
CBD Bulk p-n zero-bias B-D cap. F 0.00 0.00
CBS Bulk p-n zero-bias B-S cap. F 0.00 0.00
CJ Bulk p-n zero-bias bot. cap. F/m2 0.00 5E-4
CJSW Bulk p-n zero-bias s/w cap. F/m 0.00 5E-10
MJ Bulk p-n zero-bias bottom grad. 0.50 0.50
MJSW Bulk p-n zero-bias s/w coeff. 0.33 0.33
TT Bulk p-n transit time S 0.00 0.00
IS Bulk p-n saturation current A 1E-14 1E-14
N Bulk p-n emission coefficient 1.00 1.00
JS Bulk p-n bot. current density A/m2 0.00 1E-4
Name Parameter Units Default Values For
Lev 1-5 Lev 8
JSW or JSSW Bulk p-n bot. current density A/m 0.00 0.0
PB Bulk p-n bottom potential V 0.80 1.00
PBSW Bulk p-n sidewall potential V PB 1.00
KF Flicker-noise coefficient 0.00 0.00
AF Flicker-noise exponent 1.00 1.00
FC Forward-bias depletion coeff. 0.50 0.50
GDSNOI Channel shot noise coefficient 1.0 1.0
NLEV Noise equation selector 2.0 None
T_REL_GLOBAL Relative to current temp. °C

In addition to the 33 common parameters, the following table lists the additional
parameters used in the level 1, 2, and 3 models.
Name Parameter Units Default Level

LD Lateral diffusion length m 0.00 1,2,3
WD Lateral diffusion width m 0.00 1,2,3
KP Process transconductance A/V2 2E-5 1,2,3
VTO Zero-bias threshold voltage V 0.00 1,2,3
GAMMA Body-effect coefficient V0.5 0.00 1,2,3
PHI Surface inversion potential V 0.60 1,2,3
LAMBDA Channel-length modulation V-1 0.00 1,2
TOX Gate oxide thickness m 1E-7 1,2,3
UO Surface mobility cm2/V/s 600 2,3
NEFF Total channel charge coeff. 1.0 2
NSUB Substrate doping density cm-3 None 2,3
NSS Surface state density cm-2 None 2,3
NFS Fast surface-state density cm-2 None 2,3
XJ Metallurgical junction depth m 0.00 2,3
VMAX Max drift velocity of carriers m/s 0.00 2,3
DELTA Width effect on VTO 0.00 2,3
THETA Mobility modulation V-1 0.00 3
ETA Static feedback on VTO 0.00 3
KAPPA Saturation field factor 0.20 3
TPG Type of gate material 1.00 2,3
UCRIT Mobility degrad. critical field V/cm 1E4 2
UEXP Mobility degradation exponent 0.00 2
UTRA Mobility degrad. tr. field coeff. m/s 0.00 2
629
Model equations for levels 1, 2, and 3
Figure 26-20 Levels 1, 2, 3 MOSFET model

Definitions
Id = Drain current
VT = k•T/q
Temperature effects
to the analysis temperature from the Analysis Limits dialog box. TNOM is deter-
mined by the Global Settings TNOM value, which can be overridden with a .OP-
TIONS statement. T and Tnom may be customized for each model by specifying
the parameters T_MEASURED, T_ABS, T_REL_GLOBAL, and T_REL_LOCAL.
For details on how device temperatures and Tnom temperatures are calculated,
see the .MODEL section of the Command Statements chapter.
EG(T) = 1.16 - .000702•T•T/(T+1208)

IS(T) = IS•e(EG(Tnom)•T/Tnom-EG(T))/VT
JS(T) = JS•e(EG(Tnom)•T/Tnom-EG(T))/VT
JSSW(T) = JSSW•e(EG(Tnom)•T/Tnom-EG(T))/VT
KP(T) = KP•(T/Tnom)-1.5
UO(T) = UO•(T/Tnom)-1.5
PB(T) = PB•(T/Tnom)- 3•VT•ln((T/Tnom))-EG(Tnom)•(T/Tnom)+EG(T)
PBSW(T) = PBSW•(T/Tnom)- 3•VT•ln((T/Tnom))-EG(Tnom)•(T/Tnom)+EG(T)
PHI(T) = PHI•(T/Tnom)- 3•VT•ln((T/Tnom))-EG(Tnom)•(T/Tnom)+EG(T)

CBD(T) = CBD•(1+MJ•(.0004•(T-Tnom) + (1 - PB(T)/PB)))
CBS(T) = CBS•(1+MJ•(.0004•(T-Tnom) + (1 - PB(T)/PB)))
CJ(T) = CJ•(1+MJ•(.0004•(T-Tnom) + (1 - PB(T)/PB)))
CJSW(T) = CJSW•(1+MJ•(.0004•(T-Tnom) + (1 - PB(T)/PB))
The parasitic lead resistances have no temperature dependence.
Current equations
Only the Level 1 N-channel drain equations are shown here. The Level 2 and
Level 3 current equations are too complex for presentation in this manual. Inter-
ested users should consult reference (22) for more information.
K = KP • W/ (L - 2 • LD)
VTH = VTO + GAMMA • ((PHI - VBS)1/2- √(PHI))
Cutoff region: For Vgs < VTH
Id = 0.0
Linear region: For Vgs > VTH and Vds < (Vgs - VTH)
Id = K•(Vgs - VTH - 0.5•Vds)•Vds•(1+LAMBDA•Vds)
Saturation region: For Vgs > VTH and Vds > (Vgs - VTH)
Id = 0.5•K•(Vgs - VTH)2•(1 + LAMBDA•Vds)
Meyer model for gate capacitance
Levels 1-3 use the capacitance model proposed by Meyer. The charges are mod-
eled by three nonlinear capacitances, Cgb, Cgd, and Cgs.
Cox = COX • W • Leff
Accumulation region (Vgs < Von - PHI)

For Vgs < Von - PHI,
Cgb = Cox + CGBO • Leff
Cgs = CGSO • W
Cgd = CGDO • W
631
Depletion region (Von - PHI < Vgs < Von)
Cgb = Cox • (Von - Vgs)/PHI + CGBO • Leff
Cgs = 2/3 • Cox • ((Von - Vgs)/PHI + 1) + CGSO • W
Cgd = CGDO • W
Saturation region (Von < Vgs < Von + Vds)

Cgb = CGBO • Leff
Cgs = 2/3 • Cox + CGSO • W
Cgd = CGDO • W
Linear region:
For Vgs > Von + Vds,
Cgb = CGBO • Leff
Cgs = Cox • (1 - ((Vgs - Vds - Von)/(2•(Vgs - Von) - Vds))2) + CGSO • W
Cgd = Cox • (1 - ((Vgs - Von)/(2•(Vgs - Von) - Vds))2) + CGDO • W
Junction capacitance
If CBS=0 and CBD=0 then
Cbs = CJ(T)•AS•f1(VBS) + CJSW(T)•PS•f2(VBS) + TT•GBS
Cbd = CJ(T)•AD•f1(VBD) + CJSW(T)•PD•f2(VBD) + TT•GBD
else
Cbs = CBS(T)•f1(VBS) + CJSW(T)•PS•f2(VBS) + TT•GBS
Cbd = CBD(T)•f1(VBD) + CJSW(T)•PD•f2(VBD) + TT•GBD
GBS= DC bulk-source conductance = d(IBS)/d(VBS)

GBD= DC bulk-drain conductance = d(IBD)/d(VBD)
If VBS ≤ FC • PB(T) then

f1(VBS) = 1/(1 - VBS/PB(T))MJ
Else
f1(VBS) = (1 - FC•(1+MJ)+MJ•(VBS/PB(T)))/ (1 - FC) (1+MJ)
If VBS ≤ FC • PBSW(T) then

f2(VBS) = 1/ (1 - VBS/PBSW(T))MJSW
Else
f2(VBS) = (1 - FC•(1+MJSW)+MJSW•(VBS/PBSW(T)))/ (1 - FC) (1+MJSW)
If VBD ≤ FC • PB(T) then

f1(VBD) = 1/(1 - VBD/PB(T))MJ
Else
f1(VBD) = (1 - FC•(1+MJ)+MJ•(VBS/PB(T)))/ (1 - FC) (1+MJ)

If VBD ≤ FC • PBSW(T) then
f2(VBD) = 1/ (1 - VBD/PBSW(T))MJSW
Else
f2(VBD) = (1 - FC•(1+MJSW)+MJSW•(VBD/PBSW(T)))/ (1 - FC) (1+-MJSW)
Model parameters for BSIM1 level 4

These are the model parameters for the BSIM1 model, level 4. There are no de-
fault values. All parameter values must be specified. All parameters are binnable
except those marked with an asterisk.
Name Parameter Units

DL* Channel length reduction µ
DW* Channel width reduction µ
TOX* Gate oxide thickness µ
VDD* Supply voltage for MUS
XPART* Channel charge partitioning flag
DELL* Unused: Length reduction of S-D diff.
VFB Flat band voltage V
PHI Strong inversion surface potential V
K1 Bulk effect coefficient 1 √V
K2 Bulk effect coefficient 2
ETA VDS dependence of threshold voltage
X2E VBS dependence of ETA V-1
X3E VDS dependence of ETA V-1
MUZ* Mobility at vds=0,vgs=vth cm2/Vs
X2MZ VBS dependence of MUZ
MUS Mobility at vds=vdd,vgs=vth
X2MS VBS dependence of MUS
X3MS VDS dependence of MUS
U0 VGS dependence of mobility
X2U0 VBS dependence of U0
U1 VDS dependence of mobility
X2U1 VBS dependence of U1
X3U1 VDS dependence of U1
N0 Subthreshold slope
NB VBS dependence of subthreshold slope
ND VDS dependence of subthreshold slope
633
These are the model parameters for the BSIM 2 model, level 5. All parameters
are binnable except those marked with an asterisk.

DL* Channel length reduction µ 0.00
DW* Channel width reduction µ 0.00
TOX* Gate oxide thickness µ 0.00
VFB Flat band voltage V -1.00
PHI Strong inversion surface potential V 0.75
K1 Bulk effect coefficient 1 √V 0.80
K2 Bulk effect coefficient 2 0.00
ETA0 VDS dependence of threshold voltage 0.00
ETAB VBS dependence of ETA0 V-1 0.00
MU0* Mobility at vds=0,vgs=vth m2/V•s 400
MU0B VBS dependence of MU0 m2/V2•s 0.00
MUS0 Mobility at vds=vdd,vgs=vth m2/V•s 500
MUSB VBS dependence of MUS m2/V2•s 0.00
MU20 VDS dependence of MU in tanh term m2/V2•s 1.5
MU2G VGS dependence of MU2 m2/V3•s 0.00
MU30 VDS dependence of MU in linear term m2/V2•s 10.0
MU40 VDS dependence of MU in quad. term m2/V4•s 0.00
UA0 Linear VGS dependence of mobility m2/V2•s 0.20
UAB VBS dependence of UA m2/V3•s 0.00
UB0 Quadratic VGS dependence of mobility m2/V3•s 0.00
UBB VBS dependence of UB m2/V4•s 0.00
U10 VDS dependence of mobility m2/V2•s 0.10
U1B VBS dependence of U1 m2/V3•s 0.00
U1D VDS dependence of U1 m2/V3•s 0.00
N0 Subthreshold slope at vds=0,vbs=0 1.40
NB VBS dependence of N √V 0.50
ND VDS dependence of N V-1 0.00
VOF0 Threshold voltage offset at vds=0,vbs=0 V 1.80
VOFB VBS dependence of VOF 0.00
VOFD VDS dependence of VOF 0.00
AI0 Pre-factor of hot electron effect 0.00


AIB VBS dependence of AI V-1 0.00
BI0 Exponential factor of hot electron effect 0.00
BIB VBS dependence of BI V-1 0.00
VGHIGH Upper bound of cubic spline function µ•V 0.20
VGLOW Lower bound of cubic spline function V -0.15
VDD* Maximum VDS V 5.00
VGG* Maximum VGS V 5.00
VBB* Maximum VBS V 5.00
XPART* Channel charge partitioning flag 0.00
DELL* Unused: length reduction of S-D diff. m 0.00
635
Model parameters for BSIM3 level 8 and 49
These are the model parameters for the BSIM 3 model. This is the Berkeley
BSIM3v3.3, dated 7/29/2005. All parameters are binnable except those marked
with an asterisk.

Level* 8 Must be 8 or 49
A0 1 Bulk charge effect coefficient
A1 0 First non-saturation effect parameter
A2 1 Second non-saturation effect parameter
ACDE 1 Exponential coefficient for finite
charge thickness
ACNQSMOD* 0 AC NQS model selector
AF* 1 Flicker noise exponent for noimod = 1
AGS 0 Gate bias coefficient for channel length
ALPHA0 0 First impact ionization current parameter
ALPHA1 0 Second impact ionization current parameter
AT 33K Saturation velocity temperature coefficient
B0 0 Bulk charge effect coefficient for channel width
B1 0 Bulk charge effect width offset
BETA0 30 Second parameter for impact ionization current
BINFLAG* 0 Determines binning method: 1=HSpice binning,
0=Berkeley SPICE binning.
BINUNIT* 1 Bin units selector
CAPMOD* 3 Flag for short channel capacitance model
CBD* 0 Zero bias B-D junction capacitance
CBS* 0 Zero bias B-S junction capacitance
CDSC 240u Drain/source to channel coupling capacitance
CDSCB 0 Body-bias sensitivity of CDSC
CDSCD 0 Drain-bias sensitivity of CDSC
CF None Fringing field capacitance
CGBO* None Gate-bulk overlap capacitance/area
CGDL 0 Non-LDD region drain-gate overlap capacitance
per unit channel length
CGDO* None Gate-drain overlap capacitance/area
CGSL 0 Non-LDD region source-gate overlap
capacitance per unit channel length
CGSO* None Gate-source overlap capacitance/area
CIT 0 Interface trap capacitance
CJ* 500u Bulk bottom zero-bias capacitance/area
CJSW* 500p Bulk sidewall zero-bias capacitance/area


CJSWG* CJSW S/D gate sidewall junction capacitance
CKAPPA 600m Coefficient for lightly doped region overlap
fringing field capacitance
CLC 100n Constant term for short channel model
CLE 600m Exponential term for short channel model
DELTA 10m Effective Vds parameter
DLC* LINT Length offset fitting parameter from C-V
DROUT 560m L dependence coefficient of the DIBL
correction parameter in ROUT
DSUB DROUT DIBL coefficient exponent in subthreshold
region
DVT0 2.2 First coefficient of short-channel effect on
threshold voltage
DVT0W 0 First coefficient of narrow width effect on Vth
for small channel length
DVT1 530m Second coefficient of short-channel effect on
threshold voltage
DVT1W 5.3MEG Second coefficient of narrow width effect on
Vth for small channel length
DVT2 -32m Body bias coefficient of short-channel effect on
threshold voltage
DVT2W -32m Body bias coefficient of narrow width effect on
Vth for small channel length
DWB 0 Coefficient of Weff's substrate body bias
dependence
DWC* WINT Width offset fitting parameter from C-V
DWG 0 Coefficient of Weff's dependence
EF* 1 Frequency exponent for noise model = 2
ELM 5 Elmore constant of the channel
EM* 41MEG Saturation field
ETA0 80m DIBL coefficient in subthreshold region
ETAB -70m Body bias coefficient for the subthreshold DIBL
coefficient
GAMMA1 0 Body effect coefficient near the interface
GAMMA2 0 Body effect coefficient in the bulk
GDSNOI* 1 Channel shot noise coefficient
IJTH* 100m Diode limiting current
JS* 100u Bulk bottom saturation current density
637

JSW* 0 Bulk sidewall saturation current density
K1 0 First-order body effect coefficient
K2 0 Second-order body effect coefficient
K3 80 Narrow width effect coefficient
K3B 0 Body effect coefficient of K3
KETA -47m Body bias coefficient of the bulk charge effect
KF* 0 Flicker noise coefficient for noimod = 1
KT1 -120m Temperature coefficient for threshold voltage
KT1L 0 Channel length sensitivity of temperature
coefficient for threshold voltage
KT2 22m Body bias coefficient of the threshold voltage
temperature effect
L* 0 Channel length in meters
LINT* 0 Length offset fitting parameter from I-V without
bias
LINTNOI* 0 Lint offset for noise calculation
LL* 0 Coeff. for length dependence of length offset
LLC* LL Length reduction parameter for CV
LLN* 1 Power of length dependence for length offset
LMAX* None Maximum channel length for HSpice binning
LMIN* None Minimum channel length for HSpice binning
LREF* None Length reference for HSpice binning
LW* 0 Coeff. for width dependence for length offset
LWC* LW Length reduction parameter for CV
LWL* 0 Coefficient of length and width cross term for
length offset
LWLC* LWL Length reduction parameter for CV
LWN* 1 Power of width dependence for length offset
MJ* 500m Bulk bottom grading coefficient
MJSW* 330m Bulk sidewall grading coefficient
MJSWG* MJSW S/D gate sidewall junction capacitance grading
coefficient
MOBMOD* 1 Mobility model
MOIN 15 Coefficient for gate-bias dependent surface
NCH 1.7E17 Channel doping concentration
NFACTOR 1 Subthreshold swing factor
NGATE 0 Poly gate doping concentration
NJ* 1 Emission coefficient (takes priority over N)


NLEV* 2 PSpice noise model selector
NLX 174n Lateral non-uniform doping coefficient
NOFF 1 C-V turn-on/off parameter
NOIA* 1E20,9.9E18 Noise parameter A
NOIB* 50K,2.4K Noise parameter B
NOIC* -1.4p,1.4p Noise parameter C
NOIMOD* 1 Flag for noise model
NQSMOD* 0 Flag for NQS model
NSUB 60000T Substrate doping concentration
PARAMCHK* 0 Controls whether checking is on or off
PB* 1 Bulk bottom potential
PBSW* 1 Bulk sidewall potential
PBSWG* PBSW S/D gate sidewall junction built-in potential
PCLM 1.3 Channel length modulation parameter
PDIBLC1 390m First Rout DIBL correction parameter
PDIBLC2 8.6m Second Rout DIBL correction parameter
PDIBLCB 0 Body coefficient of DIBL parameters
PRT 0 Temperature coefficient for RDSW
PRWB 0 Body effect coefficient of RDSW
PRWG 0 Gate bias effect coefficient of RDSW
PSCBE1 424MEG First substrate current body effect parameter
PSCBE2 10u Second substrate current body effect parameter
PVAG 0 Gate dependence of Early voltage
RB* 0 Bulk resistance
RD* 0 Drain resistance
RDS* 0 Drain-source shunt resistance
RDSW 0 Parasitic resistance per unit width
RG* 0 Gate resistance
RS* 0 Source resistance
RSH* 0 Drain-source diffusion sheet resistance
T_ABS* None Absolute temperature
T_MEASURED* None Parameter measurement temperature
T_REL_GLOBAL* None Relative to current temperature
T_REL_LOCAL* None Relative to AKO model temperature
TCJ* 0 Temperature coefficient of CJ
TCJSW* 0 Temperature coefficient of CJSW
TCJSWG* 0 Temperature coefficient of CJSWG
639

TNOM* None Parameter measurement temperature
(if specified this overrides T_MEASURED)
TOX* 15n Gate oxide thickness in meters
TOXM* TOX Gate oxide thickness used in extraction
TPB* 0 Temperature coefficient of PB
TPBSW* 0 Temperature coefficient of PBSW
TPBSWG* 0 Temperature coefficient of PBSWG
TT* 0 Bulk pn transit time
U0 67m,25m Mobility at Temp = TNOM
UA 2.25n First-order mobility degradation coefficient
UA1 4.31n Temperature coefficient for UA
UB .000587f Second-order mobility degradation coefficient
UB1 -.00761f Temperature coefficient for UB
UC None Body effect mobility degradation coefficient
If mobmod=3, default = -.0465 else -4.65e-12
UC1 None Temperature coefficient for UC
If mobmod=3, default = -.0560 else -5.60E-12
UTE -1.5 Mobility temperature exponent
VBM -3 Maximum applied body bias in Vth calculation
VBX 0 Vbs at which the depletion width equals XT
VERSION* 3.3 Model version number, must be 3.3
VFB 0 Flatband voltage for DC I-V calculation
VFBCV -1 Flat-band voltage parameter for CAPMOD=0
VOFF -80m Offset voltage in the subthreshold region for
large W and L
VOFFCV 0 Coefficient for gate-bias dependent surface
VSAT 80K Saturation velocity at Temp = TNOM
VTH0 700m,-700m Threshold voltage at vbs=0 for large channel
length
VTH0M* 0 VT short-distance matching parameter
W* 0 Channel width in meters
W0 2.5u Narrow width effect parameter
WINT* 0 Width offset fitting parameter from I-V without
bias
WL* 0 Coeff. for length dependence of width offset
WLC* WL Width reduction parameter for CV
WLN* 1 Power of length dependence of width offset
WMAX* None Maximum channel width for HSpice binning


WMIN* None Minimum channel width for HSpice binning
WR 1 Width offset from Weff for Rds calculation
WREF* None Width reference for HSpice binning
WW* 0 Coefficient for width dependence of width offset
WWC* WW Width reduction parameter for CV
WWL* 0 Coefficient of length and width cross term for
width offset
WWLC* WWL Width reduction parameter for CV
WWN* 1 Power of width dependence of width offset
XJ 150n Junction depth
XPART* 0 Channel charge partitioning flag
(Neg = none, 0=40/60, 0.5=50/50 1.0=100/0)
XT 155n Doping depth
XTI* 3 Junction current temperature exponent
coefficient
641
These are the model parameters for the BSIM4.5.0 model dated July 29, 2005.
All parameters are binnable except those marked with an asterisk. Default entries
specify NMOS, PMOS default values where they differ.

LEVEL* 14 Model level. Must be 14.
A0 1 Non-uniform depletion width effect coefficient
A1 0 First non-saturation effect parameter
A2 1 Second non-saturation effect parameter
ACDE 1 Exponential coefficient for finite charge thickness
ACNQSMOD* 0 AC NQS model selector
AF* 1 Flicker noise exponent
AGIDL 0 Pre-exponential constant for GIDL
AGS 0 Gate bias coefficient of Abulk.
AIGBACC 13.6m Igb accumulation parameter
AIGBINV 12.1m Igb inversion parameter
AIGC 13.6m,9.8m Parameter for Igcs,Igcd
AIGSD 13.6m,9.8m Parameter for Igs,Igd
ALPHA0 0 First parameter of impact ionization current
ALPHA1 0 Isub parameter for length scaling
AT 33K Temperature coefficient of vsat
B0 0 First Abulk narrow width parameter
B1 0 Second Abulk narrow width parameter
BETA0 0 Substrate current model parameter
BGIDL 2.3G Exponential constant for GIDL
BIGBACC 1.71m Parameter for Igb
BIGBINV 949u Parameter for Igb
BIGC 1.71m,759u Parameter for Igc
BIGSD 1.71m,759u Parameter for Igs,d
BINFLAG* 0 Binning method: 1=HSpice, 0=Berkeley
BINUNIT* 1 Bin unit selector. 0=meters 1=microns
BVD* BVS Drain diode breakdown voltage
BVS* 10 Source diode breakdown voltage
CAPMOD* 2 Capacitance model selector
CDSC 240u Drain/source and channel coupling capacitance
CDSCB 0 Body-bias dependence of cdsc
CDSCD 0 Drain-bias dependence of cdsc
CF Computed Fringe capacitance parameter
CGBO* 0 Gate-bulk overlap capacitance per length


CGDL 0 New C-V model parameter
CGDO* 0 Gate-drain overlap capacitance per width
CGIDL 500m Parameter for body-bias dependence of GIDL
CGSL 0 New C-V model parameter
CGSO* 0 Gate-source overlap capacitance per width
CIGBACC 75m Parameter for Igb in accumulation
CIGBINV 6m Parameter for Igb in inversion
CIGC 75m,30m Parameter for Igc
CIGSD 75m,30m Parameter for Igs,d
CIT 0 Interface state capacitance
CJD* CJS Drain bottom junction capacitance per unit area
CJS* 500u Source bottom junction capacitance per unit area
CJSWD* None Drain sidewall junction capacitance per unit
periphery
CJSWGD* None Drain (gate side) sidewall junction capacitance
per unit width
CJSWGS* None Source (gate side) sidewall junction capacitance
per unit width
CJSWS* 500p Source sidewall junction capacitance per unit
periphery
CKAPPAD CKAPPAS D/G overlap C-V parameter
CKAPPAS 600m S/G overlap C-V parameter
CLC 100n Vdsat parameter for C-V model
CLE 600m Vdsat parameter for C-V model
DELTA 10m Effective Vds parameter
DIOMOD* 1 Diode IV model selector
DLC* Computed Delta L for C-V model
DLCIG* LINT Delta L for Ig model
DMCG* 0 Distance of mid-contact to gate edge
DMCGT* 0 Distance of mid-contact to gate edge in test
structures
DMCI* DMCG Distance of mid-contact to isolation
DMDG* 0 Distance of mid-contact to gate edge
DROUT 560m DIBL coefficient of output resistance
DSUB DROUT DIBL coefficient in the subthreshold region
DTOX* 0 Defined as (toxe - toxp)
DVT0 2.2 Short channel effect coeff. 0
DVT0W 0 Narrow Width coeff. 0
643

DVT1 530m Short channel effect coeff. 1
DVT1W 5.3MEG Narrow width effect coeff. 1
DVT2 -32m Short channel effect coeff. 2
DVT2W -32m Narrow width effect coeff. 2
DVTP0 0 First parameter for Vth shift due to pocket
DVTP1 0 Second parameter for Vth shift due to pocket
DWB 0 Width reduction parameter
DWC* WINT Delta W for C-V model
DWG 0 Width reduction parameter
DWJ* DWC Delta W for S/D junctions
EF* 1 Flicker noise frequency exponent
EGIDL 800m Fitting parameter for band bending
EIGBINV 1.1 Parameter for the Si bandgap for Igbinv
EM* 41MEG Flicker noise parameter
EPSROX* 3.9 Dielectric constant of the gate oxide relative to
vacuum
ETA0 80m First subthreshold region DIBL coefficient
ETAB -70m Second subthreshold region DIBL coefficient
EU 1.67,1 Mobility exponent
FNOIMOD* 1 Flicker noise model selector
FPROUT* 0 Rout degradation coefficient for pocket devices
GAMMA1 0 First Vth body coefficient
GAMMA2 0 Second Vth body coefficient
GBMIN* 1p Minimum body conductance
GDSNOI* 1 Channel shot noise coefficient
GEOMOD* 0 Geometry dependent parasitics model selector
IGBMOD* 0 Gate-to-body Ig model selector
IGCMOD* 0 Gate-to-channel Ig model selector
IJTHDFWD* IJTHSFWD Forward drain diode forward limiting current
IJTHDREV* IJTHSREV Reverse drain diode forward limiting current
IJTHSFWD* 100m Forward source diode forward limiting current
IJTHSREV* 100m Reverse source diode forward limiting current
JSD* JSS Bottom drain junction reverse saturation current
density
JSS* 100u Bottom source junction reverse saturation
current density
JSWD* JSWS Isolation edge sidewall drain junction reverse
saturation current density


JSWGD* JSWGS Gate edge drain junction reverse saturation
current density
JSWGS* 0 Gate edge source junction reverse saturation
current density
JSWS* 0 Isolation edge sidewall source junction reverse
JTSD* JTSS Drain bottom trap-assisted saturation current
density
JTSS* 0 Source bottom trap-assisted saturation current
density
JTSSWD* JTSSWS Drain STI sidewall trap-assisted saturation
current density
JTSSWGD* JTSSWGS Drain gate-edge sidewall trap-assisted saturation
current density
JTSSWGS* 0 Source gate-edge sidewall trap-assisted
JTSSWS* 0 Source STI sidewall trap-assisted saturation
current density
K1 0 Bulk effect coefficient 1
K2 0 Bulk effect coefficient 2
K2WE 0 K2 shift factor for well proximity effect
K3 80 Narrow width effect coefficient
K3B 0 Body effect coefficient of k3
KETA -47m Body-bias coefficient of non-uniform depletion
width effect
KF* 0 Flicker noise coefficient
KT1 -120m Temperature coefficient of Vth
KT1L 0 Length dependence of KT1
KT2 22m Body bias coefficient of KT1
KU0 0 Mobility degradation/enhancement coefficient
for LOD
KU0WE 0 Mobility degradation factor for well proximity
effect
KVSAT* 0 Saturation velocity degradation/enhancement
parameter for LOD
KVTH0 0 Threshold degradation/enhancement parameter
for LOD
645

KVTH0WE 0 Threshold shift factor for well proximity effect
LAMBDA 0 Velocity overshoot parameter
LC* 5n Back scattering parameter
LINT* 0 Channel length offset parameter
LINTNOI* 0 Lint offset for noise calculation
LL* 0 Coeff. of length dependence of length offset
LLC* LL Length reduction parameter for CV
LLN* 1 Power of length dependence of length offset
LLODKU0* 0 Length parameter for u0 LOD effect
LLODVTH* 0 Length parameter for VTH LOD effect
LMAX* 1 Maximum length for the model
LMIN* 0 Minimum length for the model
LODETA0* 1 ETA0 shift modification factor for stress effect
LODK2* 1 K2 shift modification factor for stress effect
LP 10n Channel length exponential factor of mobility
LPE0 174n Equivalent length of pocket region at zero bias
LPEB 0 Equivalent length of pocket region accounting
for body bias
LREF* None Length reference for HSpice binning
LW* 0 Coeff. of width dependence of length offset
LWC* LW Coefficient of width dependence for CV channel
length offset
LWL* 0 Coeff. of L*W product term for length offset
LWLC* LWL Length reduction parameter for CV
LWN* 1 Power of length dependence for length offset
MINV 0 Vgsteff fitting parameter for moderate inversion
MJD* MJS Drain bottom junction capacitance grading coeff.
MJS* 500m Source bottom junction capacitance grading
coefficient
MJSWD* MJSWS Drain sidewall junction capacitance grading
coefficient
MJSWGD* MJSWGS Drain (gate side) sidewall junction capacitance
grading coefficient
MJSWGS* MJSWS Source (gate side) sidewall junction capacitance
grading coefficient
MJSWS* 330m Source sidewall junction capacitance grading
coefficient
MOBMOD* 0 Mobility model selector


MOIN 15 Coefficient for gate-bias dependent surface
potential
NDEP 1.7E17 Depletion edge channel doping concentration
NFACTOR 1 Subthreshold swing coefficient
NGATE 0 Poly-gate doping concentration
NGCON* 1 Number of gate contacts
NIGBACC 1 Parameter for Igbacc slope
NIGBINV 3 Parameter for Igbinv slope
NIGC 1 Parameter for Igc slope
NJD* NJS Drain junction emission coefficient
NJS* 1 Source junction emission coefficient
NJTS* 20 Bottom junction non-ideality factor
NJTSSW* 20 STI sidewall junction non-ideality factor
NJTSSWG* 20 Gate-edge sidewall junction non-ideality factor
NLEV* None PSpice noise equation selector
NOFF 1 C-V turn-on/off parameter
NOIA* 6.25E41,6.188E40 First flicker noise parameter
NOIB* 3.125E26,1.5E25 Second flicker noise parameter
NOIC* 8.75G Flicker noise parameter
NSD 1E20 S/D doping concentration
NSUB 60000T Substrate doping concentration
NTNOI* 1 Thermal noise parameter
NTOX 1 Exponent for Tox ratio
PARAMCHK* 1 Model parameter checking selector
PBD* PBS Drain junction built-in potential
PBS* 1 Source junction built-in potential
PBSWD* PBSWS Drain sidewall junction capacitance built in
potential
PBSWGD* PBSWGS Drain (gate side) sidewall junction capacitance
built in potential
PBSWGS* PBSWS Source (gate side) sidewall junction capacitance
built in potential
PBSWS* 1 Source sidewall junction capacitance built in
potential
PCLM 1.3 Channel length modulation coefficient
PDIBLC1 390m First DIBL coefficient
PDIBLC2 8.6m Second DIBL coefficient
647

PDIBLCB 0 Body-effect on drain-induced barrier lowering
PDITS 0 Coefficient for drain-induced Vth shifts
PDITSD 0 Vds dependence of drain-induced Vth shifts
PDITSL* 0 Length dependence of drain-induced Vth shifts
PERMOD* 1 Pd and Ps model selector
PHIN 0 Adjusting parameter for surface potential due to
non-uniform vertical doping
PIGCD 1 Parameter for Igc partition
POXEDGE 1 Factor for the gate edge Tox
PRT 0 Temperature coefficient of parasitic resistance
PRWB 0 Body-effect on parasitic resistance
PRWG 1 Gate-bias effect on parasitic resistance
PSCBE1 424MEG First substrate current body-effect coefficient
PSCBE2 10u Second substrate current body-effect coefficient
PVAG 0 Gate dependence of output resistance parameter
RBDB* 50 Resistance between bNode and dbNode
RBDBX0* 100 Body resistance RBDBX scaling
RBDBY0* 100 Body resistance RBDBY scaling
RBODYMOD* 0 Distributed body R model selector
RBPB* 50 Resistance between bNodePrime and bNode
RBPBX0* 100 Body resistance RBPBX scaling
RBPBXL* 0 Body resistance RBPBX L scaling
RBPBXNF* 0 Body resistance RBPBX NF scaling
RBPBXW* 0 Body resistance RBPBX W scaling
RBPBY0* 100 Body resistance RBPBY scaling
RBPBYL* 0 Body resistance RBPBY L scaling
RBPBYNF* 0 Body resistance RBPBY NF scaling
RBPBYW* 0 Body resistance RBPBY W scaling
RBPD* 50 Resistance between bNodePrime and dbNode
RBPD0* 50 Body resistance RBPD scaling
RBPDL* 0 Body resistance RBPD L scaling
RBPDNF* 0 Body resistance RBPD NF scaling
RBPDW* 0 Body resistance RBPD W scaling
RBPS* 50 Resistance between bNodePrime and sbNode
RBPS0* 50 Body resistance RBPS scaling
RBPSL* 0 Body resistance RBPS L scaling
RBPSNF* 0 Body resistance RBPS NF scaling
RBPSW* 0 Body resistance RBPS W scaling


RBSB* 50 Resistance between bNode and sbNode
RBSBX0* 100 Body resistance RBSBX scaling
RBSBY0* 100 Body resistance RBSBY scaling
RBSDBXL* 0 Body resistance RBSDBX L scaling
RBSDBXNF* 0 Body resistance RBSDBX NF scaling
RBSDBXW* 0 Body resistance RBSDBX W scaling
RBSDBYL* 0 Body resistance RBSDBY L scaling
RBSDBYNF* 0 Body resistance RBSDBY NF scaling
RBSDBYW* 0 Body resistance RBSDBY W scaling
RDSMOD* 0 Bias-dependent S/D resistance model selector
RDSW 200 Source-drain resistance per width
RDSWMIN* 0 Source-drain resistance per width at high Vg
RDW 100 Drain resistance per width
RDWMIN* 0 Drain resistance per width at high Vg
RGATEMOD* 0 Gate R model selector
RNOIA* 577m First thermal noise coefficient
RNOIB* 516.4m Second thermal noise coefficient
RSH* 0 Source-drain sheet resistance
RSHG* 100m Gate sheet resistance
RSW 100 Source resistance per width
RSWMIN* 0 Source resistance per width at high Vg
SAREF* 1u Reference distance between OD edge to poly of
one side
SBREF* 1u Reference distance between OD edge to poly of
the other side
SCREF* 1u Reference distance to calculate SCA, SCB and
SCC
STETA0* 0 ETA0 shift factor related to stress effect on
VTH
STK2* 0 K2 shift factor related to stress effect on VTH
T_ABS* None Absolute temperature
T_MEASURED* None Parameter measured temperature
T_REL_GLOBAL* None Relative to current temperature
T_REL_LOCAL* None Relative to AKO model temperature
TCJ* 0 Temperature coefficient of CJ
TCJSW* 0 Temperature coefficient of CJSW
TCJSWG* 0 Temperature coefficient of CJSWG
TEMPMOD* 0 Temperature model selector
649

TKU0* 0 Temperature coefficient of KU0
TNJTS* 0 Temperature coefficient for NJTS
TNJTSSW* 0 Temperature coefficient for NJTSSW
TNJTSSWG* 0 Temperature coefficient for NJTSSWG
TNOIA* 1.5 Coefficient of channel-length dependence of
total channel thermal noise
TNOIB* 3.5 Channel-length dependence parameter for
channel thermal noise partitioning
TNOIMOD* 0 Thermal noise model selector
TNOM* None Parameter measurement temperature
(if specified this overrides T_MEASURED)
TOXE* 3n Electrical gate oxide thickness in meters
TOXM* TOXE Gate oxide thickness at which parameters are
extracted
TOXP* TOXE Physical gate oxide thickness in meters
TOXREF* 3n Target tox value
TPB* 0 Temperature coefficient of PB
TPBSW* 0 Temperature coefficient of PBSW
TPBSWG* 0 Temperature coefficient of PBSWG
TRNQSMOD* 0 Transient NQS model selector
TVFBSDOFF 0 Temperature parameter for VFBSDOFF
TVOFF 0 Temperature parameter for VOFF
U0 67m,25m Low-field mobility at Tnom
UA Multiple Linear gate dependence of mobility
1f for MOBMOD=2, else 1n
UA1 1n Temperature coefficient of UA
UB .0001f Quadratic gate dependence of mobility
UB1 -.001f Temperature coefficient of UB
UC Multiple Body-bias dependence of mobility
-.0465 for MOBMOD=1, else -.0465n
UC1 Multiple Temperature coefficient of UC
-.056 for MOBMOD=1, else -.056n
UD 100T Coulomb scattering factor of mobility
UD1 0 Temperature coefficient of UD
UP 0 Channel length linear factor of mobility
UTE -1.5 Temperature coefficient of mobility
VBM -3 Maximum body voltage
VBX 0 Vth transition body voltage


VERSION* "4.5.0" Model version parameter
VFB 0 Flat band voltage
VFBCV -1 Flat band voltage parameter for capmod=0 only
VFBSDOFF 0 S/D flatband voltage offset
VOFF -80m Threshold voltage offset
VOFFCV 0 C-V lateral-shift parameter
VOFFL* 0 Length dependence parameter for Vth offset
VSAT 80K Saturation velocity at tnom
VTH0 700m,-700m Threshold voltage
VTH0M* 0 VT short-distance matching parameter
VTL 200K Thermal velocity
VTSD* VTSS Drain bottom trap-assisted voltage dependent
parameter
VTSS* 10 Source bottom trap-assisted voltage dependent
parameter
VTSSWD* VTSSWS Drain STI sidewall trap-assisted voltage
dependent parameter
VTSSWGD* VTSSWGS Drain gate-edge sidewall trap-assisted voltage
dependent parameter
VTSSWGS* 10 Source gate-edge sidewall trap-assisted voltage
dependent parameter
VTSSWS* 10 Source STI sidewall trap-assisted voltage
dependent parameter
W0 2.5u Narrow width effect parameter
WEB* 0 Coefficient for SCB
WEC* 0 Coefficient for SCC
WINT* 0 Channel width offset parameter
WL* 0 Coeff. of length dependence for width offset
WLC* WL Width reduction parameter for CV
WLN* 1 Width reduction parameter
WLOD* 0 Width parameter for stress effect
WLODKU0* 0 Width parameter for U0 LOD effect
WLODVTH* 0 Width parameter for VTH LOD effect
WMAX* 1.0 Maximum width for the model
WMIN* 0.0 Minimum width for the model
WPEMOD* 0 Flag for WPE model (WPEMOD=1 to activate
this model)
WR 1 Width dependence of RDS
651

WREF* None Width reference for HSpice binning
WW* 0 Width reduction parameter
WWC* WW Coefficient of width dependence CV channel
width offset
WWL* 0 Coefficient of length and width product term
dependence for width offset
WWLC* WWL Coefficient of length and width product
dependence for CV channel width
WWN* 1 Power of width dependence of width offset
XGL* 0 Variation in Ldrawn
XGW* 0 Distance from gate contact center to device
edge
XJ 150n Junction depth in meters
XJBVD* XJBVS Fitting parameter for drain diode breakdown
current
XJBVS* 1 Fitting parameter for source diode breakdown
current
XL* 0 L offset for channel length due to mask/etch
effect
XN 3 Back scattering parameter
XPART* 0 Channel charge partitioning
XRCRG1 12 First fitting parameter for ias-dependent Rg
XRCRG2 1 Second fitting parameter for bias-dependent Rg
XT 155n Doping depth
XTID* XTIS Drain junction current temperature exponent
XTIS* 3 Source junction current temperature exponent
XTSD* XTSS Power dependence of JTSD on temperature
XTSS* 20m Power dependence of JTSS on temperature
XTSSWD* XTSSWS Power dependence of JTSSWD on temperature
XTSSWGD* XTSSWGS Power dependence of JTSSWGD on
temperature
XTSSWGS* 20m Power dependence of JTSSWGS on
temperature
XTSSWS* 20m Power dependence of JTSSWS on temperature
XW* 0 W offset for channel width due to mask/etch
effect

Binning
Binning is the process of adjusting model parameters for different values of

drawn channel length (L) and width (W). It is available in all BSIM versions. All
model parameters are binnable except those marked with an asterisk.
Binning in BSIM1 and BSIM2 uses the Berkeley method and uses the length and
width terms only.
Binning for both BSIM3 (Level = 8 or Level = 49) and BSIM4 (Level = 14) uses
length, width, and product terms. There are two ways to bin in these models:
Original Berkeley method

This method is used if the parameter BINFLAG is <= 0.9 or, if either the WREF
or LREF parameters are absent.
A binnable parameter X, with a nominal value of X0 is calculated for a given L

and W as follows:
Leff = L - 2*DL
Weff = W - 2*DW
X = X0 + XL/Leff + XW/Weff + XP/Leff/Weff
XL = Parameter X's length dependence

XW = Parameter X's width dependence
XP = Parameter X's product (length*width) dependence
HSPICE method
This method uses the model parameters LMIN, LMAX, WMIN, WMAX, LREF,
and WREF to provide multiple cell binning. LMIN, LMAX, WMIN, WMAX
define the cell boundary. LREF and WREF are offset values used to interpolate a
value within the cell boundaries. The model parameters are assumed to apply for
the case Leff = Lref and Weff = Wref.
When this method is used Micro-Cap expects to find multiple model statements
(or parameter sets) of the form:
.MODEL NC.1 NMOS (WMIN=1U WMAX=5U LMIN=.1U LMAX=.3U...)

.MODEL NC.2 NMOS (WMIN=5U WMAX=20U LMIN=.1U LMAX=.3U...)
.MODEL NC.3 NMOS (WMIN=1U WMAX=5U LMIN=.3U LMAX=1U...)
.MODEL NC.4 NMOS (WMIN=5U WMAX=20U LMIN=.3U LMAX=1U...)
653
Here NC is the basic model name and NC.1, NC.2, ... are the model parameters
for the W values between WMIN to WMAX and L values between LMIN to
LMAX.
Micro-Cap selects the model statement or parameter set for which the following
statements are both true:
WMIN <= W AND W <= WMAX

LMIN <= L AND L <= LMAX
If it can't find a model statement (or parameter set) for which these conditions are
true, an error message is generated.
Once the model statement (or parameter set) is selected the binnable parameters
are calculated as follows:
Leff = L - 2*DL
Weff = W - 2*DW
X = X0 + Xl*(1/Leff - 1/LREF) + Xw*(1/Weff - 1/WREF)

+ Xp / (1/Leff -1/LREF) / (1/Weff - 1/WREF)
where Xl = Parameter X's length dependence

Xw = Parameter X's width dependence
Xp = Parameter X's product (length*width) dependence
Note that the starting values of L and W are those specified on the device line for
a SPICE netlist device or in a VALUE attribute for a schematic device. Micro-
Cap allows L and W to be specified in the model statement also, but binning
always uses the L and W specified at the device level.
The units for the geometry parameters can be selected to be in microns by

setting the model parameter BINUNIT = 1. For other choices of BINUNIT, the
dimensions are in meters.
Short-distance matching for BSIM3 and BSIM4
BSIM3 and BSIM4 use the parameter VTH0M to account for short-distance
VTH matching. The device threshold is adjusted as follows:
VTH = VTH + VTH0M / SQRT(Weff*Leff)

MOSFET (EKV)
SPICE format
The EKV device format is similar to other MOSFETs but has a few more device
parameters that are not found in the other models.
SPICE format
Syntax
+ [NP|M=<no_parallel>] [NS|N=<no_series>]
+ [SCALE=<scale>] [GEO=<geometry_model>]
+ [TEMP=<temperature>]

Examples of device instance
M1 5 7 9 0 IRF350 L=1.5E-6 W=0.25 NP=10 OFF IC=25.0,8.0
MS 1 2 3 4 NSC SCALE=.8 GEO=2 NRD=1 NRS=2 AS=1E-12 AD=1E-12
Schematic format
PART attribute
<name>
Example of PART attribute

M1
VALUE attribute
[L=<length>] [W=<width>] [AD=<drainarea>] [AS=<sourcearea>]
+ [NP|M=<no_parallel>] [NS|N=<no_series>]
+ [SCALE=<scale>] [GEO=<geometry_model>]
+ [TEMP=<temperature>]
655
Examples of VALUE attribute
M=20 NRD=10 NRS=25 NRG=5
L=.35u IC=.1, 2.00
L=.4u W=2u OFF IC=0.05,1.00
MODEL attribute
<model name>
Examples of MODEL attribute

IRF350
meters. They can be specified as device attributes, model parameters, or by using
the global default values, DEFW and DEFL. Device attributes supersede model
parameters, which supersede the global DEFW and DEFL values from the Global
Settings dialog box or a local .OPTIONS statement.
The initialization [IC=<vds>[,vgs[,vbs]]] assigns initial voltages to the drain-

source, gate-source, and body-source terms in transient analysis if no operating
point is done (or if the UIC flag is set). The [OFF] keyword forces the device off
during the first iteration of the DC operating point.

<sourcearea> and <drainarea> are the diffusion areas (sq. m). Source and drain
junction capacitances may be specified directly by the model parameters CBS
and CBD. If absent, they are calculated from area and periphery terms.
The diffusion resistances may be specified directly with the model parameters
RS, RD, RG, and RB. If unspecified, they are calculated from the product of
the sheet resistivity, RSH, and the number of squares terms, <drainsquares>,
<sourcesquares>, <gatesquares>, and <bulksquares>. If these terms are absent,
or zero, and the model parameters RS, RD, RG, and RB are absent or zero, then
the parasitic resistances are not included in the model. The procedure used to
calculate the resistive values is dependent upon the ACM model parameter and is
described in the section "EKV diode parameter calculation methods".
<drainsquares> and <sourcesquares> default to 1.0. The other parameter line

values default to zero.
<no_parallel> is the number of devices in parallel. <no_series> is the number of

devices in series.

<no_parallel> multiplies the drain and source areas, the device width, the drain
and source peripheries, and divides the lead resistances RS, RD, RG, and RB.
<no_series> divides the device length.
<scale> is a scaling parameter that multiplies or divides device parameters to ac-

count for device geometric scaling.
<geometry_model> is the geometry model selector. It is used only when the

ACM (Area Calculation Method) model parameter is set to 3.
<temperature> is an optional device operating temperature. If specified, it takes

precedence over the operating temperature computed from the model parameters,
T_MEASURED, T_ABS, T_REL_GLOBAL, and T_REL_LOCAL. For more
details on how device operating temperatures and Tnom temperatures are calcu-
lated from model parameters, see the .MODEL section of the Command State-
ments chapter.
657
Model parameters for EKV level 44
Setup Parameters
Name Default Range Description
LEVEL None 44 MOSFET model level. Must be
44 for EKV.
EKVINT None None Interpolation method selector.
EKVINT = 1 selects F(v) =
ln^2(1+exp(v/2))
EKVDYN 0 0-1 EKVDYN=1 sets all intrinsic
capacitances to zero.
UPDATE None None RD,RS selector for ACM=1
SATLIM exp(4) None Ratio defining the saturation
limit. For operating point
information only.
XQC 0.4 None Charge/capacitance model
selector. XQC=.4 selects
original charge/transcapacitance
model. XQC=1 selects simpler
capacitance only model.
Process Related Parameters

COX 0.7E-3 F/m None
2
Gate oxide capacitance/area
XJ 0.1E-6 m >=1.0E-9 Junction depth
DW 0 m None Channel width correction
(normally negative)
DL 0 m None Channel length correction
(normally negative)
Basic Intrinsic Model Parameters

L DEFL m DEFL m Channel length
W DEFW m DEFW m Channel width
VTO 0.5 V None Long-channel threshold voltage
GAMMA 1.0 V1/2 >=0 Body effect parameter
PHI 0.7 >=0.1 Bulk Fermi potential
KP 50.0E-6 A/V2 None Transconductance parameter
E0 or EO 1.0E12 V/m >=1E5 Mobility reduction coefficient
UCRIT 2.0E6 V/m >=1E5 Longitudinal critical field

The following parameters are to accommodate scaling behavior of the process

and basic intrinsic model parameters, as well as statistical circuit simulation.
Note that the parameters TOX, NSUB, VFB, UO, and VMAX are only used if
COX, GAMMA and/or PHI, VTO, KP and UCRIT are not specified, respectively.
Further, a simpler mobility reduction model due to vertical field is accessible.
The mobility reduction coefficient THETA is only used if E0 is not specified.
Optional Parameters
TOX 0 m None Oxide thickness
NSUB None cm-3 None Channel doping
VFB None V None Flat-band voltage
UO None cm2/(Vs) >=0 Low-field mobility
VMAX None m/s >=0 Saturation velocity
THETA 0 V-1 >=0 Mobility reduction coefficient
Channel Length Modulation and Charge Sharing Parameters

LAMBDA 0.5 >=0 Depletion length coefficient
for channel length modulation
WETA 0.25 None Narrow-channel effect
coefficient
LETA 0.1 None Short-channel effect coefficient
Reverse Short-channel Effect Parameters

Q0 or QO 0 A·s/m2 None Reverse short channel effect
peak charge density
LK 0.29E-6 m >=1.0E-8 Reverse short channel effect
characteristic length
Impact Ionization Related Parameters

IBA 0 1/m None 1'st impact ionization coefficient
IBB 3.0E8 V/m >=1.0E8 2'nd impact ionization coeff.
IBN 1.0 >0.1 Saturation voltage factor for
impact ionization
659
Temperature Parameters
TR1 0 1/K None Linear resistance
temperature coefficient
TR2 0 1/K None Quadratic resistance
temperature coefficient
TCV 1.0E-3 V/K None Vth temperature coefficient
BEX -1.5 None Mobility temperature exponent
UCEX 0.8 None Longitudinal critical field
temperature exponent
IBBT 9.0E-4 1/K None Temperature coefficient for
IBB
TNOM 27 °C None Parameter measurement
temperature. TNOM takes
priority over T_MEASURED.
XTI 0.0 None Junction current temperature
exponent coefficient
T_MEASURED None °C >0.0 Measured temperature
T_ABS None °C >0.0 Absolute temperature
T_REL_GLOBAL None °C None Relative to current temperature
T_REL_LOCAL None °C None Relative to AKO temperature
Extrinsic Resistance Parameters

RDS ∞ None Drain-source shunt resistance
RDC 0 Ω None Drain contact resistance
RSC 0 Ω None Source contact resistance
RD 0 Ω None Drain ohmic resistance
RS 0 Ω None Source ohmic resistance
RG 0 Ω None Gate ohmic resistance
RB 0 Ω None Bulk ohmic resistance
RBSH 0 Ω / sq None Bulk sheet resistivity
RGSH 0 Ω / sq None Gate sheet resistivity
RSH 0 Ω / sq None Source / drain sheet resistivity
Extrinsic Capacitance Parameters

CGDO 0.0 F/m None Gate-drain overlap capacitance
CGSO 0.0 F/m None Gate-source overlap cap.

Extrinsic Capacitance Parameters

CGBO 0.0 F/m None Gate-bulk overlap capacitance
CBD 0.0 F None Bulk p-n zero-bias
bulk-drain capacitance
CBS 0.0 F None Bulk p-n zero-bias
bulk-source capacitance
CJ 0.0 F/m2 None Bulk p-n zero-bias bottom
capacitance
CJSW 0.0 F/m None Bulk p-n zero-bias sidewall
capacitance
CJGATE 0.0 F None Zero-bias gate-edge sidewall
bulk junction capacitance
Only used with ACM=3
Extrinsic Junction Parameters

MJ 0.50 >0.0 Bulk p-n zero-bias bottom grad.
MJSW MJ >0.0 Bulk p-n zero-bias s/w coeff.

PN Junction Parameters
ACM 0 0-3 Area calculation parameter
0=SPICE style
1=ASPEC style
2=HSPICE style
3=HSPICE and stacked devices
TT 0.00 s >=0 Bulk transit time
IS 1e-14 A >0 Bulk saturation current
N 1.00 None Bulk emission coefficient
JS 0.0 A/m2 >=0 Bulk bottom current density
JSW JSW A/m 2
>=0 Sidewall current density
PB 0.80 V None Bulk bottom potential
PBSW PB V None Sidewall potential
FC 0.50 None Forward-bias depletion
coefficient
661
Matching Parameters
AVTO 0 V·m None Area related threshold voltage
mismatch parameter
AKP 0 m None Area related gain mismatch
parameter
AGAMMA 0.0 V1/2·m None Area related body effect
mismatch parameter
Noise Parameters
KF 0 None Flicker noise coefficient
AF 1 None Flicker noise exponent
GDSNOI 1.0 None Channel shot noise coefficient
NLEV 2.0 None Noise equation selector
Geometry Parameters
SCALM 1.0 >0 Model parameter scale factor
HDIF 0 m None Length of heavily doped
diffusion, from contact to lightly
doped region (ACM=2, 3 only)
HDIF scaled=HDIF×SCALM
LD None m None Lateral diffusion into channel
from source and drain diffusion.
If LD and XJ are unspecified,
LD default=0.0. When LD is
unspecified, but XJ is specified,
LD is calculated from XJ. LD
default=0.75 × XJ.
LDIF 0 m None Length of lightly doped
diffusion adjacent to gate
(ACM=1, 2)
LDIFscaled = LDIF × SCALM
WMLT 1 None Width diffusion layer shrink
reduction factor
XJ 0 m None Metallurgical junction depth
XJscaled = XJ × SCALM

EKV diode parameter calculation methods
The EKV model employs an ACM (Area Calculation Method) parameter to select
different methods of calculating the drain and source diode parameters. The
method is similar to that employed by HSPICE.
ACM=0 specifies the original SPICE method.
ACM=1 specifies the original ASPEC method.
ACM=2 specifies an improved HSPICE method, which is based on a model

similar to the ASPEC method.
ACM=3 specifies a further HSPICE refinement that deals with capacitances

of shared sources and drains and gate edge source/drain-to-bulk periphery
capacitance and facilitates the modeling of stacked devices.
If the ACM model parameter is not set, the method defaults to the ACM=0 SPICE
model.
ACM=0 and ACM=1 models do not use HDIF. ACM=0 does not use LDIF.
The geometric element parameters AD, AS, PD, and PS are not used for the
ACM=1 model.
663
ACM=0 SPICE style diodes
Effective Areas and Peripheries

ADeff = M · AD · WMLT2 · SCALE2
ASeff = M · AS · WMLT2 · SCALE2
PDeff = M · PD · WMLT · SCALE
PSeff = M · PS · WMLT · SCALE
Source Diode Saturation Current

val = JSscaled · ASeff + JSWscaled · PSeff
If val > 0 then,

isbs = val
Otherwise,
isbs = M · IS
Drain Diode Saturation Current

val = JSscaled · ADeff + JSWscaled · PDeff
If val > 0 then,

isbd = val
Else
isbd = M · IS
Source Resistance
val = NRS·RSH
If val > 0 then,

RSeff = (val + RSC) / M
Else
RSeff = (RS + RSC) / M
Drain Resistance
val = NRD·RSH
If val > 0 then,

RDeff = (val + RDC) / M
Else
RDeff = (RD + RDC) / M

ACM=1 ASPEC style diodes
When ACM=1 the ASPEC method is used. The parameters AD, PD, AS, and PS
are not used. The units JS and CJ differ from the (ACM=0) SPICE case.

Weff = Weff · M · (Wscaled · WMLT + XWscaled)
ADeff = ASeff = Weff · WMLT2
PDeff = PSeff = Weff

If val > 0 then,

isbs = val
Otherwise,
isbs = M · IS

If val > 0 then,

isbd = val
Else
isbd = M · IS
Source Resistance
If UPDATE=0
RSeff = RS · (LDscaled + LDIFscaled) / Weff + (NRS · RSH + RSC) / M
Else if UPDATE >= 1 and LDIF=0 then
RSeff = (RS + NRS · RSH + RSC) / M
Else
RSeff = 0
Drain Resistance
If UPDATE=0
RSeff = RD · (LDscaled + LDIFscaled) / Weff + (NRD · RSH + RDC) / M
Else if UPDATE >= 1 and LDIF=0 then
RDeff = (RD + NRD · RSH + RDC) / M
Else
RDeff = 0
665
ACM=2 HSPICETM style diodes
This method uses a fold-back calculation similar to the ASPEC method, retaining
full model-parameter compatibility with the SPICE procedure. It also supports
both lightly and heavily doped diffusions (by setting the LD, LDIF, and HDIF
parameters). The units of JS, JSW, CJ, and CJSW used in SPICE are preserved,
permitting full compatibility. ACM=2 automatically generates more reasonable
diode parameter values than those for ACM=1. The ACM=2 geometry can be
generated one of two ways:
Device parameters: AD, AS, PD, and PS can be used for parasitic
generation when specified for the device. Default option values for
these parameters are not applicable.
If the diode is to be suppressed, set IS=0, AD=0, and AS=0.
The source diode can be suppressed if AS=0 is set in the element and IS=0 is set
in the model, a useful setting for shared contacts.

If AD is not specified then
ADeff = 2 · HDIFeff · Weff
Else
If AS is not specified then

ASeff = 2 · HDIFeff · Weff
Else
If PD is not specified then,

PDeff = 4 · HDIFeff + 2 · Weff
Else
If PS is not specified then,

PSeff = 4 · HDIFeff + 2 · Weff
Else

ACM=2 HSPICETM style diodes
Where the terms are defined as follows:
Weff = M · (Wscaled · WMLT + XWscaled)

HDIFscaled = HDIF · WMLT · SCALM
HDIFeff = HDIFscaled

If val > 0 then,

isbs = val
Otherwise,
isbs = M · IS

If val > 0 then,

isbd = val
Else
isbd = M · IS
Source Resistance
If NRS is specified
Else
RSeff = RSC / M + (HDIFeff · RSH + (LDscaled · LDIFscaled) · RS) / Weff
Drain Resistance
If NRD is specified
RDeff = RD · (LDscaled + LDIFscaled) / Weff + (NRD · RSH + RDC) / M
Else
RDeff = RDC / M + (HDIFeff · RSH + (LDscaled · LDIFscaled)· RD) / Weff
667
ACM=3 Improved HSPICETM style diodes
The ACM=3 method is designed to model stacked devices properly. The

CJGATE model parameter separately models drain and source periphery
capacitances along the gate edge, so PD and PS calculations do not include the
gate periphery length. CJGATE defaults to CJSW, which, in turn, defaults to 0.
The AD, AS, PD, PS calculations depend on the layout of the device, which is
determined by the value of device parameter GEO (specified in the Attribute
dialog box or in a SPICE file, on an element line). It can have the following
values:
GEO=0 (Default) Neither drain nor source is shared with another device.
GEO=1 Drain is shared with another device.
GEO=2 Source is shared with another device.
GEO=3 Both drain and source are shared with another device.

If AD is not specified then
For GEO = 0 or 2: ADeff = 2 · HDIFeff · Weff
For GEO = 1 or 3: ADeff = HDIFeff · Weff
Else
If AS is not specified then

For GEO = 0 or 1: ASeff = 2 · HDIFeff · Weff
For GEO = 2 or 3: ASeff = HDIFeff · Weff
Else
If PD is not specified, then,

For GEO = 0 or 2: PDeff = 4 · HDIFeff + Weff
For GEO = 1 or 3: PDeff = 2 · HDIFeff
Else
If PS is not specified, then,

For GEO = 0 or 1: PSeff = 4 · HDIFeff + Weff
For GEO = 2 or 3: PSeff = 2 · HDIFeff
Else

Weff and HDIFeff are calculated as follows:
Weff = M · (Wscaled · WMLT + XWscaled)

HDIFscaled = HDIF· SCALM
HDIFeff = HDIFscaled · WMLT
Effective Saturation Current Calculations

(Same as ACM=2)

If val > 0 then,

isbs = val
Otherwise,
isbs = M · IS

If val > 0 then,

isbd = val
Else
isbd = M · IS
Effective Drain and Source Resistances

(Same as ACM=2)
Source Resistance
If NRS is specified
Else
RSeff = RSC / M + (HDIFeff · RSH + (LDscaled · LDIFscaled) · RS) / Weff
Drain Resistance
If NRD is specified
RDeff = RD · (LDscaled + LDIFscaled) / Weff + (NRD · RSH + RDC) / M
Else
RDeff = RDC / M + (HDIFeff · RSH + (LDscaled · LDIFscaled)· RD) / Weff
669
Noise models
The MOSFET noise model described below is used for Level 1, 2, 3, 4, 5 and
EKV models.
It is also used for BSIM3 (Level 8) and BSIM4 (Level 14) when NLEV is
defined. If NLEV is undefined, the native Berkeley BSIM3 and BSIM4 noise
models are used.
Pin resistance thermal noise terms
I2rg = 4•k•T / RG
I2rd = 4•k•T / RD
I2rs = 4•k•T / RS
I2rb = 4•k•T / RB
Channel and shot flicker noise terms
Ichannel2 = Ishot2 + Iflicker2
Intrinsic flicker noise:

If NLEV = 0 Iflicker2 = KF•IdrainAF / ( COX • Leff2 • f )
If NLEV = 1 Iflicker2 = KF•IdrainAF / ( COX • Weff • Leff • f )
If NLEV = 2 or 3 Iflicker2 = KF•GmAF / ( COX • Weff • Leff • fAF )
Intrinsic shot noise:

If NLEV < 3 then
Ishot2 = (8/3)•k•T•gm
If NLEV = 3 then
Ishot2 = (8/3)•k•T•gm•GDSNOI•beta•(Vgs-Vth)•(1+a+a2)/(1+a)
where a = 1 - Vds/Vdsat if Vds <= Vdsat (linear region) and a = 0 elsewhere.

Short-distance matching for EKV
The EKV model provides the short-distance matching parameters AVTO for
threshold voltage, AGAMMA for gamma, and AKP for KP. The parameters are
calculated as follows:
VTH = VTH + AVTO / SQRT(Weff*Leff)
GAMMA = GAMMA + AGAMMA / SQRT(Weff*Leff)
KP = KP + AKP / SQRT(Weff*Leff)
671
MOSFET Philips Model 12
The Philips Model 12 is a compact MOSFET model, intended for digital, analog,
and RF circuit simulation in modern CMOS technologies. Its surface potential
formulation assures continuity of derivatives across operating boundaries. The
instance format is similar to other MOSFET devices but with fewer parameters.
Note that the model includes no source or drain diodes. When needed, these are
added using the JUNCAP diode model, usually as a part of a subcircuit which
includes the MOSFET.
SPICE format
Syntax for Geometrical and Binning models
+ [L=<length>] [W=<width>] [M=<no_parallel>]
Syntax for Electrical model

+ [M=<no_parallel>]

Example of device instance for Geometrical and Binning models
MS 1 2 3 4 NSC L=.25u W=1.2u M=20
Schematic format
PART attribute
<name>
Examples of PART attribute

M1
Syntax for Geometrical and Binning model VALUE attribute

[L=<length>] [W=<width>] [M=<no_parallel>]
Examples of VALUE attribute for Geometrical and Binning models

M=20 L=.4u W=2u
Syntax for Electrical model VALUE attribute

[M=<no_parallel>]
MODEL attribute
<model name>

IRF350
meters. They can be specified as device attributes or as model parameters. Device
attributes, if specified, supersede model parameters. If neither model nor device
attributes are specified, they assume the default values of W=10u, L=2u.
<no_parallel> is the number of devices in parallel. If specified, it overrides the

MULT model parameter, which multiplies the current-dependent and oxide ca-
pacitance model parameters.
The equivalent circuit of the Philips Model 12 is as follows:
Figure 26-21 Electrical model of Philips Model 12
673
Model Parameters
There are actually fourteen versions of Model 12 available for use within Micro-
Cap. The first release (1200) had only two types, electrical and geometrical. Each
of the last two versions that Philips has released (1201 and 1202) came in two va-
rieties (thermal and standard) of three variations (electrical, geometric, and bin-
ning). Although for legacy purposes each is available for use in Micro-Cap, only
the latest model, 1202.3, dated March 2006, will be documented in this manual.
First Release 1200

Micro-Cap Philips Description
12001 1200 Electrical
12002 1200 Geometrical
Second Release 1201

12013 1201 Binning
12014 1201 Electrical Thermal
12015 1201 Geometrical Thermal
12016 1201 Binning Thermal
Third Release 1202

12023 1202 Binning
12024 1202 Electrical Thermal
12025 1202 Geometrical Thermal
12026 1202 Binning Thermal
In the parameter tables that follow, the level parameter is shown first and the
remainder of the model parameters are listed in alphabetical order. Where the
NMOS and PMOS default parameters differ, they are shown as NMOS,PMOS.
Parameters marked with an asterisk are new to the Philips 1202 models and are
not available in earlier versions. Default entries specify NMOS, PMOS default
values where they differ.

Model parameters for level 12021/12024 (Philips 1202.3 Electrical)

LEVEL 12021 or 12024 Model level. Use 12021 for standard model.
Use 12024 for thermal model.
A1 6.0221,6.8583 Factor of the weak-avalanche current
A2 38.017,57.324 Exponent of the weak-avalanche current
A3 0.6407,0.4254 Factor of the drain-source voltage above
which weak-avalanche occurs
AGIDL 0 Gain factor for gate-induced leakage current
ALP 25m Factor of channel length modulation
BACC 48 Probability factor for intrinsic gate
tunnelling current in accumulation
BET 1.9215m,381.4u Gain factor
BGIDL 41 Probability factor for gate-induced drain
leakage current at reference temperature
BINV 48,87.5 Probability factor for intrinsic gate
tunnelling current in inversion
CGDO 6.392f,6.358f Oxide capacitance for the gate-drain overlap
CGIDL 0 Factor for the lateral field dependence of the
gate-induced leakage current
CGSO 6.392f,6.358f Oxide capacitance for gate-source overlap
*CS 0 Coefficient of Coulomb scattering
COX 29.8f,27.17f Oxide capacitance for the intrinsic channel
DTA 0 Temperature offset of the device
ETABET 1.3,0.5 Exponent of the temperature dependence of
the gain factor
*ETACS 0 Temperature dependence exponent of CS
ETAMOB 1.4,3 Effective field parameter for dependence on
depletion charge
ETAPH 1.35,3.75 THEPH temperature dependence exponent
ETAR 950m,400m THER temperature dependence exponent
ETASAT 1.04,860m THESAT temperature dependence exponent
ETASR 650m,500m THESR temperature dependence exponent
GATENOISE 0 Induced gate thermal noise flag (1 = include)
IGACC 0 Gain factor for intrinsic gate tunnelling
current in accumulation
IGINV 0 Gain factor for intrinsic gate tunnelling
current in inversion
IGOV 0 Gain factor for Source/Drain overlap
tunnelling current
KO 500m Body-effect factor
675
KOV 2.5 Body-effect factor for the Source/Drain

overlap extensions
KPINV 0 Inverse of body-effect factor of poly-silicon
gate
MEXP 5 Smoothing factor
MO 0 Parameter for short-channel subthreshold
slope
NFA 8.32E22,1.9E22 First coefficient of the flicker noise
NFB 25.14E6,5.04E6 Second coefficient of the flicker noise
NFC 0,362.7p Third coefficient of the flicker noise
NT .00001624f Thermal noise coefficient
NU 2 Exponent of field dependence of mobility
model
NUEXP 5.25,3.23 Exponent of the temperature dependence of
parameter NU
PHIB 950m Surface potential at the onset of strong
inversion
SDIBL 853u,35.51u Drain-induced barrier lowering parameter
SSF 12m,10m Static-feedback parameter
STA1 0 A1 temperature coefficient
STBGIDL -363.8u BGIDL temperature coefficient
STETAMOB 0 ETAMOB temperature coefficient
STPHIB -850u PHIB temperature coefficient
STVFB 500u VFB temperature coefficient
THEPH 12.9m,1m Mobility degradation parameter due to
phonon scattering
THER 81.2m,79m Coefficient of series resistance
THER1 0 Numerator of gate voltage dependent part of
series resistance
THER2 1 Denominator of gate voltage dependent part
of series resistance
THESAT 251.3m,172.8m Velocity saturation parameter due to optical/
acoustic phonon scattering
THESR 356.2m,730m Mobility degradation parameter due to
surface roughness scattering
THETH 10u,0 Coefficient of self-heating


TOX 3.2n Thickness of gate oxide layer
TR 21 Reference temperature
VFB -1.05 Flat-band voltage at reference temperature
VFBOV 0 Flat-band voltage for the Source/Drain
overlap extensions
VP 50m Characteristic voltage of channel-length
modulation

677
Model parameters for level 12022/12025 (Philips 1202.3 Geometrical)

A1R 6 Weak-avalanche current factor
A2R 38 Weak-avalanche current exponent
A3R 1 Factor of the drain-source voltage above
which weak-avalanche occurs
AGIDLR 0 Gain factor for gate-induced leakage current
ALPEXP 1 Exponent of the length dependence of ALP
ALPR 10m Factor of the channel length modulation
BACC 48 Probability factor for intrinsic gate
tunnelling current in accumulation
BETSQ 370.9u,125u Gain factor for an infinite square transistor
BGIDL 41 Probability factor for gate-induced drain
leakage current at reference temperature
BINV 48,87.5 Probability factor for intrinsic gate
tunnelling current in inversion
CGIDL 0 Factor for the lateral field dependence of the
gate-induced leakage current
CSEXP 1 Exponent of the length dependence of CS
COL .32f Gate overlap capacitance for a channel
width of 1 um
*CSR 0 Factor of the Coulomb scattering
ETABETR 1.3,500m Gain factor temperature dependence
exponent
*ETACS 0 Temperature dependence exponent of CS
ETAMOBR 1.4,3 Effective field parameter for dependence on
depletion/inversion charge
ETAPH 1.35,3.75 THEPH temperature dependence exponent
ETAR 950m,400m THER temperature dependence exponent
ETASAT 1.04,860m THESAT temperature dependence exponent
ETASR 650m,500m THESR temperature dependence exponent
FBET1 0 Relative mobility decrease due to first lateral
profile
FBET2 0 Relative mobility decrease due to second
lateral profile
GATENOISE 0 Induced gate thermal noise flag (1=include)
IGACCR 0 Gain factor for intrinsic gate tunnelling
current in accumulation

IGINVR 0 Gain factor for intrinsic gate tunnelling
current in inversion
IGOVR 0 Gain factor for Source/Drain overlap gate
tunnelling current
KOR 500m Body effect coefficient for reference device
overlap extensions
KPINV 0 Inverse body-effect factor for the poly gate
L 2u Drawn channel length in the layout
LAP 40n Effective channel length reduction per side
LMIN 150n Minimum effective channel length in
technology, used for calculation of
smoothing factor m
LP1 800n Characteristic length of 1'st lateral profile
LP2 800n Characteristic length of 2'nd lateral profile
LVAR 0 Difference between the actual and the
programmed poly-silicon gate length
MOEXP 1.34 Exponent of the length dependence of MO
MOO 0 Short-channel subthreshold slope parameter
MOR 0 Short-channel subthreshold slope per unit
length parameter
NFAR 1.573E23,3.825E24 First coefficient of the flicker noise for a
channel area of 1 um^2
NFBR 4.752G,1.015G Second coefficient of the flicker noise for a
NFCR 0,73n Third coefficient of the flicker noise for a
NU 2 Mobility field dependence exponent
NUEXP 5.25,3.23 NU temperature dependence exponent
PHIBR 950m Surface potential at strong inversion
SDIBLEXP 1.35 SDIBL length dependence exponent
SDIBLO 100u Drain-induced barrier lowering parameter
SL2KO 0 Second coefficient of the length dependence
of KO
SL2PHIB 0 Second coefficient of the length dependence
of PHIB
SLA1 0 A1 length dependence coefficient
679

SLALP 1 ALP length dependence coefficient
*SLCS 0 Coefficient of the length dependence of CS
SLETABET 0 ETABETR length dependence coefficient
SLKO 0 KO length dependence coefficient
SLPHIB 0 PHIB length dependence coefficient
SLSSF 1 SSF length dependence coefficient
SLTHESAT 1 THESAT length dependence coefficient
SSFR 6.25m Static feedback parameter
STA1 0 A1 temperature coefficient
STBGIDL -363.8u BGIDL temperature coefficient
STETAMOB 0 ETAMOB temperature coefficient
STPHIB -850u PHIB temperature coefficient
STVFB 500u VFB temperature coefficient
SWA1 0 A1 width dependence coefficient
SWALP 0 ALP width dependence coefficient
*SWCS 0 Coefficient of the width dependence of CS
SWETAMOB 0 ETAMOB width dependence coefficient
SWKO 0 KO width dependence coefficient
SWPHIB 0 PHIB width dependence coefficient
SWSSF 0 SSF width dependence coefficient
SWTHEPH 0 THEPH width dependence coefficient
SWTHER 0 THER width dependence coefficient
SWTHESAT 0 THESAT width dependence coefficient
SWTHESR 0 THESR width dependence coefficient
SWTHETH 0 THETH width dependence coefficient
THEPHR 12.9m,1m Coefficient of the mobility reduction due to
phonon scattering
series resistance
THERR 155m,80m Coefficient of the series resistance
THESATEXP 1 Exponent of length dependence of THESAT
THESATR 500m,200m Velocity saturation parameter due to
optical/acoustic phonon scattering

THESRR 400m,730m Coefficient of the mobility reduction due to
surface roughness scattering
THETHEXP 1 THETH length dependence exponent
THETHR 1m,500u Coefficient of self-heating
overlap extensions
modulation
W 10u Drawn channel width in the layout
WOT 0 Effective channel width reduction per side
WVAR 0 Difference between the actual and the
programmed field-oxide opening

681
Model parameters for level 12023/12026 (Philips 1202.3 Binning)

GATENOISE 0 Induced gate thermal noise flag
overlap extensions
KPINV 0 Inverse of body-effect factor of the
poly-silicon gate
L 2u Drawn channel length in the layout
LAP 40n Effective channel length reduction per side
LVAR 0 Difference between the actual and the
programmed poly-silicon gate length
NU 2 Mobility model field dependence exponent
Coefficients of length dependence

Name Default Parameter affected
PLA1 0 A1
PLA2 0 A2
PLA3 0 A3
PLAGIDL 0 AGIDL
PLALP 0 ALP
PLBACC 0 BACC
PLBET 0 BET
PLBGIDL 0 BGIDL
PLBINV 0 BINV
PLCGDO 0 CGDO
PLCGIDL 0 CGIDL
PLCGSO 0 CGSO
PLCOX 0 COX
PLCS 0 CS
PLETAMOB 0 ETAMOB
PLIGACC 0 IGACC
PLIGINV 0 IGINV
PLIGOV 0 IGOV
PLKO 0 KO
PLMEXP 0 MEXP
PLMO 0 MO
Coefficients of length dependence

PLNFA 0 NFA
PLNFB 0 NFB
PLNFC 0 NFC
PLPHIB 0 PHIB
PLSDIBL 0 SDIBL
PLSSF 0 SSF
PLTA1 0 STA1
PLTBGIDL 0 STBGIDL
PLTETABET 0 ETABET
*PLTETACS 0 ETACS
PLTETAMOB 0 STETAMOB
PLTETAPH 0 ETAPH
PLTETAR 0 ETAR
PLTETASAT 0 ETASAT
PLTETASR 0 ETASR
PLTHEPH 0 THEPH
PLTHER 0 THER
PLTHESAT 0 THESAT
PLTHESR 0 THESR
PLTHETH 0 THETH
PLTNUEXP 0 NUEXP
PLTPHIB 0 STPHIB
PLTVFB 0 STVFB
Coefficients of length*width product dependence

PLWA1 0 A1
PLWA2 0 A2
PLWA3 0 A3
PLWAGIDL 0 AGIDL
PLWALP 0 ALP
PLWBACC 0 BACC
PLWBET 0 BET
PLWBGIDL 0 BGIDL
PLWBINV 0 BINV
PLWCGDO 0 CGDO
PLWCGIDL 0 CGIDL
PLWCGSO 0 CGSO
683
Coefficients of length*width product dependence

PLWCOX 0 COX
*PLWCS 0 CS
PLWETAMOB 0 ETAMOB
PLWIGACC 0 IGACC
PLWIGINV 0 IGINV
PLWIGOV 0 IGOV
PLWKO 0 KO
PLWMEXP 0 MEXP
PLWMO 0 MO
PLWNFA 0 NFA
PLWNFB 0 NFB
PLWNFC 0 NFC
PLWPHIB 0 PHIB
PLWSDIBL 0 SDIBL
PLWSSF 0 SSF
PLWTA1 0 STA1
PLWTBGIDL 0 STBGIDL
PLWTETABET 0 ETABET
*PLWTETACS 0 ETACS
PLWTETAMOB 0 STETAMOB
PLWTETAPH 0 ETAPH
PLWTETAR 0 ETAR
PLWTETASAT 0 ETASAT
PLWTETASR 0 ETASR
PLWTHEPH 0 THEPH
PLWTHER 0 THER
PLWTHESAT 0 THESAT
PLWTHESR 0 THESR
PLWTHETH 0 THETH
PLWTNUEXP 0 NUEXP
PLWTPHIB 0 STPHIB
PLWTVFB 0 STVFB
Geometry independent values

Name Default Value
POA1 6.022,6.858 A1
POA2 38.02,57.32 A2
POA3 640.7m,425.4m A3

Name Default Value
POAGIDL 0 AGIDL
POALP 25m ALP
POBACC 48 BACC
POBET 1.922m,381.4u BET
POBGIDL 41 BGIDL
POBINV 48,87.5 BINV
POCGDO 6.392f,6.358f CGDO
POCGIDL 0 CGIDL
POCGSO 6.392f,6.358f CGSO
POCOX 29.8f,27.17f COX
*POCS 0 CS
POETAMOB 1.4,3 ETAMOB
POIGACC 0 IGACC
POIGINV 0 IGINV
POIGOV 0 IGOV
POKO 500m KO
POMEXP 200m MEXP
POMO 0 MO
PONFA 8.323E22,1.9E22 NFA
PONFB 25.14E6,5.043E6 NFB
PONFC 0,362.7p NFC
POPHIB 950m PHIB
POSDIBL 853u,35.51u SDIBL
POSSF 12m,10m SSF
POTA1 0 STA1
POTBGIDL -363.8u STBGIDL
POTETABET 1.3,500m ETABET
*POTETACS 0 ETACS
POTETAMOB 0 STETAMOB
POTETAPH 1.35,3.75 ETAPH
POTETAR 950m,400m ETAR
POTETASAT 1.04,860m ETASAT
POTETASR 650m,500m ETASR
POTHEPH 12.9m,1m THEPH
POTHER 81.2m,79m THER
POTHESAT 251.3m,172.8m THESAT
POTHESR 356.2m,730m THESR
POTHETH 10u,0 THETH
685

Name Default Value
POTNUEXP 5.25,3.23 NUEXP
POTPHIB -850u STPHIB
POTVFB 500u STVFB
Coefficients of width dependence

PWA1 0 A1
PWA2 0 A2
PWA3 0 A3
PWAGIDL 0 AGIDL
PWALP 0 ALP
PWBACC 0 BACC
PWBET 0 BET
PWBGIDL 0 BGIDL
PWBINV 0 BINV
PWCGDO 0 CGDO
PWCGIDL 0 CGIDL
PWCGSO 0 CGSO
PWCOX 0 COX
*PWCS 0 CS
PWETAMOB 0 ETAMOB
PWIGACC 0 IGACC
PWIGINV 0 IGINV
PWIGOV 0 IGOV
PWKO 0 KO
PWMEXP 0 MEXP
PWMO 0 MO
PWNFA 0 NFA
PWNFB 0 NFB
PWNFC 0 NFC
PWPHIB 0 PHIB
PWSDIBL 0 SDIBL
PWSSF 0 SSF
PWTA1 0 STA1
PWTBGIDL 0 STBGIDL
PWTETABET 0 ETABET
*PWTETACS 0 ETACS
PWTETAMOB 0 STETAMOB
Coefficients of width dependence

PWTETAPH 0 ETAPH
PWTETAR 0 ETAR
PWTETASAT 0 ETASAT
PWTETASR 0 ETASR
PWTHEPH 0 THEPH
PWTHER 0 THER
PWTHESAT 0 THESAT
PWTHESR 0 THESR
PWTHETH 0 THETH
PWTNUEXP 0 NUEXP
PWTPHIB 0 STPHIB
PWTVFB 0 STVFB
Remaining parameters
series resistance
overlap extensions
modulation
W 10u Drawn channel width in the layout
WVAR 0 Difference between the actual and the

687
The Philips Model 20 is an asymmetrical, surface-potential-based LDMOS
model, that describes the transition from weak to strong inversion in the channel
region, as well as from accumulation to depletion in the drift region. It provides a
good physical description of effects important for LDMOS devices, such as weak
and strong inversion in the channel region, accumulation and depletion in the
drift region, mobility reduction in both the channel and the drift region, velocity
saturation in the channel region, conductance effects of the channel region (chan-
nel length modulation, DIBL and static feedback), and weak avalanche currents
in the channel region.
SPICE format
Syntax for Electrical model
+ [M=<no_parallel>]
Syntax for Geometrical model

+ [W=<channel_width>] [WD=<drift_width>] [M=<no_parallel>]

Example of device instance for Geometrical model
MS 1 2 3 4 NSC W=20u WD=25u M=200
Schematic format
PART attribute
<name>

M1
Syntax for Geometrical VALUE attribute

[W=<channel_width>] [WD=<drift_width>] [M=<no_parallel>]
Examples of VALUE attribute for Geometrical model

W=30u WD=15u M=5

Syntax for Electrical model VALUE attribute
[M=<no_parallel>]
MODEL attribute
<model name>

MODS12
In the Geometrical model, <channel_width> and <drift_width> are the drawn

widths of the channel and drift region, before side diffusion, in meters. They can
be specified as device attributes or as model parameters. Device attributes, if
specified, supersede model parameters. If neither model nor device attributes are
specified, they assume the default values of W=20u, WD=20u.

There are two versions of Model 20 available for use within Micro-Cap, the
Electrical and the Geometric. The latest model, dated March 2006, will be docu-
mented here.
Micro-Cap Level Philips Level Description

20012 20.01 MOS20 Electrical
20012 20.01 MOS20 Geometrical
20014 20.01 MOS20 Electrical Thermal
20015 20.01 MOS20 Geometrical Thermal
20021 20.02 MOS20 Electrical
20022 20.02 MOS20 Geometrical
In the table below, the model parameters are listed in alphabetical order.
Parameters marked with an asterisk are new to the Philips 20.02 models and are
not available in the Philips 20.01 models.
689
Parameters MOSFET level 20012/20021/20014/ (Philips 20 Electrical)

A1 15 Factor of weak avalanche current, at TREF
A2 73 Exponent of weak avalanche current
A3 800m Factor of the drain-source voltage above which
weak avalanche occurs
ALP 2m Factor for channel length modulation
BET 1.4m Gain factor of channel region at TREF
BETACC 1.4m Gain factor of drift region for accumulation in
the channel region, at TREF
CGDO 0 Gate-to-drain overlap capacitance
CGSO 0 Gate-to-source overlap capacitance
COX 15f Oxide capacitance for intrinsic channel region
COXD 15f Oxide capacitance for intrinsic drift region
DTA 0 Temperature offset to ambient temperature
ETABET 1.6 Temperature scaling exponent for BET
ETABETACC 1.5 Temperature scaling exponent for BETACC
ETARD 1.5 Temperature scaling exponent for RD
ETATHE3 1 Temperature scaling exponent for THE3
*ETATHE3D 1 Temperature scaling exponent for THE3D
KO 1.6 Body factor of the channel region
KOD 1 Body factor of the drift region
LAMD 200m Ratio of the depletion layer thickness to the
effective thickness of the drift region at VSB= 0
MEXP 2 Smoothing factor from linear to saturation
*MEXPD 2 Smoothing factor from linear to quasi-saturation
MO 0 Parameter for short-channel sub-threshold slope
MSDIBL 3 Exponent for the drain-induced barrier lowering
dependence on the backgate bias
NFA 7E23 First coefficient of flicker noise
NFB 10E6 Second coefficient of flicker noise
NFC 0 Third coefficient of flicker noise
NT .00001645f Coefficient of thermal noise, at TREF
PHIB 860m Surface potential at the onset of strong inversion
in the channel region, at TREF
PHIBD 780m Surface potential at the onset of strong inversion
in the drift region, at TREF

Parameters MOSFET level 20012/20021/20014/ (Philips 20 Electrical)

RD 200 On-resistance of the drift region, at TREF
SDIBL 1m Factor for drain-induced barrier lowering
SSF 1p Factor for static feedback
STA1 0 Temperature scaling coefficient for A1
STPHIB -1.2m Temperature scaling coefficient for PHIB
STPHIBD -1.2m Temperature scaling coefficient for PHIBD
STVFB 0 Temperature scaling coefficient for VFB
STVFBD 0 Temperature scaling coefficient for the flatband
voltage of the drift region
THE1 90m Mobility reduction coefficient in the channel
region due to vertical electrical field caused by
strong inversion
THE1ACC 20m Mobility reduction coefficient in the drift region
due to the vertical electrical field caused by
accumulation
THE2 30m Mobility reduction coefficient for VSB > 0 in the
channel region due to the vertical electrical field
caused by depletion
THE3 400m Mobility reduction coefficient in the channel
region due to the horizontal electrical field
caused by velocity saturation
*THE3D 0 Mobility reduction coefficient in the drift
region due to the horizontal electrical field
caused by velocity saturation
TOX 38n Thickness of the oxide above the channel region
VFB -1 Flatband voltage of the channel region, at TREF
VFBD -100m Flatband voltage of the drift region, at TREF
VP 50m Characteristic voltage of channel length
modulation
For the electrical thermal model (Level = 20014) only there are three additional
parameters to handle thermal effects.

691
Parameters MOSFET level 20012/20022/20015 (Philips 20 Geometrical)

A1R 15 Factor of weak avalanche current of an infinitely
wide transistor, at TREF
A2 73 Exponent of weak avalanche current
A3 800m Factor of the drain-source voltage above which
weak avalanche occurs
ALP 2m Factor for channel length modulation
BETACCW 70u Gain factor of 1 um wide drift region at TREF
BETW 70u Gain factor of 1 um wide channel at TREF
CGDOW 0 Gate-to-drain overlap capacitance for a drift
region of 1 um width
CGSOW 0 Gate-to-source overlap capacitance for a
channel region of 1 um width
COXDW .75f Oxide capacitance for an intrinsic drift region of
1um width
COXW .75f Oxide capacitance for an intrinsic channel region
of 1um width
DTA 0 Temperature offset to the ambient temperature
ETABET 1.6 Temperature scaling exponent for BET
ETABETACC 1.5 Temperature scaling exponent for BETACC
ETARD 1.5 Temperature scaling exponent for RD
ETATHE3 1 Temperature scaling exponent for THE3
*ETATHE3D 1 Temperature scaling exponent for THE3D
KODR 1 Drift region body factor with W = infinite
KOR 1.6 Channel region body factor with W = infinite
LAMD 200m Quotient of the depletion layer thickness to the
effective thickness of the drift region at VSB = 0
MEXP 2 Smoothing factor from linear to saturation
*MEXPD 2 Smoothing factor from linear to quasi-saturation
MO 0 Parameter for short-channel sub-threshold slope
MSDIBL 3 Exponent for the drain-induced barrier lowering
dependence on the backgate bias
NFAW 1.4E25 First coefficient of flicker noise for a channel
NFBW 200MEG Second coefficient of flicker noise for a channel


NFCW 0 Third coefficient of flicker noise for a channel
NT .00001645f Coefficient of thermal noise, at TREF
PHIB 860m Surface potential at the onset of strong inversion
in the channel region, at TREF
PHIBD 780m Surface potential at the onset of strong inversion
in the drift region, at TREF
RDW 4K On-resistance of a 1u wide drift region at TREF
SDIBL 1m Factor for drain-induced barrier lowering
SSF 1p Factor for static feedback
STA1 0 Temperature scaling coefficient for A1
STPHIB -1.2m Temperature scaling coefficient for PHIB
STPHIBD -1.2m Temperature scaling coefficient for PHIBD
STVFB 0 Temperature scaling coefficient for VFB
STVFBD 0 Temperature scaling coefficient for the flatband
voltage of the drift region
SWA1 0 Width scaling coefficient for A1
SWKO 0 Width scaling coefficient for KO
SWKOD 0 Width scaling coefficient for the body factor of
the drift region
SWTHE1 0 Width scaling coefficient for THE1
SWTHE3D 0 Width scaling coefficient for THE3D
THE1ACC 20m Mobility reduction coefficient in drift region due
to vertical electrical field from accumulation
THE1R 90m Mobility reduction coefficient of infinitely wide
transistor, due to vertical strong-inversion field
in a channel region
THE2R 30m Mobility reduction coefficient for VSB > 0 of an
infinitely wide transistor, due to vertical
depletion field in channel region
*THE3DR 400m Mobility reduction coefficient in the drift region
due to the horizontal electrical field caused by
velocity saturation
THE3R 400m Mobility reduction coefficient in channel region
of an infinitely wide transistor due to velocity
saturation
693

TOX 38n Thickness of the oxide above the channel region
VFB -1 Flatband voltage of the channel region, at TREF
VFBD -100m Flatband voltage of the drift region, at TREF
VP 50m Characteristic voltage of channel length
modulation
W 20u Drawn width of the channel region
WD 20u Drawn width of the drift region
WDVAR 0 Width offset of the drift region
WVAR 0 Width offset of the channel region
For the geometrical thermal model (Level = 20015) only there are three addition-
al parameters to handle thermal effects.

Figure 26-22 Philips Model 20 equivalent circuit

The Philips Model 31 is a physics based device model suitable for high voltage
applications. The model describes the electrical behavior of a junction isolated
accumulation / depletion MOSFET. The model is used as the drain extension of
high voltage MOS devices, like the Lateral Double-diffused (LDMOS), Vertical
Double-diffused (VDMOS), and Extended MOS transistors.
SPICE format
Syntax
+ [M=<no_parallel>]

Example of device instance
MS 1 2 3 4 NSC M=200
Schematic format
PART attribute
<name>

M1
VALUE attribute
[M=<no_parallel>]
Examples of VALUE attribute

M=5
MODEL attribute
<model name>

LM1
695
There are two versions of Model 31 available for use within Micro-Cap, one that
includes self-heating (thermal) effects and one that does not. The latest model,
dated March 2006, will be documented here.

31001 3100.3 Model 31
31004 3100.3 Model 31 Thermal

Model parameters for MOSFET level 31001/31004 (Philips Model 31)

LEVEL 31001 Standard model =31001, Thermal model = 31004
or 31004
ACH 1 Channel resistivity temperature coefficient
ACHMOD 0 Extended temperature scaling flag (0 = Off)
ACHRON 0 Temperature coefficient of ohmic resistance at
zero bias
ACHRSAT 0 Temperature coefficient of space charge
resistance at zero bias
ACHVSAT 0 Temperature coefficient of critical drain-source
voltage for hot carriers
CGATE 1p Gate capacitance at zero bias
CSUB 500f Substrate capacitance at zero bias
DCH 1E22 Doping level channel
DSUB 1E21 Doping level substrate
PSAT 1 Velocity saturation coefficient
RON 10 Ohmic resistance at zero bias
RSAT 100 Space charge resistance at zero bias
TAUSC 1p Space charge transit time of the channel
TOX 100n Gate oxide thickness
VGAP 1.206 Bandgap voltage channel
VP 10 Pinch off voltage at zero gate and substrate voltages
VSAT 1 Critical drain-source voltage for hot carriers
VSUB 600m Substrate diffusion voltage

697
The Philips Model 40 is a physics based model for high voltage applications.
The model describes the electrical behavior of an accumulation / depletion type
MOSFET in Silicon-on-Insulator (SOI) processes. The model is used as a drain
extension of high-voltage MOS devices, like the Lateral Double-diffused MOS
(LDMOS), the Vertical Double-diffused MOS (VDMOS), and the Extended
MOS transistors.
SPICE format
Syntax
+ [M=<no_parallel>]

Example of device instance
MS 1 2 3 4 NSO M=100
Schematic format
PART attribute
<name>
Example of PART attribute

MS1
VALUE attribute
[M=<no_parallel>]
Example of VALUE attribute

M=10
MODEL attribute
<model name>
Example of MODEL attribute

LM1

There are two versions of Model 40 available for use within Micro-Cap, one that
includes self-heating (thermal) effects and one that does not. The latest model,
dated March 2006, is documented here.

40001 40.2 Model 40
40004 40.2 Model 40 Thermal
699
Model parameters for MOSFET level 40001/40004 (Philips Model 40)

ACH 0 Temperature coefficient resistivity of the channel
ACHMOD 0 Parameter to switch to extended temperature scaling
ACHRON 0 Temperature coefficient of ohmic resistance at
zero bias
ACHRSAT 0 Temperature coefficient of space charge resistance
at zero bias
ACHVSAT 0 Temperature coefficient of critical drain-source
voltage for hot carriers
CBOX 0 Wafer capacitance
CGATE 0 Gate capacitance at zero bias
DCH 1E21 Doping level channel
PSAT 1 Velocity saturation coefficient
RON 1 Ohmic resistance at zero bias
RSAT 1 Space charge resistance at zero bias
TAUSC 0 Space charge transit time of the channel
TBOX -1 Box thickness
TOX -1 Gate oxide thickness
VP -1 Pinch off voltage at zero gate and substrate voltages
VSAT 10 Critical drain-source voltage for hot carriers


MOSFET PSP Model
The PSP model is a relatively new compact MOSFET model, jointly developed
by Philips Research and Penn State University. It is a surface-potential based
MOS model, describing the key physical effects necessary to model modern
deep-submicron bulk CMOS technologies. These effects include mobility
reduction, velocity saturation, DIBL, gate current, lateral doping gradient effects,
and STI stress. The source/drain junction model, called the JUNCAP2 model, is
an integrated part of the PSP model.
SPICE format
Syntax of electrical model
+ [ABSOURCE=<bottom_area_source>]
+ [ABDRAIN=<bottom_area_drain>]
+ [LSSOURCE=<sti_edge_source>] [LSDRAIN=<sti_edge_drain>]
+ [LGSOURCE=<gate_edge_source>] [LGDRAIN=<gate_edge_drain>]
+ [AD=<drainarea>] [AS=<sourcearea>]
+ [MULT=<no_parallel>]
+ [JW=<jw>]

Syntax of geometrical and binning models
+ [L=<length>] [W=<width>]
+ [SA=<source_od-poly_distance>] [SB=<drain_od-poly_distance>]
+ [ABSOURCE=<bottom_area_source>]
+ [ABDRAIN=<bottom_area_drain>]
+ [LSSOURCE=<sti_edge_source>] [LSDRAIN=<sti_edge_drain>]
+ [LGSOURCE=<gate_edge_source>] [LGDRAIN=<gate_edge_drain>]
+ [AD=<drainarea>] [AS=<sourcearea>]
+ [MULT=<no_parallel>]
Example of electrical model instance

MS 1 2 3 4 NSO ABSOURCE=2.5p ABDRAIN=3p
Example of geometrical or binning model instance

MS 1 2 3 4 NSO W=.2u L=.09u SA=.12u SB=.12u
701
Schematic format
PART attribute
<name>

MP1, MX2
Syntax for electrical model VALUE attribute

[ABSOURCE=<bottom_area_source>]
[ABDRAIN=<bottom_area_drain>]
[LSSOURCE=<sti_edge_source>] [LSDRAIN=<sti_edge_drain>]
[LGSOURCE=<gate_edge_source>] [LGDRAIN=<gate_edge_drain>]
[AD=<drainarea>] [AS=<sourcearea>]
[PD=<drainperiphery>] [PS=<sourceperiphery>]
[MULT=<no_parallel>]
[JW=<jw>]
Syntax for geometrical and binning model VALUE attribute

[L=<l>] [W=<w>]
[SA=<source_od-poly_distance>] [SB=<drain_od-poly_distance>]
[ABSOURCE=<bottom_area_source>]
[ABDRAIN=<bottom_area_drain>]
[LSSOURCE=<sti_edge_source>] [LSDRAIN=<sti_edge_drain>]
[LGSOURCE=<gate_edge_source>] [LGDRAIN=<gate_edge_drain>]
[AD=<drainarea>] [AS=<sourcearea>]
[PD=<drainperiphery>] [PS=<sourceperiphery>]
[MULT=<no_parallel>]
Example of VALUE attribute for electrical model

ABSOURCE=5p ABDRAIN=18p LSSOURCE=.2u
Example of VALUE attribute for geometrical or binning model

MULT=10 W=.5u L=.1u SA=.2u SB=.25u
MODEL attribute
<model name>
Example of MODEL attribute

P102_N
<w> and <l> are the drawn device dimensions, before side diffusion, in meters.

<source_od-poly_distance> is the distance between OD-edge and poly at the
source side.
<drain_od-poly_distance> is the distance between OD-edge and poly at the

drain side.
<bottom_area_source> is the source junction bottom area.
<bottom_area_drain> is the drain junction bottom area.
<sti_edge_source> is the STI-edge part of source junction perimeter.
<sti_edge_drain> is the STI-edge part of drain junction perimeter.
<gate_edge_source> is the gate-edge part of source junction perimeter.
<gate_edge_drain> is the gate-edge part of drain junction perimeter.
<sourcearea> and <drainarea> are the diffusion areas (sq. m).

<jw> is the gate-edge length of source/drain junction. If specified, it overrides the

JW model parameter.
These device instance parameters can be specified as device attributes or as mod-

el parameters. Device attributes, if specified, supersede model parameters.
There are three versions of PSP model available, an electrical model, a geometri-
cal model, and a binning model. Each is based on the Philips model 102, dated
July 2006. There is no thermal version of this model.

102 102.0 PSP electrical
1020 102.0 PSP geometrical
1021 102.0 PSP binning
703
The equivalent circuit of the PSP model is as follows:
Figure 26-25 PSP model equivalent circuit
In the table below, the Level parameter is shown first and the remainder of the
model parameters are listed in alphabetical order.
Model parameters for MOSFET level 102 (PSP Electrical Model)

LEVEL 102 Model level =102
A1 1 Impact-ionization pre-factor
A2 10 Impact-ionization exponent at TR
A3 1 Saturation voltage dependence of impact
ionization
A4 0 Back-bias dependence of impact-ionization
ABDRAIN 1p Bottom area of drain junction
ABSOURCE 1p Bottom area of source junction
AD 1p Bottom area of drain junction
AGIDL 0 GIDL pre-factor


ALP 10m CLM pre-factor
ALP1 0 CLM enhancement factor above threshold
ALP2 0 CLM enhancement factor below threshold
AS 1p Bottom area of source junction
AX 3 Linear/saturation transition factor
BETN 70m Channel aspect ratio times zero-field mobility
BGIDL 41 GIDL probability factor at TR
CBBTBOT 1p Band-to-band tunneling prefactor of bottom part
CBBTGAT .001f Band-to-band tunneling prefactor of gate-edge
component
CBBTSTI .001f Band-to-band tunneling prefactor of STI-edge
component
CF 0 DIBL-parameter
CFB 0 Back bias dependence of CF
CFR 0 Outer fringe capacitance
CGBOV 0 Oxide capacitance for gate-bulk overlap
CGIDL 0 Back-bias dependence of GIDL
CGOV 1f Oxide capacitance for gate-drain/source overlap
CHIB 3.1 Tunneling barrier height
CJORBOT 1m Zero-bias capacitance per unit-of-area of bottom
component
CJORGAT 1n Zero-bias capacitance per unit-of-length of gate-
edge component
CJORSTI 1n Zero-bias capacitance per unit-of-length of STI-
edge component
COX 10f Oxide capacitance for intrinsic channel
CS 0 Coulomb scattering parameter at TR
CSRHBOT 100 Shockley-Read-Hall prefactor of bottom part
GAT 100u Shockley-Read-Hall prefactor of gate-edge part
CSRHSTI 100u Shockley-Read-Hall prefactor of STI-edge part
CT 0 Interface states factor
CTATBOT 100 Trap-assisted tunneling prefactor of bottom part
CTATGAT 100u Trap-assisted tunneling prefactor of gate edge
part
CTATSTI 100u Trap-assisted tunneling prefactor of STI-edge
part
DNSUB 0 Effective doping bias-dependence parameter
DPHIB 0 Offset parameter for PHIB
DTA 0 Temperature offset w.r.t. ambient temperature
705

FBBTRBOT 1G Normalization field at the reference temperature
for band-to-band tunneling of bottom part
FBBTRGAT 1G Normalization field at the reference temperature
for band-to-band tunneling of gate-edge part
FBBTRSTI 1G Normalization field at the reference temperature
for band-to-band tunneling of STI-edge part
FETA 1 Effective field parameter
FNT 1 Thermal noise coefficient
GC2 375m Gate current slope factor
GC3 63m Gate current curvature factor
GCO 0 Gate tunneling energy adjustment
temperature of bottom part
temperature of gate-edge part
temperature of STI-edge part
IGINV 0 Gate channel current pre-factor
IGOV 0 Gate overlap current pre-factor
JW 1u Gate-edge length of source/drain junction
LGDRAIN 1u Gate-edge length of drain junction
LGSOURCE 1u Gate-edge length of source junction
LSDRAIN 1u STI-edge length of drain junction
LSSOURCE 1u STI-edge length of source junction
tunneling of bottom part
tunneling of gate-edge part
tunneling of STI-edge part
MUE 500m Mobility reduction coefficient at TR
NEFF 5E23 Effective substrate doping
NFA 8E22 First coefficient of flicker noise
NFB 30MEG Second coefficient of flicker noise
NFC 0 Third coefficient of flicker noise


NOV 5E25 Effective doping of overlap region
NP 1E26 Gate poly-silicon doping
NSLP 50m Effective doping bias-dependence parameter
PBRBOT 4 Breakdown onset tuning parameter of bottom
part
PBRGAT 4 Breakdown onset tuning parameter of gate-edge
part
PBRSTI 4 Breakdown onset tuning parameter of STI-edge
part
PD 1u Perimeter of drain junction
PGAT 500m Grading coefficient of gate-edge part
PHIGBOT 1.16 Zero-temperature bandgap voltage of bottom
part
PHIGGAT 1.16 Zero-temperature bandgap voltage of gate-edge
part
PHIGSTI 1.16 Zero-temperature bandgap voltage of STI-edge
part
PS 1u Perimeter of source junction
QMC 1 Quantum-mechanical correction factor
RS 30 Series resistance at TR
RSB 0 Back-bias dependence of series resistance
RSG 0 Gate-bias dependence of series resistance
STA2 0 Temperature dependence of A2
STBET 1 Temperature dependence of BETN
STBGIDL 0 Temperature dependence of BGIDL
STCS 0 Temperature dependence of CS
STIG 2 Temperature dependence of IGINV and IGOV
STMUE 0 Temperature dependence of MUE
STRS 1 Temperature dependence of RS
STTHEMU 1.5 Temperature dependence of THEMU
STTHESAT 1 Temperature dependence of THESAT
707

STVFB 500u Temperature dependence of VFB
STXCOR 0 Temperature dependence of XCOR
SWGIDL 0 Flag for GIDL current, 0=turn off IGIDL
SWIGATE 0 Flag for gate current, 0=turn off IG
SWIMPACT 0 Flag for impact ionization current, 0=turn off II
SWJUNCAP 0 Flag for juncap, 0=turn off juncap
THEMU 1.5 Mobility reduction exponent at TR
THESAT 1 Velocity saturation parameter at TR
THESATB 0 Back-bias dependence of velocity saturation
THESATG 0 Gate-bias dependence of velocity saturation
TOX 2n Gate oxide thickness
TOXOV 2n Overlap oxide thickness
TR 21 Nominal (reference) temperature
bottom part
gate-edge part
STI-edge part
VBRBOT 10 Breakdown voltage of bottom part
VBRGAT 10 Breakdown voltage of gate-edge part
VBRSTI 10 Breakdown voltage of STI-edge part

Model parameters for MOSFET level 1020 (PSP Geometrical Model)

LEVEL 1020 Model level 1020 for geometrical model
A1L 0 Length dependence of A1
A1O 1 Geometry independent impact-ionization pre-factor
A1W 0 Width dependence of A1
A2O 10 Impact-ionization exponent at TR
A3O 1 Geometry independent saturation-voltage
dependence of II
A4O 0 Geometry independent back-bias dependence of II
AGIDLW 0 Width dependence of GIDL pre-factor
ALP1L1 0 Length dependence of CLM enhancement factor
above threshold
ALP1L2 0 Second order length dependence of ALP1
ALP1LEXP 500m Exponent for length dependence of ALP1
ALP1W 0 Width dependence of ALP1
ALP2L1 0 Length dependence of CLM enhancement factor
below threshold
ALP2L2 0 Second order length dependence of ALP2
ALP2LEXP 500m Exponent for length dependence of ALP2
ALP2W 0 Width dependence of ALP2
ALPL 500u Length dependence of ALP
ALPLEXP 1 Exponent for length dependence of ALP
ALPW 0 Width dependence of ALP
AXL 400m Length dependence of AX
AXO 18 Geometry independent linear/saturation transition
factor
BETW1 0 First higher-order width scaling coefficient of BETN
BETW2 0 Second higher-order width scaling coefficient of
BETN
BGIDLO 41 GIDL probability factor at TR
CBBTGAT .001f Band-to-band tunneling prefactor of gate-edge part
709

CBBTSTI .001f Band-to-band tunneling prefactor of STI-edge part
CFBO 0 Back-bias dependence of CF
CFL 0 Length dependence of DIBL-parameter
CFLEXP 2 Exponent for length dependence of CF
CFRW 0 Outer fringe capacitance for 1 um wide channel
CFW 0 Width dependence of CF
CGBOVL 0 Gate-bulk overlap oxide capacitance per sq. micron
CGIDLO 0 Back-bias dependence of GIDL
CHIBO 3.1 Tunneling barrier height
CJORBOT 1m Zero-bias capacitance per unit area of bottom part
gate-edge part
CJORSTI 1n Zero-bias capacitance per unit length of
STI-edge part
CSL 0 Length dependence of CS
CSLEXP 0 Exponent for length dependence of CS
CSLW 0 Area dependence of CS
CSO 0 Geometry independent coulomb scattering parameter
at TR
CSRHBOT 100 Shockley-Read-Hall prefactor of bottom component
CSRHGAT 100u Shockley-Read-Hall prefactor of gate-edge part
CSW 0 Width dependence of CS
CTATGAT 100u Trap-assisted tunneling prefactor of gate-edge part
CTATSTI 100u Trap-assisted tunneling prefactor of STI-edge part
CTL 0 Length dependence of interface states factor
CTLEXP 1 Exponent for length dependence of interface
states factor
CTLW 0 Area dependence of interface states factor
CTO 0 Geometry-independent interface states factor
CTW 0 Width dependence of interface states factor
DLQ 0 Effective channel length reduction for CV
DNSUBO 0 Effective doping bias-dependence parameter
DPHIBL 0 Length dependence offset of PHIB
DPHIBLEXP 1 Exponent for length dependence of offset of PHIB
DPHIBLW 0 Area dependence of offset of PHIB
DPHIBO 0 Geometry independent offset of PHIB
DPHIBW 0 Width dependence of offset of PHIB

DTA 0 Temperature offset w.r.t. ambient circuit
temperature
DWQ 0 Effective channel width reduction for CV
FBBTRBOT 1G Normalization field at the reference temperature for
band-to-band tunneling of bottom part
FBBTRGAT 1G Normalization field at the reference temperature for
band-to-band tunneling of gate-edge part
FBBTRSTI 1G Normalization field at the reference temperature for
band-to-band tunneling of STI-edge part
FBET1 0 Relative mobility decrease due to first lateral profile
FBET1W 0 Width dependence of relative mobility decrease due
to first lateral profile
FBET2 0 Relative mobility decrease due to second lateral
profile
FETAO 1 Effective field parameter
FNTO 1 Thermal noise coefficient
FOL1 0 First length dependence coefficient for short channel
body effect
FOL2 0 Second length dependence coefficient for short
channel body effect
GC2O 375m Gate current slope factor
GC3O 63m Gate current curvature factor
GCOO 0 Gate tunneling energy adjustment
IGINVLW 0 Gate channel current pre-factor for 1 um^2
channel area
IGOVW 0 Gate overlap current pre-factor for 1 um wide
channel
KUO 0 Mobility degradation/enhancement coefficient
KVSAT 0 Saturation velocity degradation/enhancement
coefficient
KVTHO 0 Threshold shift parameter
711

L 10u Design length
LAP 0 Effective channel length reduction per side
LKUO 0 Length dependence of KUO
LKVTHO 0 Length dependence of KVTHO
LLODKUO 0 Length parameter for UO stress effect
LLODVTH 0 Length parameter for VTH stress effect
LODETAO 1 Eta0 shift modification factor for stress effect
LOV 0 Overlap length for gate/drain and gate/source
overlap capacitance
LP1 10n Mobility-related characteristic length of first
lateral profile
LP1W 0 Width dependence of mobility-related characteristic
length of first lateral profile
LP2 10n Mobility-related characteristic length of second
lateral profile
LPCK 10n Char. length of lateral doping profile
LPCKW 0 Width dependence of char. length of lateral doping
profile
LVARL 0 Length dependence of LVAR
LVARO 0 Geom. independent difference between actual and
programmed gate length
LVARW 0 Width dependence of LVAR
MUEO 500m Geometry independent mobility reduction coefficient
at TR
MUEW 0 Width dependence of mobility reduction coefficient
at TR
NFALW 8E22 First coefficient of flicker noise for 1 um^2 channel
area

NFBLW 30MEG Second coefficient of flicker noise for 1 um^2
channel area
NFCLW 0 Third coefficient of flicker noise for 1 um^2
channel area
NOVO 5E25 Effective doping of overlap region
NPCK 1E24 Pocket doping level
NPCKW 0 Width dependence of pocket doping NPCK due to
segregation
NPL 0 Length dependence of gate poly-silicon doping
NPO 1E26 Geometry-independent gate poly-silicon doping
NSLPO 50m Effective doping bias-dependence parameter
NSUBO 3E23 Geometry independent substrate doping
NSUBW 0 Width dependence of background doping NSUBO
due to segregation
PBRBOT 4 Breakdown onset tuning parameter of bottom part
PBRGAT 4 Breakdown onset tuning parameter of gate-edge part
PBRSTI 4 Breakdown onset tuning parameter of STI-edge part
PGAT 500m Grading coefficient of gate-edge part
PHIGBOT 1.16 Zero-temperature bandgap voltage of bottom part
PHIGGAT 1.16 Zero-temperature bandgap voltage of gate-edge part
PHIGSTI 1.16 Zero-temperature bandgap voltage of STI-edge part
PKUO 0 Cross-term dependence of KUO
PKVTHO 0 Cross-term dependence of KVTHO
PSTI 500m Grading coefficient of STI-edge part
RSBO 0 Back-bias dependence of series resistance
RSGO 0 Gate-bias dependence of series resistance
RSW1 2.5K Source/drain series resistance for 1 um wide channel
at TR
RSW2 0 Higher-order width scaling of RS
SA 0 Distance between OD-edge to poly from one side
SAREF 1u Reference distance between OD-edge to poly from
one side
SB 0 Distance between OD-edge to poly from other side
SBREF 1u Reference distance between OD-edge to poly from
other side
713

STA2O 0 Temperature dependence of A2
STBETL 0 Length dependence of temperature dependence of
BETN
STBETLW 0 Area dependence of temperature dependence of
BETN
STBETO 1 Geometry independent temperature dependence of
BETN
STBETW 0 Width dependence of temperature dependence of
BETN
STBGIDLO 0 Temperature dependence of BGIDL
STCSO 0 Temperature dependence of CS
STETAO 0 Eta0 shift factor related to VTHO change
STIGO 2 Temperature dependence of IGINV and IGOV
STMUEO 0 Temperature dependence of MUE
STRSO 1 Temperature dependence of RS
STTHEMUO 1.5 Temperature dependence of THEMU
STTHESATL 0 Length dependence of temperature dependence of
THESAT
STTHESATLW 0 Area dependence of temperature dependence of
THESAT
STTHESATO 1 Geometry independent temperature dependence of
THESAT
STTHESATW 0 Width dependence of temperature dependence of
THESAT
STVFBL 0 Length dependence of temperature dependence of
VFB
STVFBLW 0 Area dependence of temperature dependence of
VFB
STVFBO 500u Geometry-independent temperature dependence of
VFB
STVFBW 0 Width dependence of temperature dependence of
VFB
STXCORO 0 Temperature dependence of XCOR

THEMUO 1.5 Mobility reduction exponent at TR
THESATBO 0 Back-bias dependence of velocity saturation
THESATGO 0 Gate-bias dependence of velocity saturation
THESATL 50m Length dependence of THESAT
THESATLEXP 1 Exponent for length dependence of THESAT
THESATLW 0 Area dependence of velocity saturation parameter
THESATO 0 Geometry independent velocity saturation parameter
at TR
THESATW 0 Width dependence of velocity saturation parameter
TKUO 0 Temperature dependence of KUO
TOXO 2n Gate oxide thickness
TOXOVO 2n Overlap oxide thickness
UO 50m Zero-field mobility at TR
bottom part
gate-edge part
STI-edge part
VFBL 0 Length dependence of flat-band voltage
VFBLW 0 Area dependence of flat-band voltage
VFBO -1 Geometry-independent flat-band voltage at TR
VFBW 0 Width dependence of flat-band voltage
VNSUBO 0 Effective doping bias-dependence parameter
VPO 50m CLM logarithmic dependence parameter
W 10u Design width
WBET 1n Characteristic width for width scaling of BETN
WKUO 0 Width dependence of KUO
WKVTHO 0 Width dependence of KVTHO
WLOD 0 Width parameter
715

WLODKUO 0 Width parameter for UO stress effect
WLODVTH 0 Width parameter for VTH stress effect
WSEG 10n Char. length of segregation of background doping
NSUBO
WSEGP 10n Char. length of segregation of pocket doping NPCK
WVARL 0 Length dependence of WVAR
WVARO 0 Geom. independent difference between actual and
Programmed field-oxide opening
WVARW 0 Width dependence of WVAR
XCORL 0 Length dependence of non-universality parameter
XCORLW 0 Area dependence of non-universality parameter
XCORO 0 Geometry independent non-universality parameter
XCORW 0 Width dependence of non-universality parameter

Model parameters for MOSFET level 1021 (PSP Binning Model)

LEVEL 1021 Model level 1021 for binning model
CBBTGAT .001f Band-to-band tunneling prefactor of gate-edge part
CBBTSTI .001f Band-to-band tunneling prefactor of STI-edge part
CJORBOT 1m Zero-bias capacitance per unit area of bottom part
gate-edge part
CJORSTI 1n Zero-bias capacitance per unit-of-length of
STI-edge part
CSRHBOT 100 Shockley-Read-Hall prefactor of bottom part
CSRHGAT 100u Shockley-Read-Hall prefactor of gate-edge part
CTATGAT 100u Trap-assisted tunneling prefactor of gate-edge part
CTATSTI 100u Trap-assisted tunneling prefactor of STI-edge part
DLQ 0 Effective channel length reduction for CV
DTA 0 Temperature offset w.r.t. ambient temperature
DWQ 0 Effective channel width reduction for CV
FBBTRBOT 1G Normalization field at the reference temperature for
band-to-band tunneling of bottom part
FBBTRGAT 1G Normalization field at the reference temperature for
band-to-band tunneling of gate-edge part
FBBTRSTI 1G Normalization field at the reference temperature for
band-to-band tunneling of STI-edge part
KUO 0 Mobility degradation/enhancement coefficient
KVSAT 0 Saturation velocity degradation/enhancement
coefficient
717

KVTHO 0 Threshold shift parameter
L 10u Design length
LAP 0 Effective channel length reduction per side due to
lateral diffusion of source/drain dopant ions
LKUO 0 Length dependence of KUO
LKVTHO 0 Length dependence of KVTHO
LLODKUO 0 Length parameter for UO stress effect
LLODVTH 0 Length parameter for VTH stress effect
LODETAO 1 Eta0 shift modification factor for stress effect
LVARL 0 Length dependence of difference between actual
and programmed poly-silicon gate length
LVARO 0 Geometry independent difference between actual
and programmed poly-silicon gate length
PBRBOT 4 Breakdown onset tuning parameter of bottom part
PBRGAT 4 Breakdown onset tuning parameter of gate-edge part
PBRSTI 4 Breakdown onset tuning parameter of STI-edge part
PGAT 500m Grading coefficient of gate-edge component
PHIGBOT 1.16 Zero-temperature bandgap voltage of bottom part
PHIGGAT 1.16 Zero-temperature bandgap voltage of gate-edge part
PHIGSTI 1.16 Zero-temperature bandgap voltage of STI-edge part
PKUO 0 Cross-term dependence of KUO
PKVTHO 0 Cross-term dependence of KVTHO
PLA1 0 Coefficient for the length dependence of A1
PLAGIDL 0 Coefficient for the length dependence of AGIDL

PLALP 0 Coefficient for length dependence of ALP
PLALP1 0 Coefficient for length dependence of ALP1
PLALP2 0 Coefficient for length dependence of ALP2
PLAX 0 Coefficient for length dependence of AX
PLBETN 0 Coefficient for length dependence of BETN
PLCF 0 Coefficient for length dependence of CF
PLCFR 0 Coefficient for length dependence of CFR
PLCGBOV 0 Coefficient for length dependence of CGBOV
PLCGOV 0 Coefficient for length dependence of CGOV
PLCOX 0 Coefficient for length dependence of COX
PLCS 0 Coefficient for length dependence of CS
PLCT 0 Coefficient for length dependence of CT
PLDPHIB 0 Coefficient for length dependence of DPHIB
PLIGINV 0 Coefficient for length dependence of IGINV
PLIGOV 0 Coefficient for length dependence of IGOV
PLMUE 0 Coefficient for length dependence of MUE
PLNEFF 0 Coefficient for length dependence of NEFF
PLNFA 0 Coefficient for length dependence of NFA
PLNFB 0 Coefficient for length dependence of NFB
PLNFC 0 Coefficient for length dependence of NFC
PLNOV 0 Coefficient for length dependence of NOV
PLNP 0 Coefficient for length dependence of NP
PLRS 0 Coefficient for length dependence of RS
PLSTBET 0 Coefficient for length dependence of STBET
PLSTSAT 0 Coefficient for length dependence of STSAT
PLSTVFB 0 Coefficient for length dependence of STVFB
PLSAT 0 Coefficient for length dependence of SAT
PLSATB 0 Coefficient for length dependence of SATB
PLSATG 0 Coefficient for length dependence of SATG
PLVFB 0 Coefficient for length dependence of VFB
PLWA1 0 Coefficient for L*W dependence of A1
PLWAGIDL 0 Coefficient for L*W dependence of AGIDL
PLWALP 0 Coefficient for L*W dependence of ALP
PLWALP1 0 Coefficient for L*W dependence of ALP1
PLWALP2 0 Coefficient for L*W dependence of ALP2
PLWAX 0 Coefficient for L*W dependence of AX
PLWBETN 0 Coefficient for L*W dependence of BETN
719

PLWCF 0 Coefficient for L*W dependence of CF
PLWCFR 0 Coefficient for L*W dependence of CFR
PLWCGBOV 0 Coefficient for L*W dependence of CGBOV
PLWCGOV 0 Coefficient for L*W dependence of CGOV
PLWCOX 0 Coefficient for L*W dependence of COX
PLWCS 0 Coefficient for L*W dependence of CS
PLWCT 0 Coefficient for L*W dependence of CT
PLWDPHIB 0 Coefficient for L*W dependence of DPHIB
PLWIGINV 0 Coefficient for L*W dependence of IGINV
PLWIGOV 0 Coefficient for L*W dependence of IGOV
PLWMUE 0 Coefficient for L*W dependence of MUE
PLWNEFF 0 Coefficient for L*W dependence of NEFF
PLWNFA 0 Coefficient for L*W dependence of NFA
PLWNFB 0 Coefficient for L*W dependence of NFB
PLWNFC 0 Coefficient for L*W dependence of NFC
PLWNOV 0 Coefficient for L*W dependence of NOV
PLWNP 0 Coefficient for L*W dependence of NP
PLWRS 0 Coefficient for L*W dependence of RS
PLWSTBET 0 Coefficient for L*W dependence of STBET
PLWSTSAT 0 Coefficient for L*W dependence of STTHESAT
PLWSTVFB 0 Coefficient for L*W dependence of STVFB
PLWTHESAT 0 Coefficient for L*W dependence of THESAT
PLWTHESATB 0 Coefficient for L*W dependence of THESATB
PLWTHESATG 0 Coefficient for L*W dependence of THESATG
PLWVFB 0 Coefficient for L*W dependence of VFB
PLWXCOR 0 Coefficient for L*W dependence of XCOR
PLXCOR 0 Coefficient for length dependence of XCOR
POA1 1 Geometry independent part of A1
POAGIDL 0 Geometry independent part of AGIDL
POALP 10m Geometry independent part of ALP
POALP1 0 Geometry independent part of ALP1
POALP2 0 Geometry independent part of ALP2
POAX 3 Geometry independent part of AX
POBETN 70m Geometry independent part of BETN
POBGIDL 41 Geometry independent part of BGIDL
POCF 0 Geometry independent part of CF

POCFB 0 Geometry independent part of CFB
POCFR 0 Geometry independent part of CFR
POCGBOV 0 Geometry independent part of CGBOV
POCGIDL 0 Geometry independent part of CGIDL
POCGOV 1f Geometry independent part of CGOV
POCHIB 3.1 Geometry independent part of CHIB
POCOX 10f Geometry independent part of COX
POCS 0 Geometry independent part of CS
POCT 0 Geometry independent part of CT
PODNSUB 0 Geometry independent part of DNSUB
PODPHIB 0 Geometry independent part of DPHIB
POFETA 1 Geometry independent part of FETA
POFNT 1 Geometry independent part of FNT
POGC2 375m Geometry independent part of GC2
POGC3 63m Geometry independent part of GC3
POGCO 0 Geometry independent part of GCO
POIGINV 0 Geometry independent part of IGINV
POIGOV 0 Geometry independent part of IGOV
POMUE 500m Geometry independent part of MUE
PONEFF 5E23 Geometry independent part of NEFF
PONFA 8E22 Geometry independent part of NFA
PONFB 30MEG Geometry independent part of NFB
PONFC 0 Geometry independent part of NFC
PONOV 5E25 Geometry independent part of NOV
PONP 1E26 Geometry independent part of NP
PONSLP 50m Geometry independent part of NSLP
PORS 30 Geometry independent part of RS
PORSB 0 Geometry independent part of RSB
PORSG 0 Geometry independent part of RSG
POSTA2 0 Geometry independent part of STA2
POSTBET 1 Geometry independent part of STBET
POSTBGIDL 0 Geometry independent part of STBGIDL
POSTCS 0 Geometry independent part of STCS
POSTIG 2 Geometry independent part of STIG
POSTMUE 0 Geometry independent part of STMUE
POSTRS 1 Geometry independent part of STRS
POSTTHEMU 1.5 Geometry independent part of STTHEMU
POSTTHESAT 1 Geometry independent part of STTHESAT
POSTVFB 500u Geometry independent part of STVFB
721

POSTXCOR 0 Geometry independent part of STXCOR
POTHEMU 1.5 Geometry independent part of THEMU
POTHESAT 1 Geometry independent part of THESAT
POTHESATB 0 Geometry independent part of THESATB
POTHESATG 0 Geometry independent part of THESATG
POTOX 2n Geometry independent part of TOX
POTOXOV 2n Geometry independent part of TOXOV
POVFB -1 Geometry independent part of VFB
POVNSUB 0 Geometry independent part of VNSUB
POVP 50m Geometry independent part of VP
POXCOR 0 Geometry independent part of XCOR
PWA1 0 Coefficient for width dependence of A1
PWAGIDL 0 Coefficient for width dependence of AGIDL
PWALP 0 Coefficient for width dependence of ALP
PWALP1 0 Coefficient for width dependence of ALP1
PWALP2 0 Coefficient for width dependence of ALP2
PWAX 0 Coefficient for width dependence of AX
PWBETN 0 Coefficient for width dependence of BETN
PWCF 0 Coefficient for width dependence of CF
PWCFR 0 Coefficient for width dependence of CFR
PWCGBOV 0 Coefficient for width dependence of CGBOV
PWCGOV 0 Coefficient for width dependence of CGOV
PWCOX 0 Coefficient for width dependence of COX
PWCS 0 Coefficient for width dependence of CS
PWCT 0 Coefficient for width dependence of CT
PWDPHIB 0 Coefficient for width dependence of DPHIB
PWIGINV 0 Coefficient for width dependence of IGINV
PWIGOV 0 Coefficient for width dependence of IGOV
PWMUE 0 Coefficient for width dependence of MUE
PWNEFF 0 Coefficient for width dependence of NEFF
PWNFA 0 Coefficient for width dependence of NFA
PWNFB 0 Coefficient for width dependence of NFB
PWNFC 0 Coefficient for width dependence of NFC
PWNOV 0 Coefficient for width dependence of NOV
PWNP 0 Coefficient for width dependence of NP

PWRS 0 Coefficient for width dependence of RS
PWSTBET 0 Coefficient for width dependence of STBET
PWSTTHESAT 0 Coefficient for width dependence of STTHESAT
PWSTVFB 0 Coefficient for width dependence of STVFB
PWTHESAT 0 Coefficient for width dependence of THESAT
PWTHESATB 0 Coefficient for width dependence of THESATB
PWTHESATG 0 Coefficient for width dependence of THESATG
PWVFB 0 Coefficient for width dependence of VFB
PWXCOR 0 Coefficient for width dependence of XCOR
SA 0 Distance between OD-edge to poly from one side
SAREF 1u Reference distance between OD-edge to poly
from one side
SB 0 Distance between OD-edge to poly from other side
SBREF 1u Reference distance between OD-edge to poly
from the other side
STETAO 0 Eta0 shift factor related to VTHO change
TKUO 0 Temperature dependence of KUO
bottom part
gate-edge part
STI-edge part
723

W 10u Design width
WKUO 0 Width dependence of KUO
WKVTHO 0 Width dependence of KVTHO
WLOD 0 Width parameter
WLODKUO 0 Width parameter for UO stress effect
WLODVTH 0 Width parameter for VTH stress effect
WOT 0 Effective reduction of channel width per side due to
lateral diffusion of channel-stop dopant ions
WVARO 0 Geometry independent difference between actual
and programmed field-oxide opening
WVARW 0 Width dependence of difference between actual and

N_Port
Schematic format
PART attribute
<name>
Example
SP1
FILE attribute
<file name>
The FILE attribute specifies the path and name of the N-Port parameter file.
Example
E:\MC12\data\Gg10v20m.s2p
The N_PORT device is a general device with N ports characterized by a set of

S, Y, Z, G, H, T, or ABCD parameters contained in standard Touchstone data
files.
Typically these files are provided by RF suppliers in a text file as a table of val-
ues. Here, for example, is a set of typical 2-port S parameters:
! SIEMENS Small Signal Semiconductors

! BFG194
! Si PNP RF Bipolar Junction Transistor in SOT223
! VCE = -10 V IC = -20 mA
! Common Emitter S-Parameters: August 1996
# GHz S MA R 50
! f S12 S21 S12 S22
! GHz MAG ANG MAG ANG MAG ANG MAG ANG
0.010 0.3302 -25.4 35.370 169.9 0.0053 85.3 0.9077 -10.0
0.020 0.3471 -48.2 33.679 161.6 0.0108 77.5 0.8815 -19.8
0.050 0.4525 -95.0 27.726 139.2 0.0226 61.4 0.7258 -43.7
0.100 0.5462 -131.5 19.023 128.7 0.0332 52.2 0.5077 -68.7
0.150 0.5723 -149.4 13.754 106.4 0.0394 49.1 0.3795 -84.8
0.200 0.5925 -159.8 10.787 99.1 0.0443 50.1 0.3068 -95.0
0.250 0.6023 -167.0 8.757 93.4 0.0497 51.2 0.2581 -104.8
0.300 0.6089 -172.2 7.393 89.0 0.0552 52.4 0.2298 -122.2
...
725
Micro-Cap converts the incoming parameters to Y-parameters and then imple-
ments the N_PORT device as a set of Laplace table sources. Here, for example, is
the equivalent circuit for a 4-port.
Figure 26-26 Four port equivalent circuit

The equations are:
I1 = Y12*V(1P,1M) + Y12*V(2P,2M) + Y13*V(3P,3M) + Y14*V(4P,4M)

For an example of how the N-Port device is used see the sample circuit file
NPORT4.CIR.

OPAMP
Schematic format
PART attribute
<name>
Example
OP1
MODEL attribute
<model name>
Example
LF351
There are three model levels for the OPAMP. Each succeeding level provides
increasingly more realistic models at the expense of increasingly more complex
equivalent circuits.
Level 1 is a simple voltage-controlled current source with a finite output resis-

tance and open loop gain.
Level 2 is a three stage, two pole model with slew rate limiting, finite gain, and
output resistance.
Level 3 is an enhanced Boyle model, similar to those implemented in other

SPICE programs as subcircuits. It is not, however, a macro or a subcircuit, but
a fully internal device model. It models positive and negative slew rates, finite
gain, AC and DC output resistance, input offset voltage and current, phase mar-
gin, common mode rejection, unity gain bandwidth, three types of differential
input, and realistic output voltage and current limiting.

.MODEL <model name> OPA ([model parameters])
Examples
.MODEL LM709 OPA (A=45K VOFF=.001 SRP=250K GBW=1E6)
.MODEL LF155 OPA (LEVEL=2 TYPE=1 A=50K SRP=330K)
727
Model parameters
Name Parameter Units Def. Level

LEVEL Model level(1, 2, 3) 1 1,2,3
TYPE 1=NPN 2=PNP 3=JFET 1 3
C Compensation capacitance F 30E-12 3
A DC open-loop gain 2E5 1,2,3
ROUTAC AC output resistance Ω 75 1,2,3
ROUTDC DC output resistance Ω 125 1,2,3
VOFF Input offset voltage V 0.001 3
IOFF Input offset current A 1E-9 3
SRP Maximum positive slew rate V/S 5E5 2,3
SRN Maximum negative slew rate V/S 5E5 2,3
IBIAS Input bias current A 1E-7 3
VCC Positive power supply V 15 3
VEE Negative power supply V -15 3
VPS Maximum pos. voltage swing V 13 3
VNS Maximum neg. voltage swing V -13 3
CMRR Common-mode rejection ratio 1E5 3
GBW Gain bandwidth 1E6 2,3
PM Phase margin deg. 60 2,3
PD Power dissipation W .025 3
IOSC Short circuit current A .02 3
T_REL_GLOBAL Relative to current temp. °C
VCC and VEE are the nominal power supply values at which VPS and VNS are
specified. It is possible to operate the OPAMP at other supply voltages. VEE and
VCC affect only power dissipation and the output saturation characteristics.

Model schematic and equations:
Figure 26-27 The level 1 opamp model
Figure 26-28 The level 2 opamp model
729
Figure 26-29 The level 3 opamp model with NPN inputs
Figure 26-30 The level 3 opamp model with PNP inputs

Figure 26-31 The level 3 opamp model with JFET inputs
Definitions
T Junction temperature in degrees Kelvin
VT k•T/q
BETA1 Forward beta of Q1
BETA2 Forward beta of Q2
BJT1IS Saturation current (IS) of Q1
BJT2IS Saturation current (IS) of Q2
V(A1) Voltage at node A1
V(A2) Voltage at node A2
V(CM) Voltage at node CM
I(VS1) Current through source VS1
I(VS2) Current through source VS2
I(VC) Current through source VC
I(VE) Current through source VE
I(VLP) Current through source VLP
I(VLN) Current through source VLN
V(VCC) Voltage across source VCC
V(VEE) Voltage across source VEE
I(GA) Current of source GA
I(GCM) Current of source GCM
I(F1) Current of source F1
V(E1) Voltage of source E1
V(H1) Voltage of source H1
731
Temperature effects
Temperature affects diodes, BJTs, and JFETs in the usual way as described in the
model sections for these devices. Default temperature parameters are used.
Level 1 equations
R = ROUTAC+ROUTDC
GM = A/R
Level 2 equations
R = ROUTAC+ROUTDC
GM = A1/3/ R
F1 = GBW / A (First pole)
F2 = GBW / tan(90 - PM) (Second pole)
C1 = 1/(2•π•F1•R)
C2 = 1/(2•π•F2•R)
Level 3 equations
C3 = C
C1 = 0.5*C* tan(90 - PM)
RC1 = 1/(2•π•GBW•C3)
RC2 = RC1
R2 = 1E5
GA = 1/RC1
NPN and PNP input stages

VAF = 200
NPN input
IC1 = SRP•C3/2
CE = 2•IC1/SRN-C3
PNP input
IC1 = SRN•C3/2
CE = 2•IC1/SRP-C3
BETA1 = IC1/(IBIAS+IOFF/2)
BETA2 = IC1/(IBIAS-IOFF/2)
IEE = (((BETA1+1)/BETA1)+((BETA2+1)/BETA2))•IC1
RE1 = ((BETA1+BETA2)/(BETA1+BETA2+2))•(RC1-VT/IC1)
RE2 = RE1
RP = (|VCC|+|VEE|)2/(PD-|VCC|•2•IC1-|VEE|•IEE)
RE = VAF/IEE
BJT1IS = 1E-16
BJT2IS = BJT1IS•(1+VOFF/VT)

JFET input stage
IEE = C3•SRN
CE = IEE•SRP-C3
RE = VAF/IEE
RE1 = 1
RE2 = 1
BETA1 = 0.5•GA2/IEE
BETA2 = BETA1
RP = (|VCC|+|VEE|)2/PD
ALL STAGES
RO2 = ROUTDC - ROUTAC
GCM = 1/(CMRR•RC1)
GB = RC1•A/RO2
VLP = IOSC•1000
VLN = VLP
VC = VCC - VPS
VE = -VEE + VNS
Controlled source equations

I(GA) = GA•(V(A1)-V(A2))
I(GCM)= GCM•V(CM)
I(F1) = GB•I(VS1)-GB•I(VC)+GB•I(VE)+GB•I(VLP)-GB•I(VLN)
V(E1) = (V(VCC)+V(VEE))/2
V(H1) = 1000•(I(VS2))
V(VS1) = 0.0 (Only used to measure current)
V(VS2) = 0.0 (Only used to measure current)
Level 2 and 3 models use Gain Bandwidth (GBW) as an input parameter. The
models produce an OPAMP which, in open-loop configuration, produces the
specified phase margin (PM) and a gain of -3.01 dB at F = GBW. The intersection
of the gain curve asymptote (a straight line tangent to the mid-band gain curve)
and the line F = GBW occurs at 0.0 dB.
Note PM = Phase Margin = Phase Angle + 180. To plot the phase margin of
V(OUT), the Y expression would be PH(V(OUT)) + 180.
733
Pulse source
Schematic format
PART attribute
<name>
Example
P1
MODEL attribute
<model name>
Example
RAMP
The PULSE source is similar to the SPICE PULSE independent voltage source,
except that it uses a model statement and its timing values are defined with re-
spect to T=0.

.MODEL <model name> PUL ([model parameters])
Example
.MODEL STEP PUL (VZERO=.5 VONE=4.5 P1=10n P2=20n P3=100n
+ P4=120n P5=500n)
Model parameters
VZERO Zero level V 0.0
VONE One level V 5.0
P1 Time delay to leading edge S 1.0E-7
P2 Time delay to one-level S 1.1E-7
P3 Time delay to trailing edge S 5.0E-7
P4 Time delay to zero level S 5.1E-7
P5 Repetition period S 1.0E-6

Equations
The waveform value is generated as follows:
From To Value
0 P1 VZERO
P1 P2 VZERO+((VONE-VZERO)/(P2-P1))•(T-P1)
P2 P3 VONE
P3 P4 VONE+((VZERO-VONE)/(P4-P3))•(T-P3)
P4 P5 VZERO
where From and To are T values, and T=TIME mod P5. The waveform
repeats every P5 seconds. Note that P5 ≥ P4 ≥ P3 ≥ P2 ≥ P1.
Figure 26-32 Sample waveform for model parameters

vzero=1 vone=4 P1=.1u P2=.2u P3=.4u P4=.5u P5=1u
735
Resistor
SPICE format
Syntax
R<name> <plus> <minus> [model_name]
+<resistance> [TC=<tc1>[,<tc2>]]
Example
R1 2 3 50

<plus> and <minus> are the positive and negative node numbers. The
polarity references are used only for plotting or printing the voltage across,
V(RX), and the current through, I(RX), the resistor.
Schematic format
PART attribute
<name>
Examples
R5
CARBON5
RESISTANCE attribute
<resistance>[LOT[/#][/GAUSS|/UNIFORM|WCASE]=<tolvalue>[%]]
[TC1=<tc1> [,<tc2>]]
Examples
50
50K*(1+V(6)/100)
FREQ attribute
<fexpr>
Examples
2K+10*(1+F/1e9)
MODEL attribute
[model_name]
Example
RMOD
RESISTANCE attribute
<resistance> may be a constant or an expression involving time-domain
variables. The expression is evaluated in the time domain only. Consider the
expression:
100+V(10)*2
V(10) refers to the value of the voltage on node 10 during a transient

analysis, a DC operating point calculation prior to an AC analysis, or during
a DC analysis. It does not mean the AC small signal voltage on node 10. If
the DC operating point value for node 10 was 2, the resistance would be
evaluated as 100 + 2*2 = 104. The constant value, 104, is used in AC
analysis.
FREQ attribute
If <fexpr> is used, it replaces the value determined during the operating
point. <fexpr> may be a simple number or an expression involving frequency
domain variables. The expression is evaluated during AC analysis as the
frequency changes. For example, suppose the <fexpr> attribute is this:
V(4,5)*(1+F/1e7)

In this expression, F refers to the AC analysis frequency variable and V(4,5)
refers to the AC small signal voltage between nodes 4 and 5. Note that there
<resistance> will be used in transient analysis.
Stepping effects
The RESISTANCE attribute and all of the model parameters may be stepped. If
it is stepped, it replaces <resistance>, even if <resistance> is an expression. The
stepped value may be further modified by the temperature effect.
Temperature effects
There are two different temperature factors, a quadratic factor and an exponential
factor. The quadratic factor is characterized by the model parameters TC1 and
TC2, or <tc1> and <tc2> from the parameter line. The exponential factor is char-
acterized by the model parameter TCE.
If [TC=<tc1>[,<tc2>]] is specified on the parameter line, <resistance> is multi-

plied by a temperature factor, TF.
TF = 1+<tc1>•(T-Tnom)+<tc2>•(T-Tnom)2
737
If [model_name] is used and TCE is not specified, <resistance> is multiplied by a
temperature factor, TF.
example, a spec of 3000 ppm/degree C would produce a TC1 value of 3E-3.
If [model_name] is used and TCE is specified, <value> is multiplied by a tem-

perature factor, TF.
TF = 1.01TCE•(T-Tnom)
If both [model_name] and [TC=<tc1>[,<tc2>]] are specified, [TC=<tc1>[,<tc2>]]

takes precedence.
nominal resistance was measured. T is set to the analysis temperature from the
value, which can be overridden with a .OPTIONS statement. T and Tnom may be
changed for each model by specifying values for T_MEASURED, T_ABS, T_REL_
GLOBAL, and T_REL_LOCAL. See the .MODEL command for more on how de-
vice operating temperatures and Tnom temperatures are calculated.
Monte Carlo effects

LOT and DEV Monte Carlo tolerances, available only when [model_name] is
except the T_parameters. Both forms are converted to an equivalent tolerance
factor, MF, which ultimately multiplies the value of the model parameter R.
The final resistance, rvalue, is calculated as follows:
rvalue = <value> * TF * MF * R, where R is the model parameter multiplier.

.MODEL <model_name> RES ([model parameters])
Example
.MODEL RM RES (R=2.0 LOT=10% LS=2.3n TC1=.015)
Model parameters
R Resistance multiplier 1.0
LS Series inductance H 0.0
CP Parallel capacitance F 0.0
TC1 Linear temperature coefficient °C-1 0.0
TC2 Quadratic temperature coefficient °C-2 0.0
TCE Exponential temperature coefficient %/°C 0.0
NM Noise multiplier 1.0
Figure 26-33 Resistor model

Noise effects
Resistor noise is modeled with a thermal noise current:
I = NM * sqrt(4*K*T/rvalue )
NM multiplies the resistor noise current. A value of zero eliminates the noise
contribution from all resistors that use the model.
739
S (Voltage-controlled switch)
SPICE format
S<name> <plus output node> <minus output node>
+<plus controlling node> <minus controlling node>
+<model name>
Example
S1 10 20 30 40 RELAMOD
Schematic format
PART attribute
<name>
Example
S1
MODEL attribute
<model name>
Example
RELAY
The switch can operate in two distinct operational modes, Smooth Transition and
Hysteresis.
Smooth Transition Mode:

Use this mode if you do not need input hysteresis. The smooth transition mini-
mizes convergence problems. In this mode VON and VOFF are specified. VT and
VH are ignored. The switch impedance changes smoothly from RON to ROFF
as the control voltage moves from VON to VOFF and from ROFF to RON as the
control voltage moves from VOFF to VON.
Hysteresis Mode:
Use this mode if you need input hysteresis and the circuit is not sensitive to con-
vergence problems. In the Hysteresis mode, VT and VH are specified. VON and
VOFF are ignored. The switch impedance changes abruptly from ROFF to RON
as the control voltage moves higher than VT+VH and from RON to ROFF as the
control voltage moves lower than VT-VH.

This switch is controlled by the voltage across the two input nodes and the switch
impedance is impressed across the output nodes.
RON and ROFF must be greater than zero and less than 1/Gmin.
Do not make the transition region, VON-VOFF, too small as this will cause an
excessive number of time points required to cross the region. The smallest al-
lowed values for VON-VOFF is (RELTOL•(max(VON,VOFF))+VNTOL).

.MODEL <model name> VSWITCH ([model parameters])
Example
.MODEL S1 VSWITCH (RON=1 ROFF=1K VON=1 VOFF=1.5)
.MODEL S2 VSWITCH (RON=1 ROFF=1K VT=3 VH=1)
Model parameters
RON On resistance Ohms 1
ROFF Off resistance Ohms 1E6
VON Control voltage for On state V 1
VOFF Control voltage for Off state V 0
VT Threshold voltage V None
VH Hysteresis voltage V None
Model Equations
VC = Voltage across the control nodes
VM= Mean of control voltages = (VON+VOFF)/2
VD = Difference of control voltages = VON-VOFF
k = Boltzmann's constant
T = Analysis temperature
741
Smooth Transition Mode: (Used if VON and VOFF are defined)
If VON > VOFF
If VC >= VON
RS = RON
If VC <= VOFF
RS = ROFF
If VOFF < VC < VON
RS = exp(LM + 3•LR•(VC-VM)/(2•VD) - 2•LR•(VC-VM)3/VD3)
If VON < VOFF

If VC <= VON
RS = RON
If VC >= VOFF
RS = ROFF
If VOFF > VC > VON
RS = exp(LM - 3•LR•(VC-VM)/(2•VD) + 2•LR•(VC-VM)3/VD3)
Hysteresis Mode: (Used if VON and VOFF are not defined)

If VC >= VT + VH
RS = RON
If VC <= VT - VH
RS = ROFF
Else
RS remains unchanged
Noise effects
Noise is modeled as a resistor equal to the resistance found during the DC operat-
ing point. The thermal noise current is calculated as follows:
I = sqrt(4•k•T/RS)

Sample and hold source
SPICE format
There is no equivalent Sample and Hold device in SPICE or PSpice.
Schematic format
PART attribute
<name>
Examples
S1
S10
SA
INPUT EXPR attribute

<input expression>
Examples
V(1,2)
V(10,20)*I(R1)
V(INPUT)
SAMPLE EXPR attribute

<sampling expression>
Examples
V(1,2)>1.2
V(5)>1.1 AND V(4) >1.2
I(RL)>1e-3
PERIOD attribute
<sampling period>
Examples
100ns
tmax/100
1U

PRECISION attribute
<precision>
743
Examples
1e-3
1U

This device is an ideal sample and hold. It samples <input expression> when
<sampling expression> is true or every sample period seconds. The behavioral
modes are distinguished as follows:
If <sampling expression> is specified:

<input expression> is sampled whenever <sampling expression> is >= 1.0. <sam-
pling expression> is normally a Boolean expression. The output voltage of the
source is set to the sampled value and remains constant until the next sample.
If <sampling expression> is unspecified and <sampling period> is specified:

The source samples and stores the numeric value of <input expression> once ev-
ery <sampling period> seconds, starting at 0. The output voltage of the source is
set to the sampled value and remains constant until the next sample.
Neither <sampling expression> nor <sampling period> is specified:

This will generate an error message. Either <sampling expression> or <sampling
period> must be specified.
Both <sampling expression> and <sampling period> are specified:

<input expression> is sampled whenever <sampling expression> is >= 1.0.
This is the same as the first case. The <sampling period> is ignored.
<precision> determines the degree of tracking. The smaller the number the closer
the device tracks the sampled expression. Small values of <precision> produce
smaller time steps and more accurate tracking.
Note that while <input expression> is usually a node voltage, the expression
can involve circuit variables, as in the second example. <sampling period> may
include a non-time varying expression, as in the second example, where the
<tmax> of the analysis run is used to calculate the sampling period.
See the sample file SH2.CIR for an example of how to use the sample and hold

Sine source
Schematic format
PART attribute
<name>
Examples
S1
MODEL attribute
<model name>
Example
Line60
The Sine source is similar to the SPICE SIN independent voltage source. Unlike
the SPICE source, it uses a model statement.

.MODEL <model name> SIN ([model parameters])
Example
.MODEL V1 SIN (F=1Meg A=0.6 DC=1.5)
Model parameters
F Frequency Hz 1E6
A Amplitude V 1.0
DC DC level V 0.0
PH Phase shift Radians 0.0
RS Source resistance Ω 0.001
RP Repetition period of exponential S 0.00
TAU Exponential time constant S 0.00
Model Equations
If TAU = 0 then
V = A•sin (2•π•F•TIME + PH) + DC
Else
V = A•e(-T/TAU)•sin (2•π•F•TIME + PH) + DC
where T = TIME mod RP.

745
Figure 26-34 Sample waveform for model parameters
F=2Meg A=1 DC=1 RP=2U TAU=.4U

Subcircuit call
SPICE format
X<name> [node]* <subcircuit name>
+ [PARAMS: <<parameter name>=<parameter value>>*]
+ [TEXT: <<text name>=<text value>>*]
Examples
X1 10 20 AMP
XDIFF 100 200 DIFF PARAMS: GAIN=10
Schematic format
PART attribute
<name>
Example
X1
NAME attribute
<subcircuit name>

Example
FILTER
FILE attribute
[<file name>]

Example
MYFILE.MOD
PARAMS attribute
[<<parameter name>=<parameter value>>*]

Example
CENTER=10kHz BW=1kHz
OPTIONAL attribute
[<<pin_name>=<node_name>>*]

Example
DPWR=$G_DPWR_VCC
747
TEXT: attribute
[<<text name>=<text value>>*]

Example
JEDEC="FILENAME"
[node]* are the numbers or names of the nodes specified in the .SUBCKT state-
ment. The number of nodes in the subcircuit call must be the same as the number
of nodes in the .SUBCKT statement. When the subcircuit is called, the nodes in
the call are substituted for the nodes in the body of the subcircuit in the same or-
der as they are listed in the .SUBCKT statement.
Any nodes defined in the .SUBCKT statement with the OPTIONAL keyword
may optionally follow the [node]* values. A subset of the optional nodes may be
listed, and will be assigned to the internal subckt nodes in the order specified by
the OPTIONAL statement. If you skip nodes, they must be skipped from the end
of the list. This statement is used in the Digital Library to specify optional power
nodes. Optional nodes can also be specified through the OPTIONAL attribute.
The SPICE <subcircuit name> or schematic NAME attribute defines the subcir-
cuit name. It must be the same as the name in the defining .SUBCKT statement.
The schematic FILE attribute defines the name of the file in which the .SUBCKT
statement can be found. Micro-Cap looks for the .SUBCKT statement in the fol-
lowing places in the order indicated.
• If the circuit is a schematic:

• In the Text area.
• In the file named in the FILE attribute.
• In one or more files named in a '.LIB' statement.
• In one or more files named in the default '.LIB NOM.LIB' statement.
• If the circuit is a SPICE text file:

• In the circuit description text.
• In one or more files named in a '.LIB' statement.
• In one or more files named in the default '.LIB NOM.LIB' statement.
Subcircuits are used extensively in the Component library. They are accessed via
the default .LIB statement. The SPICE keyword or schematic PARAMS attribute
let you pass multiple numeric parameters to the subcircuit. <parameter name> is
its name and <parameter value> defines the value it will assume if the parameter
is not included in the subcircuit call. For example:
.SUBCKT CLIP 1 2
+ PARAMS: LOW=0 HIGH=10
Any of these calls are legal:
X1 10 20 CLIP ;RESULTS IN LOW=0, HIGH=10

X2 10 20 CLIP PARAMS: LOW=1 HIGH=2 ;RESULTS IN LOW=1, HIGH=2
X3 10 20 CLIP PARAMS: HIGH=4 ;RESULTS IN LOW=0, HIGH=4
The SPICE keyword or schematic TEXT: attribute lets you pass text parameters
to the subcircuit. <text name> is the name of the text parameter and <text value>
defines the value it will assume if the parameter is not included in the subcircuit
call. For example:
.SUBCKT STIMULUS 1 2 3 4
+ TEXT: FILE="T1.STM"
Either of these calls are legal:

X1 10 20 30 40 STIMULUS ;RESULTS IN FILE="T1.STM"
X2 10 20 30 40 STIMULUS TEXT:FILE="P.STM" ;RESULTS IN FILE="P.STM"
Using the subckt component in a Micro-Cap schematic circuit file is easy. First,
you must enter the subcircuit into the Component library using the Component
editor. This requires entering:
• Subckt Name: Use any unique name. To avoid confusion, it should be the
same as the name used in the SUBCKT control statement, although this is not
strictly necessary.
• Shape Name: Use any suitable shape.

• Definition: Use SUBCKT.
Once these items have been entered, the pin assignments must be defined. These
define where the node numbers called out in the SUBCKT control statement go
on the shape. Pins are assigned by clicking in the Shape drawing area and naming
the pin with the subckt node number. The pin is then dragged to the desired posi-
tion on the shape. The subckt is then placed in the schematic in the usual way.
To see how subckts are called, see the sample circuit files SUBCKT1 and PLA2.
749
Switch
Schematic format
PART attribute
<name>
Example
S1
VALUE attribute
<[V | T | I]>,<n1,n2> [,<ron>[,<roff>]]
Examples
V,1,2
I,2ma,3ma
T,1ms,2ms,50,5Meg
This is the oldest of three types of switches. The newer S and W switches exhibit
a slower, but smoother transition between the off and on states.
There are three types of dependent switches; current-controlled, voltage-con-

trolled, and time-controlled.
The switch is a four-terminal device. A current sensing inductor must be con-

nected across the two input nodes when the current-controlled switch option is
used. Voltage-controlled switches are controlled by the voltage across the two
input nodes. Time dependent switches use the transient analysis time variable to
control the opening and closing of the switch.
The two controlling nodes for a time-controlled switch are not used and may be
shorted to ground or to the two output nodes to reduce the node count by two.
In transient analysis, care must be taken in choosing the time step of the simula-
tion. If the time step is too large, the switch might never turn on. There must be at
least one time point within the specified window for the switch to close or open.
The switch parameter syntax provides for a normally-on or a normally-off switch.

A normally-on switch is one that is on outside the specified window and off in-
side the window. A normally-off switch is one that is off outside the window and
on inside the window.

The optional <ron> impedance defaults to 1E-3 ohm. The optional <roff> de-
faults to 1E9 ohms. When specifying <roff>, be careful that off switch resistances
are not so low that they load your circuit, nor so high that they generate excessive
voltage by blocking the current from an inductor or a current source.
Rules of operation
If n1 < n2 then
Switch is closed (ON) when
n1 ≤ X ≤ n2
Switch is open (OFF) when
X < n1 or X > n2
If n1 > n2 then
Switch is open (OFF) when
n1 ≥ X ≥ n2
Switch is closed (ON) when
X > n1 or X < n2
If the first character in the switch parameter is V:

The switch is a voltage-controlled switch.
X = voltage across the input nodes.
n1 and n2 are voltage values.
If the first character in the switch parameter is I:

The switch is a current-controlled switch.
X = current through the inductor placed across the input nodes.
n1 and n2 are current values.
If the first character in the switch parameter is T:

The switch is a time-controlled switch.
X = TIME.
n1 and n2 are values of the TIME variable.
If the switch is closed, the switch has a resistance of <ron>. If the switch is open,
the switch has a resistance of <roff>.
751
Timer
SPICE format
There is no equivalent timer device in SPICE or PSpice.
Schematic format
PART attribute
<name>
Examples
T1
S2
INPUTEXPR attribute
<input expression>
Examples
V(1,2)>=1.3
I(R1)>=1ma AND I(R1)<=5ma
T>120ns AND V(3)>5
ELAPSED_SCALE attribute
<elapsed scale>
Examples
1E6
1
INCREMENT attribute
<increment>
Examples
-1
1
INITIAL attribute
<initial value>
Examples
0
16384

MIN attribute
<min>
Example2
0
-100
MAX attribute
<max>
Examples
16384
10
This device provides the following outputs:
• Count: The number of events since the beginning of the simulation run or
the end of the last reset pulse.
• Elapsed Time: The elapsed time since an event occurred.

• Last Time: The last time an event occurred.
An event occurs whenever <input expression> evaluates to >= 1.0.
<input expression> is typically a boolean expression defining the event condition

such as V(IN)>=3.4.
Count is initially set to <initial value> when the simulation starts and whenever
the voltage on the RESET pin exceeds 1.0.
Each time an event occurs, count is increased by <increment>. Count can be no

smaller than <min> and no larger than <max>.
Pins: There are three pins, ELAPSED, COUNT, LAST. The pins are all outputs
and are connected to voltage sources whose value reflects the elapsed time, count
value, and last event time.
The value on the ELAPSED pin is:

(Elapsed simulation time since last event occurred) · <elapsed scale>
753
The value on the COUNT pin is:
<initial value> + (Number of events since T=0 or last reset) · <increment>
The value on the LAST pin is the last simulation time (in secs) at which an event
occurred.
Note the Reset pin resets only the COUNT value. It has no effect on the
ELAPSED or LAST values.
See the sample file TIMER.CIR for an example of how to use the timer.

Transformer
Schematic format
PART attribute
<name>
Example
T1
VALUE attribute
<primary inductance>,<secondary inductance>,<coupling coefficient>
Example
.01,.0001,.98
The transformer component consists of two inductors with a mutual inductance

between them. It is entirely equivalent to two discrete inductors and a linear K
device defining a mutual inductance between them.
The coefficient of coupling is related to the mutual inductance by the following

equation.
k = M / (LP*LS)0.5

M is the mutual inductance.
k is the coefficient of coupling and -1 <= k <= 1.
LP is the primary inductance.
LS is the secondary inductance.
A resistive impedance of 1/GMIN is added between the positive input pin and the
positive output pin to avoid DC convergence problems.
A negative coupling coefficient is the equivalent of flipping the polarity of one

side of the transformer.
755
Transmission line
SPICE format
Ideal Line
T<name> <A port + node> <A port - node>
+<B port + node> <B port - node>
+[model name]
+ [Z0=<value> [TD=<value>] | [F=<value> [NL=<value>]]]
Example
T1 10 20 30 40 Z0=50 TD=3.5ns
T1 10 20 30 40 Z0=150 F=125Meg NL=0.5
T1 20 30 40 50 TLMODEL
Lossy Line
T<name> <A port + node> <A port - node>
+<B port + node> <B port - node>
+[<model name> [physical length]]
+[ LEN=<len value> R=<rvalue> L=<lvalue> G=<gvalue>
+ C=<cvalue>]
Examples
T1 20 30 40 50 LEN=1 R=.5 L=.8U C=56PF
T3 1 2 3 4 TMODEL 12.0
Schematic format
PART attribute
<name>
Example
T1
VALUE attribute for ideal line

Z0=<value> [TD=<value>] | [F=<value> [NL=<value>]]
Example for ideal line

Z0=50 TD=3.5ns
VALUE attribute for lossy line

<physical length> LEN=<len value> R=<rvalue> L=<lvalue>
G=<gvalue> C=<cvalue>
Examples for lossy line
LEN=1 R=.5 L=.8U C=56PF
R=.5 L=.8U C=56PF
MODEL attribute
<model name>
Example
RELAY

.MODEL <model name> TRN ([model parameters])
Examples
.MODEL TIDEAL TRN(Z0=50 TD=10ns)
.MODEL TLOSS TRN(C=23pF L=13nH R=.35 LEN=10)
Model parameters for ideal line

Z0 Characteristic impedance Ohms None
TD Transmission delay S None
F Frequency for NL Hz None
NL Relative wavelength 0.25
Model parameters for lossy line

R Resistance per unit length Ohms / unit None
L Inductance per unit length Henries / unit None
G Conductance per unit length Mhos / unit None
C Capacitance per unit length Farads / unit None
LEN Physical length Same as RLGC None
LEN, R, L, G, and C must use a common unit of length. For example, if C is

measured in units of farads/cm, then R must be in ohms/cm, L must be in henrys/
cm, G must be in mhos/cm, and LEN must be in centimeters.
R, L, C, and G are usually fixed values, but may also be expressions using the
frequency F, or complex frequency S (2*PI*F*j), in AC analysis only. In any
other analysis, expressions are evaluated with F = 0.
Model Equations
The ideal and lossy lines are represented by the model shown in Figure 26-35.
757
The principal difference between the two models is in the implementation of
the delay. In the ideal model, the delay is implemented as a linked-list of data
pairs (time,value) and breakpoints. The lossy line uses the SPICE3 convolution
Figure 26-35 Transmission line model
method to represent the line, producing an accurate representation of a lossy

line. The method employed is to convolve the input waveform with the impulse
response of the lossy line to produce the output waveform. The impulse response
is obtained from a presolved analytical formula. Convolution assures that any
waveform can be handled, not just steps and linear ramps.
Note that only the RLC, RC, RG, and LC lines are supported. Nonzero values for
R, L, C, and G specifying other line types will generate an error.
For both types of transmission lines, both VALUE and MODEL parameter de-
scriptions may be given. VALUE parameters replace MODEL parameters and the
final result is checked for adherence to the VALUE attribute syntax rules.
Note that either 'Z0' (Z -zero) or 'ZO' (Z-Oh) may be used in either the VALUE or
MODEL parameters.
If the length modifier, <physical length>, is given, it overrides the model param-
eter LEN, if <model name> is used.
For examples of lossless transmission line circuits see the TL1, TL2, and TL3 cir-
cuits. For lossy transmission lines see the LTRA3 and SkinEffect1 circuits. The
latter uses a frequency dependent expression for R to model skin effect.

User file source
Schematic format
PART attribute
<name>
Example
U1
FILE attribute
<file name>
Example
AMP.USR
EXPRESSION attribute
[expression]
Example
V(OUT) vs T
REPEAT attribute
[number]
Example
5
ENABLE_EXPR attribute
[enable_expression]
Examples
V(OUT)>3.2
Time>120ns
This is a voltage source whose curve comes from a text file. The curve is repeated
[number] times in transient analysis and once in AC and DC analysis.
The files contain a header and N sequential lines, each with a variable number of
data values:
759
Transient analysis:
Two data values per line: Time, Y X set to Time
Three data values per line: Time, X, Y
AC:
Three data values per line: Frequency, Real(Y), Imag(Y) X set to Frequency
Five data values per line: Frequency, Real(X), Imag(X), Real(Y), Imag(Y)
DC: Three data values per line:

DCINPUT1, X, Y
where X is the X expression value, Y is the Y expression value, and
DCINPUT1 is the value of variable1.
User files may be created externally, or by saving one or more curves or wave-
forms after an analysis run. To save a particular waveform, press F10 to invoke
the Plot Properties dialog box after the analysis is over and select the waveform
to be saved from the Save Curves section of the dialog box.
When you place a user source in a schematic, Micro-Cap reads the current data
directory to find any files with the extension *.USR. It then reads the files them-
selves to see what waveforms are available for use. It presents the files and wave-
forms as items in drop-down lists in the FILE and EXPRESSION fields so you
can easily select the file name and waveform expression.
Curves saved in user files can be displayed in an analysis by selecting them from
the Curves section of the Variables list. To invoke this list, click the right mouse
button in the Y Expression field. The X and Y parts of a waveform are stored as
CurveX and CurveY, to let you select the X and Y parts independently.
If [enable_expression] is given, it gates the source. If it is not given, the source

is always enabled. When [enable_expression] is true the waveform starts. If [en-
able_expression] becomes false, the source stays at its current output level.
See USER for an example of a simple source, USER2 for an example of two
sources, and USER3 for an example of a triggered source. Click on Info , then
on the source to examine the data file used by the source.

W (Current-controlled switch)
SPICE format
W<name> <plus output node> <minus output node>
+<controlling voltage source name> <model name>
Example
W1 10 20 V1 IREF
Schematic format
PART attribute
<name>
Example
W1
REF attribute
<controlling voltage source name>
Example
VSENSE
MODEL attribute
<model name>
Example
SW
The W switch can operate in two distinct operational modes, Smooth Transition
(default) and Hysteresis.
Smooth Transition Mode:

Use this mode if you do not need input hysteresis. The smooth transition mini-
mizes convergence problems. In this mode ION and IOFF are specified. IT and
IH are ignored. The switch impedance changes smoothly from RON to ROFF as
the control current moves from ION to IOFF and from ROFF to RON as the con-
trol current moves from IOFF to ION.
Hysteresis Mode:
Use this mode if you need input hysteresis and the circuit is not sensitive to con-
761
vergence problems. In the Hysteresis mode, IT and IH are specified. ION and
IOFF are ignored. The switch impedance changes abruptly from ROFF to RON
as the control current moves past IT+IH and from RON to ROFF as the control
current moves past IT-IH.
This switch is controlled by the current through the source defined by <control-
ling voltage source name>.
RON and ROFF must be greater than zero and less than 1/Gmin.

.MODEL <model name> ISWITCH ([model parameters])
Examples
.MODEL W1 ISWITCH (RON=1 ROFF=1K ION=1 IOFF=1.5)
.MODEL W2 ISWITCH (RON=1 ROFF=1K IT=1 IH=1.5)
Model parameters
RON On resistance Ohms 1
ROFF Off resistance Ohms 1E6
ION Control current for On state A .001
IOFF Control current for Off state A 0
IT Threshold current A None
IH Hysteresis current A None
Model Equations
IC = Controlling current
IM = Mean of control currents = (ION+IOFF)/2
ID = Difference of control currents = ION-IOFF
k = Boltzmann's constant
T = Analysis temperature

Smooth Transition Mode: (Used if ION and IOFF are defined)
If ION > IOFF
If IC >= ION
RS = RON
If IC <= IOFF
RS = ROFF
If IOFF < IC < ION
RS = exp(LM + 3•LR•(IC-IM)/(2•ID) - 2•LR•(IC-IM)3/ID3)
If ION < IOFF

If IC <= ION
RS = RON
If IC >= IOFF
RS = ROFF
If IOFF > IC > ION
RS = exp(LM - 3•LR•(IC-IM)/(2•ID) + 2•LR•(IC-IM)3/ID3)
Hysteresis Mode:(Used if ION and IOFF are not defined)

If IC >= IT + IH
RS = RON
If IC <= IT - IH
RS = ROFF
Else
RS remains unchanged
Noise effects
Noise is modeled as a resistor equal to the resistance found during the DC operat-
ing point. The thermal noise current is calculated as follows:
I = sqrt(4•k•T/RS)
763
WAV file source
Schematic format
PART attribute
<name>
Example
W1
FILE attribute
<file name>
Examples
AUDIO2.WAV
ode_to_joy.wav
CHANNEL attribute
<channel number>
Examples
0
1
SCALE attribute
<scale>
Examples
1
15
REPEAT attribute
<repeat number>
Examples
1
3
ENABLE attribute
<enable_expression>

Examples
Time>120mS
Record_flag=TRUE
<file name>: This is the name of the WAV source.

<channel number>: This is the channel number to be used.
<scale>: This is the scale factor to multiply the file values by. Increasing it
increases the volume.
<repeat number>: This is the number of times to repeat the file contents.
<enable_expression>: The file will only be played when this boolean

expression evaluates to TRUE.
During a DC operating point calculation, the value on the source is the initial
value in the file.
The Play button in the Attribute dialog box can be used to play the contents of
a WAV source. The source can also be played from Transient Analysis / F10 /
Save Curves / Play button.
For an example of the WAV source, see the circuit file WAV.CIR.
765
Z transform source
Schematic format
PART attribute
<name>
Example
E1
ZEXP attribute
<transform expression>
Example
(.10*(Z+1)*(POW(Z,2)-.70*Z+1))/((Z-.56)*(POW(Z,2)-1.16*Z+.765))
CLOCK FREQUENCY attribute

<clock frequency>
Example
24khz
The Z transform source is a voltage source whose waveform is expressed as a

complex Z variable expression. It behaves like a Laplace source with Z replaced
by EXP(S/<clock frequency>), where S = J * 2 * PI * F, and F = frequency.
See the sample file Filters\ZDOMAIN.CIR for an example of how to use the
source.

rm12 463 766 PDF

Uploaded by

Copyright:

Available Formats

rm12 463 766 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

rm12 463 766 PDF

Uploaded by

Copyright:

Available Formats

Chapter 25 Macros

What's in this chapter

• Absolute value (full wave rectifier) function (ABS)

464 Chapter 25: Macros

Figure 25-1 ABS macro circuit

There are no input parameters. This implementation uses an NFV Function

Figure 25-2 AM macro circuit

466 Chapter 25: Macros

VOut(t) = gain VIn(t)

Figure 25-3 AMP macro circuit

Figure 25-4 BALUN macro circuit

468 Chapter 25: Macros

Figure 25-5 CENTAP macro circuit

This function is implemented with the CLIP macro:

Figure 25-6 CLIP macro circuit

470 Chapter 25: Macros

Here is the circuit that implements the macro.

Figure 25-6 Comb Filter macro circuit

Figure 25-7 COMP macro circuit

472 Chapter 25: Macros

Figure 25-8 COMPARATOR macro equivalent circuit

The parameter definitions are as follows:

Figure 25-9 Constant Power macro circuit

It is implemented with a function current source whose expression is,

474 Chapter 25: Macros

Figure 25-10 CPFSK macro circuit

Figure 25-12 DC Motor macro circuit

476 Chapter 25: Macros

This function is implemented with the DELAY macro:

Figure 25-12 DELAY macro circuit

Figure 25-13 DIAC macro circuit

478 Chapter 25: Macros

VOut(t) = scale d(VIn(t))/dt

This function is implemented with the DIF macro:

Figure 25-14 DIF macro circuit

This function is implemented with the DIGPOT macro:

Figure 25-15 DIGPOT macro circuit

VOut(t) = scale Va(t)/Vb(t)

This function is implemented with the DIV macro:

Figure 25-16 DIV macro circuit

F(s) = VOut(s) /VIn(s)

This function is implemented with the F macro:

Figure 25-17 F macro circuit

The input parameter is an expression representing the complex frequency transfer

482 Chapter 25: Macros

Figure 25-18 FSK macro circuit

The input parameters are as follows:

Figure 25-19 FWeighting macro circuit

This function is implemented with two cross-referenced linear dependent VOFI

Figure 25-20 GYRATOR macro circuit

A single parameter, G, determines the impedance transformation as follows:

This function is implemented with two cross-referenced linear dependent sourc-

Figure 25-21 Ideal_Trans2 macro circuit

A single parameter, GAIN, determines the transformer voltage gain.

486 Chapter 25: Macros

This function is implemented with four cross-referenced linear dependent sourc-

Figure 25-22 Ideal_Trans3 macro circuit

VOut(t) = vinit + scale

This function is implemented with the INT macro: