BJT amplifier: basic operation

VCC
IC IC
VCC RC

RC IC

Vo
↵IE
VB VB IB

IE t
0 0.2 0.4 0.6 0.8
VBE

* In the active mode, IC changes exponentially

with VBE : IC = ↵F IES [exp(VBE /VT ) 1]
* Vo (t) = VCC IC (t) RC
) the amplitude of Vo , i.e., IbC RC , can be
made much larger than V cB.

* Note that both the input (VBE ) and output t

(Vo ) voltages have DC (“bias”) components.

M. B. Patil, IIT Bombay

BJT amplifier: basic operation
VCC
IC (mA)
RC 50
VCC

IC 40
RC
1
30
Vo
↵IE
20
VB VB IB
10 2
IE
0
t
* The gain depends on the DC (bias) value of VBE ,
the input voltage in this circuit.
* In practice, it is not possible to set the bias value 2 1
of the input voltage to the desired value (e.g.,
0.673 V).
* Even if we could set the input bias as desired,
device-to-device variation, change in temperature,
etc. would cause the gain to change.
! need a better biasing method.
* Biasing the transistor at a specific VBE is t
equivalent to biasing it at a specific IC . 0.6 0.7
VBE (Volts)
M. B. Patil, IIT Bombay
BJT amplifier biasing
saturation
5
linear
VCC /RC
5 V VCC IB5 4

IB4

Vo (Volts)
RC 3

IB3

IC
C Vo 2
B IB2
Vi RB 1
E IB1
0 0
0 VCC 0 1 2 3 4 5
VCE (Vo )
Vi (Volts)
Consider a more realistic BJT amplifier circuit, with RB added to limit the base current (and thus protect the
transistor).
* When Vi < 0.7 V, the B-E junction is not sufficiently forward biased, and the BJT is in the cut-o↵ mode
(VBE = Vi , VBC = Vi VCC )
Vi 0.7
* When Vi exceeds 0.7 V, the BJT enters the linear region, and IB ⇡ . As Vi increases, IB and
RB
IC = IB also increase, and Vo = VCC IC RC falls.
* As Vi is increased further, Vo reaches VCEsat (about 0.2 V), and the BJT enters the saturation region (both

B-E and B-C junctions are forward biased).

M. B. Patil, IIT Bombay
BJT amplifier biasing

saturation
5
linear
VCC /RC
5 V VCC IB5 4

IB4

Vo (Volts)
RC 3

IB3

IC
C Vo 2
B IB2
Vi RB 1
E IB1
0 0
0 VCC 0 1 2 3 4 5
VCE (Vo )
Vi (Volts)

M. B. Patil, IIT Bombay

BJT amplifier biasing

saturation
5
linear
VCC /RC
5 V VCC IB5 4

IB4

Vo (Volts)
RC 3

IB3

IC
C Vo 2
B IB2
Vi RB 1
E IB1
0 0
0 VCC 0 1 2 3 4 5
VCE (Vo )
Vi (Volts)

dVo
* The gain of the amplifier is given by .
dVi
* Since Vo is nearly constant for Vi < 0.7 V (due to cut-o↵) and Vi > 1.3 V (due to saturation), the circuit
will not work an an amplifier in this range.
* Further, to get a large swing in Vo without distortion, the DC bias of Vi should be at the centre of the
amplifying region, i.e., Vi ⇡ 1 V .

M. B. Patil, IIT Bombay

 BJT amplifier biasing

VCC

RC

C Vo
B
Vi RB
E
B
5

3
Vo

0
0 1 2 3 4 5
Vi
 BJT amplifier biasing

1.05
VCC B
1.03
RC Vi
1.01
C Vo
B 0.99
Vi RB
E 0.97
B
5 0.95
3.40
4
3.20
Vo
3
Vo 3.00

2 2.80

1 2.60

2.40
0 0 0.2 0.4 0.6 0.8 1
0 1 2 3 4 5
Vi t (msec)
 BJT amplifier biasing

0.80 1.05
VCC A B
0.78 1.03
RC Vi
Vi
0.76 1.01
C Vo
B 0.74 0.99
Vi RB
E 0.72 0.97
ABC
5 0.70 0.95
5.00 3.40
4
4.90 3.20
Vo Vo
3
Vo 4.80 3.00

2 4.70 2.80

1 4.60 2.60

4.50 2.40
0 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
0 1 2 3 4 5
Vi t (msec) t (msec)
 BJT amplifier biasing

0.80 1.05 1.35

VCC A B C
0.78 1.03 1.33
RC Vi Vi
Vi
0.76 1.01 1.31
C Vo
B 0.74 0.99 1.29
Vi RB
E 0.72 0.97 1.27
ABC
5 0.70 0.95 1.25
5.00 3.40 0.65
4
4.90 3.20 0.55
Vo Vo Vo
3
Vo 4.80 3.00 0.45

2 4.70 2.80 0.35

1 4.60 2.60 0.25

4.50 2.40 0.15

0 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
0 1 2 3 4 5
Vi t (msec) t (msec) t (msec)

(SEQUEL file: ee101 bjt amp1.sqproj)

M. B. Patil, IIT Bombay

BJT amplifier

saturation
5
linear
VCC /RC
5 V VCC IB5 4

IB4

Vo (Volts)
RC 3

IB3

IC
C Vo 2
B IB2
Vi RB 1
E IB1
0 0
0 VCC 0 1 2 3 4 5
VCE (Vo )
Vi (Volts)

* The key challenges in realizing this amplifier in practice are

- adjusting the input DC bias to ensure that the BJT remains in the linear (active) region with a
certain bias value of VBE (or IC ).
- mixing the input DC bias with the signal voltage.
* The first issue is addressed by using a suitable biasing scheme, and the second by using “coupling”
capacitors.

M. B. Patil, IIT Bombay

BJT amplifier: a simple biasing scheme

15 V VCC

RC
RB 1k
C

B
E

“Biasing” an amplifier ) selection of component values for a certain DC value of IC (or VBE )
(i.e., when no signal is applied).

Equivalently, we may bias an amplifier for a certain DC value of VCE , since IC and VCE are related:
VCE + IC RC = VCC .

As an example, for RC = 1 k, = 100, let us calculate RB for IC = 3.3 mA, assuming the BJT to be operating
in the active mode.
IC 3.3 mA VCC VBE 15 0.7
IB = = = 33 µA = =
100 RB RB
14.3 V
! RB = = 430 k⌦ .
33 µA

M. B. Patil, IIT Bombay

BJT amplifier: a simple biasing scheme (continued)

15 V VCC

RC
RB 1k
C

B
E

With RB = 430 k, we expect IC = 3.3 mA, assuming = 100.

However, in practice, there is a substantial variation in the value (even for the same transistor type). The
manufacturer may specify the nominal value of as 100, but the actual value may be 150, for example.

With = 150, the actual IC is,

VCC VBE (15 0.7) V
IC = ⇥ = 150 ⇥ = 5 mA ,
RB 430 k
which is significantly di↵erent than the intended value, viz., 3.3 mA.

! need a biasing scheme which is not so sensitive to .

M. B. Patil, IIT Bombay

BJT amplifier: improved biasing scheme

10 V VCC

RC
R1 3.6 k
10 k IC

IB
IE
R2
RE
2.2 k
1k
BJT amplifier: improved biasing scheme

10 V VCC

RC RC RC
R1 3.6 k
10 k IC IC IC
R1 RTh
VCC VCC
IB IB IB
IE IE IE
R2 VCC R2 VTh
RE RE RE
2.2 k
1k

R2 2.2 k
VTh = VCC = ⇥ 10 V = 1.8 V , RTh = R1 k R2 = 1.8 k
R1 + R2 10 k + 2.2 k
Assuming the BJT to be in the active mode,
KVL: VTh = RTh IB + VBE + RE IE = RTh IB + VBE + ( + 1) IB RE
VTh VBE (VTh VBE )
! IB = , IC = IB = .
RTh + ( + 1) RE RTh + ( + 1) RE
For = 100, IC =1.07 mA.
For = 200, IC =1.085 mA.
M. B. Patil, IIT Bombay
BJT amplifier: improved biasing scheme (continued)

10 V VCC

RC
R1 3.6 k
10 k IC
6V
1.8 V
IB
1.1 V
IE
R2
RE
2.2 k
1k

With IC = 1.1 mA, the various DC (“bias”) voltages are

VE = IE RE ⇡ 1.1 mA ⇥ 1 k = 1.1 V ,

VB = VE + VBE ⇡ 1.1 V + 0.7 V = 1.8 V ,

VC = VCC IC RC = 10 V 1.1 mA ⇥ 3.6 k ⇡ 6 V ,

VCE = VC VE = 6 1.1 = 4.9 V .

M. B. Patil, IIT Bombay

BJT amplifier: improved biasing scheme (continued)
10 V VCC

RC
R1 3.6 k
10 k IC

IB
IE
R2
RE
2.2 k
1k

A quick estimate of the bias values can be obtained by ignoring IB (which is fair if is large). In that case,
R2 2.2 k
VB = VCC = ⇥ 10 V = 1.8 V .
R1 + R2 10 k + 2.2 k
V E = VB VBE ⇡ 1.8 V 0.7 V = 1.1 V .
VE 1.1 V
IE = = = 1.1 mA.
RE 1k
IC = ↵ IE ⇡ IE = 1.1 mA.

VCE = VCC IC R C IE RE = 10 V (3.6 k ⇥ 1.1 mA) (1 k ⇥ 1.1 mA) ⇡ 5 V .

M. B. Patil, IIT Bombay
Adding signal to bias

VCC

RC
R1

vB

R2
RE
Adding signal to bias

VCC

RC
R1

vB

CB

vs R2
RE

b sin !t (for example) needs to be mixed with the

* As we have seen earlier, the input signal vs (t) = V
desired bias value VB so that the net voltage at the base is vB (t) = VB + Vb sin !t.
* This can be achieved by using a coupling capacitor CB .
* Let us consider a simple circuit to illustrate how a coupling capacitor works.

M. B. Patil, IIT Bombay

RC circuit with DC + AC sources

vC
A R2
vA

vs (t) R1 V0 (DC)
Vm sin !t

We are interested in the solution (currents and voltages) in the “sinusoidal steady state” when the exponential
transients have vanished and each quantity x(t) is of the form X0 (constant) + Xm sin(!t + ↵).
There are two ways to obtain the solution:
(1) Solve the circuit equations directly:
vA (t) vA (t) V0 d
+ =C (vs (t) vA (t)) .
R1 R2 dt
(2) Use the DC circuit + AC circuit approach.

M. B. Patil, IIT Bombay

Resistor in sinusoidal steady state

vR (t)

iR (t) R

M. B. Patil, IIT Bombay

Resistor in sinusoidal steady state

vR (t)

iR (t) R

bR sin (!t + ↵),

Let vR (t) = VR + vr (t) where VR = constant, vr (t) = V
iR (t) = IR + ir (t) where IR = constant, ir (t) = b
IR sin (!t + ↵).

Since vR (t) = R ⇥ iR (t), we get [VR + vr (t)] = R ⇥ [IR + ir (t)].

This relationship can be split into two:
VR = R ⇥ IR , and vr (t) = R ⇥ ir (t).
In other words, a resistor can be described by
VR vr (t)

IR R ir (t) R
DC AC

M. B. Patil, IIT Bombay

Capacitor in sinusoidal steady state

vC (t)

iC (t) C

M. B. Patil, IIT Bombay

Capacitor in sinusoidal steady state

vC (t)

iC (t) C

bC sin (!t + ↵),

Let vC (t) = VC + vc (t) where VC = constant, vc (t) = V
b
iC (t) = IC + ic (t) where IC = constant, ic (t) = IC sin (!t + ).

dvC d
Since iC (t) = C , we get [IC + ic (t)] = C (VC + vc (t)).
dt dt
This relationship can be split into two:
dVC dvc
IC = C = 0, and ic (t) = C .
dt dt
In other words, a capacitor can be described by
VC vc (t)

IC ic (t) C
DC AC

M. B. Patil, IIT Bombay

Voltage sources in sinusoidal steady state

DC voltage source:
vS (t) VS vs (t)

iS (t) IS is (t)
vS (t) = VS + 0 DC AC

AC voltage source:
vS (t) VS vs (t)

iS (t) IS is (t)
vS (t) = 0 + vs (t) DC AC

M. B. Patil, IIT Bombay

RC circuit with DC + AC sources
DC circuit AC circuit
vC VC vc
A R2 A R2 A R2
vA VA va

vs (t) R1 V0 (DC) R1 V0 vs (t) R1

Vm sin !t Vm sin !t

M. B. Patil, IIT Bombay

RC circuit with DC + AC sources
DC circuit AC circuit
vC VC vc
A R2 A R2 A R2
vA VA va

vs (t) R1 V0 (DC) R1 V0 vs (t) R1

Vm sin !t Vm sin !t

VA VA V0
DC circuit: + = 0. (1)
R1 R2
va va d
AC circuit: + =C (vs va ). (2)
R1 R2 dt
VA + v a VA + v a V 0 d
Adding (1) and (2), we get + =C (vs va ). (3)
R1 R2 dt
Compare with the equation obtained directly from the original circuit:
vA v A V0 d
+ =C (vs vA ). (4)
R1 R2 dt
Eqs. (3) and (4) are identical since vA = VA + va .
! Instead of computing vA (t) directly, we can compute VA and va (t) separately, and then use
vA (t) = VA + va (t).
M. B. Patil, IIT Bombay
Common-emitter amplifier

RC coupling
R1 capacitor
coupling
capacitor VCC
CC

CB
RL
vs R2 load
RE
resistor
CE

bypass
capacitor
Common-emitter amplifier

RC coupling RC RC
R1 capacitor R1 R1
coupling
capacitor VCC
CC CC
VCC AND
CB CB
RL
RL
vs R2 load
R2 vs R2
RE RE RE
resistor
CE CE

bypass DC circuit AC circuit

capacitor

* The coupling capacitors ensure that the signal source and the load resistor do not a↵ect the DC bias of
the amplifier. (We will see the purpose of CE a little later.)
* This enables us to bias the amplifier without worrying about what load it is going to drive.

M. B. Patil, IIT Bombay

Common-emitter amplifier: AC circuit

RC
R1

CC

CB
RL
vs R2
RE
CE
Common-emitter amplifier: AC circuit

RC
R1

CC

CB
RL
vs R2
RE
CE

* The coupling and bypass capacitors are “large” (typically, a few µF ), and at frequencies of interest, their
impedance is small.
For example, for C = 10 µF , f = 1 kHz,
1
ZC = = 16 ⌦,
2⇡ ⇥ 103 ⇥ 10 ⇥ 10 6
which is much smaller than typical values of R1 , R2 , RC , RE (a few k⌦).
) CB , CC , CE can be replaced by short circuits at the frequencies of interest.
Common-emitter amplifier: AC circuit

RC RC
R1 R1

CC

CB
RL RL RC RL
vs R2 vs R2 vs R1 R2
RE
CE

* The coupling and bypass capacitors are “large” (typically, a few µF ), and at frequencies of interest, their
impedance is small.
For example, for C = 10 µF , f = 1 kHz,
1
ZC = = 16 ⌦,
2⇡ ⇥ 103 ⇥ 10 ⇥ 10 6
which is much smaller than typical values of R1 , R2 , RC , RE (a few k⌦).
) CB , CC , CE can be replaced by short circuits at the frequencies of interest.
* The circuit can be re-drawn in a more friendly format.
* We now need to figure out the AC description of a BJT.
M. B. Patil, IIT Bombay
BJT: AC model

iC 1.1
Vm = 10 mV
Vm = 5 mV
Vm = 2 mV
B

iC (mA)
0.9
iB
vBE iE
0.7
E
vBE (t) = V0 + Vm sin !t 0.5
0 0.2 0.4 0.6 0.8 1
V0 = 0.65 V, f = 1 kHz t (msec)

M. B. Patil, IIT Bombay

BJT: AC model

iC 1.1
Vm = 10 mV
Vm = 5 mV
Vm = 2 mV
B

iC (mA)
0.9
iB
vBE iE
0.7
E
vBE (t) = V0 + Vm sin !t 0.5
0 0.2 0.4 0.6 0.8 1
V0 = 0.65 V, f = 1 kHz t (msec)

* As the vBE amplitude increases, the shape of iC (t) deviates from a sinusoid
! distortion.
* If vbe (t), i.e., the time-varying part of vBE , is kept small, iC varies linearly
with vBE . How small? Let us look at this in more detail.

M. B. Patil, IIT Bombay

BJT: small-signal model
C C
iC Vm = 10 mV
iC 1.1
Vm = 5 mV
iB ↵IE Vm = 2 mV

iC (mA)
B B 0.9
iB vBE
vBE iE
E iE 0.7
vBE (t) = V0 + Vm sin !t
E
V0 = 0.65 V, f = 1 kHz
0.5 t (msec)
Let vBE (t) = VBE + vbe (t) (bias+signal), and iC (t) = IC + ic (t). 0 0.2 0.4 0.6 0.8 1
 ✓ ◆
vBE (t)
Assuming active mode, iC (t) = ↵ iE (t) = ↵ IES exp 1 .
VT
✓ ◆
vBE (t)
Since the B-E junction is forward-biased, exp 1, and we get
VT
✓ ◆ ✓ ◆ ✓ ◆ ✓ ◆
vBE (t) VBE + vbe (t) VBE vbe (t)
iC (t) = ↵ IES exp = ↵ IES exp = ↵ IES exp ⇥ exp .
VT VT VT VT
✓ ◆
VBE
If vbe (t) = 0, iC (t) = IC (the bias value of iC ), i.e., IC = ↵ IES exp
VT
✓ ◆
vbe (t)
) iC (t) = IC exp .
VT
M. B. Patil, IIT Bombay
BJT: small-signal model

C 1.1
Vm = 10 mV
C
iC Vm = 5 mV
iC
Vm = 2 mV

iC (mA)
iB 0.9
↵IE
B B
iB vBE
0.7
vBE iE vBE (t) = V0 + Vm sin !t
E iE
V0 = 0.65 V, f = 1 kHz
E 0.5 t (msec)
0 0.2 0.4 0.6 0.8 1
✓ ◆ " #
vbe (t) x2
iC (t) = IC exp = IC 1+x + + ··· , x = vbe (t)/VT .
VT 2
If x is small, i.e., if the amplitude of vbe (t) is small compared to the thermal voltage VT , we get

vbe (t)
iC (t) = IC 1 + .
VT
We can now see that, for |vbe (t)| ⌧ VT , the relationship between iC (t) and vbe (t) is linear, as we have observed previously.

vbe (t) IC
iC (t) = IC + ic (t) = IC 1 + ) ic (t) = vbe (t)
VT VT

M. B. Patil, IIT Bombay

BJT: small-signal model

C C
iC ib ic
iC B C

iB ↵IE vbe r⇡
B B gm vbe
iB vBE
vBE iE ie
E iE

E E

IC
The relationship, ic (t) = vbe (t) can be represented by a VCCS, ic (t) = gm vbe (t),
VT
where gm = IC /VT is the “transconductance.”
For the base current, we have,
1
iB (t) = IB + ib (t) = [IC + ic (t)]

1 1
! ib (t) = ic (t) = gm vbe (t) ! vbe (t) = ( /gm ) ib (t).

The above relationship is represented by a resistance, r⇡ = /gm , connected between B and E.

The resulting model is called the ⇡-model for small-signal description of a BJT.

M. B. Patil, IIT Bombay

BJT: small-signal model

C C
iC ib ic
iC B C

iB ↵IE vbe r⇡
B B gm vbe
iB vBE
vBE iE ie
E iE

E E

* The transconductance gm depends on the biasing of the BJT, since gm = IC /VT . For IC = 1 mA,
VT ⇡ 25 mV (room temperature), gm = 1 mA/25 mV = 40 mf (milli-mho or milli-siemens).
* r⇡ also depends on IC , since r⇡ = /gm = VT /IC . For IC = 1 mA, VT ⇡ 25 mV , = 100, r⇡ = 2.5 k⌦.
* Note that the small-signal model is valid only for small vbe (small compared to VT ).

M. B. Patil, IIT Bombay

 BJT: small-signal model

C
ib ic
iC B C

vbe r⇡
B gm vbe
iB
vBE iE ie

E E

* In the above model, note that ic is independent of vce .

 BJT: small-signal model

C
ib ic 1 ib ic
iC B C B C

IC (mA)
vbe r⇡ vbe r⇡ ro
B gm vbe gm vbe
iB
vBE iE ie ie

E E E
0
0 1 2 3 4 5
VCE (V)

* In the above model, note that ic is independent of vce .

dIC
* In practice, ic does depend on vce because of the Early e↵ect, and ⇡ constant = 1/ro , where ro is
dVCE
called the output resistance.
* A more accurate model includes ro as well.

M. B. Patil, IIT Bombay

 BJT: small-signal model
insulator base emitter collector substrate
contact contact contact contact

p+ n+ n+
p+
p
epi-layer ⇠ 5 µm
ib Cµ ic n
B C
rb
n+ buried layer
C⇡ vbe r⇡ ro
gm vbe
p+ substrate > 250 µm

ie

* A few other components are required to make the small-signal model complete:
rb : base spreading resistance
C⇡ : base charging capacitance + B-E junction capacitance
Cµ : B-C junction capacitance
* The capacitances are typically in the pF range. At low frequencies, 1/!C is large, and the capacitances
can be replaced by open circuits.
* Note that the small-signal models we have described are valid in the active region only.
M. B. Patil, IIT Bombay
Common-emitter amplifier

RC coupling RC RC
R1 capacitor R1 3.6 k R1
coupling 10 k 6V
capacitor VCC 10 V
CC 1.8 V CC
VCC AND
CB 1.1 V CB
RL
RL
vs R2 load
R2 vs R2
RE 2.2 k RE RE
resistor 1k
CE CE

bypass DC circuit AC circuit

capacitor

* We have already analysed the DC (bias) circuit of this amplifier and found that VB = 1.8 V , VE = 1.1 V ,
VC = 6 V , and IC = 1.1 mA.
* We now analyse the AC (small-signal) circuit to obtain vb , ve , vc , ic .
* We will then get the complete solution by simply adding the DC and AC results, e.g., iC (t) = IC + ic (t).
* We will assume that CB , CC , CE are large enough so that, at the signal frequency (say, 1 kHz), they can
be replaced by short circuits.
M. B. Patil, IIT Bombay
Common-emitter amplifier

RC RL
vs R1 R2

E
Common-emitter amplifier

Cµ
B B C B C
rb
RC RL C⇡ vbe r⇡ ro
vs vs gm vbe RC RL R1 R2 vbe r⇡ RC RL
R1 R2 R1 R2
vs gm vbe

E E E

* The parasitic capacitances C⇡ and Cµ are in the pF range. At a signal frequency of 1 kHz, the impedance
corresponding to these capacitances is
j j
Z⇠ = ⇠ j 100 M⌦
!C 2⇡ ⇥ 103 ⇥ 10 12
! C⇡ and Cµ can be replaced by open circuits.
* For simplicity, we will assume rb to be small and ro to be large (this assumption will only slightly a↵ect the
gain computation).
* The above considerations significantly simplify the AC circuit.

M. B. Patil, IIT Bombay

Common-emitter amplifier

B C

vs R1 R2 r⇡ RC RL vo
10 k 2.2 k gm vbe 3.6 k 10 k

vo = (gm vbe ) ⇥ (RC k RL ) = (gm vs ) ⇥ (RC k RL )

vo
! ALV = voltage gain = = gm (RC k RL )
vs
(superscript L is used because the gain includes the e↵ect of RL .)
Since IC (bias current) = 1.1 mA, gm = IC /VT = 1.1 mA/25.9 mV = 42.5 mf.
! ALV = 42.5 mf ⇥ (3.6 k k 10 k) = 112.5

For vs (t) = (2 mV ) sin !t, the AC output voltage is,

vo = ALV vs = (112.5) (2 mV ) sin !t = (225 mV ) sin !t

The AC collector current is,

ic = gm vbe = gm vs = 42.5 mf ⇥ (2 mV ) sin !t = 85 sin !t µA.

M. B. Patil, IIT Bombay

Common-emitter amplifier

RC
R1 3.6 k
10 k
CC
VCC
CB 10 V
RL
vs R2
2.2 k RE
1k CE

For vs (t) = (2 mV ) sin !t, we can now obtain expressions for the instantaneous currents and voltages:
vC (t) = VC + vc (t) = VC + vo (t) = 6 V (225 mV ) sin !t .
iC (t) = IC + ic (t) = 1.1 mA + 0.085 sin !t mA .

Note that the above procedure (DC + AC analysis) can be used only if the small-signal approximation (i.e., |vbe | ⌧ VT ) is
valid. In the above example, the amplitude of vbe is 2 mV , which is much smaller than VT (about 25 mV).
For vs (t) = (20 mV ) sin !t, for example, the small-signal approximation will not hold, and a numerical simulation will be
required to obtain the currents and voltages of interest.
In practice, such a situation is anyway not prevalent (because it gives rise to distortion in the output voltage) except in special
types of amplifiers.

M. B. Patil, IIT Bombay

Frequency response of common-emitter amplifier
CB , CC , CE : short circuit
B rb C C⇡ , Cµ : open circuit
103
CB Cµ CC
C⇡ r⇡ ro
gm vbe

Gain
102
vs R1 R2 E RC RL

RE CE 101
101 102 103 104 105 106 107 108
Frequency (Hz)

108
* CB , CE , CC are large capacitances ! 1/!C is negligibly
small (short circuit) except at low frequencies. C = 1 pF
106
* C⇡ , Cµ are small capacitances ! 1/!C is very large (open
circuit) except at high frequencies.

|Z| (⌦)
* In the “mid-band” range (which we have considered so far), 104
the large capacitances behave like short circuits, and the
small capacitances like open circuits. In this range, the gain C = 10 µF
102
is independent of frequency.

100 1
10 102 103 104 105 106 107 108
f (Hz) M. B. Patil, IIT Bombay
General representation of an amplifier

source
resistance

Rs Ro
load
source vs vi vo
Rin resistance
(signal) AV vi RL

amplifier

* An amplifier is represented by a voltage gain, an input resistance Rin , and an output resistance Ro . For a
voltage-to-voltage amplifier, a large Rin and a small Ro are desirable.
* The above representation involves AC quantities only, i.e., it describes the AC equivalent circuit of the
amplifier.
* The DC bias of the circuit can a↵ect parameter values in the AC equivalent circuit (AV , Rin , Ro ). For
example, for the common-emitter amplifier, AV / gm = IC /VT , IC being the DC (bias) value of the
collector current.
* Suppose we are given an amplifier as a “black box” and asked to find AV , Rin , and Ro . What experiments
would give us this information?

M. B. Patil, IIT Bombay

Voltage gain AV

source
resistance

Rs Ro il
load
source vs vi vo
Rin resistance
(signal) AV vi RL

amplifier

If RL ! 1, il ! 0, and vo ! AV vi .

We can remove RL (i.e., replace it with an open circuit), measure vi and vo , then use AV = vo /vi .

M. B. Patil, IIT Bombay

Input resistance Rin

source
resistance

R s ii Ro
load
source vs vi vo
Rin resistance
(signal) AV vi RL

amplifier

Measurement of vi and ii yields Rin = vi /ii .

M. B. Patil, IIT Bombay

Output resistance Ro

io

Rs Ro Rs Ro

vs vi Rin vo RL vi Rin vo
AV vi

Method 1:
If vs ! 0, AV vi ! 0.
Now, connect a test source vo , and measure io ! Ro = vo /io .
(This method works fine on paper, but it is difficult to use experimentally.)

M. B. Patil, IIT Bombay

Output resistance Ro

Rs Ro

vs vi Rin vo
AV vi RL

Method 2:
RL
vo = AV v i .
RL + Ro
If RL ! 1, vo1 = AV vi .
1 1
If RL = Ro , vo2 = AV vi = vo1 .
2 2
Procedure:
Measure vo1 with RL ! 1 (i.e., RL removed).
Vary RL and observe vo .
When vo is equal to vo1 /2, measure RL (after removing it).
Ro is the same as the measured resistance.
M. B. Patil, IIT Bombay
Common-emitter amplifier

RC coupling B C
3.6 k capacitor
R1
coupling 10 k
capacitor
CC
VCC vs vi R1 R2 r⇡ RC vo RL
10 V gm vbe
CB
RL
E
vs R2 load
2.2 k RE
1k resistor amplifier
CE

bypass
capacitor

M. B. Patil, IIT Bombay

Common-emitter amplifier

RC coupling B C
3.6 k capacitor
R1
coupling 10 k
capacitor
CC
VCC vs vi R1 R2 r⇡ RC vo RL
10 V gm vbe
CB
RL
E
vs R2 load
2.2 k RE
1k resistor amplifier
CE

bypass
capacitor
vo
AV = , with RL ! 1.
vi
gm vbe RC
AV = = gm R C = 42.5 mf ⇥ 3.6 k = 153.
vi
The input resistance of the amplifier is, by inspection, Rin = (R1 k R2 ) k r⇡ .
r⇡ = /gm = 100/42.5 mf = 2.35 k ! Rin = 1 k.
The output resistance is RC (by “Method 1” seen previously).

M. B. Patil, IIT Bombay

Common-emitter amplifier with partial bypass

RC coupling RC
R1 capacitor R1
coupling
capacitor
CC

CB VCC VCC
RL
RE1 RE
R2 load R2
vs resistor
RE2
CE

bypass DC circuit
capacitor

M. B. Patil, IIT Bombay

Common-emitter amplifier with partial bypass

RC coupling RC
R1 capacitor R1
coupling
capacitor
CC

CB VCC VCC
RL
RE1 RE
R2 load R2
vs resistor
RE2
CE

bypass DC circuit
capacitor

* For DC computation, CE is open, and the DC analysis is therefore identical to our earlier amplifier, with
RE RE 1 + RE 2 .
* Bypassing a part of RE (as opposed to all of it) does have an impact on the voltage gain (see next slide).

M. B. Patil, IIT Bombay

Common-emitter amplifier with partial bypass

B C
RC coupling
R1 capacitor ib ib
coupling
capacitor r⇡
CC gm vbe
CB VCC vs R1 R2 RC RL
E
RL
RE1 ( + 1) ib
R2 load
vs resistor
RE1

RE2
CE bypass
capacitor AC circuit

Again, assume that, at the frequency of operation, CB , CC , CE can be replaced by short circuits, and the BJT
parasitic capacitances by open circuits.
vs
vs = ib r⇡ + ( + 1) ib RE 1 ! ib = .
r⇡ + ( + 1) RE 1
vo (RC k RL ) (RC k RL )
vo = ib ⇥ (RC k RL ) ! = ⇡ if r⇡ ⌧ ( + 1) RE 1 .
vs r⇡ + ( + 1) RE 1 RE 1
Note: RE 1 gets multiplied by ( + 1).
M. B. Patil, IIT Bombay
Common-emitter amplifier with partial bypass

B C
RC coupling
R1 capacitor ib ib
coupling
capacitor r⇡
CC gm vbe
CB VCC vs R1 R2 RC RL
E
RL
RE1 ( + 1) ib
R2 load
vs resistor
RE1

RE2
CE bypass
capacitor AC circuit

vbe r ⇡ ib r⇡
= =
vs r⇡ ib + RE ( + 1) ib r⇡ + RE ( + 1)
The small-signal condition, viz., |vbe (t)| ⌧ VT now implies
r⇡ r⇡ + RE ( + 1)
|vs | ⌧ VT or |vs | ⌧ VT ⇥ , which is much larger than VT .
r⇡ + RE ( + 1) r⇡
! Although the gain is reduced, partial emitter bypass allows larger input voltages to be applied without
causing distortion in vo (t). (For comparison, we required |vs | ⌧ VT for the CE amplifier.)
M. B. Patil, IIT Bombay