Chapter 9

Output Stages And Power Amplifiers

Low Output Resistance – no loss of gain

Small-Signal Not applicable

Total-Harmonic Distortion (fraction of %)

Efficiency

Temperature Requirements
Collector current waveforms for transistors operating in (a) class A, (b) class B,
(c) class AB, and (d) class C amplifier stages.
 An emitter follower (Q1) biased with a
Class A
constant current I supplied by transistor Q2.

Transfer Characteristics

Transfer characteristic of the emitter follower. This linear

characteristic is obtained by neglecting the change in vBE1
with iL. The maximum positive output is determined by
the saturation of Q1. In the negative direction, the limit of
the linear region is determined either by Q1 turning off or
by Q2 saturating, depending on the values of I and RL.
Class A

Transfer Crossover distortion can be eliminated by biasing the transistors at a small,

Characteristics non-zero current.
A bias Voltage VBB is applied between Qn and Qp.
For vi = 0, vo = 0, and a voltage VBB/2 appears across the base-emitter juncti
of each transistor.

VBB
2⋅ VT
iN iP IQ IS⋅ e

VBB is selected to result the required quiscent current IQ

VBB
vo vi + − v BEN
2

iN iP + iL
 iN   iP   iQ 
v BEN + v EBP VBB VT ⋅ ln   + VT  
⋅ ln 2⋅ VT ⋅ ln  
 IS   IS   IS 
2 2
iN IQ

2 2
iN − iL⋅ iN − IQ 0
Class A

Transfer Characteristics
From figure 9.3 we can see that

v omax VCC − VCE1sat

In the negative direction, the limite of the linear region is

determined either by Q1 turning off

v Omin −I⋅ RL

or by Q2 saturating

v Omin −VCC + VCE2sat

Depending on the values of I and RL. The absolutely lowest

output voltage is that given by the previous equation and is
achieved provided that the bias current I is greater than the
magnitude of the corresponding load current

−VCC + VCE2sat
I≥
RL
Class A

Transfer Characteristics

Exercises D9.1 and D9.2

Class A
1 2

Signal Waveforms

vo( t ) 0 vcE1( t ) 1

1 0
0 5 10 0 5 10
t t

2
1

ic1( t ) 1
pD1 ( t ) 0.5

0
0 5 10 0
0 5 10
t
t
Class A

Power Dissipation

P VCC ⋅ I

Largest Power Dissipation When vo = 0

Q1 must be able to withsatnd a continuous dissipation of VCC*I

The power dissipation of Q1 depends on the value of RL.

If RL is infinite, iC1 = I and the dissipation in Q1 depends on vo.
Maximum power dissipation will occur when vo = -VCC since vCE1 will be 2VCC.
pD1 = 2VCC*I. This condition would not normally persist for a prolonged interval, so
the design need not be that conservative. The average pD1 = VCC*I

When RL is zero a positive voltage would result in a theoretically infinite current (large
practical value) would flow through Q1. Short-circuit protection is necessary.
Class A

Power Conversion Efficiency

load_pow er( PL)

η
supply_pow er ( PS)

2
1 Vo Vo average voltage
PL ⋅
2 RL

PS 2 ⋅ VCC ⋅ I

2
1 Vo 1  Vo   Vo 
η ⋅ ⋅ ⋅ 
4 I ⋅ RL ⋅ VCC 4  I ⋅ RL   VCC 

Vo ≤ VCC Vo ≤ I ⋅ RL

maximum efficiency is obtained when

Vo VCC I ⋅ RL
Class A

Exercise 9.4

−3
Vopeak := 8 I := 100 ⋅ 10 RL := 100 VCC := 10

2
 Vopeak 
 
PL :=
 2  PL = 0.32
100

Pplus := VCC ⋅ I Pplus = 1

Pminus := VCC ⋅ I Pminus = 1 PS := Pplus + Pminus

PL
η := η = 0.16
PS
 Biasing the Class B Output

• No DC current is used to bias this configuration.

• Activated when the input voltage is greater than the Vbe

for the transistors.

• npn Transistor operates when positive, pnp when negative.

• At a zero input voltage, we get no output voltage.

Class A

Power Conversion Efficiency

CLASS A
Many class A amplifiers use the same transistor(s) for both halves of the audio
waveform. In this configuration, the output transistor(s) always has current
flowing through it, even if it has no audio signal (the output transistors never 'turn
off'). The current flowing through it is D.C.
A pure class 'A' amplifier is very inefficient and generally runs very hot even
when there is no audio output. The current flowing through the output
transistor(s) (with no audio signal) may be as much as the current which will be
driven through the speaker load at FULL audio output power. Many people
believe class 'A' amps to sound better than other configurations (and this may
have been true at some point in time) but a well designed amplifier won't have
any 'sound' and even the most critical 'ear' would be hard-pressed to tell one
design from another.
NOTE: Some class A amplifiers use complimentary (separate transistors for
positive and negative halves of the waveform) transistors for their output stage.
Class B CLASS 'B'
A class 'B' amplifier uses complimentary transistors
Circuit Operation
for each half of the waveform.
A true class 'B' amplifier is NOT generally used for
audio. In a class 'B' amplifier, there is a small part of
the waveform which will be distorted. You should
remember that it takes approximately .6 volts
(measured from base to emitter) to get a bipolar
transistor to start conducting. In a pure class 'B'
amplifier, the output transistors are not "biased" to an
'on' state of operation. This means that the the part
of the waveform which falls within this .6 volt window
will not be reproduced accurately.
The output transistors for each half of the waveform
(positive and negative) will each have a .6 volt area
in which they will not be conducting. The distorted
part of the waveform is called 'crossover' or 'notch'
Class B output stage. distortion. Remember that distortion is any unwanted
variation in a signal (compared to the original signal).
The diagram below shows what crossover distortion
looks like.
 Class B

Circuit Operation

Transfer characteristic for the class B output stage in Fig. 9.5.

 Operation
When the input voltage rises to be large enough
to overcome the Vbe, it will begin to cause an
output voltage to appear. This occurs because
Qn begins to act like an emitter follower and Qp
shuts off. The input will be followed on the
emitter until the transistor reaches saturation.
The maximum input voltage is equal to the
following:
v imax VCC − VCENsat

The same thing will begin to happen if the input voltage is negative
by more than the Veb of the transistor. This causes the Qp to act
like an emitter follower and Qn turns off. This will continue to
behave this way until saturation occurs at a minimum input voltage
of: v
imin −V + V
cc ECPsat
 Emitter Follower Configuration (Chapter 4)

vb ( β + 1) ( re + par( RL, ro) )

vs RS + ( β + 1) ( re + par( RL, ro ) )

vo (
par ro , RL )
vb (
re + par ro , RL )

Rs will be small for most

configurations, so the vb/vs will
Characteristics of the Emitter Follower: be a little less than unity. The
same is true for re, so vo/vb will
•High Input Resistance
be a little less than unity making
•Low Output Resistance our vo/vs a little less than unity.
•Near Unity Gain
Transfer Characteristic
 Push-Pull Nature of Class B

• Push: The npn transistor will push the current to ground

when the input is positive.

• Pull: The pnp transistor will pull the current from the
ground when the input is negative.
 Crossover Distortion

The Crossover Distortion is due to the dead band of input

voltages from -.5V to .5V. This causes the Class B output
stage to be a bad audio amplifier. For large input signals,
the crossover distortion is limited, but at small input
signals, it is most pronounced.
 Graph of Crossover Distortion

Illustrating how the dead band in the class B transfer characteristic results in crossover distortion.
 Power Efficiency
Load Power: 2 Since each transistor is only conducting for
1 Vop one-half of the time, the power drawn from
PL ⋅
2 RL each source will be the same.

1 Vop
2 Ps ⋅ ⋅ VCC
1 Vop π RL
⋅
PL 2 RL
η This efficiency will be at a max when
2⋅ Ps 1 Vop Vop is at a max. Since Vop cannot
2 ⋅ ⋅ VCC exceed Vcc, the maximum efficiency
π RL will occur at pi/4.

π Vop π This will be approximately 78.5%,

η ⋅ η max much greater than the 25% for
4 VCC 4
Class A.
Class AB

Circuit Operation
Crossover distortion can be eliminated by biasing the transistors at a small,
non-zero current.
A bias Voltage VBB is applied between Qn and Qp.
For vi = 0, vo = 0, and a voltage VBB/2 appears across the base-emitter junction
of each transistor.

VBB
2⋅VT
iN iP IQ IS⋅e

VBB is selected to result the required quiscent current IQ

VBB
vo vi + − v BEN
2

iN iP + iL
 iN   iP   iQ 
v BEN + v EBP VBB VT ⋅ln   + VT ⋅ln I  2⋅VT ⋅ln  
 IS   S  IS 
2 2
iN IQ

2 2
iN − iL⋅iN − IQ 0
Class AB

Output Resistance
Class AB
Calv in College - ENGR 332
Class AB Output Stage Amplifier
Exercise 9.6
Consider the class AB circuit (illustrated below ) w ith Vcc= 15 V, IQ= 2 m A, RL=100 ohm s.
Determ ine VBB. Determ ine the values of iL, iN, iP, vBEN, vEBP, vI, vO/vI, Rout, and vo/vi versus
vO for vO varying from -10 to 10V.
Note that vO/vI is the larg e sig nal voltag e g ain and vo/vi is the increm ental g ain obtained as
RL/(RL+ Rout). T he increm ental g ain is eq ual to the slope of the transfer curve.
Assum e QN and QP to be m atched, w ith IS= 10E-13.
Class AB

Exercise 9.6

under quiescent conditions iN=iP=IQ vO=vI=0

Solving for VBB
− 13 −3
VBB := 1 IS := 10 VT := 0.025 IQ := 2⋅ 10 RL := 100

Given VBB
2
VT
IQ IS⋅ e

VBB := Find( VBB) i := 0 .. 100 VBB = 1.186

Class AB

Exercise 9.6

vO
i i
vO := −10 + iL :=
i 5 i RL

iLi 0

10 0 10
vOi
Class AB Solving for iN

initial guesses iN := 0.02 iLD := 0.02 IQ := 0.002

Exercise 9.6 Given
2 2
iN − iLD⋅ iN − IQ 0

iNN( IQ, iLD) := Find( iN)

i := 0 .. 100 IQ := 0.002 iLD := iL

i i i

i (
iN := iNN IQ , iL
i i ) 3
1 .10
100
−5 10
iN = 4.997× 10 iN i ⋅ 1000
10
1
0.1
0.01
10 5 0 5 10
vOi
iP := iN − iLD
i i i 3
1 .10
100
10
iP i⋅ 1000
1
0.1
0.01
10 5 0
 iNi  vOi
vBEN := VT⋅ ln  
i  IS 
Class AB

Exercise 9.6 3
1 .10
100
10
iP i⋅ 1000
1
0.1
0.01
10 5 0
 iNi  vOi
vBEN := VT⋅ ln  
i
 IS 

0.6
vBENi

 iPi  0.5
vEBP := VT⋅ ln   10 5 0
i
 IS  vOi

vEBPi 0.6

10 5 0
Class AB
VBB
Exercise 9.6 vI := vO + vBEN −
i i i 2

0
vIi

vO 10
i 10 5 0
vOvI :=
i vI
i vOi

vOvIi 0.5

0
10 5 0
vOi
Class AB

Exercise 9.6

vOvI i 0.5

0
10 5 0
vO i
VT
Rout :=
i iP + iN
i i
5

Rout i

0
10 5 0 5 10
RL
vovi := vO i
i RL + Rout
i

vovii

0.95
Simplified internal circuit of the LM380 IC power amplifier (Courtesy National
Semiconductor Corporation.)
Small-signal analysis of the circuit in Fig. 9.30. The circled numbers indicate the order of
the analysis steps.
Structure of a power op amp. The circuit consists of an op amp followed by a class AB
buffer similar to that discussed in Section 9.7. The output current capability of the buffer,
consisting of Q1, Q2, Q3, and Q4, is further boosted by Q5 and Q6.
The bridge amplifier configuration.
Double-diffused vertical MOS transistor (DMOS).
Typical iD-vGS characteristic for a power MOSFET.
A class AB amplifier with MOS output transistors and BJT drivers. Resistor R3 is adjusted to
provide temperature compensation while R1 is adjusted to yield to the desired value of
quiescent current in the output transistors.