Physical Resistor (phy_res)

A physical resistor consists of a two terminal resistor (tied between t1 and t2) and two diodes (tied between t1-t0 and t2-t0). The diodes are junction diodes. Under normal operation, the two diodes are reverse biased, but the parameter subtype can reverse the direction of the diodes. If you do not specify t0, ground is assumed. The instance parameters always override model parameters. If you do not specify the instance resistance value, it is calculated from the model parameters.

This device is supported within altergroups.

If R(inst) is not given and R(model) is given,

R(inst) = R(model).

Otherwise,

R(inst) = Rsh * (L - 2 * etchl) / (W - 2 * etch).

If the polynomial coefficients vector (coeffs=[c1 c2 ...]) is specified, the resistor is nonlinear. When nonlinform is set to g, the conductance is

G(V) = dI / dV

= (1 + c1 * V + c2 * V^2 + ...) / R(inst)

where

V = V(t1) - V(t2)

Here V is the controlling voltage across the resistor. It is also the controlling voltage when the model parameter polyarg is set to diff. In this form, the physical resistor is symmetric with respect to V(t1) and V(t2). The branch current as a function of the applied voltage is given by

I(V) = (V / R(inst)) * (1 + 1/2 * c1 * V + 1/3 * c2 * V^2 + ...)

where ck is the kth entry in the coefficient vector.

If the model parameter polyarg is set to sum, then the controlling voltage is defined as

Vsum = ( (V(t1) - V(t0)) + (V(t2) - V(t0)) )/2

Here, Vsum is the controlling voltage between the resistor and the substrate, t0. In this case, the device becomes asymmetric with respect to V(t1) and V(t2). The branch current as a function of the applied voltage for this case is given by

I(Vsum) = (V / R(inst)) * (1 + c1 * Vsum + c2 * Vsum^2 + ...)

The large-signal conductance is given by

G(Vsum) = I/V = (1 + c1 * Vsum + c2 * Vsum^2 + ...) / R(inst)

Note, since the device is asymmetrical, the small-signal model is more complicated then a simple conductance.

When nonlinform is set to r, the resistance is

R(V) = dV / dI

= R(inst) * (1 + c1 * V + c2 * V^2 + ...).

The resistance as a function of temperature is given by:

R(T) = R(tnom) * [1 + tc1 * (T - tnom) + tc2 * (T - tnom)^2].

where

T = trise(inst) + temp

if trise(inst) is given, and

T = trise(model) + temp

otherwise.

If you do not specify the junction leakage current (is) and js is specified, the leakage current is calculated from js and the device dimensions.

is = js * 0.5 * (L - 2 * etchl) * (W - 2 * etch)

If you specify the instance capacitance or the linear model capacitance, linear capacitors are used between t1-t0 and t2-t0. Otherwise, nonlinear junction capacitors are used and the zero-bias capacitance values are calculated from the model parameters.

If C(inst) is not given and C(model) is given,

C(inst) = C(model).

Otherwise,

C(inst) = 0.5 * Cj * (L - 2 * etchlc) * (W - 2 * etchc) + Cjsw * (W + L - 2 * etchc - 2 * etchlc).

If the capacitance is nonlinear, the temperature model for the junction capacitance is used. Otherwise, the following equation is used.

C(T) = C(tnom) * [1 + tc1c * (T - tnom) + tc2c * (T - tnom)^2].

Thermal Noise

where:

R is the resistor value and T is the temperature.

Shot Noise for the Junctions

Where I is the current flow through the junction.

Flicker Noise for the Resistor

Where KF, AF, FEXP, WEEXP, LEEXP, WDEXP and LDEXP are model parameters, I is the current, W and L are the device width and length, Weff=W-2ETCH and Leff=L-2ETCHL.

Flicker Noise for the Junctions

Where:

Where KF, AF and FEXP are model parameters, I is the current, and f is the frequency.

Sample Instance Statement

res1 (net9 vcc) resphy l=1e-3 w=2e-6

Sample Model Statement

model resphy phy_res rsh=85 tc1=1.53e-3 tc2=4.67e-7 etch=0 cj=1.33e-3 cjsw=3.15e-10 tc1c=9.26e-4

Instance Syntax

Name  1  2  [0] ModelName parameter=value ...

Instance Parameters

r (Ω)	Resistance.
c (F)	Linear capacitance.
l (m)	Line length.
w (m)	Line width.
region=normal	Estimated operating region. Spectre generates output number (0-1) in a rawfile. Possible values are normal or breakdown.
tc1=0 1/C	Linear temperature coefficient of resistor.
tc2=0 C-2	Quadratic temperature coefficient of resistor.
tc1c=0 1/C	Linear temperature coefficient of linear capacitor.
tc2c=0 C-2	Quadratic temperature coefficient of linear capacitor.
trise (C)	Temperature rise from ambient.
isnoisy=yes	Should resistor generate noise. Possible values are no or yes.
m=1	Multiplicity factor.

The w and l parameters are scaled by the option parameters scale and scalem. The values of w and l printed by Spectre are those given in the input file. These values may not have the correct units if the scaling factors are not unity. The correct effective resistor dimensions are stored in the output parameters. You can obtain these dimensions with the info statement. You can delete the diodes from the device by either setting is=0 or subtype=poly. You can also set both mj and mjsw to zero to make the capacitance linear but still calculated from the device geometry. If subtype=poly, the linear capacitors will always be used irrespective of the values of mj and mjsw.

Model Syntax

model modelName phy_res parameter=value ...

Model Parameters

Substrate type parameters

subtype=p

Substrate type. Possible values are n, p or poly.

Resistance parameters

r=∞ Ω	Default resistance.
rsh=∞ Ω/sqr
	Sheet resistance.
minr=0.1 Ω	Minimum resistance.
coeffs=[...]	Vector of polynomial conductance or resistance coefficients.
nonlinform=g	The form of the nonlinear resistance. Possible values are g or r.
polyarg=diff	Polynomial model argument type. Possible values are sum or diff.

Temperature effects parameters

tc1=0 1/C	Linear temperature coefficient of resistor.
tc2=0 C-2	Quadratic temperature coefficient of resistor.
tc1c=0 C-2	Linear temperature coefficient of linear capacitor.
tc2c=0 C-2	Quadratic temperature coefficient of linear capacitor.
tnom (C)	Parameters measurement temperature. Default set by options.
trise=0 C	Temperature rise from ambient.

Junction diode model parameters

is (A)	Saturation current.
js=0 A/m2	Saturation current density.
n=1	Emission coefficient.
eg=1.11 V	Band gap.
xti=3	Saturation current temperature exponent.
imelt=`imax’ A	Explosion current, diode is linearized beyond this current to aid convergence.
jmelt=`jmelt’ A/m2
	Explosion current density, diode is linearized beyond this current to aid convergence.
imax=1 A	Maximum current, currents above this limit generate a warning.
jmax=1e8 A/m2	Maximum current density, currents above this limit generate a warning.
dskip=yes	Use simple piece-wise linear model for diode currents below 0.1*iabstol. Possible values are no or yes.
bvj=∞ V	Junction reverse breakdown voltage.

Junction capacitance model parameters

c=0 F	Default linear capacitance.
cj=0 F/m2	Zero-bias junction bottom capacitance density.
cjsw=0 F/m	Zero-bias junction sidewall capacitance density.
mj=1/2	Junction bottom grading coefficient.
mjsw=1/3	Junction sidewall grading coefficient.
pb=0.8 V	Junction bottom built-in potential.
pbsw=0.8 V	Junction sidewall built-in potential.
fc=0.5	Junction bottom capacitor forward-bias threshold.
fcsw=0.5	Junction sidewall capacitor forward-bias threshold.
tt=0 s	Transit time.

Device size parameters

l=∞ m	Default line length.
w=1e-6 m	Default line width.
etch=0 m	Narrowing due to etching.
etchl=0 m	Length reduction due to etching.
etchc=etch m	Narrowing due to etching for capacitances.
etchlc=etchl m	Length reduction due to etching for capacitances.
scaler=1	Resistance scaling factor.
scalec=1	Capacitance scaling factor.

Noise model parameters

kf=0	Flicker (1/f) noise coefficient.
af=1	Flicker (1/f) noise exponent.
wdexp=1	Flicker (1/f) noise W exponent.
ldexp=1	Flicker (1/f) noise L exponent.
weexp=0	Flicker (1/f) noise W effective exponent.
leexp=0	Flicker (1/f) noise L effective exponent.
fexp=1	Flicker (1/f) noise frequency exponent.

DC-mismatch model parameters

mr=0.0 Ω2	Resistor mismatch dependence.
mrl=0.0 Ω2/m^mrlp
	Resistor mismatch length dependence.
mrlp=0.0	Resistor mismatch length power dependence.
mrw=0.0 Ω2/m^mrwp
	Resistor mismatch width dependence.
mrwp=0.0	Resistor mismatch width power dependence.
mrlw1=0.0 Ω2/m^(2 mrlw1p)
	Resistor mismatch area 1 dependence.
mrlw1p=0.0	Resistor mismatch area 1 power dependence.
mrlw2=0.0 Ω2/m^(2 mrlw2p)
	Resistor mismatch area 2 dependence.
mrlw2p=0.0	Resistor mismatch area 2 power dependence.

Output Parameters

leff (m)	Effective line length.
weff (m)	Effective line width.
iseff (A)	Effective saturation current.
reff (Ω)	Effective resistance.
ceff (F)	Effective zero-bias capacitance.
meff	Effective multiplicity factor (m-factor).

Operating-Point Parameters

subtype=p	Substrate type. Possible values are n, p or poly.
region=normal	Estimated operating region. Spectre generates output number (0-1) in a rawfile. Possible values are normal or breakdown.
i (A)	Current through the resistor.
capd1 (F)	Capacitance at the positive node.
capd2 (F)	Capacitance at the negative node.
id1 (A)	Current between nodes t1 and t0.
id2 (A)	Current between nodes t2 and t0.
res (Ω)	Resistance between nodes t1 and t2.
resd1 (Ω)	Resistance between nodes t1 and t0.
resd2 (Ω)	Resistance between nodes t2 and t0.
pwr (W)	Power at op point.

Related Topics

Two Terminal Resistor (resistor)

R2 Model (r2)

Passive Components

Return to top