Professional Documents
Culture Documents
Diseño y Aplicación de Circuitos Electrónicos PDF
Diseño y Aplicación de Circuitos Electrónicos PDF
J. G. Gift
Brent Maundy
Electronic
Circuit Design
and Application
Electronic Circuit Design and Application
Stephan J. G. Gift • Brent Maundy
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
S.J.G. Gift dedicates the book to his parents, his wife Sandra,
and two sons Allister and Christopher.
Preface
Text Philosophy
This book treats with the design and application of a broad range of analog
electronic circuits in a comprehensive and clear manner. The discussion of
design and application in the text, we believe, brings out the inherently
interesting nature of electronics. There are several books on the market that
vii
viii Preface
deal extensively with analog electronic circuits, but most tend to give theory,
explanations, and analysis of circuit behavior and generally do not enable the
reader to design complete real-world functioning circuits or systems. This text
addresses this shortcoming while treating comprehensively with the subject. It
first provides a foundation in the theory and operation of basic electronic
devices including the diode, the bipolar junction transistor, the field effect
transistor, and the operational amplifier. It then presents detailed instruction
on the design of working real-world electronic circuits of varying levels of
complexity including feedback amplifiers, power amplifiers, regulated power
supplies, filters, oscillators, and waveform generators. Upon completion of the
book, the reader will understand the operation of and be able to confidently
design a broad range of functioning analog electronic circuits and systems.
With the proliferation of electronic systems today, the need for electronic
design engineers is greater than before, and this book we believe fills an
important niche in the world of analog electronics. It is intended primarily as
an undergraduate text for University and College students enrolled in electri-
cal, electronic, and computer engineering programs. It should also prove
useful for graduate students and practicing engineers who need to design
practical systems to meet research or real-world needs.
Text Features
This book enables the reader to analyze a variety of circuits, to develop a deep
understanding of their operation, and to design and optimize a range of
working circuits and systems. Many examples help the reader to quickly
become familiar with key design parameters and design methodology for
each class of circuits. Each chapter starts with fundamental ideas and develops
them step-by-step into a broad range of applications of real-world circuits and
systems. Each chapter ends with several circuit applications with a full
discussion of design methods. Also, at the end of each chapter, there are
research projects that are intended to stimulate the interests of the reader and
encourage varying levels of investigation and experiment.
The attractive features of the book include the following:
• Upon completion of the text, the reader will be able to confidently design
important analog electronic circuits.
Text Overview
Closing Remarks
We have prepared a text which we believe enables the reader to enjoy the
wonderful world of electronics while learning and applying design rules that
lead to practical working electronic systems. We have included many inter-
esting ideas and research projects throughout the text that will hopefully excite
the reader. We hope this book will satisfy most of the electronic circuit design
and application needs of students and practicing engineers and invite feedback
regarding its contents.
We would like to express our sincere thanks to our students who provided
useful feedback over the years while using this material in classroom instruc-
tion and to our Universities for providing the support that enabled the prepa-
ration of this text. We wish to also thank Mr Itanie Gordon for drawing most
of the diagrams in the text.
xi
Contents
1 Semiconductor Diode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Theory of Semiconductors . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Energy Levels . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.2 Intrinsic (Pure) Semiconductor . . . . . . . . . . . . . 4
1.1.3 Extrinsic (Impure) Semiconductor . . . . . . . . . . 5
1.2 Current Flow in Semiconductor Diodes . . . . . . . . . . . . . 7
1.2.1 Zero Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.2 Reverse-Biased Diode . . . . . . . . . . . . . . . . . . . 8
1.2.3 Forward-Biased Diode . . . . . . . . . . . . . . . . . . . 9
1.3 General Characteristic of a Diode . . . . . . . . . . . . . . . . . 9
1.3.1 Diode Specifications . . . . . . . . . . . . . . . . . . . . 12
1.4 Diode Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4.1 Zener Diodes . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4.2 Signal Diodes . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.3 Power Diodes . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.4 Varactor Diodes . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.5 Light-Emitting Diodes . . . . . . . . . . . . . . . . . . . 14
1.4.6 Photodiodes . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.4.7 PIN Diodes . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.4.8 Schottky Diodes . . . . . . . . . . . . . . . . . . . . . . . 16
1.5 Diode Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.5.1 DC Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.5.2 Clippers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.5.3 Clampers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.5.4 Half-Wave Rectifier . . . . . . . . . . . . . . . . . . . . . 22
1.5.5 Full-Wave Rectifier . . . . . . . . . . . . . . . . . . . . . 25
1.5.6 Bridge Rectifier . . . . . . . . . . . . . . . . . . . . . . . . 27
1.5.7 Zener Diode Regulators . . . . . . . . . . . . . . . . . . 27
1.6 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2 Bipolar Junction Transistor . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.1 Transistor Construction and Operation . . . . . . . . . . . . . . 41
2.2 Transistor Configurations . . . . . . . . . . . . . . . . . . . . . . . 43
2.2.1 Common Emitter Configuration . . . . . . . . . . . . 45
xiii
xiv Contents
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
10 Power Supplies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
10.1 Basic System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
10.2 Rectification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
10.3 Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
10.4 Average DC Output Voltage . . . . . . . . . . . . . . . . . . . . . 377
10.5 Bipolar Unregulated Power Supplies . . . . . . . . . . . . . . . 378
10.6 Voltage Multipliers . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
10.7 Voltage Regulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
10.7.1 Ripple and Regulation . . . . . . . . . . . . . . . . . . . 381
10.7.2 Zener Diode Regulator . . . . . . . . . . . . . . . . . . 382
10.7.3 Simple Series Transistor Regulator . . . . . . . . . . 384
10.7.4 Series Feedback Voltage Regulators . . . . . . . . . 387
10.7.5 Protection Circuits . . . . . . . . . . . . . . . . . . . . . . 394
10.7.6 IC Voltage Regulators . . . . . . . . . . . . . . . . . . . 396
10.7.7 Simple Approach to Regulated
Power Supplies . . . . . . . . . . . . . . . . . . . . . . . . 399
10.8 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
11 Active Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
11.1 Introduction to Filters . . . . . . . . . . . . . . . . . . . . . . . . . . 411
11.2 Basic First-Order LP Filter . . . . . . . . . . . . . . . . . . . . . . 412
11.2.1 Low-Pass Filter with Gain . . . . . . . . . . . . . . . . 414
11.3 Low-Pass Second-Order Filter . . . . . . . . . . . . . . . . . . . 416
11.3.1 Sallen-Key or Voltage-Controlled
Voltage Source (VCVS) Topology . . . . . . . . . . 417
11.3.2 Low-Pass Multiple Feedback Topology . . . . . . 422
11.4 Higher-Order Low-Pass Filters . . . . . . . . . . . . . . . . . . . 423
11.4.1 Third-Order Low-Pass Unity-Gain Filter . . . . . . 424
11.5 High-Pass First-Order Filter-Butterworth Response . . . . 428
11.5.1 First-Order High-Pass Filter with Gain . . . . . . . 429
11.6 High-Pass Second-Order Filter . . . . . . . . . . . . . . . . . . . 431
11.6.1 Sallen-Key or Voltage-Controlled
Voltage Source (VCVS) Topology . . . . . . . . . . 431
11.6.2 High-Pass Second-Order Multiple
Feedback Filter . . . . . . . . . . . . . . . . . . . . . . . . 435
11.7 Higher-Order High-Pass Filters . . . . . . . . . . . . . . . . . . . 436
11.7.1 Third-Order High-Pass Unity-Gain Filter . . . . . 436
11.7.2 Band-Pass Filter . . . . . . . . . . . . . . . . . . . . . . . 438
11.7.3 Sallen-Key Band-Pass Filter . . . . . . . . . . . . . . . 440
11.7.4 Multiple Feedback Band-Pass Filter . . . . . . . . . 441
11.7.5 Wien Band-Pass Filter . . . . . . . . . . . . . . . . . . . 442
11.8 Band-Stop Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
11.8.1 Twin-T Notch Filter . . . . . . . . . . . . . . . . . . . . 443
Contents xix
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579
Author Biographies
Stephan J. G. Gift graduated with BSc (First Class Honours) and PhD
degrees in Electrical Engineering from the University of the West Indies in
Trinidad and Tobago. He is Professor of Electrical Engineering and former
Head of the Department of Electrical and Computer Engineering and Dean of
the Faculty of Engineering at the University of the West Indies. He is currently
Pro Vice Chancellor of Graduate Studies and Research at this University. He
has published many papers in electronic circuit design and has performed
reviews for many international journals. He is a senior member of the IEEE
and a past president and Fellow of the Association of Professional Engineers
of Trinidad and Tobago.
xxi
Semiconductor Diode
1
The electrons in the outermost shell, referred conduction, and such materials are referred to as
to as valence electrons, determine the chemical conductors. The electric attraction between the
properties of the element. This number is equal to electron cloud of conduction electrons and the
the number of the group in the periodic table of atomic nuclei in the structure constitutes the
elements to which the atom belongs. Thus, lith- metallic bond. Examples of conductors include
ium with atomic 3, sodium with atomic number the metals such as silver, gold, copper, zinc, alu-
11 and potassium with atomic number 19 all have minium and lead in which the valence electrons
one valence electron and therefore belong to are freely available for conduction.
Group I in the periodic table. Where the conduction band and the valence
band do not overlap as in Fig. 1.4, an energy gap
or forbidden band exists between them. In this
1.1.1 Energy Levels case, a discrete amount of energy is required to
move the valence electron from its band in the
In a material, the valence electrons are bound in a structure into the conduction band. If the gap is
crystalline structure. However, it may be possible narrow (requiring a relatively small amount of
in some materials if sufficient energy is applied energy to overcome atomic attraction as provided,
for these electrons to break free of these inter- e.g. by ambient light and heat), then the material
atomic or inter-molecular bonds and be capable of in which this occurs is called a semiconductor.
moving about in a conduction mode. Thus, these Examples of semiconductors and the required
electrons can move from the valence band where energy for the generation of free carriers are ger-
they are bound to the conductions band where manium with gap energy Eg ¼ 0.67 eV, silicon
they are free. These two bands may or may not with Eg ¼ 1.11 eV, gallium (III) arsenide (GaAs)
overlap. with Eg ¼ 1.43 eV and cadmium sulphide (CdS)
When the bands overlap as in Fig. 1.3, there is with Eg ¼ 1.73 eV. Some electrons are therefore
no energy gap between the valence electrons and available for conduction through excitation
the conduction or free electrons, and hence in though not as many as in conductors.
materials where this occurs, the valence electrons Where the conduction band and the valence
are loosely bound to the nucleus and can freely band are widely separated as in Fig. 1.5, the
move about the structure. This movement of energy gap is large with Eg 1. An extremely
valence/conduction electrons, usually under the large amount of energy is required to move the
influence of an applied electric potential, is
Energy
Energy
Conduction Band
Conduction Band
} Region of Overlap
Eg
} Energy Gap
Fig. 1.3 Overlap of valence and conduction bands in Fig. 1.4 Separation of valence and conduction bands in a
conductors semiconductor
4 1 Semiconductor Diode
Energy
Si Si Si
Conduction Band
electrons
forming
bond
Eg
} Energy Gap Si Si
valence
Si
electrons
Valence Band
Si Si Si
electrons each having a negative charge e. In the electrons, and the holes behave as positively
solid-state structure, silicon and germanium form a charged particles. Holes and electrons move
lattice in which all atoms are equidistant from their through a crystal by either diffusion or drift.
immediate neighbours as shown in Fig. 1.6. Each Under normal circumstances, thermal energy
atom having four neighbours meets this equidistant causes random electron movement in a semicon-
requirement. In the crystal lattice, each atom forms ductor with no net flow of charge. However, if an
a covalent bond with its four neighbours, each increased concentration of electron or holes occurs
bond involving two valence electrons, one from at one end of the semiconductor by some mecha-
each atom. Thus, each bond pair traverses an orbit nism, this gives rise to a net flow referred to as a
around both the parent atom and the neighbouring DIFFUSION current away from that area. DRIFT
atom, thus enabling each atom to have effectively a occurs when there is net charge movement under
stable state of eight electrons in its valence shell. the action of an externally applied electric field.
This arrangement provides the lattice with a The velocity of the movement is called DRIFT
strong, stable, regular structure. VELOCITY, and this movement of holes and
Despite the strength of the bonds, energy from electrons through the crystal lattice is referred to
ambient light and heat is sufficient to overcome as DRIFT CURRENT. The electrons travel faster
the energy gap Eg and to cause some of the bonds than the holes since while the electrons undergo
to be broken and the electrons to be free. When a direct translation due to the field, the movement
covalent bond is broken, a hole-electron pair is of the holes in the opposite direction of the
created, and both the hole and the free electron are electrons involves a series of discontinuous elec-
known as charge carriers. They can thereafter tron movements into and out of covalent bonds.
move under the action of an applied electric Conduction of current in a pure semiconductor is
field and contribute to electrical conduction. If known as intrinsic conduction. Since the charge
the temperature of the material increases, thermal carriers are thermally generated, the concentration
energy causes the breaking of additional covalent of these carriers increases with temperature. As a
bonds and the consequent release of additional result the conductivity of intrinsic semiconductors
electrons. As a result the conductivity of the crys- increases with temperature at the approximate rate
tal increases. This means that as indicated earlier, of 5% per degree Celsius for germanium and 7%
a semiconductor material has a negative tempera- per degree Celsius for silicon.
ture coefficient of resistance.
When an electron escapes from a covalent
bond, there is an absence of an electron, which, 1.1.3 Extrinsic (Impure)
since this constitutes a missing negative charge Semiconductor
e, is equivalent to a positive charge e+. Such an
area in the crystal lattice is called a HOLE. A hole The electrical characteristics of intrinsic semicon-
exerts an attractive force on an electron, and it can ductor material can be significantly changed by
be filled by an electron that has been thermally the introduction of a small number (1 in 107) of
liberated form a covalent bond. This process is certain atoms of impurity into the semiconductor
called RECOMBINATION, and it causes a con- material. This process of introducing impurity
tinual loss of holes and free electrons. At a given atoms into pure semiconductor crystal is called
temperature, a dynamic equilibrium is reached DOPING. A treated crystal is said to be DOPED
such that the rate of generation of electrons and and is called EXTRINSIC material. Since the
holes is equal to the rate of recombination, this number of impurity atoms is very much smaller
resulting in a constant number of holes and free than the number of atoms of the crystal, the crys-
electrons. Both the hole and the electron are tal lattice is essentially undisturbed, and four
known as charge carriers, though the electron crystal atoms surround each impurity atom. Two
conduction (electron current) is different from types of extrinsic materials of great importance in
hole conduction (hole current). The motion of semiconductor device fabrication are n-type and
holes is in a direction opposite to the motion of p-type. Both types are formed by the addition of a
6 1 Semiconductor Diode
specific amount of impurity atoms into a silicon the number of the free electrons in the lattice greatly
or germanium base. exceeds the number of holes, and therefore the
crystal is called n-type. The impurity atom is called
N-Type Material a DONOR atom because it donates a free electron
N-type material shown in Fig. 1.7 is formed by to the crystal lattice. In n-type material, electrons
the addition of impurity atoms of elements from are the majority charge carriers, and holes are the
Group 5 of the periodic table which have five minority charge carriers. The above discussion also
valence electrons. Examples of these elements holds for a germanium crystal lattice. Note that the
are arsenic, antimony and phosphorus. In a silicon crystal is still neutral overall. The effect of the
crystal, each of the impurity atoms, for example, doping process on the crystal conductivity can be
phosphorus, will take the place of one of the described using the energy band diagram. As a
silicon atoms in the crystal lattice, and because result of the doping, a donor energy level appears
the number of impurity atoms is small, the crystal in the forbidden band with gap energy Eg signifi-
lattice will be essentially undisturbed. As a result, cantly less than that possessed by the intrinsic
four silicon atoms surround each impurity atom, material. Thermal energy is sufficient at room tem-
and each phosphorus atom will establish covalent perature to move these electrons into the conduc-
bonds with the four neighbouring silicon atoms. tion bond, thereby increasing the conductivity of
This means that one valence electron in the phos- the semiconductor. At room temperature in intrin-
phorus atom is unused. This surplus electron is sic silicon, there is approximately one free electron
only loosely bound to its parent atom. It easily for every 1012 atoms (1 to 107 for germanium).
breaks free and is available for conduction. The For a doping level of 1 donor atom in every 107
impurity atom has therefore donated a conduction silicon atoms, the ratio 1012:107 indicates an
election to the crystal lattice. A free electron is increase in free electrons by a ratio 100,000:1.
created in the silicon crystal lattice without the
creation of a corresponding hole. (A positive P-Type Material
charge does remain in the parent atom, but since The p-type material shown in Fig. 1.8 is formed
no bond is broken, the attraction of this charge for by the addition of impurity atoms of elements
free electrons is not as a great as when a bond is from Group 3 instead of Group 5 of the periodic
broken.) Hole-electron pairs are however still pro- table, these having three valence electrons.
duced by thermal energy. Because of the impurity, Examples of these elements are boron, gallium
Si Si Si Si Si Si
Fifth valence
electron of missing
phosporus electron
Si Ph Si Si B Si
boron
Phosphorus impurity
Impurity
Si Si Si Si Si Si
Fig. 1.7 Formation of n-type material Fig. 1.8 Formation of p-type material
1.2 Current Flow in Semiconductor Diodes 7
and indium. When boron impurity atoms are pointing in the direction p to n. The terminal on
introduced into the silicon crystal lattice, each the p side is referred to as the anode, while that on
boron atom will again take the place of a silicon the n side is referred to as the cathode. An external
atom in the lattice with its three valence electrons. voltage applied to the two terminals is referred to
It will form covalent bonds with three of the four as a bias. Below we consider the situations of zero
neighbouring silicon atoms. Since only three bias (where the external voltage is zero), reverse
bonds are formed, one hole (position for bond bias (where the external voltage is applied such
formation) is introduced into the lattice for each that the anode is negative relative to the cathode)
impurity atom. This hole is free to conduct as and forward bias (where the external voltage is
thermally generated holes, which continue to be such that the anode is positive relative to the
produced. Because of the impurity, the number of cathode).
free holes is the crystal lattice is much greater
than the electrons, and hence the material is
called p-type. The impurity atoms are called 1.2.1 Zero Bias
ACCEPTOR atoms. Again the p-type material is
electrically neutral. HOLES are the majority Consider the case where the externally applied
charge carriers, and electrons are the minority voltage is zero as shown in Fig. 1.11. In both
charge carriers in p-type semiconductor material. regions, there is a high probability that minority
charge carriers will meet and recombine with
majority charge carriers and as a result the life-
time of the minority charge carriers is short.
1.2 Current Flow in Semiconductor
While there is random movement of free electrons
Diodes
and holes in the lattice, the initial high concentra-
tion of electrons in the n-type and holes in the
Consider p-type and n-type semiconductor
p-type results in a net movement of electrons
materials that are brought into contact, forming
from n-type to p-type and holes from p-type to
a junction between them as shown in Fig. 1.9.
n-type. This process is known as DIFFUSION
Recall that both regions contain charge carriers of
and indicates the tendency for charge carriers to
either sign though in the n-type region electrons
move away from areas of high concentration to
are the majority carriers and in the p-type region
lower concentration areas.
holes are the majority carriers. Such an arrange-
As the n-type region loses negative charge
ment of p-type and n-type material in contact is
carriers and gains positive charge carriers and
referred to as a semiconductor diode. The symbol
the p-type region loses positive charge carriers
for the diode is shown in Fig. 1.10, with the arrow
and gains negative charge carriers, the n-type
region close to the junction becomes depleted of
electrons and positively charged, and the p-type
region close to the junction becomes depleted of
holes and negatively charged. The region in the
vicinity of the junction therefore becomes
Fig. 1.9 P-type and n-type material in contact forming a depleted of majority carriers and is called the
semiconductor diode DEPLETION REGION. It has a relatively
high resistivity and is of the order of 0.001 mm
Cathode wide. The migration also causes a build-up of
positive charges in the n-type region close to the
junction and negative charges in the p-type close
to the junction. This results in a potential across
Anode the junction called a BARRIER POTENTIAL.
It is directed such that further majority change
Fig. 1.10 Symbol of the diode diffusion across the junction is reduced though
8 1 Semiconductor Diode
majority carriers with sufficient energy are able 1.2.2 Reverse-Biased Diode
to overcome the barrier potential and migrate to
the other side. Minority charge carriers The reverse bias condition exists when an exter-
(electrons in the p-type material and holes in nal voltage VD is applied across the p-n junction
the n-type material) that enter the depletion with the positive potential being connected to the
region are swept across the junction to the n-type material and the negative potential being
other side by the action of the barrier potential. connected to the p-type material. This is shown
A dynamic equilibrium is reached such that the in Fig. 1.12. Under the action of this potential,
majority charge carrier or DIFFUSION current majority charge carriers are attracted away from
and minority charge currents are equal and hence the junction, thereby further depleting this
the net current across the junction is zero. It region. This increases both the magnitude of
should be noted that the crystal is electrically the potential barrier and the width of the deple-
neutral both before and after diffusion. The net tion region. Fewer majority carriers have suffi-
flow of charge in any direction for a semicon- cient energy to break through the barrier
ductor diode is zero, when the externally applied potential, and hence the majority charge current
voltage is zero. decreases eventually going to zero with
1.3 General Characteristic of a Diode 9
increasing reverse bias. The minority charge cur- n-type material towards the junction. This move-
rent, which was aided by the barrier potential, ment of holes and electrons towards the junction
increases quickly reaching its maximum value results in a reduction of the width of the depletion
where all the minority charges which are thermally layer and consequently the magnitude of the
generated are swept across the junction. This cur- potential barrier.
rent is called the reverse saturation current and is Majority carriers from each material can now
indicated as flowing from n to p in Fig. 1.12, cross the junction under the action of the electric
according to conventional current flow. (Note field. Even though the minority charge current
that this current comprises holes flowing from n flow remained constant, the increase in majority
to p and electrons flowing from p to n.) With a high current flow results in a net increase in conven-
enough reverse bias, the current flowing across the tional current across the junction from p-type
junction rises to a maximum equal to the minority material to n-type material (holes from p to n
charge current or REVERSE SATURATION and electrons from n to p). This current increases
CURRENT. The reverse saturation current for rapidly with increase in the forward bias VD
germanium is typically 1–2 μA at room tempera- according to an exponential curve as shown in
ture, while that for silicon is much lower at Fig. 1.14. The curve shows a sharp corner at
10–20 nA. This current approximately doubles in about VD ¼ 0.7 V for silicon and VD ¼ 0.3 V
value for every 10 C rise in temperature. for germanium. Below this voltage the current
through the diode is quite small, while above
this voltage, there is large current flow. This volt-
1.2.3 Forward-Biased Diode age can therefore be viewed as a threshold voltage
above which the diode is on and VD is constant
The forward-biased condition exists when an and below which the diode is off.
external voltage VD is applied across the p-n
junction with the negative potential being
connected to the n-type material and the positive 1.3 General Characteristic
potential being connected to the p-type material. of a Diode
This is shown in Fig. 1.13. Under these
conditions, holes (majority carriers) are repelled From solid-state physics, the relationship
from the p-type material towards the junction, and between diode voltage VD and diode current ID
electrons (majority carriers) are repelled from the is given by
18
16
14
12
10
2
No Bias
qV D VD
I D ¼ I o e mkT 1 ð1:1Þ I D ¼ I o emV T ð1:4Þ
18
16
14
12
10
4 ΔID
2
No Bias
ΔVD
-40 -30 -20 -10 0 0.5 1
VD(V)
0.3 0.7
-0.1µA
-0.2µA
Reverse Bias Region
VD
conductor (contact resistance). We use the desig-
dI D I emV T 1
¼ o ¼ ðI þ I o Þ nation rF to denote the diode forward resistance
dV D mV T mV T D
comprising rD, body resistance and contact
I
D , ID Io ð1:5Þ resistance.
mV T
Breakdown Region
Hence, If the reverse bias voltage across a diode is
increased beyond a certain value, the reverse cur-
dV D mV T 26 mV
rd ¼ ¼ ¼ ð1:6Þ rent increases rapidly, and the junction is said to
dI D ID I D ðmAÞ have broken down as shown in Fig. 1.16. This
critical voltage is called the BREAKDOWN
Example 1.1 VOLTAGE of the junction. Two effects cause
Determine the dynamic resistance of a this breakdown:
(a) The ZENER EFFECT occurs when the
semiconductor diode for 1 mA current through
electric field across the junction is strong
the diode.
enough to break covalent bonds in the lattice
and thereby generate carriers. This effect
Solution Using Eq. (1.6), the dynamic resistance
occurs primarily at low levels of breakdown
for a 1 mA current flow is given by r D ¼
voltage or ZENER VOLTAGE VZ
1 mA ¼ 26 Ω.
26 mV
corresponding to high doping levels in the
p- and n-type materials.
In addition to the dynamic resistance, there are (b) The AVALANCH EFFECT occurs when
resistance of the semiconductor material of the the charge carriers are accelerated by the
diode and resistance between the diode material electric field of the reverse bias to the extent
(body resistance) and the external metallic where they create carriers by collision.
12 1 Semiconductor Diode
10
1.3.1 Diode Specifications
8
Fig. 1.16 Diode characteristic showing Zener region Diodes are often rated according to their power
handling capability. The physical construction of
the diode determines its rating, and this informa-
tion is included in the manufacturer’s specifica-
For voltages less than about 5 volts, the Zener tion. Another diode rating is the current carrying
breakdown mechanism prevails, while for capacity or forward current. The instantaneous
voltages above 5 volts, the avalanche breakdown power dissipated by a diode is given by the rela-
mechanism is dominant. While the Zener break- tion PD ¼ IDVD.
down mechanism is a significant contributor to
the breakdown process only at low levels of VZ,
the breakdown region is called the ZENER 1.4 Diode Types
region, and diodes designed to operate in this
part of the p-n junction characteristic are called There is a variety of diodes available for
ZENER DIODES (see Sect. 1.4.1). Such diodes electronic circuit design and application. These
exploit the steep characteristic gradient to main- include Zener diodes, signal diodes, power
tain a constant voltage even with changing diodes, varactor diodes, light-emitting diodes,
reverse diode current. In general, the Zener region photodiodes, PIN diodes and Schottky diodes.
of the semiconductor diode must be avoided if the Each of these is briefly discussed in what
diode is not to be destroyed. For Zener diodes follows.
designed to operate in this region, current
limiting must be introduced (usually by the inclu-
sion of a resistor) in order to prevent diode failure. 1.4.1 Zener Diodes
The maximum reverse bias potential that can be
applied before breakdown is called the peak The Zener diode is a diode designed to be
operated in the Zener region as shown in
Fig. 1.18. Its symbol is shown in Fig. 1.17, and
1.4 Diode Types 13
Example 1.2
IZ(max) A Zener diode has VZ ¼ 8.2 V at 25oC. If the
diode has TC ¼ 0.05%/oC, determine the Zener
Fig. 1.18 Diode characteristics showing Zener operating voltage at 75oC.
region
Solution Using equation (1.7), ΔV Z ¼ VZ
100 T C
it exploits the steep slope of the diode character- 8:2
istic in the reverse bias region. As can be seen ðT 1 T 0 Þ ¼ 100 0:05 ð75 25Þ ¼ 0:205 V:
from the diode characteristic in Fig. 1.18, the
voltage in this region is almost constant for vary- Therefore VZ at 75 C is VZ (75 C) ¼ VZ + 0.205 ¼
o o
ing current. The actual Zener voltage VZ is deter- 8.2 + 0.205 ¼ 8.205 V.
mined by the doping levels in the Zener diode
with increased doping resulting in a reduction of In operating the Zener diode, the reverse cur-
VZ. Zener diode voltages VZ vary from 1.8 volts to rent through the device must not fall below a
about 200 volts with a tolerance of about 5% minimum value IZmn as shown in Fig. 1.16 in
and power ratings from ¼ W to about 50 W. order that the Zener voltage does not fall below
Silicon is usually preferred in the manufacture its operating value VZ. Also, the reverse current
of Zener diodes due to its higher current and must not exceed the maximum value IZmx set by
temperature capability. the Zener power rating.
In the on state, Zener diodes have a low
dynamic resistance rZ which is the reciprocal of Example 1.3
the slope of the Zener curve given by rZ ¼ ΔVZ/ If the 8.2 V Zener used in Example 1.2 has a
ΔIZ, and the equivalent circuit may be shown as a power rating of 500 mW, determine the maxi-
battery of voltage VZ in series with rZ. In most mum reverse current IZmx.
applications, however, rZ is quite small as com-
pared with other circuit resistance and hence can
14 1 Semiconductor Diode
CURRENT
Fig. 1.25 Symbol for PIN diode
SATURATION
CURRENT
0 VOLTAGE
2
Fig. 1.26 Symbol for Schottky diode
increases. The symbol for this diode is shown in In the circuit shown in Fig. 1.27, find the current
Fig. 1.26. I flowing through the resistor.
1.5.2 Clippers
1.5.1 DC Circuits
There is a class of diode networks called clippers
In general, a diode is in the “ON” state if the
that have the ability to “clip” off a portion of a
current established by the applied source is such
signal without affecting the remaining part of the
that its direction matches that of the arrow in the
alternating waveform. Such circuits are useful in
diode symbol. This generally corresponds to
protecting electronic equipment against voltage
VD 0.7 V for silicon and VD 0.3 V for
surges. The basic circuit of the clipper is shown
germanium. Analysing diode circuits with DC
in Fig. 1.28.
inputs is quite simple and involves applying stan-
As the input voltage goes more positive than
dard circuit laws.
the reference voltage Vref, the diode D1 turns
on. Current then flows from the input signal
Example 1.5
source through the diode and causes a voltage
drop across R1 such that the excess input voltage
is reduced. The result is that the output voltage is
limited to a maximum value equal to (Vref + VD).
For an input voltage that is less positive than Vref
or of negative polarity, diode D1 turns off, and the
full input signal waveform appears at the output
terminals of the circuit. Figure 1.29 shows the
clipped output waveform for an input sinusoidal
voltage waveform.
Fig. 1.27 Simple diode circuit
Time
Time
If the diode and the polarity of the reference maximum value on both half-cycles. The two
voltage are reversed, the input voltage signal reference voltages can be of different magnitude.
will be clipped on the negative rather than the If they are both set to zero as shown in Fig. 1.32,
positive half-cycle. If bi-directional diodes are the output amplitude is limited to 0.7 volts.
placed in parallel as shown in Fig. 1.30, then This circuit is used to protect some electronic
both the positive and the negative half-cycles circuits. The circuit can be further modified with
of the input waveform will be clipped as shown the inclusion of Zener diodes to replace the ref-
in Fig. 1.31. The result is that the amplitude of erence voltages.
the input signal waveform can be restricted to a
1.5 Diode Circuits 19
Example 1.7
Sketch the output waveform of the circuit shown
in Fig. 1.35. The input is a symmetrical square
wave input signal of amplitude 12 volts.
Fig. 1.34 Solution waveform for Example 1.6 The output voltage Vo is then Vo ¼ 6.8 + 0.7 ¼
7.5 volts. When the input voltage goes to
Example 1.6 12 volts, the normal diode turns off, and hence
Sketch the output waveform of the circuit shown so does the Zener. The output voltage is then
in Fig. 1.33. The input is a symmetrical square 12 volts. The output waveform is shown
wave input signal of amplitude 12 volts. Fig. 1.36.
Solution For the circuit in Fig. 1.33, as the input Example 1.8
voltage goes to +12 V, the forward voltage of the Sketch the output waveform of the circuit shown
diode and the battery voltage are both exceeded, in Fig. 1.37. The input is a sine wave input signal
and conduction through the diode occurs. The of peak amplitude 10 volts.
output voltage Vo is then Vo ¼ 3 + 0.7 ¼ 3.7 V.
When the input voltage goes to 12 V, the diode Solution For the circuit in Fig. 1.37, on the pos-
turns off, and hence there is no conduction. The itive half-cycle as the input voltage amplitude
output voltage is then 12 volts. The output increases, the forward voltage of the Zener
waveform is shown Fig. 1.34. diode is exceeded, and conduction through the
device occurs. The peak output voltage Vo is
20 1 Semiconductor Diode
1.5.3 Clampers
-4.7
-10
Vin – 0.7
Solution For the circuit in Fig. 1.44, as the anoutputvoltageofVo ¼ 127.3¼ 19.3volts.
input voltage goes to +12 volts, the capacitor The effect is that the output waveform is shifted
charges to VC ¼ 12 4 0.7 ¼ 7.3 volts negative by 7.3 volts while still preserving the
through the forward-biased diode. The output original waveform shape (Fig. 1.45).
voltage Vo is across the series battery-diode con-
nection giving Vo ¼ 4.7 volts. When the input
voltage goes to 12 volts, the normal diode
turns off, and the input voltage is in series with
the voltage across the capacitor. This results in
1.5 Diode Circuits 23
VS
VDC
A
B ΔV
VDC
ID Large
current
pulse
ΔtD t
Δt
T
pffiffiffi
Such a waveform is only suitable for applications V DC ¼ 2V S . Since the capacitor voltage adds to
such as battery charging where only the unidirec- the transformer voltage when the diode turns off, the
tional nature of the voltage is important. In most PIV is increased
pffiffiffi to twice the peak secondary volt-
equipment operated from such a supply however, age, i.e. 2 2V S . A large load current causes C to
the voltage variations will appear as hum and noise discharge more rapidly, and this results in a greater
at the output of the equipment. In general, the DC variation in the output voltage. While increasing
output of a rectifier is required to have little varia- C will reduce this variation, the maximum value of
tion and a minimum of ripple. capacitance that can be employed is limited since as
One method of achieving this is to connect a the capacitance value is increased, an increasing
reservoir capacitor C across the output of the current is required to recharge the capacitor. This
rectifier. The resulting supply shown in means that the amplitude of the current pulses in
Fig. 1.48 is referred to as half-wave unregulated Fig. 1.49 will increase and appropriately rated
DC power supply. Each time VS goes sufficiently diodes will have to be employed to handle this larger
positive, the diode conducts, and the voltage pffiffiffi forward current. The fluctuating unidirectional volt-
across the capacitor increases to V pk ¼ 2V S . age appearing across the load may be regarded as a
This corresponds to point A in Fig. 1.49. When DC voltage with an AC voltage component. This
VS falls below the capacitor voltage, the diode AC voltage is known as the ripple voltage.
turns off, and the capacitor discharges into the In order to calculate the peak-to-peak value
load at a rate determined by the time constant ΔV of the output ripple and the average output
CRL seconds where RL is the load resistance. As voltage, certain approximations need to be made.
the discharge occurs, the capacitor voltage VDC is Specifically, from Fig. 1.49 for the half-wave
represented by the curve AB in Fig. 1.49. As VS rectifier, the diode is off for most of the period
rises above the capacitor voltage at point B, the T. During this time the capacitor delivers a load
voltage at the anode of the diode exceeds the current IDC and therefore discharges in the pro-
cathode voltage causing the diode to conduct. cess. The decay is exponential but can be
The capacitor is recharged and the cycle is approximated as a linear decay. Also, the load
repeated. The charging current ID through diode current is assumed to be approximately constant
D1 takes the form of large pulses as shown in during the discharge. For a capacitor, the relation-
Fig. 1.49, which the diode must be able to handle. ship I ¼ CdV/dt holds. This can be approximated
For small load currents, the capacitor discharge by I CΔV/Δt where ΔV is a small change in
between charging pulses is small, and hence the capacitor voltage and Δt is the associated small
average output voltage is only slightly less than
1.5 Diode Circuits 25
time interval over which this voltage change a factor of 1.5, i. e.; Irms(sec) ¼ 1.5IDC(max) ¼
occurs. For the half-wave rectifier, 1.5 A. Choose a diode with IAV IDC(max) ¼ 1 A
pffiffiffi
ΔV ΔV and PIV > V pk ¼ 12 2 ¼ 16:8 V: Capacitor C
I DC ¼ C ¼C ð1:8Þ is given by C ¼ fIΔV ¼ 602:5
1
¼ 6666 μF: A
Δt
DC
TH
rough indication of the output voltage under
where ΔV is the change in voltage of the reservoir full-load conditions is given by VDC ¼ Vpk
capacitor and TH is the period of the half-wave ΔV/2 ¼ 16.8 2.5/2 16.
rectified waveform. From this, the peak-peak out-
put ripple is given by
I DC T H I DC 1.5.5 Full-Wave Rectifier
ΔV ¼ ¼ ð1:9Þ
C Cf
where f is the frequency of the input voltage. In a full-wave rectifier, both half-cycles of the
From this, the average DC output voltage is input waveform are utilized to produce a unidi-
rectional load current. The basic circuit is shown
V DC ¼ V pk ΔV=2 ð1:10Þ in Fig. 1.50 and can be viewed as two half-wave
rectifiers connected in parallel across the load. It
utilizes a centre-tapped transformer having equal
voltages VS with diodes D1 and D2 connected to
Example 1.10 the load as shown. During the half-cycle of the
Design a half-wave unregulated power supply input waveform where point A is made positive
delivering 12 volts DC at 1 A with less than with respect to G and point B is made negative
2.5 V peak-to-peak ripple. with respect to G, diode D1 conducts, while D2 is
reversed-biased. As a result, current flows
Solution Since the peak transformer voltage is through A and D1 from the transformer winding
greater than the root mean square value by a into the load resulting in the polarity across the
factor of 1.414, it is reasonable to choose the load as shown (Fig. 1.50).
transformer voltage Vrms to be equal to the Similarly, when the input signal is such that
required DC output voltage, i.e. Vrms ¼ VDC ¼ point B is positive with respect to G and point A is
12 V. Also, to prevent transformer heating, the negative relative to G, diode D2 is now forward-
transformer secondary current rating should be biased, while D1 is reversed-biased, and so cur-
larger than the maximum load current by at least rent flows through B and D2 in the load in the
VDC
DV
VDC
ID
pffiffiffi
same direction as before. The waveform of the The peak load voltage is 2V S . The inclusion
output voltage is shown in Fig. 1.51. During the of a reservoir capacitor as shown in Fig. 1.52
conduction of either diode, the full transformer reduces the ripple as shown in Fig. 1.53. It
voltage is applied p
across
ffiffiffi the other, and therefore functions in the same way as in the half-wave
the PIV is again 2 2V S . rectifier. However, since two half-wave rectifiers
1.5 Diode Circuits 27
operating over alternate half-cycles are than the maximum load current by at least a factor
involved, the capacitor is recharged twice per of 1.5, i.e.Irms(sec) ¼ 1.5IDC(max) ¼ 1.5 A.
input cycle instead of one. The result is that this Choose a diode with IAV IDC(max) ¼ 1 A and
pffiffiffi
circuit has fewer ripples than the half-wave recti- PIV > V pk ¼ 15 2 ¼ 21:2 V . Capacitor C is
fier. It however requires a centre-tapped trans- given by C ¼ 2fI DCΔV ¼ 2601:5 ¼ μF . A rough
1
former and two diodes for its implementation.
indication of the output voltage under full-load
For a full-wave rectifier, the period TF of the
conditions is given by VDC ¼ Vpk ΔV/
rectified waveform is half that of the half-wave
2 ¼ 16.8 2.5/2 16.
rectifier and therefore TF ¼ 1/2f. Hence for the
full-wave rectifier, the peak-peak output ripple is
given by
1.5.6 Bridge Rectifier
I DC T F I DC
ΔV ¼ ¼ ð1:11Þ
C 2f Full-wave rectification can be accomplished by
an alternative circuit called a bridge rectifier
Example 1.11 shown in Fig. 1.54. The bridge rectifier circuit
Design a full-wave unregulated power supply uses four diodes in a bridge arrangement that
delivering 15 volts DC at 1 A with less than obviates the need for a centre-tapped transformer.
1.5 V peak-to-peak ripple. Use a centre-tapped During those half-cycles of the input that make
transformer. point A positive with respect to point B, diodes
D2 and D4 conduct, while D1 and D3 are
Solution Since the peak transformer voltage is non-conducting. Current therefore flows from
greater than the root mean square value by a point A to point B via D2, the load and D4.
factor of 1.414, it is reasonable to choose the When A is negative relative to B, D1 and D3
transformer voltage Vrms to be equal to the conduct, and D2 and D4 do not conduct. Current
required DC output voltage, i.e.Vrms¼VDC ¼ 15 V. then flows from point B to point A via D3, the
Also, to prevent transformer heating, the trans- load and D1. The resulting current flows in the
former secondary current rating should be larger same direction through the load resulting in a
The maximum load current ILmx is set by the Finally, the minimum load resistor that can be
minimum Zener current IZmn below which the connected to the regulator is given by RLmn ¼
Zener voltage begins to fall and is given by 5
¼ 25
5
¼ 200 Ω.
I Lmx
I Lmx ¼ I Zmx I Zmn ð1:14Þ
Variable Supply Voltage and Fixed Load
We now consider the case involving a fixed load
Example 1.12 and a variable supply voltage. In this situation, the
A voltage regulator is being designed using a varying input voltage will cause a corresponding
6.3 volt Zener diode that has a minimum current variation in the current through the associated
requirement of 5 mA and ½Watt power rating. resistor and hence the Zener diode. Note that
The system shown in Fig. 1.57 will be powered since VZ is constant, the load current does not
from a 22 volt supply, and the load is variable. experience any change. When the input voltage
Calculate the required series resistance and the falls, the current through the Zener also falls but
maximum current available from the regulator. must not be allowed to fall below the minimum
Zener current that enables it to function on the
Solution Using (1.12), the maximum steep part of the slope of the breakdown charac-
current that can be passed by the Zener is given teristic and thereby maintain a constant voltage.
by PD ¼ VZIZmx. This gives I Zmx ¼ VPDZ ¼ 0:5
6:3 ¼ This means that R must be selected such that the
79:4 mA: From (1.13), R ¼ 79:4 mA ¼ 198 Ω .
226:3 minimum (or greater) Zener current IZmn and the
Therefore the maximum available load current load current IL flow with the input voltage at its
is from (1.14) given by ILmx ¼ IZmx IZmn ¼ minimum Vimn. Thus
79.4 5 ¼ 74.4 mA. In practice, the need for a V imn V Z
safety margin will require R to be greater than R¼ ð1:15Þ
I Zmn þ I L
198 Ω such that the power dissipated in the Zener
is substantially less than the rated value. This will When the supply voltage is at its maximum,
of course result in the maximum load current the Zener current will also be at its maximum, and
being less than 74.4 mA. therefore the Zener power rating must exceed the
maximum dissipation that will occur, i.e.
Example 1.13
PD V Z I Z max ð1:16Þ
A 5 volt Zener diode with a minimum current
requirement of 10 mA is to be used in a voltage
regulator. The supply voltage is 18 volts, and the Example 1.14
variable load current has a maximum value of A 5 volt Zener diode has a minimum current
25 mA. Calculate the minimum power rating of requirement of 5 mA and is to be used in a voltage
the Zener diode, the required series resistance and regulator. The supply voltage is 25 volts 10%,
the minimum load resistance that can be and the fixed load current is 20 mA. Calculate the
connected to the regulator. series resistance required, the minimum power
rating of the Zener diode and the minimum
Solution Since the load current is variable, the instantaneous power dissipated in the Zener.
Zener must be able to accommodate the maxi-
mum load current (should the load resistor be Solution The minimum value of the input volt-
removed) along with the minimum Zener current. age is 0.9Vi ¼ 22.5 V. Therefore (1.15) gives
This gives a maximum Zener current of imn V Z
R ¼ VI Zmn þI L ¼ 5þ20 ¼ 25 mA ¼ 700 Ω. The max-
22:55 17:5
IZmx ¼ ILmx + IZmn ¼ 25 + 10 ¼ 35 mA. Hence
imum value of the input voltage is 1.1Vi ¼ 27.5 V.
the minimum power rating PDmn of the Zener is
This produces the maximum current IRmx through
PDmn ¼ VZIZmx ¼ 5 V 35 mA ¼ 175 mW. The
the series resistor given by I R max ¼ 27:55 ¼
series resistor is given by R ¼ 35 mA ¼ 371 Ω .
185 R
22:5
700 ¼ 32 mA. Since 20 mA flows into the fixed
30 1 Semiconductor Diode
supply. The minimum voltage across C1 is 1.3 Volt Supply from 5 Volts
16.3 1 ¼ 15.3 V. Allowing for transformer The circuit shown in Fig. 1.59 operates in the
resistance, we use 15 V. Since resistor R1 must manner of a Zener diode regulator. It provides
allow a total of 20 mA to flow, its value is given 1.3 volts from the widely available 5 volt supply
by R1 ¼ (15 9)/20 mA ¼ 300 Ω. The maximum as a replacement for small low-voltage cells. The
pffiffiffi
input voltage is 12 2 0:7 ¼ 16:3 V. Hence the resistor R1 limits the current in the green LED D1
maximum current through the Zener is to a few milliamps of say 10 mA which results in
(16.3 9)/300 Ω ¼ 24.3 mA. Hence the maxi- a diode voltage of about 2 volts across D1. This
mum power dissipation in the Zener is gives R1 ¼ (5 2)/10 mA ¼ 300 Ω. In order to
9 V 24.3 mA ¼ 218.7 mW. A 500 mW Zener reduce this 2 volts to the desired 1.3 volts, a signal
will provide good safety margin. This circuit is diode D2 such as the 1N4148 is included in the
inherently short-circuit protected because of R1. circuit as shown. A small capacitor C1 ¼ 10 μF is
Thus, in the event of a short circuit at the output, placed at the output in order to provide some
the full voltage of capacitor C1 will be applied basic filtering for noise signals above about
across R1 which will now be in parallel with C1. 1 kHz. At this frequency, the reactance of the
Under this condition, the maximum current capacitor is given by 1/2πfC ¼ 1/
through this resistor will be 16.3 V/ 2π 103 10 106 ¼ 16 Ω which short-
300 Ω ¼ 54.3 mA. The maximum power that circuits signals above this frequency.
will be dissipated in this resistor is therefore
16.3 V 54.3 mA ¼ 886 mW. A 2 W resistor 9 Volt Supply from 12 Volts
for R1 will provide an adequate safety margin. There are many portable devices that require
The LED D3 will indicate when the system is on 9 volts for their operation. The circuit in
and together with R2 will discharge capacitor C1 Fig. 1.60 converts the 12 volts available from
when the system is off. Using a green LED for D3 the cigarette lighter in an automobile to 9 volts.
and a LED current of 10 mA, resistor Each of the four silicon diodes D1 to D4 drops 0.7
R2 ¼ (15 2.1)/10 mA ¼ 1.3 k. Finally, capaci- volts resulting in an output of just over 9 volts.
tor C2 is used to remove high-frequency noise
from the supply output, and a value of 0.1 μF Ideas for Exploration (i) Modify the circuit in
will have a low reactance at frequencies of 10 kHz order to reduce the output of a 9 volt battery to
and above. 5 volts.
Battery Charger
A 12 volt battery charger circuit is shown in
1.6 Applications Fig. 1.61. The circuit involves a 24 volt centre-
tapped 5 ampere transformer operating into a full-
The circuits below are some simple applications wave rectifier whose output is filtered by a
of diodes in real-world systems that can be easily 2000 μF capacitor C. The final circuit output is
designed and implemented. delivered to the battery under charge through a
series 10 Ω variable power resistor VR1 and an
ammeter. At the start of charging, the variable
metre or lux) and voltage output is given by For example, for a current of 10 mA, an input
lux ¼ 1333Vo. For example, a voltage from 0 to voltage of 12 volts will require R1 ¼ (12 2)/
0.75 volts will correspond to light intensity of 10 mA ¼ 1 k. Here we have used an approximate
0–1000 lux. Hence, the voltmeter V connected LED voltage of 2 volts. A blown fuse
across the resistor can be calibrated to read light disconnects the supply from the load and the
intensity. green LED which goes off. This results in the
full supply voltage being applied across the red
Ideas for Exploration (i) Use a simple analog LED in series with the signal diode, thereby
multimeter as the voltmeter and calibrate the scale causing this LED to turn on.
to read light intensity in units of lux.
Diode Tester
Blown Fuse Indicator This circuit, shown in Fig. 1.65, enables the
The circuit in Fig. 1.64 indicates when a fuse is polarity of unmarked semiconductor diodes to
blown. During normal operation, the green LED be determined. It is powered by a low-voltage
is on, and the voltage across this device is insuf- transformer T1. Resistor R1 limits the current
ficient to turn on the red LED in series with a flowing through the diodes. When the diode
normal signal diode D1. Therefore, the red LED under test is connected with the anode at A and
remains off. The current through the green LED the cathode at B, it will allow current to flow
is limited by resistor R1 which can be of the order from A to B and through diodes D1 and D2
of kilo-ohms depending on the supply voltage. causing the green LED D1 to light. If the test
diode is connected in the reverse, then current
will flow from terminal B to A and through
Fig. 1.64 Blown fuse indicator Fig. 1.66 Emergency telephone line light
Ideas for Exploration (i) Use this circuit to half-cycle when terminal A is negative with
monitor the output voltage of the battery charger respect to terminal B, diode D2 turns on charging
pffiffiffi
circuit discussed earlier. capacitor C2 to 6 2 ¼ 8:5 V. Diode D1 turns off
during this time. The result is that the voltage
Zener Diode Regulator Using a Voltage pffiffiffi
across terminals C and D is 2 6 2 ¼ 17 V .
Doubler For a supply current of 25 mA, allowing a
The circuit in Fig. 1.69 is that of Zener diode minimum Zener current of 5 mA, then the total
voltage regulator that produces 12 volts DC at Zener current must be 30 mA. For 30 mA into the
25 mA from a 6 volt transformer. It operates 12 V Zener D3, allowing a 1 V voltage drop on
using two half-wave rectifiers whose outputs are each capacitor, each capacitor must be C ¼ I/
connected in series. Thus, during the half-cycle in fΔV ¼ 30 mA/60 1 ¼ 500 μF. With this 1 V
which the transformer terminal A is positive with voltage drop on each capacitor, the value of R1
respect to transformer terminal B, diode D1 turns that allows the flow of 30 mA into the 12 V Zener
pffiffiffi
on and charges capacitor C1 to 6 2 ¼ 8:5 V with diode D3 is given by R1 ¼ (17 1 1 12)/
the polarity shown. During this time diode D2 is 30 mA ¼ 100 Ω. The maximum current into the
reversed-biased and therefore off. During the Zener is (17 12)/100 ¼ 50 mA, and therefore
36 1 Semiconductor Diode
the maximum power dissipated in the Zener diode mA ¼ 218 Ω. For a 20 volt output,
is PD ¼ 12 V 50 mA ¼ 600 mW. Hence, a 1 W 20 ¼ (1 + R1/R2) 2.5. Using R2 ¼ 2 k, then
Zener is chosen for safe operation. R1 ¼ 14 k. Capacitor C2 ¼ 1 μF removes any
high-frequency noise and residual ripple.
Ideas for Exploration (i) Implement an LED
“ON” indicator at the output of the system using Ideas for Exploration (i) Re-design the circuit
an LED and a series resistor. (ii) Re-design the to allow for a higher Zener current. This enables
system to deliver 9 volts at the output by changing the system to supply more current to an
D3 and R1. external load.
passage of such a wave induces an electrical (i) The formation of n-type and p-type
signal in the antenna that flows into the material
inductor-capacitor tank circuit to ground. For a (ii) Increased conductivity in n-type and
signal whose frequency is at the resonant fre- p-type material
quency f ¼ 2π p1 ffiffiffiffi
LC
ffi of the LC circuit, there is reso- (iii) Why there is conduction across a
nance such that a large potential difference forward-biased p-n junction
develops across the tank circuit. This signal is (iv) Why there is little or no conduction
rectified by the germanium diode, and the carrier across a reverse-biased p-n junction
component of the signal is filtered out by the (v) What happens when n-type and p-type
capacitor associated with the piezoelectric ear- material are brought into contact
phone. The remaining signal is the original (vi) The effect of temperature on the con-
audio component, which drives the high imped- ductivity of intrinsic semiconductor
ance earphone. material
The coil and the variable capacitor must enable (vii) Why a conductor conducts better than a
changing the resonant frequency across the full semiconductor
medium wave band. A 365 pF variable capacitor (viii) Why a semiconductor conducts better
and a 300 μH inductor are suitable. The coil can than an insulator
be constructed using a ferrite rod with 100 turns (ix) The cause of the barrier potential in an
of no 26SWG (25AWG) enamelled copper wire. np junction
The germanium diode 1N34A is suitable for the 2. Determine the dynamic resistance of a semi-
detection stage as its turn-on voltage is only conductor diode for 0.5 mA current through
0.3 V. The piezoelectric earphone converts the the diode.
electrical signal to sound, and its high impedance 3. A Zener diode has VZ ¼ 6.8 V at 25oC. If the
ensures that frequency selectivity is not reduced. diode has TC ¼ 0.03%/oC, determine the Zener
The resistor R1 ’ 10 k to 47 k provides a DC path voltage at 63oC.
for the proper operation of the diode. For effective 4. An 10 V Zener has a power rating of 400 mW.
operation, the antenna should be about 30 m long, Determine the maximum reverse current IZmx.
about 7 m above ground and insulated from its 5. An LED is driven by a 9 volt battery in series
supports. with a resistor R1. Determine the value of R1 if
The circuit must also have a good ground to the LED is green.
the physical Earth using a solid copper rod driven 6. In the circuit of problem 5, if the LED is
into the ground. The circuit should be able to replaced by a silicon diode, what is the value
receive nearby AM stations, with difficulty of the current for the same value of R1?
being experienced with more distant ones. 7. Sketch the output waveform for each of the
circuits shown in Fig. 1.72. In each case Vi is a
Ideas for Exploration (i) Wind the antenna into symmetrical square wave input signal of
a coil of wire and see how that affects reception. amplitude 10 volts.
(ii) Replace the ferrite rod by a 1 inch former, 8. For the circuits in problem 7, sketch the output
thereby realizing an air-cored inductor. (iii) Exper- waveform for a sine wave input signal of
iment with coils with a reduced number of turns to amplitude 15 volts.
attempt short wave reception (1.7–30 MHz). 9. Sketch the following circuits:
(i) Full-wave rectifier using a centre-
tapped transformer
(ii) Full-wave bridge rectifier operating into
Problems a load
(iii) Clipping circuit using one Zener diode
1. Briefly explain the following:
38 1 Semiconductor Diode
(iv) Clamping circuit with a peak positive input is a symmetrical square wave input
output of +0.7 volts signal of amplitude 12 volts.
10. Sketch the output waveform for each of the 11. With the aid of a circuit diagram, explain the
circuits shown in Fig. 1.73. In each case the operation of a half-wave rectifier operating
Problems 39
into a resistive load. Draw a voltage/time The supply voltage is 24 volts and the load
diagram of the output waveform. Explain current is variable. Calculate the required
the effect of a filter capacitor being placed series resistance and the minimum load resis-
across the load. tor that can be connected to the supply.
12. With the aid of a circuit diagram, explain 19. A 5 volt Zener diode has a minimum current
the operation of a full-wave rectifier requirement of 10 mA and is to be used in a
operating into a resistive load. Draw a volt- voltage regulator. The supply voltage is
age/time diagram of the output waveform. 20 volts 10% and the fixed load current
Explain the effect of a filter capacitor being is 20 mA. Calculate the series resistance
placed across the load. required, the minimum power rating of the
13. Compare the advantages and disadvantages Zener diode and the minimum instanta-
of using a bridge rectifier with a normal neous power dissipated in the Zener.
transformer or two diodes and a centre- 20. A 12 volt Zener diode has a minimum current
tapped transformer to realize a full-wave requirement of 6 mA and is to be used in a
rectifier. voltage regulator. The supply voltage is
14. Design a half-wave unregulated power sup- 30 volts 5% and the fixed load current is
ply delivering 15 volts DC at 0.5 A with less 15 mA. Calculate the series resistance
than 2 V peak-to-peak ripple. required and the maximum instantaneous
15. Using a centre-tapped transformer, design a power dissipated in the Zener diode.
full-wave unregulated power supply deliv- 21. A 12 volt Zener diode with a minimum
ering 12 volts DC at 2 A with less than 1 V current requirement of 4 mA and a power
peak-to-peak ripple. rating of 600 mW is to be used in a voltage
16. Re-design the circuit in problem 16 using a regulator. The supply voltage is 20 volts
15 volt transformer and a bridge rectifier. and the load current is variable. Calculate
17. A 6 volt Zener diode with a minimum cur- the required series resistance and explain
rent requirement of 5 mA is to be used in a what occurs if the supply is short-circuited.
voltage regulator. The supply voltage is 22. A 9 volt Zener diode has a minimum current
22 volts and the variable load current has a requirement of 3 mA and is to be used in a
maximum value of 15 mA. Calculate the voltage regulator. The supply voltage is
required series resistance and the minimum 24 volts 5% and the variable load current
power rating of the Zener diode. has a maximum value of 18 mA. Calculate
18. A 16 volt Zener diode with a minimum cur- the series resistance required and the mini-
rent requirement of 5 mA and a power rating mum power rating of the Zener diode.
of 800 mW is to be used in a voltage regulator.
23. A 6 volt Zener diode with a 1 W power rating 28. The circuit shown in Fig. 1.74 is that of a
has a minimum current requirement of 5 mA mains wiring fault detector. Determine the
and is to be used in a voltage regulator. condition of the leds (on/off) for the follow-
The supply voltage is 18 volts 5% and the ing fault conditions: (i) live disconnected;
load current is variable. Calculate the series (ii) live and neutral interchanged; (iii)
resistance required and the maximum current ground disconnected; (iv) live and ground
that can be delivered by the supply. interchanged; (v) neutral disconnected; and
24. Design a Zener regulated power supply to (vi) all connections good.
deliver 12 volts at a current of 25 mA using 29. Resistor R1 and LED D1 in Fig. 1.75 consti-
an 18 volt transformer and a bridge rectifier. tute a blown fuse indicator for a power sup-
25. Using the basic configuration of Fig. 1.60, ply. Explain the operation of the arrangement
design a circuit to deliver (i) 3 V from a 5 V and determine a suitable value for the
DC supply and (ii) 6 V from a 12 V DC resistor.
supply. 30. Discuss how the circuit shown in Fig. 1.76
26. Determine how to arrange the components in can be used for testing fuses in an automobile
the polarity indicator circuit of Fig. 1.62 if without their removal and determine a suit-
only one LED is available. able value of R1.
27. Using the basic Zener regulator configuration
and a BZX55C2V0 2 volt Zener, design a
circuit that supplies 1.3 volts from a 5 V
supply that can replace 1.3 V mercury cells.
Use an 1N4148 silicon diode to effect the
voltage reduction from 2 V to 1.3 V.
Bipolar Junction Transistor
2
The bipolar junction transistor or BJT is a device p-type, n-type and p-type material in direct con-
capable of amplifying a voltage or current, some- tact is shown in Fig. 2.1b. In the n-p-n structure,
thing that diodes are not able to do. This the n-type material referred to as the emitter
amplifying characteristic makes the BJT suitable region of the transistor is heavily doped, while
for a wide range of applications. The device was the base and collector regions are lightly doped.
invented in 1947 by Walter H. Brattain, John In normal operation as an amplifier, the emitter-
Bardeen and William Shockley who were base junction is forward-biased by applying a
awarded the Nobel Prize in Physics in 1956 for potential VBE, and the collector-base junction is
this invention. It revolutionized the electronics reverse-biased by applying a potential VCB. For
industry by enabling miniaturization of electronic the n-p-n transistor, electrons (IE) are injected
circuits and increased equipment portability. This from the emitter into the base region because of
chapter discusses the characteristics of the BJT the forward bias across the base-emitter junction.
and its use in elementary amplifier circuits. At the Most diffuse across the base to the collector
end of it, the student will be able to: where they (IC) are collected by the reverse bias
potential across the collector-base junction. A few
• Explain the basic operation of a BJT electrons (IB) flow out to the base connection
• Describe the various configurations because of the positive potential there. This pro-
• Describe the operation of a basic BJT amplifier cess of carriers being generated in the emitter,
• Analyse single-stage transistor amplifiers traversing the base and being collected by the
• Design single-stage transistor amplifiers collector is referred to as transistor action. The
p-n-p transistor behaves similarly with holes
replacing electrons as the majority carriers. The
symbols for n-p-n and p-n-p transistors are shown
2.1 Transistor Construction in Fig. 2.2.
and Operation In the arrangement in Fig. 2.3, VBE forward-
biases the base-emitter junction as before, but
The bipolar junction transistor or BJT is basically instead of a voltage VCB directly across the
a two-diode structure consisting of either p-n-p or collector-base junction, a voltage VCE is applied
n-p-n. The former is called a p-n-p transistor, across the collector and emitter, thereby referring
while the latter is called an n-p-n transistor. The both potentials to a common reference at the
n-p-n structure involving n-type, p-type and emitter. VCE is chosen sufficiently larger than
n-type material in direct contact is shown in VBE to ensure that the collector-base junction is
Fig. 2.1a, while the p-n-p structure consisting of reverse-biased. This is called the common emitter
# Springer Nature Switzerland AG 2021 41
S. J. G. Gift, B. Maundy, Electronic Circuit Design and Application,
https://doi.org/10.1007/978-3-030-46989-4_2
42 2 Bipolar Junction Transistor
(a) (b)
Fig. 2.1 NPN and PNP transistors. (a) NPN transistor. (b) PNP transistor
collected by the collector, the collector current IC quite small relative to IC and, like the reverse
is very nearly equal to the emitter current IE, saturation current IO for a diode, is temperature
sensitive. Its effect can be quite critical in some
I C ¼ αI E ð2:3Þ
designs. However, in modern devices, ICO is often
where α 1. α is referred to as the common base sufficiently small that it can be ignored.
current gain. While the collector current IC
comprises mainly majority carriers from the emit-
ter, there is a small component that is made up of 2.2 Transistor Configurations
minority carriers. This minority carrier compo-
nent is called leakage current ICO. Its value is There are three transistor configurations, which
provide useful amplification. These are the
44 2 Bipolar Junction Transistor
Fig. 2.6 Transistor configurations. (a) Common emitter (CE). (b) Common base (CB). (c) Common Collector (CC)
common emitter, common base and common col- The common base configuration shown in
lector configurations. These configurations using Fig. 2.6b is so-called because the base terminal
npn transistors are shown in Fig. 2.6. with input is common to both the input and the output. It
voltage Vi and output voltage Vo. The common corresponds to the circuit in Fig. 2.1 used to
emitter configuration shown in Fig. 2.6a is in introduce transistor operation. The common col-
fact that considered in Fig. 2.3 in discussing tran- lector configuration shown in Fig. 2.6c has the
sistor action above. The name common emitter collector common to both the input and the output
derives from the fact that the emitter terminal through the supply VCC which is a short circuit
is common to both input and output circuits. for signals.
2.2 Transistor Configurations 45
2.2.1 Common Emitter Configuration curves involving current-voltage plots are avail-
able that may be used. These curves give infor-
Consider the common emitter configuration mation on the value of current flowing into or out
shown in Fig. 2.7. The input signal is applied at of one terminal in response to current into the
the base and the output signal taken at the collec- base terminal or voltage across the base-emitter
tor as shown. This configuration can be viewed as terminals. Three sets of characteristics can be
fundamental to transistor operation since the other plotted:
two configurations can be derived from this one (i) Input characteristics
using negative feedback. It is the most frequently (ii) Mutual and transfer characteristics
used configuration. (iii) Output characteristics
In the analysis and design of transistor circuits
using this configuration, static characteristic They can be experimentally determined using
the circuit shown in Fig. 2.8. This circuit
comprises a variable DC supply VCC that supplies
the collector-emitter circuit and a variable DC
supply VBB which supplies the base-emitter cir-
cuit. Voltmeter A measures the base-emitter volt-
age VBE, while voltmeter B measures the
collector-emitter voltage VCE. The milliammeter
measures the collector current IC, while the
microammeter measures the base current IB.
Each of three sets of characteristics will now be
considered.
60
1
40
20 0 10 20 30 IB (µA)
IB = 200µA
10 20
IB = 150µA
IB = 100µA
5 10
IB = 50µA
0 0.2 0.4 0.6 0.8 1.0 VBE (V)
voltage. The family of curves with base current Fig. 2.13 Common emitter output characteristics for
as the varying parameter is shown in Fig. 2.12. fixed values of VBE
The low value of the slope of these curves is yet
another indication of the limited effect of VCE on transistor, since, for a given value of collector-
IC. The slope at the approximately flat portion of emitter voltage VCE, the change in collector cur-
these curves represents the output admittance hoe rent ΔIC and the corresponding change in base
of the transistor and is given by current ΔIB can be obtained by projection on the
curves. The family of curves with base-emitter
ΔI C
hoe ¼ j ð2:9Þ voltage as the varying parameter is shown in
ΔV CE I B constant
Fig. 2.13.
where ΔIC and ΔVCE are small changes in IC and Output impedance as well as transconductance
VCE. The output impedance of the transistor in can be determined using these curves. The flat
this configuration is the reciprocal of hoe. These portion of the output curves are all asymptotic to
output characteristics can also be used to deter- straight lines, which intercept the collector-
mine the (short circuit) current gain hfe of the emitter voltage axis at approximately the same
48 2 Bipolar Junction Transistor
point as shown in Fig. 2.14. The magnitude of the emitter-base junction is forward-biased and the
collector-emitter voltage where this interception collector-base junction is reverse-biased, the cut-
occurs is called the early voltage VA. off region in which the emitter-base junction is
VA is typically in the range 50 < VA < 300 not forward-biased and the saturation region in
volts. It enables the output resistance of the tran- which the collector-base junction is not reverse-
sistor in this configuration to be determined by biased. As in the silicon diode, large movements
of minority carriers across the base-emitter junc-
1 V A þ V CEq
¼ ð2:10Þ tion will occur only if the emitter-base junction is
hoe I Cq
forward-biased. If therefore the base-emitter volt-
where VCEq and ICq are the quiescent voltage and age VBE < 0.7 V the threshold voltage, little or no
current, respectively. conduction takes place (only leakage current
The basic output characteristics shown in made up of minority carriers), and the transistor
Fig. 2.15 illustrate three regions of transistor is said to be cut-off. Thus for cut-off operation,
operation: the active region in which the VBE < 0.7 V (Si), IB ¼ 0 and IC ¼ ICEO. If
VBE 0.7 V, majority carriers are injected from
the emitter into the base region (where they
become minority carriers), and their subsequent
Ic
diffusion across the base into the collector
depends on having an adequate reverse bias on
the collector-base junction. If this junction is not
reverse-biased, injection of carriers into the base
VCE will take place from the collector, and under these
VA circumstances, the transistor is saturated. Thus,
for saturated operation, VBE 0.7 V (Si) and the
Fig. 2.14 Early voltage VA collector-base junction not reverse-biased. The
Saturaon Acve
Ic Region Region
(mA)
12 Ib = 120μA
Ib = 100μA
10
Ib = 80μA
8
Ib = 60μA
6
Ib = 40μA
4
Ib = 20μA
2
Ib = 0μA
0 VCE(V)
2 4 6 8 10 12
cut-off region
IE (mA) IC (mA)
10
VCB = 20 V VCB = 20 V
9
8 8
VCB = 10 V VCB = 10 V
7 7
VCB = 1 V 6 VCB = 1 V
6
5 5
4 4
3 3
2 2
1 1
IE = 7 mA
7
IE = 6 mA
6
IE = 5 mA
Saturaon region
5
IE = 4 mA
4
IE = 3 mA
3
IE = 2mA
2
IE = 1 mA
1
IE = 0 mA
-1 0 5 10 15 20 VCB (V)
Cutoff region
IE (mA)
10
9
8
7
6
5
4
3
2
1
0 10 20 30 40 50 IB (μA)
Fig. 2.21 Common collector amplifier
Fig. 2.22 Current transfer characteristic
Note that
ΔI E ΔI C þ ΔI B ΔV E
hfc ¼ ¼ ¼ hfe þ 1 ð2:21Þ AV ¼ j constant ð2:22Þ
ΔI B ΔI B ΔV B V ce
This characteristic is reasonably linear and is This value is approximately one with no volt-
independent of VCE. The voltage transfer charac- age amplification taking place. This characteristic
teristic shown in Fig. 2.23 indicates the manner in is even more linear than the current transfer char-
which the emitter voltage changes in response to acteristic and is attributable to negative feedback
the base voltage, the collector-emitter voltage to be discussed in Chap. 7. Also, the voltage
kept constant. The slope of the characteristic characteristic like the current characteristic is vir-
gives the short-circuit voltage gain AV defined as tually independent of changes in VCE.
2.3 Common Emitter Amplifier 53
0.7V
IC
Vi
IC
ICq
Fig. 2.27 No waveform distortion with biasing beyond the knee in Fig. 2.9. A practical biasing
scheme that overcomes these problems and the
selection of ICq and RL in order to achieve opti-
Vi mum performance as part of the design process is
discussed next.
Fixed Biasing
The basic scheme is shown in Fig. 2.29. It utilizes
the fact that there is no knee on the common
emitter transfer characteristic and control of IC
using IB is quite practical. This approach is
VC implemented in Fig. 2.29 where the bias voltage
is applied to the base-emitter junction via resistor
RB. The effect of this is to fix IB at some value.
VCq Thus, for the base circuit, applying Kirchhoff’s
voltage law
V BB ¼ I B RB þ V BE ð2:23Þ
The problem of DC bias through the signal collector resistor, and therefore the collector volt-
source is addressed by the removal of the signal age is at VCC the supply voltage. When VCE ¼ 0,
source from the position in which it is located in I C ¼ VRCCL . This corresponds to the transistor in
Fig. 2.26 and applying it to the input through a saturation with the maximum collector current
capacitor C, referred to as a coupling capacitor. flowing. There is usually a small voltage across
This capacitor blocks the DC coming from VBB the collector-emitter terminals giving the
from flowing into the signal source. This DC collector-emitter saturation voltage VCEsat 0.2
could both damage the signal source and upset V. Hence, biasing sets VCE and IC at some
the bias conditions in transistor Tr1. At the signal operating point Q on the load line.
frequencies, the capacitor is assumed to be large In order to amplify a sinusoidal signal Vi, the
such that its impedance is negligible and hence signal is superimposed on the DC bias conditions
does not interfere with signal flow into the at the base of the transistor as shown in Fig. 2.29.
transistor. Later we shall discuss choosing this For the npn transistor, as Vi goes positive
value. corresponding to conventional current flowing
A second important equation is that defining into the base, the base-emitter voltage is
the output circuit including VCE and IC for the increased, and hence the collector current IC
BJT. Applying Kirchhoff’s voltage law to the increases. The voltage drop across RL therefore
output circuit, we get increases, and since VCC is fixed, VCE is lowered.
This is manifested as a movement up the load
V CC ¼ I C RL þ V CE ð2:26Þ
line. As Vi goes negative, the base-emitter voltage
It is called the static load line for the circuit and is reduced, and hence the collector current
represents the locus of all the possible operating decreases with a resulting movement down the
VCE/IC points for the device in the particular cir- load line. When VCE is reduced to VCEsat(200
cuit under consideration. This line is plotted on mV), the transistor is saturated. When VCE is
the common emitter output characteristics as increased such that VCE ¼ VCC(IC ¼ 0), the tran-
shown in Fig. 2.30. From (2.26) when IC ¼ 0, sistor is cut-off. These values of VCE represent the
VCE ¼ VCC, and the transistor is cut-off. In such maximum and minimum voltages possible across
circumstances, there is no voltage drop across the the transistor.
Ic (mA) DC Load
Line
IC =VCC/RL 14 Ib = 120μA
12 Ib = 100μA
10 Ib = 80μA
8 Q-point Ib = 60μA
6
Ib = 40μA
4
Ib = 20μA
2
Ib = 0μA
VCE(V)
0 2 4 6 8 10 12
VCC = ½ V CC VCE = VCC
the maximum change in VCE is VCC. Therefore, 1.93 MΩ. The capacitors are chosen to be large
for maximum symmetrical swing, we let say 50 μF. Hence with RB ¼ 1.93 MΩ, ICq ¼ 1
VCEq ¼ VCC/2, that is, set the quiescent collector mA, VCEq ¼ 10 V and the npn silicon transistor is
voltage to be half of the supply voltage VCC. In biased for maximum symmetrical swing.
such a case, ICq ¼ ICmax/2 where ICmax ¼ VCC/RL. A signal voltage can now be applied at the
This will enable VCE to increase by VCC/2 to VCC input through coupling capacitor C1; the output
and decrease by the same amount VCC/2 to zero. voltage is available at the collector through cou-
Also, instead of using two DC supplies VCC and pling capacitor C2. Oscilloscope traces of the
VBB, a single supply or VCC can be used as shown input and output are shown in Fig. 2.32. The
in Fig. 2.31 with RB connected to VCC. Then in the input capacitor C1 presents a reactance X C1 ¼
bias-setting Eq. (2.25), VBB must be replaced by 1=2πfC 1 to the input signal of frequency f. If C1
VCC. Capacitor C2, like C1, is a coupling capacitor is sufficiently large, then this reactance is negligi-
that couples the output signal to an external load ble for signal frequencies and therefore is practi-
while preventing the DC voltage at the collector cally a short circuit to such signals. It however
of Tr1 from interfering with or be interfered by blocks the flow of direct current, which would
that load. change the transistor bias current as well as pos-
sibly affect the signal source. This also applies to
Example 2.1 the output capacitor C2. Methods of determining
Using the circuit of Fig. 2.31, design a fixed-bias these values are discussed in Chap. 6.
common emitter amplifier with a 20 volt supply The mutual characteristic shown in Fig. 2.33
and an npn silicon transistor having β ¼ 100. illustrates the action of the transistor in the biased
Design for maximum symmetrical swing. common emitter configuration. The Q point
established by the bias is shown as ICq and VBEq.
Solution Choose ICq ¼ 1 mA for effective tran- The input signal Vi causes a variation ΔVBE in VBE
sistor operation. For maximum symmetrical which results in a corresponding change ΔIC in
swing, VCEq ¼ VCC/2. Hence VCEq ¼ 20/ IC. Thus a change ΔVBE in VBE produces a change
ΔIC in the collector current IC that results in a
58 2 Bipolar Junction Transistor
Fig. 2.32 Input and output signals of the common emitter amplifier
Ic
ICq ΔIC
VBE
VBEq
ΔVBE
change ΔVC ¼ ΔICRL in the collector (output) common emitter voltage amplifier can now be
voltage VC. The negative sign indicates that as found. It is given by
ΔVBE increases, ΔIC increases and hence the volt-
ΔI C RL
age drop across RL increases. This corresponds to AV ¼ ¼ 40 I C RL ¼ gm RL ð2:31Þ
ΔV BE
a lowering of the transistor collector voltage VC
by ΔVC which is the output voltage Vo. It should which is approximately the same for all
be noted that the input signal can also be a current transistors. Again, the negative sign indicates
Ii into the transistor base resulting in a change ΔIB inversion of the output voltage relative to the
in IB. This results in a change ΔIC in IC and hence input signal, i.e. these two signals are 180 out
an output voltage Vo ¼ ΔVC. However, most of phase as shown in Fig. 2.32. The input imped-
input signals are in the form of a voltage rather ance Zi at the base of the common emitter ampli-
than a current, and hence, the focus is on input fier is the ratio of the input voltage change ΔVBE
voltage signals. to the corresponding input current change ΔIBE
flowing into the base circuit. Thus, noting that
Amplification IC ¼ βIB and hence ΔIC ¼ βΔIB,
Now ΔIC/ΔVBE is (approximately) the transcon-
ΔV BE ΔV BE β
ductance gm of the transistor, and in order to Zi ¼ ¼ ¼ ð2:32Þ
evaluate the voltage gain ΔICRL/ΔVBE, we need ΔI B ΔI C =β gm
to consider the diode equation That is,
qV D
I D ¼ I o e mkT 1 ð2:27Þ β
Zi ¼ ð2:33Þ
40I C
Differentiating ID with respect to VD gives
dI D q qV D Example 2.2
¼ I o e mkT
dV D mkT For the circuit developed in Example 2.1, deter-
n qV D o
q ð2:28Þ mine the voltage gain and input impedance.
¼ I o e mkT 1 þ I o
mkT
q
¼ ðI þ I o Þ Solution For the circuit of Fig. 2.31 RL ¼ 10 k
mkT D
and IC ¼ 1 mA. Hence using (2.31) AV ¼ 40
Noting that Io is small, for values of ID such that ICRL ¼ 40 1 103 10 103 ¼ 400.
ID Io, then Using the peak values of the output and input
voltages from Fig. 2.32, we find that the measured
dI D q
¼ I ð2:29Þ voltage gain is 350. Thus, the formula for gain
dV D mkT D
is only an approximate one. Using β ¼ 100, the
For the transistor, ID IE IC and VD VBE. input impedance Zi is from (2.33) given by Z i ¼
Also, for m 1 and T ¼ 300 K (room tempera- 100
¼ 2:5 kΩ. Thus, in the common emitter
401103
q
ture) mkT ¼ 40 V1 . Therefore using (2.29), we amplifying mode, the input impedance of the
get transistor is of moderate value only.
dI D ΔI C The amplified signal produced by this circuit
¼ ¼ 40 I C ð2:30Þ while having reasonable fidelity for small-signal
dV D ΔV BE
inputs does possess some level of distortion. Con-
Hence gm ¼ 40 IC A/V or gm ¼ 40 IC mA/V sider Fig. 2.33. It can be seen that equal changes
where IC is the current value in milliamperes.
ΔVBE of VBE do not produce equal changes
Note that this is approximately the same for all
ΔIC of IC. The positive-going change +ΔVBE
transistors. The voltage gain AV of the simple in VBE produces a larger collector current change
than the negative-going change ΔVBE. This
60 2 Bipolar Junction Transistor
arises because of the non-linearity of the IC/VBE 25 ¼ IB 1.5 106 + 0.7. This gives IB ¼ 16.6
characteristic and results in the output voltage μA. Hence the collector current is
having a larger negative-going amplitude than IC ¼ 100 16.6 106 ¼ 1.7 mA. From this,
the positive-going amplitude for equal amplitude V(5 k) ¼ 1.7 mA 5 k ¼ 8.5 V and therefore
bipolar inputs. This constitutes signal distortion. the collector-emitter voltage VCE ¼ 25 V
The effect can be reduced by (i) increasing ICq (5 k) ¼ 25 8.5 ¼ 16.5 V.
such that the Q point is at an increasingly linear
part of the characteristic or (ii) reducing the vari-
ation of IC and thereby restricting operation to an
2.4 Alternative Biasing Methods
approximately linear part of the characteristic.
Operating in the latter manner is called small-
As we have discussed, in order that a BJT be able
signal operation.
to amplify a signal, it must be turned on so that a
suitable operating point is established. The sim-
Circuit Analysis
plest method and the one already discussed is
It is sometimes necessary to analyse a given
called fixed biasing (or constant current biasing).
amplifier circuit in order to determine the voltages
The name is derived from the fact that the base
and currents in that circuit.
current IB is the parameter that is fixed (by fixing
RB and VCC). Since IC ¼ βIB, fixing IB results in a
Example 2.3
particular value of IC being established in the
Determine the collector current and voltage in the
transistor. There are however several problems
fixed-bias amplifier of Fig. 2.34, assuming tran-
with this technique. Firstly, it is evident that
sistor current gain of 100.
since IC ¼ βIB and IB is fixed, then IC is dependent
upon the value of β. For a specific transistor such
Solution In analysing this circuit, maximum
as the 2N3904, the value of β varies widely from
symmetrical swing capability cannot be assumed.
one device to another. Therefore, if the transistor
Therefore, the analysis is started by determining
is changed, even though it is the same number, β
the base current IB. Thus, applying Kirchhoff’s
changes and hence so does IC.
voltage law to the base circuit, we have
β also varies from one transistor type to
another. For example, the 2N3904 has β typically
at 150, while the 2N2222 has β at about 50. β also
changes as a result of collector current changes as
shown in Fig. 2.35. Finally, β is also temperature-
dependent as can be seen in Fig. 2.35 and in
response to a change of temperature may more
than double its value. The effect of all β changes
is a change in collector current. This results in a
movement of the Q point, which reduces the
maximum symmetrical swing and in extreme
circumstances may push the transistor into satu-
ration or close to cut-off.
Example 2.4
Given a DC supply of 10 V, a load resistor of 1 k
and a silicon pnp transistor with β ¼ 100 at room
Fig. 2.34 Circuit for Example 2.3 temperature, (a) bias the transistor for maximum
2.4 Alternative Biasing Methods 61
30
10
7.0 IC
5.0
0.1 0.3 0.5 0.7 1.0 3.0 5.0 7.0 10
Solution
Since VRE is made very much larger than VBE and of this design rule. In general therefore, a good
VBq ¼ VBE + VRE, it follows that rule of thumb that is a reasonable compromise
between large voltage swing and high quies-
V Bq V RE ð2:44Þ
cent current stability is V RE ¼ 10
1
V CC .
Inequality (2.43) therefore becomes 3. RE RB/β where RB ¼ R1R2/(R1 + R2) in order
to minimize changes in IC due to changes in β.
V Bq R V This criterion can be generally realized by
1 þ 2 RE =β ð2:45Þ
R2 R1 RE choosing I R ¼ 10 1
I C where IR is the current
along the potential divider R1, R2.
Now VBq/R2 is the current IR through resistor R2
4. VBq ¼ VRE + VBE ¼ IRR2 therefore, R2 ¼ VBq/
of the potential divider, while VRE/RE is the col-
IR R1 ¼ (VCC VBq)/IR
lector current IC. Thus (2.45) can be written
5. For maximum symmetrical swing, V RL ¼
V CC V RE
R
I R 1 þ 2 I C =β ¼ I B ð2:46Þ 2 , and therefore RL ¼ V CC2IV
Cq
RE
. VRE is
R1 held constant by the capacity CE.
6. Select large (of order of tens of μF) capacitor
This is satisfied if
values for good low-frequency response.
IR IB ð2:47Þ 7. Calculate the resulting voltage gain using
A V ¼ gm R L .
Inequality (2.47) requires that the current through
the potential divider is much larger than the base Example 2.5
current of the transistor which the divider Design a voltage divider-biased common emitter
supplies. This ensures that the potential divider amplifier using an 18 V supply and an npn silicon
voltage VBq is approximately fixed. Inequality transistor having β ¼ 100.
(2.47) is approximately satisfied if
I R 10 I B ð2:48Þ Solution Choose ICq ¼ 1 mA for good β and
frequency response. Using rule-of-thumb (2.44)
Since β 100 for most (small-signal) transistors VRE ¼ 18/10 ¼ 1.8 V. Use VRE ¼ 2 V. Hence,
and IC ¼ βIB, then RE ¼ 2/1 mA ¼ 2 k and RL ¼ 1 182 mA2 ¼ 8 k .
VBq ¼ VRE + VBE ¼ 2.7 volts, and from (2.52),
I R I C =10 ð2:49Þ
IR ¼ ICq/10 ¼ 0.1 mA. Therefore, R2 ¼
where IR is the current flowing through the 0:1 mA ¼ 27 k and R1 ¼ 0:1 mA ¼ 153 k . Choose
2:7 volts 182:7
resistors R1 and R2 of the potential divider. Note CE ¼ 47 μF. The resulting voltage gain is
however that IR cannot be too large since these AV ¼ 40ICRL ¼ 40 1 mA 8 k ¼ 320.
resistors affect the input impedance of the ampli- To complete the design, large coupling
fier and may also represent a drain on the power capacitors, say 47 μF, are used at the input and
supply. A simple design procedure for a common output. Note that the DC voltage at the emitter is
emitter amplifier can now be outlined: kept constant by the capacitor C3 and hence max-
1. Choose VCC and ICq. VCC may depend on the imum peak-to-peak symmetrical swing is reduced
available supply. ICq is usually in the range from VCC to VCC VRE. A further point is that the
0.1 mA to 10 mA depending on noise, transis- resistors R1 and R1 are effectively in parallel with
tor current gain and frequency response the input signal. Thus, while their values must be
characteristics. sufficiently low to ensure that IR 10IB for bias
2. To minimize effects of VBE changes, stability, they cannot be too low since they will
VRE 5VBE which for silicon is VRE 3.5 load down the input signal.
volts. The required output voltage swing and Consider now the analysis of a voltage
available supply voltage may preclude the use divider-biased common emitter amplifier circuit.
64 2 Bipolar Junction Transistor
Solution In analysing this circuit, once again 2:6 V. This gives VE ¼ VB 0.7 ¼ 1.9V. There-
maximum symmetrical swing cannot be assumed, fore, the collector current is IC ¼ 1.9/
and therefore the voltages and currents in the 1 k ¼ 1.9 mA which gives V(6 k) ¼ 1.9 mA 6
collector circuit are at this point unknown. Anal- k ¼ 11.4 V. Hence, collector voltage
ysis must hence start at the base circuit where VC ¼ 26 V(6 k) ¼ 26 11.4 ¼ 14.6 V. Finally,
note that the collector voltage VCE is given by
VCE ¼ VC VE ¼ 14.6 1.9 ¼ 12.7 V.
Example 2.8
Determine the collector current and voltage for Biasing Using a Bipolar Supply It is possible to
the circuit shown in Fig. 2.40 assuming β ¼ 100. bias the common emitter amplifier using a bipolar
power supply as shown in Fig. 2.41. Here the
Solution Using Kirchhoff’s voltage law, two collector of the transistor goes to the positive
equations for VC can be written. They are supply +VCC through RL as before. However, the
VC ¼ IB 800 k + 0.7 and VC ¼ 20 IC 2 emitter isn’t returned to ground but instead goes
k ¼ 20 β IB 2 k. Setting the right-hand side to a negative supply VCC through a resistor RE
of these two equations equal and solving for IB which together set the collector current value. The
yield IB ¼ 19.3 μA, and therefore, IC ¼ βIB ¼ 1.9 resistor RB which supplies base current to the
mA. From this, VC ¼ 20 IC transistor is connected to ground, thereby setting
2 k ¼ 20 1.9 mA 2 k ¼ 16.2 V. the base at an approximately zero potential.
66 2 Bipolar Junction Transistor
Capacitors C1 and C2 are the usual input and Fig. 2.42 Common base
output coupling capacitors, while C3 is the amplifier
decoupling capacitor that grounds the emitter for
signals.
Example 2.9
Design a common emitter amplifier using the
configuration in Fig. 2.41 with a
15V bipolar
supply and a transistor with β ¼ 110.
Fig. 2.44 Voltage divider-biased amplifiers. (a) Common emitter; (b) common base
Example 2.10
Convert the common emitter amplifier of Example
2.5 shown in Fig. 2.45 to a common base amplifier.
The results of this circuit under test are shown biasing technique must be made. Firstly, the emit-
in Fig. 2.47, where waveforms of the input and ter voltage, unlike the common emitter amplifier,
output voltage are shown. Before discussing the is not fixed; it varies according to Vi. However,
operation of the circuit, two points relating to the this change is generally small (of the order of
hundreds of mV) as compared to the changes in
the collector voltage, and therefore maximum
symmetrical swing calculations, which assume
fixed VE, are still approximately correct. Sec-
ondly, in the common emitter design, in order to
prevent loading down the input signal, R1 and R2
could not be too low. In this configuration, this
restriction is removed since there is no signal
input at the base of the transistor so that IR may
be made larger than IC/10 in order to improve the
bias stabilizing action of the circuit. However,
note that an increased IR represents an increased
demand on the circuit power supply.
Example 2.12
Improve the bias stability in the common base
amplifier in Example 2.11.
Fig. 2.47 Input and output signals for common base amplifier
2.5 Common Base Amplifier 69
transistor base, a higher current IR through R1 and Thus, the input impedance of the common base
R2 is permissible since the associated lower circuit is a factor β lower than that of the common
values of R1 and R2 can be accommodated. emitter configuration given in (2.32).
Hence choose IR ¼ ICq/2 ¼ 0.5 mA. This value
is five times higher than the value used before but Example 2.13
is not so high as to increase the current drain on For the circuit of Example 2.11, find the voltage
the system power supply unduly. Hence R2 ¼ 2.7/ gain and the input impedance.
0.5 mA ¼ 5.4 k and R1 ¼ (20 2.7)/
0.5 mA ¼ 34.6 k. Solution AV ¼ + 40ICRL ¼ 40 1 9 k ¼ 360
Regarding the operation of the circuit, note and Zi ¼ 1/40IC ¼ 1/40 1 103 ¼ 25 Ω.
that the capacitor C1 grounds the base of the As is evident, the input impedance of the com-
transistor for signals so that the input signal mon base configuration is quite low.
appears directly across the emitter-base junction.
As Vi goes positive, the emitter-base voltage of Constant Current or Fixed Bias
the BJT is reduced, and this results in a reduction The common base amplifier using constant cur-
of the transistor collector current. The voltage rent or fixed bias is shown in Fig. 2.48. This
drop across RL is therefore reduced resulting in circuit is slightly different from the common
an increase in the collector voltage of the BJT, emitter fixed-bias circuit because of the presence
i.e. Vo goes positive. As Vi goes negative, the of RE which improves the bias stability of the
emitter-base voltage of the BJT is increased, and common base circuit. An example illustrating
this results in an increase of IC. The voltage the design procedure follows.
across RL increases, and hence, the collector
voltage of the BJT goes down, i.e. Vo goes nega- Example 2.14
tive. The overall result is that Vi and Vo are in Design a common base amplifier using fixed bias
phase and there is no phase inversion of the and a silicon npn transistor having β ¼ 100. Use
voltage as occurs in the common emitter VCC ¼ 22volts.
amplifier.
Let us now evaluate the voltage gain AV, which Solution Following the procedure outlined for
using Eq. (2.30), is given by the common emitter case, choose ICq ¼ 1 mA.
V o ΔI C RL
AV ¼ ¼ ¼ 40I C RL ¼ gm RL ð2:50Þ
Vi ΔV BE
Then VRE ¼ 22/10 ¼ 2.2 volts. Use VRE ¼ 2 volts. voltage swing available at the collector of the
Therefore RE ¼ 2/1 ¼ 2 k. For maximum sym- transistor is approximately 18 volts since the emit-
metrical swing, V CEq ¼ V RL ¼ ð22 2Þ=2 ¼ ter voltage experiences only small variations due to
10 V . Therefore RL ¼ 10/1 mA ¼ 10 k. Now the signal. Therefore, for maximum symmetrical
V CC ¼ I B RB þ V RE þ V BEq giving I C RB =β ¼ swing, RL ¼ 218=2
mA ¼ 4:5 k . Capacitors C2 andC3
V CC V BEq V RE ¼ 22 0:7 2 ¼ 19:3 volts. are chosen to be large for good low-frequency
HenceRB ¼ 19.3 100/1 mA ¼ 1.93 M. response. Voltage gain is given by
ChooseC1, C3 and C2 large, say 100 μF. AV ¼ +40ICRL ¼ 40 2 mA 4.5 k ¼ 360.
Input impedance is given by Zi ¼ 1/40IC ¼ 1/
Biasing Using a Bipolar Supply 40 2 mA ¼ 12.5 Ω.
Once again it is possible to convert the common
emitter amplifier using bipolar supplies shown in
Fig. 2.41 to a common base configuration by 2.6 Common Collector Amplifier
grounding the input capacitor C1 and applying
the input signal to an ungrounded capacitor C3 as The common collector amplifier is shown in
shown in Fig. 2.49. The output is taken from Fig. 2.50. This is the third configuration in
capacitor C2. In this case both base resistor RB which the BJT can be used, and this too needs
and capacitor C1 can be completely removed and to be biased in order that a complete signal cycle
the transistor base terminal directly grounded. can be reproduced. Once again, a battery VBB is
used to establish a quiescent collector current in
Example 2.15 the transistor before an input signal is applied.
Using the configuration shown in Fig. 2.49, This biasing arrangement is shown Fig. 2.51.
design a common base amplifier to operate from The signal input Vi then causes changes in this
a
18 V supply. Determine (i) the voltage gain current such that a voltage change appears across
and (ii) input impedance of the circuit. resistor RE. Similar to the common emitter and
common base circuits, this approach allows DC
Solution Let the collector current be 2 mA. Since current to flow through the signal source. Some of
the transistor base is at ground potential, it follows the biasing techniques employed in the common
that the voltage at the transistor emitter is 0.7 V. emitter and common base designs are applicable
Hence RE ¼ (0.7 (18))/2 mA ¼ 8.7 k. The here but with some modifications.
Fig. 2.49 Common base amplifier with bipolar supply Fig. 2.50 Common collector amplifier
2.6 Common Collector Amplifier 71
Example 2.16
Design a common collector voltage divider-
biased amplifier using an 18 volt power supply
Fig. 2.52 Voltage divider-biased common collector and an npn silicon transistor with β ¼ 100.
amplifier
Solution Choose ICq ¼ 2 mA and VRE ¼ 18/2 ¼ 9
Voltage Divider Biasing volts. Then RE ¼ 9/2 mA ¼ 4.5 k. Choose IR ¼ 2/
Voltage divider biasing applied to the common 10 ¼ 0.2 mA. Then R2 ¼ 0.7/0.2 mA ¼ 48.5 k
collector configuration is shown in Fig. 2.52. R1 and R1 ¼ (18 9.7)/0.2 mA ¼ 41.5 k. Choose
and R2 fix the base voltage, and RE provides C1 and C2 large, say 100 μF.
current stabilizing action (feedback). Note that In this circuit, Vo is in phase with Vi. In fact,
there is no collector resistor; the transistor collec- since VBE is approximately fixed, this results in
tor is connected directly to the supply. It is there- Vo Vi. As a result, this circuit is sometimes
fore effectively grounded for signals through the referred to an emitter follower since the emitter
DC supply. Signal input is applied at the base and follows the input. From circuit theory, the output
output is taken at the emitter. A simply design impedance Zo of this circuit can be found by
procedure is outlined. short-circuiting the input and applying a voltage
1. Choose VCC and ICq. VCC may depend on the V and measuring the corresponding current I at
available supply. ICq is influenced by the load the output as shown in Fig. 2.53.
72 2 Bipolar Junction Transistor
Example 2.18
Determine the input impedance for the common
collector circuit in Example 2.16.
actual input impedance is obtained by placing Zi The peak negative output voltage is hence
in parallel with RB giving 453 k//502 k.
RE RL
The common collector amplifier with its high V O ¼ I L RL ¼ I Cq ð2:60Þ
RE þ RL
input impedance and low output impedance is
excellent for serving as a buffer where a high- which can be written as
impedance source must drive a low-impedance
load. When a load RL is connected however, the V O ¼ I Cq RE kRL ð2:61Þ
peak-to-peak output voltage swing of the ampli-
If we use VCEq ¼ VCC/2, then
fier is reduced on the negative half-cycle for an
npn (and the positive half-cycle for a pnp transis- V CC =2
I Cq ¼ ð2:62Þ
tor). For the circuit shown in Fig. 2.55, on the RE
positive half-cycle, the transistor supplies current
to the load, and the peak output is limited only by giving V o ¼ V2CC RERþR
L
L
. In order to maximize Vo,
the power supply rail. On the negative half-cycle RE should be made as small as is practicable so
however, the transistor must sink and not source that RL/(RE + RL) ! 1. Note however from (2.62)
current and the current flowing in from RL and that this increases the quiescent current in the
must be sunk by RE. As this occurs, the transistor transistor. A better approach is to replace RE by
must progressively turn off since this current is a constant current source. This will be discussed
now being supplied by RL. Eventually when the in Chap. 5.
transistor is completely off, the maximum current
flows in from RL. To determine the maximum Biasing Using a Bipolar Supply
value of the peak negative swing at the output at Biasing of the common collector amplifier using a
this time, we first note that the DC voltage VC bipolar supply is shown in Fig. 2.56. Here the
across the output coupling capacitor C2 is emitter is connected to the negative supply
VC ¼ ICqRE since no DC flows in RL. Therefore through resistor RE, while the base bias is sup-
when the transistor is cut-off on the negative half- plied via resistor RB which is connected to
cycle, the capacitor voltage is dropped across RE ground. C1 and C2 are the usual input and output
and the load RL, and this results in a load current coupling capacitors. Note that the use of a bipolar
IL out of the load RL and into RE of supply in this configuration with the transistor
base held at an approximately zero potential
VC I Cq RE
IL ¼ ¼ ð2:59Þ
RE þ RL RE þ RL
ensures that the output at the emitter is biased for one or more of the device terminals. It is usually
maximum symmetrical swing and permits a referred to as continuous collector current. The
greater output voltage swing since the emitter maximum collector-emitter voltage VCEO is the
can approach the positive and negative supply maximum voltage that can be applied between the
voltages. collector and emitter terminals of the transistor
when the base terminal is open circuited. It is
Example 2.20 governed by the process of avalanche breakdown.
Design a common collector amplifier using a In fact, beyond VCEO of Fig. 2.57 is a region
bipolar supply
12 V and a general purpose referred to as the breakdown region, which
npn silicon transistor having β ¼ 150. represents catastrophic failure of the transistor.
The vertical line on the characteristic corresponds
Solution Choose ICq ¼ 5 mA in order to drive to VCEsat, which is the minimum allowed
external loads. Then RE ¼ (0.7 (12))/ collector-emitter voltage of the BJT if it is not to
5 mA ¼ 2.3 k. Choose RB ¼ 20 k. enter the non-linear saturation region. There is
also the parameter VCBO which is the maximum
voltage that can be applied between the collector
and base terminals when the emitter is open
2.7 Transistor Operating Limits
circuited.
and Specifications
The approximate horizontal line near the VCE
axis corresponds to ICEO, which is the leakage
Each transistor has a region of operation in which
current. The region beneath this line is the cut-
it can function safely. This region is called the
off region that must also be avoided during (lin-
safe operating area (SOA) and is illustrated in the
ear) amplifier operation. The maximum power
(CE) output characteristic of Fig. 2.57. The maxi-
dissipation PDmax in the transistor is given by
mum collector current ICmax that a transistor can
carry is set by the current carrying capacity of the PD max ¼ V CE I C ð2:63Þ
fine wire used to connect the semiconductor
regions to the terminal loads. Excessive current This is the equation of a hyperbola that completes
will melt these wires resulting in an open circuit at the boundary of the SOA. It is especially
Ic (mA)
70μA
60μA
Ic max 50
50μA
40
40μA
Saturation
region
30 30μA
20μA
20
10μA
10
IB = 0μA
0 5 10 15 20 VCE (V)
0.3 V
Cutoff ICEO VCE max
VCE sat region
important in power transistors where considerable employs collector-base feedback biasing and a
power is dissipated during operation. Any rise in collector current of 2 mA. Using the design
temperature of the device results in a movement procedure for collector-base feedback biasing,
of the hyperbola towards the origin. The effect is the value of RL for maximum symmetrical
a reduction of the SOA with temperature increase. swing is given by RL ¼ (9/2)/2 mA ¼ 2.25 k.
Appropriate strategies to mitigate these effects Using a current gain of 100 for the 2N3904
must be introduced in utilizing power transistors, transistor, the feedback resistor RB is given
the main strategy being the mounting of the by RB ¼ ð4:5 0:7Þ= 2100mA
¼ 190 k. The coupling
power transistor on a heatsink. This is fully capacitors C1 and C2 are about 10 μF. The circuit
discussed in Chap. 10 on power amplifiers. provides a gain of better than 100 which when
Finally, the parameter fT (sometimes referred to converted to decibels is 20 log 100 ¼ 40 dB. This
as the gain/bandwidth product) is the frequency at means the output from the amplifier is greater
which the current gain hfe of the transistor falls to than 200 mV, which is sufficient to drive a
unity. It gives an indication of the transistor per- power amplifier. An interesting point here is that
formance at high frequencies. the low output impedance of the microphone
reduces the effect of the feedback via RB on the
signal and hence allows for a higher gain without
2.8 Applications the need for splitting RB and connecting a capaci-
tor to ground.
In this section, several simple circuit Ideas for Exploration: (i) Use a loudspeaker
implementations using the BJT are discussed. with an impedance of 4–16 ohms to replace the
These are intended to enable the design of actual dynamic microphone at the input. This device like
working circuits using BJTs. the dynamic microphone operates on the principle
of electromagnetic induction. While the micro-
Microphone Preamplifier 1 phone converts sound energy to an electrical sig-
A preamplifier is an amplifier that increases the nal, the loudspeaker converts an electrical signal
amplitude of a low-level signal such that it can to sound. However, its role is here reversed such
drive a power amplifier. The circuit in Fig. 2.58 that it is used as a microphone where it converts
provides amplification for a dynamic microphone sound to an electrical signal which is connected to
using a 9 V battery. Such microphones operate on the input of the preamplifier. (ii) Use an electret
the principle of electromagnetic induction and microphone to replace the dynamic microphone.
provide a nominal output of the order of 2 mV An electret or condenser microphone requires
at a low output impedance. The transistor external power being fed through a resistor for
its operation as shown in Fig. 2.59. The output
signal is coupled to the input of the preamplifier.
Because of the voltage at the output of this micro-
phone, a non-polarized coupling capacitor should
be used at the input of the preamplifier.
Microphone Preamplifier 2
Another dynamic microphone preamplifier is
shown in Fig. 2.60. It comprises a common emit-
ter amplifier using a small-signal transistor. Using
the design technique developed in this chapter,
resistors R1 ¼ 73 k and R2 ¼ 17 k set the base
voltage of Tr1 to about 1.7 volts. This along with
Fig. 2.58 Microphone preamplifier 1 R3 ¼ 1 k establishes a quiescent current of 1 mA.
76 2 Bipolar Junction Transistor
Biasing Tr1 for maximum symmetrical swing The circuit in Fig. 2.61 protects speakers at turn-
requires that resistor R4 ¼ 4 k. The capacitor on from the DC that may exist at the output of a
values are large,C1 ¼ C2 ¼ 10 μF, power amplifier. It consists of a small-signal npn
C3 ¼ 100 μF, to allow adequate low-frequency transistor such as the 2N3904 connected as an
response. emitter follower. The input to the emitter follower
Ideas for Exploration: (i) Use a loudspeaker is driven by the voltage across a capacitor C2
with an impedance of 4–16 ohms to replace the whose potential increases at turn-on because of
dynamic microphone. (ii) Use an electret micro- charge current flowing in through resistor VR1.
phone to replace the dynamic microphone. Again, The output at the emitter of the transistor drives a
because of the voltage at the output of this micro- relay coil the contacts of which connect the
phone, a non-polarized coupling capacitor should speaker to the output of the amplifier. The supply
be used at the input of the preamplifier. voltage for the circuit is derived from the supply
voltage of the amplifier. Upon turn-on of the
Speaker Protection Circuit amplifier, capacitor C2 charges up via potentiom-
eter VR1. Depending on the adjusted value of VR1,
the time for the capacitor to charge is sufficient to
allow the output of the amplifier to stabilize.
Values of C2 ¼ 100 μF and VR1 ¼ 50 kΩ will
provide a charge time of the order of
VR1 C2 ¼ 50 103 100 106 ¼ 5s.
When the capacitor voltage attains a value of
about 9 volts, the relay is activated, and the
contacts connect the speaker to the amplifier out-
put. The extent of the delay can be adjusted by
varying the potentiometer. Diode D1 which can
be an IN4001 diode protects the transistor from
back emfs generated by the relay coil, while
Fig. 2.59 Electret microphone
capacitor C1 provides nominal filtering to the
system supply voltage. A value of about 0.1 μF is capacitor C1 ¼ 1000 μF provides the unregula-
suitable. ted input. The peak voltage on the filter
pffiffiffi
Ideas for Exploration: (i) Re-design the sys- capacitor is 12 2 1:4 ¼ 15:6 V. The maximum
tem using a 2N3906 pnp small-signal transistor. peak-to-peak ripple across the capacitor is
This would require reversal of the polarity of the given by ΔV ¼ I/2fC ¼ 100 103/
power supply. (ii) Investigate the action of the 2 60 1000 106 ¼ 0.8 V giving a mini-
diode in protecting the transistor. mum unregulated voltage 15.6 0.8 ¼ 14.8 V.
Using an 8.2 V Zener and a red LED as an
Bench Power Supply on-indicator with a forward voltage of 1.8 V,
Figure 2.62 provides a fixed output of about the total voltage across the series arrangement is
9 volts and a maximum current of about 8.2 + 1.8 ¼ 10 V. The voltage output at the
250 mA. It uses a Zener diode regulator and emitter of the transistor would be 10 V less the
boosts the current output with a transistor in 0.7 V of the transistor base-emitter junction giv-
emitter follower configuration. In order to ensure ing about 9.3 V. For a Zener operating current of
that sufficient voltage is available to deliver 10 mA, resistor R1 is given by (14.8 10)/
current to the associated Zener, a basic full-wave 10 mA ¼ 480 Ω. The 2N3055 power transistor
rectifier using a 12 volt transformer, a diode boosts the current output capacity of the Zener
bridge comprising D1 to D4 and a filter diode by the factor of the current gain of the
78 2 Bipolar Junction Transistor
transistor which, for a 2N3055, is about 50. The approximately constant. Any further small
maximum current available is calculated using the increase in the load current further turns on Tr2,
minimum allowable current for the Zener which and this reduces the current to the Zener. The
is about 5 mA. Hence the available current from effect is that the Zener voltage falls and a constant
the Zener regulator is 10 5 ¼ 5 mA, and voltage supply becomes a constant current sup-
therefore the maximum output current is ply. Then, R2 ¼ 0.7/ILmx where ILmx is the maxi-
5 mA β ¼ 5 mA 50 ¼ 250 mA where mum load current.
β ¼ 50 is the current gain of the transistor. The Ideas for Exploration: (i) Plot the voltage-
fuse provides short-circuit protection. current characteristic for the modified bench sup-
Ideas for Exploration: (i) Modify the system to ply circuit to show the effect of the electronic
produce a variable bench supply by introducing a fuse. (ii) Introduce a small capacitor across the
10 k potentiometer across the series LED-Zener output of the supply to improve filtering.
arrangement and connecting the wiper of the
potentiometer to the transistor base. Identify any Blown Fuse Indicator
problems with this arrangement. (ii) Use a tran- This circuit gives a visual indication when a pro-
sistor with a higher gain and explain if this tection fuse is blown. It is placed between the
provides any performance improvement. (iii) Uti- power supply and the circuit being protected and
lize the TL431A programmable Zener introduced comprises a pnp transistor arranged as shown in
in Chap. 1 to design the supply instead of the Fig. 2.64. During normal operation with a contin-
Zener diode. uous fuse, the voltage across the base-emitter
junction of the transistor is approximately zero.
Electronic Fuse The transistor is therefore off and the LED does
This circuit adds electronic over-current protec- not light. If the fuse is blown because of an
tion to the bench supply just discussed. It consists over-current, then voltage drop across the base-
of a small-signal npn silicon transistor Tr2 with a emitter junction causes Tr1 to turn on. Current
resistor R2 between the base and emitter of the thereby flows through the LED which now turns
transistor. The arrangement is connected to the on. When Tr1 is on, the full supply voltage is
supply as shown in Fig. 2.63. As increased load dropped across the LED and resistor. Thus, for a
current flows through R2, the voltage drop across 24 volt supply and allowing an (red) LED voltage
this resistor increases. When this value equals of 1.8 V and a current of 10 mA, R3 ¼ (24 1.8)/
0.7 V, transistor Tr2 turns on drawing current 10 mA ¼ 2.2 kΩ. Assuming a transistor current
away from the base of transistor Tr1, thereby gain of 100, the base current of Tr1 is given by
causing the supply current to remain 10 mA/100 ¼ 0.1 mA, and this flows through R1
and R2. Hence, R1 + R2 ¼ (24 0.7)/ that the telephone is in use. The diode bridge
0.1 mA ¼ 233 kΩ. To ensure transistor satura- ensures that the DC signal into the transistor
tion, use 200 k. Then noting that R2 is a load on from the telephone line is always of the same
the supply, set R1 ¼ R2 ¼ 100 k. polarity. When the telephone is on hook, the
Ideas for Exploration: (i) Introduce a green telephone circuit is open, and 48 volts available
LED in series with resistor R2. During normal from the telephone exchange will be across tip
operation there would be sufficient current (T) and ring (R). At this time since the emitter of
through this resistor to turn this LED on but it the transistor is at VCC, the resistor values R1, R2
goes off when the fuse is blown. This would and R3 must be chosen such that the voltage at the
complement the action of the red LED. base of the transistor is greater than VCC 0.7
volts to ensure that the transistor is off. The LED
Telephone Monitor would therefore not be illuminated. When the
This circuit shown in Fig. 2.65 turns on an LED handset is lifted, the telephone circuit is closed,
when the telephone handset is lifted indicating and as a result of line resistance, the voltage
80 2 Bipolar Junction Transistor
across tip and ring will drop. With suitable values held at a fixed voltage 1.4 V using two forward-
for resistors R1, R2 and R3, the voltage at the base biased diodes in series instead of a voltage
of the transistor will drop to a value such that the divider. This arrangement provides a more stable
voltage across the base-emitter junction of the voltage at the base. Current through these diodes
transistor exceeds 0.7 volts, thereby turning on is supplied from a 15 volt supply through a resis-
the transistor and illuminating the LED. Thus for tor R5 ¼ 3.3 k. Hence the current through diodes
VCC ¼ 3 volts, R1 ¼ 470 k, R2 ¼ 1 M and D1 and D2 is (15 1.4)/3.3 k ¼ 4 mA. Capacitor
R3 ¼ 100 k, then the voltage at the base of the C5 ensures that the transistor base is grounded for
transistor is given by R1 þRR32 þR3 48 ¼ 3 volts. At signals. Choosing a collector current of 1 mA,
this voltage with the emitter voltage also at then R6 ¼ (1.4 0.7)/1 mA ¼ 700 Ω. For maxi-
3 volts, it follows that the base-emitter voltage is mum symmetrical swing, the quiescent voltage
zero and the transistor is off. When the phone is across R4 is made equal to the quiescent
lifted off the hook, the telephone circuit is closed. collector-emitter voltage giving V CE ¼ V R4 ¼
Current flows such that a voltage drop occurs ð15 0:7Þ=2 ¼ 7:2 V . Hence R4 ¼ 7.2/
along the line, thereby reducing the 48 volts 1 mA ¼ 7.2 k. Input and output coupling
across tip and ring. This reduces the base voltage capacitors C1 to C4 are chosen to be large values
and causes the transistor to turn on lighting the for good low-frequency response. The input
LED. Resistor R4 limits the current through the impedance at the emitter is 1/40IC ¼ 1/40 1
LED. A value of about 100 Ω is suitable. mA ¼ 25 Ω which is quite low. By choosing
Ideas for Exploration: (i) Introduce an LED signal input resistors R1 to R3 to be much greater
that indicates the on-hook condition. than 25 Ω say 22 k, very good signal mixing
occurs at the transistor emitter with very low
Audio Mixer channel interaction (cross-talk).
This circuit for an audio mixer in Fig. 2.66 uses a Ideas for Exploration: (i) Introduce 10 k
common base amplifier to enable the mixing of potentiometers in each input circuit in order to
signals from several sources. It consists of a com- adjust individual channel gain. (ii) Sacrifice max-
mon base amplifier in which the transistor base is imum symmetrical swing for increased output
signal amplitude by increasing the value of resis- While the current through the Zener is 1 mA, the
tor R3. (iii) Re-design the circuit using a red or power transistor can allow significantly higher
green LED instead of the two signal diodes. currents, thereby increasing the power rating of
the arrangement beyond the power rating of the
Power Zener Diode Zener. For example, if the transistor current is set
The power rating of a Zener diode can be at 1A, then the power rating of the arrangement
increased by coupling the Zener with a power becomes approximately PD0 ¼ 5 V 1 A ¼ 5 W.
transistor as shown in Fig. 2.67. Resistor R1 sets Note that the base current of the power transistor
the current through the Zener at 0.7/R1. For 1 mA flows through the Zener in addition to the 1 mA
we get R1 ¼ 700 Ω. The effective Zener voltage supplied to the resistor, since this latter current is
VZ0 across the arrangement is VZ0 ¼ VZ + 0.7, set by the base-emitter voltage of the transistor.
which for a 4.3 V Zener gives VZ0 ¼ 4.3 + 0.7 ¼ 5 V. Ideas for Exploration: (i) Use this power
Zener to design a high-current Zener diode
regulator. (ii) Compare this regulator with that
providing the bench power supply. Zener
regulators are referred to as shunt regulators as
the regulating device (Zener) is in parallel with
the load, while the earlier regulator in the bench
supply is a series regulator since the regulating
element (transistor) is in series with the load. (iii)
Implement this circuit using a pnp transistor.
small and hence most of I flows through R2. occurs. As a result, no bias voltage is dropped
Therefore, the effective Zener voltage VZ0 across across R2, and hence the transistor remains off.
the arrangement is V Z 0 ¼ 1 þ RR21 ðV Z þ 0:7Þ: The LED D1 does not light. An open diode will
cause no conduction with either connection and
Thus using R1 ¼ R2 ¼ 6.8 k and a 6.8 V Zener hence no lighting of the LED. A shorted diode
gives VZ0 ¼ (1 + 1)(6.8 + 0.7) ¼ 15 V. The will cause conduction and hence LED lighting
resulting current through R1 and R2 is I ¼ 7.5/ with both connections. The values of R1 and R2
6.8 k ¼ 1.1 mA. The power rating of the adjusted must be chosen such that the transistor is turned
diode will then be a function of the power rating on when diode conduction takes place. R1 ¼ 10 k
of the associated transistor. Note that if the Zener and R2 ¼ 2.2 k will give a voltage at the transistor
is replaced by a short circuit, the voltage output base of 2.2 9/(10 + 2.2) ¼ 1.6 V. This value is
becomes V Z 0 ¼ 1 þ RR21 0:7. This particular cir- more than sufficient to turn on the transistor and
cuit arrangement is sometimes referred to as a Vbe will fall as base current is drawn from the junction
multiplier or amplified diode (see Chap. 5). of the two resistors. A value for resistor R3 of 1 k
Ideas for Exploration: (i) Investigate the use of will limit the current in the transistor to less than
this adjustable Zener in the design of a variable 9 V/1 k ¼ 9 mA.
voltage regulated supply. Ideas for Exploration: (i) Investigate the use of
the diode tester to check transistors. Noting that a
Diode Tester transistor is made up of two back-to-back diodes,
A transistor can be used in the testing of diodes as it follows that both the base-emitter and base-
shown in Fig. 2.69. The diode to be tested is collector junctions of a transistor should have a
connected in the base biasing circuit. An LED low forward resistance and a high reverse resis-
along with a current limiting resistor is connected tance, while the collector-emitter terminals
in the collector circuit. When a good diode is should have high resistance in both connections.
connected to the test terminals with the correct Both npn and pnp transistor can be tested but with
polarity, conduction through the diode and R1 appropriate polarities.
occurs, and a fraction of the supply voltage is
dropped across resistor R2.This turns on Tr1 and Dry Soil Indicator
hence the LED. If the connection is reversed, the The circuit in Fig. 2.70 functions as a dry soil
diode becomes reverse-biased, and no conduction detector. The two leads connected to the base and
emitter terminals of the transistor are placed in the
soil whose moisture level is to be monitored. If
the water level is sufficient, the resistance
between the two probes will be low. As a result,
the voltage drop across the base-emitter junction transistor giving a collector current that is
is insufficient to turn on the transistor, most of the measured by the milliammeter. Upon
supply voltage being dropped across resistor R1. questioning, if the candidate lies, the body resis-
As the soil becomes dry, the soil resistance will tance may fall resulting in an increased base
increase, and hence the voltage drop across the current and hence a corresponding increase in
base-emitter junction will eventually be sufficient collector current. This increase is measured by
to turn on the transistor. The LED D1 in the the meter and may be an indication that the
collector circuit will therefore turn on. Some individual may be lying. Almost any small-
experimentation may be necessary to determine signal npn transistor such as the 2N3904 can be
a suitable value for R1. A reasonable value to start used. The resistor R1 ¼ 1 k in series with the
with is R1 ¼ 220 k. Since the supply voltage is milliammeter limits the current to less than
only 3 V, a current limiting resistor in series with 9 mA in the event that the probes are short-
the LED may not be necessary. circuited. Note however that since R1 is in the
Ideas for Exploration: (i) Try the circuit with collector circuit, it has very little influence on the
different types of soil. collector current. When in use, the range of the
meter should be adjusted to accommodate the
Lie Detector resulting current.
A lie detector is an instrument that attempts to Ideas for Exploration: (i) Experiment with
detect when an individual is telling a lie by mea- various individuals to evaluate the effectiveness
suring changes in body characteristics such of the circuit. (ii) Research other body
as skin resistance. It is based on the idea that characteristics used in lie detection including
skin resistance changes with emotional state, breathing rate, pulse rate and blood pressure.
particularly as a result of perspiring. This resis-
tance change can be easily detected and used as Research Project 1
an indicator of lying. The lie detector circuit This project involves the use of the transistor to
shown in Fig. 2.71 utilizes a transistor Tr1 with improve the performance of the simple radio
a milliammeter in the collector circuit. The receiver of Chap. 1. The circuit is shown in
probes connected to the collector and base Fig. 2.72. Here the base-emitter junction of a
terminals and held firmly in both hands by the germanium transistor Tr1 such as the NTE101
candidate under test. Under normal conditions, replaces the germanium diode in order to produce
the body resistance will allow a level of base signal demodulation. The output of the tank cir-
current to flow. This is amplified by the cuit drives the base of the transistor, while the
Drain Drain
n-type p-type
Gate Gate
p-type
n-type
n-type
Channel
p-type
Channel
Source Source
Source Source
constant at a value referred to as the saturation The reverse bias at the gate-channel p-n junc-
drain current with no further increase resulting tion can be increased by the application of a
from an increasing drain-source voltage. negative potential to the gate terminal relative to
The channel is now said to be “pinched off”, the source. Thus, instead of VGS ¼ 0 as before, the
and the FET is operating in the pinch-off or con- gate is disconnected from the source and a nega-
stant current region. Note that in this region the tive bias applied at the gate with respect to the
channel is not closed off completely since this source as shown in Fig. 3.6. This reverse bias
would cause the drain current ID to go to zero. widens the existing depletion layer and as a result
The drain-source voltage at which pinch-off decreases channel width as well as increases
occurs corresponds in magnitude to what is called channel resistance, even before VDS is applied.
the pinch-off voltage VP. If VDS is increased still Then as VDS is increased from zero, the decreased
further, avalanche breakdown will eventually channel width means that pinch-off will occur at a
occur as in all p-n junctions. The onset of ava- lower value of VDS and a lower value of saturation
lanche breakdown is very rapid and may cause drain current. Further negative increases in VGS
irreparable damage to the junction. The resulting
ID/VDS characteristic is shown in Fig. 3.5.
Fig. 3.4 Pinch-off in n-channel JFET Fig. 3.6 JFET with negative gate-source voltage
ID
VGS =0
B C
IDSS
Breakdown
Ohmic Pinch-off region
region
A region
o Vp VDS
Pinch-Off Values
Ohmic Saturaon Region
ID (mA) Region VGS = 0V
VGS = -1V
VGS = -2V
VGS = -3V
VGS = -4V
0 5 10 15 20 25 30 VDS (V)
Pinch-Off Values
Ohmic Saturaon Region
ID (mA) Region VGS = 0V
VGS = 1V
VGS = 2V
VGS = 3V
VGS = 4V
0 -5 -10 -15 -20 -25 -30 VDS (V)
VDS = VGS - VP
SA T
IDS
Linear region Saturaon region
VGS0 =0
IDSS
VGS1<VGS0
VGS2<VGS1
VGS3<VGS2
ID (mA)
6
5
4
3
2
1
VGS=VP
-6 -5 -4 -3 -2 -1 0 1 2 3
VGS
Fig. 3.13 N-channel
Fig. 3.12 Circuit for the determination of the
characteristics of the n-channel JFET specific positive voltage. Using power supply P1
and the voltmeter between the gate and source,
device can be operated as a voltage-dependent VGS is varied. As VGS is varied, the corresponding
resistance, i.e. as a resistance whose value VDS/ value of drain current ID is recorded using the
ID depends on the value of VGS. Here the gate- milliammeter in the drain circuit for various
source voltage on the FET determines the channel values of VGS. This is repeated for various fixed
resistance according to values of VDS. The result is the mutual character-
ro istic for the device.
rd ¼ ð3:2Þ
ð1 V GS =V P Þ2 This curve shows directly the effect of gate-
source voltage on saturation drain current and
where ro is the channel resistance rd with indicates that the JFET is a voltage-controlled
VGS ¼ 0 V. device. Mutual characteristics for n-channel and
The mutual or transfer characteristic of a FET p-channel JFETS are shown in Figs. 3.13 and
is a graph of drain current against gate-source 3.14, respectively. From these curves, it can be
voltage for constant values of drain-source volt- seen that for |VDS| > |VP| where ID assumes its
age above pinch-off obtained using the circuit of saturation value, as VGS is changed from zero,
Fig. 3.12. Using power supply P2 and the voltme- the saturation drain current decreases from IDSS.
ter between the drain and source, VDS is set at a At some value VGS ¼ VP, the drain current goes
3.3 JFET Parameters 95
Fig. 3.15 JFET configurations: (a) common source, (b) common drain and (c) common gate
3.4 Using the JFET as an Amplifier 97
(a) (b)
Fig. 3.16 (a) JFET fixed-bias configuration; (b) design solution
98 3 Field-Effect Transistor
into the gate. The solution is shown in Fig. 3.16 positive voltage developed across RS as ID flows
(b). Input and output coupling capacitors must be results in the gate (0 V) being at a lower potential
used because of the voltages present on the gate than the source (+IDRS) as in the fixed-bias con-
and drain. Using (3.10), gm is given by figuration. VGG is effectively replaced by
VGS ¼ IDRS. In order to determine the value of
gm ¼ 2I DSS
VP 1 V GS
VP ¼ 2: 10
8 1 2:34
8 RS to produce a desired ID, Shockley’s equation is
¼ 1:77 mA=V: again used. Thus, from (3.4),
2
Hence the voltage gain AV is given by V GS
I D ¼ I DSS 1
Av ¼ gmRL ¼ 1.72 1.2 k ¼ 2. This of VP
course is quite low as compared with the BJT and 2
I D RS
results from the comparatively low value of FET ¼ I DSS 1 þ ð3:12Þ
VP
transconductance.
Re-arranging, this yields
Self-Bias rffiffiffiffiffiffiffiffi
One disadvantage of the fixed-bias method is the ID VP
RS ¼ 1 ð3:13Þ
need for two voltage supplies. The voltage source I DSS ID
VGG can be eliminated by the inclusion of RS in
the source circuit of the FET as shown in
Fig. 3.17. Example 3.5
Using the FET of Example 3.4, find RS to replace
This method of biasing is referred to as self- VGG.
bias and operates in the following manner: Since rffiffiffiffiffiffiffiffiffiffiffiffiffiffi !
the gate is connected to ground through RG, the 5 mA
Solution From (3.13), RS ¼ 1
10 mA
8 8
¼ :2929 ¼ 469 Ω:
5 mA 5 103
Example 3.6
2
Using Shockley’s equation I D ¼ I DSS 1 VVGSP
and a 20 volt supply, design a common source
amplifier using an n-channel JFET having a
Fig. 3.17 Self-biased JFET amplifier pinch-off voltage of 3 volts and IDSS ¼ 8 mA.
3.4 Using the JFET as an Amplifier 99
Use self-bias and a 2 mA quiescent current. Show Hence the voltage gain AV is given by
pffiffiffiffiffiffiffiffiffi
that gm ¼ 2 IVD IPDSS and use this to calculate the Av ¼ gmRL ¼ 2.67 4.6 k ¼ 12.3.
voltage gain of your circuit.
Voltage Divider Biasing
Solution Using IDSS ¼ 8 mA, VP ¼ 3 V A third FET biasing scheme is the voltage divider
and ID ¼ 2 mA in Shockley’s equation gives bias as shown in Fig. 3.19. Since there is no gate
qffiffiffiffiffiffi qffiffiffiffiffiffiffiffi current,
V GS ¼ 1 IIDSS
D
mA ð3Þ
V P ¼ 1 28 mA
R2
¼ ð1 0:5Þð3Þ ¼ 1:5 V: VG ¼ V ð3:14Þ
R1 þ R2 DD
Fig. 3.18 Solution to Example 3.6 Fig. 3.19 FET voltage divider bias
100 3 Field-Effect Transistor
Example 3.9
Design a source follower circuit using a JFET
with IDSS ¼ 8 mA, VP ¼ 4 V and a 12 V supply.
Fig. 3.21 JFET common drain amplifier Fig. 3.22 JFET common drain biased with bipolar supply
ID ¼ 2 mA, then RS ¼ 6 V/2 mA ¼ 3 k. For symmetrical swing, the source voltage was set at
qffiffiffiffiffiffiffiffi
ID ¼ 2 mA, V GS ¼ 1 28 mA half the supply voltage giving RS ¼ 6V/ID. Now
mA ð4 VÞ ¼
for RS connected to 12 V, for maximum sym-
2 V. Hence VG ¼ 6 2 ¼ 4 V. From this it
metrical swing, the source voltage is set at 0 V
follows that R1 ¼ 2 M and R2 ¼ 1 M will produce
giving a voltage drop across RS of
this gate voltage.
0 (12) ¼ 12 V. Therefore RS is calculated
from RS ¼ 12/2 mA ¼ 6 k. All other component
Biasing Using Bipolar Supply
values remain the same.
The common drain amplifier of Fig. 3.21 using
voltage divider biasing can be improved by
returning RS to a negative supply voltage VDD
as shown in Fig. 3.22. This results in a larger 3.4.3 Common Gate Configuration
value for RS and hence greater bias stability as
well as a larger output voltage swing since the The last configuration of the FET is the common
available peak-to-peak voltage swing increases gate configuration shown in Fig. 3.15(c). Here
from VDD to 2VDD. the input signal is applied to the source and the
output signal taken from the drain of the FET.
Example 3.10 The gate is grounded and hence common to both
For the common drain amplifier in Example 3.9, input and output. This configuration is analo-
re-design the system to use a bipolar supply of gous to the common base BJT amplifier and is
12 V and return RS to the negative supply volt- frequently used in high-frequency amplifiers and
age instead of ground. where low input impedance is required. A prac-
tical implementation of this configuration in an
Solution The modified system is shown in amplifier circuit is shown in Fig. 3.23. The
Fig. 3.22. Since RS is being connected to 12 V design of this circuit follows the same procedure
instead of ground, its value has to be as the common source circuit discussed in Sect.
re-calculated. From the design in Example 3.9 3.4.1. In the self-bias case, the resistor RG is now
where RS was connected to ground, for maximum not necessary, and the gate can be connected
102 3 Field-Effect Transistor
(a) (b)
DRAIN D
n+
METALLIZATION P-TYPE
LAYER SUBSTRATE
Substrate
G
GATE n SS
DIFFUSED
CHANNEL
SOURCE S n+
VDD VDD
D D
+
+
VGS=0V VGS=VGG +
+
+
G G +
+
SS +
SS
p
+
p
+
p +
+
+
p
+
S n S n
(a) (b)
Fig. 3.27 N-channel depletion MOSFET action
ID (mA)
VGS = 0V
0
4 8 12 16 20 VDS (V)
If the gate-source voltage is made negative cause a greater drain current to flow for a given
relative to the source as shown in Fig. 3.27(b), drain-source voltage and means that a larger value
electrons are repelled out of the channel into the of drain-source voltage is required to cause pinch-
n+ region. This reduces the number of free off. Therefore, the overall effect of the gate-
electrons which are available for conduction in source voltage increase is an increase in pinch-
the channel region and so the channel resistance is off voltage and saturation current. This mode is
increased. As a result, pinch-off occurs at a lower referred to as the enhancement mode. Thus, the
drain-source voltage and saturation current. gate-source voltage controls the saturation drain
If the gate-source voltage is changed from zero current of the MOSFET. Moreover, because of
and made positive, electrons will be attracted into the silicon dioxide layer, an n-channel depletion
the channel from the heavily doped n+ regions. MOSFET operates in both an enhancement and a
The number of conduction electrons is increased depletion mode, and it is possible to change the
and so the channel resistance is reduced. This will polarity of the gate-source voltage. The transfer
and output characteristics are shown in Figs. 3.29
and 3.30, respectively, and are similar to those of
ID (mA) the JFET except that the JFET has only the deple-
9
8 tion mode. Shockley’s equation given by I D ¼
2
7 I DSS 1 VVGSP also defines the behaviour of the
6 depletion MOSFET.
5 The p-channel depletion MOSFET is shown in
4 Fig. 3.31. It operates in a manner similar to the
3
n-channel depletion MOSFET except that all
polarities are reversed. The associated drain
2
characteristics are shown in Fig. 3.32 with
1 reversed voltage polarities. The mutual character-
istic is shown in Fig. 3.33, and here too the
-6 -5 -4 -3 -2 -1
Vp
0 1 2 3 VGS polarity change in the gate-source voltage can
be seen. As previously observed, the JFET is
Fig. 3.29 N-channel mutual characteristics also a depletion mode or normally on device. It
ID (mA)
ENHANCEMENT
10
VGS = +1V
MODE
9
8
7 VGS = 0V
6
5 VGS =-1V
DEPLETION
4
MODE
3 VGS = -2V
2
VGS = -3V
1
VGS = -4V (off)
0 4 8 12 16 VDS (V)
DRAIN D
p-
METALLIZATION n-TYPE
LAYER SUBSTRATE
Substrate
G
GATE p SS
DIFFUSED
CHANNEL
SOURCE S p-
ENHANCEMENT
depletion MOSFET 10
VGS = -1V
MODE
9
8
7 VGS = 0V
6
5 VGS = +1V
DEPLETION
4
MODE
3 VGS = +2V
2
VGS = +3V
1
VGS = +4V (off)
0 -4 -8 -12 -16 VDS (V)
and aluminium contacts made as shown in the source, there is essentially no conduction through
diagram. Like the enhancement-type MOSFET, the channel with the application of a drain-source
the substrate and source are connected together in voltage. In fact for VGS 0, the channel current is
order to ensure that the channel substrate junction of the order of a few nano-amperes (since the
is reverse-biased. Unlike the depletion-type channel consists of two back-to-back diodes).
MOSFET, no channel exists between the two With a positive voltage applied between gate
highly doped drain and source regions. Because and source as shown in Fig. 3.36, a virtual chan-
of this if no potential is applied between gate and nel is created between the drain and the source.
The positive voltage attracts negative carriers into
the p-type substrate from the n-type regions, and
ID (mA) this results in significantly increased conductivity
9 between drain and source.
8
7
6
5
4
3
2
1
SiO2
INSULATING
LAYER
DRAIN D
n+
METALLIZATION
LAYER
P-TYPE Substrate
G
GATE SUBSTRATE SS
SOURCE S n+
VDD
D
n+
Induced
Channel
VGS=VGG
G
SS
p
VGG
S n+
The minimum positive gate-source voltage increases, and therefore, pinch-off occurs at the
that is necessary for the creation of the virtual higher values of VDS and ID as shown in Fig. 3.37.
channel so that conduction is possible is called Thus, the drain current of the enhancement-type
the threshold voltage. Its value at which the drain MOSFET can be controlled by the gate-source
current ID reaches some defined value, say 10 μA, voltage.
is often quoted and is typically about 2 volts. The enhancement-type MOSFET has a mutual
Thus, the positive voltage enhances the drain characteristic Fig. 3.38 that is very different from
current, and such a device is called an the JFET and the depletion-type MOSFET. The
ENHANCEMENT-type MOSFET. As VGS is main difference arises because of the need for the
made positive, the drain current ID increases gate-source voltage to exceed a threshold value in
slowly then rapidly. order that channel conduction occurs. The defin-
An increase in the gate-source voltage above ing equation is different from the Shockley and is
the threshold value will attract more electrons into given by
the channel region. Once the virtual channel has
been formed, the drain current which flows is I D ¼ k ðV GS V T Þ2 ð3:17Þ
influenced by the magnitude of both the gate-
where k is a constant. It is found using
source and drain-source voltages. For a given
VGS > VT as VDS is increased, the width of the I DðonÞ
k¼ 2 ð3:18Þ
depletion region increases until pinch-off occurs, V GSðonÞ V T
thereby causing saturation of the drain current as
occurred in the depletion MOSFET. As VGS is where I DðonÞ and V GSðonÞ are particular drain current
increased above VT, channel conductivity and associated gate-source voltage values for the
3.6 MOSFET Amplifiers 109
VGS = +7V
8
VGS = +6V
4 VGS = +5V
VGS = +4V
VGS = +3V VGS =VT= +2V
4 8 12 16 20 VDS (V)
8
VGS = -6V
VGS = -5V
4
VGS = -4V
VGS = -3V
VGS =VT= -2V
0 -4 -8 -12 -16 -20 VDS (V)
ID (mA)
6
VT
-6 -4 -2 0 VGS (V)
Example 3.13
Using self-bias, design a biasing network for a
common source n-channel depletion-type
MOSFET having IDSS ¼ 8 mA and VP ¼ 8 V.
(a) (b)
Solution The circuit is shown in Fig. 3.42.
Fig. 3.41 Symbol for the enhancement MOSFET (a)
Choosing a quiescent drain current of
n-channel and (b) p-channel
2 mA, Shockley’s equation gives V GS ¼
qffiffiffiffiffiffi qffiffiffiffiffiffiffiffi
1 IIDSS
D
mA ð8 VÞ ¼ 4 V:
V P ¼ 1 28 mA
3.6.1 Depletion-Type MOSFET
Common Source Amplifier Hence RS ¼ 4 V/2 mA ¼ 2 k. We determine RD
in order to achieve maximum symmetrical swing.
The depletion-type MOSFET which also obeys Therefore, RD ¼ (VDD IDRS)/ID. This yields
Shockley’s equation can be biased in a similar RD ¼ (20 4)/2 mA ¼ 4 k. Finally, resistor RG
manner to the JFET. This is shown Fig. 3.42 can be set at 1 M because of the high input
where self-bias is used. impedance of the device.
3.6 MOSFET Amplifiers 111
Fig. 3.43 MOSFET voltage divider bias divider biasing for a drain current of 5 mA.
Design for a maximum symmetrical swing with
VDD ¼ 18 volts.
A depletion-type common source amplifier
using voltage divider bias is shown in Fig. 3.43. Solution Using VG ¼ VDD/10 ¼ 1.8 V,
The design approach is similar to that used in the for IDSS ¼ 6 mA, VP ¼ 3 V and
JFET amplifier in Fig. 3.17. In Fig. 3.43 since ID ¼ 5 mA, Shockley’s equation gives RS ¼
there is no gate current, qffiffiffiffiffi
3
1:8
6 þ 1:8
3 1 5 ¼ 631 Ω. Note IDRS ¼ 5
R2
VG ¼ V ð3:19Þ 0.631 ¼ 3.2 V. Hence for maximum symmetrical
R1 þ R2 DD
swing RL ¼ ðV DD 3:2
5
Þ=2
¼ ð183:2
5
Þ=2
¼ 1:48 k .
and
Since VG ¼ 1.8 volts, then for R2 ¼ 200 k and
V GS ¼ V G I D RS ð3:20Þ R1 ¼ 18 200 k/1.8 200 k ¼ 1.8 M.
drain current using Shockley’s equation, and the The design of this circuit follows the same
gate voltage set by the voltage divider is then procedure as the common source circuits
given by V G ¼ V 2DD þ V GS . Capacitors C1 and C2 discussed earlier. In the self-bias case, the resistor
are the input and output coupling capacitors. RG is no longer necessary, and the gate can be
connected directly to ground as shown in
Example 3.15 Fig. 3.45. In the voltage divider bias, the gate is
Using the circuit shown in Fig. 3.44, design a grounded using the input capacitor. In both cases
common drain amplifier using an n-channel the signal is injected at the source using a capaci-
depletion MOSFET having IDSS ¼ 7 mA and tor, and the output signal is taken at the drain.
VP ¼ 5 volts. Design for maximum symmetri- Hence, the common source amplifier can be
cal swing with VDD ¼ 20 volts. converted to a common gate amplifier by ground-
ing the input coupling capacitor and lifting the
Solution Since VDD ¼ 20 V, then the quiescent grounded end of the bypass capacitor and there
source voltage must be 10 V for maximum sym- applying the input signal.
metrical swing. Hence if we choose ID ¼ 4 mA,
then RS ¼ 10 V/4 mA ¼ 2.5 k. For ID ¼ 4 mA, Example 3.16
qffiffiffiffiffiffiffiffi
Design a common gate amplifier using an
V GS ¼ 1 48 mA mA ð4 VÞ ¼ 1:2 V . Hence n-channel depletion MOSFET with IDSS ¼ 2 mA
VG ¼ 10 1.2 ¼ 8.8 V. Resistors R1 ¼ 880 k and VP ¼ 3 volts. Design for maximum sym-
and R2 ¼ 1.12 M will produce this gate voltage. metrical swing using a 24 V supply.
Example 3.20
Using the configuration in Fig. 3.49, design a
voltage divider-biased common drain amplifier
using an enhancement-type MOSFET having Solution The design procedure used in Example
VT ¼ 2 V, k ¼ 0.25 mA/V2 and use VDD ¼ 20 V. 3.19 for the voltage divider-biased common
source amplifier is employed to produce the
Solution The circuit is shown in Fig. 3.49. design shown in Fig. 3.48. However, now capac-
Setting RG1 ¼ RG2 ¼ 1 M this gives VG ¼ 10 V. itor C1 is grounded and the input signal applied at
For a quiescent drain current of 2 mA, V GS ¼ capacitor CS which is ungrounded. The resulting
qffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi circuit is shown in Fig. 3.51.
V T þ IkD ¼ 2 þ 0:252 mA=VmA
2 ¼ 4:8 V . Hence,
Fig. 3.50 Common gate amplifier Fig. 3.51 Solution for Example 3.21
116 3 Field-Effect Transistor
and the device turns on. This voltage is trans- gate-source voltage is increased until the thresh-
ferred to the low-voltage DC motor which then old voltage of the MOSFET is exceeded and the
operates. The speed of the motor increases as the device turns on lighting the lamp. The brightness
voltage is increased. The value of the potentiom- of the lamp is controlled by the amount of cur-
eter is not critical, and the high input impedance rent flowing through the MOSFET which
of the FET allows the use of a high-value potenti- increases as the voltage is increased. The value
ometer such as 100 k or higher. of the potentiometer is not critical, and the high
input impedance of the MOSFET allows the use
Ideas for Exploration of a high-value potentiometer such as 100 k or
higher. If VR1 ¼ 500 k, then for R1 ¼ 47 k, the
(i) Replace the motor by a 9 volt lamp in order minimum voltage on the gate will be 1 V which
to realize one form of a lamp dimmer. is just below the threshold voltage for the
(ii) Remove the motor and use the system as a MOSFET.
variable bench supply by taking the voltage
output from the source of the MOSFET. How Ideas for Exploration (i) Replace the lamp by a
does the threshold voltage of the MOSFET 12 volt DC motor and experiment with the
affect the operation of the circuit? resulting motor speed controller.
Ideas for Exploration over a wide range, and at the resonant frequency
(i) Derive the equation t ¼ R1 C 1 ln VI
for the of the LC circuit, the current in the microammeter
VT
will increase sharply. The signal frequency and
discharge time of the capacitor through the
amplitude are adjusted until the maximum current
resistor, given the initial voltage VI and the
increase occurs corresponding to almost full-scale
threshold voltage of the MOSFET VT.
deflection on the meter. At resonance, 2πfL ¼ 1/
(ii) Vary the resistor and capacitor values to
2πfC giving L ¼ 109/395f2 in Henrys where f is
achieve different delay times.
the frequency reading on the signal oscillator.
(iii) Experiment with different loads between the
supply and the drain of the MOSFET.
Ideas for Exploration
Research Project 1 (i) Test the system with some known
This project involves the design of a circuit for inductances.
checking inductance values. It utilizes the princi- (ii) Does the use of the JFET confer any particu-
ple of series resonance which occurs in a series lar advantage to the circuit?
connection of an inductor and a capacitor when,
at a particular frequency, the magnitude of the Research Project 2
reactance of the capacitor is equal to the magni- This second research project investigates the
tude of the reactance of the capacitor. In such a internal structure of the MOSFET and in particu-
case, the voltages across these two components lar identifies and utilizes the internal capacitance
cancel and the circuit impedance is at a minimum. between the gate and the source. The data sheet of
Current flow through the circuit is therefore at a the IRF1405 provides information on the gate-
maximum. The circuit in Fig. 3.62 operates on source capacitance for this device. The project
this basis. The JFET is connected in the source involves utilizing this device in a touch-sensitive
follower mode with a signal of variable frequency switch that exploits this junction capacitance as
applied at the input. The output drives a capacitor shown in Fig. 3.63. It employs the IRF1405
C3 ¼ 0.01 μF and the inductor L to be tested with enhancement MOSFET in the common source
a microammeter inserted to measure current. In configuration. The relay is a 12 V device whose
operation, a sinusoidal signal from an oscillator is contacts connect a lamp or other appliance to the
passed into the system. Potentiometer VR1 mains supply. Diode D1 protects the MOSFET
enables the adjustment of the amplitude of the from back emf from the relay coil. When the
signal. The frequency of the signal is adjusted contacts A are bridged by touch, charge is
In Chaps. 2 and 3, the BJT and FET were and two output terminals. The BJT is considered
introduced and their operation discussed. The first.
need for biasing was established and various bias- There are four external variables, the input
ing schemes presented. Also, the operation of voltage and current Vi and Ii and output voltage
the BJT and FET as amplifiers each in three and current Vo and Io. Since operation takes place
configurations was discussed. While the methods over the linear range of transistor characteristics
utilized gave reasonably good answers, more pre- (small-signal operation), Vi, Ii, Vo and Io can be
cise design requires the use of BJT and FET related to each other by constant parameters. Thus
models or equivalent circuits. The models we Vi and Io can be related to Vo and Ii by four
adopt are the hybrid parameter or h-parameter constants, called hybrid or h-parameters giving
model and the y-parameter model. These transis-
V i ¼ hi I i þ ho V o ð4:1Þ
tor parameters can in general be obtained from
manufacturer’s data sheets. At the end of this I O ¼ h f I i þ ho V o ð4:2Þ
chapter, the student will be able to:
which can be represented as shown in Fig. 4.2.
• Explain h- and y-parameters (Note that there are other ways of relating the four
• Use h- and y-parameters in the analysis of variables. See Sect. 4.5.)
transistor circuits Under small-signal conditions,
Vi
hi ¼ j input impedance with short
I i V o ¼0
‐circuited output ð4:3Þ
4.1 H-Parameters and the BJT
Io
hf ¼ j forward current gain with short
I i V o ¼0
Small-signal amplifiers are amplifiers that are ‐circuited output ð4:4Þ
assumed to operate over the linear range of the
Vi
active device (BJT or FET) by using only small- hr ¼ j voltage feedback ratio with open
V o I i ¼0
signal voltages and currents. Therefore the AC ‐circuited input ð4:5Þ
operation may be deduced by representing the
Io
transistor by an equivalent circuit consisting ho ¼ j output admitance with open
V o I i ¼0
of linear signal generators and circuit elements. ‐circuited input ð4:6Þ
In deducing this equivalent circuit, the transistor
is regarded as an active two-port network as The values of these parameters depend on the
shown in Fig. 4.1, i.e. it has two input terminals transistor configuration. In order to distinguish
among the three configurations, i.e. common Now for the common emitter configuration, hie
emitter (CE), common base (CB) and common is the ratio of a small change in input voltage
collector (CC), a second subscript is added to (ΔVi) to the resulting small change in input cur-
each h-parameter. Thus for the CE configuration, rent (ΔIi) which for this configuration is the input
hi becomes hie; for the CB, it is hib; for the CC, it impedance. Hence from Eq. (2.33) in Chap. 2,
is hic; and so on for the other parameters.
hfe
hie ¼ ð4:7Þ
40I C
hfe ¼ β ð4:8Þ
Fig. 4.1 Transistor as an active two-port network The actual value of hfe is a function of the
transistor collector current IC, temperature and
signal frequency. Figure 4.3 shows a typical plot
of hfe against collector current IC(mA).
It can be seen that hfe initially increases with
collector current but reaches a maximum at
around 20 mA. Manufactures generally state the
maximum value of hfe and the associated collector
current at which it attains this value. A plot of
current gain against temperature is shown in
Fig. 4.4 where hfe increases with temperature by
Fig. 4.2 Transistor h-parameter equivalent circuit as much as twice its value over a temperature
hfe
150
C urrent G ain
100
50
IC (mA)
0
1 10 100 1000
Collector Current
Fig. 4.3 Transistor current gain against collector current
4.2 Analysis of a Common Emitter Amplifier Using H-Parameters 129
change from 20 to 80 C. Finally, the current gain 4.2 Analysis of a Common Emitter
hfe of a transistor varies with signal frequency, its Amplifier Using H-Parameters
value falling off at higher frequencies. This is
illustrated in Fig. 4.5. fβ is the frequency at We are now ready to use the h-parameter equiva-
which the current gain falls by 3 dB of its lent circuit to analyse an amplifier circuit. Specif-
low-frequency value and is referred to as the ically, we wish to study the CE amplifier
cut-off frequency of the transistor. The frequency configuration. Now the h-parameter model is
at which the current gain falls to unity is called the valid only for linear systems and since the tran-
transition frequency ft. sistor characteristics are nonlinear, the signals
must be restricted to small voltage and current
150 variations about the Q point such that the transis-
tor operates approximately linearly. In such
circumstances, the h-parameters can be used to
Current Gain
100
analyse transistor networks for the small time-
varying components of signals and not the DC
50
component.
Consider the basic CE circuit shown in
0
-100 -50 0 50 100 150 200 Fig. 4.6 in which the biasing components have
Temperature (K) been omitted for simplicity. In an AC equivalent
circuit, the battery is a short circuit to AC. Thus
Fig. 4.4 Transistor current gain against temperature
replacing the transistor by the h-parameter equiv-
alent circuit yields Fig. 4.6(a) which becomes
|hFE|
Fig. 4.6(b) after rearranging. This is called the
1000 AC equivalent circuit for the CE amplifier.
Since hre 0 and assuming 1/hoe RL, then
100 this circuit can be reduced to that shown in
Fig. 4.7.
All of the linear circuit laws including those of
10 Ohm, Thevenin, Norton, Kirchhoff and the
Superposition theorem can now be applied since
1 the non-linear transistor has been linearized about
0.1 1 10 100 1000 10000 the operating point. We can now derive an
Frequency (MHz) expression for the voltage gain Av of the amplifier,
where Av ¼ Vo/Vi.
Fig. 4.5 Transistor current gain against frequency
(a) (b)
Fig. 4.6 Common emitter amplifier and its equivalent circuit
130 4 BJT and FET Models
(a) (b)
Fig. 4.8 Fixed-bias common emitter amplifier with equivalent circuit
4.2 Analysis of a Common Emitter Amplifier Using H-Parameters 131
fixed bias case, but the input impedance becomes resistor RS connected in series with the input as
Zi ¼ R1kR2khie. The current gain changes since shown in Fig. 4.15, the equivalent circuit is
some of the signal current is lost in the bias shown in Fig. 4.16. RS in conjunction with the
resistors R1 and R2. The fraction that flows into resistors R1, R2 and hie form a potential divider
the base Ib/Ii is using the current divider theorem and hence introduce voltage signal alternation.
given by The voltage gain AV is given by
Ib R1 kR2 Vo Vo Vb
¼ ð4:19Þ AV ¼ ¼ ð4:21Þ
I i hie þ R1 kR2 Vi Vb Vi
Io Io Ib R1 kR2 V o hfe RL
Ai ¼ ¼ ¼ hfe ð4:20Þ ¼ ð4:22Þ
Ii Ib Ii hie þ R1 kR2 Vb hie
where
Fig. 4.16 Equivalent circuit of common emitter amplifier with input resistor
27 k==173 k
Fig. 4.17 Voltage divider-biased common emitter 2:5 kþ27 k==173 k ¼ 90: From (4.24) the input
amplifier impedance is given by Zi ¼ R1//R2//hie ¼ 27 k//
173 k//2.5 k ¼ 2.3 k. The output impedance is
Z 0i ¼ R1 ==R2 ==hie ð4:24Þ simply the 9k load resistor.
Fig. 4.19 Common emitter amplifier with partially From Fig. 4.20, this impedance is clearly in
decoupled emitter resistor parallel with R1 and R2 and hence
Z i ¼ R1 kR2
hie þ 1 þ hfe Re ð4:33Þ amplifier in Chap. 2 yields RE ¼ 2.4 k, RL ¼ 10.8 k,
R1 ¼ 209 k and R2 ¼ 31 k. Using Eq. (4.29) for
Comparing this circuit with the CE amplifier the voltage gain of the partially bypassed config-
with completely bypassed emitter resistor, firstly, uration and noting that re ¼ 1/40IC ¼ 25 Ω,
h R
the voltage gain magnitude goes from fehie L ¼ 50 ¼ 10800
Re þ25 giving Re ¼ 191 Ω. Thus the emitter
RL RL
re to re þRe , i.e. the presence of Re significantly resistor RE ¼ 2.4 k must be split into RE ¼ 2.2 k
reduces the voltage gain. Secondly the input and Re ¼ 191 Ω. The input impedance is given by
impedance changes from hie to hie + (1 + hfe)Re Zi ¼ 31 kk209 kk{2.5 k + (1 + 100)
(ignoring the effect of R1 and R2), i.e. Re causes a 0.191 k} ¼ 31 k//209 k//21.8 k ¼ 12.1 k
significant increase in the input impedance,
looking into the transistor base. In practice, how- Bootstrapping
ever, R1 and R2 limit the input impedance In order to reduce the effect of R1 and R2 and
increase. thereby realize higher input impedance,
bootstrapping can be employed in the partially
Example 4.4 bypassed circuit of Fig. 4.19. The technique
In the circuit of Fig. 4.17, consider the case where involves (i) the introduction of resistor RB
the emitter resistor is split such that only 1.6 k is between the transistor base and the junction of
bypassed. Determine the voltage gain and input R1 and R2 and (ii) the introduction of a capacitor
impedance. CB between the transistor emitter and the junction
of R1, R2 and RB as shown in Fig. 4.21. As usual,
Solution Using the circuit values of Fig. 4.17 all capacitors are assumed to be large enough to
and the nomenclature of Fig. 4.19, Re ¼ 400 and be short circuit to AC. Using the simplified model
RE ¼ 1.6 k. Then from (4.29) the voltage gain is of the BJT, the equivalent circuit representing the
9
given by AV ¼ 0:4þ0:025 ¼ 21 . This is signifi- bootstrapped circuit is shown in Fig. 4.22. Let
cantly reduced from the original value of AV ¼ Z i ¼ R1 kR2 kRe ¼ R0e . From KVL,
9
0:025 ¼ 360 as a result of the presence of the I b hie ¼ I x RB ð4:34Þ
unbypassed 400 Ω resistor. This resistor also
affects the input resistance which becomes giving
Zi ¼ 27 kk173 kk{2.5 k + (1 + 100)
0.4 k} ¼ 27 k//173 k//42.9 k ¼ 11 k.
Example 4.5
Using the partially bypassed configuration in
Fig. 4.19, design a common emitter amplifier
having a gain of 50, using a 2N3904 npn transis-
tor and a 24 V power supply. Determine the input
impedance.
V o ¼ hfe I b RL ð4:39Þ
This reduces to
hfe RL
AV ¼ , RB hie ð4:41Þ
hie þ 1 þ hfe R0e
V i ¼ I b hie ð4:42Þ
V o ¼ hfe I b I F RL ð4:43Þ
I F ¼ ðV o V i Þ=RF ð4:44Þ
where Yb ¼ Ib/Vi and YF ¼ IF/Vi are the input
Substituting (4.44) in (4.43), we get admittances making up Yi. From (4.42), Yb ¼ Ib/
Vi ¼ 1/hie. From (4.44),
RL RF hie hfe RF
V o ¼ Ib I F ¼ ðV o V i Þ=RF ¼ ðAV 1ÞV i =RF ð4:50Þ
RL þ RF RF
hie hfe RF Hence,
¼ I b RL ==RF ð4:45Þ
RF
Y F ¼ I F =V i ¼ ð1 AV Þ=RF ð4:51Þ
Hence the voltage gain is given by
Therefore
V o RL ==RF hie hfe RF =RF
AV ¼ ¼ Ii 1 1 AV
Vi hie Yi ¼ ¼ Yb þ YF ¼ þ ð4:52Þ
Vi hie RF
hfe RF
¼ RL ==RF 1 =RF ð4:46Þ This corresponds to two impedances in paral-
hie
lel given by
Since hfeRF/hie 1, (4.46) reduces to
Z i ¼ 1=Y i ¼ hie ==RF =ð1 AV Þ ð4:53Þ
AV ¼ hfe RL ==RF =hie ð4:47Þ
(a) (b)
Fig. 4.31 Common collector amplifier circuit (a) and equivalent circuit (b)
4.3 Analysis of the Common Collector Amplifier Using H-Parameters 141
kk10 k) ¼ 844 k which is significantly larger than Hence the voltage gain is given by
the original value of 47 k.
V o I b hfe RL hfe RL RL
Av ¼ ¼ ¼ ¼ ð4:76Þ
Vi I b hie hie re
4.4 Analysis of the CB Amplifier
Note that the voltage gain for the common
Using H-Parameters
base amplifier has the same magnitude as that
for the common emitter amplifier, but unlike the
Here we analyse the third configuration, namely,
common emitter amplifier, the output and input
the common base amplifier, using common emitter
voltages are in phase. In the common emitter
h-parameters. A biased common base amplifier is
amplifier, the output voltage is 180 out of
shown in Fig. 4.36. Assuming hreVce 0 and 1/hoe
phase with the input voltage (represented by a
is large, the equivalent circuit reduces to that shown
negative sign), while in the common base ampli-
in Fig. 4.37. The input voltage Vi is given by
fier the output voltage is in phase with the input
V i ¼ I b hie ð4:74Þ voltage.
To calculate the input impedance, Zi, we eval-
while uate the input current I flowing into the emitter of
V o ¼ I b hfe RL ð4:75Þ the BJT:
I i ¼ I b 1 þ hfe ð4:77Þ
V i ¼ I b hie ð4:78Þ
hence
Vi hie
Zi ¼ ¼ ¼ re ð4:79Þ
Ii 1 þ hfe
Fig. 4.38 Current-driven common base amplifier Fig. 4.40 Common base amplifier for Example 4.11
(a) (b)
Fig. 4.44 JFET common source amplifier (a) and equivalent circuit (b)
Chap. 3.
If part of the resistor RS of the FET is
un-bypassed or if capacitor C3 is removed, then
the equivalent circuit becomes that shown in
Fig. 4.45. If we assume rd RL, then
V o ¼ gm V GS RL ð4:92Þ
V i ¼ V GS þ gm V GS RS ð4:93Þ
Fig. 4.45 Equivalent circuit of common source amplifier
with un-bypassed source resistor
Hence
Vo gm V GS RL gm RL Thus, as in the BJT, inclusion of an
¼ ¼ ð4:94Þ un-bypassed source resistor RS reduces the volt-
Vi ð1 þ gm RS ÞV GS 1 þ gm RS
age gain of the common source FET amplifier.
4.7 Analysis of the Common Drain JFET Amplifier 147
Solution Using Eq. (4.91) the voltage gain is Hence the voltage gain AV becomes
given by AV ¼ gmRL ¼ 5 0.966 ¼ 4.83.
Vo gm V GS RS
The input impedance is set by resistor AV ¼ ¼
RG ¼ 1 MΩ. V i ð1 þ gm RS ÞV GS
gm RS
¼ ð4:97Þ
Example 4.13 1 þ gm RS
If the bypass capacitor in Example 4.12 is
If gmRS 1, then AV 1. That is, the gain of
removed, determine the changed value of the
the common drain amplifier is approximately
voltage gain.
unity which means that the source follows
the gate.
Solution Using Eq. (4.94), the changed voltage
In order to evaluate the output impedance Zo
gain is given by
looking in at the source, we find the ratio of the
gm RL 5 0:966 open-circuit output voltage and short-circuit cur-
AV ¼ ¼ ¼ 1:45:
1 þ gm RS 1 þ 5 0:468 rent, i.e. Zo ¼ Vos/Isc. The open-circuit voltage is
given by
V oc ¼ AV V i ð4:98Þ
4.7 Analysis of the Common Drain and the short-circuit current is given by
JFET Amplifier
I sc ¼ gm V GS ¼ gm V i ð4:99Þ
The basic circuit of a common drain or source
since, when the output is short-circuited, VGS ¼ Vi.
follower JFET amplifier is shown in Fig. 4.46.
Hence
The equivalent circuit is shown in Fig. 4.47. The
output voltage is given by V os gm RS Vi
Zo ¼ ¼
I sc 1 þ gm RS gm V i
RS
¼ ð4:100Þ
1 þ gm RS
Example 4.14
For the circuit of Fig. 4.46, R1 ¼ 2 MΩ,
Fig. 4.46 JFET common drain amplifier R2 ¼ 1 MΩ, RS ¼ 3 kΩ and gm ¼ 5 mA/V find
148 4 BJT and FET Models
the voltage gain, input impedance and output is taken from the drain with the gate grounded.
impedance. This circuit is generally used in high-frequency
applications. It is analogous to the common base
Solution Using (4.97) the voltage gain is BJT amplifier. For this circuit, the output and
gm RS
given by AV ¼ 1þg RS ¼ 1þ53 ¼ 0:94: The input
53 input voltages are given by
m
amplifier. The input impedance Zi at the source is positive and negative for this kind of MOSFET.
given by The equivalent circuit is similar to that for the
JFET and is also shown in Fig. 4.49.
Vi V GS 1
Zi ¼ ¼ ¼ ð4:104Þ The voltage gain is easily shown to be
Ii gm V GS gm
Vo
Actually, the resistor RS appears in parallel AV ¼ ¼ gm rd ==RL gm RL , r d RL
Vi
with 1/gm giving ð4:106Þ
1
Zi ¼ ==RS ð4:105Þ If an un-bypassed bias resistor RS is included
gm
in the source circuit, then the voltage gain
becomes
Example 4.15
gm r d ==RL
For the circuit shown in Fig. 4.48 with RL ¼ 966 Ω, AV ¼ ð4:107Þ
1 þ gm R S
RS ¼ 468 Ω and gm ¼ 5 mA/V find the voltage
gain and the input impedance.
Example 4.16
Solution Using Eq. (4.103), the voltage gain is For the circuit shown in Fig. 4.49 with RL ¼ 5 k,
given by AV ¼ gmRL ¼ 5 0.966 ¼ 4.83. From RG ¼ 10 MΩ and gm ¼ 5 mA/V find the voltage
(4.105) the input impedance given by Zi ¼ 1/gm// gain and the input impedance.
RS ¼ 1/5//0.468 ¼ 140 Ω.
Solution Using Eq. (4.106), the voltage gain is
given by AV ¼ gmRL ¼ 5 5 ¼ 25. The
4.9 Depletion MOSFET Common input impedance is set by resistor RG ¼ 10 MΩ.
Source Amplifier
This device may be operated in the same man-
The depletion MOSFET operates in a manner ner as a JFET as shown in Fig. 4.50. Here RS is
similar to the JFET. In the circuit shown in included in order to reduce the drain current to a
Fig. 4.49, a common source MOSFET is consid- value less than IDSS. Because of the bypass capac-
ered. The circuit uses no gate-source bias and itor C3, the equivalent circuit is the same as that in
therefore the drain current is equal to IDSS. This Fig. 4.49, and therefore the expressions for volt-
is allowable since the gate voltage can be both age gain and input impedance are also the same.
Fig. 4.53 Depletion MOSFET Common Gate Amplifier and Equivalent Circuit
Another biasing arrangement that can be used 1 1 1
Vo þ ¼ Vi gm ð4:119Þ
to bias the enhancement MOSFET is the drain- r d ==RD RG RG
gate feedback bias shown in Fig. 4.56. The equiv-
alent circuit for this configuration is shown in from which the voltage gain AV is given by
Fig. 4.57. In order to determine the voltage gain
of the circuit, we write the node equation for the
drain terminal:
Vo
I i ¼ gm V gs þ ð4:117Þ
r d ==RD
Vi Vo Vo
¼ gm V i þ ð4:118Þ
RG r d ==RD
which yields
R gm RG gm
Vo
1 1 Solution From (4.116) the transconductance gm
AV ¼ ¼ 1G ¼ is given by gm ¼ 2k(VGS VGST) ¼ 2 0.31
V i R þ r ==R
1 1
R ==r ==R
G d D G d D
103(10 3.5) ¼ 4.03 mA/V. From (4.120), the
1 expression for the voltage gain of the common
gm RG ==r d ==RD , gm ð4:120Þ
RG source amplifier in Fig. 4.58 is given by AV ’
gmRD, RD RG, rd. Therefore AV ¼ 4.03 5 k
In order to find the input impedance to this
¼ 20.2. From (4.124), the input impedance is
circuit, we use I i ¼ V iRV o
to get RG þRD
G
Z i ¼ 1þg RD , r d RD . This gives Z i ¼ 1þ20:2 ¼
8M
m
V o ¼ V i I i RG ð4:121Þ 377 k.
Substituting (4.121) in (4.117) gives
Vi RG 4.13 Enhancement MOSFET
I i ¼ gm V i þ Ii ð4:122Þ
r d ==RD r d ==RD Common Drain Amplifier
RG 1
Ii 1 þ ¼ V i gm þ The circuit for an enhancement MOSFET common
r d ==RD r d ==RD
drain amplifier is shown in Fig. 4.59, while the
ð4:123Þ equivalent circuit is shown in Fig. 4.60. Apart from
the need to exceed the threshold voltage in order to
Therefore, the input impedance Zi ¼ Vi/Ii is
turn on the MOSFET, the arrangement is similar to
given by
the common drain circuit using the depletion
RG þ r d ==RD MOSFET and therefore has similar characteristics.
Z i ¼ V i =I i ¼ ð4:124Þ
1 þ gm r d ==RD For this circuit,
V o ¼ gm V GS RS ð4:125Þ
Example 4.17
V i ¼ V GS þ V o ¼ V GS ð1 þ gm RS Þ ð4:126Þ
The MOSFET shown in Fig. 4.58. has
k ¼ 0.31 103 A/V2, VGST ¼ 3.5 V and Therefore the voltage gain is therefore given by
VGSq ¼ 10 V. Determine the voltage gain and
input impedance of the circuit. Vo gm V GS RS
AV ¼ ¼
V i ð1 þ gm RS ÞV GS
gm RS
¼ ð4:127Þ
1 þ gm RS
Fig. 4.60 Equivalent circuit of common drain amplifier using enhancement MOSFET
(a) (b)
Fig. 4.61 Enhancement MOSFET Common Gate Amplifier and Equivalent Circuit
The voltage gain is given by drain on the battery. For maximum symmetrical
swing, V R1 ¼ 9=2 ¼ 4:5 V. Hence R1 ¼ 4.5 V/
Vo
AV ¼ ¼ gm R L ð4:130Þ 0.3 mA ¼ 15 k. The associated base current is 0.3/
Vi
100 ¼ 3 uA and therefore R2 ¼ (4.5 0.7)/
and the input impedance is given by 3 μA ¼ 1.3 M. The input resistor Rm sets the
sensitivity of the circuit. The relationship between
1
Zi ¼ ==RS ð4:131Þ Vi and the meter current Im can be found using the
gm equivalent circuit for the system shown in
Fig. 4.63. After manipulation, the input resistor
V
Rm is given by Rm ¼ 0:9 I fsdfsd where Vfsd is the
Fig. 4.68 EMOSFET source follower the milliammeter. Hence the current gain β is
given by β ¼ IC mA 103/10 μA 106 ¼ 100IC
where IC is the actual reading on the milliammeter
in milliamperes. Therefore, the current gain of the
transistor being tested is obtained by multiplying
the reading on the milliammeter by 100. Resistor
R3 ¼ 120 Ω and diode D1 protect the meter in the
event that a shorted transistor is introduced.
R2 ¼ 330 Ω limits the battery current under such
a condition.
Problems
the partially bypassed configuration. Use this 10. Using a supply voltage of 16 volts, design a
technique to increase the input impedance of collector-base feedback biased common
the design in question 6. emitter circuit with fixed gain of 12 using
8. Using the configuration in Fig. 4.74, design a an emitter resistor and a transistor with
common emitter amplifier having a gain of hfe ¼ 150.
25, using an npn transistor and a 12 V power 11. For the common collector amplifier shown in
supply. Fig. 4.76 determine the voltage gain, input
9. The common emitter amplifier in Fig. 4.75 is impedance and output impedance.
biased for maximum symmetrical swing. 12. Improve the input impedance of the common
Determine the voltage gain for the circuit collector amplifier in question 11 using
and the current gain of the transistor. bootstrapping.
13. For the common base amplifier shown in
Fig. 4.77, determine the voltage gain, input
impedance and output impedance.
Fig. 4.74 Circuit for Question 8 Fig. 4.76 Circuit for Question 11
Fig. 4.75 Circuit for Question 9 Fig. 4.77 Circuit for Question 13
160 4 BJT and FET Models
14. For the circuit shown in Fig. 4.78, draw the V, find the voltage gain, input impedance and
equivalent circuit. For RL ¼ 2 k, RS ¼ 1 k, output impedance.
RG ¼ 2 M and gm ¼ 8 mA/V, calculate the 18. For the common gate amplifier shown in
voltage gain and input impedance of the Fig. 4.80 if RL ¼ 3.2 k, RS ¼ 2.5 k and
circuit. gm ¼ 5 mA/V, find the voltage gain and the
15. If the bypass capacitor is removed from the input impedance.
circuit, determine the new value of the 19. The common source depletion mode
voltage gain. MOSFET in Fig. 4.81 has RL ¼ 10 k,
16. Draw the equivalent circuit of the common RG ¼ 10 M and gm ¼ 9 mA/V. Draw the
drain JFET amplifier in Fig. 4.79 and derive equivalent circuit and determine the
its voltage gain. voltage gain.
17. For the circuit of Fig. 4.79, R1 ¼ 1.8 MΩ, 20. Draw the equivalent circuit for the circuit in
R2 ¼ 1.5 MΩ, RS ¼ 2.2 kΩ and gm ¼ 4 mA/ Fig. 4.81 with a bias resistor inserted in the
source circuit.
Fig. 4.79 Circuit for Question 16 Fig. 4.81 Circuit for Question 19
Bibliography 161
In this chapter, several transistor circuit required than is achievable using a single device,
configurations are considered. These circuits gen- then a cascade connection of two single stage
erally involve more than one transistor and can be amplifiers can be used.
implemented in a wide range of applications as Consider the block diagram of a cascade
well as perform some special functions. In this amplifier shown in Fig. 5.1. Here the output of
chapter, the design and application of many of amplifier A1 feeds directly into the input of ampli-
these circuits will be discussed. After completing fier A2. The overall gain G of the cascaded
the chapter, the reader will be able to: system is
Vo Vo V1
• Explain the operation of cascaded amplifiers, G¼ ¼ ¼ G1 G2 ð5:1Þ
Vi V1 Vi
in particular two common emitter amplifiers,
and determine overall voltage gain and other where G1 is the gain of amplifier A1 and G2 is the
characteristics gain amplifier A2. Equation (5.1) indicates that
• Design several multitransistor amplifier the gain G of the cascaded amplifier is the product
circuits using BJTs and FETs of the gains of the individual amplifiers. This
• Explain the operation of and utilize a range of holds true for any number of amplifiers. How-
special transistor circuits in practical ever, Eq. (5.1) assumes no interaction between
applications the cascaded amplifiers, that is, the gain amplifier
• Design simple electronic switches using BJTs A1 is unaffected by amplifier A2 and vice versa.
and FETs This assumption holds true for operational
• Utilize the JFET as a voltage-controlled amplifiers since the output impedance of these
resistor devices is quite low. However, it does not neces-
sarily hold for voltage amplifiers based on dis-
crete BJTs where the output impedance may be
5.1 Cascaded Amplifiers higher. In multi-stage amplifiers using BJTs,
there is considerable interaction between these
The single transistor amplifier has been consid- stages. In particular, the input impedance of the
ered in Chaps. 2, 3 and 4. The BJT single transis- succeeding stage changes the effective value of
tor amplifier produces a higher voltage gain than the load of the preceding stage thereby changing
its FET counterpart, because of the higher its gain.
transconductance of the BJT as compared with Consider the two-stage amplifier in Fig. 5.2
the FET. In either case, if a higher voltage gain is comprising two cascaded common emitter
amplifiers. Here capacitor C2 couples the output G ¼ G1 G2 ¼ gm1 gm2 RL1 RL2 ð5:4Þ
of the first amplifier built around transistor Tr1 to
the input of the second amplifier built around
transistor Tr2. Because this capacitor and other This gain can be as large as 1000 and higher.
resistors are involved, this type of coupling is Thus, despite the loading effect of Tr2, the cas-
referred to as RC coupling. The voltage gain G1 cading of two BJT amplifier stages can produce
of the first stage for a signal going from the base high gains.
to the emitter of Tr1 is given by
Example 5.1
G1 ¼ gm1 RL1 ð5:2Þ Draw the equivalent circuit of the two-stage
amplifier in Fig. 5.3. Determine the voltage gain
where RL1 is the effective load of Tr1. Because of
Vo/Vi and hence the current gain Io/Ii. Assume
the presence of Tr2, this load is made up of RL1
hfe ¼ 100 for both transistors.
along with the load presented by the input to Tr2
which involves R3, R4 and hie2 associated with
Solution The equivalent circuit is shown in
Tr2. [It will be shown shortly that RL1 ¼
Fig. 5.4. The voltage gain is given by VVoi ¼
RL1 ==R3 ==R4 ==hie2 .] Thus the voltage gain of
Tr1 is reduced because of the loading effect AV1 AV2 where AV1 is the voltage gain of Tr1
resulting from the input to Tr2. However, the and AV2 is the voltage gain of Tr2. In order to
gain G2 of Tr2 is given by proceed, the values of hie1 for Tr1 and hie2 for Tr2
must be determined. Let VB1 and VB2 be the DC
G2 ¼ gm2 RL2 ð5:3Þ voltages at the bases of the two transistors. Then,
because of the voltage divider, V B1 ¼
and the overall gain G is given by
153þ47 20 ¼ 4:7 V . Hence VE1 ¼ 4 and
47
AV1 ¼ 40IC1RL1 ¼ 40 1 2.8 ¼ 112. Vo/Vs and hence the current gain Io/Is. Assume
The load RL2 of the second transistor is hfe ¼ 120 for both transistors.
RL2 ¼ 10 k//8 k ¼ 4.4 k. Therefore, the voltage
gain AV2 of Tr2 is given by Solution The equivalent circuit is shown in
AV2 ¼ 40IC2RL2 ¼ 40 0.5 4.4 ¼ 88. Fig. 5.6. The voltage gain is given by VV oS ¼
Hence the overall voltage gain AV is ρ AV1 AV2 where ρ is the factor by which
AV ¼ AV1 AV2 ¼ 112 88 ¼ 9856. In the input signal is attenuated before it arrives at
order to determine the current gain Io/Ii, we note the base of Tr1, AV1 is the voltage gain of Tr1 and
that VVoi ¼ IIoi RZLi where RL is the 8 k resistor into AV2 is the voltage gain of Tr2. In order to proceed,
which Io flows and Zi is the input impedance at the the values of hie1 and hie2 must be determined. Let
base of Tr1. Zi at the base of Tr1 is Zi ¼ 47 k// VB1 and VB2 be the DC voltages at the bases of the
153 k//hie1 ¼ 47 k//153 k//2.5 k ¼ 2.3 k. two transistors. Then, because of the voltage
Rearranging gives IIoi ¼ VVoi RZLi ¼ 9856 2:38k ¼
k divider, V B1 ¼ 300
37
30 ¼ 3:7 V: Hence VE1 ¼ 3
2834. and IC1 ¼ 3/6 ¼ 0.5 mA. Thus hie1 ¼ 400:5120
¼ 6k.
Similarly, V B2 ¼ 150 30 ¼ 4:7V
23:5
giving
Example 5.2 VE2 ¼ 4, IC2 ¼ 4/(3.6 + 0.4) ¼ 1mA and hie2 ¼
Draw the equivalent circuit of the two-stage
401 ¼ 3 k . The load RL1 of Tr1 is RL1 ¼ 6 k//
120
amplifier in Fig. 5.5. Determine the voltage gain
166 5 Multiple Transistor and Special Circuits
Here the common emitter amplifier Tr1 is Solution In order to lower the voltage gain to
followed by a common collector amplifier Tr2. 50, an un-bypassed resistor R6 is introduced in the
A feature of this circuit is the use of the collector emitter of Tr1 as shown in Fig. 5.8. The voltage
of Tr1 to bias transistor Tr2 by direct connection, gain now becomes AV ¼ R6Rþr 3
e
where re ¼ 1/
thereby eliminating the need for biasing 40IC ¼ 1/40 ¼ 25 Ω. Hence 50 ¼ R69þ25
k
giving
components and capacitive coupling between
R6 ¼ 155 Ω.
the two transistors. Such an arrangement is
referred to as direct coupling and, when used
An emitter follower can also be used to pre-
throughout an amplifier, confers the advantage
serve the gain of a single FET amplifier while
of enabling the amplifier to respond down to
providing a reduced output impedance. Such a
DC. Since the input impedance Zi2 ¼
circuit is shown in Fig. 5.9. The FET Tr1 is a
[hie2 + (1 + hfe2)R5] to Tr2 is high, the load on
common source amplifier with a high input
Tr1 is small so that the effective collector resis-
impedance, the exact value for which is
tance for Tr1 is R3//Zi2 R3. Hence the gain of the
established by bias resistor R1.
first stage is preserved at G1 ¼ gm1R3. Since the
second stage is an emitter follower, the gain of
Example 5.5
this stage is approximately unity and therefore the
Design a two-stage amplifier using the topology
overall gain G is G ¼ gmR3. This circuit has the
of Fig. 5.9.
additional advantage of low output impedance
because of the common collector output ampli-
fier. Its value Zo is given by
hie2 þ R3
Zo ¼ ==R5 ð5:5Þ
1 þ hfe
Example 5.3
Using the two-stage design in Fig. 5.7, design a
two-stage amplifier having the ability to drive
low-impedance loads.
Solution The design of the first stage follows the Fig. 5.8 Common emitter amplifier with reduced gain
steps for the standard common emitter stage. For
a 20-volt supply, the design approach of section
(Sect. 2.4 in Chap. 2) yields R1 ¼ 174 k,
R2 ¼ 26 k, R3 ¼ 9 k and R4 ¼ 2 k. For this circuit,
the collector voltage of Tr1 is 11volts. This is
therefore the base voltage of Tr2 resulting in an
emitter voltage for Tr2 of 10.3 volts. Therefore if
we choose a current of 10 mA for Tr2 in order to
ensure a large current drive capability, then
R5 ¼ 10.3/10 mA ¼ 1 k.
Example 5.4
Modify the design in Example 5.3 in order to set
Fig. 5.9 Common source JFET amplifier with emitter
the voltage gain at 50. follower output
168 5 Multiple Transistor and Special Circuits
As a result, bias stability for the two transistors is important for Tr2 than Tr1. Applying the design
achieved since any quiescent current change in techniques of Chap. 2 for a common emitter
either transistor is corrected by the loop. Note that amplifier, let the emitter voltage be VCC/
for signals, capacitor C3 grounds the emitter of 10 ¼ 20/10 ¼ 2 V. Choosing 2 mA quiescent
Tr2 thereby disconnecting the feedback loop. current for this stage, then R3 ¼ 2 V/2 mA ¼ 1 k.
The equivalent circuit is shown in Fig. 5.13. The For maximum symmetrical swing, the voltage
voltage gain of the first stage is AV1 ¼ drop across R2 is (20 2)/2 ¼ 9 V. Hence
40I C1 R1 khie2 , while the gain of the second R2 ¼ 9 V/2 mA ¼ 4.5 k. The voltage at the base
stage is AV2 ¼ 40I C2 R2 . Hence the overall of Tr2 is the same voltage at the collector of Tr1
gain is the product of these two gains, and this which is 2.7 V. Choosing a quiescent current in
can be quite high. Tr1 of 1 mA, then R1 ¼ (20 2.7)/1 mA ¼ 17.3 k.
Finally, R4 is calculated from (2 0.7)
Example 5.7 R4 ¼ 1 mA/100 where the current gain of Tr1 is
Using the configuration in Fig. 5.12, design a taken as 100. This gives R4 ¼ 130 k. The gain for
direct coupled amplifier with high gain. this circuit is about 76 dB.
I C ¼ β1 I B1 þ β2 I B2 I b2 ¼ 1 þ hfe1 I b1 ð5:14Þ
¼ β1 I B1 þ β2 I E1
ð5:10Þ and therefore (5.13) becomes
¼ β1 I B1 þ β2 ð1 þ β1 ÞI B1
β1 β2 I B1 V i ¼ I b1 hie1 þ 1 þ hfe1 hie2 ð5:15Þ
But
I b2 ¼ 1 þ hfe1 I b1 ð5:25Þ
Hence
V i ¼I b1 hie1 þ 1þhfe1 hie2 þ 1þhfe1 1þhfe2 RE
ð5:26Þ
Fig. 5.19 Circuit for Example 5.10 Hence the voltage gain is given by
174 5 Multiple Transistor and Special Circuits
1 þ hfe1 1 þ hfe2 I b1 RE
AV ¼ V o =V i ¼
hie1 þ 1 þ hfe1 hie2 þ 1 þ hfe1 1 þ hfe2 RE I b1
ð5:28Þ
This reduces to
1 þ hfe1 1 þ hfe2 RE
AV ¼ V o =V i ¼
hie1 þ 1 þ hfe1 hie2 þ 1 þ hfe1 1 þ hfe2 RE
ð5:29Þ
Z i ¼ V i =I b1
¼ hie1 þ 1 þhfe1 hie2 þ 1 þ hfe1
Fig. 5.21 Common collector amplifier using Darlington
1 þ hfe2 RE
pair
’ hfe1 hfe2 RE ð5:30Þ
where the term in square brackets in the denomi- Solution The 2N3904 transistor in the first stage
nator is the input impedance at the base of Tr2. of the feedback pair has hoe1 of 1 to 40 μmhos.
Using the modified Eq. (5.33) for Ib2, the expres- Using hoe1 ¼ 20 μmhos, then 1/hoe1 ¼ 1/
sion for the voltage gain becomes 20 106 ¼ 50 k which is greater than the load
5.3 Feedback Pair 177
Fig. 5.28 Series feedback pair Fig. 5.30 Practical feedback pair with gain
hfe1 hfe2 R5 þ 1 þ hfe1 þ hfe1 hfe2 R4
AV ¼
hie1 þ 1 þ hfe1 þ hfe1 hfe2 R4
ð5:57Þ
AV ¼ 1 þ R5 =R4 ð5:58Þ
Example 5.16
V o ¼ hfe2
I b2 R5 Design a feedback pair using Fig. 5.30 to provide
þ 1 þ hfe1 I b1 hfe2 I b2 R4 ð5:54Þ
a fixed gain of 5 using a 20-volt supply.
Noting that Ib2 ¼ hfe1Ib1, (5.39) reduces to
Solution Let the collector current of each tran-
V o ¼ hfe1h
fe2 I b1 R5 sistor be 1 mA. Then R3 ¼ 0.7 V/1 mA ¼ 700 Ω.
þ 1 þ hfe1 I b1 þ hfe1 hfe2 I b1 R4 ð5:55Þ For bias stability of Tr1, let the emitter voltage be
VCC/10 ¼ 20/10 ¼ 2 V. Then, noting that the
At the input, collector currents of both transistors flow into
V i ¼ I b1hie1 R4, the value of this resistor is R4 ¼ 2 V/
2 mA ¼ 1 k. For maximum symmetrical swing,
þ 1 þ hfe1 I b1 þ hfe1 hfe2 I b1 R4 ð5:56Þ
the voltage at the collector of Tr2 which is the
Therefore, the voltage gain AV ¼ Vo/Vi is given output must be 2 + (20–2)/2 ¼ 11. From this,
by R5 ¼ 9 V/1 mA ¼ 9 k. This gives a gain of
180 5 Multiple Transistor and Special Circuits
Example 5.17
Determine the gain of the JFET-BJT feedback practical current source has a finite output resis-
arrangement in Fig. 5.33. tance R as shown in Fig. 5.34. The value of R is
generally high: R ! 1 corresponds to an ideal
Solution Using AV ¼ 1 + R5/R4, the gain of this current source. While the terminals of a voltage
amplifier is AV ¼ 1 + 104/100 ¼ 101. source should in general not be short-circuited
lest an infinite current flows, the terminals AB
of the current source should in general not be
5.4 Current Sources open-circuited lest an infinite voltage develops
across its terminals. Current sources are widely
A current source is a current supply that maintains used in electronic circuits design and are particu-
its current value, regardless of the load. This larly important in integrated circuit design. They
implies infinite output impedance. However, a can be realized using BJTs and FETS.
5.4 Current Sources 181
(a) (b)
Fig. 5.34 Practical current
source Fig. 5.35 BJT current source with (a) voltage source and
(b) Zener diode
Example 5.18
Calculate the constant current IC in the circuit of
Fig. 5.35 (b) for VZ ¼ 6.2, RZ ¼ 1 k and RE ¼ 5.5 k,
and determine the current through the Zener
BJT Current Source 1 diode.
A constant current source using a BJT is shown in
Fig. 5.35. It consists in Fig. 5.35 (a) of a transistor Solution The voltage VRE across RE is given by
Tr1 with a resistor RE connecting the emitter to VRE ¼ 6.2 0.7 ¼ 5.5 V. Hence IC ¼ 5.5 V/
ground and its base fixed at a potential VBB by a 5.5 k ¼ 1 mA. The current IZ in the Zener is given
suitable voltage source as shown in Fig. 5.35 (a). by IZ ¼ (15 6.2)/1 k ¼ 8.8 mA.
It follows therefore that
V RE ¼ ðV BB V BE Þ ¼ I E RE ð5:59Þ BJT Current Source 2
Another BJT constant current source utilizes two
where IE is the emitter current in the transistor. BJTs. The configuration is shown in Fig. 5.36.
Therefore The base emitter function of Tr2 maintains a
voltage of 0.7 V across the resistor R1 thereby
I E ¼ ðV BB 0:7Þ=RE I C ð5:60Þ yielding a current
I R ¼ 0:7=R1 ð5:61Þ
(a) (b)
Fig. 5.37 JFET constant current source
The circuit shown in Fig. 5.38 is called a current Now Tr1 and Tr2 again have the same collector
mirror (based on a circuit developed by Bob currents IC and base currents IB. For Tr3 the
Widlar) since the current established in Tr1 via emitter current is given by
resistor R1 is mirrored in transistor Tr2 and hence I E3 ¼ βI B þ 2I B ¼ ð2 þ βÞI B ð5:68Þ
flows through R regardless of its value. The cir-
cuit can therefore function as an effective con- Hence,
stant current source. The current through R1 is
I E3 2þβ
given by I B3 ¼ ¼ I IB ð5:69Þ
ð1 þ β Þ 1 þ β B
V cc V BE
I1 ¼ ð5:64Þ Therefore
R1
R1
V R1 ¼ I R1 R1 ¼ 0:7 ð5:75Þ
R2
Therefore
V Z ¼ V R1 þ V R2
R
¼ 0:7 þ 0:7 1
R2
R ð5:76Þ
¼ 0:7 1 þ 1
R2
R
¼ 1 þ 1 V BE
R2
Thus the VBE of Tr1 is multiplied by a factor Fig. 5.42 Cascode amplifier
(1 + R1/R2), setting a voltage VZ across the tran-
sistor. The circuit therefore functions as a Zener hie1
r e1 ð5:78Þ
diode. Note that any increase in IR will increase hfe1
the voltage across R2 which in turn will tend to
The gain of the second-stage Tr2 is given by
increase VBE. This will then turn the transistor
further on thereby increasing IC and hence divert- hfe2 RL RL
ing the additional current through the transistor. AV2 ¼ þ ð5:79Þ
hie2 r e2
from the reduction of the voltage change at the Solution The practical circuit is shown in
collector of Tr1. Fig. 5.43. Here the diodes D1 and D2 fix the
voltage at the base of Tr2 at 1.4 volts, and RD
Example 5.24 supplies current to the diodes. Choose this current
Using the configuration shown in Fig. 5.43, to be about 5 mA, and then RD ¼ 241:4
5 mA ¼ 4:5 k .
design a cascode amplifier circuit. Use Choose ICq(Tr2) ¼ 1 mA . The maximum peak-
VCC ¼ 24 V. to-peak voltage swing for this transistor is
approximately 24 1.4 ¼ 22.6 volts . Therefore,
for maximum symmetrical swing, RL ¼
2 =1 mA ¼ 11:3 k . Using β(Tr1) ¼ 100, RB ¼
22:6
240:7
ð1 mA=100Þ ¼ 2:3 M . Voltage gain for this circuit is
AV ¼ 40ICRL ¼ 40 1 11.3 ¼ 452 .
5.8 Improved Emitter Follower implemented around Tr2 with diodes D1 and D2
setting the fixed voltage across resistor R3 and the
The emitter follower has a low output impedance transistor base-emitter junction. We choose
which enables it to drive low-impedance loads ICq ¼ 2 mA so that the circuit can drive
without a reduction of the output voltage. The low-impedance loads. Then R3 ¼ 0.7 volts/
configuration also has high current drive capabil- 2 mA ¼ 350 Ω. Resistor R2 supplies current to
ity and is therefore widely used in the output stage the diodes, and if we allow about 5 mA, then
of amplifiers. One drawback of the circuit R2 ¼ (15 1.4)/5 mA ¼ 2.7 k. Finally, resistor
discussed in Fig. 2.55 in Chap. 2 is its inability R1 supplies bias current into Tr1 and a value of
to deliver full output voltage swing on both half- about 10 k ensures that the resulting offset voltage
cycles of a sinusoidal waveform. Specifically, an is small.
emitter follower using an npn transistor can
deliver an output voltage into a load RL limited An even better arrangement is the complimen-
only by the power supply voltage. However, on tary emitter follower circuit shown in Fig. 5.47
the negative half-cycle, the output voltage VO is
limited to
V O ¼ I Cq RE kRL ð5:81Þ
Fig. 5.45 Improved emitter follower Fig. 5.47 Complimentary symmetry emitter follower
188 5 Multiple Transistor and Special Circuits
utilizing complimentary transistors. Here the input to Tr2 is set to zero, as the input signal Vi1 to
emitter resistor or constant current source is Tr1 increases positively, the base-emitter voltage
replaced by another transistor in emitter follower drop Vbe1 of Tr1 increases and Vbe2 of Tr2
configuration but having opposite polarity. On the decreases. The result of this is an increase in the
positive half-cycle, the npn transistor as before collector current through Tr1 which, because of
delivers a positive-going output voltage and the constant current source, causes a
sources a high current, while on the negative corresponding decrease in the collector current
half-cycle, the pnp transistor delivers a negative- of Tr2. Therefore, the collector voltage of Tr1
going output voltage and sinks a high current. goes down while that of Tr2 goes up, by the
Proper operation requires that a bias voltage be same magnitude. If Vi1 increases negatively, the
applied between the two transistor base terminals reverse occurs resulting in a decrease in the col-
so that the two transistors are turned on during the lector current of Tr1 and an increase in the collec-
quiescent state. This circuit is discussed more tor current of Tr2. This in turn results in an
fully in Chap. 10. increase in the voltage at the collector of Tr1 and
a decrease in the voltage at the collector of Tr2.
The effect of the input signal Vi1 then is to
5.9 Differential Amplifier produce an inverted signal voltage at the collector
of Tr1 and an in-phase signal voltage at the col-
One of the most important and widely used lector of Tr2. For signals Vi1 and Vi2 at both
circuits is the differential amplifier or “long tail” inputs, (Vi1 Vi2) is the differential input voltage,
pair shown in Fig. 5.48. It comprises two and (Vo1 Vo2) is the differential output voltage.
transistors of the same polarity connected at Since there are two inputs and two outputs, the
their emitters where a constant current source is amplifier is said to have a double-ended input and
connected to supply bias current to each. Each a double-ended output. It is important to note that
transistor is essentially in a common emitter mode the system does not respond to a common mode
with an input signal supplied to its base and an signal, i.e. a signal common to both inputs. To see
output signal taken at its collector. The system this, we tie both inputs together and apply a
operates in the following manner: Assuming the signal, the constant current source prevents any
change in collector current and hence there is no
output signal in response. The circuit may be
simplified by replacing the constant current
source with a resistor from the connected emitters
to the negative supply. Also, if an output is
needed from one collector only, then the resistor
in the collector of the unused output may be
removed and replaced by a short circuit. This
modified circuit is shown in Fig. 5.49. A more
complete discussion of the differential amplifier is
presented in Chap. 8.
Ic (mA)
Saturation
Region Ib = 120µA
14
IC = VCC/RL 12 Ib = 100µA
10 Ib = 80µA
8 Ib = 60µA
6 Q-point
Ib = 40µA
Fig. 5.49 Modified differential amplifier 4
Ib = 20µA
2 Cut-off
across it is quite small or in cut-off (fully off) in Ib = 0µA Region
VCE(V)
which case the current through the device is vir- 0 2 4 6 8 10 12
VCC = ½ VCC VCE = VCC
tually zero. The basic circuit is shown in
Fig. 5.50. Fig. 5.51 Transistor output characteristics showing satu-
As we have seen before in the common emitter ration and cutoff regions
configuration, with the load resistor RL collector
current and collector-emitter voltage operate is never in saturation or in cut-off so that full
along a load line such that if the base current reproduction of the amplified signal is assured.
(or base-emitter voltage) is at zero, the transistor However when operated as a switch, the transistor
is off and the collector voltage is at its maximum is driven into either the fully ON state by supply-
value the supply voltage. The collector current is ing sufficient base current or the fully OFF state
then the very small collector leakage current ICEO. by reducing the base current to zero. This
If the base current is increased, then the collector switching can be conveniently done by the appli-
current increases correspondingly thereby caus- cation of a positive-going voltage pulse to the
ing the collector-emitter voltage to fall. If the base transistor base terminal through the base resistor
current is increased sufficiently, the collector- RB. When the voltage pulse is at zero, no base
emitter voltage falls to a very small value referred current flows into the transistor, and therefore it is
to as the saturation voltage VCEsat which is typi- off. Since there is no collector current, no voltage
cally VCEsat 0.2 V with VBEsat 0.7 V. The therefore appears across RL. When the voltage
collector current is then at its maximum value pulse goes positive, base current is injected into
IC VCC/RL limited only by the resistor RL, and the transistor via the resistor RB, and the transistor
the transistor is said to be in saturation. These is switched on. As a result almost the full supply
ON-OFF regions are shown in the output voltage appears across the load resistor. The value
characteristics in Fig. 5.51. of resistor RB determines the actual value of base
Thus when functioning as an amplifier, the current that flows and must therefore be carefully
transistor operates in the active region where it chosen. The input voltage that switches the
190 5 Multiple Transistor and Special Circuits
VBE(SAT)
Base-
emitter 90%
voltage
10%
0
Time
tON tOFF
VCC
90%
Collector-
emitter
voltage
10%
VCE (SAT)
0
td ts Time
tf tr
collector-emitter voltage from VCC to 0.9VCC is between the removal of the voltage pulse and
referred to as delay time td. During the time the the rise of the collector-emitter voltage to
transistor was off with the base at zero volts, the 0.1VCC is the storage delay tS. After this time the
collector-base junction capacitor CCB was transistor begins to turn off and the collector
charged to VCC through the load resistor RL, and voltage rises. The rate of rise is limited by the
this capacitor voltage appears across the transistor recharging of the collector-base junction capaci-
collector. Therefore, as the collector voltage falls tor which takes place through the load resistor RL.
in response to the base-emitter voltage, this junc- The time taken for the collector voltage to rise
tion capacitor must be discharged through the from 0.1VCC to 0.9VCC is the rise time tR, and the
transistor in order that the collector voltage sum of the storage time and the rise time is called
continues to fall. This further delays the decrease the turn-off time toff. A plot of transistor switching
of the collector-emitter voltage which eventually voltages is shown in Fig. 5.53. Typical values for
reaches its saturation value VCEsat. The time taken the turn-on times is 65 nanoseconds and turn-off
for the collector voltage to fall from 0.9VCC to time is 240 nanoseconds.
0.1VCC is referred to as the fall-time tf, and the
sum of the delay time and the fall time is the turn- BJT Switching Applications
on time ton. Several applications of the transistor switch are
After the collector voltage attains its saturation possible. Two examples are shown in Fig. 5.54
value, the collector current cannot be further and Fig. 5.55. In the first, the transistor is used to
increased, and therefore any additional increase switch on a small incandescent lamp. Here the
in the base current will result in excess charge lamp replaces the load resistor in the basic circuit.
storage in the base region of the transistor. When When the input voltage goes high, the transistor is
the input voltage pulse is reduced to zero, the turned on, and the full supply voltage is placed
need to remove this excess charge results in a across the lamp causing it to light. The current
delay in the turning off of the transistor. The through the lamp on turn-on needs to be known so
base-emitter capacitor CBE must be discharged that the base resistor can be determined. The
through RB so that the base-emitter voltage can transistor must be able to handle the current
fall and turn-off can begin. The time duration load, and many small to medium power
192 5 Multiple Transistor and Special Circuits
ID(mA)
6 VGS=0V
5 Ohmic region
-0.5V
4
3
-1V
2
-1.5V
1
-2V
-2.5V
0 1 2 3 4 5 6 7 8 VDS(volts)
S
V
G
microcontroller or other logic circuit output in 4
order to control high-power systems. However,
5V
the +5 volts may be insufficient to overcome the 3 -0.
threshold voltage of the MOSFET to turn it
on. This problem may be solved by cascading a
2
BJT switch and a MOSFET switch as shown in -1V
Fig. 5.59. Turning off the BJT applies +12 volts
-1.5V
to the gate of the MOSFET and turns it on. The 1
-2V
input of the MOSFET can be damaged by elec-
-2.5V
trostatic discharge. In order to protect the gate
from over-voltage, a Zener diode clipping circuit 0 0.1 0.2 0.3 0.4 0.5 VDS(volts)
D1 can be used as shown in Fig. 5.59. An over
Fig. 5.61 Ohmic or linear region of JFET
voltage at the input will force the Zener diode into
conduction thereby limiting the gate voltage. The
excessive voltage is then dropped across R2. drain-source voltage less than the pinch-off volt-
age represented on the locus, the slope of the
curves and hence the channel resistance vary
with the gate-source voltage.
5.12 Voltage-Controlled Resistor As VGS is made more negative, the curves
become flatter corresponding to a higher resis-
As has been already observed, the FET operates tance. An approximate value of this resistance
as a linear resistor in the ohmic region to the left can be obtained from Shockley’s equation
of the pinch-off locus in Fig. 5.60. By varying the
2
gate-source voltage, the value of this resistance V
I D ¼ I DSS 1 GS ð5:87Þ
can be varied thereby making the FET a voltage- VP
controlled resistor. This action is evident in
Fig. 5.61 for an n-channel JFET where for This resistance rd is given by
5.13 Applications 195
V DS V DS 1
rd ¼ ¼ ð5:88Þ
ID I DSS
2
1 VVP
of 2 mA for Tr2 to enable sufficient drive current voltages, both transistors will be off and hence so
for external loads, then R3 ¼ 3.8 V/2 mA ¼ 1.9 k. will both LEDs. When P1 goes to a high state that
exceeds the turn-on voltage of Tr1 (0.7 V) and the
Ideas for Exploration (i) Re-design the circuit red LED (1.7 V) which is approximately 2.4 V,
using the 2N3906 pnp transistor, and compare the then the red LED will turn on. Tr2 will remain off.
performance of the two circuits. When P1 goes to a low voltage that is less than
VCC 2.8, then there will be sufficient voltage
Logic Probe across the GREEN LED (2.1 V) and the base-
This circuit shown in Fig. 5.65 is a logic probe emitter junction of Tr2 to turn them on. Resistors
that enables the determination of the state of a R3 and R4 serve to limit the current, and resistors
logic signal which can attain either a high state R1 and R2 limit the base current into the
(+5 volts) or a low state (0 volts). Probe P2 is transistors. Using a supply voltage of
connected to the earth of the logic circuit, while VCC ¼ 3 volts, maximum input voltage of
probe P1 is connected to the logic signal. When 5 volts and maximum LED current of 10 mA,
there is no signal at P1 (floating probe), providing then R3 ¼ (5 0.7 1.7)/10 mA ¼ 260 Ω where
VCC is less than the sum of the voltage drops the LED voltage is 1.7 V. For low input voltage of
across the LEDs and the transistor base-emitter zero and allowing an LED current of 2 mA, then
R4 ¼ (3 0.7 2.1)/2 mA ¼ 100 Ω. Resistors R1
and R2 can be about 1 k each in order to protect
the transistor inputs from over-voltage.
selected according to the design rules in Chap. 3. configuration similar to that of Sect. 5.7 using
Note that the gate of Tr1 is held at zero volts by R1 BJTs but biased in a more conventional manner.
and current through R2 causes the voltage at the Transistor Tr1 operates in the normal common
source of Tr1 to go positive thereby reverse- emitter mode. Choosing a quiescent current of
biasing the gate-source junction of Tr1 and limit- 2 mA and using one tenth of the supply voltage
ing the drain current to a selected value. The as the emitter voltage of Tr1 (2 V), resistor
voltage VG(Tr2) at the gate of Tr2 set by R3 and R5 ¼ 1 k. Allowing 1 V across Tr1 for proper
R4 is equal to the drain-source voltage across Tr1. operation, for maximum symmetrical swing, the
Thus, if the voltage at the source of Tr1 is ΔV, voltage across R4 and Tr2 is (20 3)/2 ¼ 8.5 V.
then the voltage at the source of Tr2 is Therefore R4 ¼ 8.5/2 ¼ 4.3 k. Since the resistor
VG(Tr2) + ΔV. Finally, the load resistor R5 is chain R1-R2-R3 is supplying base current to two
determined to ensure maximum symmetrical transistor, we let the current I down this chain be
swing. one fifth (instead of one tenth) of the collector
current giving I ¼ 0.4 mA. Since the emitter of
Ideas for Exploration (i) Modify the circuit by Tr1 is at 2 V, the base of Tr1 is at 2.7 V and hence
using a JFET for Tr2 that has a sufficient gate- R3 ¼ 2.7 V/0.4 mA ¼ 6.8 k. Since the voltage at
source voltage to enable self-biasing of Tr2 and the emitter of Tr2 is 3 V, the base of Tr2 is 3.7 V
the operation of Tr1 in the saturation region. This and therefore the voltage across R2 is 1 V. Hence
would mean that components R3, R4 and C2 R2 ¼ 1 V/0.4 mA ¼ 2.5 k. The voltage across R1
would not be needed and the gate of Tr2 would is 20 3.7 ¼ 16.3 V and hence R1 ¼ 16.3/
be connected to the source of Tr1. 0.4 mA ¼ 40.8 k. The gain of this circuit is
given by 40 2 mA 4.3 k ¼ 344 ¼ 51 dB,
BJT Cascode Amplifier and its upper cut-off frequency was measured at
The amplifier shown in Fig. 5.68 utilizes two 19 MHz.
binary junction transistors in a cascode
Ideas for Exploration (i) Modify the circuit by
using a JFET for Tr2 that has a sufficient gate-
source voltage to enable self-biasing of Tr2 and
the operation of Tr2 in the active region. This
would mean that capacitor C2 would not be
needed though it can be retained to provide sup-
ply filtering to the base of Tr1. The gate of Tr2
would then be connected to the emitter of Tr1.
Touch-Sensitive Indicator
The next system to be discussed is a circuit that
switches on when touched. The circuit is shown
in Fig. 5.69. It consists of two transistors arranged
such that the emitter of the first is connected to the
base of the second (Darlington pair). There are
current limiting resistors in the collector of each
transistor with the second transistor including an
LED in its collector. Touching the base and sup-
ply voltage simultaneously will cause a small
current to flow in the base of Tr1 where it is
amplified by a factor of the transistor current
Fig. 5.68 BJT cascode amplifier gain. This amplified current now flows into the
5.13 Applications 199
base of Tr2 and is amplified further by a factor current gain. This amplification is continued in
equal to the transistor current gain. The result is Tr3 with the result that a tiny current into the
that a tiny current into the circuit at the base of Tr1 circuit at the base of Tr1 is amplified by a factor
is amplified by a factor of about 100 100 ¼ 104 of about 100 100 100 ¼ 106 results in a
and results in a significant current flowing in the significant current flowing in the collector of Tr3.
collector of Tr2. The LED in the collector of Tr2 is The LED in the collector of Tr3 is therefore
therefore illuminated. R2 is determined by the illuminated. R3 is determined by the current
current level required to turn on the LED. Thus, level required to turn on the LED. Thus, for a
for a 9-volt supply and allowing about 15 mA to 9-volt supply and allowing about 15 mA to acti-
activate the LED, R2 ¼ (9 2.1)/15 mA ¼ 460 Ω. vate the LED, R3 ¼ (9 2.1)/15 mA ¼ 460 Ω.
Using the minimum value of the transistor current Using the minimum value of the transistor current
gain, the maximum current in Tr1 is given by gain, the maximum current in Tr2 is given by
15 mA/100 ¼ 0.15 mA. Hence R1 ¼ 9/ 15 mA/100 ¼ 0.15 mA. Hence R2 ¼ 9/
0.15 mA ¼ 60 k. A value of 47 k is suitable. 0.15 mA ¼ 60 k. A value of 47 k is suitable.
Similarly, R1 ¼ 9/0.0015 mA ¼ 6 M, and a value
Ideas for Exploration (i) Replace resistor R2 of 1 M is suitable.
and the LED by a relay coil (with protection
diode), and use the circuit to switch on a mains Nickel-Cadmium Battery Charger
connected device. The circuit on Fig. 5.71 is that of a nickel-cad-
mium battery charger. These batteries require a
Electromagnetic Field Detector constant current for a fixed period for proper
This circuit is an enhancement of that in Fig. 5.69 recharging. The circuit comprises an unregulated
and detects electromagnetic fields associated with supply from a 40 V centre-tapped transformer that
pffiffiffi
house wiring, static electricity and other sources. provides about 20 2 ¼ 28 V across the filter
The circuit is shown in Fig. 5.70 and bears some capacitor C1. A value of 1000 uF ensures very
similarity to a Darlington pair. It consists of three little fall in voltage as currents up to 100 mA are
transistors arranged such that the emitter of the drawn from the supply. Diodes D3 and D4 provide
first is connected to the base of the second and the a fixed voltage at the base of Tr1 such that there is
emitter of the second connected to the base of the a 0.7 V across resistor R2. This results in a con-
third. There are current-limiting resistors in the stant current of 0.7/R2 in the collector of Tr1. The
collector of each transistor with the final transistor value of R2 is determined by the constant current
including an LED in its collector. Any current to be supplied. For 50 mA, R2 ¼ 0.7/50 mA ¼ 14 Ω.
induced in the base of Tr1 is amplified by a factor Resistor R1 supplies current to diodes D3 and D4
of the transistor current gain. This amplified cur- and to the base of Tr1. For 5 mA through these
rent now flows into the base of Tr2 and is diodes, R1 ¼ (28 1.4)/5 mA ¼ 5.3 k. This
amplified further by a factor equal to the transistor constant current source can accommodate
200 5 Multiple Transistor and Special Circuits
be driven by the output of a power amplifier gains. Tr1 collector voltage of 5 V results in the
where the output voltage goes up to 50 V emitter voltage of Tr2 as 4.3 V. For 2 mA in Tr2,
corresponding to power amplifiers with outputs R4 ¼ 4.3 V/2 mA ¼ 2.2 k. Capacitors C1 and C2
just over 100 Watts. are coupling capacitors, while capacitor C3
grounds the base of Tr1 for operation as a com-
Speaker to Microphone Converter mon base amplifier.
The circuit in Fig. 5.73 converts a loudspeaker
into a very sensitive microphone. Transistor Tr1 is Ideas for Exploration (i) Compare this circuit
connected in the common base mode with with the speaker-to-microphone circuit of
collector-base feedback biasing while Tr2 is Chap. 2; (ii) introduce bootstrapping of a fraction
connected as an emitter follower to provide a of R3 in order to increase the gain of the circuit
low output impedance. A current of 1 mA in Tr1 thereby making the system an even more sensi-
means that R3 ¼ 4 k results in 5 V at the collector tive microphone.
of the transistor. Resistor R1 ¼ 500 Ω ensures that
the signal input is not shorted to ground. This Sound to Light Converter
means that the transistor emitter is at 0.5 V. This circuit in Fig. 5.74 converts sound to light
Hence R2 ¼ (9 4 0.7 0.5)/5 μA ¼ 760 k. signals. An electret microphone at the input
A transistor current gain of 200 is assumed. converts sound into an electrical signal that is
Resistor R2 may need to be adjusted for different fed into transistor Tr1 via capacitor C1. Tr1 is
connected as a simple common emitter amplifier
with collector-base feedback. The collector cur-
rent is set at 2 mA to ensure that sufficient drive
current is available to drive the four connected
transistors. For maximum symmetrical swing, the
voltage across R3 is 4.5 V. Hence R3 ¼ 4.5 V/
2 mA ¼ 2.2 k. Resistor R2 ¼ (4.5 0.7)/
0.02 mA ¼ 190 k where a current gain of 100 is
used. The collector of Tr1 is direct coupled to
Tr2-Tr5, each of which has an LED in the collec-
tor and a 1 k resistor in the emitter. This ensures
that the input impedance to these transistor
amplifiers is high (better than 100 k) thereby
allowing Tr1 to drive several such transistors.
Fig. 5.73 Speaker to microphone converter This resistor also serves to define the current in
each of these transistors in response to the signal VR2 zeros the meter for zero input voltage. For a
from Tr1. These leds vary in brightness in 0.5 V signal at the gate of the FET, most of this is
response to the input from the microphone. dropped across the meter circuit. Hence the resis-
tance VR1 in series with the meter is given by
Ideas for Exploration (i) Replace the micro- VR1 ¼ 0.5 V/100 μA ¼ 5 k. This potentiometer
phone by the output from the auxiliary output must be adjusted for full-scale deflection with
from a cell phone, and observe the effect of 0.5 V at the input.
playing music through the system. The lights
should respond to the varying music signal Ideas for Exploration (i) Re-design the system
amplitude. to operate from a 9 V battery for portability.
Fig. 5.76 AC
millivoltmeter
Research Project 1
The circuits considered up to this point generally Fig. 5.77 Low-power amplifier
utilize small signal transistors. This project
involves the design of an amplifier using the Tr2. This will produce about 50 mW into an
design principals presented thus far that will 8 ohm loudspeaker. Some experimentation here
accept signal from a source such as the auxiliary with the value of R3 may be necessary. The power
output of a cell phone and drive a loudspeaker to transistor can be a 2N3055. It should be mounted
give audible output. This is a prelude to the par- on a small heatsink in order to dissipate any heat
ticularly interesting subject of power amplifier produced. Note also that the collector current of
design which will be fully introduced in the power transistor flows through the speaker
Chap. 9. The circuit is shown in Fig. 5.77. It resulting in a permanent displacement of the
comprises a common emitter amplifier using a speaker cone. It should be noted that speakers
small signal transistor Tr1, driving a power tran- with small diameters do not operate well under
sistor Tr2 also connected in the common emitter this condition.
mode. Tr2 drives an 8 ohm loudspeaker
connected in its collector circuit. Resistor R3 Ideas for Exploration (i) Replace the power
should result in about 120 mA in the collector of transistor by a power Darlington such as the
204 5 Multiple Transistor and Special Circuits
MJ3000. Resistor R3 would have to be increased germanium diode is employed here as the input
for the same value of collector current. The signal source to the low-power amplifier of
Darlington will provide a higher input impedance research project 1 above. The combined system
which reduces loading on Tr1 and thereby is shown in Fig. 5.79. Capacitor C3 must be about
increase the sensitivity of the system, i.e. a 470 pF and filters the carrier from the modulated
lower signal voltage will drive the system; radio frequency signal. Capacitor C4 provides
(ii) introduce a volume control by including a power supply decoupling such that the inductance
potentiometer VR1 ¼ 10k at the input. The signal associated with the power leads does not affect
source then drives the potentiometer and the the system performance. Its value can be about
potentiometer wiper is connected to the amplifier 100uF. Resistor R1 in the previous system is here
input as shown in Fig. 5.78; (iii) using the replaced by potentiometer VR1 ¼ 10 k.
MOSFET amplifier of Fig. 3.57 in Chap. 3,
replace the power BJT and implement the power Ideas for Exploration (i) Experiment by reduc-
amplifier. ing the length of the antenna; (ii) introduce a radio
frequency signal booster as shown in Fig. 5.80.
Research Project 2 This simple circuit uses an MPF102 or other
This project is a further development of the crys- suitable JFET in the common source mode. Resis-
tal radio project. The system from Chap. 2 where tor R1 ¼ 1 M ensures that the gate is grounded for
a germanium transistor was used instead of a JFET biasing. C1 ¼ C2 ¼ 1000 pF are coupling
Problems 205
capacitors which can be small since the circuit is replace the long-wire antenna in the earlier
operating at high frequencies (540 kHz to forms of the crystal radio in Chaps. 1 and 2, and
1600 kHz). check if reception is improved.
shown in Fig. 5.8 in order to lower the volt- supply voltage of 26 V, and calculate the gain
age gain to 25. of your circuit.
6. Design a two-stage amplifier having a JFET 9. Explain the principle of bootstrapping, and
input stage using the topology of Fig. 5.9 and use the technique to increase the gain of the
a 25-volt supply. Use a JFET with circuit in Question 8.
VP ¼ 5 volts and IDSS ¼ 5 mA. Determine 10. Replace the transistor in the first stage of
the voltage gain of the circuit, and state the Fig. 5.85 by a JFET, and re-design the circuit.
main advantage of the JFET stage. 11. Design a common emitter amplifier using a
7. Design a low input impedance two-stage pnp Darlington package having βD ¼ 2000
amplifier using the circuit of Fig. 5.10 and a and a 30-volt supply. Calculate the voltage
20-volt supply. Determine the input and out- gain of your circuit.
put impedances as well as the voltage gain. 12. Introduce partial bypassing of the emitter
8. Using the configuration in Fig. 5.85, design a resistor in Question 11 in order to achieve a
direct coupled amplifier with high gain. Use a gain of 50.
Problems 207
13. Show that the current gain of a Darlington 17. Derive the voltage gain of a feedback pair in a
pair is approximately equal to the product of common emitter configuration, and show that
the current gains of the two transistors it is significantly higher than either the
making up the pair. Darlington pair or a single BJT.
14. Design a common collector amplifier using a 18. Design a circuit using BJTs in a feedback pair
Darlington package having βD ¼ 1000 and a to have a gain of 8. Use a supply voltage of
32-volt supply. Calculate the input imped- 15 volts.
ance of your circuit. 19. Repeat the design in Question 18 with a JFET
15. Apply bootstrapping in order to increase the as the input transistor.
input impedance of the circuit and estimate 20. Using the configuration in Fig. 5.86, design a
the value of this impedance. constant current source that produces a cur-
16. Show that current gain of a feedback pair is rent of I ¼ 5 mA from a 15-volt supply.
approximately equal to the product of the 21. Using the basic circuit of Question 20, show
current gains of the two transistors making how a pnp transistor can be used to supply a
up the pair. constant current to a grounded load.
208 5 Multiple Transistor and Special Circuits
The discussions in the previous chapters be large such that these capacitors represented
concerned the mid-frequency performance of an short-circuits to the signal currents passing
amplifier. At these frequencies, the coupling and through them. If however the signal frequency is
bypass capacitors pass the signals virtually unim- sufficiently low, the reactance of these capacitors
peded, while the transistor junction capacitors are becomes significant and can affect signal levels in
considered to be open circuits. The BJT and FET the circuit. In this section, the circuits are
models provided useful tools with which to ana- analysed in order to determine the effect of the
lyse these circuits. The performance at low and capacitors so that appropriate values may be cho-
high frequencies however requires further consid- sen for optimum circuit response. In preparation
eration. In the case of the low frequencies, the for doing so, we consider the RC circuit in
effect of the coupling and bypass capacitors needs Fig. 6.1 with an input sinusoidal signal Vi and
to be determined while at high frequencies the an output signal Vo. The reactance of the capacitor
response which is largely determined by transis- in the complex frequency domain s is 1/sC where
tor junction capacitances needs to be ascertained. s ¼ jω and ω is the angular frequency that is
In this chapter therefore, more complex equiva- related to frequency f by ω ¼ 2πf. The transfer
lent circuits are introduced in order to examine function Vo/Vi for the circuit is given by
the full frequency response characteristics of
Vo R sCR
BJTs and FETs. While the analysis is done AV ¼ ¼ ¼ ð6:1Þ
V i R þ 1=sC 1 þ sCR
using the JFET, it applies in general to the
MOSFET also. After completing the chapter, the
reader will be able to The magnitude of this transfer function is
given by
• Determine the low-frequency response of tran-
sistor amplifiers sCR jsCRj
jAV j ¼ ¼
• Determine the high-frequency response of 1 þ sCR j1 þ sCRj
transistor amplifiers ωCR
¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð6:2Þ
1 þ ω2 C 2 R2
0.1
20dB/dec Therefore, the voltage gain AV is given by
Vo hfe RL I b
AV ¼ ¼
0.01 Vi ðhie þ 1=sC ÞI b
hfe RL sC i
Fig. 6.2 Frequency response plot of CR circuit ¼ ð6:8Þ
1 þ sC i hie
jAV jdB ¼ 20 log ωCR In order to plot the gain magnitude |AV| against
¼ 20 log ω þ 20 log CR ð6:3Þ frequency f, consider the following:
This can be written as hfe RL sC i hfe RL sC i
jAV j ¼ ¼
1 þ sC i hie j1 þ sC i hie j
jAV jdB ¼ 20 log f þ 20 log 2πCR
¼ 20 log f 20 log f o ð6:4Þ hfe RL ωCi
¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð6:9Þ
1 þ ω2 Ci 2 hie 2
where fo ¼ 1/2πRC corresponding to ωo ¼ 1/RC.
Equation (6.4) is that of a straight line on a |AV| When ωCihie > > 1, then
dB logf graph with slope +20 dB/ decade and |
hfe RL ωCi hfe RL
AV|dB ¼ 0 at f ¼ fo as shown in Fig. 6.2. On the jAV j ! ¼ ð6:10Þ
ωC i hie hie
same graph, we have |AV| ¼ 1 for ωCR > 1 which
is equivalent to f > fo. At f ¼ fo, which is the common emitter mid-frequency gain
when Ci is treated as a short-circuit derived in
ωCR 1
jAV j ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffi Chap. 4. The magnitude of this gain converted to
1þω C R 2 2 2 1 þj
pffiffiffi decibels is given by
¼ 1= 2 ¼ 0:707 ¼ 3 dB ð6:5Þ
hfe RL
jAV jdB ¼ 20 log ð6:11Þ
hie
This response is that of a high-pass filter and is
Now let ωL ¼ 1/Cihie which means fL ¼ 1/
essentially the response of input and output cou-
2πCihie. Then ωCihie > > 1 corresponds to
pling capacitors at low frequencies. We now con-
ω > > ωL or f > > fL. For ωCihie < < 1 which
sider these responses.
corresponds to ω < < ωL or f < < fL, then
6.1 BJT Low-Frequency Response 213
Example 6.2
Determine CE for the common emitter amplifier
in Fig. 6.12 in order to produce a lower cut-off
frequency of less than 50 Hz. Assume all cou-
pling capacitors are large.
Example 6.3
6.2 FET Low-Frequency Response
Determine the capacitors Ci, CE and Co in the
circuit of Fig. 6.13. To give fL ¼ 50 Hz. Assume
The determination of the coupling and bypass
hfe ¼ 100.
capacitors in a common source FET amplifier is
similar to the common emitter BJT amplifier.
Solution For this circuit, IC ¼ 1 mA and hence
Consider the common source JFET amplifier
re ¼ 25 Ω and hie ¼ 2.5 k. Using (6.21), Ci ¼ shown in Fig. 6.14. Similar to the BJT, the
1
2πZ i 50 where Zi ¼ R1//R2//hie ¼ 26 k//174 k// lower cur-off frequency set by Ci is given by
2.5 k ¼ 2.25 k. Hence C i ¼ 2π2:251
k50 ¼ 1
1:4 μF . Using (6.29), C o ¼ 2π ð9 kþ12 kÞ50 ¼
1 fL ¼ ð6:38Þ
2πRG C i
0:15 μF. Finally using (6.37), CE ¼ 2π2550
1
¼
where RG replaces hie and again Ci and RG
127 μF. Since CE is the largest capacitor, we use
together form a high-pass filter. Similarly, the
this to determine the lower cut-off frequency and
lower cut-off frequency created by the output
choose it to be CE ¼ 150 μF. We then select the
coupling capacitor Co is given by
other two capacitors to be about ten times the
calculated values so that the resulting cut-off 1
fL ¼ ð6:39Þ
frequencies are well below fL ¼ 50 Hz. Hence 2π ðRL þ RX ÞC o
Ci ¼ 15 μF and Co ¼ 2 μF.
218 6 Frequency Response of Transistor Amplifiers
Fig. 6.14 Common source JFET amplifier The most widely used model for transistor high-
frequency behaviour is the hybrid-pi equivalent
circuit shown in Fig. 6.17. This model was first
published by L.J. Giaceletto in the RCA Review
of 1954. The resistor r bb0 is the ohmic base resis-
tance, r cb0 is the collector-base resistor which is
quite large (several megohms) while capacitors
C b0 e and C b0 c are the emitter-base and collector-
base junction capacitors.
At low frequencies, since r bb0 is small and r cb0
is large, the hybrid-pi equivalent circuit becomes
Fig. 6.15 Equivalent circuit of common source JFET
amplifier that shown in Fig. 6.18. This is very similar to the
h-parameter low-frequency equivalent circuit of
Chap. 4 where
In order to determine the bypass capacitor CS,
hie ¼ r bb0 þ r b0 e ð6:45Þ
the equivalent circuit shown in Fig. 6.15 needs to
be analysed. Thus 1=hoe ¼ r ce ð6:46Þ
V i ¼ V GS þ gm V GS RS ==1=sC S and
RS
¼ V GS þ gm V GS ð6:40Þ gm vb0 e ¼ hfe ib ð6:47Þ
1 þ sC S RS
f
jAjdB ¼ 20 log k 20 log
fo |A|
¼ 20 log f þ 20 log k k
þ 20 log f o , f
>> f o ð6:51Þ
k/10
Using these equations, |A|dB is plotted against
logf. For frequencies f well below the pole fre-
quency such that f < < fo, the response
approaches the horizontal line |A|dB ¼ 20 log k. k/100
f
0.1f0 f0 10f0 100f0
For frequencies well above the pole frequency
such that f > > fo, the response approaches the Fig. 6.20 Frequency response plot of single stage
line whose slope is 20 dB/decade. These lines amplifier
are shown in Fig. 6.20. Note that at f ¼ fo, both
lines intersect since they both have the same
magnitude value |A|dB ¼ 20 log k. For the transfer
function, at f ¼ fo,
k k
jAj ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffi
1 þ ð f = f o Þ2 1þ1
pffiffiffi
¼ k= 2 ¼ 0:707 k ð6:52Þ
1 1
fβ ¼ ð6:60Þ re ¼ ð6:67Þ
2πr b0 e ðC b0 e þ Cb0 c Þ 40I c
But
hfe
r b0 e hie ¼ ¼ βo r e ð6:66Þ IC
40I c
Fig. 6.23 Variations of fT with collector current IC
where
Fig. 6.22 Frequency response plot of the magnitude of β Fig. 6.24 Amplifier with feedback admittance Yf
222 6 Frequency Response of Transistor Amplifiers
Y a ¼ I a =V i ð6:69Þ
Fig. 6.25 Equivalent amplifier circuit
while the output admittance is
ðV i V o ÞY f þ I b
Y b ¼ I b =V o ð6:70Þ Y out ¼
Vo
With Yf connected, the resulting input admit- 1
¼ Yb þ Y f 1
tance Yin becomes AV
and the resulting output admittance Yout is given Thus, the output admittance Yout of the ampli-
by fier with feedback Yf is the sum of the basic
admittance Yb and the admittance Yf of the feed-
Y out ¼ I o =V o ð6:72Þ back admittance providing the magnitude of the
gain of the amplifier is significantly greater than
From (6.71),
unity. Results (6.75) and (6.77) constitute
Ii I f þ Ia Miller’s theorem. Based on this theorem, the
Y in ¼ ¼ ð6:73Þ
Vi Vi amplifier circuit of Fig. 6.24 can be re-drawn in
the equivalent form of Fig. 6.25 in which the
But feedback admittance Yf is replaced by an admit-
I f ¼ ðV i V o ÞY f ð6:74Þ
Yf(1 AV) at the input and an admittance
tance
Y f 1 A1V at the output. This result enables a
Therefore,
more simplified analysis of the original amplifier.
ðV i V o ÞY f þ I a
Y in ¼
Vi
¼ Y a þ ð1 AV ÞY f ð6:75Þ 6.5 Common Emitter Amplifier
Based on (6.75), the input admittance Yin seen The results that have been derived will now be
at the amplifier input is the input admittance Ya of used to analyse the circuit of the common emitter
the basic amplifier in parallel with an admittance amplifier shown in Fig. 6.26.
whose value is that of the feedback admittance The equivalent circuit is shown in Fig. 6.27 in
multiplied by the factor (1 AV). If the amplifier which the collector-emitter resistor rce ¼ 1/hoe is
is inverting, then AV ! AV and the multiplying assumed to be very large compared with RL and
factor (1 AV) becomes (1 + AV). therefore has been omitted. Since Cb0 e is
At the output of the amplifier, we have connected between b and c in order to apply
Miller’s Theorem, we determine the voltage
Io I f þ Ib
Y out ¼ ¼ ð6:76Þ gain AV between b and c given by
Vo Vo
Vo
Substituting for If using (6.74) gives AV ¼ ð6:78Þ
V b0 e
6.5 Common Emitter Amplifier 223
Now Y f ð1 1=AV Þ ¼ jωC cb0 1 þ 1=gm R0L
jωC cb0 ,gm R0L 1 ð6:84Þ
V o ¼ gm V b0 e R0L ð6:79Þ
which is equivalent to a capacitance Ccb0 at the
where RL0 is RL including the effect of Ccb0 .
output. The capacitance C M 0 at the input is called
Therefore
the Miller capacitance and the resulting equiva-
gm V b0 e R0L lent circuit is shown in Fig. 6.28.
AV ¼ ¼ gm R0L ð6:80Þ
V b0 e Because of the multiplying effect of the
(1 + gmR0L) factor, its value is comparable to C b0 e
Since the feedback admittance Yf is between the base emitter junction.
Y f ¼ jωC cb ð6:81Þ Evaluating the transfer function Vo/Vb for the
amplifier,
using Miller’s Theorem, Yf across the collector-
V o ¼ gm V b0 e R0L ð6:85Þ
base junction can be replaced by Yf(1 AV) at the
input and Yf(1 1/AV) at the output. At the input, where
Y f ð1 AV Þ ¼ jωC cb0 1 þ gm R0L ð6:82Þ R0L ¼ RL ==1=sC b0 e ð6:86Þ
which is equivalent to a capacitance C M 0 given by r b0 e ==1=sC M
V b0 e ¼ V ð6:87Þ
r b0 b þ r b0 e ==1=sC M b
CM 0 ¼ Ccb0 1 þ gm R0L ð6:83Þ
where
At the output,
224 6 Frequency Response of Transistor Amplifiers
C M ¼ CM 0 þ C b0 e ð6:88Þ Vo Vo VG
¼
VS VG VS
Hence r b0 e
rbb0 þRE þrb0 e
r b0 e ==1=sC M
¼ gm R0L ρ
Vo V V 0
¼ o b e ¼ gm R0L 1 þ sC M ðr bb0 þ RG Þ==r b0 e
V b V b0 e V b r bb0 þ r b0 e ==1=sC M
r b0 e 1=sC M
ð6:93Þ
0 r b0 e þ 1=sC M gm R0L r b0 e
¼ gm RL ¼ where
r b0 e 1=sC M r bb0 þ r b0 e þ r bb0 r b0 e sC M
r bb0 þ
r b0 e þ 1=sC M
RA ==RB
gm R0L r
r bb0 þ r b0 e
ρ¼ ð6:94Þ
¼ RS þ RA ==RB
1 þ sC M r bb0 ==r b0 e
ð6:89Þ From (6.42),
RL
The circuit to the left of b involving the source R0L ¼ RL ==1=sC cb0 ¼ ð6:95Þ
voltage VS as well as RS, RA and RB can be 1 þ sRL C cb0
replaced by the Thevenin equivalent VG in series If Ccb0 is small, then the pole created by RLCcb0
with RG as shown in Fig. 6.29 where is at a high frequency and can be ignored, in
RG ¼ R1 ==R2 ==RS ð6:90Þ which case R0L ! RL in (6.95).
Thus from (6.93), transfer function Vo/VS for
and the common emitter amplifier can be written as
R1 ==R2 Vo k
VG ¼ V ð6:91Þ ¼
RS þ R1 ==R2 S V S 1 þ jf = f o
Vo /VG
If the gain is reduced by reducing RL, the cut-off
(log scale) frequency is increased. A large transition fre-
k quency fT reduces Cbe in (6.68) and therefore
also improves the frequency response of the
amplifier. Finally, a small Ccb also improves the
f (log scale) frequency response by decreasing the Miller
fo capacitance CM.
Thus, in order to realize a high cut-off fre-
quency fo in the amplifier design, the following
Fig. 6.30 Frequency response plot of the common emit-
steps can be taken:
ter amplifier
+20volts
173k 9k
Co
Vo
600Ω Ci
VS 27k 2k
CE Fig. 6.33 Equivalent circuit of amplifier in Fig. 6.32
Now
Fig. 6.32 Common emitter amplifier with partially I b0 ¼ V b0 e =Z b0 e ð6:102Þ
bypassed emitter resistor
where
Z b0 e ¼ r b0 e ==1=sC b0 e ð6:103Þ
6.6 Common Emitter Amplifier
with Local Series Feedback Therefore
V b0
¼ f1 þ Re ðgm þ yb0 e ÞgZ b0 e
I b0
1 þ R e ð gm þ y b 0 e Þ
¼ ð6:106Þ
yb 0 e
across b0 and ground where Fig. 6.36 Effect of local series feedback on bandwidth
gm RL
CM ¼ C0 ð6:109Þ
1 þ Re ðgm þ yb0 e Þ b c gm RL r b0 e
k¼ ð6:113Þ
RS þ r bb0 þ r b0 e þ Re ð1 þ gm r b0 e Þ
and a capacitor Cb0 c across c and ground. Also,
from (6.104) and
V b0 1
V b0 e ¼ ð6:110Þ fo ¼
1 þ Re ðgm þ yb0 e Þ 2πC b0 e rb0 e ðRe þ αÞ=½RS þ r b0 b þ rb0 e ð1 þ gm Re Þ
ð6:114Þ
The resulting equivalent circuit is shown in
Fig. 6.35. with
The transfer function VO/VS is given by
α ¼ ðRS þ r bb0 Þð1 þ gm RL C b0 c =C b0 e Þ ð6:115Þ
Vo gm RL
¼
V S 1 þ Re ðgm þ yb0 e Þ þ ðRS þ r bb0 Þðyb0 e þ sC b0 e gm RL Þ From (6.114), for Re < < α, as Re increases, fo
ð6:111Þ increases as shown in Fig. 6.36. Also, the gain-
bandwidth product kfo is
Substituting for yb0 e using (6.105), (6.111)
gm RL r b0 e
becomes kf o ¼
2πCb0 e r b0 e ðRe þ αÞ
Vo k gm RL
¼ ð6:112Þ ¼
V S 1 þ jf = f o 2π ½Re Cb0 e þ ðRS þ r bb0 ÞðC b0 e þ C b0 c gm RL Þ
where ð6:116Þ
228 6 Frequency Response of Transistor Amplifiers
where
gm r b0 e RS RF RL
Rm ¼
ðRS þ r bb0 þ r b0 e ÞðRF þ RL Þ þ gm r b0 e RS RL
ð6:118Þ
Gain
RL Þ þ C b0 c gm RL RF ð6:119Þ (dB)
Decreasing RF
This reduces to
1
fo ¼ ðr bb0 þRS Þ
2πr b0 e ðRS þr ½C b0 e þ
bb0 þr b0 e Þþgm ðRL ==RF Þr b0 e RS =RF
Reducing RF corresponds to increasing feed- Fig. 6.39 Effect of local shunt feedback on bandwidth
back. Eventually RF < < RL and fo becomes
1 From (6.121), fo increases as RF is reduced,
fo ¼ ðrbb0 þRS Þ i.e. the bandwidth increases with increased feed-
2πr b0 e ðRS þr ½Cb0 e þ C b0 c gm RF
bb0 þr b0 e Þþgm r b0 e RS back as shown in Fig. 6.39. It is convenient to
ð6:121Þ identify the current gain AI ¼ IL/IS for the circuit
which is easily obtained by dividing Vo/IS by the
6.8 High-Frequency Response of the Cascode Amplifier 229
load resistor RL. Thus, the current gain-bandwidth frequency response of the common-base amplifier
product AIfo for this circuit is given by compared with the common emitter amplifier, the
frequency effect associated with re2 can be
gm RS RF
AI f o ¼ ignored. The DC collector current of Tr1 is
ðr bb0 þ RS Þ½Cb0 e ðRF þ RL Þ þ C b0 e gm RF RL
approximately that of Tr2. Therefore
ð6:122Þ
gm1 1=r e2 ð6:124Þ
which is a maximum when RS ! 1. That is, the
common emitter amplifier with local shunt feed- and hence the gain of the common emitter stage is
back is best driven from a current source in order unity. Hence
to improve the frequency response characteristics.
C0M ¼ ð1 þ 1ÞCcb1 ¼ 2Ccb0 1 ð6:125Þ
Therefore
Vo V V c1e1
¼ o
V S V e2b2 V S
rb0 e
rbb0 þRS þrb0 e
¼
1 þ sC M ðr bb0 þ RS Þ==r b0 e
gm R L
ð6:129Þ
ð1 þ sC cb2 RL Þ
ZL
Vo re þRS
¼
V S 1 þ sC 0 RS re
b e RS þre
RL
r þR
¼h
e iS
1 þ sC b0 e RRSSþr
re
e
ð1 þ sC b0 c RL Þ
ð6:136Þ
Therefore ð6:143Þ
6.11 High-Frequency Response of a Common Source FET Amplifier 233
This reduces to
Vo RE 1 þ gm r b0 e
¼
V b c RE þ r b e 1 þ gm r b0 e ==RE
0 0
1 þ sC b0 e 1þgr r 0
m be
C b0 e rb0 e ==RE
ð6:144Þ
1 þ s 1þg r 0 ==RE
m be
But
RE 1 þ gm r b 0 e
RE þ r b0 e 1 þ gm r b0 e ==RE
RE ð1 þ gm r b0 e Þ
¼
RE þ r b0 e þ gm r b0 e RE
R ð1 þ gm r b0 e Þ
¼
E 1, RE > Fig. 6.49 Common collector amplifier
RE 1 þ RrE þ gm r b0 e
1
>
1
ð6:145Þ fU ¼ ð6:148Þ
gm 2π ðRS þ r bb0 ÞC cb0
Hence
Example 6.7
Vo 1 þ sC b0 e 1þgr r 0 Determine the upper cut-off frequency for the
m be
Finally, since the input impedance seen at the the upper cut-off frequency fU is given by
transistor base is high, the ratio V b0 c =V i 1
fU ¼
1=sC cb0
¼ 1þsC 1
. Then the voltage 2π ðRS þ r bb0 ÞCcb0
RS þrbb0 þ1=sC cb0 cb0 ðRS þr bb0 Þ
gain is given by 1
¼ ¼ 568 MHz
2π ð50 þ 20Þ 4 1012
Vo 1 1 þ sC b0 e 1þg r 0 r
¼ : m be
Fig. 6.54 Common gate gm Ri
FET amplifier V gs 1 þ
1 þ sRi C gs
1 þ gm Ri þ sRi C gs
¼ V gs
1 þ sRi C gs
1
¼ V i ð6:153Þ
1 þ sRi C gs
Hence
Vo V V gs
¼ o:
V i V gs V i
gm R L 1
¼ :
1 þ sC gd RL 1 þ gm Ri þ sRi C gs
ð6:154Þ
This reduces to
Vo g R 1
¼ m L :
Vi 1 þ gm Ri 1 þ sC gd RL 1 þ sC gs Ri ==1=gm
ð6:155Þ
gm RS
If gmRS > > 1, then 1þg RS 1 and in (6.160)
m
Vo/Vgd ! 1. This suggests a very high-frequency
response for the common drain amplifier when
driven by a low source impedance Ri (in which
case Vgd Vi) and has a high value of FET source
resistor RS. Note however that a non-zero Ri in
conjunction with Cgs and the small Miller capaci-
tance resulting from Cgd form a low-pass filter
which limits the eventual frequency response of
this circuit.
Fig. 6.56 Common drain FET amplifier
|A| |A|
(dB) (dB)
k k
f f
f01 10f01 102f01 103f01 104f01 f01 10f01 102f01 103f01 104f01
(f 02) (f 02) (f 03)
(a) (b)
Fig. 6.58 Frequency plots of two (2)- and three (3)-pole system
Example 6.10
The transfer function of a three-pole amplifier
|A|
system is given in eq. (6.162).
104
104
AV ðjωÞ ¼ 103
ð1 þ jf =10Þ 1 þ jf =103 1 þ jf =104
102
ð6:162Þ
101
Plot the asymptotic curve for this system,
indicating the low-frequency gain and the posi- 100
f
101 102 103 104 105
tion of each pole.
10-1
Table 6.1 Results of frequency response tests of CE amplifier with local series feedback
Resistor Re 100 Ω 200 Ω 300 Ω 400 Ω 500 Ω
Bandwidth 1.9 MHz 2.9 MHz 3.7 MHz 4.4 MHz 4.9 MHz
Voltage gain 36.2 dB 31.4 dB 28.3 dB 26 dB 24.3 dB
Table 6.2 Results of frequency response tests of CE amplifier with local shunt feedback
Resistor Rf 100 k 50 k 20 k 10 k 1k
Bandwidth 1 MHz 2 MHz 5 MHz 9.6 MHz 87.2 MHz
Voltage gain 18.6 dB 13.2 dB 5.6 dB 0 dB 20 dB
+20volts
+30volts
RB 10k Co
5k Vo
2M Vo
Ci
Ci Vi
Vi
50k
+VCC=20
RL=9k
R1=174k
Co
Vo
Ci
Vi
R2=27k
CE
RE=2k
Fig. 6.66 Circuit for Question 2
swing, determine a suitable value of coupling Fig. 6.68 Circuit for Question 4
capacitor to realize a maximum lower cut-off
frequency of 70 Hz.
3. Determine the output coupling capacitor Co +VCC=20
for the circuit shown in Fig. 6.67 in order that
the amplifier has a low-frequency cut-off fre-
RL=9k
quency of 50 Hz. Assume transistor β ¼ 100. R1=174k
Co
4. Determine CE for the common emitter ampli- Vo
fier in Fig. 6.68 in order to produce a lower Ci
Vi
cut-off frequency of less than 25 Hz. Assume
all coupling capacitors are large.
5. Determine the capacitors Ci, CS and Co in the Rx=12k
R2=27k
circuit of Fig. 6.69 to give fL ¼ 80 Hz. CE
RE=2k
Assume hfe ¼ 200.
6. Evaluate the lower cut-off frequency
resulting from each of the capacitors shown Fig. 6.69 Circuit for Question 5
in Fig. 6.70.
242 6 Frequency Response of Transistor Amplifiers
+30volts +20volts
17.3k 12k
12k 1μF
Co Vo
Ci Vo
100μF 80Ω
Vi
100μF
68k 2.7k
2M 3k 3k VS
CS
+20volts +30volts
173k 9k 14.3k
Co
Vo Vi
600Ω Ci 47μF 47μF
Vo
15.7k
12k
VS 27k 2k
CE
Fig. 6.71 Circuit for Question 9 Fig. 6.73 Circuit for Question 11
7. A silicon NPN transistor has fT ¼ 200 MHz. 12. Find the cut-off frequency of the common
Determine Cb0 e for a collector current of source amplifier shown in Fig. 6.74. The
2 mA. JFET has the following characteristics:
8. State and verify Miller’s theorem. Cgs ¼ 70 pF, Cgd ¼ 8 pF, rd ¼ 150 kΩ,
9. For the common emitter amplifier shown in gm ¼ 10 mA/V.
Fig. 6.71, determine the upper cut-off fre- 13. Draw the equivalent circuit of the circuit in
quency. The transistor used has fT ¼ 250 Fig. 6.74.
MHz, Ccb0 ¼ 7 pF and β ¼ 125. 14. For a common gate JFET amplifier with
10. Determine the upper cut-off frequency for the RS ¼ 5 k and RL ¼ 6 k, using a JFET having
common base amplifier shown in Fig. 6.72 Cgd ¼ 4 pF, find the upper cut-off frequency
where a 2 N3904 transistor is used. of the circuit.
11. Determine the upper cut-off frequency for the 15. Explain the excellent wideband
common collector amplifier shown in characteristics of the Cascode amplifier
Fig. 6.73 where a 2 N3904 transistor is used. 16. Sketch a common emitter amplifier in which
local series feedback is used to improve the
Bibliography 243
Bibliography
Fig. 6.74 Circuit for Question 12 S.S. Hakim, Feedback Circuit Analysis (Iliffe Books Ltd.,
London, 1966)
J. Millman, C.C. Halkias, Integrated Electronics: Analog
and Digital Circuits and Systems (McGraw Hill,
amplifier bandwidth. For the maximum band- New York, 1972)
A.S. Sedra, K.C. Smith, Microelectronic Circuits, 6th edn.
width, what should be the nature of the (Oxford University Press, Oxford, 2011)
source resistance?
Feedback Amplifiers
7
A feedback amplifier is one in which a portion of of the amplifier with feedback (closed-loop gain)
the output is fed back to the system input where it compared with the gain of the amplifier without
is combined with the input signal. The basic con- feedback (open-loop gain). Feedback can however
cept is shown in Fig. 7.1. At the amplifier input, result in amplifier instability, and therefore appro-
the feedback signal which may be a voltage or priate steps need to be taken to prevent this.
current is combined with the input signal which is In this chapter we introduce the concept of
also a voltage or current through a summing or feedback and apply it to amplifier systems. The
mixing network, and the resulting signal is passed resulting circuits display a range of improved
into the amplifier system. The summing network characteristics, which are discussed. At the end
is either a series circuit which mixes feedback of the chapter, the student will be able to:
voltage with source voltage or a shunt circuit
which mixes feedback current with source current • Explain the feedback concept
while the feedback network is most often a pas- • Discuss the different types of amplifiers and
sive network, usually resistive. feedback
The basic feedback principle in feedback • Design feedback amplifiers
systems is the sampling of a portion of the output
signal such that when mixed with the input
improves the overall characteristics of the ampli-
fier. When any increase in the output signal results 7.1 Classification of Amplifiers
in a feedback signal into the input in such a way as
to cause a decrease in the output signal, the ampli- Amplifier systems may be conveniently classified
fier is said to have negative feedback. Negative into four types: voltage amplifiers, current
feedback is generally useful because it can amplifiers, transconductance amplifiers and
improve the characteristics of the various amplifier trans-resistance amplifiers. This classification is
topologies presented earlier. For example, feed- based on the magnitudes of the input and output
back can produce a significant improvement in impedances of an amplifier relative to the source
the linearity and frequency response of the ampli- and load impedances, respectively.
fier and lower distortion and noise. Another advan-
tage of negative feedback is that it stabilizes the
gain of the amplifier against variations in then 7.1.1 Voltage Amplifier
device characteristics of the transistors or other
active devices used in the amplifier. The price The voltage amplifier accepts a voltage at its input
paid for these improvements is the lowered gain and delivers a voltage at its output. It is
# Springer Nature Switzerland AG 2021 245
S. J. G. Gift, B. Maundy, Electronic Circuit Design and Application,
https://doi.org/10.1007/978-3-030-46989-4_7
246 7 Feedback Amplifiers
SUMMING
JUNCTION
INPUT + OUTPUT
Σ Amplifier
_
Feedback
Network Fig. 7.3 Norton equivalent circuit of a current amplifier
source resistance Rs, i.e. Ri Rs. For these dropped across the output impedance. Then, the
conditions, the input voltage Vi ’ Vs. In order to input current Ii ’ Is and the output voltage
effectively deliver the output current to a load, the Vo ’ RMIi. As a result the voltage at the output
output resistance Ro of the system must be high so is proportional to the input current, and the factor
that most of the current flows into the load RL and of proportionality is independent of the
little flows into the shunt output Ro resulting in an magnitudes of the source and load resistances.
output current Io ’ GMVi. Under these conditions, Ideally, the input and output resistances of a
the output current is proportional to the input trans-resistance amplifier are both zero.
voltage, and the proportional factor is indepen- Table 7.1 summarizes the characteristics of the
dent of the magnitudes of the source and load four amplifier types.
resistances. Ideally, the input and output
resistances of a transconductance amplifier are
both infinite.
7.2 Feedback Amplifier Topologies
Vi ¼ Vs V f ð7:1Þ
Therefore, system voltage gain with feedback In the case of the trans-resistance amplifier of
A f ¼ VV os is given by Fig. 7.8, the output signal being sampled is a
voltage and again the sampling is accomplished
Av
Af ¼ ð7:4Þ by connecting across the output. However, the
1 þ βAv sampled voltage must first be converted to a cur-
The result (7.4) shows that voltage-series feed- rent since it must be subtracted from the input
back around a voltage amplifier results in the signal current. The subtraction is effected by
amplifier gain reduced by the factor (1 + βAv). connecting the feedback current in shunt with
An example of the application of voltage-series the input current. Note that the output voltage
feedback is shown in the simplified two-transistor must be inverted relative to the voltage associated
circuit of Fig. 7.7 consisting of a common emitter with the input current for current subtraction to
amplifier Tr1 driving another common emitter take place. The result is voltage-shunt feedback.
amplifier Tr2. A portion of the output voltage at From Fig. 7.8
the collector of Tr2 is returned through the feed- Ii ¼ Is I f ð7:5Þ
back network RFRE to the emitter of Tr1 where it
is subtracted from the input voltage. where If ¼ βVo. The output voltage is given by
7.2 Feedback Amplifier Topologies 249
I o ¼ Ai I i ¼ Ai I s I f ¼ Ai I s Ai βI o ð7:14Þ
This gives
I o ð1 þ βAi Þ ¼ Ai I s ð7:15Þ
Ai
Af ¼ ð7:16Þ
1 þ βAi
In the presence of a large amount of negative Thus, while the amplifier open-loop gain
feedback, the loop gain Aβ is much greater than changes from 1000 by 20%, the closed-loop
one and therefore (7.23) reduces to amplifier changes by only 0.2%. This represents
a 100-fold improvement. The factor D ¼ 1 + Aβ is
A 1 sometimes called the de-sensitivity.
Af ¼ ð7:24Þ
βA β
Equation (7.28) shows that the magnitude of the k 0 ¼ k=ð1 þ βkÞ ð7:33Þ
fractional change in gain with feedback |dAf/Af| is
and
reduced by the factor 1 + Aβ Aβ or the loop
gain, compared to the fractional change without f o 0 ¼ f o ð1 þ βk Þ ð7:34Þ
feedback |dA/A|.
7.6 Feedback and Harmonic Distortion 253
Is If Ri RmIi RL Vo
Vi
Rof R'of
Rif
Fig. 7.19 Thevenin Equivalent of Voltage-Series Feed-
back Amplifier Fig. 7.20 Voltage-Shunt Feedback Amplifier
7.7 Input Resistance 255
and Rm is the open-loop gain with no load. There- From (7.50) it can be seen that the input resistance
fore RM ! Rm as RL ! 1. Substituting (7.44) in of the amplifier is increased by the factor of the
(7.43) and rearranging, we get loop gain as a result of the negative feedback.
Vi Vi Ri
Rif ¼ ¼ ¼ ð7:46Þ
I s I i ð1 þ βRM Þ 1 þ βRM
7.7.4 Current-Shunt Feedback
From (7.46) it can be seen that the input resistance
of the amplifier is lowered by the factor of the The Norton equivalent of the current-shunt
loop gain as a result of the negative feedback. feedback amplifier is shown in Fig. 7.22. Since
the amplifier input resistance is being consi-
dered, the source resistance is once again set to
7.7.3 Current-Series Feedback infinity (removed). In order to determine the
amplifier input resistance Rif from Fig. 7.22, we
The Thevenin equivalent input-Norton equivalent have
output representation of the current-series feed- I s ¼ I f þ I i ¼ βI o þ I i ð7:51Þ
back amplifier is shown in Fig. 7.21. Since the
amplifier input resistance is being considered, the and
source resistance is again set to zero. In order to
Ai I i Ro
determine the amplifier input resistance Rif from Io ¼ ¼ AI I i ð7:52Þ
Fig. 7.21, we have Ro þ RL
V s ¼ V f þ V i ¼ βI o þ V i ð7:47Þ where
Io Ai Ro
and AI ¼ ¼ ð7:53Þ
I i Ro þ RL
Gm V i Ro
Io ¼ ¼ GM V i ð7:48Þ is the open-loop current gain of the amplifier,
Ro þ RL
taking the effect of the external load into account,
where and Ii is the open-loop gain with no load. There-
fore AI ! Ai as RL ! 0. Substituting (7.52) in
Io G R
GM ¼ ¼ m o ð7:49Þ (7.51) and rearranging, we get
V i Ro þ RL
Vi Vi Ri
is the open-loop transconductance of the ampli- Rif ¼ ¼ ¼ ð7:54Þ
I s I i ð1 þ βAI Þ 1 þ βAI
fier, taking the effect of the external load into
account, and Gm is the open-loop gain with no From (7.54) it can be seen that the input
load. Therefore GM ! Gm as RL ! 0. Substituting resistance of the amplifier is decreased by the
(7.48) in (7.47) and rearranging, we get factor of the loop gain as a result of the negative
feedback.
V s V i ð1 þ βGM Þ
Rif ¼ ¼
Ii Ii
¼ Ri ð1 þ βGM Þ ð7:50Þ
Fig. 7.21 Current-Series Feedback Amplifier Fig. 7.22 Current-shunt feedback amplifier
256 7 Feedback Amplifiers
7.8 Output Resistance This result confirms that the output resistance Rof
of the amplifier with feedback is reduced. It is
Negative feedback also directly influences the reduced by the factor 1 + βAv. Rof is the output
output resistance of an amplifier. The nature of resistance seen by the load RL. If RL is treated as
that influence depends on the method of deriving part of the amplifier, then the output resistance
the feedback. In particular, if the feedback is with feedback becomes Rof in parallel with RL,
acquired by sampling an output voltage, then that is
regardless of the method of application of the Rof RL
feedback, the output resistance is reduced in an R0of ¼ ð7:57Þ
Rof þ RL
attempt to stabilize that output voltage. If on the
other hand the feedback is acquired by sampling Substituting for Rof from (7.56) yields
an output current, then the effect is an increase in
1þβAv RL
R
output resistance in an attempt to maintain the Ro RL
R0of ¼ ¼ ð7:58Þ
Ro þ RL þ βAv RL
1þβAv þ RL
Ro
output current. Again, the effect is independent
of the method of application of the feedback
signal. In the following sections, the effect of This reduces to
negative feedback on amplifier output resistance
Ro RL =ðRo þ RL Þ
is examined and quantified. R0of ¼
1 þ βAv RL =ðRo þ RL Þ
R0o
¼ ð7:59Þ
7.8.1 Voltage-Series Feedback 1 þ βAV
2. For shunt mixing, set the input voltage to zero. feedback analysis to hold is that there must be
This corresponds to short-circuiting the input no signal feed-forward through the feedback net-
terminal. work. The emitter current of Tr1 is approximately
equal to the input current and flows forward
These steps ensure that the negative feedback through the feedback network RE//RF. However,
is reduced to zero while still taking into account this current is generally small as compared with
the loading of the feedback circuitry on the basic the current IF from the output. In order to deter-
amplifier. mine the basic amplifier without feedback, we
first determine the input circuit of the amplifier
by setting the output voltage to zero. The effect of
7.9.1 Voltage-Series Feedback this is that RF appears in parallel with RE. Follow-
ing this, the output circuit is determined by setting
Consider the voltage amplifier comprising two the input current to zero. This corresponds to
common emitter amplifiers that are cascaded to opening the connection of the junction of RE
produce a higher overall voltage gain as shown in and RF to the emitter of Tr1 with the result that
Fig. 7.24. The biasing circuitry is omitted for RF appears in series with RE. The resulting open-
simplicity. We first note that because of the signal loop voltage amplifier is shown in Fig. 7.25. The
inversion produced by each stage, the output feedback factor β is given by
voltage at the collector of the second stage is in Vf RE
phase with the input voltage at the base of the first β¼ ¼ ð7:67Þ
V o RF þ RE
stage. We note also that a signal injected at the
emitter of the first stage will produce an inverted
output signal and therefore the feedback signal The amplifier open-loop gain can now be
must be applied at the emitter of Tr1. Voltage found in the usual way following which feedback
sampling is thus effected by sampling the output analysis can proceed. It should be noted that
voltage at the collector of the second stage and there is (local) current-series feedback in the first
applying series feedback mixing with the input stage of the amplifier because of the presence of
voltage at the emitter of the first stage. The β the un-bypassed emitter resistance. This kind of
network therefore comprises resistors RF and RE. feedback is analysed shortly. An example
One of the assumptions necessary for the follows.
Fig. 7.24 Voltage-series feedback amplifier Fig. 7.25 Open-loop amplifier circuit with feedback net-
work loading
7.9 Analysis of Feedback Amplifiers 259
The effective load RL2 of Tr2 is RL2 ¼ 5.6 k// Another voltage amplifier in which voltage-
(5.6 + 0.47) k ¼ 2.91 k. Hence the voltage series feedback is employed is the emitter fol-
gain AV2 of the second stage is AV2 ¼ lower. In this circuit the full output voltage
h R
fehieL2 ¼ 1002:91 k
¼ 116:4 . Therefore, constitutes the feedback voltage and is subtracted
2:5 k
from the input voltage. The error voltage is that
the overall open-loop voltage gain AV is
dropped across the base emitter junction of the
AV ¼ AV1 AV2 ¼ 4.27 116.4 ¼ 497.
transistor. It is possible to apply the feedback
Now β ¼ RFRþR E
¼ 5600þ470
470
¼ 0:077 . Also
E analysis to this circuit to arrive at the gain expres-
1 + βAV ¼ 1 + 0.077 497 ¼ 39.3. There- sion and other characteristics. This analysis
fore the closed-loop voltage gain AVf is shows that the emitter follower can be viewed as
found to be AVf ¼ 1þβA AV
V
¼ 39:3
497
¼ 12:6 . a common emitter amplifier with collector resistor
h R
This is quite close to the value for very RE and open-loop gain AV ¼ fehie E ¼ gm RE around
large open-loop gain given by which 100% voltage-series feedback
AVf(AV ! 1) ¼ 1/β ¼ 12.9. corresponding to β ¼ 1 is applied. The resulting
260 7 Feedback Amplifiers
open-loop current gain can now be found. Noting RF ¼ 2.5 + 101 0.1//1.2 ¼ 11.8 k. Hence
I c1 ¼ 17:3þ11:8 ¼ 0:59 ; I b1 ¼ hfe ¼ 100
that the collector current is much larger than the I b2 17:3 I c1
If ¼
RE
I ð7:69Þ result is AI ¼ II oS ¼ 100 0:59 100
RF þ RE o 0:34 ¼ 2006 . The feedback factor is β ¼
From this the feedback factor beta is given by RE
RF þRE ¼ 1:2þ0:2
0:1
¼ 0:08 and 1 + βAI ¼ 1 +
If 0.08 2006 ¼ 161.5. This gives the closed-
RE
β¼ ¼ ð7:70Þ loop current gain AIf ¼ 1þβA
AI
¼ 161:5
2006
¼ 12:4.
I o RF þ RE I
Example 7.4
Ri
1þβAI ¼ 161:5
855
¼ 5:3 Ω. As a current amplifier
For the current amplifier shown in Fig. 7.29, with feedback, the input resistance is small
determine the closed-loop current gain, input as expected. To enable the circuit to accept a
resistance and output resistance. For the voltage signal input, a resistor RS Rif can
transistors hfe ¼ 100, hie ¼ 2.5k, hre ¼ hoe ¼ 0. be added in series with the input. For a
Determine RS to give a voltage gain of 15. voltage gain of 15, V o =V s ¼ I o R2 =I S RS ¼
AIF RR2S ¼ 12:4 9RkS ¼ 15: This gives
Solution RS ¼ 7.4 k.
(iii) The open-loop output resistance Ro looking
(i) The open-loop current gain must first be in at the collector of the second transistor is
determined. This is given by AI ¼ II oS ¼ Ro ¼ 1 since hoe ¼ 0. Hence the closed-
I o I b2 I c1 I b1
. From the circuit Io
¼ hfe ¼ loop output resistance Rof looking at the
I b2 I c1 I b1 I S I b2
collector is Rof ¼ Ro(1 + βAI) ¼ 1. It
¼
100 ; II b2c1 R1RþR
where (noting that
1
i2 follows therefore that the output resistance
RF RB) we have Ri2 ¼ hie + (1 + hfe)RE// of the amplifier is Rof//R2 ¼ R2 ¼ 9 k.
h
AVf ¼
V o I o RL
¼ ¼ GMf RL hie þR
fe
E
¼ 2:5þ2
100
¼ 22:22 mA=V. β ¼ RE ¼
Vs Vs
2 k. And hence 1 + βGM ¼ 1 2
h R RL (22.22) ¼ 45.44. Therefore GMf ¼ 1þβG GM
¼
¼ fe L ¼ ð7:75Þ
hie þ 1 þ hfe RE r e þ RE 22:22
M
Example 7.5
For the transconductance amplifier of Fig. 7.32
7.9.4 Voltage-Shunt Feedback
determine the closed-loop transconductance GMf,
the closed-loop input impedance and the closed-
The single transistor amplifier circuit of Fig. 7.33 is
loop output impedance.
an example of voltage-shunt feedback. The sam-
pled output voltage at the collector of the transistor
Solution We proceed in the manner developed
develops a feedback current in the feedback resistor
in the case of the other two amplifiers. Thus the
RF. Because of the inverting action of the configu-
open-loop transconductance is GM ¼ VI os ¼ ration, this current is subtracted from the input
signal at the base of the transistor. The rules of
feedback analysis give the input circuit of the
open-loop amplifier by shorting the output node
Fig. 7.32 Transconductance amplifier example Fig. 7.33 Voltage-shunt feedback amplifier
264 7 Feedback Amplifiers
Ao β
1 þ RR1 : 1þA oβ
Af ¼ ð7:90Þ
1 þ jf = f o ð1 þ Ao βÞ
where
β ¼ R1 =ðR1 þ R2 Þ ð7:91Þ
Fig. 7.38 Frequency response plot showing loop gain Fig. 7.39 Diagram for Example 7.9
268 7 Feedback Amplifiers
closed ‐ loop gain (dB) ¼ 80 dB 20 dB ¼ input stage made up of the differential amplifier
60 dB. Tr1 and Tr2, the intermediate common emitter
The open-loop characteristic shown in stage Tr3 and the emitter follower output stage
Fig. 7.39 corresponding to a voltage amplifier Tr4. Signals applied at the input of the differential
with a single pole characteristic will have a maxi- amplifier Tr1Tr2 are amplified and passed via R3
mum phase shift between output and input of to the second stage Tr3. This common emitter
90 degrees. It can be shown that such a system amplifier provides the bulk of the voltage gain.
will not become unstable with feedback and can The output stage Tr4 is an emitter follower that
accommodate 100% feedback. As a result, it is provides a low-impedance output drive. In this
sometimes referred to as being unity gain stable. configuration, the loading effects of the feedback
However, steps have to be taken with (multistage) network are negligible.
amplifiers in order to realize conditions that
ensure system stability when feedback is applied. Differential Amplifier
These methods are discussed in Sect. 7.12. Also, In the differential amplifier stage, two transistors
the analysis developed is directly applicable to Tr1 and Tr2 are connected at their emitters where
practical operational amplifiers since the assumed an approximately constant current I through resis-
high input impedance and low output impedance tor R2 is connected to supply bias current to each.
that characterize a good voltage amplifier are Each transistor is basically in a common emitter
easily met by modern operational amplifiers. mode with an input supplied to its base and an
These will be more extensively discussed in the output taken from its collector. At the two base
next chapter. terminals of the differential amplifier are the input
signal Vi = vi1 at the base of Tr1 and the feedback
signal Vf = vi2 at the base of Tr2. The output vo1 is
7.11 Transistor Feedback Amplifier taken at the collector of Tr1 where resistor R3 is
connected. It is possible to obtain an output vo2 at
The feedback concept can be applied in the the collector of Tr2 by the inclusion of another
design of a voltage amplifier system using dis- resistor R3. Thus vi1 vi2 is the differential input
crete devices. One basic topology is shown in Fig. voltage and vo1 is the single-ended output voltage.
7.40. It comprises three amplifying stages: the Since there are two inputs and one output, the
amplifier is said to have a double-ended input
and a single-ended output.
In examining the small-signal behaviour of
this circuit, the output voltage vo1 at the collector
of Tr1 due to each input acting alone, that is, with
the opposite input grounded, will be determined
and then the superposition principle will be
applied to determine the output due to both acting
simultaneously. Consider the case where input
2 is grounded (vi2 = 0) and a small signal applied
to input 1. We assume perfectly matched
transistors Tr1 and Tr2 and therefore that IC1 = IC2.
Since Tr1 is essentially a common emitter ampli-
fier, vi1 is amplified and inverted. For amplifica-
tion of vi1, Tr2 functions as a common base
amplifier. Hence the input impedance re of Tr2
is the emitter resistance of Tr1. Therefore, the
voltage gain of Tr1 is
Fig. 7.40 Transistor feedback amplifier
7.11 Transistor Feedback Amplifier 269
Thus for the output vo1 of Tr1, from (7.103) the Av2 ¼ 40I C3 RL ð7:112Þ
component due to vi1 is 2r RL
vi1 and from (7.104)
RL
e
where IC3 is the current in transistor Tr3 and RL is
that due to vi2 is 2re vi2 . Using the superposition
the load made up of R4 in parallel with the input
principle, impedance of Tr4. Since Tr4 is in the emitter
RL follower configuration, its input impedance is
vo1 ¼ ðv vi2 Þ ð7:105Þ high and therefore does not significantly affect
2r e i1
R4. Hence RL R4.
Hence the gain [single-ended out/differential in]
is given by Emitter Follower
vo1 R The output stage of the amplifier consists of an
¼ L ð7:106Þ emitter follower whose gain is approximately
vi1 vi2 2r e
one. It presents low output impedance and
If a resistor R3 is included in the collector of Tr2, provides the amplifier with the ability to deliver
then by similar arguments the output vo2 from the
270 7 Feedback Amplifiers
current without loading down the amplifier and given by AVf ¼ 10 ¼ 1 + R7/R6. For R6 ¼ 1 k then
reducing the open-loop gain. R7 ¼ 9 k. In this design, a compensation capacitor
The overall gain of the open-loop amplifier is between the collector and base of Tr3 is generally
given by the product of the open loop gain of the needed for stability. Its value, which is likely to be
three stages. The closed-loop amplifier is shown in the range 1 pF 100 pF, is adjusted for a flat
in Fig. 7.40. The negative feedback being applied frequency response (see Example 7.12).
is voltage-series feedback as in the non-inverting The open-loop voltage gain is determined as
operational amplifier. For a sufficiently large follows: The voltage gain AV1 of the differential
open-loop gain, the closed-loop gain is given by stage is AV1 ¼ R3 ==h
2re
ie3
where re is associated with
R6 Tr1 and Tr2 and hie3 is associated with Tr3. re ¼ 1/
AV ¼ 1 þ ð7:113Þ 40IC ¼ 1/40 0.5 ¼ 50 Ω and hie3 ¼ 100/40 2
R7
mA ¼ 1.25 k. Therefore AV1 ¼ 1:4 k==1:25
250
k
¼ 10.
In order that the circuit be stable, a small
The voltage gain AV2 of the second and main
capacitor must be placed from the collector of
voltage gain stage is given by AV2 ¼ 40ICRL
transistor Tr3 to its base. This is called compensa-
where IC ¼ 2 mA and RL is the load resistance
tion and is discussed in Sect. 7.12.
of Tr3. This load is R4 ¼ 12 k in parallel with the
input impedance of the emitter follower Tr4. The
Example 7.10
input impedance of Tr4 is higher than
Design a feedback amplifier with a closed-loop
hfeR5 ¼ 100 5 k ¼ 500 k. Therefore RL ¼ 12
gain of 10 using the configuration shown in
k//500 k 12 k and AV2 ¼ 40 2 12 ¼ 960.
Fig. 7.40. Use small signal transistors having
Therefore, the open-loop gain of the amplifier is
hfe ¼ 100 and a 25-volt supply. Justify all
AV ¼ AV1AV2 ¼ 10 960 ¼ 9600. Using the
steps in your design. Calculate the open-loop
voltage gain of your amplifier. complete formula, AVf ¼ 1þβA AV
V
¼ 1þ0:19600
9600
¼
9:99 . This is very close to 10 and therefore
Solution Let the current in Tr1 and Tr2 be justifies our use of AVf ¼ 1 + R7/R6.
0.5 mA. This is a reasonable current that results The design of this amplifier is not complete.
in good transistor frequency response and current The circuit has at least three poles, each of which
gain. This results in 1 mA flowing through resistor can introduce significant phase shift such that
R2. The voltage at the base of Tr1 is approximately instability can occur upon application of feed-
0 volts and hence R2 ¼ 250:7 back. Steps must therefore be taken to ensure
1 mA ¼ 24:3 k. The base
current of Tr1 flowing through R1 is 0.5 mA/ that the system with feedback is stable. This is
100 ¼ 5 μA. Allowing a maximum 50 mV drop the subject of the next section.
across R1 gives R1 ¼ 50 103/5 106 ¼ 10 k.
Since the base-emitter voltage of Tr3 is across R3,
R3 ¼ 0.7/0.5 mA ¼ 1.4 k. The voltage at the 7.12 Stability and Compensation
output of the amplifier must be zero under quies-
cent conditions. The voltage at the base of Tr4 is In a negative feedback amplifier with open-loop
therefore 0.7 volts. This gives the voltage drop gain A and feedback factor β, the closed-loop gain
across R4 as 25 0.7 ¼ 24.3 volts. We choose a is given by A f ¼ 1þβA
A
. If the magnitude of the
current 2 mAin Tr3 which is somewhat higher loop gain Aβ is very large, i.e. |Aβ| 1, then
than the currents in the differential stage. Then Af ! 1/β which is the closed-loop low-frequency
R4 ¼ 24.3/2 mA ¼ 12 k. In order to drive external gain of an amplifier. This analysis is however
loads, we choose an even larger current of 5 mA in generally inadequate since A and sometimes β
Tr4. Then R5 ¼ 25/5 mA ¼ 5 k. For a closed-loop are frequency dependent. This means that at cer-
gain of 10, we assume that the open-loop gain is tain frequencies |Aβ| may not be very much
sufficiently large such that the closed-loop gain is greater than unity. Under such circumstances,
7.12 Stability and Compensation 271
the situation needs to be further examined. Thus, A simple approach to investigating stability
the closed-loop gain becomes involves the separate construction of Bode plots
for the amplifier transfer function A( jω) and the
AðsÞ
A f ðsÞ ¼ ð7:114Þ factor 1/β( jω) which represents the closed-loop
1 þ βðsÞAðsÞ gain for high-gain systems. Noting that
For physical frequencies s ¼ jω and the closed- 20 log jAβj ¼ 20 log jAðjωÞj
loop gain takes the form
1
20 log ð7:118Þ
A f ðjωÞ ¼
AðjωÞ
ð7:115Þ βðjωÞ
1 þ βðjωÞAðjωÞ
we can see that the difference between the two
The loop gain now becomes A( jω)β( jω) which is graphs is actually the loop gain expressed in
a complex function that can be represented by its dB. Therefore, the stability of the feedback sys-
magnitude and phase tem can be assessed by determining the difference
between the |A| plot and the |1/β| plot. For |
AðjωÞβðjωÞ ¼ jAðjωÞβðjωÞjejφðωÞ ð7:116Þ
Aβ| ¼ 1, then 20 log |Aβ| ¼ 0 and hence the
where φ(ω) is the phase shift of a signal exiting point of intersection of the two curves
the feedback network relative to a signal entering corresponds to the point at which |Aβ| ¼ 1. An
the amplifier. The stability of the closed-loop example will illustrate the method:
system depends on the loop-gain. If the magni- Consider a three-pole amplifier whose open-
tude of the loop-gain is unity and the phase shift is loop transfer function is given by
180 , then
105
AðjωÞ ¼
jAðjωÞβðjωÞjejφðωÞ ¼ 1:ej180 ¼ 1 ð7:117Þ 1 þ j 10f 3 1 þ j 10f 4 1 þ j 10f 5
ðjωÞ ð7:119Þ
Hence the closed-loop gain becomes A11 ! 1.
This suggests that the system will produce an The magnitude |A( jω)| at a frequency f(Hz) is
output without an input and is therefore unstable. found from
If the magnitude of the loop gain is less than one
when φ ¼ 180 , the system will not sustain the
output without an input and is stable. The differ-
ence in dB between unity loop gain |Aβ|
corresponding to 0 dB and the loop gain value dB
at ϕ ¼ 180 is the Gain Margin which is the
amount by which the gain can be increased before
instability ensues. The difference between the
phase at the frequency at which the loop gain
magnitude is unity and ϕ ¼ 180 is the 0 f (180°) Gain
Phase Margin. It is the amount by which the Margin f (log scale)
loop gain is reduced. In order to ensure stability of open-loop gain response intersects the closed-
the closed-loop system with acceptable levels of loop plot at a relative slope of 20 dB/dec. In
frequency response peaking, the magnitude and order to illustrate the method, consider the open-
phase responses of the open-loop amplifier and loop 20 log |A| response of Fig. 7.44 having three
the feedback network can be adjusted such that poles at fp1, fp2 and fp3 and a closed 20 log |1/β|
ϕ < 180 when |Aβ| ¼ 1 or alternatively that | plot as shown. The |1/β| plot intersects the |A| plot
Aβ| < 1 whenϕ ¼ 180 . This process is referred at a slope greater than20 dB/dec, and our rule-
to as Compensation. The three methods to be of-thumb for a stable feedback amplifier is not
discussed are (i) dominant pole compensation, satisfied. In order to correct this, a new pole is
(ii) miller compensation and (iii) lead introduced at the maximum frequencyf1 that
compensation. causes the intersection of the changed |A| and
the |1/β| plot at a relative slope of 20 dB/dec
Dominant Pole Compensation and at a frequency before the effect of the lowest
This is perhaps the simplest method and involves pole at fp1 (curve a). The closed-loop amplifier is
the introduction of a pole at a frequency that is now stable. For general purpose operational
lower than all the system poles such that the amplifiers that must be stable for all values of
effects of these poles on the open-loop magnitude feedback (unity-gain stable), the pole at f1 would
response is minimal and therefore the changed have to be moved to fD so that the modified plot
274 7 Feedback Amplifiers
Vo gm2 sC f RA RB
¼ gm1
Vi 1 þ s C A RA þ C B RB þ C f ðgm2 RA RB þ RA þ RB Þ þ s2 C A CB þ C f ðCA þ C B Þ RA RB
ð7:123Þ
where gm1 is the transconductance of the differ- located at a frequency fz ¼ gm2/2πCfwhich for
ential stage and gm2 is the transconductance of the typical transistor parameters is generally at a
second stage. The zero in this transfer function is high frequency and away from the main system
7.12 Stability and Compensation 275
poles. Considering the denominator polynomial where ω0p1 and ω0p2 are the representative poles of
and noting that it is second-order, it can be the system. One pole is usually dominant,
approximated by i.e. ω0p1 << ω0p2 . Therefore D(s) becomes
s s s s2
DðsÞ ¼ 1 þ 0 1þ 0 DðsÞ 1 þ þ 0 0 ð7:125Þ
ωp1 ωp2 0
ωp1 ωp1 ωp2
1 1 s2
¼1þs 0 þ 0 þ 0 0 ð7:124Þ Comparing the denominator of (7.123) and
ωp1 ωp2 ωp1 ωp2
(7.125) and equating coefficients yields
1 1
f 0p1 ¼
2πðC A RA þ CB RB þ C f ðgm2 RA RB þ RA þ RB Þ 2πgm2 RB C f RA
gm2 C f
f 0p2 ð7:126Þ
C A C B þ C f ðC A þ C B Þ
An increase in Cf reduces fp1 but increases fp2. and is usually much smaller than the value Cc
This effect is called pole-splitting and allows used in the pole-shifting method of compensation
movement of fp1 to a dominant frequency while applied at node A in Fig. 7.45.
also moving fp2 to a higher frequency away from In order to determine the value of Cf needed to
the first pole as shown in Fig. 7.47 (curve c). compensate a particular amplifier, we note that
The overall effect is that the available (7.123) can be written as
compensated open-loop gain is increased. It
sC f
should also be noted in (7.126) that the Vo g m1 RA g m2 R B 1 g
ðsÞ ¼ ð7:127Þ
m2
capacitanceCf has effectively been multiplied by Vi 1 þ ωp1 1 þ ωp2
s s
the Miller-effect factor gm2RB which is the
low-frequency gain of the stage. The effect over-
At frequencies close to the dominant pole, fre-
all is the increase of the value of the capacitance
quency (7.127) becomes
in parallel with RA to C 0f ¼ gm2 RB C f such that
f 0p1 can be written as f 0p1 ¼ 1=2πC 0f RA . The
capacitance Cf therefore need not be very large
feedback resistor as shown in Fig. 7.48. Experi- high frequencies by capacitor C2. The final stage
mental adjustment of the value to maximize is an emitter follower that prevents loading of R4
bandwidth while minimizing frequency response and gives the amplifier a low-impedance output.
peaking is a simple approach to utilizing the Resistor R7 along with R1is the overall feedback
technique. loop. Capacitor C3 is necessary for DC blocking
in the feedback loop. C1 and C4 are coupling
capacitors and C5 is required for stability.
7.13 Three-Transistor Feedback
Example 7.13
Amplifier
Design a transistor amplifier with a gain of
10 using the configuration of Fig. 7.49. Use a
The three-transistor circuit shown in Fig. 7.49 is
30 volt supply and transistors with current gain
another useful voltage amplifier configuration. It
of 100.
is a single-ended design that requires capacitor
coupling. The first stage is a common emitter
Solution We choose currents of 1 mA for Tr1,
amplifier that provides modest gain. Emitter resis-
2 mA for Tr2 and 5 mA for Tr3. We also select
tor R1 is included to allow the application of
R1 ¼ 220 Ω. This resistor should not be too large
voltage-series feedback around the amplifier.
otherwise the gain of this first stage would be
The second stage is another common emitter
unduly reduced and it should be large enough to
amplifier which provides the main voltage gain.
accommodate the negative feedback implementa-
Resistor R6 is a feedback loop that provides DC
tion. The design starts at stage 2 where the voltage
bias to Tr1. The loop is disabled at medium and
swing is greatest. For bias stability, let the voltage
at the emitter or Tr2 be about one tenth of the
supply voltage giving 3 volts. Hence R3 ¼ 3/
2 mA ¼ 1.5 k. For maximum symmetrical
swing in Tr2, R4 ¼ ð303 Þ=2
2 mA ¼ 6:75 k . Using
Kirchoff’s voltage law around the internal feed-
back loop, 3 ¼ IB1R6 + 0.7 + IC1R1 where IB1 and
IC1 are the base and collector currents of Tr1.
Using IB1 ¼ 1 mA/100 ¼ 0.01 mA, IC1 ¼ 1 mA
and R1 ¼ 220 Ω we get R6 ¼ 208 k. Since the
Fig. 7.48 Voltage amplifier with lead compensation
Ideas for Exploration: (i) Use the giving R9 + VRZ ’ 1 V/0.1 mA ¼ 10 k. Use
bootstrapping technique to increase the input R9 ¼ 8.2 k and potentiometer VRZ ¼ 5 k which
impedance of the amplifier. (ii) Use the is varied to zero the meter. Each transistor in the
bootstrapping technique to increase the gain of pair passes 50uA. These low currents ensure that
the second stage. the system can be operated from penlight batteries
over a long period before battery replacement.
DC Voltmeter Using Differential Amplifier The voltage across the base-emitter junction of
The circuit of a DC voltmeter is shown in the MPSA65 at 0.5 mA is about 1 V, and hence
Fig. 7.51. It uses a differential amplifier at the R8 ¼ 1 V/0.05 mA ¼ 20 k. Full-scale deflection
input followed by a pnp Darlington pair such as of the meter must occur for an input voltage of
the MPSA65 in the common emitter mode. The 100 mV. Hence 100 mV/(R10 + VRC) ¼ 100 μA
differential amplifier provides modest gain while giving R10 + VRC ¼ 1 k. Set R10 ¼ 680 Ω and
the Darlington pair produces higher gain. The potentiometer VRC ¼ 1 k which is used to cali-
meter is placed in a feedback loop around the brate the meter. The upper ranges of .3 V, 1 V,
amplifier such that the current through the meter 3 V, 10 V, 30 V and 100 V require input resistors
is defined by the input voltage and a resistor R1 ¼ 20k, R2 ¼ 90 k, R3 ¼ 290 k, R4 ¼ 1 M,
value. Thus, the meter current Im develops a volt- R5 ¼ 3 M, R6 ¼ 10 M as shown. These in con-
age Vx ¼ ImR10, (VRC ¼ 0) at the inverting input junction with R7 ¼ 10 k will attenuate the input
which is subtracted from the input voltage Vi at voltages such that 100 mV will appear at the input
the non-inverting input giving an error signal to the differential amplifier on each range.
(Vi Vx). This error signal is multiplied by the Ideas for Exploration: (i) Explain the mecha-
amplifier open-loop gain A giving an output nism whereby VRZ can zero the meter and VRC
Vo ¼ A(Vi Vx). Providing the open-loop gain can be used to calibrate the meter. (ii) Modify the
A of the system is large, the difference is circuit so that it can measure ac signals by placing
minimized by the feedback loop such that the meter in a diode bridge.
Im ¼ Vi/R10.
Current through the Darlington pair is set at RIAA Preamplifier
0.5 mA giving R11 ¼ 1.5 V/0.5 mA ¼ 3 k, while The signal from a phonograph disk (record) has a
that through the differential pair is set at 100uA characteristic in which the high frequencies have
280 7 Feedback Amplifiers
a higher amplitude than the low frequencies. the RIAA equalization characteristic. The design
Because of this, the preamplifier for such a signal of the network is algebraically quite tedious. The
must possess a frequency response characteristic values given in the diagram achieve close adher-
that is exactly the inverse of this. This means that ence to the required curve. The standard input
the amplifier must have a higher gain for low resistance for magnetic pickups is 47 k as
frequencies and a lower gain for high frequencies. achieved in the diagram.
The Record Industry Association of America Ideas for Exploration: (i) Refer to the paper
(RIAA) has developed a standard for magnetic “On RIAA Equalization Networks” by Stanley
phonograph cartridges that establishes the exact Lipshitz, Journal of the Audio Engineering Soci-
nature of this equalization characteristic. This is ety, Journal of the Audio Engineering Society 27
shown in Fig. 7.52 where two poles and one zero (6): 458–481, 1979. (ii) Use bootstrapping in the
are indicated. second stage to increase open-loop voltage gain.
The poles are at 50 Hz and 2100 Hz and the
zero is at 500 Hz. A practical implementation of JFET-BJT Preamplifier
this is shown n Fig. 7.53. It is based on the three- A circuit for a preamplifier using a JFET at the
transistor preamplifier of Sect. 7.13 with a CR input is shown in Fig. 7.54. This configuration is
network in the signal feedback loop which creates similar to the three BJT circuit except that the
7.14 Applications 281
input BJT is replaced by a JFET. This gives the 0.7 V. Hence R3 + R4 ¼ 9.7/1 mA ¼ 9.7 k. Since
circuit a very high input impedance. The JFET is the current in Tr1 is 0.5 mA, R2 ¼ 0.7 V/
connected in the common source mode. The out- 0.5 mA ¼ 1.2 k. From Shockley’s equation, for
put drives Tr2 which is a pnp BJT in the common a drain current of 0.5 mA in Tr1, VGS ¼ 3 V.
emitter mode while the final stage is an emitter Hence R7 ¼ 3/0.5 mA ¼ 6 k. If the circuit has to
follower to buffer the second stage and also pro- have a closed-loop gain of 11, then R6 ¼ 600 Ω.
vide a low-impedance output. The circuit uses a This value of resistor R6 which forms part of the
bipolar supply which of course enhances the signal feedback loop limits the gain of this stage.
available output voltage swing. Resistor This is overcome by bootstrapping resistor R3
R1 ¼ 10M connects the gate to ground and with capacitor C5 which increases the value of
enables Tr1 to be self-biased. Resistor R7 provides R3 for signals thereby increasing the gain of Tr2.
DC feedback that establishes quiescent current From Eq. (7.131), the value of the compensation
stability throughout the system. This feedback is capacitor is given by C4 ¼ 2π gf mAV where gm ¼ 1/
c
reduced for signals since resistor R6 is connected {(1/gm(Tr1)) + R6} ¼ 1/(100 + 600) ¼ 1.4 103
from the source of Tr1 to ground for signals via and gm(Tr1) ¼ 10 mA/V is used. Using fc ¼ 2
C3. This feedback arrangement is voltage-series MHz and AV ¼ 11, C 4 ¼ 7002π210
1
¼ 10pF.
6
11
feedback and hence, for sufficiently high open-
Ideas for Exploration: (i) Replace the
loop gain, the closed-loop gain of the system for
bootstrapping arrangement in the collector of
signals is 1 + R7/R6. C4 is the Miller compensa-
Tr2 by a constant current source.
tion capacitor. Because of the DC feedback loop
from the output to the source of Tr1, the design
AC Millivoltmeter
process for this circuit starts at the output.
The circuit shown in Fig. 7.55 is that of a wide-
Let the currents be 0.5 mA in Tr1, 1 mA in Tr2
band AC millivoltmeter for the measurement of
and 3 mA in Tr3. For a 9 V supply, let the
signals in the millivolt range. It uses two common
emitter of Tr3 be 0 volts for maximum symmetri-
emitter amplifiers so that the sensitivity of the
cal swing. Then, noting that the bias currents of
system exceeds that of the single transistor con-
the JFET and the emitter follower flow into R5,
figuration discussed in Chap. 4. The input transis-
the value of this resistor is given by R5 ¼ 9 V/
tor Tr1 is bootstrapped and therefore provides a
3.5 mA ¼ 2.6 k. The voltage at the base of
high input impedance to the circuit. This stage
Tr3which is connected to the collector of Tr2 is
drives a common emitter amplifier Tr2, the
282 7 Feedback Amplifiers
Fig. 7.55 AC
Millivoltmeter
mode while the second is in the common base Ideas for Exploration: (i) Try to increase the
mode. The output stage is another emitter fol- bandwidth of the system by selecting suitable
lower. In addition to these wideband stages, the high-frequency transistors that have low junction
system has feedback from output to the inverting capacitances.
input. The result is very wideband operation.
With a 15 V supply, using R1 ¼ 12 kand Research Project 1
R2 ¼ 4.7 k sets the voltage at the base of Tr1 at The circuit shown in Fig. 7.57 is that of an audio
4.2 V. Let the current in each transistor in the compressor or constant volume amplifier. It
differential pair be 1.5 mA for good high- provides a near constant amplitude output signal
frequency response. Then R3 ¼ (4.2 0.7)/ for widely varying input signal amplitude by
3 mA ¼ 1.2 k. Since the base voltage of Tr2 is incorporating automatic gain control. Such a cir-
about 4.2 V, the maximum peak-to-peak swing cuit is useful in reducing the dynamic range of
available at Tr3 is approximately (15 4.2)/ music signals in order to enable better processing
2 ¼ 5.4 V. Hence, for maximum symmetrical of these signals particularly in the recording
swing in Tr2, R4 ¼ 5.4/1.5 mA ¼ 3.6 k. In order industry. The first stage is an emitter follower
to drive transmission lines which typically have that is biased for maximum symmetrical swing.
characteristic impedances of between 50 ohms The second stage is a JFET that is used as a
and 300 ohms, let the current in the emitter fol- voltage-controlled resistor. With a 15 V supply,
lower output stage be 10 mA. The voltage at the the drain of this JFET is set at approximately 2 V
base of Tr3 is (15 5.4) ¼ 9.6 V, and therefore by resistors R4 and R5 which form a potential
the resistance in the emitter of this transistor is divider. Resistor R6 ¼ 220 k grounds the gate of
8.9/10 mA ¼ 890 Ω. In order to get 4.2 V at the the JFET and R7 ¼ 2.2 k along with the drain
base of Tr2, for R6 ¼ 470 Ω thenR5 ¼ 520 Ω. The voltage set the drain current which positions the
closed-loop gain of the system is set by 1 + R5/R operating point in the ohmic region. The drain-
where R is the value of VR1 ¼ 1 k added to the source resistance in conjunction with R7 provides
value of R7 ¼ 47 Ω, in parallel with R6. At a gain the dynamic adjustment that results in automatic
of 17 dB, the simulated circuit had a bandwidth of gain control. The output at the source of the FET
11 MHz. All capacitors need to be large. drives the common emitter amplifier Tr3 while the
emitter follower Tr4 buffers the final output.
not now be necessary as base bias would be Ideas for Exploration: (i) Use a feedback pair
provided by the filter network resistor R1. The in place of the single transistor. This enables gain
basic system is shown in Fig. 7.59 and the to be introduced.
associated transfer function with C1 ¼ C2 ¼ Cand
R1 ¼ R, R2 ¼ R/2 is given by Research Project 4
2 A tone control circuit adjusts the amplitude of
f
j high (treble) and low (bass) frequencies in an
Vo fc
¼ pffiffiffi f f 2 ð7:134Þ audio signal in order to improve the listening
Vi
1 þ 2j f þ j f quality perception of the signal. It is usually
c c
placed just before the power amplifier in the
where the corner frequency is audio signal chain. The most widely used circuit
pffiffiffi for accomplishing this is the Baxandall tone con-
2
fc ¼ ð7:135Þ trol circuit. An implementation of this is shown in
2πRC Fig. 7.60. The stage driving the Baxandall tone
This is a Butterworth response for the high-pass control network at A is an emitter follower while
filter. For a lower cut-off frequency of 50 Hz, the two-stage amplifier following the network at
using C ¼ 1 μF, Eq. (7.135) gives R ¼ 4.5 k. B is a standard design with feedback biasing
286 7 Feedback Amplifiers
discussed in Chap. 5. Feedback from the 5. Explain what is meant by “gain stabilization”
two-stage amplifier is applied to the network at in a negative feedback amplifier and why it is
C. Potentiometer VRB lifts/cuts the bass important in the design of such amplifiers.
frequencies, while potentiometer VRT lifts/cuts 6. Show that the magnitude of the fractional
the treble frequencies. The project involves change in the gain of an open-loop amplifier
researching Baxandall tone control theory and is reduced by a factor equal to the loop gain in
designing the network. the closed-loop amplifier.
Ideas for Exploration: (i) Use bootstrapping to 7. An amplifier with gain 1500 has a gain
increase the gain in the second stage. This will change of 15% due to aging. Calculate the
provide increased feedback and hence lower change in gain of the feedback amplifier if the
distortion. feedback factor β ¼ 0.2.
8. An amplifier with a transfer function A ¼
k
1þjf = f o has a low-frequency open-loop gain
Problems of k ¼ 120, 000 and an open-loop bandwidth
of fo ¼ 1 Hz. For a closed-loop gain of
1. Draw and label the equivalent voltage ampli-
100, calculate the closed-loop bandwidth of
fier. Indicate the levels of the input and out-
the amplifier.
put resistances relative to the source and load
9. Show that negative feedback reduces the dis-
resistances and show how negative feedback
tortion in the output stage of an amplifier by a
can be applied around the amplifier.
factor equal to the loop gain.
2. Draw and label the equivalent current ampli-
10. Discuss the effect of negative feedback on the
fier. Indicate the levels of the input and out-
input and output resistances of the four clas-
put resistances relative to the source and load
ses of amplifiers.
resistances, and show how negative feedback
11. Show that the bandwidth of a feedback
can be applied around the amplifier.
amplifier increases with the application of
3. State four advantages of negative feedback in
negative feedback.
amplifiers, and show how such feedback can
12. Evaluate the voltage gain, input resistance
be implemented around a trans-resistance
and output resistance for the voltage feed-
amplifier.
back amplifier in Fig. 7.61. For the transistors
4. Determine the amount of negative feedback
hfe ¼ 120, hie ¼ 5 k, hre ¼ hoe ¼ 0.
in an operational amplifier having an open-
13. For the current amplifier shown in Fig. 7.62,
loop gain of 110 dB and a closed-loop gain of
determine the closed-loop current gain, input
20 dB.
Bibliography
J. Millman, C.C. Halkias, Integrated Electronics: Analog
and Digital Circuits and Systems (McGraw Hill,
New York, 1972)
A.S. Sedra, K.C. Smith, Microelectronic Circuits, 6th edn.
(Oxford University Press, Oxford, 2011)
Operational Amplifiers
8
V 0 ¼ Ad ðV 1 V 2 Þ ¼ Ad V D ð8:1Þ
The operational amplifier or op-amp is one of the
most widely used linear integrated circuits today.
Its great popularity arises from its versatility, A positive-going signal at the input defined as
usefulness, low cost and ease of use. It was the non-inverting (+) input produces a positive-
introduced in the 1940s mainly for use in analog going signal at the output, while a positive-going
computers where it performed mathematical signal at the input defined as the inverting ()
operations (hence the name) including addition, input produces a negative-going signal at the
multiplication, integration and differentiation. output.
The device later found ready application in a The ideal operational amplifier possesses sev-
wide range of circuits and functions, many of eral important properties. Its voltage gain is infi-
which will be discussed in this chapter. At the nite (very large), i.e. Ad ¼ 1, and is constant for
end of the chapter, the student will be able to: all frequencies which means that the operational
amplifier has infinite bandwidth. It has zero out-
• Understand the operation of the device put impedance and infinite input impedance. The
• Derive the mathematical relations that govern ideal operational amplifier also has the property
its operations that VO ¼ 0 when VD ¼ 0. In other words, no
• Use the device in a range of linear circuit offset voltage is present at its output terminal
applications when the differential input voltage is zero. The
operational amplifier can therefore be represented
by the ideal voltage-controlled voltage source
shown in Fig. 8.2. Because Ad ¼ 1, the differen-
8.1 Introduction tial input voltage in an ideal operational amplifier
is zero. Also, since the input resistance in either
The operational amplifier in Fig. 8.1 is a direct- input terminal is infinite, there is no current flow
coupled voltage amplifier that performs certain into or out of either terminal.
mathematical operations. It has differential input The operational amplifier is an active device
and a single-ended output. It responds only to the typically powered by a bipolar or dual-polarity
voltage difference VD between the two inputs supply V. This enables both input and output to
given by VD ¼ V1 V2 and not to their common be referred to ground or zero potential.
potential. It amplifies this differential input volt- Depending on the technology, it is constructed
age by a large factor Ad which is the differential from bipolar transistors and/or field-effect
gain of the amplifier and delivers an output volt- transistors. The operational amplifier may operate
age given by
# Springer Nature Switzerland AG 2021 289
S. J. G. Gift, B. Maundy, Electronic Circuit Design and Application,
https://doi.org/10.1007/978-3-030-46989-4_8
290 8 Operational Amplifiers
VO R
¼ 2 ð8:9Þ
Vi R1
C ¼ 1=ð2π f L R1 Þ ð8:12Þ
Example 8.4
A non-inverting amplifier has R1 ¼ 2 k and
R2 ¼ 22 k; determine the gain of the system.
C ¼ 1=ð2π f L RB Þ ð8:27Þ
Fig. 8.8 Non-inverting amplifier with bootstrapping The resulting circuit shown in Fig. 8.10 has
high input impedance, low output impedance and
Hence the input impedance Zi ¼ Vi/IB to the a unity-gain closed-loop transfer function. The
bootstrapped amplifier is given by high input impedance Zi ¼ 1 is again reduced
by any bias resistor that may be required at the
Z i ¼ V i =I B ¼ ð1 þ AβÞRB ð8:26Þ
input.
For a 741 operational amplifier, RB is typically Again, for AC signals, a high input impedance
100 k with A ¼ 105 at low frequencies. C2 sets the may be restored by using bootstrapping as shown
lower operating frequency below which the tech- in Fig. 8.11. From Eq. (8.26), since β ¼ 1, the
nique becomes ineffective. Because of the avail- input impedance is given by Zi ¼ (1 + A)RB. For a
ability of high input impedance op-amps with low 741, A ¼ 105 at low frequencies, and therefore the
bias current requirements, this technique may not low-frequency input impedance is given by
be necessary since the value of RB can be quite Zi ¼ (1 + 105) 100 k ¼ 103M which is an
high in such devices. Also, since bootstrapping is extremely high value!
only effective for AC signals, C1 is introduced to
eliminate DC.
8.5 Summing Amplifier
Non-inverting AC Amplifier
For the amplification of AC signals, a capacitor As mentioned in Sect. 8.1, the operational ampli-
C must be added in series at the input as shown in fier derives its name from the fact that it is capable
Fig. 8.9. Its value is set by the resulting lower of performing mathematical operations. One such
cut-off frequency fL and is given by operation is the ability to perform the sum of
input signals with the added advantage of having
gains larger than one. To see how this is possible,
8.5 Summing Amplifier 295
Rf Rf Rf Rf
VO ¼ V1 þ V2 þ V3 þ þ Vn
R1 R2 R3 Rn
ð8:31Þ
Fig. 8.10 Voltage follower Hence the output voltage is the inverted sum of
the scaled input voltages. Because each input
resistor is connected to a virtual earth, it follows
that the input resistance seen by each source
Vk(k ¼ 1, 2, 3, . . .n) is Rk. This makes the sum-
ming amplifier very useful for mixing audio
signals from different sources. As in the single
input inverting amplifier, the circuit performs
inversion of the inputs.
The overall gain of the circuit is set by Rf,
while the individual channel gains are set by the
individual input resistors R1, R2, R3, . . ., Rn which
Fig. 8.11 High input impedance AC voltage follower are also the input impedances of the respective
inputs. Note that if R1 ¼ R2 ¼ R3 ¼ . . . ¼ Rn ¼ Ri,
we examine an extension of the inverting ampli- then
fier circuit shown in Fig. 8.12. Here, several Rf
(n) input signals drive input resistors connected V0 ¼ ðV þ V 2 þ V 3 þ . . . þ V n Þ ð8:32Þ
Ri 1
to the inverting terminal. By noting that the
inverting node of the operational amplifier Further if R1 ¼ R2 ¼ R3 ¼ . . . ¼ Rn ¼ Rf, then
because of negative feedback acts as a virtual
ground, then KCL dictates that V O ¼ ðV 1 þ V 2 þ V 3 þ þ V n Þ ð8:33Þ
Fig. 8.14 Diagram for Example 8.7 Fig. 8.16 Diagram for Example 8.9
8.6 The Differential Amplifier 297
at the non-inverting terminal of the operational Solution From (8.35) this condition occurs when
amplifier. Hence the we have equal resistances in the circuit of
output voltage is given by
Fig. 8.13. In that case choose R1 ¼ R2 ¼ 10 k
V ^O ¼ RaRþR
b
1 þ R2
R1 V 2 . Adding the two results
b
and Ra ¼ Rb ¼ 10 k.
from superposition V O ¼ V 0O þ V ^O yields the
result Example 8.11
! Design a circuit to realize Vo ¼ 3V2 2 V1.
R2 1 þ RR2
VO ¼ V1 V2 ð8:34Þ
R1 1 þ RRab Solution Right away the circuit of Fig. 8.18 and
Eq. (8.36) come to mind. We will use Fig. 8.18
because of the advantages previously mentioned.
Two interesting scenarios arise out of the cir- Hence the design equations are RR21 ¼ 2. Note 1 þ
cuit of Fig. 8.17. The first occurs when the resis-
tance ratios are all equal. That is, RR21 ¼ RRba . In that
R2
R1 ¼ 3 is satisfied by default. Letting R1 ¼ 1 k
case (8.34) simplifies to yields R2 ¼ 2 k which is the desired result.
R2 Example 8.12
VO ¼ ðV V 2 Þ ð8:35Þ
R1 1 For the special case of the difference amplifier
shown in Fig. 8.19, show that the relationship
and the output is thus proportional to the true
between the output and the inputs is V O ¼
difference of two inputs (but inverted).
The second case occurs when Ra ¼ 0 and RR21 ðV 1 V 2 Þ.
Rb ¼ 1. This is shown in Fig. 8.18. In that case
Vo is given by Solution The differential amplifier amplifies the
difference between two signals to be amplified.
R R
VO ¼ 1 þ 2 V2 2 V1 ð8:36Þ The potential at the non-inverting terminal is not
R1 R1 zero, but instead is set by the potential divider
arrangement of R1 and R2 and is given by
298 8 Operational Amplifiers
Exercise 8.1
Show that in the realization of Vo ¼ 2 V1 3 V2,
only Fig. 8.17 can be used and one solution is to
choose Ra ¼ Rb ¼ R1 ¼ 1 k and R2 ¼ 3 k.
Fig. 8.24 A
transimpedance amplifier
Table 8.1 A comparison of various op-amp technologies with the ideal op-amp
Bipolar JFET CMOS Ideal op-amp
Ri 6MΩ 1013Ω 1012Ω 1
Ro <100 Ω <100 Ω <100 Ω 0
Ad 103 106 103 106 103 106 1
i1 ’ i2 <1pA 0
40 μA 1pA
Vos 25 μV 5 mV ’0 0
amplifiers, this current is virtually nonexistent and Ri and Ro have been included in the inverting
can be neglected altogether. Real op-amps also configuration as shown in Fig. 8.30.
have offset voltages Vos not modelled in As before, using KCL yields
Fig. 8.29, but its effect is to cause the output to
Va Vi Va Vo Va 0
be nonzero when the differential input is zero. þ þ ¼0 ð8:57Þ
R1 R2 Ri
Finally, the open-loop voltage gain varies across
op-amps but is normally large. Typically values V o ðAo V a Þ V o V a
range from 1000 to 107. The ubiquitous 741 has a þ ¼0 ð8:58Þ
Ro R2
typical voltage gain of 105.
In all operational amplifiers, the gain Ad is Solving (8.57) and (8.58) yields
actually a frequency-dependent quantity, that is, Vo R 1
Ad Ad(s), but for the purposes of the following ¼ 2 ð8:59Þ
Vi R1 ðRo þR2 Þ R 1 þR 2 þR i
1 1 1
section, we shall assume it is constant for all 1þ Ao R Ro
frequencies. In other words, Ad(s) ¼ Ao where 2
Ao where
Av ¼ ð8:60Þ
1 þ jf = f o
8.13 Frequency Effects 305
f o ¼ f o Ao β ð8:66Þ
f o ¼ f o Ao = 1 A f ðDC Þ ð8:77Þ
Fig. 8.36 Frequency response of inverting amplifier
The maximum feedback 20 log Ao which
occurs for β ¼ 1 results in Af(DC) ¼ 0. This
which for large loop gains such that Aoβ > > 1
corresponds to R1 ! 1 and/or R2 ¼ 0.
becomes
Finally, in the inverting mode, it is possible to
R2 replace R1 and Vi by a current source such that the
A f ðDC Þ ¼ ð8:73Þ
R1 input signal is a current Ii as shown in Fig. 8.37.
Let R2 ¼ RF. The circuit now functions as a
The full Eq. (8.71) for large loop gains reduces transimpedance amplifier (current in, voltage
to out), but the feedback remains voltage-series.
R2 =R1 The transfer function RM Vo/Ii is given by
Av ¼ , Ao β >> 1 ð8:74Þ
1 þ jf =f o RF :Ao =ð1 þ Ao Þ
RM ¼ ð8:78Þ
1 þ jf = f o ð1 þ Ao Þ
where
which reduces to
f o ¼ f o Ao β ð8:75Þ
RF
is the closed-loop bandwidth and the RM ¼ ð8:79Þ
1 þ jf =f o
low-frequency closed-loop gain A f VVoi ðDC Þ is
given by where
A f ðDC Þ ¼ R2 =R1 ¼ 1 1=β ð8:76Þ f o ¼ f o Ao ð8:80Þ
The expression for the closed-loop bandwidth and the low-frequency closed-loop
(8.75) of the inverting configuration is exactly the transimpedance RM(DC) is given by
same as the expression (8.66) for the closed-loop
bandwidth of the non-inverting configuration. RM ðDC Þ ¼ RF ð8:81Þ
Since the magnitude of the low-frequency gain
R2/R1 is less than that of the non-inverting config-
This is the well-known voltage-to-current con-
uration 1 + R2/R1, the frequency response plot is
verter. Since the feedback mode is still voltage-
shifted downward relative to that of the
series and the impedance of the current source is
non-inverting configuration as shown in
(ideally) infinite, it follows that R1 ! 1 and
Fig. 8.36.
hence β ! 1. This configuration therefore
corresponds to maximum voltage-series feedback
It follows also that the closed-loop bandwidth
with 20 log AodB, and as is evident from (8.80),
for the inverting configuration also varies as β is
the closed-loop bandwidth f o is the GBP of the
varied and hence as the low-frequency gain
associated amplifier. It is interesting to note that
Af(DC) is varied. Specifically, using (8.76) in
because of the high impedance of the signal
(8.75) gives
source, the closed-loop bandwidth f o is
8.14 Non-ideal Effects 307
independent of the closed-loop low-frequency FET transistors. The input bias current IB is typi-
transimpedance RF and hence can be varied with- cally in the range 10 nA to 1 μA for bipolar
out affecting the bandwidth. op-amps and <100pA for FET input op-amps.
Although typically in the nanoampere to micro-
ampere range, input offset currents cause
problems by creating unwanted voltage drops
8.14 Non-ideal Effects
across input resistors. The overall effect is to
cause the output of the op-amp to have a small
Apart from bandwidth issues, there are other
offset voltage depending on the amplifier
factors or non-ideal effects that greatly influence
configuration used.
the performance of an op-amp. These include
offset voltages and currents, common mode rejec-
Common Mode Rejection Ratio (CMRR)
tion ratio, dynamic range and slewing. In consid-
CMRR is the ratio of the differential gain of the
ering some of these, we begin with some
op-amp to the common mode gain expressed in
definitions.
dB. Stated mathematically it is
Offset Voltages (Vos) Offset voltages in an A
op-amp occur because of mismatch in the input CMRR ¼ d ð8:84Þ
Acm
stages which results in an output voltage even
when the differential input voltage is zero. Common mode gain Acm is the ratio of the
Hence an offset voltage can be viewed as the output voltage of an op-amp to the input voltage
voltage that must exist between the two input applied to both inputs simultaneously, i.e. with
terminals of the op-amp in order to make the the inputs of the op-amp electrically connected.
output voltage zero. Such voltages can have Under these conditions, there ought to be no
values of both polarities. A typical range of output voltage from the op-amp as it ideally
values might be 10 μV to 10 mV for bipolar responds only to a differential voltage. The prin-
op-amps and as high as 50 mV for FET input cipal reason for an output in these circumstances
op-amps. is that the gain at one input is slightly different
from that of the other. This is an undesirable
Input Offset Currents (Ios) Offset currents exist effect since the objective of having differential
because there is a finite current flowing in/out of inputs is to cancel any common mode signals
the input terminals that is required to bias the that may be present. In op-amp circuits
input transistors of the op-amp. By definition, Ios employing negative feedback, common mode
is equal to the difference of the input bias currents gain is usually not a problem, but in instrumenta-
or tion amplifiers without overall feedback, it can
become problematic. Because Ad is frequency
I os ¼ jI B1 I B2 j ð8:82Þ dependent, it means that the CMRR is also fre-
where I B1 and I B2 are the input bias currents quency dependent. Manufactures usually specify
which may flow into or out of the op-amp. Some- only the low-frequency value of the CMRR as its
times a manufacture may specify an op-amp’s value decreases with increasing frequency.
input bias current (IB) which is the average of Expressed in dB the CMRR is given by
the two currents IB1 and IB2. That is, A
CMRRdB ¼ 20 log 10 d ð8:85Þ
I B1 þ I B2 Acm
IB ¼ ð8:83Þ
2
For a typical op-amp, CMRR is in the range
Bias currents are worse for op-amps made with 60–70 dB. More high-performance op-amps
bipolar input transistors than for those made with however have CMRR of the order of 120 dB.
308 8 Operational Amplifiers
In some op-amps, it is possible to null the output voltage lies in the range 110 mV Vo
output voltage produced by the offset sources (c) 110 mV and with the nominal value in the
using an external voltage or as in the case of the range 55 mV Vo(nom, c) 55 mV. Therefore,
741 op-amp a potentiometer connected across in this case compensation can be omitted, but it
two offset “null” points. For the 741 these are will not be the case for all op-amps.
pins 1 and 5 on the 8-pin DIP package with the
potentiometer wiper connected to the negative
supply. Note however that temperature changes
8.14.2 CMRR and PSRR
in the components will eventually cause the out-
put voltage to change with temperature. For other
As stated earlier the CMRR is a measure of how
op-amps laser trimming techniques are used to
well an op-amp amplifies differential signals and
keep the output offset voltage to a minimum.
rejects common mode ones as given by the ratio
Example 8.16 A A
Using the specifications for the LF351 JFET input CMRR ¼ d ¼ o ð8:89Þ
Acm Acm
op-amp, determine the output offset voltage for
the uncompensated and compensated inverting where we shall assume that we are working at low
amplifier with resistors R2 ¼ 10 k and R1 ¼ 1 k. frequencies so that Ad ¼ Ao can be used. An
equation that fully describes the effects of both
Solution From the data sheet of the LF351 differential and common mode gain is therefore
op-amp, it lists maximum values of Vos ¼ 10 given by
mV, IB ¼ 200 pA and Ios ¼ 100 pA all at
V þ V2
25 C. We shall assume room temperature opera- V o ¼ Ad ðV 1 V 2 Þ þ Acm 1
2
tion. For the uncompensated op-amp, Rc ¼ 0, but
¼ Ad V D þ Acm V cm : ð8:90Þ
IB2 is not specified as required by (8.88). We can,
however, make a best guess estimate by using In Eq. (8.90), the factor of ½ is there because
(8.82) and (8.83) to solve for IB2. Solving yields under common mode conditions when
two values for IB2 of 450pA and 350pA, but let us V1 ¼ V2 ¼ Vcm, then Vo ¼ AcmVcm which is
use the smaller of the two numbers. Substitution consistent with the definition of common mode
of the relevant numbers into (8.88) yields gain. Equation (8.90) can therefore be rewritten as
V oðncÞ ¼ 10 kΩð10 pAÞ þ 350 pAð10 kΩÞ h i
V
V o ¼ Ao V D þ cm ð8:91Þ
10kΩ CMRR
þ ð10 mVÞ 1 þ
1kΩ or
¼ 0:1 μV þ 3:5 μV 110 mV h i
1
V o ffi Ao V 1 V 2 1 ð8:92Þ
CMRR
In other words, the output offset voltage lies
within the range 110 mV V0(nc) 110 mV. if V1 ’ V2. Both representations are equally
Clearly for the uncompensated amplifier, the off- valid, but (8.92) indicates that we can simply add
set voltage is the term that dominates because Ios a voltage source of value V2/CMRR to the positive
and IB are so small. With that in mind if the input on an op-amp to model its effect. Such a
nominal value of Vos(nom) ¼ 5 mV is used, model is shown in Fig. 8.39.
then the output voltage lies in the range 55 In a similar manner, we can describe the PSRR
mV Vo(nom, nc) 55 mV. For the compensated by the equation Vo ¼ AoVD + SΔVsupplyAo
amplifier, the situation does not change much or
since the second term in (8.88) is zero. Thus the
310 8 Operational Amplifiers
Fig. 8.39 Op-amp model that accounts for both CMRR 8.14.4 Slew Rate
and PSRR
CMRR ¼ ¼ RG
ð8:95Þ change of voltage which we define as the slew
Acm 1þRR
1þRa
2
1 rate of
Rb
dV o I
Clearly if R1 ¼ Ra and R2 ¼ Rb as stated SRþ ¼ ¼ b ð8:97Þ
dt Cc
before, then the CMRR ! 1. In practice this
will not be the case because of resistance The positive sign is included in (8.97) to show
tolerances. However, it is not uncommon to easily that it represents the positive slew rate. By a
achieve CMRRs in excess of 100 dB. Like similar reasoning, if the polarity of the input
8.14 Non-ideal Effects 311
Fig. 8.40 Simplified model of a three-stage op-amp Fig. 8.41 An alternative op-amp model
differential voltage were reversed, then Ib maximum slope at its zero crossings and is
would flow out of Cc generating a negative slew given by
rate of
dV o ðt Þ
¼ V a ω ¼ 2πV a f ð8:99Þ
dV o I dt
SR ¼ ¼ b ð8:98Þ max
dt Cc
For linear op-amp operation, the slew rate
The primary reason there are two different given in (8.99) must not exceed the slew rate of
values for positive and negative slew rates is the op-amp; otherwise slew rate limiting occurs.
that sometimes the input stage is not completely This imposes an upper bound on the frequency of
symmetric and different values of output current operation of the op-amp for a given sinusoidal
Io exist for negative and positive differential output amplitude. That bound is sometimes
inputs. A typical value is therefore quoted by referred to as the full power bandwidth (BWp)
most manufacturers with a square wave or step and is given by
input being the most commonly used test signal.
The topology shown in Fig. 8.40 is not the SR
BW p ¼ : ð8:100Þ
only topology used in designing op-amps. New 2πV a
high slew rate op-amps such as the LT1364 Equation (8.100) indicates that bandwidth can
employ a different topology that offers higher be traded for output amplitude and vice versa to
slew rates than the approach of Fig. 8.40 and is avoid slew rate limiting. As an example, consider
shown in Fig. 8.41. This is achieved by sensing an op-amp with a slew rate of SR ¼ 13 V/μs and
currents that are proportional to the difference of maximum peak output voltage Va ¼ 14 V being
the differential inputs. Buffers provide high- powered from a 15 V supply. The full power
impedance inputs to the op-amp. Resistor bandwidth can then be calculated to be
R limits the maximum current that can flow BWp ¼ 13 106/(2π 14) ¼ 148 kHz. Its
through the buffers. The sensed currents are then theoretical peak-to-peak output swing versus fre-
copied, and their difference is used to drive a quency will therefore resemble the plot shown in
high-impedance node labelled z in Fig. 8.41 to Fig. 8.42. Therefore in order to operate this
achieve high gain in the op-amp. The voltage at op-amp at 1 MHz, the maximum peak output
that node is buffered to preserve the high gain. voltage Va that can be delivered using (8.100) is
The high slew rates come for the fact that Cc can given by Va ¼ SR/2πBWp ¼ 13 106/
be made small, while the sensed currents I1, 2 can 2 π 106 ¼ 2.1 V. Finally, manufactures
be large. often specify a plot like that in Fig. 8.41 but
If the output of the op-amp is a sine wave seldom provide BWp in their specifications
described by the equation Vo(t) ¼ Va sin ωt because it depends on the supply voltage, gain
where Va is the amplitude and ω is the angular and load conditions.
frequency, then the output signal achieves its
312 8 Operational Amplifiers
4.5
3.5
2.5
2
10 0 10 2 10 4 10 6 10 8
Frequency in Hertz
Inverting Amplifier
If an input resistor R1 is now added to the
inverting terminal and a voltage source drives
this amplifier, an inverting voltage amplifier Fig. 8.46 CFA inverting amplifier
results as shown in Fig. 8.46 For this configura-
tion, noting that the input resistance at the
inverting terminal is zero, then Ii ¼ Vi/R1, and f o ¼ f o ð1 þ Z o βÞ ¼ f o ð1 þ Z o =R2 Þ ð8:114Þ
therefore substituting for Ii in Eq. (8.109) yields which is the same as the tras-resistance case in Eq.
the transfer function given by (8.110 ). Note that while this value is dependent on
Vo R2 =R1 R2, it is not dependent on R1. Thus, the closed-loop
¼ ð8:112Þ bandwidth of the CFA in the inverting configura-
V i 1 þ jf = f o ð1 þ Z o βÞ
tion is independent of resistor R1. This fact makes
which for low frequencies becomes the CFA very attractive compared to the op-amp
whose gain and bandwidth are directly related by
Vo R
¼ 2 ð8:113Þ the gain bandwidth product. It follows that resistor
Vi R1
R1 may be changed to vary the CFAs closed-loop
gain via R2/R1 and the bandwidth will remain
Equation (8.113) indicates that the expression constant provided R2 is fixed.
R2/R1 for the closed-loop gain of the inverting
amplifier for the CFA is the same as that for the Non-inverting Amplifier
op-amp. The 3 dB closed-loop bandwidth is To utilize the CFA in a non-inverting application,
given by let us consider the circuit of Fig. 8.47 with
8.15 The Current Feedback Amplifier 315
After substitution for Z and manipulation, we As in the case of the op-amp, if R1 ! 1, then
get the closed-loop gain is unity, and the bandwidth is
given of course by Eq. (8.120). While in the case
Zoβ
Vo 1 þ R
R
: 1þZ β of the op-amp, we can go further and set R2 ¼ 0 to
¼ ð8:117Þ
1 o
15
-5
-10
R 1=222.2
R 1=2k
-15
R 1=924
R 1=
-20
-25
10 1 10 2 10 3 10 4 10 5 10 6 10 7 10 8
Frequency in Hertz
Differential Amplifier
The CFA can also be used as a differential ampli-
fier in much the same way as its op-amp counter-
Fig. 8.49 The CFA connected as a buffer part. In that case the circuit shown in Fig. 8.51 can
be used, and the output is given by
R2
Vo ¼ ðV V 2 Þ ð8:121Þ
R1 1
Integrator
The most notable exception to the CFA’s general
use is as an integrator. The problem that the CFA
Fig. 8.50 The CFA used as a mixer has in the integrator configuration is that the
8.16 Applications 317
10
Closed Loop Gain in dB
-5
-10 R 1=222.2
R 1=2k
-15 R 1=924
R 1=
-20
-25
10 1 10 2 10 3 10 4 10 5 10 6 10 7 10 8
Frequency in Hertz
ment the gain can be varied from 1 to 101. low-frequency response down to 20 Hz, the
(ii) Introduce offset nulling in order to minimize input coupling capacitor value is given by
offset voltage at the output. Consult the C ¼ 1/2 π 20 105 ¼ 0.08 μF. A value of
manufacturer’s literature on implementing this. C ¼ 0.1 μF is used. A coupling capacitor at the
output may be necessary if the amplifier is
8.16 Applications 319
Bootstrapped AC Amplifier
In the circuit in Fig. 8.56, by moving the resistor
RB to the position shown in Fig. 8.57, this resistor
is effectively bootstrapped. As a result, the
Fig. 8.54 DC non-inverting amplifier
low-frequency input impedance goes from
Zi ¼ 100 k to Zi ¼ 100 k loopgain ¼ 100
k 103 ¼ 100 M where loop gain is given by
C
open-loop gain/closed-loop gain. Capacitor C2 is
calculated by ensuring that its reactance at the
lowest frequency under consideration is small
compared with RB. Using 1 Hz we have C2 ¼ 1/
2 π 1 105 ¼ 1.6 μF. The gain of the system
continues to be 101.
Fig. 8.56 AC non-inverting amplifier with 100% DC Fig. 8.59 (a) AC voltage follower; (b) bootstrapped volt-
feedback age follower
Inverting Amplifier
This circuit in Fig. 8.60 (a) is an inverting ampli-
fier. It has a gain R2/R1 ¼ 100 k/
10 k ¼ 10. Unlike the non-inverting case,
the input impedance is R1 ¼ 10 k which is com-
paratively low. The output signal is out of phase
with the input signal. In order to reduce any
voltage offset at the output, a resistor R3 ¼ R1//
Fig. 8.57 Bootstrapped AC amplifier R2 ¼ 9 k is placed between the non-inverting
terminal and ground as shown in Fig. 8.60 (b).
Fig. 8.62 Modified inverting amplifier given by C1 ¼ 1/2 π 20 104 ¼ 0.8 μF. A
value of 1 μF is suitable for both capacitors.
Note however that at low gains, R3 will be a load
on the op-amp and hence may have to be Ideas for Exploration (i) Convert the circuit
increased. into a variable gain inverting amplifier by
replacing resistor R2 by a 1 M potentiometer in
AC Inverting Amplifier series with a 10 k resistor as shown in Fig. 8.67
This inverting amplifier processes AC by the for the DC case. With this arrangement the gain
inclusion of coupling capacitors C1 and C2 as can be varied from 1 to 100.
shown in Fig. 8.63. These capacitors block any
DC and ensure that only AC passes through the Differential Amplifier
amplifier. Using a lower cut-off frequency of The circuit in Fig. 8.64 is that of a differential
20 Hz and noting the input impedance to the amplifier. It has a gain R2/R1 ¼ 100 k/
amplifier is R1 ¼ 10 k, then capacitor C1 is 10 k ¼ 10 and an output given by Vo ¼ 10
322 8 Operational Amplifiers
(V1 V2). If all resistors are made equal say R11 ¼ 3.3 k protect the meter from current over-
100 k, then the circuit would have unity gain load. The resistor chain R1 to R9 forms an attenu-
and would be that of a subtractor with output ator that provides a voltage range 0.1 V to 1000 V
given by Vo ¼ V2 V1. in nine ranges. These values are as follows:
R1 ¼ (680 k + 120 k), R2 ¼ 100 k, R3 ¼ (68 k + 12
DC Voltmeter k), R4 ¼ 10 k, R5 ¼ (6.8 k + 1.2 k), R6 ¼ 1 k,
This circuit shown in Fig. 8.65 is that of a DC R7 ¼ (680 Ω + 120 Ω), R8 ¼ 100 Ω and
voltmeter. It uses the 741 op-amp in a voltage-to- R9 ¼ 100 Ω. Finally, any offset voltage must be
current configuration since the meter is included reduced to zero, and this is done using potentiom-
in a current-series feedback loop. Thus, the cur- eter VR1, while the input leads are shorted
rent Im through the meter M is sampled by resistor together.
Rm across which a feedback voltage is developed
that is applied in series with the input voltage Vi. Ideas for Exploration (i) Convert the circuit
The result is that the ratio Im/Vi is stabilized into a current meter by connecting a resistor RC
because of the feedback and the current Im across the input terminals as shown in Fig. 8.66.
through the meter is simply Im ¼ Vi/Rm. The The current to be measured is passed through this
input impedance to the non-inverting input is resistor, and the choice of RC is given by RC ¼ 0.1
already high without feedback (1 M) and is V/Imx where Imx is the maximum current. Thus,
increased by the feedback. With Rm ¼ 100 Ω, for a 1 mA range, RC ¼ 0.1/1 mA ¼ 100 Ω, or for
full-scale deflection with a 1 mA milliammeter a 100uA range, RC ¼ 0.1/100 uA ¼ 1 k.
occurs with Vi ¼ 100 mV. Diodes D1 and D2
along with resistor R10 ¼ 10 k protect the input High-Impedance DC Voltmeter
of the op-amp from excessive voltage, while The circuit shown in Fig. 8.67 is a high-
diodes D3 and D4 in conjunction with resistor impedance voltmeter using a MOSFET op-amp
8.16 Applications 323
coefficient and operates from 400 μA to 10 mA along with R9 ¼ 1 k protects the meter by limiting
with a low dynamic resistance. the voltage across the meter during over-range
conditions.
Linear Scale Ohmmeter
The circuit in Fig. 8.69 is that of an ohmmeter that Ideas for Exploration (i) The CA3140 op-amp
has a linear scale. In this circuit, a stable reference is a direct replacement for the 741 and can operate
voltage from the 5.6 V Zener D1 (with current on voltages down to 4 V either single or dual
supplied by R1 ¼ 2.2 k) is developed via the supply. Use one of these op-amps to replace the
transistor across resistor R2 ¼ 1 k. This voltage 741 and use another to design a split supply to
which is approximately 5 V drives the inverting enable operation from a single 9 V battery.
amplifier through range resistors R3 ¼ 1 k,
R4 ¼ 10 k, R5 ¼ 100 k and R6 ¼ 1 M. These AC Voltmeter
along with the feedback resistor Rx which is the The circuit shown in Fig. 8.70 is that of a simple
resistor to be measured set the gain of the ampli- AC voltmeter. The meter is placed in a diode
fier. The output of the op-amp drives a 1mA meter bridge D1 to D4 to allow the measurement of
in series with R7 ¼ 2.7 k and VR1 ¼ 5 k. The AC signals. Diode D5 protects the meter from
output of the op-amp is given by V o ¼ 5 RRRx V over-current, while capacitor C2 ¼ 10 μF
where RR represents one of the range resistors. It provides filtering to remove ripples from the cur-
follows that the output voltage is proportional to rent through the meter. Offset nulling may not be
Rx the resistance to be measured. In order to necessary in this circuit because of the presence
calibrate the instrument, resistor Rx is set equal of the diodes. Resistors Rm1 to Rm4 enable varia-
to the value of the selected range, for example, tion of voltage ranges. Thus using Im ¼ Vi/Rm, for
Rx ¼ 10 k on the 10 k range (R4 ¼ 10 k selected). Vi ¼ 0.1 V and Im ¼ 1 mA, then Rm1 ¼ Vi/Im ¼ 0.1
Then the op-amp will be set for unity gain and V/1 mA ¼ 100 Ω. Similarly, Rm2 ¼ 1 k for Vi ¼ 1
will therefore output approximately 5 V. VR1 is V, Rm3 ¼ 10 k for Vi ¼ 10 V and Rm4 ¼ 100 k for
adjusted for full-scale deflection on the 1 mA Vi ¼ 100 V. Resistor RB ¼ 100 k provides bias to
meter. If high precision resistors (2% tolerance) the non-inverting input of the op-amp, and capac-
are used for the range resistors, then all the ranges itor C1 ¼ 1 μF couples the AC signal to the input.
will be calibrated following this procedure on one
range. Resistor R8 ¼ 1 k reduces offset voltage of Ideas for Exploration (i) Convert this circuit to
the op-amp. Zener diode D2 with voltage 5.6 V an AC/DC instrument by removing C1 and
8.16 Applications 325
Ideas for Exploration (i) Refer to the paper “On Fig. 8.79 Scratch filter using op-amp
RIAA Equalization Networks” by Stanley
Lipshitz, Journal of the Audio Engineering Soci-
ety, Volume 27 Issue 6 pp. 458–481, June 1979.
(ii) Use a different CR network to achieve the
RIAA equalization.
Research Project 4
A common problem when playing old records is
scratches on the vinyl surface which generate
high-frequency noise. This project involves the Fig. 8.80 Rumble filter using op-amp
design of a scratch filter to remove this noise as
was discussed in Chap. 7, but here using a Research Project 5
741 op-amp. The system is shown in Fig. 8.79. Using the voltage follower op-amp configuration,
It comprises two connected RC networks with explore the development of a rumble filter that
one capacitor grounded and the other connected removes unwanted low-frequency (less than
to the output of the buffer. The transfer function 50 Hz) components from an audio signal as
for this system with R1 ¼ R2 ¼ R and discussed in Chap. 7. The two filter network
C1 ¼ C, C2 ¼ 2C is given by resistors and the two network capacitors in
Vo 1 Fig. 8.79 must be interchanged. Resistor R1
¼ pffiffiffiffi 2 ð8:124Þ which is part of the filter network also provides
Vi
1 þ 2j f
þ j f
fc fc bias to the non-inverting input of the op-amp. The
op-amp filter is shown in Fig. 8.80 and the
where associated transfer function with C1 ¼ C2 ¼ C
1 and R1 ¼ R, R2 ¼ R/2 is given by
f c ¼ pffiffiffi ð8:125Þ
2 2πRC 2
f
Vo j fc
¼ pffiffiffi f f 2 ð8:126Þ
Vi
This is a second-order response referred to as a 1 þ 2j f þ j f
c c
Butterworth response that has the characteristic of
being maximally flat. With an appropriately where the corner frequency is
placed corner frequency fc, unwanted high- pffiffiffi
2
frequency signals produced by vinyl scratches fc ¼ ð8:127Þ
can be eliminated. For an upper cut-off frequency 2πRC
of 10 kHz, using C ¼ 2.2 nF, Eq. (8.125) gives
R ¼ 4.7 k. This is a Butterworth response for the high-
pass filter.
Ideas for Exploration (i) Minimize DC offset
by introducing a resistor in the feedback connec- Ideas for Exploration (i) Minimize DC offset
tion between the output of the op-amp and the by introducing a resistor in the feedback
inverting terminal.
330 8 Operational Amplifiers
connection between the output of the op-amp and 6. A non-inverting amplifier has R1 ¼ 5 k and
the inverting terminal. R2 ¼ 25 k. Determine the gain of the system.
7. Configure an op-amp circuit to realize the
operation VO ¼ 5 V1 2 V2. If a third
input was added to the differential amplifier
Problems
such that VO ¼ 5 V1 ‐ 2 V2 + 3 V3, realize
such a circuit using a single op-amp.
1. For each of the circuits shown in Fig. 8.81,
8. Design a non-inverting amplifier with a gain
determine Vo.
of 12 using an op-amp that operates from a
2. Design an inverting amplifier with a voltage
single-ended 15 V supply.
gain of 5.
9. Design an inverting amplifier with a gain of
3. Design an inverting amplifier with a voltage
9 using an op-amp that operates from a
gain of 8 and an input resistance of 10 k.
single-ended 18 V supply.
4. Design a non-inverting amplifier with a volt-
10. Show that the closed-loop bandwidth of a
age gain of 8.
unity-gain stable op-amp is equal to the
5. Design a non-inverting amplifier with a volt-
open-loop bandwidth multiplied by the loop
age gain of 8 and an input resistance of
gain. Such an operational amplifier has a
150 kΩ.
Problems 331
cycle. This same arrangement may be used as a transistor is always on as movement up and down
power amplifier by replacing the small signal the load line occurs in response to an input signal.
transistor by a power transistor as shown in The signal power in RL, PL is given by
Fig. 9.1.
A signal at the input of the amplifier causes the ZT
1
collector current of the transistor to change, and PL ðAC Þ ¼ iC 2 RL dt ð9:1Þ
T
hence an output voltage is developed across RL 0
the load resistor. Class A operation follows from
The peak value of the maximum collector AC
the fact that a DC bias is established in the tran-
current is given by
sistor with the device operating along the DC
“load line” as shown in Fig. 9.2. The Q point V cc
iCpk ð max Þ ¼ ð9:2Þ
represents the DC voltage and current in the tran- 2RL
sistor when no signal is present. As is evident, the
and occurs when the transistor is symmetrically
biased. Under these conditions, the maximum AC
power to the load becomes
iCpk ð max Þ 2
PL ðAC Þmax ¼ pffiffiffi RL
2
2
V V 2
¼ pffifficcffi RL ¼ cc ð9:3Þ
2 2R L 8RL
Ic (mA)
VCC/RL
Q-point
ICQ
VCEQ VCE(V)
VCC
imum symmetrical swing, the load resistance speaker is replaced by a 15-ohm speaker, the
would need to be 6 V/100 mA ¼ 60 Ω. It is power out increases from 75 mW to 141 mW.
possible to achieve this with the 8-ohm speaker
by using a transformer, and this is discussed in This basic power amplifier system is subject to
the next section. For the current arrangement, the an effect referred to as thermal runaway. As the
maximum symmetrical change ΔIC in the collec- system operates, the transistor heats up and the
tor current is ΔICpk ¼ 100 mA, i.e. a minimum IC collector current increases. This causes increased
of zero and a maximum IC of 200 mA. While this heat dissipation in the transistor which causes
causes only a small change in the collector volt- further increase in collector current. This self-
age of the transistor, power is delivered to the reinforcing cycle may eventually result in the
failure of the transistor. Various strategies are
speaker and is given by PL ðACÞ ¼ available to prevent this phenomenon from occur-
2 2
ΔI Cpk ring. A simple method is to use collector-base
pffiffi RL ¼ 0:1ffiffi
p 8 ¼ 40 mW . There is
2 2
feedback biasing which requires that the bias
however significant power dissipation in the resistor RB be returned to the collector of the
transistor which may have to be mounted on a transistor instead of the power supply. The result
heatsink in order to prevent overheating. This is that any increase in collector current with tem-
approach was explored in a research project in perature will cause an increased voltage drop
Chap. 5, and we can now outline the design across the speaker which in turn results in a
process. reduction in the collector voltage of the transistor.
The result is that the base current flowing through
Example 9.1 RB is reduced thereby reducing the collector cur-
Using the configuration in Fig. 9.3 and a transis- rent. Higher values of speaker resistance enhance
tor with a current gain of 50, design a low-power this collector current stabilizing action.
336 9 Power Amplifiers
Further improvement of this circuit can be where PAC is the output power delivered to the
realized by replacing the power transistor with a load and PDC is the input DC power from the
power Darlington or power MOSFET (which supply. In general, not all of PDC is converted to
does not suffer from thermal runaway) that results PAC, i.e. η < 100%, and the component of the
in a higher input impedance and the addition of a input power that is not converted to AC output
preamplifier stage to increase the sensitivity of the power is dissipated as heat in the active output
overall system. This enhanced arrangement using devices.
a power Darlington is shown in Fig. 9.4.
Because of the low resistance of the speaker, Example 9.3
the voltage drop across it is low (0.15 8 ¼ A power amplifier has an efficiency of 20% and
1.2 V), and therefore most of the supply voltage delivers an output power of 25 W to a resistive
would be dropped across the transistor. This does load. Calculate the power dissipated and the DC
not allow for maximum (symmetrical) swing of input power to the amplifier.
the collector voltage. This can be corrected by the
use of a step-down transformer in the collector Solution η ¼ PPDC
AC
100 . ∴PDC ¼ PηAC 100 ¼
circuit such that a greater impedance is reflected 2500
¼ 125 W . Hence power dissipated Pdis is
20
in the collector circuit of the output transistor.
given by Pdis ¼ 125 W – 25 W ¼ 100 W.
This will be discussed under transformer-coupled
amplifiers.
Example 9.4
A large signal amplifier delivers an output power
of 5 W and consumes 20 W from the associated
9.2.1 Efficiency Calculations DC supply. Calculate its efficiency.
Power amplifiers act as power converters as they
Solution
deliver large amounts of power to a load. They are
driven by an input signal and convert power
PAC 5
available from a power supply (input or DC η¼ 100 ¼ 100 ¼ 25%:
PDC 20
power) to useful signal output power in the load
(AC power). The efficiency of the amplifier η is a
measure of the effectiveness of this conversion Power considerations are extremely important
and is given by in large-signal amplifiers because of the large
voltages and currents involved. For example, a
PAC
η¼ 100% ð9:4Þ power transistor will be destroyed if its power
PDC
dissipation capacity is exceeded. In small-signal
amplifiers, it was noted that the Q-point must lie
9.3 Transformer-Coupled Class A Amplifier 337
where V(t) is the voltage across the device, i(t) is This means that the maximum efficiency of the
the current through it, and T is the period of the class A common emitter amplifier is 25%.
time-varying portion of V or i. For the fixed-bias
Class A amplifier, the average power PDC deliv-
ered by the power supply, neglecting the small 9.3 Transformer-Coupled Class A
power associated with the bias, is given by Amplifier
Z
1 T
PDC ¼ V i dt In order to increase the efficiency, the DC power
T o cc C
Z T dissipated in the load RL may be reduced by
1 ð9:6Þ
¼ V ðI þ ic ðt ÞÞdt coupling the load to the transistor using a step-
T o cc c down transformer as shown in Fig. 9.5.
¼ V cc I c Assuming the resistance of the transformer
primary is zero, then the DC load line is vertical,
where ic is the signal component and Ic the DC passing through Vcc on the VCE axis as shown in
component of iC. The power dissipated in the Fig. 10.4. Because of the transformer, the load RL
transistor Pc is calculated using seen by the transistor collector is given
PC ¼ PDC PAC RL ¼ a2 RL ð9:11Þ
Z
1 T2
¼ PDC i R dt where a is the turns ratio of the transformer. With
T o C L
Z ð9:7Þ this arrangement, a larger output voltage swing
1 T
¼ PDC ðI þ ic Þ2 RL dt for the same collector current change can be
T o c
Z
1 T2
¼ PDC I 2c RL i R dt
T o c L
realized as compared with the case where RL was into 8 ohms and operating from a 20-volt supply.
placed in the collector circuit. In fact, the AC load Use a power Darlington with β ¼ 1000.
line intercepts VCE axis at a value greater than Vcc.
The value of RL determines the slope of the AC Solution The power into the load is given by
2
load line. If RL is chosen such that VCEpk ¼ 2VCC
PL ¼ ppkffiffi2 =RL where Vpk is the peak voltage
V
and ICpk ¼ 2ICq, then maximum symmetrical
swing is possible. Then, input power is given by across the speaker from the output of the trans-
VCCICq and output power (AC) is given by Vpccffiffi2 former secondary. Thus, for 1 W output into
2
8 ohms, 1 ¼ ppkffiffi2 =8 which yields Vpk ¼ 4 V
V
pCqffiffi. Hence
I
2
and Ipk ¼ 4 V/8 ¼ 0.5 A. Since the power supply
V cc I Cq
ηmax ¼ 100 ¼ 50% ð9:12Þ is 20 volts, the peak change in collector voltage is
2V cc I Cq
ΔVCEpk ¼ 20 V. This is the input signal voltage to
The output voltage in this arrangement is able the transformer primary in the collector. Hence
to exceed the supply voltage because of the the transformer turns ratio a to realize this is
inductive effect of the transformer primary as a ¼ ΔVCEpk/Vpk ¼ 20/4 ¼ 5. (Alternatively, the
shown in the AC load line of Fig. 9.6. Thus, the power supply voltage may be varied to accommo-
maximum efficiency of a transformer-coupled date a different transformer turns ratio.) The peak
amplifier is 50%, which is twice as good as the signal current into the transformer primary is
efficiency of the class A amplifier with the load in ΔICpk ¼ Ipk/a ¼ 0.5 A/5 ¼ 100 mA. Therefore
the collector where there was DC power dissipa- ICq ¼ ΔICpk ¼ 100 mA. Hence, noting that the
tion in the load. Again, for the transformer- Darlington transistor has a base-emitter voltage of
coupled case, the maximum transistor dissipation 1.4 V, the fixed-bias resistor is calculated from
occurs when there is no signal present and is 20 ¼ RBIC/β + 1.4 ¼ RB 0.1 A/30 + 1.4 giving
given by RB ¼ 186 k. Note that the power supplied to the
system is VCCICq ¼ 20 V 0.1 A ¼ 2 W which
V cc 2 is double the output power of 1 W corresponding
PC ¼ V CEq I Cq ¼ ð9:13Þ
RL to 50% efficiency.
since VCEq ¼ VCC and I Cq ¼ V CC =RL . A preamplifier can also be used with this sys-
tem to increase sensitivity as shown in Fig. 9.7. In
Example 9.5 this configuration, resistor R2 ensures bias stabil-
Using the configuration shown in Fig. 9.5, design ity in the output stage and R3 provides feedback
a transformer-coupled amplifier delivering 1 W bias to Tr1.
9.4 Class B Push-Pull Amplifier 339
Fig. 9.8 Complimentary symmetry emitter follower similar action takes place on the negative half-
cycle with Tr2 already on. The overall effect is
that the crossover distortion is eliminated as a
smooth transition from one transistor to the
6
other occurs.
Input A simple method of forward biasing Tr1 and
3
Voltage /V
PD ¼ Pcc PL
2 1 ð9:20Þ
¼ V cc I c max RL ðI c max Þ2
π 2
In order to determine the value of Icmax for
maximum dissipation, PD is differentiated with
respect to Icmax and the result set to zero giving
i.e.
2 V cc
I c max ¼ ð9:21Þ
π RL
100 ¼ 80 μA. This gives the base current in R1 is the resistance being driven by the transistor
as 80 μA and hence R1 ¼ ð9=20:7 Þ
80 μA ¼ 47:5 k .
emitter which is 8 ohms. For a current in
the output of 20 mA say, then ρ ¼ 1þ40208
40208
¼
Because the value of R1 depends so directly on
the current gain of Tr1, its value may have to be 0:86 , R2b ! R'2b ¼ R2b/(1 0.86) ¼ 7.4R2b.
adjusted in order to set the amplifier quiescent Using R2b ¼ 240 Ω then R'2b ¼ 7.4 240 ¼ 1.8 k.
output voltage at 4.5 volts for maximum symmet- Note that resistor R2a is part of the load of the
rical output swing. Capacitor C2 must be suffi- amplifier output and is in parallel with the
ciently large in order to achieve a satisfactory speaker.
344 9 Power Amplifiers
Example 9.8
Design a 10 W power amplifier using the config-
uration in Fig. 9.14 along with the variable Zener
for biasing.
through D3 and D4, R5 ¼ (35 1.4)/3 mA ¼ 11.2 Tr6 is introduced in the circuit. This stage
k. For a 20 Hz low-frequency response, C2 is provides further gain and allows the application
given by C2 ¼ 1/2π 20 8 ¼ 995 μF. A of voltage-series feedback at the emitter of Tr6 via
value of 1000 μF is therefore chosen. The input R2, R1 and C1. The Zener regulator D3, D4
impedance at the base of Tr1 is approximately hie provides regulated bias for the base of Tr6
for Tr1 which is hie ¼ hfe/40IC ¼ 100/(40 5 through potential divider R9 and R10. This config-
mA) ¼ 500 Ω. The feedback reduces it further. uration requires a compensation capacitor, C4
Therefore the minimum value of the input capac- whose value is given by C 4 ¼ 2π gf mAV where
c
itor C1 is given by C1 ¼ 1/2π 20 500 ¼ 15.9 μF. gm ¼ 1/(re(Tr6) + R1).
A value of at least 20 μF is chosen. The presence
of R1 further reduces this value. Darlingtons Tr2 Example 9.9
and Tr3 must be power transistors having a col- Design a 25 W power amplifier using the config-
lector current rating of at least 1.6 A the peak load uration in Fig. 9.16.
current, a collector emitter voltage BVCEO of at
least 35 V and power rating of 1/5 the amplifier Solution For maximum symmetrical output, the
output power. Complimentary Darlington quiescent voltage at the output of the power ampli-
transistors TIP 121 and TIP126 (80 V, 5 A, fier is VCC/2. For P ¼ 25 W into 8 Ω, using
65 W) are suitable. Transistors Tr1 and Tr4 can pffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
V pk ¼ 2PRL , V pk ¼ 2 25 8 ¼ 20 V . In
be small signal transistors but must have a break- order to allow for transistor saturation losses, the
down voltage of greater than 35 volts. MPSA05 peak voltage swing is set at 20 + 5 ¼ 25 V. Hence
and MPSA55 are adequate. The VBE multiplier VCC ¼ 2 25 ¼ 50 V. The peak output current is
must provide at least 4 VBE drops across Tr2 and given by Ipk ¼ 20/8 ¼ 2.5 A. Using a current gain
Tr3. Let IRa ¼ 1 mA. Then Ra(Tr5) ¼ 0.7 V/ of 1000 for each Darlington pair, the maximum
1 mA ¼ 700 Ω. output current results in maximum base currents
for the Darlington of IB(Tr2, 3) given by Ib(Tr2,
Hence 1 þ RRba ¼ 2:4 0:7 ¼ 3:4, Rb ¼ 2:4, Rb ¼
Ra
3) ¼ 2.5 A/1000 ¼ 2.5 mA. This current flows
2:4 k. A 5 k potentiometer can be used for Rb to into the collector of Tr1 during negative-going
enable adjustment of the quiescent current in Tr2 half-cycles or is supplied from the constant current
and Tr3. A 2 N3904 transistor can be used. Resis- flowing out of the collector of Tr4 during positive-
tor R1 at the input sets the overall voltage gain of going half-cycles. However, in order to ensure
the power amplifier. For a voltage gain of that Tr4 can always supply the necessary current,
20, R1 R2/10 ¼ 16.8 k (see Chap. 8). Because the constant current is set at 5 mA which is twice
of the moderate open-loop gain of the circuit, R1 the requirement for the base current to the
may have to be reduced somewhat in order to Darlington pairs. Resistors R3 and R4 are chosen
achieve the desired gain of 10. Under test, the to maximize bias current stability in the output
circuit performed as expected delivering 15 volts stage while minimizing power dissipation from
peak into an 8-ohm load which is equivalent to signal current. Since higher output currents are
14 W. Resistor R2 was adjusted to 496 k in order involved, the power dissipation is a greater prob-
to set the DC output voltage at 17.5 V. The 5 k lem than before. These resistors would therefore
potentiometer was adjusted to set the bias current have to be smaller than in the low-power amplifier
in the output stage at about 14 mA. Finally the and a value between 0.2 Ω and 0.3 Ω for each of
circuit gain was set at 20 by adjusting these resistors is reasonable. The quiescent current
R1 ¼ 16 k. The frequency response was measured in Tr1 is equal to the current in the current source
at 100 Hz to 20 kHz. which is 5 mA. Hence, taking the current gain of
An improved non-inverting version of the Tr1 to be 100, the base current Ib(Tr1) ¼ 5 mA/
power amplifier in Fig. 9.14 is shown Fig. 9.16. 100 ¼ 50 mA. Since this current is supplied by
An additional common emitter amplifying stage Tr6 the quiescent current in this transistor is set at
9.6 Medium-Power Amplifier Design 347
say 0.5 mA. Hence R7 ¼ 0.7/0.5 mA ¼ 1.4 k. for D3 and D4, the Zener regulator would pro-
Resistor R6 is calculated from R6 ¼ 0.7/ vide 36 V. Hence R9 ¼ (36 22)/0.25 mA ¼ 56
5 mA ¼ 140 Ω. To allow about 3 mA through k. Allowing 2 mA into D3 and D4 for proper
diodes D1 and D2, R5 ¼ (50 1.4)/3 mA ¼ 16.2 k. operation, then R8 ¼ (50 36)/2 mA ¼ 7 k.
For a 20 Hz low-frequency response into the The input impedance of the amplifier is less than
8 Ω load, C2 is given by C2 ¼ 1/2π 20 8 ¼ 995 R9//R10 which is 56 k//88 k ¼ 32 k. Therefore
μF. A value of 1000 μF is therefore chosen. The the value of the input capacitor C3 can be found
VBE multiplier must provide at least 4VBE drops using C3 ¼ 1/2π 20 32 k ¼ 0.25 μF. A
across Tr2 and Tr3. Let IRa ¼ 1 mA. Then value of 1 μF is then chosen since the input
Ra(Tr5) ¼ 0.7V/1 mA ¼ 700 Ω. Hence 1 þ RRba ¼ impedance at the base of Tr6 while high because
2:4
¼ 3:4 RRab ¼ 2:4, Rb ¼ 1:7 k. A 5 k potentiometer of voltage-series feedback introduced by R2 and
0:7
R1 reduces the overall input impedance below
for Rb can be used to allow adjustment of the
32 k. The gain of the amplifier is given by
quiescent current in the output stage. Resistor R2
1 + R2/R1. Since R2 ¼ 4.7 k then for a gain of
along with resistor R1 provides feedback to the
20, R1 ¼ 247 Ω. A small filter capacitor C6 of
emitter of the input transistor Tr6. Let R2 ¼ 4.7 k
about 0.1 μF can be used to reduce the effect of
say. Then the voltage at the emitter of Tr6 is
long power supply leads, while C5 ¼ 1 μF
VE(Tr6) ¼ 25 0.5 mA(4.7 k) ¼ 22.7 V. This
improves the regulation of the Zener diodes.
sets the voltage at the base of Tr6 at
For an adequate low-frequency response, capac-
VB(Tr6) ¼ 22.7 0.7 ¼ 22 V. Using a current of
itor C1 must be chosen such that its reactance is
0.25 mA through resistors R9 and R10 to ensure
equal to R1 at 20 Hz. Thus C1 ¼ 1/
bias voltage stability for Tr6, then R10 ¼ 22/
2 π 20 247 ¼ 32 μF. Capacitor C4
0.25 mA ¼ 88 k. Using two 18 V Zener diodes
introduces Miller compensation for circuit
348 9 Power Amplifiers
stability. For Tr6, IC ¼ 0.5 mA and hence overall harmonic distortion performance of the
re(Tr6) ¼ 1/40IC ¼ 50 Ω. Therefore using amplifier particularly at high frequencies.
fc ¼ 500 kHz, AV ¼ 20 and gm ¼ 1/
(re(Tr6) + R1) ¼ 1/(50 + 247) ¼ 1/297 in equa-
tion (9.37), the values for capacitor C4 is given
9.7 High-Power Amplifier Design
by C 4 ¼ 2π gf m AV ¼ 2972π510
1
5
20
¼ 54 pF:
C
Darlingtons Tr2 and Tr3 must be power In the power amplifiers considered thus far, nei-
transistors having a collector current rating of ther the input nor the output stages permit direct
at least 2.5 A the peak load current, a collector coupling. An enhanced power amplifier circuit
emitter voltage BVCEO of at least 50 V and that overcomes these problems is shown in
power rating of 1/5 the amplifier output power. Fig. 9.17. It comprises three amplifying stages:
Transistors TIP142 and TIP147 (100 V, 10 A, the input stage made up of the differential ampli-
125 W) can be used. Transistors Tr1 and Tr4 can fier Tr1 and Tr2, the intermediate common emitter
be small-signal transistors but must have BVCEO stage Tr3 with current source load Tr4 and the
of greater than 50 V. MPSA06 and MPSA56 complimentary symmetry emitter follower output
(80 V, 500 mA, 1.5 W) are adequate. These stage using Darlington pairs Tr5 and Tr6. In the
transistors can also be used for Tr5 and Tr6. A differential amplifier input stage, two pnp
simple but useful modification of this circuit is transistors Tr1 and Tr2 are connected at their
the use of a Cascode amplifier in the second emitters with resistor R2 supplying current to
stage. This arrangement removes distortion these emitters from a Zener-regulated supply fed
caused by early effect and thereby enhances the from the supply rail. Each transistor is basically in
a common emitter mode with resistor R3 in the
during its operation. For good transistor opera- voltage across Tr7 of 0.7 mA (1 k + 1 k + 5
tion, let the quiescent current in each transistor of k) ¼ 4.9 volts. This is more than the 2.8 volts
the differential pair be 1 mA. The total current required to turn on the Darlington pairs. The
through R2 is therefore 2 mA. This current is minimum voltage across Tr7 (corresponding to
provided by the Zener diode D1 which is selected VR1 ¼ 0) is 0.7 mA (3.3 k) ¼ 2.3 V. There-
to be about half of the supply voltage. A Zener fore, the potentiometer can be used to set the bias
with voltage of 24 V is available. The base of Tr1 current in the output stage to a value that
is at an approximately zero potential and hence R2 minimizes cross-over distortion. This value is
is given by R2 ¼ (24 0.7)/2 mA ¼ 11.7 kΩ. likely to lie in the range 10 mA to 100 mA. The
Resistor R4 supplies current to the Zener of about capacitor C2 influences the low-frequency perfor-
5 mA as well as the 2 mA to the differential pair. mance of the amplifier. A value of 100 μF sets
Hence R4 ¼ (50 24)/7 mA ¼ 3.7 kΩ. The base- a pole at fp ¼ 1/(2π100 106 520) ¼ 3 Hz.
emitter voltage of Tr3 appears across R3 and The capacitor C1 also influences low-frequency
therefore R3 ¼ 0.7/1 mA ¼ 700 Ω. Resistor R1 roll-off. Using the input resistance as 10 k (since
supplies base current to Tr1. It should be low in feedback will increase the input impedance into
order to minimize the voltage drop arising from Tr1), a value of 10 μF for this capacitor will result
the base current flow while high enough to pro- in a pole at fp ¼ 1/2π10 106 10000 ¼ 1.59 Hz.
vide a high input impedance to the circuit. A Capacitor C3 is determined using equation (9.37).
value of about 10 k or higher is a reasonable The choice of frequency fc is critical. It must be
compromise. For a peak output voltage of 40 V, chosen high enough such that adequate feedback
the peak output current is 40/8 ¼ 5 A. Using is available at all frequencies of interest. How-
Darlington current gain of 1000, the maximum ever, setting this frequency too high will increase
peak base current of each Darlington is the risk of instability arising from increased phase
5/1000 ¼ 5 mA. Since this current is sourced shift around the feedback loop. For audio
from the collector current of Tr3, we therefore amplifiers, fc is likely to be in the range 200 kHz
choose a quiescent current of 10 mA for this to 2 MHz using fc ¼ 500 kHz and gm ¼ 1/2re ¼ 1/
transistor. The collector current of Tr4 must also 50, where re ¼ 1/40IC, we get C3 ¼ 50 1
be set at 10 mA. This is accomplished by noting 1
¼ 318 pF. The final value of C3 can
2π5105 20
that the voltage across diode D2 is dropped across
be experimentally determined by adjusting for
R5 through which the 10 mA also flows giving
minimum square wave overshoot.
R5 ¼ 0.7/10 mA ¼ 70 Ω. Resistor R6 is chosen
such that about 5 mA flows through diodes D2
Suggested transistors are as follows: Power
and D\3 to turn them fully on. Hence
Darlingtons MJ11021/MJ11022 (250 V, 15 A,
R6 ¼ (50 1.4)/5 mA ¼ 9.7 k. In order to
175 W); Tr3/Tr4 MJE340/MJE350 (300 V,
minimize DC offset at the output, feedback resis-
500 mA, 20 W); Tr1/Tr2 2 N5400 pnp (140 V,
tor R12 is chosen to be the same value 10 k as the
600 mA, 625 mW).
input resistor R1. Resistor R11 then sets the
This amplifier can be improved in several
closed-loop gain. A value of 520 Ω gives a
ways. The first improvement is to replace the
closed-loop gain of 20. For the VBE multiplier,
resistor R2 supplying current to the differential
the current in this stage is the 10 mA collector
pair by a constant current source as shown in
current of Tr3. Because of the base-emitter volt-
Fig. 9.18. The effect of this is to improve the
age of Tr7 across R7, a value of 1 k results in a
ripple rejection capability of the amplifier so that
current through this resistor of 0.7/1 k ¼ 0.7 mA.
power supply changes do not get into the ampli-
This is approximately seven times larger than the
fier via the emitters of Tr1 and Tr2. A second
base current of Tr7 which is 10 mA/
improvement is the inclusion of small resistors
100 ¼ 0.1 mA. If we set R8 ¼ 1 k, then a 5 k
Re in the emitters of Tr1 and Tr2. These resistors
potentiometer for VR1 will produce a maximum
reduce the distortion produced in this stage,
352 9 Power Amplifiers
Example 9.11
The amplifier circuit shown in Fig. 9.20 uses 9.8 High-Power MOSFET Amplifier
feedback pairs in the output stage and can deliver
60 W into 8 ohms. Analyse the circuit to deter- A high-quality power amplifier may be designed
mine the quiescent currents in the input and inter- using power MOSFETS instead of power
mediate stages as well as the closed-loop gain. transistors in the output stage. Power MOSFETS
Investigate the purpose of the 100 μF capacitor became available in the 1970s and enable the
connected between the output and the junction of exploitation of the high input impedance of the
the two 2.7 k resistors. field effect transistor while being able to handle
significant power levels. They can be used to
Solution The voltage at the emitters of the input replace the Darlington Pairs in Fig. 9.17 in the
transistors is approximately +0.7 V. Noting that output stage. These devices respond to very high
9.8 High-Power MOSFET Amplifier 353
pffiffi 2
frequencies and are more easily biased. Another ðV opk = 2Þ
¼ 150 where Rspk ¼ 8 Ω. This gives
advantage of these devices is their extremely high Rspk
input impedance completely isolates the second Vopk ¼ 49 V. The corresponding peak output
stage transistor collector from the effects of the current is 49/8 ¼ 6.13 A. Resistors R20 to R23
amplifier load. A completely symmetrical ampli- provide output stage quiescent current stability.
fier design using power MOSFETS is shown in As before, since they are in the output current
Fig. 9.21. Note that four MOSFETS are used, two path they cannot be large. Values of 0.22 Ω will
n-channel and two p-channel devices each pair be used and, because the current is approximately
sharing the load current on alternatively positive shared, will result in a peak voltage drop of less
and negative signal cycles. Along with power than 1 V. On the peak positive half-cycle with
MOSFETS at the output, this circuit uses a com- maximum current flowing into the load, the volt-
plimentary symmetry differential amplifier at the age drop in R20 (or R21) added to the gate-source
input and complimentary Cascode intermediate FET voltage of 2 V gives the voltage at the col-
voltage gain stages. The result is a fully symmet- lector of Tr7 as 49 + 1 + 2 ¼ 52 V. If we allow
rical and very linear circuit that produces low about 5 V across the active load in order that it is
levels of harmonic distortion. The dominant pole fully operational, then the positive supply voltage
is set by the capacitance at the gate-drain needs to be about +60 V. Similarly, we choose a
junctions of the MOSFETS with all other poles negative supply voltage of 60 V. The FET gate
at high frequencies. resistors R16 R19 are necessary for MOSFET
stability and each needs to be about 220 Ω.
Example 9.12 Diodes D5, D7, D6 and D8 protect the gate source
Design a power amplifier that delivers 150 W junctions from over-voltage with Zeners D5 and
average power into an 8-ohm load using the con- D6 limiting the voltage to 12 V. Potentiometer
figuration in Fig. 9.21. VR1 is adjusted in order to set the FET bias cur-
rent. Note that a Vbe multiplier is not necessary.
Solution In order to produce 150 W into 8 ohms, Suitable power MOSFETS are 2SK1058/2SJ162
the peak output voltage Vopk is given by (160 V, 7 A, 100 W).
354 9 Power Amplifiers
+Vcc
R12 C7
R6 R8
Tr8 D3
R13
C5 Tr7
D1 R16 Tr10
R9 Tr9
D5
Tr1 Tr2 R17 R21
C1 D7 R20 L1
R25
R2 R3
R4 R5
R24 D8 R22 R26 R27
R1
Tr3 Tr4 vR1 R23
D6 R18 Tr11 C4
R10 C2
Tr12
C6 R14 R19
D2 C3 Tr6
Tr5 D4
R7 R11
R15 C8
-Vcc
At the input, the complimentary symmetry transistors to be 0.5 mA, then using diodes D1
differential amplifier has a reduced current and D2 at 24 V, it follows that
flowing in the bias resistor R1 since Tr1 and Tr3 R9 ¼ R10 ¼ (24 0.7)/1 mA ¼ 23.3 k. R8 and
approximately provide each other with the neces- R11 supply current to the Zener diodes. This cur-
sary base current. This resistor can therefore rent needs to be about 5 mA. Hence
assume a higher value while still keeping the R8 ¼ R11 ¼ (60 24)/(5 + 1) ¼ 6 k. The Cascode
transistor bases at an approximately zero poten- amplifiers Tr5–Tr6 and Tr8–Tr7 using MJE340/
tial. Suitable transistors are 2 N5551/2 N5401 MJE350 high-voltage transistors for Tr6 and Tr7
(160 V, 600 mA). We let R1 ¼ R25 for minimum must operate with enough current to charge the
DC offset and set the value at 33 k. Hence with input (gate-drain) capacitance of the MOSFETS
R24 ¼ 1 k, the overall closed-loop gain is 34. This which varies with drain-source voltage. Using
gives a sensitivity of about 1 V rms for full out- gate capacitance of 500 pF and a current of
put into 8 ohms. The input transistors all have 15 mA, the slew rate will be SR ¼ 215 500 pF ¼
mA
emitter degeneration resistors R2 to R5. These 60 V=μs. This is sufficient to avoid slewing dis-
resistors balance the current in Tr1–Tr2 and Tr3– tortion for peak output signals at 20 kHz which is
Tr4, improve the frequency response and reduce SR ¼ 2π 49 20, 000 ¼ 6.2 V/μs. Resistors
distortion. However, the gain of this stage is R12 and R15 introduce local feedback in the
reduced. Values of 100 ohms represent a reason- Cascode amplifiers thereby improving linearity.
able compromise. Diodes D1 and D2 provide A value for these resistors of 82 ohms results in a
regulated supplies for the currents into Tr1–Tr2 voltage drop across it of 15 mA 82 ¼ 1.23 V.
and Tr3–Tr4. If we set the currents in these Therefore, the voltage between the base of Tr5
9.9 IC Power Amplifiers 355
connected in the inverting configuration (signal intended primarily for use in portable devices.
applied to the inverting input with non-inverting Fig. 9.24 shows it in a typical application.
input grounded). The load is connected across the
IC outputs which are out of phase. Thus, while the Medium-Power IC Amplifier: LM3886
output of one IC swings positive towards the For higher output power levels than can be deliv-
supply voltage, the output of the other swings ered by the LM386, the LM3886 shown in
negative towards ground. Fig. 9.25 is available. This is a high performance
audio power amplifier that can deliver 56 W con-
USB-Powered Amplifier: LM4871 tinuous power into an 8 ohm load with low har-
The LM4871 uses the bridge configuration in monic distortion between 20 Hz and 20 kHz. It
Fig. 9.23 in its internal structure to deliver 3 W operates from 10 V to 42 V and can supply
into a 3-ohm load from a 5-volt supply such as is up to 4 A. The device is fully protected against
available from a computer USB port. It requires over-temperature, over-current and voltage
few external components not including output transients. It operates essentially as a power
coupling capacitor. It possesses thermal shut- op-amp, and therefore the gain is given by AV ¼
down protection and is unity gain stable. The 1 þ RR21 . The peak output voltage is about 5 V less
amplifier gain can be set externally and is
than the supply voltage.
the supply voltage. The IC is used to deliver drive set to 20 k. Resistor R1 is also set at 20 k. From
current (source and sink) to complimentary the data sheet, the shutdown resistor RM is calcu-
Darlington pairs Tr1 and Tr2 such as the lated using RM ¼ (Vref 2.9)/ISD. With Vref ¼ 6.8
MJ11021 and MJ11022 (250 V, 15 A, 175 W). V and ISD ¼ 1.5 mA, RM ¼ 2.6 k. Capacitors, C1
Transistor Tr3 is a VBE multiplier that sets the bias and C2, are selected to be 10 μF for adequate
current in Tr1 and Tr2. The resistors Rf and Ri set low-frequency response. The compensation
the closed-loop gain at AV ¼ 1 + Rf/Ri. For stabil- capacitor Cc is set at a nominal value of 30 pF
ity, AV must be greater than about 20. The capaci- with the final value determined experimentally.
tor CC provides compensation and must be The power supply pins to the IC must be
optimized experimentally. The IC driver has a decoupled with C3 ¼ C4 ¼ 0.1 μF capacitors. In
shutdown mode and is thermally protected. order to deliver this level of power to the speaker,
Power supply decoupling of the IC using C3 and the Darlington output stage needs to comprise a
C4 is essential. parallel arrangement of four sets of complimen-
tary Darlington pairs MJ11021/MJ11022 as
Example 9.14 shown in Fig. 9.27. Each pair will deliver a por-
Configure the LME49811 IC to produce 200 W tion of the output load current with all four pairs
into 8 ohms with a gain of 21. being driven by the LME49811. Adequate
heatsinking of these transistors as well as the
Solution In order to produce 200 W into 8 ohms, LME49811 is necessary for successful operation
the peak output voltage is given by V Opk ¼ of this high-power system.
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 8 200 ¼ 56:6 V. The peak output current
is then IOpk ¼ 56.6/8 ¼ 7.1 A. The specifications
for the LME49811 indicate that in order to deliver 9.10 Amplifier Accessories
an output voltage of 56.6 V, the supply voltage
must be 80 V. The components of the VBE A power amplifier is often used to drive inductive
multiplier are determined as done previously. and capacitive loads such as loud speakers and
For a gain of 21, Ri is set to 1 k and hence Rf is significant lengths of connecting cable. If the
9.10 Amplifier Accessories 359
amplifier is to remain stable under such that turns on transistors Tra and Trb. These in turn
conditions, appropriate networks need to be reduce the drive current to the base of the output
connected at the output of the power amplifier. transistors and as a result limit the current in these
The two networks normally used are a shunt devices.
Zobel network for maintaining stability into
inductive loads and a parallel connected inductor
and resistor in series with the output for preserv-
9.10.1 Heatsink Design
ing stability when driving capacitive loads.
A capacitive load connected to the output of a
In the class B amplifier, almost 25% of the input
power amplifier can result in the amplifier going
power from the power supply is converted into
unstable. This occurs since the output resistance
heat, most of which is dissipated in the output
of the amplifier in conjunction with the capacitor
stage. As a result, during operation power
form what is effectively a low-pass filter at the
amplifiers dissipate a large amount of heat in the
output of the amplifier. This network introduces a
power transistors. A characteristic of transistors is
pole into the transfer function of the amplifier
that as the temperature is increased, their power-
which causes phase shift that pushes the closed-
handling capability decreases. A graphical illus-
loop system towards instability. In order to pre-
tration of this is presented in Fig. 9.29. Here the
vent this, an inductor is placed between the ampli-
curve ABCD defines the safe operating area of the
fier output and the capacitive load such that the
transistor. Portion AB represents the maximum
capacitance is isolated from the amplifier. Its
current that the transistor can handle, portion
value is chosen such that the upper cut-off fre-
CD is set by the maximum collector-emitter volt-
quency created by the inductor working in con-
age of the transistor and portion BC is the hyper-
junction with the lowest load resistance presented
bola defined by VCEIC ¼ PD where PD is the
by the speaker load is above the highest fre-
transistor power rating at room temperature
quency to be amplified. Additionally, a small
(25oC).
resistor is placed in parallel with this inductor in
As the temperature of the case of the transistor
order to provide damping. This network is
and hence the collector junction temperature
referred to as a Thiele network. Typical values
increases, the power rating of the transistor
are 2 μH in parallel with a 1 Ω resistor of suitable
decreases. The result is that the hyperbola
wattage. (See L1 and R27 in Fig. 9.21)
represented by curve BC moves towards the
The voice coil of a loud speaker presents an
axes as shown thereby decreasing the safe
inductive load to a power amplifier that can excite
operating area of the transistor. The transistor is
instability at high frequencies as the inductive
then said to be de-rated. A curve showing the
reactance increases. This problem can be resolved
transistor de-rating as a function of case tempera-
by placing a Zobel network across the speaker.
ture is shown in Fig. 9.30. In order to prevent this
This network comprises a resistor in series with a
transistor de-rating, the dissipated heat must be
capacitor, and its effect is to cause the inductive
conducted away such that the transistor junction
speaker load to appear resistive across the fre-
temperature is kept low. To aid in heat conduction
quency of operation of the amplifier. The resistor
away from the transistor, the device must be
is usually in the range 4.7 Ω 10 Ω, while the
mounted on a piece of metal called a heatsink.
capacitor is almost always 0.1 μF. (See R26 and
This along with the metal case of the transistor
C4 in Fig. 9.21).
provides a large surface area from which the heat
It may be necessary to provide protection for
can be radiated. This results in a lower transistor
the output transistors in a power amplifier. This is
junction temperature as compared with the tran-
to prevent transistor burn-out due to current over-
sistor operating without the heatsink. The heat
load. A simple mechanism to achieve this is the
path is from the transistor junction to the case,
circuit shown in Fig. 9.28. This is incorporated in
from the case to the heatsink and finally from the
the output stage of the amplifier. A large current
heatsink to the surrounding atmosphere.
flow through resistors RE produces a voltage drop
360 9 Power Amplifiers
Example 9.15
Increasing Determine the maximum allowable dissipation at
Temperature
100oC in a power transistor rated at 125 W at
25oC if de-rating occurs above this temperature
C by a de-rating factor of 0.5 W/oC.
60
40
20
T J ¼ T A þ PD ðθJC þ θCS þ θSA Þ ð9:39Þ Solution From the previous example, the tran-
sistor was able to dissipate 87.5 W at a junction
where PD is the power to be dissipated, TJ is the temperature of 100oC. Thus, at this temperature
transistor junction temperature, TA is the ambient the transistor can safely dissipate 50 W. From
temperature in the immediate vicinity of the Equation (10.37), θSA ¼ (TJ TA)/PD θJC
transistor, θJC(oC/W) is the transistor thermal θCS ¼ (100 25)/50 0.4 0.2 ¼ 0.9oC/W.
resistance between junction and case, θCS is the
insulator thermal resistance between case and
heatsink and θSA is the heatsink thermal resis-
tance between heatsink and atmosphere. θJC is 9.11 Applications
set by the transistor manufacturer, while θCS is
determined by the transistor mounting A variety of power amplifiers and associated
arrangements. If a good thermal compound is systems are presented below.
used between the transistor case and the
heatsink, θCS 0.2oC/W. It is however often 1.5 W Low-Power Amplifier
necessary to isolate the transistor case from the The circuit shown in Fig. 9.31 is an improved
heatsink by introducing a mica wafer between non-inverting version of the low-power amplifier
them. This electrically isolates the transistor but with bootstrapping in Fig. 9.13a and is similar to
raises the thermal resistance to typically the configuration in Fig. 9.16 (improved medium
θCS 2oC/W. The last parameter θSA is the power). An additional common emitter
main heatsink specification and is determined amplifying stage Tr4 is introduced in the circuit.
using Equation (9.39). This stage provides further gain and allows the
application of voltage-series feedback at the emit-
Example 9.16 ter of Tr4. The resistors R1 and R2 provide a fixed
In a power amplifier design, a power transistor is voltage for the base of Tr4 and resistor R8 and
expected to dissipate 50 W. Using the transistor in capacitor C5 provide filtering. This configuration
the previous example, determine the thermal requires a compensation capacitor C4 whose
resistance of the heatsink required for safe opera- value is given by C 4 ¼ 2π gf mAV where gm ¼ 1/
c
tion of this transistor. Use θJC ¼ 0.4oC/W and (re(Tr4) + R9) and AV is the closed-loop gain of
θCS 2oC/W. the amplifier.
362 9 Power Amplifiers
For maximum symmetrical output, the quies- current stability in the output stage while
cent voltage at the output of the power amplifier is minimizing power dissipation from signal cur-
VCC/2. For P ¼ 1.5 W into 8 Ω, using V pk ¼ rent. Since only moderate output currents are
pffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2PRL , V pk ¼ 2 1:5 8 ¼ 4:9 V . In order involved, the power dissipation is the less serious
to allow for transistor saturation losses, the peak issue, and therefore these resistors can be selected
voltage swing is set at 6 V, and therefore for bias stability. A value of about 1 Ω for each of
VCC ¼ 2 6 ¼ 12 V. The peak output current these resistors is reasonable. Taking the current
is given by Ipk ¼ 4.9/8 ¼ 0.6 A. Using a current gain of Tr1 to be 100, the base current Ib(Tr1) ¼ 12
gain of 100 for each output transistor, the maxi- mA/100 ¼ 120 μA. Since this current is supplied
mum output current results in maximum base by Tr4 the quiescent current in this transistor is set
currents for each given by Ib(Tr2, 3) ¼ 0.6 A/ at say 0.5 mA. Hence R3 ¼ 0.7/0.5 mA ¼ 1.4 k.
100 ¼ 6 mA. This current flows into the collector For a 20 Hz low-frequency response into the
of Tr1 during negative-going half-cycles or is 8 Ω load, C2 is given by C2 ¼ 1/2π 20 8 ¼ 995
supplied by the bootstrapped resistor R4 during μF. A value of 1000 μF is therefore chosen. The
positive-going half-cycles. In order to ensure that VBE multiplier must provide at least 2VBE ¼ 1.4 V
this current is always available, the collector cur- drops across the base-emitter junctions of Tr2 and
rent of Tr1 is set at 12 mA which is twice the Tr3. Let IRa ¼ 1 mA. Then Ra(Tr5) ¼ 0.7 volts/
requirement for the base current to the output 1 mA ¼ 700 Ω. Hence 1 þ RRba ¼ 1:4 0:7 ¼ 2: Ra ¼
Rb
pffiffiffiffiffiffiffiffi
example, for P ¼ 100 W, R1 ¼ 40 100 ¼
400 Ω . The potential divider formed by resistor
R3 and potentiometer VR1 ¼ 1 k sets the voltage
to be applied to the base-emitter junction of Tr1
(noting that R1 is small). We allow for a 1 volt
drop across the LED and set the potentiometer at
a value of say 700 ohms when triggering occurs
pffiffiffi pffiffiffi
P1
for a given V pk ¼ 4 P. Then R43 þ1000 ¼ 700
0:7
from
pffiffiffi
which R3 ¼ 2000 2 P 1 . For example, for
pffiffiffiffiffiffiffiffi
P ¼ 100 W, R3 ¼ 2000 2 100 1 ¼ 38 k .
Fig. 9.33 Amplifier Clipping Indicator
R2 ¼ 10 k assists in turning on Tr2.
on. They remain in this condition until the voltage Ideas for Exploration (i) Investigate the possi-
across the arrangement goes to zero or the current bility of including a green LED in series with R3
through them goes to zero. When the transistors such that it illuminates when there is no clipping
are turned on, current flows through the LED but goes off when clipping occurs since then the
which then illuminates. Resistor R1 limits the current through R3 goes to zero.
current through the LED to about 100 mA. Thus
R1 ¼ Vpk/100 mA where Vpk is the maximum Power Operational Amplifiers
peak output voltage of the amplifier. For an A very simple method of realizing a power ampli-
amplifier delivering output power P into an fier is to use an operational amplifier and boost
pffiffiffi
8-ohm speaker load, V pk ¼ 4 P . Therefore the output current as shown in Fig. 9.34. The
pffiffiffi pffiffiffi
R1 ¼ V pk =100 mA ¼ 4 P=0:1 ¼ 40 P . For output of the operational amplifier drives a
9.11 Applications 365
complimentary emitter follower with a resistor R3 significantly reduced. For example, with
connected between the output of the operational R3 ¼ 100 Ω the limiting current is 0.7 V/
amplifier and the emitters of the two transistors. If 100 ¼ 7 mA above which the transistors take
the output current is less than 0.7 V/R3, then the over. It is important to note that while the bulk
transistors remain off, and the output current is of the output current is supplied by the transistors,
supplied by the operational amplifier. As the cur- base current to the transistors must be supplied by
rent exceeds 0.7 V/R3, the voltage drop across R3 the operational amplifier. This limits the amount
exceeds 0.7 V and the transistors turn on Tr1 for of current that can be delivered to about 1 ampere
positive-going signals and Tr2 for negative-going since with a transistor current gain of say 100, the
signals and supply the current. Crossover distor- base current demand for 1 A output is 10 mA
tion is suppressed by the operation of resistor R3 which is the maximum current typically available
and the inclusion of the booster transistors in the from an operational amplifier. Further enhance-
feedback loop of the operational amplifier ensures ment of the output current capacity can be
that distortion introduced by these transistors is achieved by the use of complimentary feedback
pairs in place of the two transistors. The current
gain for these pairs is better than 1000 in which
case as much as 10 A can be delivered using this
approach. In the system, RB ¼ 100 k provides
bias to the non-inverting input of the operational
amplifier and R1 and R2 set the gain at 1 + R2/R1.
For a 741 operational amplifier operating from
15 V supplies, the peak output voltage is
pffiffi 2
ð12= 2Þ
about 12 V giving a power of P ¼ 8 ¼
9 W into 8 ohms and 18 W into 4 ohms.
Another operational amplifier current booster
circuit is shown in Fig. 9.35. The transistors are
driven by current flowing into the power leads of
Fig. 9.34 Operational amplifier power system 1 the operational amplifier in response to a signal.
pffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
One advantage of this circuit over the previous using V pk ¼ 2PRL , V pk ¼ 2 75 8 ¼
current booster is R3 and R4 can be selected such 34:6V: Allowing for transistor saturation and
that a quiescent current flows in the transistors. other losses, the supply is selected to be
The result is that there is little or no crossover VCC ¼ 40 V.
distortion.
Using a 741 op-amp, R3 ¼ R4 ¼ 400 Ω pro- Ideas for Exploration (i) Replace the bootstrap
duced a quiescent current in the transistors of arrangement in the collector of Tr3 by a transistor
20 mA. Setting RB ¼ 100 k, R2 ¼ 9 k and constant current source; (ii) Introduce a transistor
R1 ¼ 1 k, a 1 V input resulted in a 10 V output. constant current source to supply the emitters of
Like the first current booster circuit, enhancement the differential pair Tr1 and Tr2. What improve-
of the output current capacity can be achieved by ment does this provide?
the use of complementary power Darlington pairs
such as the MJ3001/MJ2501 (80 V, 10 A) or the Research Project 1
TIP142/TIP147 (100 V, 10 A) in place of the two This power amplifier is intended to deliver
transistors. The current gain for these Darlington extremely high-power output. It must produce
pairs can exceed 1000 enabling the circuit to 250 W into 8 ohms or 500 W into 4 ohms. The
deliver currents approaching 10 A. In the system, project requires the scaling up of the power
RB provides bias to the non-inverting input of the MOSFET amplifier described in Example 9.12
operational amplifier, and R1 and R2 set the gain in order to produce these power levels. For
at 1 + R2/R1. Using this circuit, an OPA452 oper- pffiffiffiffiffiffiffiffiffiffiffi
P ¼ 250 W into 8 Ω, using V pk ¼ 2PRL ,
ational amplifier operating from 35 V supply pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
V pk ¼ 2 250 8 ¼ 63:2 V . Allowing for
can deliver 50 W into 8 ohms and 100 W into
transistor saturation and other losses, the supply
4 ohms.
is set at 75 V, and therefore VCC ¼ 75 V.
The number of MOSFETS in the output stage
Ideas for Exploration (i) While both the opera-
must at least be doubled compared to the 150 W
tional amplifier and the complimentary emitter
circuit therefore giving eight devices: four
follower can be operated from the same bipolar
n-channel and four p-channel. These can be
supply, it is important that the supply is
Renesas 2SK2221 and 2SJ352 or Magnatec
decoupled at the transistors and at the power
BUZ901 and BUZ906. This is a major project
pins to the operational amplifier. Also, separate
that should be undertaken with great care.
ground connectors must be used for the speaker
load and the operational amplifier ground. This
Ideas for Exploration (i) Refer to the article
prevents “ground loops” from creating spurious
“High-Power Audio Frequency Amplifier”,
results. (ii) Explore the use of the bridge configu-
Elektor Electronics, May 1986, p60.
ration in order to increase the power output of the
systems. (iii) Reconfigure this circuit to operate in
Research Project 2
the inverting configuration and examine its per-
Peter Walker and the Acoustical Manufacturing
formance using simulation.
Company introduced the current dumping ampli-
fier in 1975. It uses the system described in the
75 W Power Amplifier Using Feedback Pairs
op-amp power amplifier project in Fig. 9.34
This power amplifier in Fig. 9.36 utilizes com-
where transistors are turned on to “dump” current
plementary feedback pairs in the output stage.
into an external load. However, while the op-amp
The design follows the procedures previously
power amplifier uses negative feedback to reduce
discussed. Resistors R13 ¼ 560 Ω and
the distortion arising from the dumping action,
R14 ¼ 560 Ω provide current to Tr6 and Tr7 and
the system developed by this company is claimed
also assist in the proper operation of the output
to cancel this distortion completely. The project
transistors Tr5 and Tr8. For P ¼ 75 W into 8 Ω,
requires the reader to implement the scheme in an
9.11 Applications 367
transistors Tr1 and Tr2 is not present in the output therefore there is little power loss in the
signal. switching devices and as a result such amplifiers
have very high efficiencies (better than 90%). As
Ideas for Exploration (i) Refer to the articles a consequence, they require only small heatsink
“Current Dumping Audio Amplifier”, area since relatively low power is dissipated. The
P.J. Walker, Wireless World, December 1975, basic scheme is shown in Fig. 9.38. The input
p560 and Current Dumping Analysis by signal is compared with a triangular wave of a
H.S. Malvar, Wireless World, March 1981, p69. higher frequency of about 200 kHz using a com-
(ii) Examine other distortion cancelling schemes parator. The output of the comparator is a pulse
used by amplifier manufacturers or published in width modulated waveform whose pulse widths
the audio engineering literature. vary according to the signal amplitude.
This waveform then drives the push-pull
Research Project 3 MOSFET output stage, switching it fully on or
This project involves the design of a class D fully off for each pulse. The output of this
amplifier. This is a class of amplifiers that stage is applied to the loudspeaker through a
operates by switching on and off the output to low-pass LC filter to remove the unwanted
the external load. When on, the voltage across switching components and recover the amplified
the switching device is low and when off, the analog signal.
current through the switching device is very low,
Problems 369
Ideas for Exploration (i) Refer to International 7. Explain the phenomenon of crossover distor-
Rectifier Application Note AN-1071, “Class D tion and the reason it occurs. Suggest one
Amplifier Basics” by Jun Honda and Jonathan method of overcoming this problem.
Adams. (ii) Consider using the International Rec- 8. Design a low-power amplifier using the
tifier IRS2092 Class D Audio Amplifier Control- topology of Fig. 9.12 and a 12-volt supply.
ler IC that contains the control and switching 9. For the amplifier circuit shown in Fig. 9.42,
elements for driving the power MOSFETS. determine the quiescent current and the base
voltage in transistor Tr3 if the output voltage
is 4.5 volts. Develop a design procedure for
this configuration.
Problems
10. What would be the effect on amplifier perfor-
mance of increasing the input and output
1. Using the configuration in Fig. 9.39 and a
capacitors? Discuss the use of a variable
transistor with a current gain of 50, design a
Zener circuit to adjust the output bias current.
low-power amplifier to deliver 60 mW into an
11. Design a 500 mW amplifier using the circuit
8-ohm load using a 6-volt supply (Fig. 9.39).
of Fig. 9.43. Introduce bootstrapping to
2. If the 8-ohm speaker in the above amplifier is
improve the performance of this circuit.
replaced by a 15-ohm speaker, calculate the
output power.
3. A power amplifier has an efficiency of 30%
and delivers an output power of 15 W to a
resistive load. Calculate the power dissipated
and the DC input power to the amplifier.
4. A large signal amplifier delivers an output
power of 25 W and consumes 75 W from the
associated DC supply. Calculate its efficiency.
5. Using the configuration shown in Fig. 9.40,
design a transformer-coupled amplifier deliv-
ering 1.5 W into 8 ohms and operating from a
16-volt supply. Use a power Darlington with Fig. 9.40 Circuit for Question 5
β ¼ 2000.
6. Using the configuration shown in Fig. 9.41,
design a transformer-coupled amplifier deliv-
ering 2 W into 8 ohms and operating from a
24-volt supply. Use a power Darlington with
β ¼ 1000.
Fig. 9.39 Circuit for Question 1 Fig. 9.41 Circuit for Question 6
370 9 Power Amplifiers
19. In a power amplifier design a power transistor 21. An audio power amplifier can be represented
is expected to dissipate 75 W. Using a in a simplified model as an amplifier A1 with
transistor that can dissipate 100 W at a junc- relatively low distortion d1 driving an output
tion temperature of 100oC determine the stage A2 with significant distortion d2
thermal resistance of the heatsink required (Fig. 9.46). Positive feedback is applied
for safe operation of this transistor. Use around A1 via β1 and negative feedback is
θJC ¼ 0.4oC/W and θCS 2oC/W. applied around the system through β2. Show
20. As a research project, modify the configura- that when A1β1 ¼ 1, the output is given by
tion in Fig. 9.45 to include another transistor V o ¼ Vβ i þ Ad1 β1 . This means that in the ampli-
2 2
stage at the input as in Fig. 9.16. fier output signal Vo, the already low
372 9 Power Amplifiers
distortion d1 is further reduced by the factor Motorola Product Application Note AN-483B, Comple-
A1β2 and the major distortion d2 completely mentary Darlington Output Transistor in Audio
Amplifiers, April 1974
vanishes! D. Self, Audio Power Amplifier Design, 6th edn. (Focal
Press, 2013)
G. Slone, Randy, High-Power Audio Amplifier Construc-
tion Manual (McGraw-Hill, New York, 1999)
Bibliography M. Yunik, Design of Modern Transistor Circuits (Prentice
Hall, New Jersey, 1973)
B. Cordell, Designing Audio Power Amplifiers (Mc Graw
Hill, New York, 2011)
Power Supplies
10
An electrical power source is essential for the unregulated output of the filter to a regulated one
operation of most electronic equipment. Some where the DC voltage is very stable and ripple-
portable equipment such as radio receivers and free. The power supply comprising only of trans-
pocket calculators may be battery operated, but former, rectifier and filter is referred to as an
most electronic equipment requires an electrical unregulated supply. When a regulator is added,
supply. In this chapter, we examine the operation the system is referred to as a regulated power
and design of linear power supplies that convert supply.
an AC voltage from the public mains supply into
a stable DC voltage. At the end of the chapter, the
student will be able to: 10.2 Rectification
• Explain the operation of linear regulators The principle of rectification has been introduced
• Design several forms of linear regulated power in Chap. 1 where diode applications were
supplies discussed. In this chapter, more quantitative anal-
• Utilize IC regulators in regulated power ysis is done so that design issues may be consid-
supplies ered. The circuit shown in Fig. 10.2 is that of a
• Design short-circuit protection circuits half-wave rectifier with a transformer to reduce
the mains voltage to a more desirable level. The
diode permits current flow in one direction only.
The voltage VS represents the secondary voltage
10.1 Basic System of the transformer and is given by VS ¼ Vm sin ωt
where ω ¼ 2πf. For f ¼ 60 Hz, ω ¼ 377 rad/s.
A regulated power supply consists of several The output waveform is shown in Fig. 10.3. Its
sub-systems, a transformer, a rectifier, a filter average value is the ratio of the area under one
and a voltage regulator, as shown in Fig. 10.1. cycle of the half-wave rectified voltage waveform
The transformer converts the high AC mains volt- and the waveform period T ¼ 1/60 given by
age to a lower usable voltage while also providing
isolation from the mains. The rectifier converts RT
V m sin 377tdt
the low AC voltage from the transformer to a 0 Vm
varying DC voltage, while the filter removes the V AV ¼ ¼ ð10:1Þ
1=60 π
ripple from the rectified voltage, thereby
converting it from high ripple content to low Note that the area under half of the output
ripple content. Finally the regulator converts the voltage cycle is zero. The supply output is
# Springer Nature Switzerland AG 2021 373
S. J. G. Gift, B. Maundy, Electronic Circuit Design and Application,
https://doi.org/10.1007/978-3-030-46989-4_10
374 10 Power Supplies
RT
V m sin 377tdt
VDC 0 2V m
V AV ¼ ¼ ð10:2Þ
1=120 π
VS
A
VDC B
A
B ΔV
VDC
ID Ipk
ΔtD t
Δt
T
load current is approximately constant, then the too large, the diode peak current rating could be
average current pulse Ipk through the diode is exceeded causing damage to the diode (and/or the
given by transformer). Voltage and current waveforms are
shown in Fig. 10.9. Because of the capacitor
I pk Δt D ¼ I DC T ð10:5Þ
voltage, the diode is subjected to a maximum
from which reverse voltage of 2Vs, and therefore the diode
PIV rating must be PIV ¼ 2Vs.
I pk ¼
T
I ð10:6Þ The filtering of full-wave rectifiers, either that
Δt D DC using the centre-tapped transformer or the bridge
rectifier version, can be similarly accomplished
Since T ΔtD, the magnitude of the current
using a reservoir capacitor as shown in
pulse is quite large. Moreover because ΔtD
Figs. 10.10 and 10.11. Because of the availability
decreases as capacitor size increases, then the
of both cycles in the rectifier output, the capacitor
peak diode current increases. If the capacitor is
10.4 Average DC Output Voltage 377
is recharged twice as frequently as in the half- is 2Vs, while that for the bridge rectifier is Vs. The
wave rectifier, and therefore the ripple content is output voltage and current waveforms for these
approximately half of that in the half-wave circuits are shown in Fig. 10.12.
rectifier.
Also, since the capacitor discharge is reduced,
the diode current flow during recharge is also
10.4 Average DC Output Voltage
reduced. The peak inverse voltage for the full-
wave rectifier using the centre-tapped transformer
In order to calculate the peak-to-peak value of
the output ripple and the average output voltage,
certain approximations need to be made. Specif-
ically, from Fig. 10.8 for the half-wave rectifier,
the diode is off for most of the period T. During
this time, the capacitor delivers a load current IDC
and therefore discharges in the process. The decay
is exponential but can be approximated as a linear
decay. Also, the load current is assumed to be
approximately constant during the discharge. For
a capacitor, the relationship I ¼ CdV/dt holds.
Fig. 10.10 Full-wave rectifier with capacitive filter This can be approximated by I CΔV/Δt where
Fig. 10.11 Full-wave rectifier using diode bridge and capacitive filter
VS
ΔV
VDC
ID
ΔtD
Δt
T
rectifier, the peak-peak output ripple is given by tion of the output voltage under full-load
I DC T I DC conditions is given by VDC ¼ Vpk ΔV/
ΔV R ¼ ¼ ð10:9Þ 2 ¼ 17 1.5/2 16. In both designs, diode
C 2Cf
voltage drops and transformer winding resistance
From this, the average DC output voltage is reduce the output voltage.
V DC ¼ V pk ΔV=2 ð10:10Þ
and make terminal 1 positive with respect to the (80 V centre tap) is used. When rectified the
pffiffiffi
centre tap, while diodes D3 and D4 provide nega- peak output will be 40 2 ¼ 56:6 V . The maxi-
tive full-wave rectification and make terminal mum peak-to-peak ripple is 5 V, and allowing for
2 negative relative to the centre tap. Capacitors diode voltage drop and transformer resistance will
C1 and C2, respectively, filter the positive and give an average output voltage of about
negative voltages. It is possible to realize a bipo- 50 VDC. The transformer current rating is cho-
lar unregulated supply using a single winding sen as 5 A which is sufficient to deliver the 5 A
transformer. This elegant circuit is shown in peak current requirement into 8 ohms. The filter
Fig. 10.14. Each section is effectively a half- capacitors are given by C ¼ 2f ΔV I
¼ 2605
5A
¼
wave rectifier and therefore would in general 8333 μF. Use C ¼ 10,000 μF with 75 VDC work-
produce twice as much ripple as the circuit of
ing voltage. A diode bridge with piv 200 V and
Fig. 10.13. However, it does not require a 5 A rating is used to provide the rectification.
centre-tapped transformer, and this is its main
advantage. Moreover, this circuit can function as
a voltage doubler (as seen in the Zener diode
example of Fig. 1.69 in Chap. 1) by moving the 10.6 Voltage Multipliers
ground to either the +V or the V terminal.
It is possible to produce a high DC voltage using a
Example 10.3 low-voltage transformer. This can be done using a
Design a full-wave unregulated bipolar power combination of diodes and capacitors along
supply delivering 50 V DC at 5 A with less with the transformer. Such circuits rely on
than 5 V peak-to-peak ripple to supply the capacitor charge storage and are therefore not
100 W power amplifier in Example 9.10 of able to deliver high currents. In the circuit of
Chap. 9. Fig. 10.15, when the AC input is negative-
going such that diode D1 is forward-biased,
Solution The basic system is shown in capacitorpCffiffiffi1 is charged up to the peak transformer
Fig. 10.13 (excluding loads RL1 and RL2). In voltage 2V s with the polarity shown. During
order to produce 50 V, a 40 V transformer this time D2 is reverse-biased. When the
380 10 Power Supplies
pffiffiffi
transformer input reverses and is positive going, 2 2p Vffiffisffi across capacitor C2 charges capacitor C3
D1 is reverse-biased, but D2 is now forward- to 3 2V s . After a few cycles, the p output
ffiffiffi voltage
biased allowing capacitor C2 to charge
pffiffiffi up to the across C3 rises to approximately 3 2V s .
sum of p the
ffiffiffi transformer voltage 2V s and the Any number of multiplications of the basic
voltage 2V s on C1, both of which are in series. input voltage is realized in the multiplier circuit
After
pffiffiffi a few cycles, capacitor C2 charges up to of Fig. 10.17.
pffiffiffi The peak transformer secondary
2 2V s making this circuit a voltage doubler. voltage 2V s is multiplied by the number of
When a load is connected, however, C2 will be cascaded stagespresulting
ffiffiffi in an output voltage at
constantly discharging, and therefore, p the
ffiffiffi output terminal 1 of n 2V s relative to terminal A where
voltage will in general be less than 2 2V s and n corresponds to the number of diodes in the
will have high ripple content. Increasing the size circuit path. As can be seen from the circuit, n is
of the capacitors will reduce but not eliminate the always an even number. If point B is grounded
problem. instead of A, then pffiffiffi the output is taken at terminal
The principle of the voltage doubler may be 2 and is still n 2V s, but now npisffiffiffian odd number.
extended to voltage tripling, quadrupling and Thus, any multiplication of 2V s is possible
beyond. In the voltage tripler circuit of using this circuit. The diode and capacitor voltage
Fig. 10.16, during the half-cycle when terminal ratings
pffiffiffimust be twice the transformer output volt-
A is positive relative to terminal
pffiffiffi B, diode D1 turns age 2V s . Again, the capacitor values and diode
on, and C1 is charged to 2V s . When the trans- current ratings depend on the output current.
former voltage reverses such that terminal B is
positive relative to terminal A, diode D2 conducts
and the transformer voltage in series with the 10.7 Voltage Regulators
voltage
pffiffiffi across C1 together charge capacitor C2
to 2 2V s . When A is again positive relative Unregulated power supplies have certain undesir-
to B, diode D3 conducts, and the series combina- able characteristics. Firstly, the DC output voltage
tion of the transformer voltage and the voltage decreases, and the AC ripple increases as load
10.7 Voltage Regulators 381
current increases. Secondly, the output voltage This parameter is however difficult to calculate
changes significantly in response to a changing because of the difficulty in determining the rms
input voltage resulting from changing line volt- ripple voltage in the output. A ripple measure that
age. These disadvantages can be minimized by is more easily evaluated is the ripple factor r
adding a voltage regulator to the unregulated sup- given by
ply. The resulting supply is called a regulated
Peak to peak ripple in output voltage
power supply. The function of the regulator is r¼
Average or DC output voltage
therefore to maintain a constant output voltage
under conditions of changing load current and ð10:12Þ
changing line voltage. There are two types of The peak-to-peak output voltage ripple
voltage regulators: (a) the shunt regulator in involved in this specification is easily determined
which a control element is in parallel (shunt) using an oscilloscope. Both percent ripple and
with the load and (b) the series regulator in ripple factor measures should be as low as possi-
which a control element is in series with the load. ble. Another factor of importance in a power
In a shunt regulator, the regulating device is supply is load regulation which specifies the
placed in parallel with the load and works in change in the DC output voltage arising from
conjunction with a resistor in series with the changing load current. This factor is given by
load and the unregulated supply. The regulator
operates by varying the current through the con- V DC ðno loadÞ V DC ðfull loadÞ
Load regulation ¼ 100%
trol element depending on the load current. This V DC ðfull loadÞ
results in a varying voltage drop across the series ð10:13Þ
resistor such that the load voltage remains con-
stant. One example of a shunt regulator is the where VDC (no load) is the average value of the
output voltage when the external load resistance
Zener diode regulator which was introduced in
Chap. 1. In a series regulator, the regulating is removed and VDC (full load) is the average
device is placed in series with the load and the value of the output voltage when the external
load resistance is at a minimum. In practical
unregulated supply, and the voltage across the
control element is adjusted so that the load volt- circuits, diode resistance and transformer winding
age remains constant. Examples of these types of resistance will result in voltage drops that cause
the output voltage to be reduced with increasing
regulators will be discussed in this chapter.
load. It is desirable to reduce this drop and
thereby have a low load regulation. Line regula-
tion specifies the change in the DC output voltage
10.7.1 Ripple and Regulation resulting from a change in the line or input volt-
age. It is a measure of effectiveness of the regula-
The performance of a voltage regulator can be tion (output voltage stability) in the presence of
specified by several parameters. One of these changing input voltage. It is given by
gives an indication of the ripple content of the
ΔV out
DC output. The output of a rectifier contains Line regulation ¼ 100% ð10:14Þ
considerable ripple, which would have to be ΔV in
reduced by the regulator in order to make the where ΔVout is the change in the output voltage of
DC voltage useable. The quality of the resulting the regulator and ΔVin is the associated change in
output may be expressed using percent ripple R the input voltage to the regulator with constant
which is defined as load. Ideally, both line and load regulation should
Ripple voltage ðrmsÞ be zero, and in practice, the values are less than
R¼ 100% 0.01% for good regulators. The ratio ΔVout/ΔVin
Average or DC output voltage
is sometimes referred to as stability factor.
ð10:11Þ
382 10 Power Supplies
Zener Region
prevented from rising too high, the voltage across
the Zener in this condition is approximately con-
stant at the breakdown or Zener voltage, and the
dynamic resistance rd is quite low, ranging from a
few ohms to about 50 ohms. In this region, the
voltage across the Zener is approximately con- IZ(max)
stant, and when operated here, the device
provides a constant voltage output. Fig. 10.18 Zener diode characteristic
The current through the Zener in this region is
limited by the power rating PD of the diode and
must be operated such that VzIz < PD. If
Iz(max) ¼ PD/Vz is exceeded, the diode will be
destroyed. Also, Iz must not fall below the mini-
mum value Iz(min); otherwise, the Zener voltage
will fall. This corresponds to the knee of the curve
in Fig. 10.18. Zener diodes are available in
voltages ranging from about 2.4 V to about
200 V. The power ratings range from about
1/4 w to over 100 W. Fig. 10.19 Zener diode regulator
The Zener diode operates as an effective volt-
age regulator. The basic system is shown in value of IL must be known. When IL is at maxi-
Fig. 10.19. A (varying) voltage Vi is applied to mum corresponding to minimum RL, and Vi at a
the Zener diode through a resistor R that limits the minimum, the value of R must be such that
current. The voltage VZ across the Zener is then Iz Izmin.
applied to a load RL. If Vi increases, say, more Thus,
current flows through the Zener but based on its
V i min V Z
characteristic, VZ stays approximately constant. I L max I Z min ð10:15Þ
R
Similarly, if Vi decreases, the current though the
Zener drops but again VZ remains constant. Thus, i.e.
the voltage VZ is approximately constant despite
V i min V Z
changes in the input voltage or load current. The R ð10:16Þ
I L max þ I Z min
load RL therefore sees a regulated (constant) volt-
age providing Izmin Iz Izmax. This design of a When Vi rises and IL goes to zero (load
Zener-regulated power supply involves the deter- removed), the current through the Zener is at its
mination of R. In order to do so, the minimum and maximum, and the required power rating PD of
maximum values of Vi as well as the maximum the diode must satisfy
10.7 Voltage Regulators 383
Example 10.3
A voltage from an unregulated supply varies
between a minimum value of 16 volts and a
maximum value of 20 volts. Using this unregu-
Fig. 10.20 Equivalent circuit of Zener diode regulator lated supply, design a Zener-regulated supply that
delivers 10 volts DC with a current capacity of
10 mA.
V i max V Z
PD > V Z ð10:17Þ
R
Solution The first step is to choose a 10 volt
Zener-regulated power supplies are useful for Zener for the circuit of Fig. 10.19. The power
low-power applications involving less than a few rating will be determined later. To determine
watts since under the condition where no load R we consider the case where Zener current is
current is being supplied, the Zener must accom- at a minimum. This occurs when the input
modate the full load current. voltage is at a minimum, in this case 16 volts,
In order to determine the ripple at the output of and the load current is at a maximum at 10 mA.
the Zener diode regulator, the dynamic resistance We determine from diode specifications the
of the Zener must be considered. The dynamic minimum diode current for proper Zener opera-
resistance is the reciprocal of the slope of the tion. Assume a value of 5 mA. The voltage
diode characteristic in the operating region. It is across R is (16–10) volts and the current through
an indication of the resistance of the diode to R is (10 + 5) mA. R is therefore given by R ¼
small voltage and current changes. The value is ð1610Þ
typically in the range 5 Ω to 50 Ω. The actual ð10þ5Þ 103 ¼ 400 Ω. Note that if the load were
circuit under small-signal conditions is equivalent suddenly removed, then the current through D
to that shown Fig. 10.20. For input ripple voltage would rise to 15 mA when the input voltage is
vi, the output ripple vR is given by at 16 volts. To determine the diode power
rating, we calculate the maximum current
r d ==RL through the diode. This occurs when the unregu-
vR ¼ vi
R þ r d ==RL ð10:18Þ lated input voltage is at a maximum of 20 volts
rd
¼ vi and the load current is at a minimum (zero)
R þ rd corresponding to a removal of the load. The volt-
where rd RL. Hence to minimize vR, R should age across R is then (20 10) ¼ 10 V, and hence,
be as large as possible, for example, by using a the current ID through D is given by I D ¼ 400
10
¼
large resistor or a constant current diode. Hence in 25 mA. This current flows through R, and hence
(10.16), equality is used with Izmin set at a value the required Zener power rating PD is given
determined by the specification sheet. If the load by PD ¼ ID. VD ¼ 25 mA 10 V ¼ 0.25 W.
is variable and Vi is fixed, then Vimin ¼ Vi. If Vi is We choose a 1/2 W diode to provide a safety
variable and the load is fixed, then at no time will margin and for longer diode life. In this design,
current flowing through R all flow into the Zener the input ripple voltage is (20 16) ¼ 4 V pk-to-
since some current will always be flowing into the pk. Hence, assuming rd ¼ 10 Ω, the output ripple
load. Hence (10.17) becomes voltage is vR ¼ 400þ10
10
4 ¼ 0:1 V . The ripple
factor r is then r ¼ V DC ¼ 0:1
vR
10 ¼ 0:01.
V ið max Þ V Z
PD > V Z IL ð10:19Þ
R
384 10 Power Supplies
Fig. 10.25 Transistor regulator with electronic protection Fig. 10.26 Changed position for protection transistor
This circuit has the disadvantage of causing a load current increases, the current through D2
change in the regulated output voltage as current decreases with a consequent fall in the
flows through RX. This problem can be overcome voltage across D2. Eventually D2 turns off and
by placing the protection transistor in the position all of the current through R2 is supplied by Tr1.
shown in Fig. 10.26. In this position the This results in a maximum current through Tr2
regulated output is unaffected by the voltage into the load given by IL(max) ¼ hfe2(VZ 0.7)/R2
drop across RX. where hfe2 is the current gain of Tr2. The
The second transistor introduced to provide circuit now converts from a constant voltage
electronic protection can provide that protection source to a constant current source, and as a
in a different manner as shown in Fig. 10.27. The result any further decrease in load resistance
introduced transistor is a pnp transistor along with results in a fall in the output voltage. This may
resistor R2 and signal diode D2. The collector be described as a constant current characteristic.
current IC1 of Tr1 flows out of the base of Tr2 Under short-circuit conditions, the maximum
into R2. This current results in a collector current current IL(max) continues to flow into through the
IC2 in Tr2 which flows through diode D2 also short circuit. It is possible to reduce this short-
into R2. The current flowing in R2 is therefore circuit current by the introduction of another sig-
made up of the collector currents of the two nal diode D3 as shown in Fig. 10.28. When the
transistors and is given by (VZ 0.7)/R2. As maximum output current flows and the regulator
10.7 Voltage Regulators 387
(why?). Let the minimum current through the a constant current source. An excellent method
Zener be 5 mA for good Zener performance. of implementing this scheme is the use of the
Also, choose IC1 ¼ 1 mA as a reasonable operating constant current diode made using a JFET. This
current for Tr1. Hence, R4 ¼ V i ð min ÞV z
¼ is a two-terminal device made, for example, by
pffiffi 5 mA
Siliconix that can simply replace the resistor
12 215:6 V i ð min Þð9þ0:7Þ
5 mA ¼ 2 k: Also, R3 ¼ I L =hfe þI C ðTr1 Þ ¼
pffiffi feeding the Zener. Finally, a small capacitor
12 219:7
100 mA=100þ1 mA ¼ 3 k2: The current flowing placed across the circuit output removes output
through R2 and R1 must be at least ten times noise and reduces the supply impedance at high
the base current of Tr2. The base current IB1 of Tr1 frequencies. In order to handle a large load cur-
is approximately 1 mA/100 ¼ 10 μA. Hence, rent, a second transistor Tr3 can be used along with
Tr2 in a Darlington arrangement as shown in
9
R1 þR2 10 10 μA , R1 + R2 90 kΩ, V o ¼
Fig. 10.32. This reduces the pass transistor base
ðV Z þ 0:7Þ 1 þ RR21 and RR21 ¼ 6:3
9
1 ¼ 0:43 . current demand that may otherwise cause resistor
R3 to be too low in value and thereby result in too
Thus for R1 ¼ 10 kΩ, R2 ¼ 4.3 kΩ and note that
large a standing current in the error transistor Tr1.
for these values R1 + R2 90 kΩ as required.
The calculation would now involve the combined
Finally, C ¼ 120Δ
I L max
V ¼ 1201 ¼ 833 μF.
100 mA
transistor current gain hfe2hfe3. Further ripple
reduction can be achieved by applying
There are several ways to improve the perfor- bootstrapping to the load resistor R3 of the error
mance of this circuit. A great improvement can transistor Tr1. This increases the error transistor
be realized by supplying the Zener from the gain which increases the applied feedback.
regulated output instead of the unregulated
input by connecting resistor R4 to the regulated Example 10.7
output. This reduces the ripple across the Zener Using the circuit shown in Fig. 10.32, design a
diode and hence at the output. A second +20 volt, 1.5 A regulator that is driven by an
improvement is to place a capacitor across the unregulated supply having 2 volt peak-peak
Zener diode, again reducing the ripple. The input ripple. Use a 22 volt transformer and a
value of this capacitor is that value that has an bridge rectifier to provide the unregulated input.
impedance comparable to or lower than the Assume the power transistor has a gain of 25 and
dynamic resistance of the Zener. A third method the other transistors have gains of 150. Justify all
of performance improvement is to replace R4 by design steps.
10.7 Voltage Regulators 391
R
Vo ¼ Vz 1 þ 2 ð10:36Þ
R1
Fig. 10.39 Variable transistor voltage regulator
Note that the op-amp must be supplied from
the unregulated supply. (Why?) Similar to the 5 mA ¼ 1.5 kΩ. From (10.36), 7:5 1 þ RR21 ¼
improved transistor case, the circuit can be
improved by supplying the Zener diode from the 15 . Hence, R1 ¼ 10 k ¼ R2. C ¼ 2ΔVf
I
¼
regulated output (Fig. 10.37). All the other 2
2260 ¼ 8333 μF.
improvement techniques (except bootstrapping)
are applicable here. The current limiting Variable Output Voltage
techniques are also applicable. The basic feedback voltage regulator can be
modified to produce a variable output voltage.
Example 10.12 In such a situation since the regulated output
Using the circuit shown in Fig. 10.38, design a will now be variable, the Zener diode which
+15 volt, 2 A regulator that is driven by an unreg- provides the reference voltage will have to be
ulated supply having 2 volt peak-peak input rip- supplied from the unregulated input voltage and
ple. Use a 20 volt transformer and a full-wave not the regulated output if its current supply is to
rectifier to provide the unregulated input. Assume be maintained.
the power transistor has a gain of 150. In the case of the transistor feedback regulator,
one approach is shown in Fig. 10.39. Here the
Solution Choose VZ ¼ 7.5 V. Set IZ ¼ 5 mA for Zener diode is supplied from the unregulated
good Zener operation. Hence, R3 ¼ (15 7.5)/ voltage through a constant current diode to mini-
mize ripple across the Zener diode. Additionally,
394 10 Power Supplies
a large filter capacitor is placed in parallel with immediately destroyed by excessive current
the Zener diode to further reduce ripple. The flow. Various protection schemes can be used to
variable output voltage is produced by varying prevent this. One circuit that can be used is a
the feedback through adjustment of VR1 in current limiting circuit shown in Fig. 10.41. Tran-
Fig. 10.39 which replaced feedback resistor R2. sistor Tr3 and R5 are introduced such that the load
The minimum voltage is VZ, while the current flows through R5 and thereby develops a
maximum voltage must be at least three volts voltage across the base-emitter junction of Tr3.
less than the minimum unregulated voltage For a sufficiently large load current, this voltage
in order to ensure that the pass transistor is prop- exceeds the turn-on voltage of the transistor
erly biased. Thus, ðV Z þ V BE Þ 1 þ VR 1
¼ (0.7 V). Tr3 therefore turns on and diverts base
R1
current away from the base of Tr1, thereby limit-
ðV mn ðunregÞ 3Þ is the design equation. ing the load current. The design equations are
With respect to the op-amp feedback regulator, simple: Let the maximum load current be IL(max).
one simple approach to achieving variable Then at turn on IL(max)R5 ¼ 0.7 V giving
regulated voltage is shown in Fig. 10.40. Here
again the Zener diode is supplied from the unregu- 0:7
R5 ¼ ð10:37Þ
lated voltage through a constant current diode to I Lð max Þ
minimize ripple, and a large filter capacitor C1 is
placed in parallel with the Zener diode. The vari- This current limiting circuit at switch-on
able output voltage is produced by varying the converts the regulator from a constant voltage
reference voltage through adjustment of VR1 in
Fig. 10.40. This approach has the advantage that
the feedback around the system is unaltered,
thereby maintaining the system characteristics as
the voltage is varied. It is important to note that the
output of this system cannot go to zero because of
saturation voltage loss within the op-amp. Once
again, the maximum voltage must be set at about
3 V below the minimum unregulated voltage.
In this linear regulator, if the output is short- Fig. 10.41 Transistor voltage regulator with electronic
circuited to ground, the pass transistor will be protection
Vout
R6 ¼ 317 Ω.
Protection for the op-amp regulator follows
similar principles as seen in Fig. 10.45. Here
transistor Tr2 is used to introduce electronic
Fig. 10.46 Three-terminal IC regulator
short-circuit protection as in the discrete transistor
case. When load current flows such that the volt-
voltages and currents are available. The design
age drop across R4 equals 0.7 V, then Tr2 turns on
of a fully regulated power supply using an IC
and draws current away from the pass transistor
regulator is quite straightforward.
Tr1, thereby limiting the load current drawn from
the supply. Resistor R5 ensures that the output
Three-Terminal Voltage Regulator
voltage of the op-amp decreases as current
The simplest of the IC regulators is probably the
through Tr2 increases, thereby reducing the out-
three-terminal device shown in Fig. 10.46. An
put voltage of the regulated supply.
unregulated voltage is supplied to the input termi-
nal, a regulated output is available at the output
terminal, and the third terminal is grounded. Both
10.7.6 IC Voltage Regulators positive and negative voltage devices are
available.
Instead of building regulators using transistors
and operational amplifiers, IC voltage regulators Fixed Positive Voltage Regulators
that greatly simplify the process are available. Fixed positive voltages are available in the 7800
These ICs contain all the elements of the basic series of regulators. The voltages range from
block including reference voltage, error amplifier, 5 volts to 24 volts as shown in Table 10.1. As
series pass device and overload protection cir- an example, the 7815 IC regulator is shown in
cuitry. They provide excellent voltage regulation Fig. 10.47. The output voltage is +15 volts. A
in three basic formats, fixed positive voltage, positive rectified input voltage is supplied at Vi
fixed negative voltage and adjustable output with capacitor C1 providing filtering (reservoir
voltages. The overall power supply consists of capacitor). The output is filtered by capacitor C2
an unregulated supply providing the input voltage which is a smaller capacitor, designed to remove
to the IC regulator. A wide range of output mostly high-frequency noise. The ground
10.7 Voltage Regulators 397
terminal is connected to ground. If the supply input voltage specification for the regulator.
transformer has an output voltage of Vs rms, Some specifications include the following:
pffiffiffi
then the peak voltage into the regulator is 2V s ,
pffiffiffi Output Voltage: Regulated Output Voltage
and the minimum voltage is V mn ¼ 2V s ΔV
where ΔV is the peak-to-peak ripple. Line Regulation: Typically, 0.1%
Based on the maximum load current ILmx, the Load Regulation: Typically, 0.1%
reservoir capacitor C1 is given by Current Limit: The maximum current that flows at
the output of the regulator into a short circuit.
I Lð max Þ Drop Out Voltage: This is the minimum voltage
C1 ¼ ð10:42Þ
2f ΔV across the input and output terminal of the
regulator that must be maintained if regulating
Note that the minimum value of the unregu-
action is to be sustained.
lated input voltage must exceed the minimum
Ripple Rejection Ratio: Ratio in dB of input
ripple to output ripple.
Output Resistance: The DC output resistance of
Table 10.1 7800 series positive voltage regulators the regulator.
Output voltage Minimum input voltage
IC (V ) (V ) Fixed Negative Voltage Regulators
7805 +5 7.3 Similar to the 7800 series positive voltage
7806 +6 8.3 regulators, the 7900 series provides negative
7808 +8 10.5
regulated voltages. Similar specifications apply
7810 +10 12.5
with the list of available voltages shown in
7812 +12 14.6
Table 10.2. The connection of a 5 volt regulator
7815 +15 17.7
7818 +18 21.0 is shown in Fig. 10.48.
7824 +24 27.1
Adjustable Voltage Regulators
Adjustable voltage regulators allow the user to
change the output voltage of the regulator to a
desired level. The LM317 is an example of such a
regulator. Its output voltage can be adjusted over
the range 1.2 volts to 37 volts. The circuit con-
nection is shown in Fig. 10.49. The output volt-
age Vo is given by
R
V o ¼ V ref 1 þ 2 þ I adj R2 ð10:43Þ
R1
Fig. 10.47 Three-terminal positive fixed voltage
regulator
Fig. 10.50 Adjustable regulated power supply using the LM317 IC regulator
10.7 Voltage Regulators 399
10.7.7 Simple Approach to Regulated feedback loop of the TL431A and therefore
the
Power Supplies output voltage is given by V o ¼ 1 þ R1 2:5 V.
R2
20 volt output, 20 ¼ (1 + R2/R1) 2.5. Using variable reference voltage to the reference input
R1 ¼ 2 k, then R2 ¼ 14 k. Capacitor C2 ¼ 100 μF of the op-amp. In order that zero output voltage be
reduces ripple across the Zener, and C3 ¼ 1 μF possible, the negative supply terminal of the
removes any high-frequency noise and residual op-amp cannot go to zero (why?) but must be
ripple. taken to a negative voltage. A negative voltage
of 6.8 volts is furnished by Zener diode D5
which is powered via R3 by the half-wave rectifier
D3 C2. The total voltage between the power
10.8 Applications pffiffiffi
supply terminals of the op-amp is 15 2 þ 6:8 ¼
In this section, several regulated power supply 28 V which is less than the maximum rated supply
designs are presented. voltage of 2 22 ¼ 44 V for the LM741. A
Darlington pair comprising Tr1 and Tr2 is used
Variable Voltage Regulated Supply and can be made up using a 2 N3053 and a
We here explore the design of a variable voltage 2 N3055, respectively, or a MJ000 power
regulated power supply using the basic op-amp Darlington. Potentiometer VR2 permits variation
configuration of Fig. 10.37. The system must be of the current limit. Tr3 can be a small-signal
adjustable down to zero and up to about 15 volts transistor 2 N3904. Rectifying diodes D1 and D2
and must have adjustable current limiting from must have piv ratings of better than 2 15
pffiffiffi
about 100 mA to 1 A. In order to secure a 15 volt 2 ¼ 42 V and current rating higher than 1 A.
output from a regulator, a 30 volt centre-tapped The 1 N5401 diode is rated at 100 V and 3 A and
transformer rated at twice the required current is can be used. Since I ¼ 1 A and f ¼ 60 Hz, in
used. This would enable the design of a full-wave order to realize a peak-to-peak ripple voltage of
rectifier using two rectifying diodes that provides about 2.5 volts, capacitor C1 must be C 1 ¼
sufficient voltage to ensure pass transistor opera- I
¼ 2602:5
1
¼ 3333 μF . The negative supply
2f ΔV
tion. One system topology is shown in Fig. 10.53.
to the op-amp must deliver at least 2 mA to the
It is an enhancement of the circuit of Fig. 10.37.
op-amp. Hence using I ¼ 0.002 A and f ¼ 60 Hz,
In order to achieve variable output, the Zener
in order to realize a pk-to-pk ripple voltage of
diode D4 is supplied from the unregulated input
about 2.5 volts, capacitor C2 must be C 2 ¼
through a constant current diode D6 for improved
ripple reduction. The Zener voltage supplies a
I
¼ 0:002
f ΔV 601 ¼ 33 μF . Use C2 ¼ 100 μF. Diode
potentiometer VR1 which in turn supplies a D3 can be an 1 N4002 with piv ¼ 100 V and
current rating of 1 A. Resistor R3 must supply the (80 V, 50 mA) operational amplifier to replace
Zener and the op-amp with current. Allowing the LM741. (ii) Utilize the LM338A regulator IC
10 mA through D5 in order to accomplish in the design of a variable regulated power
pffiffiffi
this, then R3 ¼ 15 2 6:8 =10 mA ¼ 1:4 k: A supply.
5.6 V Zener is used for the reference Zener D4.
This value will be amplified by the op-amp to Variable Power Supply with Zener
produce the desired output voltage. VR1 is a Stabilization
10 k potentiometer to provide the voltage varia- This regulated supply utilizing a 741 op-amp
tion. Capacitor C3 should have a reactance that is includes variable output voltage from 5 V to
comparable to the dynamic resistance of the Zener 20 V at 1 A while supplying the Zener from the
which is of the order of tens of ohms. A 10 uF regulated output. This is achieved by varying the
capacitor has a reactance at f ¼ 120 Hz of resistance supplying the 3.9 V Zener while also
1/2π120 10 ¼ 133 Ω which is acceptable. The varying the applied feedback. The circuit is
constant current diode D6 is a Siliconix J509 which shown in Fig. 10.54. The essence of the approach
supplies about 3 mA to D4. The high impedance of is to vary both the amount of feedback and the
D6 and the low impedance of D4 in parallel with C3 value of the resistor supplying the Zener. This
ensure that the ripple content at the non-inverting latter action will ensure that as the output voltage
input at the op amp is very low. A choice of changes, the current through the Zener will
R1 ¼ 10 k and R2 ¼ 18 k means that the maximum remain approximately constant. Because of the
saturation loss of the op-amp, the output is not
output of the system is 5:6 1 þ 18 10 ¼ 15:7 V just
above the desired output voltage level of 15. Resis- allowed to fall below 5 V. With an input unregu-
tor R5 sets the maximum current of say 1.1 A and is lated voltage of 30 V, the maximum voltage is
given by R5 ¼ 0.7/1.1 A ¼ 0.6 Ω. The 10 ohm 20 V. The output of the Zener is passed through a
potentiometer VR2 allows variation of the current low-pass filter R4 C2 to reduce any noise, and a
limit to a minimum value of Imin ¼ 0.7/10.6 ¼ 66 capacitor C3 is implemented across the output
mA. The resistor R4 ¼ 330 Ω assists in lowering also for purposes of filtering. Both can be about
the output voltage during a current overload con- 10 μF. A Darlington pair comprising a 2 N3053
dition. Finally, capacitor C4 ¼ 0.01 μF is intended medium power transistor and a 2 N3055 power
to reduce high-frequency noise at the output of the transistor or an MJ3000 is used as the series
power supply. device.
Ideas for Exploration (i) Re-design the system Ideas for Exploration (i) Re-design the system
to deliver a higher voltage using the OPA452 to deliver a higher voltage using the OPA452
(80 V, 50 mA) operational amplifier to replace Therefore, the positive Zener voltage with respect
the 741. to the zero potential is VZ ¼ IR1 + 0.7 ¼ 1.4R1/
RV + 0.7, and the negative Zener voltage with
Discrete Voltage Regulator respect to the zero reference is VZ ¼ (IR1 +
This Zener diode regulator enhanced by an emit- 0.7) ¼ (1.4R1/RV + 0.7). Thus, a symmetrical
ter follower enables the adjustment of the Zener supply is available with easy adjustment using
voltage to compensate for the voltage drop caused RV. Variable resistor RV should comprise a vari-
by emitter follower action. The follower is really able component RVa and a fixed component Rb
a feedback pair with modest gain where the sec- such that RV has a minimum value RV ¼ Rb. For
ond transistor is a pnp Darlington pair such as the example, let R1 ¼ 6.8 k, RVa ¼ 10 k and Rb ¼ 1 k.
MJ2501. Thus, with a 15 V Zener, the voltage Then noting that RVmax ¼ (10 k + 1 k) ¼ 11 k and
that would be available at the output of Tr2 would RVmin ¼ 1 k, then VZmin ¼ (1.4 6.8 k/
be 14.3 V because of the VBE drop. Thus if 11 k) + 0.7 ¼ 1.6 V and VZmax ¼ (1.4 6.8 k/
R1 ¼ 12 k and R2 ¼ 600 Ω, then Vo ¼ (1 + 600/ 1 k) + 0.7 ¼ 10.2 V. This means that the bipolar
12000) 14.3 ¼ 15 V. The JFET is connected as Zener voltages are variable from 1.6 V to
a constant current diode supplying the Zener 10.2 V using RV.
diode. Capacitor C1 ¼ 10 μF provides filtering
for the Zener (Fig. 10.55). Ideas for Exploration (i) Use this configuration
to design an adjustable Zener-regulated bipolar
Ideas for Exploration (i) Re-design the system power supply.
to provide a negative voltage with respect to
ground. (ii) Explore the introduction of short- Simple Feedback Power Supply
circuit protection. This feedback regulator delivers 6 V at 100 mA
from a 12 V supply. It does not use a reference
Bipolar Zener voltage Zener but instead uses the error transistor
The circuit in Fig. 10.56 is essentially two com- base-emitter junction as the reference voltage.
plimentary feedback pairs connected as VBE The circuit is shown in Fig. 10.57. The defining
R1 ¼ R1 þR2 where Vo is the regulated
Vo
multipliers with voltage adjustment by a single equation is 0:7
resistor. Since variable resistor RV is across the
base-emitter junctions of Tr1 and Tr2, it follows
that the current I through RV is I ¼ 1:4V RV .
output voltage and R1 ¼ R1a + VR1. For Vo ¼ 6 V, Fig. 10.58 Regulated power supply using OPA549
if R2 ¼ 4.7 k then R1 ¼ 620 Ω. A good approach power op-amp
is to set R1a ¼ 470 Ω and potentiometer VR1 ¼ 1 k
in series with R1a in order to adjust the output in a regulated output of 1 V to 25 V at a maximum
current of 8 A. Current limit is set by resistor
voltage to the desired voltage. A load current of
100 mA produces a base current requirement in RCL ¼ I75 LM
7:5 kΩ where ILM is the current
Tr1 of 100 mA/50 ¼ 2 mA where β ¼ 50 is used. limit in amperes. For example, for a current limit
With a collector current in Tr2 of 2 mA, then
of 3 A, RCL ¼ 75 3 7:5 k ¼ 17:5 k.
R3 ¼ (12 6 0.7)/(2 mA + 2 mA) ¼ 1.3 k.
This circuit can be adapted to a variety of
Ideas for Exploration (i) Re-design the system
applications using an unregulated input. If an
to deliver a negative voltage relative to ground.
unregulated supply is used, R3 can be
bootstrapped to realize a higher gain in Tr2 and
Dual Tracking Power Supply
therefore greater ripple-reducing feedback.
This project involves the design of the dual track-
R4 ¼ 1 k provides about 5 mA to the LED to
ing power supply shown in Fig. 10.59. It has one
indicate the on-condition.
positive and one negative output of equal voltage.
The positive voltage is adjustable from 0 to 20 V,
Ideas for Exploration (i) Add electronic protec-
and the negative supply automatically tracks the
tion to this circuit. (ii)Use this configuration to
positive voltage. The supply can deliver up to
design an adjustable Zener-regulated bipolar
1 A. The basic system for each voltage polarity
power supply. (ii) Use a 20 V transformer and
is essentially the op-amp system discussed earlier
an unregulated half-wave rectifier to supply the
in this chapter with certain modifications. A 24 V
circuit.
transformer is used to supply each regulator. A
low-cost high-voltage replacement for the
High-Current Variable Power Supply Using
741 op-amp is used: It is the OPA452. This
OPA549 Power Op-Amp
op-amp operates from up to 40 V, can deliver
The OPA549 is a high-voltage, high-current oper-
50 mA, has a GBP of 1.8 MHz and is unity-gain
ational amplifier that can be used as a variable
stable. The pass transistors are MJ3001 and
regulated power supply. It is simply connected as
MJ2501 complementary power Darlington
an amplifier with a gain of 10 with DC at the input
which are 80 V, 10 A, 150 W devices with
and the amplified DC at the output. The op-amp is
β ¼ 1000. These ensure that the current drawn
powered from a 30 V unregulated supply. This is
from each op-amp is no more than 1 A/
shown in Fig. 10.58, where a variable DC voltage
1000 ¼ 1 mA. The feedback components
of 0.1 V to 2.5 V at the non-inverting input results
404 10 Power Supplies
Ideas for Exploration (i) Simulate this system Ideas for Exploration (i) Examine whether
and that using bootstrapping and compare the resistor R3 is necessary in this application.
load regulation and line regulation performances (ii) Use an OPA452 to produce higher dual
of the two systems. voltages.
the current to the base of the feedback pair, and using energy storage devices such as an
thereby converting the supply from a constant inductor and/or a capacitor. Its main advantage
voltage supply into a constant current supply. over the linear regulators discussed in this chapter
Under short-circuit conditions, the full unregu- is its high efficiency which is better than 90%.
lated supply voltage is across Tr2 and with One class of switch-mode power supply is the
450 mA flowing through the device, some step down or buck converter shown in
10 W of power would be dissipated in this Fig. 10.63. The operation can be divided into
device. It must therefore be mounted on a suit- two phases: a charge phase and a discharge
able heatsink. Capacitor C2 ¼ 100 μF provides phase. In the charge phase, the switch S is closed,
additional filtering at the output, and resistor and energy is stored in the inductor L. In the
R7 ¼ 1 k discharges this capacitor when the discharge phase, the switch is open, and the
supply is turned off. energy stored in the inductor is transferred to the
output capacitor C and load through the diode
Ideas for Exploration (i) Introduce an LED in D. The capacitor maintains the load voltage dur-
series with R7 to indicate the ON condition. ing the charging phase of the inductor which is
(ii) Convert the regulated supply into a fully vari- central to the energy transfer process. Since there
able supply by replacing the resistor string by a is no build-up of current in the inductor, it follows
potentiometer of value about 10 k. The wiper of that the change in inductor current ΔIC during the
the potentiometer then goes to the base of Tr3. charging phase must be equal to the change in
(iii) Introduce fold-back current limiting to this inductor current ΔID during the discharge phase,
circuit. i.e. |ΔIC| ¼ |ΔID|. From this it can be shown that
the relation between the input voltage Vi and the
Research Project 4 output voltage Vo is given by Vo ¼ DVi where
This research project involves the investigation of D ¼ TON/TS, TON is the on-time of the switch and
a switch-mode power supply. Such a supply TS is the switching period. Since D < 1, then
converts one voltage level to another by Vo < Vi which represents a reduction of the input
switching devices on and off at high frequency voltage as generally occurs with the series
regulators discussed in this chapter.
A simple implementation of this system is
shown in Fig. 10.64. A variable duty cycle square
wave is generated by the 555 timer, and this is
used to switch on and off a p-channel MOSFET
IRF4905. Potentiometer VR1 ¼ 10 k varies the
duty cycle of the square wave and hence the
output voltage. The output voltage in this circuit
varies with changing load since no feedback is
Fig. 10.63 Step down buck converter present. An improved circuit is shown in
Vi Tr3 Vo
R3
Tr2 R2
C Tr1 R4
D
R1
Vi Vo
Tr2
R3
R1
Tr1
C
D R2
20. Indicate other methods of improving the per- 25. Outline the operation of a switch-mode buck
formance of a regulator. regulator.
21. Design a +12 V regulated supply using regu-
lator IC from the 7800 series.
22. Design a 15 V regulated bipolar supply
using regulator ICs from 7800 and 7900 Bibliography
series.
23. Using the circuit of Fig. 10.51, design a vari- J. Millman, C.C. Halkias, Integrated Electronics: Analog
able voltage regulated supply that can deliver and Digital Circuits and Systems (Mc Graw Hill Book
0–9 V at 100 mA. Company, New York, 1972)
R. Coughlin, F. Driscoll, Operational Amplifiers and Lin-
24. Design a variable voltage bench power sup- ear Integrated Circuits, 5th edn. (Prentice Hall, New
ply using a Darlington pair and the topology Jersey, 1998)
of Fig. 10.71.
Active Filters
11
A filter is an electrical network that passes signals all-pass filters. A low-pass filter passes signals
within a specified band of frequencies while with constant amplitude from DC up to a fre-
attenuating those signals that fall outside of this quency fc referred to as the cut-off frequency.
band. Passive filters utilize passive components, The signal output above fc is attenuated. This is
namely, resistors, capacitors and inductors, while shown in Fig. 11.1 where the magnitude of the
active filters contain passive as well as active output voltage of the ideal (solid line) and practi-
components such as transistors and operational cal (dashed line) responses are plotted against the
amplifiers. Passive filters have the advantage of frequency. The ideal response shows the pass-
not requiring an external power supply as do band, the range of transmitted frequencies, and
active filters. However, they often utilize the stop-band, the range of attenuated frequencies
inductors which tend to be bulky and costly, and the cut-off or break frequency fc. In the prac-
whereas active filters utilize mainly resistors and tical response, fc is the frequency at which the
capacitors. Additionally, active filters can pro- magnitude of the output voltage falls to 0.707 of
duce signal gain and have high input and low its low-frequency value which is 3 dB. Here
output impedances which allow simple cascading fc ¼ 1 Hz for the normalized response.
of systems with little or no interaction between A high-pass filter attenuates signals from DC
stages. This chapter discusses the principles of up to a frequency fc while passing all signals with
active filter operation and treats with several constant amplitude above fc. The normalized
types and configurations of active filters. At the response is shown in Fig. 11.2 where the ideal
end of the chapter, the student will be able to: and practical responses with pass-band and stop-
band are indicated. A band-pass filter passes a
• Understand the principles of operation of the band of frequencies, while frequencies outside
basic types of filters of that band are attenuated. This is shown in
• Explain the operation of a range of active Fig. 11.3 where the ideal and practical responses
filters are indicated. A band-stop filter stops a band of
• Design a range of active filters frequencies while passing all frequencies outside
that band. It performs in a manner that is exactly
opposite to the band-pass filter. The band-stop
response is shown in Fig. 11.4 with ideal and
11.1 Introduction to Filters practical responses. Finally, an all-pass filter
passes all frequencies with a constant amplitude
There are five basic types of filters: low-pass, output but with a changing phase. This is shown
high-pass, band-pass, band-stop (or notch) and in Fig. 11.5.
# Springer Nature Switzerland AG 2021 411
S. J. G. Gift, B. Maundy, Electronic Circuit Design and Application,
https://doi.org/10.1007/978-3-030-46989-4_11
412 11 Active Filters
Vo 1=sC 1
11.2 Basic First-Order LP Filter Af ¼ ¼ ¼
V i R þ 1=sC 1 þ jωRC
We first consider a low pass of the simplest kind ¼
1
ð11:1Þ
shown in Fig. 11.6(a). It is a non-inverting 1 þ jf = f c
low-pass active filter using an RC network feeding
where
into a unity-gain operational amplifier. A resistor
Ros is sometimes included in the feedback loop of f c ¼ 1=2πRC ð11:2Þ
the op-amp in order to minimize DC offset at the
output as shown in Fig. 11.6(b). For the circuit,
11.2 Basic First-Order LP Filter 413
0.5
0
0
-45
Phase (deg)
-90
-135
-180
10-2 10-1 100 101 102
Frequency (Hz)
(a) (b)
(a) (b)
1 þ RR21
A f ðsÞ ¼ ð11:12Þ
1 þ RCs
Comparing coefficients in (11.11) and (11.12)
Fig. 11.10 Solution for Example 11.2 gives
R2
Example 11.2 Ao ¼ 1 þ ð11:13Þ
R1
Design a first-order non-inverting low-pass filter
with a gain of 10 and a cut-off frequency of 5 kHz a1 =2π f c ¼ RC ð11:14Þ
using Fig. 11.8.
The design steps are as follows:
1. Select C usually between 100 pF and 1 μF.
Solution Choose C ¼ 0.01 μF. Then from
2. Set the coefficient a1 ¼ 1.
fc ¼ 1/2πRC, resistor R is found to be R ¼ 1/
3. Determine R using R ¼ a1/2πfcC.
(2 π 5 103 0.01 106) ¼ 3.2 k. Since
4. Select R1 and find R2 using R2 ¼ R1(Ao 1).
1 + R2/R1 ¼ 10, for R1 ¼ 1 k, then R2 ¼ 9 k. The
solution is shown in Fig. 11.10.
For the first-order inverting low-pass filter, the
basic first-order system can be written as
Design Procedure
In preparation for the discussion on higher-order Ao
AðsÞ ¼ ð11:15Þ
filters and their realization, we wish to develop a 1 þ a1 ðs=ωc Þ
more ordered procedure for designing first-order
filters. Thus, the basic first-order system can be The transfer function for the actual system is
written as given by equation (11.16):
R2
Ao
AðsÞ ¼ ð11:11Þ A f ðsÞ ¼ R1
ð11:16Þ
1 þ a1 ðs=ωc Þ 1 þ R2 Cs
where Ao ¼ k is the low-frequency gain, ωc ¼ 2πfc Comparing coefficients in (11.15) and (11.16)
is the cut-off frequency and here a1 ¼ 1. For gives
416 11 Active Filters
(a) (b)
Ao
AðsÞ ¼ ð11:36Þ
1 þ a1 ðs=ωc Þ þ b1 ðs=ωc Þ2
yields
Ao ¼ k ð11:37Þ
b1 =ωc 2 ¼ R2 C 2 ð11:39Þ
Fig. 11.15 Sallen-Key second-order low-pass filter pffiffiffi
with gain For a Butterworth filter, a1 ¼ 2 and b1 ¼ 1.
Hence from (11.39),
characteristics unaffected. This can be seen by ωc ¼ 1=RC, f c ¼ 1=2πRC ð11:40Þ
simply substituting αR1, αR2 and C1/α, C2/α in
Eqs. (11.30) and (11.32) to see that the equations and from (11.38)
are unchanged. Also, the system can be designed pffiffiffi
2¼3k ð11:41Þ
for gain as shown in Fig. 11.15 where
k ¼ 1 þ Rb =Ra ð11:33Þ giving
pffiffiffi
The process of developing the associated k ¼3 2 ¼ 1:6 ð11:42Þ
design procedure follows that used for the unity-
For minimum offset
gain case.
Ra Rb R
R1 þ R2 ¼ Ra ==Rb ¼ ¼ b ð11:43Þ
Example 11.4 Ra þ Rb k
Develop a design procedure for the VCVS
Hence
Butterworth second-order low-pass filter with
gain for the special case when R1 ¼ R2 and Rb ¼ kðR1 þ R2 Þ ¼ 3:2R ð11:44Þ
C1 ¼ C2.
Also
Solution Using nodal analysis, the transfer func-
Ra Rb ðk 1Þ
tion for the filter with gain is given by R1 þ R2 ¼ Ra ==Rb ¼ ¼ Ra
Ra þ Rb k
Af ¼
Vo ð11:45Þ
Vi
k Hence
¼
1 þ sðC 1 R1 þ C 1 R2 þ C 2 R1 kC2 R1 Þ þ s2 R1 R2 C 1 C 2
kðR1 þ R2 Þ
ð11:34Þ Ra ¼ ¼ 3:2 R=0:6 ¼ 5:3 R ð11:46Þ
k1
For the case where R1 ¼ R2 ¼ R and C1 ¼ C2 ¼ C, These equations enable the determination of Ra
then (11.34) reduces to and Rb to satisfy both the gain requirement and
the minimum offset requirement. We can now
Vo k
Af ¼ ¼ ð11:35Þ state a design procedure for a specific cut-off
V i 1 þ sð3 kÞCR þ s2 R2 C2
frequency:
Comparing coefficients with (11.19)
420 11 Active Filters
Design Procedure 1
Aðjf Þ ¼ pffiffiffi 2 ð11:48Þ
1þj 2 f
þ j f
1. Select C1 ¼ C2 ¼ C usually between 100 pF fc fc
and 1 μF.
In the general second-order transfer function
2. Determine R ¼ 1/2πfcC.
(11.23) of Fig. 11.14, if we let R1 ¼ R2 ¼ R and
3. Set R1 ¼ R2 ¼ R.
C1 ¼ C2/2 ¼ C, then for s ¼ jω, the transfer
4. Find Ra and Rb using (11.46) and (11.44).
function (a) reduces to
Simplified Design Procedure for Second-Order Vo 1
ðjωÞ ¼ ð11:49Þ
Low-Pass Unity-Gain Butterworth Filter Vi 1 þ 2RCjω þ 2R2 C 2 ðjωÞ2
A simplified design procedure for the unity-gain
second-order Butterworth filter can be developed. Substituting ω ¼ 2πf and
Consider the transfer function (11.33) for a pffiffiffi
low-pass second-order filter: f c ¼ 1=2 2πRC ð11:50Þ
Ao yields
AðsÞ ¼ ð11:47Þ
2
1 þ a1 ðs=ωc Þ þ b1 ðs=ωc Þ Vo 1
ðjf Þ ¼ pffiffiffi 2 ð11:51Þ
Vi
1þj 2 f
þ j f
For unity gain, Ao ¼ 1 and forpaffiffiffi Butterworth fc fc
Clearly, transfer function (11.51) is identical to Solution Choose C1 ¼ 0.01 μF and then
pffiffiffi
transfer function (11.48), the transfer function of C2 ¼ 0.02 μF. Using (11.36), R ¼ 1=2 2π
a second-order unity-gain Butterworth filter. 10 103 0:01 106 ¼ 1:1 k and hence
Hence a simplified design procedure for a unity- R1 ¼ R2 ¼ 1.1 k. Ros ¼ 1.1 k 2 ¼ 2.2 k.
gain second-order Butterworth filter is as follows:
1. Choose the desired cut-off frequency fc. Example 11.6
2. Choose C1, usually an available value between Design a VCVS second-order unity-gain
about 100 pF and 1 μF. low-pass Chebyshev filter with a cut-off fre-
3. Let C2 ¼ 2C1. quency of 10 kHz and a ripple width
pffiffiffi
4. Determine R using R ¼ 1=2 2πfC 1 and then RWdB ¼ 2 dB.
set R1 ¼ R2 ¼ R.
5. For minimum offset use Ros ¼ 2R. Solution For this case, fc ¼ 10 kHz. Select
C1 ¼ 0.01 μF. From Table 11.2, the coefficients
Example 11.5 for RWdB ¼ 2 dB are a1 ¼ 1.1813, b1 ¼ 1.7775.
Using the configuration shown in Fig. 11.14, Then from (11.31), C2 (4 1.7775
re-design the second-order Butterworth filter of 0.01 106)/(1.1813)2 ¼ 0.05 μF. Select
Example 11.5 using the simplified approach. C2 ¼ 0.1 μF. Hence from (11.30) R1 ¼
422 11 Active Filters
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
1:1813þ ð1:1813Þ2 ð41:77750:01106 =0:1106 Þ
4π10103 0:01106
¼
1598 Ω and from (11.32) R2 ¼
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
1:1813 ð1:1813Þ2 ð41:77750:01106 =0:1106 Þ
4π10103 0:01106
¼
282 Ω. Finally, for minimum offset,
Ros ¼ 1598 + 282 ¼ 1880 Ω.
Example 11.7
Design a VCVS second-order unity-gain Fig. 11.16 Multiple feedback low-pass filter
low-pass Bessel filter with a cut-off frequency of
10 kHz. Noting that Vb ¼ 0 because node b is a virtual
earth and eliminating Va, we get
Solution For this case, fc ¼ 10 kHz. Select R2
Vo
C1 ¼ 0.01 μF. From Table 11.1, the Bessel ¼ R1
Vi 1 þ C 1 R2 þ R3 þ R1 s þ C1 C2 R2 R3 s2
R2 R3
coefficients are a1 ¼ 1.3617, b1 ¼ 0.6180.
Then C2 (4 0.6180 0.01 106)/
ð11:54Þ
(1.3617)2 ¼ 0.01333 μF. Select C2 ¼
0.02 μF. Hence R1 Comparing the coefficients of (11.54) with the
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1:3617þ ð1:3617Þ ð40:61800:0110 =0:0210 Þ
2 6 6
general second-order low-pass transfer function
¼ 4π10103 0:01106
¼ 1700 Ω and R2 ¼ AðsÞ ¼
Ao
ð11:55Þ
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1:3617 ð1:3617Þ2 ð40:61800:01106 =0:02106 Þ 1 þ a1 ðs=ωc Þ þ b1 ðs=ωc Þ2
4π10103 0:01106
¼
458 Ω. Finally, for minimum offset, (where a negative sign is introduced to account
Ros ¼ 1700 + 458 ¼ 2158 Ω. for the inversion produced by the MFB configu-
ration) yields
Ao ¼ R2 =R1 ð11:56Þ
11.3.2 Low-Pass Multiple Feedback
Topology a1
¼ C1 R2 þ R3 þ
R2 R3
ð11:57Þ
ωc R1
Another available topology for implementation of
second-order low-pass filters is the multiple feed- b1
¼ C 1 C 2 R2 R3 ð11:58Þ
back (MFB) topology shown in Fig. 11.16. In this ωc 2
circuit there are multiple feedback paths, and the Solving for R1,R2, R3 in terms of C1 and C2, we
active element is operating in a high-gain mode. get
The amplifier introduces 180 of phase shift in pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
addition to that introduced by the network a1 þ a1 2 4b1 ð1 þ Ao ÞC 1 =C 2
R1 ¼ ð11:59Þ
elements. Assuming an ideal op-amp as the active 4π f c C 1 Ao
element, then for nodes a and b,
In order to ensure a positive value under the
Vi Va V Vb Va Vo square root and hence a real solution for R1, then
¼ V a sC 2 þ a þ
R1 R3 R2
ð11:52Þ 4b1 ð1 þ Ao Þ
C2 C1 ð11:60Þ
a1 2
Va Vb
¼ ðV b V o ÞsC 1 ð11:53Þ Using (11.59) in (11.56) and (11.58), we get
R3
11.4 Higher-Order Low-Pass Filters 423
R3 ¼ 4π2 b1
2 . analysed and used in filter implementation. In this
f c C 1 C 2 R2
5. Scale values if necessary. chapter, the factorizing approach is pursued in the
development of higher-order filters.
Example 11.8 Therefore, the approach adopted here in
Design a MFB second-order low-pass realizing higher-order low-pass filters is to cas-
Butterworth filter with a gain of 10 and a cut-off cade first- and/or second-order filters in order to
frequency of 10 kHz. approximate the higher-order filter. Thus, in order
to realize a fourth-order filter, two second-order
Solution For this case, fc ¼ 10 kHz. Select filters are cascaded. In order to realize an odd-
C1 ¼ 0.001 μF. From Table 11.1, the order filter, at least one first-order must be cas-
Butterworth coefficients are a1 ¼ 1.4142, caded with one or more second-order filters. For
example, in order to synthesize a third-order filter,
b1 ¼ 1. Then C 2 C 1 4b1 ða1þA oÞ
1
2
one first-order and one second-order filters are
C2 0.001 106 4 1 (1 + 10)/ cascaded. Similarly, if a fifth-order filter is
pffiffiffi2
2 ¼ 0:022 μF . Select C2 ¼ 0.1 μF. Hence required, then one first-order and two second-
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
pffiffi 2 order filters must be cascaded.
1:4142þ ð 2Þ ð41110:001106 =0:1106 Þ
R1 ¼ The transfer function of one stage is given by
4π10103 0:001106 10
¼ 2.12 k, R2 ¼ AoR1 ¼ 2.12 k and R3 ¼ Ao
AðsÞ ¼ ð11:63Þ
1
¼ 119.5 Ω. 1 þ a1 ðs=ωc Þ þ b1 ðs=ωc Þ2
4πð104 Þ 0:001106 0:1106 21:2103
2
Ao
AðsÞ ¼
∏i 1 þ ai ðs=ωc Þ þ bi ðs=ωc Þ2
Ao
¼
2
1 þ a1 ðs=ωc Þ þ b1 ðs=ωc Þ 1 þ a2 ðs=ωc Þ þ b2 ðs=ωc Þ2 . . . 1 þ an ðs=ωc Þ þ bn ðs=ωc Þ2
ð11:65Þ
The multiplication of the factors in the denom- first- and second-order stages is shown diagram-
inator of the transfer function yields an nth-order matically in Fig. 11.17 up to the sixth order. As
polynomial where n represents the order of the can be seen, a filter with even order comprises
filter. The filter coefficients ai and bi determine only second-order stages, while a filter with odd
the filter characteristics such as Butterworth, order has a first-order stage included.
Chebyshev and Bessel, the types being consid-
ered in this discussion. These are tabulated in
Table 11.1 up to order 10. This cascading of
Fig. 11.17 Cascading first- and second-order stages to realize higher-order filters
11.4 Higher-Order Low-Pass Filters 425
Partial Filter 1
The first-order system is
1
AðsÞ ¼ ð11:68Þ
1 þ a1 ðs=ωc Þ
Fig. 11.21 Completed third-order low-pass Butterworth filter for Example 11.21
The transfer function for the second-order non- For this case, fc ¼ 20 kHz. Select C3 ¼ 1 nF.
inverting unity-gain low-pass filter in Fig. 11.20 From Table 11.1, a1 ¼ 1, b1 ¼ 0. Hence
is given by R3 ¼ a1/2πfcC3 ¼ 1/2π 20 103 109 ¼ 8 k.
1
Af ¼ Partial Filter 2
1 þ sC 1 ðR1 þ R2 Þ þ s2 R1 R2 C 1 C 2
For this case, fc ¼ 20 kHz. Select C1 ¼ 0.001 μF.
ð11:72Þ From Table 11.1, a2 ¼ 1, b2 ¼ 1. Then
C2 4b2C1/a22 ¼ 4 1 0.001 106 ¼
Comparing coefficients for (11.63) and (11.62)
0.004 μF. Select C2 ¼ 0.005 μF. Hence R1 ¼
yields pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1þ 1ð410:001106 =0:005106 Þ
¼ 5.8 k and R2
a2 =ωc ¼ C 1 ðR1 þ R2 Þ ð11:73Þ p
3
0:001106
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4π2010
1 1ð410:001106 =0:005106 Þ
¼ ¼ 2.2 k.
b2 =ωc 2 ¼ R1 R2 C 1 C 2 ð11:74Þ 4π20103 0:001106
Finally, for minimum offset, Ros ¼ 5.8 k + 2.2 k
From this, the design steps are as follows: ¼ 8 k.
1. Select C1 usually between 100 pF and 1 μF.
2. For the specific type of filter (Butterworth, The completed filter is shown in Fig. 11.21.
Chebyshev, Bessel) and the specific filter
requirements, obtain the coefficients a2 and Simplified Design Procedure for Third-Order
b2 from the relevant table. Low-Pass Unity-Gain Butterworth Filter
3. Determine C2 using C2 4b2C1/a22. A simplified design procedure for the unity-gain
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
a þ a2 2 4b2 C 1 =C 2 third-order Butterworth filter can be developed.
4. Find R1 and R2 using R1 ¼ 2 4π f c C 1
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Consider the transfer function (11.75) for a
a2 a2 2 4b2 C 1 =C 2
and R2 ¼ . low-pass third-order filter:
4π f C1 c
Partial Filter 1
11.4 Higher-Order Low-Pass Filters 427
This transfer function can be implemented using Butterworth filter with a cut-off frequency of
the configuration shown in Fig. 11.18 which com- 9 kHz.
prises cascaded first- and second-order systems.
For this system, the transfer function is given by Solution Choose C1 ¼ 0.005 μF and then
C3 ¼ 2C1 ¼ 0.01 μF and C2 ¼ 2C3 ¼ 0.02 μF.
1 1
Af ¼ : Using (11.70), R ¼ 1/2πfcC3 ¼ 1/
1 þ sR3 C 3 1 þ sC 1 ðR1 þ R2 Þ þ s2 C 1 C 2 R1 R2
2π 9 103 0.01 106 ¼ 1.8 k. Hence
ð11:77Þ R1 ¼ R2 ¼ R3 ¼ 1.8 k.
Let R1 ¼ R, R2 ¼ R, R3 ¼ R and C1 ¼ C3/2 and
C2 ¼ 2C3. Then Fig. 11.18 becomes Fig. 11.22 Example 11.11
and (11.77) becomes Design a VCVS fourth-order unity-gain low-pass
Chebyshev filter with a cut-off frequency of
1 1 20 kHz and 1 dB ripple width.
Af ¼ :
1 þ sRC 3 1 þ sC 3 R þ s2 ðC 3 RÞ2
1 1 Solution The design of each of the two partial
¼ : ð11:78Þ filters is considered separately.
1 þ jf = f c 1 þ jf = f c þ ðjf = f c Þ2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0:3039þ ð0:3039Þ2 ð41:16970:001106 =0:1106 Þ For this case, fc ¼ 50 kHz. Select C1 ¼ 820
¼ 4π20103 0:001106 pF. From Table 11.1, a2 ¼ 1.6180, b2 ¼ 1. Then
¼ 2 k and R2 ¼ C2 4b2C1/a22 ¼ 4 1 820 1012/
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0:3039 ð0:3039Þ2 ð41:16970:001106 =0:1106 Þ (1.618)2 ¼ 1.26 nF. Select C2 ¼ 1.5 nF. Hence
4π20103 0:001106
¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1:618þ ð1:618Þ2 ð410:820109 =1:5109 Þ
360 Ω. Finally, for minimum offset, Ros ¼ 2 k + R1 ¼ 4π50103 0:820109
360 Ω ¼ 2360 Ω. ¼ 4.4 k and R2 ¼
The completed filter is shown in Fig. 11.23. pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1:618 ð1:618Þ ð410:82010 =1:510 Þ
2 9 9
4π50103 0:820109
¼ 1.9 k.
Example 11.12 Finally, for minimum offset, Ros ¼ 4.4 k + 1.9 k
Design a VCVS fifth-order unity-gain low-pass ¼ 6.3 k.
Butterworth filter with a cut-off frequency of
50 kHz. Partial Filter 3
For this case, fc ¼ 50 kHz. Select C1 ¼ 330
Solution The design of each of the three partial pF. From Table 11.1, a3 ¼ 0.6180, b3 ¼ 1. Then
filters is considered separately. C2 4b3C1/a32 ¼ 4 1 330 1012/
(0.6180)2 ¼ 3.5 nF. Select C2 ¼ 4.7 nF. Hence
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Partial Filter 1 0:618þ ð0:618Þ2 ð410:330109 =4:7109 Þ
R1 ¼
For this case, fc ¼ 50 kHz. Select C ¼ 1 nF. From 4π50103 0:330109
Table 11.1, a1 ¼ 1, b1 ¼ 0. Hence R ¼ a1/ ¼ 4.5 k and R2 ¼
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2πfcC ¼ 1/2π 50 103 109 ¼ 3.2 k. 1:618 ð1:618Þ ð410:33010 =4:710 Þ
2 9 9
4π50103 0:330109
¼ 1.5 k.
Partial Filter 2
11.5 High-Pass First-Order Filter-Butterworth Response 429
Finally, for minimum offset, Ros ¼ 4.5 k + 1.5 k included in the feedback loop in order to reduce
¼ 6 k. DC offset at the output as shown in Fig. 11.25(b).
The completed filter is shown in Fig. 11.24. For the circuit
Vo R sCR
Af ¼ ðsÞ ¼ ¼
Vi R þ 1=sC 1 þ sCR
11.5 High-Pass First-Order jf = f c
Filter-Butterworth Response ¼ ð11:80Þ
1 þ jf = f c
High-pass filters can be generated from low-pass where
filters by changing resistors to capacities and
capacities to resistors in the filter network. This f c ¼ 1=2πRC ð11:81Þ
corresponds to the transformation s ! 1/s or
Note that for f fc, |Af| ! 1 and for f fc, |
jð f = f c Þ ! jð f =1 f Þ . Thus, the first-order low-pass
c Af| ! 0. The frequency response plot for this
filter given by 1
1þjf = f c is transformed such that circuit is shown in Fig. 11.26, where the cut-off
jf = f c
1
1þjf = f c ! 1
1þjf =1f
¼ 1þjf = f c . A basic first-order frequency is fc ¼ 1/2πRC. At f ¼ fc, the filter
c
transfer function becomes
high-pass filter is shown in Fig. 11.25(a). It
employs a CR network to provide the high-pass jf = f c j
filtering, the output of which feeds into a unity- A f ðf cÞ ¼ ¼ ð11:82Þ
1 þ jf = f c 1 þ j
gain buffer. It is obtained from the first-order
low-pass filter by interchanging R and C in the Hence, A f ð f c Þ ¼ j ¼ p1ffiffi ¼ 0:707 ¼
1þj 2
filter network. A resistor ROS ¼ R may be
3 dB. Thus, at the break frequency f ¼ fc the
magnitude response falls by 3 dB from its
mid-band value as shown in Fig. 11.26.
Example 11.13
Design an RC high-pass first-order Butterworth
filter with a cut-off frequency of 1 kHz.
The high-pass first-order filter with gain is shown Hence the transfer function is given by
in Fig. 11.27 where the active device is modified R2 sCR1 jf = f c
to produce a gain greater than unity. The transfer Af ¼ : ¼ k ð11:85Þ
R1 1 þ sCR1 1 þ jf = f c
function is given by
where k ¼ R2/R1 is the gain and
Vo kjf = f c
Af ¼ ¼ ð11:83Þ
V i 1 þ jf = f c f c ¼ 1=2πR1 C ð11:86Þ
Design Procedure
As we did for the low-pass filters, we wish to
write down an ordered procedure for designing
first-order high-pass filters. Thus, the basic first-
order system can be written as
A1
AðsÞ ¼ ð11:87Þ
1 þ a1 =ðs=ωc Þ
(a) (b)
For the first-order non-inverting high-pass As in the case of second-order low-pass filters, a
filter, the transfer function is given by (see earlier) high-pass filter can be optimized to meet at least
one of the following three requirements:
1 þ RR21 (i) maximum flatness in the pass-band;
A f ðsÞ ¼ ð11:89Þ
s þ 1=RC (ii) maximum sharpness in the transition from
pass-band to stop-band; and (iii) linear phase
Comparing coefficients in (11.89) and (11.88) response. In order to allow this optimization of
gives the filter, the transfer function must contain com-
R2 plex poles and hence, for a second-order filter,
A1 ¼ 1 þ ð11:90Þ can take the following form:
R1
1 A1 s2
a 1 ωc ¼ ð11:91Þ AðsÞ ¼ ð11:96Þ
RC s2 þ a 1 ωc s þ b 1 ωc 2
The design steps are as follows: where A1 is the low-frequency gain in the pass-
1. Select C usually between 100 pF and 1 μF. band and a1 and b1 are positive real filter
2. Set the coefficient a1 ¼ 1. coefficients that define the location of the complex
3. Determine R using R ¼ 1/a12πfcC. poles of the filter and hence the characteristics of the
4. Select R1 and find R2 using R2 ¼ R1(A1 1). filter. The angular frequency ωc ¼ 2πfc represents
the cut-off frequency of the filter. This transfer
For the first-order inverting low-pass filter, the function can be normalized by setting ωc ¼ 1 giving
transfer function is given by (see earlier)
A1 s2
AðsÞ ¼ ð11:97Þ
1 þ a1 s þ b1
R2
sR1
s2
AðsÞ ¼ ð11:92Þ
s þ 1=R1 C 1
Only the Butterworth, Chebyshev and Bessel
Comparing coefficients in (11.92) with the responses are considered. These filter responses
general transfer function occur for second (and higher) order and not for
first-order where complex poles are not possible
AðsÞ ¼
A1 s
ð11:93Þ and therefore all three filter types have identical
s þ a1 ω c responses for a first-order high-pass filter. The
amplitude responses of first-, second- and
gives equation (11.94):
higher-order high-pass Butterworth filters using
R2 the normalized characteristic are shown in
A1 ¼ ð11:94Þ
R1 Fig. 11.29. The second-order high-pass filter
transfer function (11.97) can be realized around
1
a 1 ωc ¼ ð11:95Þ a single active device using two topologies, the
R1 C1
Sallen-Key and the multiple feedback.
The design steps are as follows:
1. Select C1 usually between 100 pF and 1 μF.
2. Set the coefficient a1 ¼ 1.
3. Determine R1 using R1 ¼ 1/a12πfcC1.
432 11 Active Filters
s2 C 1 C 2 R1 R2
A ðsÞ ¼
1 þ sR2 ðC 1 þ C 2 Þ þ s2 R1 R2 C 1 C 2
ð11:100Þ
s2
A f ðsÞ ¼ ð11:101Þ
s2 þ 2
R1 C s þ C2 R1 R
1 2
Fig. 11.30 Second-order high-pass unity-gain filter
Comparing coefficients for (11.101) and
(11.96) yields
11.6.1 Sallen-Key or A1 ¼ 1 ð11:102Þ
Voltage-Controlled Voltage
Source (VCVS) Topology 2
a 1 ωc ¼ ð11:103Þ
R1 C
The circuit shown in Fig. 11.30 is the VCVS or 1
Sallen-Key implementation of a second-order high- b1 ω c 2 ¼ ð11:104Þ
R1 R2 C2
pass unity-gain filter. Comparing this with the
second-order low-pass filter, it can be seen that the From (11.103),
resistors and capacitors of the filter network are
1
interchanged. Using KCL at nodes A and B, we have R1 ¼ ð11:105Þ
a1 π f c C
ðV i V A ÞsC1 ¼ ðV A V B ÞsC2 þ ðV A V o Þ=R2
and from (11.104),
ð11:98Þ
a1
R2 ¼ ð11:106Þ
ðV A V B ÞsC 2 ¼ V B =R1 ð11:99Þ b1 4πfC
where VB ¼ Vo. These equations yield We can now state a design procedure for a
specific cut-off frequency:
11.6 High-Pass Second-Order Filter 433
yields
A1 ¼ k ð11:111Þ
Example 11.15
a1 ωc ¼ ð3 kÞCR ð11:112Þ
Design a VCVS second-order unity-gain high-
pass Butterworth filter with a cut-off frequency b 1 ωc 2 ¼ R 2 C 2 ð11:113Þ
of 10 kHz.
pffiffiffi
For a Butterworth filter, a1 ¼ 2 and b1 ¼ 1.
Solution For this case, fc ¼ 10 kHz. Select C1 ¼ Hence from (11.113),
pffiffiffi
C2 ¼ 0.01 μF. From Table 11.1, a1 ¼ 2, b1 ¼
ωc ¼ 1=RC, f c ¼ 1=2πRC ð11:114Þ
1: Then R1 ¼ π10103 p1ffiffi20:01106 ¼ 2.3 k and
pffiffi
R2 ¼ 4π10103 0:0110
2
6 ¼ 1.1 k. Finally, for and from (11.112)
minimum offset, Ros ¼ 2.3 k. pffiffiffi
2¼3k ð11:115Þ
The system can be designed for gain as shown
in Fig. 11.31. where giving
Rb pffiffiffi
k ¼1þ ð11:107Þ k ¼3 2 ¼ 1:6 ð11:116Þ
Ra
For minimum offset
The process of developing the associated
design procedure follows that used for the unity-
gain case.
Example 11.16
434 11 Active Filters
Ra Rb R s2 C 1 C 2 R1 R2
R ¼ Ra ==Rb ¼ ¼ b ð11:117Þ Af ¼
Ra þ Rb k 1 þ sR2 ðC 1 þ C 2 Þ þ s2 R1 R2 C1 C2
Hence ð11:123Þ
Solution For this case, fc ¼ 10 kHz. Select For simplicity of the analysis, let C1 ¼ C3 ¼ C.
C1 ¼ C2 ¼ 0.001 μF. From Table 11.2, the Then (11.128) reduces to
coefficients for RWdB ¼ 1 dB are a1 ¼ 1.3022, C 2
b1 ¼ 1.5515. Then from (11.105) and (11.106), C2 s
A f ðsÞ ¼ ð11:129Þ
R1 ¼ π10103 1:30220:00110
1
6 ¼ 24 k and R2 ¼ s2 þ 2CþC 2
R2 C 2 C s þ R1 R21C2 C
1:3022
¼ 6.7 k. Finally, for
4π10103 1:55150:001106
minimum offset, Ros ¼ 24 k. Comparing the coefficients of (11.129) with
the general second-order high-pass transfer
Example 11.20 function
Design a VCVS second-order unity-gain high- A1 s2
pass Bessel filter with a cut-off frequency of AðsÞ ¼ ð11:130Þ
s2 þ a1 ω c s þ b1 ω c 2
10 kHz.
(where a negative sign is introduced to account
Solution For this case, fc ¼ 10 kHz. Select for the inversion produced by the MFB configu-
C1 ¼ C2 ¼ 0.001 μF. From Table 11.1, the Bessel ration) yields
coefficients are a1 ¼ 1.3617, b1 ¼ 0.6180. Then
C
R1 ¼ π10103 1:36170:00110
1
6 ¼ 23.4 k and R2 ¼
A1 ¼ ð11:131Þ
C2
1:3617
¼ 17.5 k. Finally, for
4π10103 0:61800:001106
C 2 þ 2C
minimum offset, Ros ¼ 23.4 k. a 1 ωc ¼ ð11:132Þ
R2 C2 C
1
b 1 ωc 2 ¼ ð11:133Þ
R1 R2 C 2 C
11.6.2 High-Pass Second-Order
Multiple Feedback Filter Solving for R1, R2 in terms of C and C2, we get
a1
Another available topology for implementation of R1 ¼ ð11:134Þ
b1 2π f c ðC 2 þ 2C Þ
second-order high-pass filters is the multiple feed-
back (MFB) topology shown in Fig. 11.32. In
this circuit the resistors and capacitors are
interchanged compared to the corresponding
low-pass MFB circuit.
Assuming an ideal op-amp, then for nodes a
and b,
ðV i V a ÞsC 1 ¼ ðV a V o ÞsC 2 þ ðV a V b Þ
Va
sC 3 þ ð11:126Þ
R1
ðV b V o Þ
ðV a V b ÞsC 3 ¼ ð11:127Þ
R2
Fig. 11.32 High-pass second-order multiple feedback
Noting that Vb ¼ 0 because node b is a virtual
filter
earth and eliminating Va, we get
436 11 Active Filters
ðC2 þ 2C Þ
R2 ¼ ð11:135Þ
a1 2π f c C 2 C
Design Procedure
1. Select C usually between 100 pF and 1 μF.
2. Determine C2 using C2 ¼ C/A1.
3. For the specific type of filter (Butterworth,
Chebyshev, Bessel) and the specific filter Fig. 11.33 Third-order high-pass filter
requirements, obtain the coefficients a1 and
b1 from the relevant table. A1 s
4. Find R1 and R2 using R1 ¼ b1 2π f aðC1 2 þ2CÞ and AðsÞ ¼ ð11:137Þ
c s þ a 1 ωc
R2 ¼ að1C2π2 þ2C Þ
f C2 C.
c Therefore, the general transfer function
5. Scale values if necessary. representing this cascade of second- and first-
order transfer functions is given by
Example 11.21
Design a MFB second-order high-pass s2
A ðsÞ ¼ A 1 ∏ i 2
Butterworth filter with a gain of 10 and a cut-off s þ ai ω c s þ bi ω c 2
frequency of 1 kHz. s2 s2
¼ A1 2
s þ a1 ω c s þ b1 ω c 2 s2 þ a 2 ω c s þ b 2 ω c 2
Solution For this case, fc ¼ 1 kHz. Select C1
s2
¼ C3 ¼ C ¼ 0.01 μF. Then C2 ¼ 0.01 μF/10 ... 2 ð11:138Þ
s þ a n ωc s þ b n ωc 2
¼ 0.001 μF. From Table 11.1, the Butterworth
coefficients are a1 ¼ 1.4142, b1 ¼ 1. Then R1
¼ b1 2π f aðC1 2 þ2CÞ ¼ 2π103 ð0:001þ0:02
1:4142
Þ106
¼
c
ðC 2 þ2CÞ
10.7 k and R2 ¼ a1 2π f c C 2 C ¼ 11.7.1 Third-Order High-Pass
ð0:001þ0:02Þ10 6 Unity-Gain Filter
1:41422π103 0:001106 0:01106
¼ 236 k.
In order to develop a third-order unity-gain high-
pass filter, we cascade first- and second-order
11.7 Higher-Order High-Pass Filters filters such that
s s2
We come now to higher-order filters in which the AðsÞ ¼ A1 : 2
s þ a 1 ωc ð s þ a 2 ωc s þ b 2 ωc 2 Þ
transfer function has a denominator polynomial
that is of degree n. We again use the method of ð11:139Þ
factoring the higher-order transfer function A(s) Note again that b1 ¼ 0 for the first-order part of
into first- and second-order transfer functions the transfer function. For the unity-gain system
A1(s) and A2(s), each of which is then synthesized under consideration, A1 ¼ 1. Hence (11.139) can
and cascaded. be written as
The transfer function of one stage is given by
s s2
A1 s2 AðsÞ ¼ : 2
AðsÞ ¼ 2 ð11:136Þ s þ a 1 ω c ð s þ a 2 ωc s þ b 2 ωc 2 Þ
s þ a 1 ωc s þ b 1 ωc 2
ð11:140Þ
In a first-order transfer function, the coefficient
b1 is zero giving the transfer function The circuit of this third-order system is shown
in Fig. 11.33. It comprises a unity-gain
non-inverting first-order system cascaded with a
11.7 Higher-Order High-Pass Filters 437
s2 Partial Filter 1
AðsÞ ¼ ð11:144Þ For this case, fc ¼ 20 kHz. Select C3 ¼ 1 nF.
ðs2 þ a 2 ωc s þ b 2 ωc 2 Þ
From Table 11.1, a1 ¼ 1, b1 ¼ 0. Hence R3 ¼ 1/
The transfer function for the second-order non- 2πfca1C3 ¼ 1/2π 20 103 109 ¼ 8 k.
inverting unity-gain low-pass filter in Fig. 11.33
is given by (C1 ¼ C2 ¼ C) Partial Filter 2
s2
A f ðsÞ ¼ ð11:145Þ
s2 þ 2
R1 C s þ C2 R11 R2
For unity gain, A1 ¼ 1 and for a Butterworth Solution Choose C ¼ 0.01 μF and then C1 ¼
response from Table 11.1, a1 ¼ 1 and a2 ¼ 1, C2 ¼ C3 ¼ 0.01 μF. Using (11.154), R3 ¼ 1/
b2 ¼ 1. Then setting s ¼ jω, (11.150) becomes 2πfcC ¼ 1/2π 9 103 0.01 106 ¼
1.8 k. Hence R1 ¼ 2R3 ¼ 3.6 k and R2 ¼ R3/
ðjf = f c Þ2 jf = f c 2 ¼ 1.8 k/2 ¼ 900 Ω.
A¼ : ð11:151Þ
1 þ jf = f c þ ðjf = f c Þ2 1 þ jf = f c
Example 11.24
where fc is the cut-off frequency. This transfer Design a VCVS third-order unity-gain Bessel
function can be implemented using the configura- high-pass filter using Fig. 11.33 with a cut-off
tion shown in Fig. 11.33 which comprises cas- frequency of 1 kHz.
caded first- and second-order systems. For this
system, the transfer function is given by Solution The design of each of the two partial
filters is considered separately.
s2 C 1 C 2 R 1 R 2 sC 3 R3
Af ¼ :
1 þ sR2 ðC 1 þ C 2 Þ þ s2 C 1 C 2 R1 R2 1 þ sR3 C 3
Partial Filter 1
ð11:152Þ For this case, fc ¼ 1 kHz. Select C3 ¼ 0.1 μF.
From Table 11.1, a1 ¼ 0.756, b1 ¼ 0. Hence
Let C1 ¼ C2 ¼ C3 ¼ C and R1 ¼ 2R3 and
R3 ¼ 1/2πfca1C¼ 1/2π 103 0.756 0.1
R2 ¼ R3/2. Then (11.152) becomes
s2 ðCR3 Þ2 sCR3
Af ¼ :
1 þ sR3 C þ s2 ðCR3 Þ2 1 þ sCR3
ðjf = f c Þ2 jf = f c
¼ :
2 1 þ jf = f
1 þ jf = f c þ ðjf = f c Þ c
ð11:153Þ
where
106 ¼ 2.1 k.
11.7 Higher-Order High-Pass Filters 439
k
G¼ ð11:169Þ
3k
Eliminating VA yields
In the design process, a value of C is selected, s
and then using (11.167), R is determined for a Vo CR1
¼ 2 ð11:172Þ
given fo. Following this, either a particular Vi s þ 2s
CR3 þ CR2 R1 þR 2
1 R2 R3
Example 11.25
Design a Sallen-Key second-order band-pass fil-
ter with a centre frequency of 5 kHz and a quality
factor of 5.
Solving for R1, R2, R3 yields (In the case where the centre frequency gain G is
not specified, then G can be selected to ensure that
Q
R1 ¼ ð11:177Þ the realizability condition is satisfied with equality.
2π f o GC
Then resistor R2 can be removed from the circuit.)
2Q If the realizability condition is not met, then the
R3 ¼ ð11:178Þ
ωo C requirement C1 ¼ C2 would have to be relaxed.
1 G
G2 ¼ ¼ 4π f o CQ 1 2 ð11:179Þ
R2 2Q
11.7.5 Wien Band-Pass Filter
where G2 is conductance. This must be positive
and therefore the following realizability condition A third band-pass filter is shown in Fig. 11.42. It
must be satisfied: utilizes a Wien network which has a band-pass
response. At fo ¼ 1/2πRC, the output of the net-
G G
1 2 0) 1 ð11:180Þ work is at a maximum of 1/3 of the input voltage.
2Q 2Q2 By placing this network in a positive feedback
loop, the band-pass response is sharpened. The
If the realizability condition is satisfied such
transfer function is given by
that G/2Q2 ¼ 1, then G2 ¼ 0 which means
R2 ! 1(open circuit). In these circumstances,
resistor R2 is removed from the circuit.
Example 11.26
Design a second-order MFB band-pass filter with
f0 ¼ 1 kHz, Q ¼ 7 and G ¼ 10.
Q¼
1
ð11:183Þ where ωo ¼ 2πfo is the notch frequency and Q is
3 α RR23 the quality factor. The normalized response for
this filter is shown in Fig. 11.43 for varying
R2 values of Q.
G¼α Q ð11:184Þ
R1
Vo s2 þ ω o 2
¼ 2 ð11:186Þ
V i s þ s ωQo þ ωo 2
where
1 1
ωo ¼ , f ¼ ð11:187Þ
RC o 2πRC
1
Q¼ ð11:188Þ
4ð 1 ρÞ
Vo s 2 þ ωo 2
¼k 2 ð11:190Þ 11.8.3 All-Pass Filters
Vi s þ s ωQo þ ωo 2
1 ωso dϕ d
Vo t gr ¼ ¼ 2 tan 1 ω=ωo
ðsÞ ¼ ð11:195Þ dω dω
Vi 1 þ ωso
1=π f o
¼ ð11:199Þ
The amplitude response is given by 1 þ ð f = f o Þ2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
V o 1 s=ωo 1 þ ðω=ωo Þ2
¼ ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼1 This has a value
V i 1 þ s=ωo
1 þ ðω=ωo Þ2
t gr ¼ 1=π f o , f f o ð11:200Þ
ð11:196Þ
which is the maximum group delay occurring at
which is constant for all frequencies. Note that at low frequencies. A plot of tgr against f is shown in
low frequencies, Fig. 11.47.
Magnitude
0.5
0
0
-45
Phase (deg)
-90
-135
-180
0.01fo 0.1fo fo 10fo 100fo
Frequency (Hz)
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
V o ð1 sCRÞ 1 þ ðωCRÞ2
¼ ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 1
V i 1 þ sCR
1 þ ðωCRÞ2
ð11:210Þ
V HP RR1x s2
¼ R7 ð2R1 þRx Þ
ð11:218Þ
Vi s2 þ CRR s þ C21R2
1 ðR6 þR7 Þ
of general form
V HP ks2
¼ 2 ð11:219Þ
Vi s þ s ωQo þ ωo 2
Rx
of general form G¼ ¼ kQ ð11:227Þ
βð2R1 þ Rx Þ
V BP sωo G Q
¼ 2 ð11:221Þ
Vi s þ s ωQo þ ωo 2 Example 11.31
Using the state variable filter, design a band-pass
V LP RR21 C21R2 filter with a centre frequency fo ¼ 10 kHz, a
¼ R7 ð2R1 þRx Þ
ð11:222Þ
Vi s2 þ CRR s þ C21R2 quality factor Q ¼ 50 and a centre frequency
1 ðR6 þR7 Þ
gain G ¼ 10.
of general form
Solution Choosing C ¼ 0.001 μF, then R ¼ 1/
V LP kωo 2
¼ 2 ð11:223Þ 2π 10 103 0.001 106 ¼ 15.9 kΩ.
Vi s þ s ωQo þ ωo 2
G ¼ kQ giving 10 ¼ 50 k from which k ¼ 1/5.
This yields Rx ¼ 10 kΩ and R1 ¼ 50 kΩ. Finally,
From these for β ¼ R7/(R6 + R7),
Q ¼ 1/β(2 + k) which becomes 50 ¼ 1/β(2 + 0.2)
f 0 ¼ 1=2πRC ð11:224Þ which yields β ¼ 1/110. For R7 ¼ 1 kΩ then
R6 ¼ 109 kΩ.
k ¼ Rx =R1 ð11:225Þ
R1 1
Q¼ ¼ ð11:226Þ 11.9.1 Modified State Variable Filter
βð2R1 þ Rx Þ βð2 þ kÞ
We wish to exploit the excellent summing
properties of the virtual earth of an op-amp in the
inverting configuration. Thus, in the state variable
11.9 State Variable Filter 449
filter configuration
R in Fig. 11.50, instead of adding high-pass, band-pass and low-pass outputs.
the term a1 x(t)dt derived from the band-pass Moreover, the availability of quad op-amp
output at theR non-inverting terminal of the op-amp, packages such as the LM148 series or the superior
the term x(t)dt is inverted (sign changed) using OPA4134 facilitates the implementation of this
R
an inverting amplifier and the term a1 x(t)dt added system.
at the inverting terminal using the normal summing For this system, the transfer functions are as
process via the virtual earth. The resulting modified follows:
block diagram is shown in Fig. 11.52, and the
modified circuit is shown in Fig. 11.53. VN R1
VR1 s 2
þ 1
2
VR3 C 2
R1 R2 1
V LP VR1 VR2 C 2 VR3 2
11.10 Applications
¼ 2 ð11:231Þ
Vi s þ VR2RVR
2
3C
s þ VR 12 C2
3
Several filter applications are presented in this
For this system, section.
All-Pass Notch
Example 11.32 Figure 11.56 shows a notch filter made using
Design a universal filter with a cut-off frequency two first-order all-pass filters. The output of the
of 20 kHz, a quality factor of 100 and a centre second all-pass filter is V A ¼ 1Ts1þTs 1þTs V i ¼
: 1Ts
12TsþT 2 s2
V i where T ¼ RC. This output is Rx1 ¼ Rx2 ¼ 10 k. In the circuit the potentiome-
1þ2TsþT 2 s2
ter VR ¼ 1 k has been added to enable an
applied to one end of the potential divider adjustment for the deepest possible notch.
comprising two series connected equal value Finally, for a notch frequency fo ¼ 1 kHz,
resistors Rx1 ¼ Rx2 (Assume initially VR ¼ 0). choose C ¼ 0.01 μF and therefore R ¼ 1/2π
The input signal is applied at the other end 103 0.01 106 ¼ 15.9 k.
of these two resistors and op-amp 3 connected
as a buffer takes its input from the junction Ideas for Exploration (i) Replace fixed resistors
of these two resistors. The voltage across R by ganged variable resistors in series with fixed
these resistors is therefore V A V i ¼ resistors such that the notch frequency can be
continuously varied.
12TsþT 2 s2
2 2 1 Vi ¼
4Ts
2 2 V i . Half of
1þ2TsþT s 1þ2TsþT s
this is dropped across each resistor Rx1 and is Variable Wide-Band Filter
given by V Rx1 ¼ 1þ2TsþT
2Ts
2 2 Vi
s
. Therefore, The circuit in Fig. 11.57 is a variable wide-band
filter whose cut-off frequencies are made variable
the voltage Vo at the output Vo of op-amp 3 is
2 2
by including variable resistors in series with fixed
given by V o ¼ V i þ V Rx1 ¼ 1þ2Ts2TsþT
2 2
1þ2TsþT s
s
Vi ¼ resistors. Each filter uses equal value resistors and
capacitors and unity gain for simplicity. In order
1 þ T 2 s2
V i:
1 þ 2Ts þ T s 2 2 to realize a Butterworth response, the gain of each
op-amp must be set to 1.6. Such a response is
This goes to zero at fo ¼ 1/2πRC and is especially useful in audio systems as a scratch and
non-zero at all other frequencies. This represents rumble filter to remove unwanted high and low
a notch filter with notch frequency fo ¼ 1/2πRC frequencies. The Butterworth response ensures
and Q ¼ 1/2. We can choose RA ¼ 10 k and minimum amplitude distortion in the pass-band.
The dual-ganged potentiometers are 100 k linear 1 to 10. The circuit is assembled around a quad
type. op-amp such as the OPA2134.
Ideas for Exploration (i) Adjust the gains to 1.6 Ideas for Exploration (i) Use a bank of
for Butterworth response. (ii) Compare the per- switched capacitors to increase the frequency
formance of the 741 op-amp with the OPA134 range.
op-amp in this application.
Research Project 1
Universal Filter This research project involves the analysis and
The circuit in Fig. 11.58 is a universal filter based optimization of the variable frequency band-
on the modified sate variable filter. With the values pass filter in Fig. 11.59. It uses cascaded
selected (C ¼ 0.01 μF and the ganged potentiom- all-pass filters in a feedback loop such that at the
eter VR3 ¼ 100 k), the resonant frequency ( fo ¼ 1/ frequency where the phase shift is 180 , the feed-
2πCVR3) is variable from about 160 Hz to about back around the input op-amp goes to zero such
16 kHz, the quality factor (Q ¼ (VR2 + 1 k)/R2) is that the gain is at its maximum. At all other
variable from 0.1 to 100, and the centre frequency frequencies, the gain is reduced by the feedback.
gain (G ¼ R1/(VR1 + 1 k)) is variable from about The loop gain is kept less than unity to ensure
there is no oscillation. Potentiometer VR1 ¼ 20 k
Research Project 2
Fig. 11.61 Circuit for Question 7
This project involves research into and the design
of an active cross-over system used in audio
speaker systems. A diagrammatic representation
of such a system is shown in Fig. 11.60. This
system separates the high frequencies and the
low frequencies of a line level audio signal com-
ing from a preamplifier using a high-pass filter
and a low-pass filter, respectively. The output of
the high-pass filter drives a power amplifier which
delivers power to the tweeter in the speaker enclo- Fig. 11.62 Circuit for Question 12
sure, while the output of the low-pass filter drives
a separate power amplifier which drives the 4. Show that in a first-order low-pass filter, the
woofer in the speaker system. Such an approach amplitude response is down 3 dB at the
allows the signal separation to take place at low cut-off frequency and determine the phase
signal levels, thereby avoiding the use of the angle of the output relative to the input at
expensive components found in passive cross- that frequency.
over filters used in speaker systems. 5. Using the Sallen-Key configuration, design a
second-order low-pass Butterworth filter
Ideas for Exploration (i) Examine the introduc- with unity gain and a cut-off frequency of
tion of an active band-pass filter that separates the 50 kHz. Ensure that offset voltage is
mid-band frequencies and used to drive a third minimized.
power amplifier that delivers power to a 6. Show that in the Sallen-Key second-order
mid-range speaker. low-pass Butterworth filter, for the case
where the network resistors are equal and
the network capacitors are equal, the gain
Problems must be 1.6. Hence repeat the design of
Question 5 using this modified approach.
7. Show that in the Sallen-Key second-order
1. Design a first-order low-pass non-inverting unity-gain low-pass filter configuration
filter with unity gain and a break frequency shown in Fig. 11.61, if R1 ¼ R2 and
of 5 kHz. C2 ¼ 2C1, then the transfer function reduces
2. Design a first-order non-inverting low-pass to a second-order Butterworth response.
filter with gain of 8 and a cut-off frequency of 8. Design a VCVS second-order unity-gain
17 kHz. low-pass Chebyshev filter with a cut-off fre-
3. Design a first-order low-pass filter with gain quency of 22 kHz and a ripple width
of 12 and a cut-off frequency of 25 kHz using RWdB ¼ 1dB.
the inverting configuration.
454 11 Active Filters
9. Design a VCVS second-order unity-gain 19. Using the Sallen-Key configuration, design a
low-pass Bessel filter with a cut-off fre- second-order high-pass Butterworth filter
quency of 550Hz. with unity gain and a cut-off frequency of
10. Design a MFB second-order low-pass 5 kHz.
Butterworth filter with a gain of 8 and a 20. Show that in the Sallen-Key second-order
cut-off frequency of 3 kHz. high-pass Butterworth filter, for the case
11. Design a VCVS third-order unity-gain where the network resistors are equal and
low-pass Butterworth filter with a cut-off fre- the network capacitors are equal, the gain
quency of 33 kHz. What is the ultimate roll- must be 1.6. Hence repeat the design of
off rate of such a filter? Question 19 using this modified approach.
12. Show that in the Sallen-Key third-order 21. Show that in the Sallen-Key second-order
unity-gain low-pass filter configuration unity-gain high-pass filter configuration
shown in Fig. 11.62, if R1 ¼ R2 ¼ R3 and shown in Fig. 11.63, if R1 ¼ 2R2 and
C2 ¼ 2C3 and C1 ¼ C3/2, then the transfer C1 ¼ C2, then the transfer function reduces
function reduces to a third-order Butterworth to a second-order high-pass Butterworth
response. Hence repeat Question 11 using response.
this simplified design procedure. 22. Design a VCVS second-order unity-gain
13. Design a VCVS fourth-order unity-gain high-pass Chebyshev filter with a cut-off fre-
low-pass Chebyshev filter with a cut-off quency of 14 kHz and a ripple width
frequency of 12 kHz and 3 dB ripple width. RWdB ¼ 1dB.
14. Design a VCVS fifth-order unity-gain 23. Design a VCVS second-order unity-gain
low-pass Butterworth filter with a cut-off fre- high-pass Bessel filter with a cut-off fre-
quency of 1200 Hz. quency of 5 kHz.
15. Design a first-order high-pass non-inverting 24. Design a MFB second-order high-pass
filter with unity gain and a break frequency of Butterworth filter with a gain of 6 and a
5 kHz. cut-off frequency of 2 kHz.
16. Design a first-order non-inverting high-pass
filter with gain of 8 and a cut-off frequency of
17 kHz.
17. Design a first-order high-pass filter with gain
of 15 and a cut-off frequency of 10 kHz using
the inverting configuration.
18. Show that in a first-order high-pass filter, the
amplitude response is down 3 dB at the
cut-off frequency and determine the phase
angle of the output relative to the input at
that frequency. Fig. 11.64 Circuit for Question 26
Fig. 11.63 Circuit for Question 21 Fig. 11.65 Circuit for Question 29
Bibliography 455
and
1
G¼ ð12:8Þ
3
The magnitude and phase responses can be deter-
mined analytically by setting s ¼ jω in (12.6)
giving
V wo jωCR
ðjωÞ ¼
V wi 1 þ 3CRjω þ ðjωÞ2 C 2 R2
Fig. 12.3 Wien network jf = f o
¼ ð12:9Þ
1 þ 3jf = f o þ ðjf = f o Þ2
shown in Fig. 12.2b form a bridge across which a
null occurs at fo. The responses are shown in Fig. 12.4. Clearly at
In analysing this system, consider the Wien the frequency fo, the phase response shows that
network as shown in Fig. 12.3. The transfer func- there is no phase shift between Vwo and Vwi. Also,
tion Vwo/Vwi for this network is given by at this frequency, |Vwo/Vwi| ¼ 1/3.
From (12.9) for Vwo to be in phase with Vwi,
V wo Z2 since the numerator is imaginary, then the denom-
¼
V wi Z 1 þ Z 2 inator must also be imaginary. Hence, the real
1
R sC = R þ sC1 part of the denominator must be zero, that is
¼ ð12:5Þ
R þ sC
1
þ R sC
1
= R þ sC
1
1 ω2 C 2 R 2 ¼ 0 ð12:10Þ
This reduces to
i.e.
V wo sCR
¼ ð12:6Þ 1 1
V wi 1 þ 3sCR þ s2 C 2 R2 ω0 ¼ or f0 ¼ ð12:11Þ
CR 2πCR
This response is called a band pass response and This gives
has centre frequency ωo ¼ 2πfo and pass band
gain G given by V wo jωCR 1
ðω Þ ¼ ¼ ð12:12Þ
V wi 0 3jωCR 3
1 1
ω0 ¼ ,f ¼ ð12:7Þ
CR o 2πCR Thus, a signal Vwi entering the network produces
a signal Vwo at the output and at the frequency fo;
460 12 Oscillators
0.35
0.3
0.25
Magnitude
0.2
0.15
0.1
0.05
0
90
45
Phase (deg)
-45
-90
10-3 10-2 10-1 100 101 102
Frequency (Hz)
where β is the transfer function of the Wien net- Solution Using the configuration in Fig. 12.5,
work given in (12.9) and A is the amplifier gain. we first choose a reasonable value for C say
From (12.13), we have C ¼ 0.01 μF. Then, for oscillation at f ¼ 1 kHz
using Eq. (12.17) the value of R is given by R ¼
jωCRA ¼ 1 þ 3jωCR ðωCRÞ2 ð12:14Þ 1
2π103 0:01106
¼ 15:9 k: Also, from (12.16),
from which the gain of the op-amp is given by 1 þ RRBA ¼ 3. If
we choose RA ¼ 1 k then RB ¼ 2 k. In order to
1 ðωCRÞ2 þ jωCRð3 AÞ ¼ 0 ð12:15Þ
sustain oscillation, we make R2 slightly greater
Hence equating real and imaginary parts to zero, than 2 k say 2.01 k
we have A 3 ¼ 0 and 1 ω2C2R2 ¼ 0 giving This circuit was simulated in the laboratory
using an LF351 operational amplifier and a
A¼3 ð12:16Þ 15-volt supply. The frequency of the resulting
sinusoidal waveform was 968 Hz and the
12.2 RC Oscillators 461
1
2R:2sC =ð2Rþ2sC
1
Þ
Solution For the network, Vo
Vi ¼ 2R:1=2sC ¼
1
RþsC þ2Rþ1=2sC
2sCR
1þ7sCRþ4s2 C 2 R2
¼ 14ω2 R2sCR
C þ7sCR
2 2 where s ¼ jω. For
Vo to be in phase with Vi, 1 4ω2R2C2 ¼ 0 which
gives ωo2 ¼ 1/4C2R2 or fo ¼ 1/4πCR. Hence,
V i ð f o Þ ¼ 7 . An oscillator using this network
Vo 2
1
α¼ ð12:23Þ
ωCR
Fig. 12.8 Phase shift oscillator system In order that the phase shift network contributes
180 at fo, the imaginary term in (12.22) must be
additional 180 . At the frequency fo at which the zero, i.e.
phase shift introduced by the RC network is pre-
cisely 180 , the total phase shift around the sys- 6α α3 ¼ 0 ð12:24Þ
tem will be zero. This frequency fo will be the giving
frequency at which the circuit will oscillate,
provided that the magnitude of the loop gain is 1 1
ω0 ¼ pffiffiffi or f 0 ¼ pffiffiffi ð12:25Þ
sufficiently large, i.e. greater than or equal to one. 6CR 2π 6CR
For the lead phase shift network shown in
Fig. 12.9, the transfer function Vo/Vi is given by By substituting ωo ¼ 2πfo in (12.22), we get
Vo 1
Vo Vo VB VA
¼ ð f oÞ ¼ ð12:26Þ
Vi VB VA Vi Vi 29
R ZB ZA In order to satisfy the Barkhausen criterion, then
¼ ð12:18Þ
R jX C Z B jX C Z A jX C A ¼ 29, i.e. the magnitude of the gain of the
associated amplifier must be 29 or greater.
where
A practical implementation of the phase shift
1 oscillator using the lead network is shown in
XC ¼ ð12:19Þ
2πfC Fig. 12.10. An inverting op-amp is used to pro-
vide 180 phase shift as well as the gain of 29.
RðR jX C Þ
ZB ¼ ð12:20Þ Note that the input resistor R of the op-amp is one
2R jX C of the resistors of the phase shift network. This is
RðZ B jX C Þ possible since the inverting input of the op-amp is
ZA ¼ ð12:21Þ a virtual earth. The phase shift oscillator is not
R þ Z B jX C
very useful as a variable frequency oscillator
since frequency variation requires variation of
After substitution and extensive manipulation, three components simultaneously. It can however
the transfer function becomes be used as a fixed frequency oscillator.
Vo 1
¼ ð12:22Þ Example 12.3
V i 1 5α2 jð6α α3 Þ
Design a phase shift oscillator to operate at a
where frequency of 1 kHz.
jX C ðR þ Z B Þ
12.2.3 Phase Shift Oscillator-Lag ZA ¼ ð12:30Þ
R þ Z B jX C
Network
After substitution and considerable manipulation,
The phase shift oscillator may be constructed the transfer function becomes
using three lag RC sections instead of the lead
Vo 1
RC sections used previously. The basic system is ¼ ð12:31Þ
V i 1 5α2 þ jð6α α3 Þ
shown in Fig. 12.11.
Neglecting amplifier loading the inverting where
amplifier shifts the signal by 180 , while the lag
network shifts negatively by an additional factor. α ¼ ωCR ð12:32Þ
At a frequency fo at which this shift is 180 ,
For a 180 phase shift contribution at ωo, the
with a loop gain of one or greater, oscillation
imaginary term in (12.31) must be zero, i.e.
results. For the phase shift lag network shown in
Fig. 12.12, the transfer function Vo/Vi is given by 6α α3 ¼ 0 ð12:33Þ
Vo Vo VB VA
¼ giving
Vi VB VA Vi
pffiffiffi pffiffiffi
jX C ZB ZA 6 6
¼ ð12:27Þ ω0 ¼ or f 0 ¼ ð12:34Þ
R jX C Z B þ R Z A þ R CR 2πCR
Fig. 12.13 Phase shift lag oscillator with buffered network output
pffiffi
Vo 1 Eq. (12.34) yields R ¼ 2π103 0:110
6
6 ¼ 4 k . For
ð f 0Þ ¼ ð12:35Þ
Vi 29
RI ¼ 2.5 k, then RF ¼ 72.5 k. This circuit based
on the LF351 was simulated and produced a
In order to satisfy the Barkhausen criterion, sinusoidal waveform of frequency 965 Hz and
A ¼ 29 which means that the associated amplitude 1.22 volts peak. An interesting feature
amplifier must have a gain whose magnitude is of this circuit is that the three RC low-pass
29 or greater. In order to prevent the inverting sections filter out harmonics from the signal and
amplifier from loading the network and thereby hence the distortion at the output of the voltage
affect its oscillating frequency, the phase shift follower is low.
oscillator is implemented using the configuration
shown in Fig. 12.13 in which a buffer is included
at the output of the third RC section. Again, an 12.2.4 Buffered Phase Shift Oscillator
inverting operational amplifier is used to provide
180 phase shift and 29 gain. A dual op-amp The buffered phase shift oscillator shown in
such as the LF353 is convenient for the imple- Fig. 12.14 involves buffering all RC sections of
mentation of this circuit. the phase shift lag network such that there is no
interaction between the sections. The effect of this
Example 12.4 is the simplification of the analysis since there is
Design a phase shift lag oscillator with an no interaction between successive RC sections.
oscillating frequency of 1 kHz. The transfer function for each section is given
by 1/(1 + sRC), and hence the Barkhausen
Solution Using the circuit of Fig. 12.13, for the criterion gives
phase shift network, choose C ¼ 0.1 μF. Then
12.2 RC Oscillators 465
pffiffi
1 Then Eq. (12.39) gives R ¼ 2π5104 0:0110
3
6 ¼
Aβ ¼ ¼1 ð12:36Þ
ð1 þ sCRÞ3 5:5 k. The gain of the amplifier is given by RR21 ¼ 8.
This reduces for s ¼ jω to We choose R1 ¼ 5 k giving R2 ¼ 40 k. We use
R2 ¼ 42 k in order to sustain oscillation. This
A ¼ 1 3ω2 C 2 R2 þ jωCR 3 ω2 C 2 R2 circuit was simulated, and it oscillated at a fre-
ð12:37Þ quency of 4.92 kHz with amplitude of 1.9
volts peak.
For A real, it follows that the imaginary compo-
nent on the right must be zero, i.e.
12.2.5 Multiphase Sinusoidal
3 ω2 C 2 R 2 ¼ 0 ð12:38Þ
Oscillator
giving
pffiffiffi pffiffiffi Another type of phase shift oscillator is the multi-
3 3 phase sinusoidal oscillator shown in Fig. 12.15.
ω0 ¼ or f 0 ¼ ð12:39Þ
CR 2πCR This circuit uses active phase shift first-order
elements to produce 360 around the loop. Each
This is the frequency of oscillation. Substituting
of the elements is identical, and therefore three
this in (12.37) gives A ¼ 8, i.e. each RC
equal amplitude outputs are produced each
section produces an attenuation of 1/2 at fo
shifted in phase from the other by 120 . This
resulting in a total attenuation of 1/8. This circuit
kind of oscillator is particularly useful in
is easily implemented using a quad operational
communications systems. The transfer function
amplifier. It is particularly useful for low distor-
for each section is given by k/(1 + sRC), and
tion oscillation as each RC network filters the
applying the Barkhausen criterion gives
harmonics from the signal. Also, since the ampli-
fier gain is only 8, more feedback can be 3
R=R1
employed around this amplifier than for the case Aβ ¼ ¼1 ð12:40Þ
1 þ sRC
where the gain is 29.
For s ¼ jω, this reduces to
Example 12.5
Design a buffered phase shift oscillator to oscil- ð1 þ jωτÞ3 þ k3 ¼ 0 ð12:41Þ
late at 5 kHz. Use an LF347 quad op-amp and a
15-volt power supply. where k ¼ R/R1 and τ ¼ RC. This becomes
1 þ k 3 3ω2 τ2 þ jω 3 ω2 τ2 ¼ 0 ð12:42Þ
Solution Using the circuit in Fig. 12.14, for
oscillation at 5 kHz, we choose C ¼ 0.01 μF.
3 ω2 τ 2 ¼ 0 ð12:43Þ
from which
pffiffiffi pffiffiffi
3 3
ωo ¼ or f o ¼ ð12:44Þ
RC 2πRC
Substituting this in (12.42) gives k3 ¼ 8, and
therefore k ¼ 2, i.e. each stage must have a gain
of 2.
Reff ¼ r s Q2 ð12:60Þ
V
AV ¼ o ¼ gm r d kjX L kjX C ð12:51Þ
Vi where Q is the quality factor of the inductor given
where rd is the output resistance of the FET. The by
high FET input impedance prevents transformer ωo L
loading. Now, for the circuit, Q¼ ð12:61Þ
R
The equations are adjusted by replacing rd by rd||
di di di
Vo ¼ L 1 þ M 2 L 1 ð12:52Þ Reff ¼ r d. The frequency of oscillation is therefore
dt dt dt unaffected while the loop gain condition (12.59)
di2 di di becomes
V f ¼ L M 1 M 1 ð12:53Þ
dt dt dt M
g r ¼1 ð12:62Þ
since i2 0 and where M is the mutual induc- L m d
tance between the two windings, i1 is the current
in the transformer primary and i2 is the current in Example 12.8
the transformer secondary. From (12.52) to Design an LC resonant oscillator to operate at
(12.53) 31.8 kHz. Use a 2N3819 JFET and a 15-volt
V f M supply.
β¼ ¼ ð12:54Þ
Vo L
Solution For the 2N3819, VP ¼ 3 V, IDSS ¼ 10
Hence mA and rd ¼ 50 k. Choose ID ¼ 1 mA for the
JFET operating current. Then using Shockley’s
M qffiffiffiffiffiffi qffiffiffiffi
Aβ ¼ gm r d kjX L k jX C ð12:55Þ
L equation, RS ¼ I DSS 1 I D ¼
ID VP
10 1
1
3
M rd ¼ 2:1 k, CS is chosen to have a reactance that
Aβ ¼ g 1
L m 1 þ jX kjX
r
is low (one tenth or less) compared with RS at the
L C
oscillation in the simulation, M was set at 10 mH In this implementation, the amplifier produces
and the gain of the amplifier reduced by zero phase shift at fo, and, hence, no further signal
bypassing only half of RS. The system oscillated inversion is required. The signal is fed back to the
at 29.3 kHz with amplitude of 24.2 volts peak non-inverting input of the current feedback
to peak. amplifier.
The LC resonant oscillator may be
implemented using a special current feedback Example 12.9
amplifier, the AD844, in which access to the Design an LC resonant oscillator using the
compensation node is available. This is shown AD844 to operate at 1 kHz.
in Fig. 12.19. The Barkhausen criterion gives
Solution For a frequency of 1 kHz choose
r z kjX L kjX C
Aβ ¼ ρ C ¼ 0.005 μF. Then using (12.66), we get
R3 þ r x r
L ¼ 5 H. Condition (12.67) gives R1RþR
1
: z ¼
ρ rz 2 R3 þr x
¼ ¼ 1∠0o ð12:63Þ 1. For the AD844, rz ¼ 3 106 Ω and rx ¼ 50 Ω.
R3 þ r x 1 þ r z jðX L X C Þ
XL XC Choose R1 ¼ 1 kΩ, R2 ¼ 2 kΩ and hence
where R3 ¼ 500 kΩ. This circuit was simulated and
the frequency of oscillation was measured at
R1 980 Hz. The primary advantage of this implemen-
ρ¼ ð12:64Þ
R1 þ R2 tation is that a coupling transformer is not
required.
and rx and rz are parameters of the AD844. From
(12.63), it follows that the imaginary component
of Aβ must be zero, i.e.
12.3.2 Colpitts and Hartley Oscillators
XL ¼ XC ð12:65Þ
The Colpitts and Hartley oscillators can be
giving
represented in the general form shown in
1 Fig. 12.20. The active device may be a FET or
fo ¼ pffiffiffiffiffiffi ð12:66Þ
2π LC operational amplifier with high input impedance.
Its gain with no load is Av and its output imped-
At the frequency of oscillation, Aβ(fo) ¼ 1, i.e. ance Ro. The equivalent circuit is shown. The load
R1 rz impedance ZL consists of Z2 in parallel with a
¼1 ð12:67Þ
R1 þ R2 R3 þ r x series combination Z1 and Z3. Feedback to the
Example 12.11
Design a Colpitts oscillator to oscillate at
31.8 kHz using a 2N3819 JFET. Use a supply
voltage of 15 volts.
where Z1 and Z2 are capacitors C and Z3 is an The LC oscillator shown in Fig. 12.24 utilizes an
inductor L in series with a capacitor CT. LC tank circuit in the positive feedback loop of an
Substituting Z1 ¼ jXC, Z2 ¼ jXC and Z3 ¼ j amplifier. At the resonant frequency of the circuit,
(XL + XCT) in (12.83) results in the impedance becomes (i) very large such that
AV X C X C positive feedback is maximum and (ii) resistive
Aβ ¼ such that the phase shift of the feedback signal is
jRo ð2X C þ X L þ X CT Þ X C ðX C þ X L þ X CT Þ
zero. Hence, since the phase shift produced by the
ð12:84Þ
amplifier is zero, the conditions for oscillation can
For zero phase shift around the loop, then be met. The loop gain is given by
2X C þ X L þ X CT ¼ 0 ð12:85Þ Z LC R þ R2
Aβ ¼ : 1 ð12:87Þ
R þ Z LC R1
Substituting this in (12.84) results in the condition
Av 1. Note that (12.85) yields where
qffiffiffiffiffiffiffiffiffiffiffiffiffi jX L ðjX C Þ
Z LC ¼ ð12:88Þ
CT þ 2
C
1 jX L jX C
fo ¼ pffiffiffiffiffiffi ð12:86Þ
2π LC
At the resonant frequency ωo
The capacitor CT enables easy tuning of the
jX L jX C ¼ 0 ð12:89Þ
oscillator.
giving
Example 12.12
1 1
Evaluate the frequency of oscillation of the ωo ¼ pffiffiffiffiffiffi or f o ¼ pffiffiffiffiffiffi ð12:90Þ
resulting Clapp oscillator by the introduction of LC 2π LC
a capacitor CT ¼ 0.01 μF in the Colpitts oscillator
At this frequency, from (12.87) the loop gain
of Example 12.11.
goes to
12.4 Crystal Oscillators 473
R2
Aβ ¼ 1 þ 1 ð12:91Þ
R1
Example 12.13
Design a simple LC oscillator to oscillate at a
frequency of 1 kHz, using an LF351 op-amp.
Fig. 12.25 Equivalent circuit of crystal
1 1
Z cry ¼ þ Rs þ sLs
sC p
sC s
1 ω2 C s Ls þ jωC s Rs
¼
Rs C s C p þ jω C s þ Cp ω2 C s C p Ls
ð12:92Þ
j ð ω2 L s C s 1 Þ
Z cry ¼ : ð12:93Þ
ω C s þ C p ω2 C p C s L s
Fig. 12.27 Crystal oscillator
From Eq. (12.92), the crystal has two resonant
frequencies. The series-resonant frequency
ωs corresponds to Zcry ¼ 0 which is obtained by
setting the numerator in (12.93) to zero. This
R Z cry
gives Aβ ¼ 1þ 2 ð12:96Þ
R3 R þ Z cry
1 1
ωp ¼ pffiffiffiffiffiffiffiffiffiffi or f p ¼ pffiffiffiffiffiffiffiffiffiffi ð12:94Þ Near parallel resonance, which is set for a given
Ls C s 2π Ls Cs Z
crystal, |Zcry| is large resulting in RþZcrycry 1, and
The parallel resonance frequency ωp corresponds the phase shift is zero. At other frequencies, |Zcry|
to Zcry ¼ 1 which is obtained by setting the falls and Aβ ¼ 1 is not met. Hence, for oscillation
denominator in (11.89) to zero. This results in at the crystal resonant frequency fo
rffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 C R2
ωp ¼ pffiffiffiffiffiffiffiffiffiffi 1 þ s or Aβ 1 þ 1 ð12:97Þ
Ls C s C p R3
Note that fp > fs. From (12.94), it is clear that the Example 12.14
series-resonant frequency is determined solely by Design a crystal oscillator to oscillate at 100 kHz
the parameters Ls and Cs, which are well defined using a suitable operational amplifier.
by the crystal. From (12.95), however, the paral-
lel resonant frequency is dependent on the elec- Solution Using a suitable operational amplifier
trode capacitance Cp that is susceptible to and a 100 kHz crystal, the circuit was configured
variations. using R1 ¼ 1 kΩ, R2 ¼ 1 kΩ and R3 ¼ 1 kΩ.
Note that the amplifier gain A ¼ 2 ensures that
Aβ > 1 and hence that oscillation occurs.
12.4.1 Crystal Oscillator Using
an Op-Amp
12.4.2 Miller Oscillator
In the circuit shown in Fig. 12.27, the crystal,
along with resistor R, provides positive feedback A simple crystal oscillator making use of the
around the operational amplifier. The loop gain is natural oscillations of the crystal is the Miller
given by oscillator shown in Fig. 12.28. The output of the
12.4 Crystal Oscillators 475
jX L jX C ¼ 0 ð12:98Þ
giving
1 1
ωo ¼ pffiffiffiffiffiffi or f o ¼ pffiffiffiffiffiffi ð12:99Þ
LC 2π LC
Example 12.15
Design a Miller oscillator using a 2N3918 JFET
to oscillate at 100 kHz. Use a 15-volt supply. Fig. 12.28 Miller oscillator
the FET is adjusted until a stable output signal is in Fig. 12.35 is an all-pass filter. Its transfer func-
obtained. This circuit works very well and tion is given by H ðsÞ ¼ 1Ts
1þTs where T ¼ RC. It has
maintains constant output amplitude over a wide the unique feature of unity gain and varying phase
frequency range. For proper operation the ampli- with changing frequency. The phase varies from
tude of the signal across the FET must maintain it 0 to 180 , and hence the Barkhausen criterion
in the ohmic region. can be easily met by combining three all-pass
stages as shown.
The Barkhausen criterion gives
12.7 Oscillator Creation
1 Ts 3
Aβ ¼ ¼1 ð12:100Þ
Having discussed the principle of operation of an 1 þ Ts
oscillator and the Barkhausen criterion, it is pos- which is equivalent to
sible to create “new” oscillator structures by
employing these principles. jAβð jωo Þj ¼ 1 ð12:101Þ
and
Oscillator 1 A simple illustration of this is the
use of an all-pass filter network to realize an ∠ðAβð jωo ÞÞ ¼ 2π ð12:102Þ
oscillator structure. Each of the cascaded stages
Since |H(s)|¼1 in the all-pass network, condition
(12.101) is satisfied. From condition (12.102)
3 2 tan 1 ðωo T Þ ¼ 2π ð12:103Þ
Exercise 12.1
Show that for four all-pass networks, the fre-
quency of oscillation is f o ¼ 2πRC
1
.
where C T ¼ CC22þC
C3
3
. Using R1 ¼ 470 k and
R2 ¼ 4.7 k to bias the transistor, with coupling
capacitors C1 ¼ C4 ¼ 0.1 μF, then for C2 ¼ C3 ¼ 1
nF and L1 ¼ 1 mH, we get f o ¼
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
¼ 225 kHz . The frequency of
3 9
2π 10 0:510
oscillation using simulation software was
measured at 215 kHz. Note that using (12.112),
(12.111) becomes ω2 L1 C 2 ¼ LL11CCT2 ¼ CCT2 > 1 .
Therefore, the Barkhausen criterion is generally
satisfied.
Fig. 12.38 Oscillator 3
Exercise 12.3 Show that if X1 and X3 are
capacitors and X2 is an inductor, while X1 and
X2 have different signs as required, the system
will not oscillate.
12.8 Applications
be determined and also gives an indication of the JFET. From its specifications, the 2N3819 for
stability of the circuit, particularly under varying VDS < VP has a channel resistance rd that varies
loads. The sine-wave generator is also used for between 100 Ω and 3.6 k for VGS varying from
troubleshooting as it can be injected into a circuit 0 V to 2.5 V. Therefore R2 ¼ 20 k and
at different points and monitored at others. R1 ¼ 8 k result in gain variation from 3.4 when
Another extremely important application of rd ¼ 100 Ω to 2.7 when rd ¼ 3.6 k, inclusive of
sine-wave generators is in communication the critical value 3. The gate-biasing resistor R5 is
systems. In radio transmitters, information to be set at 1 M. Capacitor C1 is chosen to provide
transmitted is superimposed on a sinusoidal sig- adequate filtering of the output ripple and thereby
nal called a carrier. The resulting modulated car- produces a DC voltage. Since R5 is large, this
rier signal is then amplified and radiated through capacitor need not be very large. It cannot be
radio antennas as radio waves. In radio receivers, too large otherwise the time constant R5C1 will
this radio wave induces an electrical signal in a prevent a quick response to varying amplitude
receiving antenna. It is then mixed with a sinusoi- changes. A compromise value of 2 μF is used.
dal signal produced by a local oscillator and the The diodes are small signal devices such as
resulting lower-frequency signal processed by the 1N4148. The output of the Wien bridge oscillator
receiver circuitry to extract the information. This is low and the output amplifier opa3 is used to
mixing technique has long formed the basis of boost this signal by a factor of two by setting
super heterodyne receivers. Sinusoidal oscillators R3 ¼ R4 ¼ 10 k. The control amplifier opa2 is
are also used in standard frequency sources and in set at a gain of 6 with R7 ¼ 10 k and R6 ¼ 2 k in
the production of synthesized music. order to provide sufficient drive for the half-wave
rectifier. The potentiometer VR2 ¼ 10 k provides
Wide-Range Wien Bridge Oscillator a variable signal level to the control amplifier.
The Wien bridge oscillator circuit in Fig. 12.40 The circuit is operated from a 15 V supply.
uses the LF351 JFET input op-amp and is Resistors are 2% tolerance and all capacitors are
intended for laboratory use. It covers the fre- polystyrene.
quency range 15 Hz 150 kHz in four bands.
It is a simple system and the frequency is easily Ideas for Exploration: (i) Introduce another
varied over a wide frequency range using the potentiometer in order to provide a variable out-
switched banks of capacitors and the ganged var- put signal amplitude from the oscillator.
iable potentiometer. The components of the Wien
network giving the frequency ranges shown are Wien Bridge Oscillator Using a 741 Op-Amp
selected using Eq. (12.11). Fixed resistors Ra ¼ 1 The circuit in Fig. 12.41 uses the 741 op-amp in a
k are used to set the minimum value of R, and a simple Wien bridge oscillator. It uses the Wien
dual 10 k linear potentiometer is used for VRa as network components as the previous case. An
the remaining part of R. This arrangement R53 thermistor (if available) and resistor
provides continuous variation within a band set R1 ¼ 470 Ω produce the necessary gain of three
by a pair of equal capacitors C. Values of C of with amplitude stabilization as a result of the
1 μF, 0.1 μF, 0.01 μFand 0.001 μF establish four action of the thermistor. Note however that the
bands. By switching in different equal pairs of GBP of the 741 limits the upper frequency of
C using the two-pole multi-throw switch, four operation of the circuit to less than 1 MHz/
frequency bands can be realized, each allowing 3 ¼ 333 kHz.
a decade change in frequency by varying the
potentiometer. These bands are 15 Hz 150 Hz, Ideas for Exploration: (i) The R53 thermistor is
150 Hz 1.5 kHz, 1.5 kHz 15 kHz and difficult to find and appears to be no longer avail-
15 kHz 150 kHz. able. Use another approach to amplitude stabili-
The gain of the oscillator amplifier is set by R1 zation such as the low power incandescent lamp
and R2 along with the channel resistance of the as shown in Fig. 12.42. In operation,
482 12 Oscillators
Crystal Oscillator 1
The circuit in Fig. 12.52 is an oscillator that
employs a series-resonant crystal in a positive
feedback loop around two transistor stages. The
first stage around Tr1 is a common base amplifier
with the output of the crystal feeding into the
transistor emitter. The output at the collector of
Tr1 is fed into the emitter follower stage Tr2, the
Fig. 12.51 Colpitts BJT output of which feeds into the series-resonant
crystal. Thus, at resonance the impedance of the
Colpitts Oscillator Using a BJT crystal is at a minimum, and positive feedback is
The circuit shown in Fig. 12.51 is a Colpitts at a maximum. At all other frequencies, the
oscillator implemented using a BJT. Here the impedance is higher and therefore oscillation
transistor is again biased using collector-base occurs at the series-resonant frequency of the
feedback biasing with R1 ¼ 470 k and crystal. Resistors R1 ¼ 47 k and R2 ¼ 12 kset
R2 ¼ 4.7 k. Capacitor C1 ¼ 0.1 μF couples the the base voltage of Tr1 at 3 V. Therefore with
output of the transistor amplifier to the input of R4 ¼ 1.2 k, the current in the first stage is 2.3/
the LC network. The output of the network is 1.2 k ¼ 1.9 mA. With R3 ¼ 2.2 k, the voltage at
directly connected to the input of the transistor the base of the emitter follower is 15 (1.9
with no coupling capacitor being necessary. With mA 2.2 k) ¼ 10.8 V. Resistor R5 ¼ 2.2 k sets
L ¼ 10 mH and C ¼ 5 nF, the oscillating fre- the current in Tr2 at 10.1/2.2 k ¼ 4.6 mA. Capac-
pffiffiffi pffiffiffiffiffiffi
quency is given by f ¼ 2=2π LC ¼ 31 kHz. itor C2 works in conjunction with the crystal and
is typically about 100 pF. This circuit operates
Ideas for Exploration: (i) Examine the possibil- with a wide range of crystals from 100 kHz to
ity of frequency variation by using a variable 10 MHz.
12.8 Applications 487
output signal from the network into the inverting
this transfer function is given by
VVoi ðjωÞ
¼
ω¼ωo terminal of the amplifier results in zero phase shift
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
11
2
¼ 0. Thus at ω ¼ 1/RC corresponding around the system and with amplification satisfies
ð11Þ þ4 2
to f ¼ 1/2πRC, the signal transmission through Aβ 1. Use the circuit shown in Fig. 12.56 where
the network is zero. Therefore, at this single fre- there is no positive feedback with adjustable R3 in
quency, there is no negative feedback around the order to produce oscillation.
op-amp, and the positive feedback around the
system produces oscillation at this frequency. At Research Project 2: Sinusoidal Oscillator
all other frequencies, the network produces an Using an Inverter
output such that negative feedback signal is The oscillator shown in Fig. 12.57 satisfies the
applied to the amplifier. Thus the condition Barkhausen criterion by using a NAND gate
Aβ 1 around the positive feedback loop is connected as an inverter producing 180 phase
operational at fo ¼ 1/2πRC where A is the ampli- shift along with amplification. The output of the
fier gain at fo and β ¼ RA/(RB + RA). Using NAND gate drives a phase-lag network that
R ¼ 2 k and C ¼ 0.01 μF, the oscillating fre- produces the additional 180 phase shift. The
quency is given by fo ¼ 1/2πRC ¼ 1/ result is continuous oscillation. The square wave
2π 2000 0.01 106 ¼ 8 kHz. For out of the inverter is filtered by the three-stage RC
sustained oscillation, RA ¼ 1 k and RB ¼ 10 k phase shift network thereby removing harmonics
giving β ¼ 1 k/((1 k + 10 k) ¼ 0.09. This value and producing a sine-wave output. The frequency
pffiffiffi
may have to be adjusted, but increasing values of of the sine wave is f ¼ 6=2πRC , and the
β will generally result in increased distortion of amplitude of this signal is given by Vsin ¼ 4Vsq/
the signal.
Problems 489
Problems
50 kHz using a 2N3819 JFET. Use a supply 20. Design a Miller oscillator using a 2N3918
voltage of 24 volts. JFET to oscillate at 110 kHz. Use a 25-volt
17. Determine the required capacitor, CT, to supply.
adjust the frequency of the Colpitts
oscillators in Question 16 to 55 kHz.
18. Using the circuit in Fig. 12.24, design a sim- Bibliography
ple LC oscillator to oscillate at a frequency of
5 kHz. J. Millman, C.C. Halkias, Integrated Electronics: Analog
and Digital Circuits and Systems (Mc Graw Hill Book
19. Design a crystal oscillator to oscillate
Company, New York, 1972)
at 75 kHz using an OP42 operational U. Tietze, C. Schenk, Advanced Electronic Circuits
amplifier. (Springer-Verlag, Berlin, 1978)
Waveform Generators and Non-linear
Circuits 13
In this chapter, we consider circuits that involve comparator supply rails VCC and VSS, respec-
non-linear operation of the operational amplifier. tively. The symbol for a comparator is also iden-
These can be used to realize waveform generators tical to an op-amp. In fact, because of its high
which are circuits that produce a variety of non- open-loop gain, an op-amp can operate as a com-
sinusoidal waveforms. They are fundamentally parator by omitting feedback. To understand the
instrumentation building blocks used for signal operation of a comparator, consider the circuit
generation and test and measurement. Typically, shown in Fig. 13.1.
waveform generators produce square waves, tri- Here Vi the input voltage is applied at the
angular waves, pulses and in some cases arbitrary inverting input, while the reference voltage VR is
waveforms over a range of frequencies with con- applied at the non-inverting input of the compar-
stant amplitude. Here we discuss comparators, ator. If Vi < VR, then the differential voltage
op-amp-based free-running (astable) and VD ¼ Vi VR is negative and hence the output
one-shot (monostable) multivibrator circuits as of the comparator is positive, i.e. Vo ¼ Vomx, and
well as precision rectifiers and other non-linear if Vi > VR, then VD ¼ Vi VR is positive and the
circuits. At the end of the chapter, the student will output of the comparator is negative,
be able to: i.e. Vo ¼ Vomn. This comparator would be
operating in the inverting mode. If the input
• Explain the operation of several non-linear terminals are reversed such that Vi is applied at
circuits the non-inverting input while the reference volt-
• Explain the operation of a wide range of wave- age VR is applied at the inverting input of the
form generators comparator, then if Vi < VR, then the differential
• Design a wide range of non-linear circuits and voltage VD ¼ Vi VR is negative and hence the
waveform generators output of the comparator is negative,
i.e. Vo ¼ Vomn, and if Vi > VR, then VD ¼ Vi VR
is positive and the output of the comparator is
positive, i.e. Vo ¼ Vomx. This comparator would
13.1 The Comparator be operating in the non-inverting mode.
If VR ¼ 0, then the comparator is referred to as
A comparator is very similar to an op-amp except an inverting or non-inverting zero-crossing detec-
that the differential gain is made larger with the tor. This action for VR ¼ 0 for a triangular wave
addition of internal positive feedback circuitry. input is shown in Fig. 13.2 for a non-inverting
As a result, the output has two states Vomx and comparator and in Fig. 13.3 for an inverting
Vomn whose value typically approaches the comparator.
# Springer Nature Switzerland AG 2021 493
S. J. G. Gift, B. Maundy, Electronic Circuit Design and Application,
https://doi.org/10.1007/978-3-030-46989-4_13
494 13 Waveform Generators and Non-linear Circuits
single-ended power supply since both inputs will voltage, and therefore the transistor will be off.
have voltages at values between ground and the When the temperature is above 35 C, the output
supply voltage. Under normal temperature of the op-amp will be high. Therefore, the red
conditions, the output voltage of the temperature LED will be on with current limited by
transducer falls below the reference voltage, and R3 ¼ 1 k to about 10 mA, and the green LED
the op-amp output is therefore saturated at close will be off. The high op-amp output will turn on
to zero. When the temperature rises above some the transistor thereby causing the relay to switch
specific value, the output of the sensor exceeds on power to the fan. Resistor R4 ¼ 10 k ensures
the reference voltage, and the op-amp output transistor saturation.
saturates at close to the supply voltage. This out-
put is used to switch on an indicator using a relay. If noise is present on the input signal, this can
lead to spurious or false switching of the compar-
Example 13.3 ator. In order to overcome this and other potential
Using the basic circuit shown in Fig. 13.4, design problems, the comparator’s performance can be
a system to turn on a fan when the ambient enhanced if positive feedback is added in the form
temperature in a room exceeds 35 C. of resistors Ra and Rb as shown in Fig. 13.5.
For this circuit the output voltage Vo in terms
Solution Here an op-amp is connected as a com- of the input Vi can be determined by writing the
parator. The output of the LM35 temperature equation
sensor drives the non-inverting input of the
Ra
op-amp, while the inverting terminal is connected V o ¼ Ad V Vi ð13:1Þ
to the output of a potential divider arrangement Ra þ Rb o
comprising R1 ¼ 10 k and VR1 ¼ 2 k. With a
where Ad is the differential gain of the compara-
12 V supply, this means that the wiper of the
tor. Solving for Vo gives
potentiometer can be varied from 0 to 2 V. For a
threshold temperature of 35 C, the potentiometer Ad
Vo ¼ V ð13:2Þ
is set to 10 mV 35 ¼ 350 mV. When the Ad β 1 i
temperature is below 35 C, the output of the
where β a constant is defined as β ¼ RaRþRa
. Since
sensor will be less than 350 mV, and therefore b
the output of the op-amp will be low. Therefore, β is always less than one, it follows that the
the green LED will be on with current limited by denominator of (13.2) could be less than one or
R2 ¼ 1 k to about 10 mA, and the red LED will be greater than one. For the case 0 Adβ 1, the
off. The base of the transistor will also be at a low gain of the comparator is negative and larger than
496 13 Waveform Generators and Non-linear Circuits
Ad β 1
V imx ¼ V omx ð13:3Þ
Ad
and
Ad β 1
V imn ¼ V omn ð13:4Þ
Ad
Fig. 13.6 Transfer characteristic of the comparator when Note that if a voltage VR is applied to the
β¼0 grounded end of resistor Ra by removing it from
ground, the transfer characteristic of Fig. 13.7
Ad which corresponds to the case of the simple shifts to the right by an amount VR if VR > 0 and
comparator in Fig. 13.1. When Adβ ¼ 0 which to the left by an equal amount if VR < 0. The new
corresponds to setting Ra ¼ 0 or Rb ! 1, VR ¼ 0 switching points can be found by adding VR to
and the comparator has the transfer characteristic Eqs. (13.3) and (13.4). Finally, an op-amp can
shown in Fig. 13.6. For values of 0 Adβ 1 the function as a comparator or a Schmitt trigger.
slope of the transfer characteristic is 1βA
Ad
d
> However, its switching times between output
Ad . states will typically not be as good as a dedicated
For Adβ > 1 the situation changes as the slope comparator.
of the transfer characteristic becomes positive and
less than Ad. Under this condition the comparator Example 13.4
is referred to as a Schmitt trigger with a transfer Design a Schmitt trigger circuit with a switching
characteristic shown in Fig. 13.7. From Fig. 13.7, threshold of 5 V using the LF411 op-amp
the transfer characteristic has a unique shape in powered by 15 V supplies. For the op-amp
that for Vi > Vimx the output voltage is Vo ¼ Vomn. assume |Vomx| ¼ |Vomn| ¼ 14.2 V and Ad ¼ 200,
However, if Vi starts off being greater than Vimx 000. Verify your result using PSPICE with an
and is decreased to Vimx and less, the output input signal Vi ¼ 10 sin (2π103t)V.
remains at Vomn until Vi < Vimn is reached. When
this occurs, Vo ¼ Vomx. The converse is true, i.e. if Solution To find β we use Eq. (13.3) with
we start with Vi being less than Vimn and increas- Vimx ¼ 5 V. Solving yields β ¼ 0.352. Choosing
ing to Vi and greater, the output will eventually Ra ¼ 1 k yields Rb ¼ 1.84 k. The PSPICE simu-
switch from Vomx to Vomn. Thus, the Schmitt trig- lation result is shown in Fig. 13.8 verifying that
ger has two distinct switching points, Vimx and the circuit triggers at 5 V.
Vimn given by
13.2 Square-Wave Generation 497
(a) (b)
Fig. 13.9 (a) A square-wave generator with a clamped output, (b) output voltage Vo and capacitor voltage Vc waveforms
while D1 is off. For this half-cycle the time con- be minimized by using a comparator for A1 in
stant τ1 ¼ R1aC1. Conversely for the negative place of an op-amp. Since the comparator is
half-cycle D1 is on while D2 is off, and the time designed for high gain and not in a feedback
constant is now τ2 ¼ R1bC1. The expression for configuration, its slew rates can be very high.
the periods T1 and T2 are therefore given by Also, as a general rule of thumb, if the output of
the square-wave generator is required to have a
1þβ
T 1 ¼ τ1 ln particular rise time tr, then it is related to the
1β amplifiers slew rate by
R
¼ τ1 ln 1 þ 2 a ð13:8Þ 2V o
Rb t r ¼ 0:8 ð13:10Þ
SR
1þβ Here rise, time is defined as the time it takes
T 2 ¼ τ2 ln
1β the output to get from 10% to 90% of the asymp-
R tote. To minimize the capacitance, two back-to-
¼ τ2 ln 1 þ 2 a ð13:9Þ back diode-connected transistors can be
Rb
employed in place of the Zener diodes. The effec-
tive Zener voltage is then set by the emitter-base
The only disadvantage to using the diodes in breakdown voltage of the transistors used, and the
this manner is that the output voltage is reduced nodal capacitance is considerably less than that of
by one diode drop. As a general rule of thumb, the a commercial Zener diodes.
square-wave generator is useful to frequencies of An alternative method of generating unsym-
about 1/10 of the GBP of the amplifier used. As metrical square waves is to remove the lower end
the frequency is increased beyond that slew rate of resistor Ra from ground and apply a voltage VR
effects and the capacitance seen at the output of to the lower end of resistor Ra. For this configura-
A1 due to Zener diodes alters the output rise and tion, show that the new periods T1 and T2 are
fall times. The effect of slew rate limitations can given by
13.2 Square-Wave Generation 499
ð1 þ βÞ αð1 βÞ reasonable response time of 200 ns and its opera-
T 1 ¼ τ ln ð13:11Þ tion over a wide voltage range. The final circuit is
ð1 βÞ αð1 βÞ
shown in Fig. 13.11. Note the presence of the
and 4 kΩ resistor which is there because the LM111
has an open collector output. A PSPICE simula-
ð1 þ βÞ þ αð1 βÞ
T 2 ¼ τ ln ð13:12Þ tion of the output waveform is shown in
ð1 βÞ þ αð1 βÞ
Fig. 13.12.
where α ¼ VR/Vo is a constant whose magnitude
is less than one but greater than zero. Note that for Example 13.6
α ¼ 0, the equation above revert back to Modify the previous design so that the circuit
Eq. (13.7) and for 0 < α < 1, T1 > T2 and for now operates from a single 12 V supply and
1 < α < 0, T2 > T1. generates a 5 V TTL output at the same
frequency.
Example 13.5
Design a square-wave oscillator to have an oscil-
lation frequency of 100 kHz, with an output of
5 V using a 12 V bipolar supply.
Solution To accomplish this we recognized that frequency. To understand why, we recall from
the original circuit oscillates about zero volts. Fourier series that a square wave is made up of
Therefore, for single supply operation, the new the sum of sinusoidal waves. That is, the output
circuit must oscillate about a new virtual ground voltage Vo(t) of Fig. 13.9 can be described by
which in this case must be Vo/2 or 2.5 V. The
4A 1 1 1
bottom plate of C1 and the bottom end of Ra must V o ðt Þ ¼ sin ω0 t þ sin 3ω0 t þ sin 5ω0 t þ sin 7ω0 t þ
π 3 5 7
therefore be supplied with 2.5 V. At this point the ð13:13Þ
designer has several options open to them. If the
2.5 V source is readily available, the design is where A is the peak amplitude of the square wave.
complete. If not, then one must improvise. Once Examining Eq. (13.13) closely shows us that the
simple method is to use a potential divider com- third harmonic has an attenuation
prising of the 1 k and the 3.8 k resistor to gener- αvo(3ωo) ¼ 9.54 dB greater than the fundamental.
ate the desired voltage that is then buffered. Such We only need to consider the third harmonic since
a solution is shown Fig. 13.13. If the number of the fifth and seventh harmonics, etc. have greater
components is an important factor, the circuit of attenuation. For a unity gain second-order
Fig. 13.13 will also work if the op-amp is bandpass filter tuned to ωo, the attenuation for a
removed and the potential divider directly given Q to the third harmonic is
connected to the 1200 pF capacitor and the 10 k !
resistor, but virtual “2.5 V” will oscillate. Low 3
αBP ð3ω0 Þ ¼ 20 log 10 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
resistor values will reduce the oscillation, but they 9 þ 64Q2
increase the power consumption of the circuit. If ð13:14Þ
the buffer is employed, the resistor values can be
increased to reduce the power consumption. The If Vo(t) is now passed through the bandpass
PSPICE simulation of the output waveform is filter, the total attenuation of the third harmonic is
shown in Fig. 13.14. now αT ð3ω0 Þ ¼ αvo ð3ω0 Þ þ αBP ð3ω0 Þ .
Table 13.1 shows the attenuation for the third
harmonic as a function of Q.
13.2.1 Sine Wave Generation from Hence it can be seen that only a Q of about
a Square-Wave Input 10 is sufficient to achieve almost 40 dB of attenu-
ation for the third harmonic. Thus, a state variable
Low distortion sine waves can be easily generated filter, Tow-Thomas or Antonio’s GIC, could be
from a square-wave input by using a simple used to generate a low distortion sine wave.
bandpass filter to extract the fundamental Finally, note that the resulting sine wave will
13.3 Triangular Wave Generation 501
Table 13.1 Attenuation of the third-order harmonic in a square wave to various values of the bandpass filter Q
Q αBP(3ωo) αvo(3ωo) αT (3ωo)
10 28.25 9.54 38.07
15 32.04 9.54 41.60
20 34.54 9.54 44.08
have an amplitude that is 4/π times that of the the capacitor must be constant so that the voltage
input signal. An example of the filtering action of developed across the capacitor is the integral of a
the second-order bandpass filter is shown in the constant. Hence a current source whose current is
MATLAB simulation Fig. 13.15 where a square a function of the output will generate a triangular
wave of input frequency 1 kHz and 1 V peak-to- output waveform. An example of this is shown in
peak is the input and the output is a sine wave of Fig. 13.16. Here transistors Q1 and Q2 which
amplitude 1.27 V. could be bipolar or CMOS act as current sources
and current sinks. The operation of this circuit is
therefore conceptually similar to that of Fig. 13.9.
That is, during the positive upswing diode, D2 is
13.3 Triangular Wave Generation
forward-biased (D1 is reversed biased) so that
transistor Q2 is effectively off. Transistor Q1
In the previous section we noted that the capacitor
therefore acts as a current source charging C1
voltage was exponential with a time constant. To
with a constant current I C1 ’ ðV Z 3 V BE1 Þ=R1 .
generate a triangular waveform which has con-
When the capacitor voltage Vc(t) reaches the
stant slope, the current charging or discharging of
502 13 Waveform Generators and Non-linear Circuits
threshold voltage +βVo, the output of A1 goes role of diodes D1 and D2 is therefore to steer the
negative and D1 becomes forward-biased with current source or sink depending on the output
D2 now reversed-biased. Transistor Q1 therefore state. Resistor R serves to limit the current
switches off and Q2 switches on acting as a cur- through Zener diodes Z1 and Z2. To compute the
rent sink connected to C1 with constant current period of oscillation, we need to evaluate each
I C2 ’ ðV Z 4 V BE2 Þ=R2. Eventually, the capacitor half-cycle period. Assigning t ¼ 0 as the starting
voltage Vc(t) reaches the threshold voltage βVo point at which the capacitor voltage is βVo, the
and the output of the amplifier A1 goes positive voltage across the capacitor Vc(t) is given by,
and the cycle repeats as shown in Fig. 13.17. The
13.3 Triangular Wave Generation 503
Z t
1 Example 13.7
V c ðt Þ ¼ I C1 dt ð13:15Þ Design a square-wave/sawtooth oscillator that
C 0
oscillates at a frequency of 3.2 kHz.
for 0 t T1. At t ¼ T1 the half-cycle is
completed as Vc1(T1) ¼ + βVo is reached. But Solution We begin, by choosing a typical
from (13.15) op-amp, the TL082 which has a bandwidth of
4 MHz. Next we choose Vz3 ¼ Vz4 ¼ 4.7 V
T 1 V Z 3 V BE1
V c ðT 1 Þ ¼ βV o ð13:16Þ using the 1N750 Zener diodes and R1 ¼ R2 ¼ 1 k.
C R1
Since the average Zener diode current is about
1A, we can choose R3 ¼ 2 k, which limits the
Zener diode current to approximately 10 mA.
Solving for T1 yields T 1 ¼ 2CβV o R1
V Z 3 V BE1 .
Diodes D1 and D2 can be regular 1N4148 diodes,
In a similar manner half-cycle
T2 can be
period and we can set R4 and R5 to both be 1 k. Next, we
solved as T 2 ’ 2CβV o V Z V
R2
BE
. The total choose C1 ¼ 0.22 μF, and with Vo ’ 13.4 V, it
4 2
V z3 V BE
period T ¼ T1 + T2 and hence the frequency of follows that β ¼ 4CV o R1 f
or β ’ 0.1. Since β ¼
osc
oscillation is therefore given by Ra
let us choose Ra ¼ 1 k which results in
Ra þRb ,
VR β T1 ¼
2R1 C 1 Ra
ð13:23Þ
V o2 ¼ þ V o1 ð13:20Þ
1β 1β Rb
αRb
f osc ¼
4C 1 Ra ½R1 þ αð1 αÞRC
αRb
ffi ð13:26Þ
4R1 C1 Ra
T 3T
2 T 2
if R1
RC where RC is the total potentiometer
resistance.
" 2 #
Rb VS Example 13.10
f osc ¼ 1 ð13:30Þ Design a triangular wave generator to generate a
4R1 C 1 Ra VO
positive-going sawtooth wave of 5 V pk-pk
which is a non-linear function of frequency. amplitude with a frequency of 5 kHz of a 15 V
supply.
Example 13.9
Determine the frequency of oscillation of Solution We first note that for a positive ramp
Fig. 13.18 if the control voltage changes from function, VS must be negative. Since we require
VS ¼ 0 V, 2 V, 3 V. You may assume that the output Vo ¼ 5 V we can employ 1N437
Zener diodes for D1 and D2 as before to clamp the
C1 ¼ 0.1 μF, R1 ¼ 1 k and RRba ¼ 6.
output to the desired value. Furthermore let us
choose VS ¼ 4.8 V so that according to
Solution With Vo ffi 13.4 V, f osc ¼ (13.29), δ ¼ 2% and hence T1 < < T2. Employing
6
410000:1106
1 13:4
Vs
. For Vs ¼ 0 V, Eq. (13.30) with fosc ¼ 5 kHz, let us choose
fosc ¼ 15 kHz. For Vs ¼ 2 V, f osc ¼ Ra ¼ Rb ¼ 10 k. Hence, we can solve for the
time constant R1C1 ¼ 3.92 μs. Choosing
6
410000:1106
1 13:4
2
¼ 14:66 kHz . Finally,
C1 ¼ 1000 pF yields R1 ¼ 3.92 k. We can use a
for Vs ¼ 3 V, f osc ¼ 410000:110
6
6
3.9 k resistor. The last step involves the choice of
1 13:4 ¼ 14:25 kHz.
3
amplifiers. If the design application is not critical,
then a general purpose op-amps such as the
LF411 op-amp can be employed for A1 and A2.
The final circuit and PSPICE simulation results
13.3.2 Sawtooth Generation
are shown in Figs. 13.21 and 13.22. Note that the
output of the generator has a negative DC offset.
A natural extension of the triangular wave gener-
Why?
ator is that of a sawtooth waveform. Sawtooth
waveforms are used where a repeatable ramp
Another means of altering the duty cycle is to
function whose single positive or negative slope
employ the circuit of Fig. 13.23 which uses the
has a predefined period T. In fact, the sawtooth
circuit of Fig. 13.10 to alter the time constants.
waveform can be regarded as a triangular wave-
Here we choose R1b > > R1a so that the integrator
form with one slope of the triangular waveform
time constant during the positive cycle is more.
being set at 1. That is either T1 ¼ 0 or T2 ¼ 0.
As such, the triangular wave generator of
Fig. 13.18 can be used, but its duty cycle must
be set to near 0% for a positive-going sawtooth 13.3.3 Voltage-Controlled Oscillators
waveform and near 100% for a negative-going
sawtooth waveform. A voltage-controlled oscillator (VCO) is a circuit
whose frequency of oscillation can be directly
13.3 Triangular Wave Generation 507
exploiting the IC/VBE characteristics of transistors shaping resistor is crucial because it determines
Q1 and Q2 to approximate the sine function over a the distortion levels of the sine wave produced.
limited input range. It can be recognized that Note in Fig. 13.25 amplifier A4 generates VM for
transistors Q1 Q4 form a differential amplifier MOS switches M1 and M2. It is equally possible
with the inverting terminal grounded and a signif- to use JFETs instead of MOS transistors with the
icantly reduced input signal applied to Q1. Here addition of some bias circuitry. Figure 13.26
the 47 Ω resistors act as degeneration resistors, shows the outcome of a PSPICE simulation of
and transistors Q3,4 form a current mirror that the circuit of Fig. 13.25 using LF411 op-amps,
copies the collect current of Q1 and subtracts it R1 ¼ 10 k, C1 ¼ 12 nF and transistors Q1,2 and
from the collector current of Q2. The 100 Ω resis- Q3,4 as general purpose 2N3904/2N3906
tor forces the node labeled a to be low impedance transistors. Input voltage VM is varied from 1 to
and converts the output current to a voltage that is 10 V.
then amplified by amplifier A5. Two variable
resistors are included that offer symmetry adjust- A direct alternative means to construct a linear
ment and wave shaping. The value of the wave- sinusoidal VCO requires designing a system
510 13 Waveform Generators and Non-linear Circuits
whose imaginary axis poles move linearly with a If R1 ¼ R then the characteristic equation is
control voltage Vc. Most such commercial VCOs given by C 1 C2 R2 s2 þ RðC1 C 2 Þs
are made using transconductors connected in a K 1 K 2 V C ¼ 0 . Setting C1 ¼ C2 and solving for
2
filter configuration whose poles are controlled the frequency of oscillation yields
via a transconductance. To better understand this rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
suppose the characteristic equation of a system or 1 K 1K2
f osc ¼ V ð13:35Þ
a bandpass filter was of the form s2 + a1s + a0. R C1 C2 C
Next we arrange it so that a1 ¼ 0 and then set
In practice R1 will be slightly less than R to get
a0 ¼ KV 2C where K is a constant. Setting a1 ¼ 0
this circuit to oscillate, and amplitude control is
ensures that the poles are on the jω axis, resulting
pffiffiffiffiffiffiffiffiffiffi required.
in a frequency of oscillation f osc ¼ KV C . A
simple circuit that accomplishes this is shown in
Example 13.12
Fig. 13.27 consisting of an op-amp and two
The variable frequency oscillator of Fig. 13.27 is
multipliers. Each multiplier is assumed to have a
set up using a TL082 op-amp and two multipliers
multiplier constant Ki, i ¼ 1, 2. The two
(AD534) whose K1 ¼ K2 ¼ 0.1. You may assume
multipliers are needed to generate the term Vc2.
that C1 ¼ C2 ¼ 1 nF and R ¼ 1 k. Determine the
If only one multiplier was used, then the fre-
frequency of oscillation if Vc is varied from 1 to
quency of oscillation would vary with the square
0.5 V.
root of Vc. To determine the frequency of oscilla-
tion, we must compute the loop gain and set it
Solution Using Eq. (13.35), the frequency of
equal to one. Breaking the loop at the point qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
indicated by the the loop gain of this circuit oscillation is given as f osc ¼ 1000
1 0:10:1
11018 c
V ,
can be computed as which for Vc ¼ 1 V yields fosc ¼ 100 kHz and
for Vc ¼ 0.5 V yields fosc ¼ 70.7 kHz.
K 1 K 2 V 2C
LG ¼
C 1 C 2 R1 R2 s2 þ C1 R1 R C2 R2 s þ R1 R
¼1 13.4 Monostable Multivibrators
ð13:34Þ
A monostable multivibrator produces a single
pulse for a fixed period of time T. Unlike the
13.4 Monostable Multivibrators 511
free-running or astable multivibrators discussed since the voltage at the inverting terminal is
earlier on, the monostable multivibrator has one greater than zero, the output is initially at Vo
stable state and a quasi-stable state. It is during and so is the voltage across the capacitor C. If a
the quasi-stable state that a pulse is generated for negative going pulse of sufficient amplitude is
a time T before returning to the stable state. To now applied to Vi, VR subsequently follows it
generate the pulse, however, requires that a trig- going negative first then positive. When this
ger input be applied to the circuit whose period happens the output of A1 switches to +Vo and
must be less than the desired period T. One such the capacitor voltage changes to Vo with the
simple but fairly accurate monostable pulse gen- node voltage Va ¼ 2Vo. At once the C begins to
erator and its associated waveforms are shown in discharge through R with time constant τ ¼ RC as
Figs. 13.28 and 13.29, respectively. D1 is open. Eventually Va falls to VR and the
The purpose of R1 and R2 is to establish a output switches back to Vo at which point Va
positive reference voltage V R ¼ R1RþR 1
2
V cc at the tries to follow but is clamped by D1 to 0.6 V.
inverting terminal of amplifier A1. In the stable With the amplifier output at Vo and Va at 0.6 V,
state, the voltage Va at the non-inverting terminal C again discharges through R until Va reaches
is zero because the capacitor is open circuit. Note zero. Since the capacitor discharges during
512 13 Waveform Generators and Non-linear Circuits
Example 13.13
A monostable multivibrator is constructed using
the circuit of Fig. 13.28. The values chosen were
R1 ¼ 1 k, R2 ¼ 2 k, R ¼ 1 k, C ¼ 0.1 μF and
Vcc ¼ 15 V. Determine the period of this
multivibrator if Vz ¼ 10 V.
Example 13.14
A monostable multivibrator is constructed using
the circuit of Fig. 13.30. The values chosen were
R1 ¼ 1.5 k, R2 ¼ 3 k, R ¼ 1 k and C ¼ 0.1 μF
with Vcc ¼ 15 V. Determine the period of this
multivibrator if Vz ¼ 10 V.
flip-flop whose state can be forced to reset by a high, the output returns to the stable low state and
reset pin. The output of the flip-flop is taken from the cycle is complete. The associated waveforms
Q where it is fed to an amplifier A3 that inverts Q are shown in Fig. 13.33.
to provide Q and be able to drive a load. The If the saturation voltage of Q1 is ignored, the
block diagram of the 555 timer is shown in capacitor voltage is governed by
Fig. 13.32 being used as a monostable
V c ðt Þ ffi V cc 1 eRC
1
multivibrator. In the stable state, S is low because ð13:38Þ
the trigger input Vi is larger than V3cc . The input
R is also low because the control voltage which is Since switching occurs when V c ðT Þ ¼ 2V3cc , it
equal to 2V3cc is larger than the threshold voltage Vc follows that
at the capacitor terminal. The output of the flip-
2V cc
ffi V cc 1 eRC
T
flop Q is subsequently high forcing Q1 on so that ð13:39Þ
3
C is shorted to ground. At this point the main
output Vo is approximately zero volts. Assuming from which the period T can be solved as
now that a trigger pulse comes along and its value
T ’ ln ð3ÞR1 C 1 ¼ 1:1R1 C 1 ð13:40Þ
is less than 2V3cc for a short period of time. When
this occurs, comparator A2 changes state, and Practical pulse widths can range from a few
S goes high forcing Q low and turning off the microseconds to several seconds with the
discharge transistor Q1. Capacitor C1 immedi- 555 possessing good temperature stability. The
ately begins to charge towards Vcc as the output propagation time from an input pulse to when
Vo remains high. Upon Vc exceeding 2V3cc , the the output is generated is in the order of tens of
output of the comparator A1 changes state forcing nanoseconds to a few hundred nanoseconds
R to go high and subsequently setting Q high depending on the technology and manufacturer.
which shorts out the capacitor C1 once more via Note in the circuit of Fig. 13.32, an optional
discharge transistor Q1. Immediately as Q goes 0.01 μF capacitor C is present to ensure that the
13.4 Monostable Multivibrators 515
T 2 ¼ 0:69Rb C ð13:42Þ
Example 13.16 2:88 kHz. The duty cycle D of the output wave-
þRb
An astable 555 oscillator is constructed using the form is D ¼ RRaaþ2R b
¼ 1000þ22000
1000þ2000
¼ 0:6 or 60%.
following components, Ra ¼ 1 k, Rb ¼ 2 k and
capacitor C1 ¼ 0.1 μF. Calculate the output fre-
quency from the 555 oscillator and the duty cycle
13.5 Precision Rectifiers
of the output waveform.
A non-linear op-amp function that is of consider-
Solution From Eq. (13.43) the frequency of
able importance is a precision rectifier. This func-
oscillation is f osc ¼ ð1 kþ221:44
kÞ0:1106
¼ tion arises since ordinary silicon diodes cannot
rectify signals of amplitude less than 0.7 volts, the
13.5 Precision Rectifiers 517
diode threshold voltage. Using the op-amp, a the negative half-cycle of the input signal, Vi goes
circuit that functions like an ideal diode that is negative, and therefore the voltage Vo1 goes posi-
capable of rectifying signals of a few millivolts tive. This causes diode D2 to be forward-biased.
can be realized. Such a circuit is called a precision Since the anode of diode D1 is connected to the
rectifier and can be classified as either a linear inverting terminal and is therefore at ground
half-wave rectifier or a full-wave rectifier. The potential, the effect of Vo1 going positive is to
linear half-wave rectifier delivers an output signal reverse-bias this diode. The input current I ¼ Vi/
that depends on the amplitude and polarity of the R therefore flows through Rf resulting in an output
input signal. It is also called a precision half-wave voltage Vo ¼ Vi since Rf ¼ Ri. The op-amp
rectifier. The precision full-wave rectifier delivers effectively operates like an inverter. Note that
an output signal that depends on the amplitude since during this half-cycle Vi is negative, the
but not the polarity of the input signal. It is some- inversion results in a positive-going signal. The
times referred to as an absolute value circuit. diode turn-on voltage of D2 is virtually eliminated
since small voltages at the input of the op-amp
results in amplified voltage Vo1 at the output of
the op-amp which turns on diode D2. Because this
13.5.1 Linear Half-Wave Rectifier
diode is within the feedback loop of the op-amp,
voltage Vo1 assumes the necessary level required
Linear half-wave rectifier circuits deliver one-half
to turn on this diode. The result is that signals at
cycle of a signal and eliminate the other half-cycle
the level of millivolts can be rectified. For the
while holding the output at zero volts. A circuit
case where both diodes are reversed, then
for achieving this is shown in Fig. 13.36. It
positive-going input signals are transmitted and
involves an inverting amplifier in which two
inverted, while the output is zero for all negative-
diodes are added thereby converting the amplifier
going signals.
into a precision half-wave rectifier. The operation
of the circuit is straightforward. During a positive
half-cycle of the input signal, Vi goes positive and
13.5.2 Signal Polarity Separator
therefore the voltage Vo1 at the output of the
op-amp goes negative. This causes diode D1 to
An interesting variation of the half-wave rectifier
conduct as a result of which the Vo1 is limited to
is the signal polarity separator shown in
0.7 V, the anode of the diode being connected
Fig. 13.37. In this circuit, signals of different
to the virtual earth at the inverting terminal. Diode
polarities are conveniently separated. Thus,
D2 is reverse-biased by Vo1. The input current
when the input signal Vi is positive, the output
I ¼ Vi/R flows into the input resistor R through
Voa of the op-amp goes negative thereby forward-
D1 into the output of the op-amp. The output
biasing diode D1 and reverse-biasing diode D2.
voltage Vo is therefore at zero potential. During
This causes current Vi/R to flow through the input
resistor R, through the feedback resistor RF1 ¼ R
and through forward-biased diode D1 into the
output of the op-amp. This results in an output
voltage Vo1 ¼ Vi with Vo2 ¼ 0 since D2 is off.
When the input signal Vi goes negative, the output
Voa of the op-amp goes positive thereby forward-
biasing diode D2 and reverse-biasing diode D1.
This causes current Vi/R to flow out of the output
of the op-amp, through forward-biased diode D2,
through feedback resistor RF2 ¼ R and through
Fig. 13.36 Inverting Linear Half-Wave Rectifier: Posi- the input resistor R into the input signal source.
tive Output This results in an output voltage Vo2 ¼ Vi with
518 13 Waveform Generators and Non-linear Circuits
Fig. 13.40 AC to DC
converter
that a quad op-amp IC containing four devices R3, the voltage from the junction into the
such as the LM324 may be used to implement this inverting terminal here can be varied from
system. approximately 0 100 mV corresponding to
the 0 10 V variation at the wiper of the poten-
Ideas for Exploration (i) Introduce a fifth tiometer. This level is comparable to the output of
op-amp comparator driving a green LED with the microphone. With a suitable voltage at the
the inverting terminal connected to the 5.6 V inverting terminal, if the output of the micro-
Zener reference voltage and the non-inverting phone exceeds this level, the output of the
terminal connected to the junction of R7 and R8. op-amp goes high switching on transistor Tr1
This will add a fifth battery voltage level indica- via R4 ¼ 10 k. This in turn switches on Tr2
tion of VB ¼ VM/β ¼ 5.6 V/0.4 ¼ 14 V; which further turns on Tr1, and the regenerative
(ii) replace R6 by a 10 k potentiometer so that action rapidly turns on both devices thereby
the threshold battery voltages can be varied if activating the relay. These transistors can be
desired; (iii) implement the original system MPSA05 and MPSA55 complimentary pair. The
using the 339 quad comparator which contains activated relay turns on the mains supply to an
four independent comparators. alarm. Should the output of the microphone be
reduced causing the op-amp output to go low, the
Sound-Activated Switch transistors remain on. They can only be turned off
The circuit of a sound activated switch is shown by opening the reset switch SW1 which interrupts
in Fig. 13.43. The output of a microphone drives the current flow through the transistors.
the non-inverting input of an op-amp, while the
inverting input is supplied with a reference volt- Ideas for Exploration This connection of
age which can be varied. Resistor R1 ¼ 2 k and transistors constitutes a thyristor which is
the potentiometer VR1 ¼ 10 k are connected to a discussed in Chap. 14. Explore the use of an
12 V supply. This enables the voltage at the wiper actual thyristor in place of the two transistors.
of the potentiometer to be variable between 0 and
10 V. This voltage is further reduced by another High-Speed Half-Wave Rectifier
potential divider arrangement comprising The half-wave precision rectifier of Fig. 13.44 is
R2 ¼ 100 k and R3 ¼ 1 k the junction of which implemented here using the AD844 current feed-
is connected to the inverting input of the op-amp. back amplifier (CFA). This device has a high slew
Because of the attenuation produced by R2 and rate of 2000 V/μs and a high closed-loop
13.6 Applications 523
in order that the system switch to +Vsat, the volt- k, then f ¼ 1/0.693(R1 + 2(R2 + VR1))C1 yields
age at the non-inverting terminal must go to zero. oscillation between 1 and 12 Hz as VR1 is varied.
This occurs when ðVR2sat Þ ¼ RV1i giving V i ¼ Thus, at the output on pin 3, a square-wave volt-
age signal is delivered to the base of the transistor
RR12 ðV sat Þ, and therefore the input Vi must rise
through resistor R3 ¼ 10 k. This forces a base
above the threshold value V i ¼ RR12 V sat for current into the transistor which, if the transistor
switching to occur. As in the inverting Schmitt is in good working order, will cause the flow of
trigger, there is a switching band centered on zero collector current that results in the turning on of
with switching levels at RR12 V sat . the LED. Resistor R4 ¼ 1 k limits the current
through the LED to 9 V/1 k ¼ 9 mA.
Ideas for Exploration (i) Apply a bias voltage
to the inverting input in order to shift the Ideas for Exploration (i) Introduce a switch
switching band to the left or right on the diagram. such that transistors of both polarities can be
easily tested.
Transistor Tester
The circuit shown in Fig. 13.47 is used to test npn Sawtooth Wave Generator
BJTs. Here, using the design information The circuit shown in Fig. 13.48 uses a 555 timer
presented in this chapter, the 555 timer is to generate a sawtooth waveform. It utilizes the
configured as an astable multivibrator. With 555 timer in the astable mode but charges the
R1 ¼ 1 k, R2 ¼ 100 Ω, C1 ¼ 100 μF and VR1 ¼ 10 capacitor using a constant current source. This
13.6 Applications 525
417 μA ¼ 24 k. Use R3 ¼ 10 k to ensure that the Tr2 turns on and capacitor C4 charges via diode
transistor turns on fully. D1 to the supply voltage. When the output goes
high, transistor Tr1 turns on, diode D1 turns off,
Ideas for Exploration (i) Replace the and diode D2 turns on. The series combination of
Darlington pair by a MOSFET such as the the 12 V at the output of the emitter followers and
IRF511. the charged capacitor C4 charges up capacitor C5
to twice the supply voltage or 24 V. A 75 mA
Voltage Doubler load current will cause the voltage across capaci-
The circuit in Fig. 13.51 converts a DC voltage to tor C5 to fall by 75 103/10 106 7 103 ¼ 1
twice its value with the same polarity. The output V before being recharged. Resistor R3 ¼ 220 Ω
of the 555 astable multivibrator with frequency limits the base current to the transistors while
7 kHz (as set by R1 ¼ 100 Ω, R2 ¼ 10 k, ensuring that they are fully turned on.
C1 ¼ 0.01 μF) drives a complimentary pair of
transistors (such as the MPSA05 and the Ideas for Exploration (i) Rearrange
MPSA55) connected in the emitter follower con- components D1, D2 so that the voltage across C5
figuration. With the arrangement of diodes D1 and is negative thereby realizing 12 V suitable for
D2 along with capacitors C4 ¼ 10 μF and powering an op-amp. The resulting circuit would
C5 ¼ 10 μF, as the output goes low, transistor then be a positive to negative DC converter.
13.6 Applications 527
Research Project 2
This research project involves the investigation of
the operation of a monostable multivibrator using
transistors and the design of an operating circuit.
Fig. 13.53 Astable multivibrator using transistors The basic configuration is shown in Fig. 13.54. It
comprises two common emitter stages with AC
and DC coupling such that there is one stable state
system to oscillate continuously as it switches
to which the system always reverts and one quasi-
back and forth between two quasi-stable states.
stable state into which the system can be triggered
Suppose Tr1 is on. Just prior while Tr1 was off,
by an external pulse. The stable state is Tr2 on
capacitor C1 would have charged up to the supply
and Tr1 off. If a negative-going pulse is applied to
voltage through R3 and the base-emitter junction
the base of Tr2 turning it off, the collector voltage
of Tr2 which would have been forward-biased.
rises driving base current into Tr1 thereby turning
With Tr1 on, the collector of Tr1 would go to
it on. Capacitor C1, which was charged through
almost zero volts and the voltage on C1 would be
R3 to voltage VCC while Tr1 was off and Tr1 on,
applied to the base-emitter junction of Tr2 taking
applies VCC to the base-emitter junction of Tr2
it to VCC thereby turning Tr2 off. As Tr2 turns
ensuring that it turns off. C1 now charges through
off, its collector voltage rises at a rate limited by
R1 until its voltage gets to 0.7 V turning on Tr2.
the need for C2 to charge through R4. During this
The collector voltage of Tr2 then falls and the
time C2 charges up to +VCC through R4 and C1
base current drive through R2 to the base of Tr1
charges through R1 until it reaches 0.7 V causing
goes to zero and Tr1 turns off. The system
Tr2 to turn on. As Tr2 turns on, the collector of
remains in this state until it is triggered again.
Tr2 would go to almost zero volts and the voltage
Note that a positive-going signal at the base of
on C2 would be applied to the base-emitter junc-
Tr1 will also trigger the system to move from the
tion of Tr1 taking it to VCC thereby turning Tr1
stable state into the quasi-stable state from which
off. As Tr1 turns off, its collector voltage rises at a
it returns after a fixed time T ¼ 0.693R1C1.
rate limited by the need for C1 to charge through
R3. During this time C1 charges up to +VCC
Ideas for Exploration (i) Introduce a speed-up
through R3 and C2 charges through R2 until it
capacitor across R2 that reduces the turn-off time
reaches 0.7 V causing Tr1 to turn on and the
of Tr1 by rapidly removing charge stored in the
cycle repeats itself. The signal out at either tran-
base; (ii) convert this system into a bistable
sistor collector is a square wave. The time T1 that
multivibrator by removing C1 and connecting R1
Tr1 is off can be shown to be T1 ¼ 0.693R1C1 and
in place of C1 as was done at Tr2 with C1 and R1
time T2 that Tr2 is off is T2 ¼ 0.693R2C2. Hence
to produce the monostable multivibrator.
the frequency of the square-wave output from
Problems 529
Fig. 13.60 Circuit for Question 10 period of this multivibrator if the voltage of
each Zener diode is VZ ¼ 8.2 V.
C1 ¼ C2 ¼ 0.003 μF and R2 ¼ 2.5 k. Deter- 17. Show how a JFET can be used to improve the
mine the frequency of oscillation if Vc is operation of the monostable multivibrator in
varied from 1 to 0.4 V. Fig. 13.63.
16. A monostable multivibrator is constructed 18. Describe the operation of the 555 timer and
using the circuit of Fig. 13.63. The values its use as an astable multivibrator. Hence
chosen were R1 ¼ 2 k, R2 ¼ 3 k, R ¼ 1 k, design an astable multivibrator using the
C ¼ 0.3 μF and Vcc ¼ 15 V. Determine the 555 to oscillate at 10 kHz. Show how a
diode can be used to achieve 50% duty cycle.
532 13 Waveform Generators and Non-linear Circuits
In previous chapters we discussed several semi- One specification of the device is the sensi-
conductor devices including the diode, the BJT, tivity which is the resistance at a specified illumi-
the FET, the operational amplifier and the current nation level. A typical curve containing this
feedback amplifier. There are several other information is shown in Fig. 14.2 where resis-
devices that are available to the designer and we tance is plotted against light intensity expressed
will discuss some of these in this chapter. At the in lux. For applications involving on-off
end of the chapter the student will be able to responses such as fire detectors, cells with steep
slopes are suitable since then a small change in
• Demonstrate knowledge of a variety of special illumination results in a large change in
devices resistance.
• Use these devices in a range of circuits For applications involving signals of varying
levels such as exposure control for cameras, cells
with low slopes are more appropriate. Another
specification is the dark resistance – the resistance
14.1 Light-Dependent Resistor of the cell under zero illumination. This
establishes the maximum current that can be
A light-dependent resistor (LDR) or photoresistor expected for a given voltage applied to the cell.
is a resistor that is sensitive to light. Specifically, This parameter is typically in the range 500 k to
the device lowers its resistance significantly in 20 M. The power rating of the cell in Watts is yet
response to exposure to light. It is sometimes another specification, and maximum cell voltage
referred to as a photoconductor, cadmium sulphide lies between 100 and 300 V. This voltage must
cell or photocell. It is made up of a semiconductor never be exceeded.
of high resistance. Energy absorption from incident Typical applications of the photoconductive
light stimulates bound electrons into the conduc- cell are shown in Fig. 14.3. The circuit in
tion band such that the resulting electron-hole pairs Fig. 14.3a produces an increasing output voltage
increase the conductivity of the material and con- level for increasing light intensity levels and is
sequently lower the resistance. They are used in suitable for ambient light measurement systems
many applications including camera light metres, used in camera exposure metres and brightness
street lights, clock radios, security alarms and out- control circuits. The circuit in Fig. 14.3b is a DC
door clocks. An example of an LDR is the ORP12. relay system that can be utilized in a range of
This device has a dark resistance of about 1 M applications including smoke detectors and night
which falls to about 5 k in bright light. The symbol lights. Light above a certain intensity causes the
for the LDR is shown in Fig. 14.1. resistance of the LDR to fall sufficiently such that
# Springer Nature Switzerland AG 2021 535
S. J. G. Gift, B. Maundy, Electronic Circuit Design and Application,
https://doi.org/10.1007/978-3-030-46989-4_14
536 14 Special Devices
the voltage across the base-emitter junction of the relay that can be used to switch on a desired
the transistor falls below 0.7 volts thereby turning system.
off the transistor. If the light falls below this
intensity, the resistance of the LDR increases Example 14.1
thereby turning on the transistor and activating Design a system using a photocell and an op-amp
as a comparator to switch on a light when ambient
light falls below a certain level.
(a) (b)
A photodiode is a diode that generates a cur- reverse saturation current arising from exposure to
rent or a voltage when exposed to light. Its sym- incident light continues to flow for zero reverse
bol is shown in Fig. 14.6. It operates as a reverse- bias. In curves 2 and 3 for short-circuited diode
biased pn junction and therefore experiences a terminals, a photo current ISC that is proportional to
reverse current arising from thermally generated the light intensity flows through the diode, while
carriers. As the reverse-bias voltage is increased, for open-circuited terminals, a voltage VOC appears
this current increases to its saturation value when across the diode terminals. ISC and VOC are
the current flow equals the rate of thermal gener- indicated on Fig. 14.7.
ation of the carriers. When such a junction is As is evident from the curves, the reverse cur-
exposed to increased illumination, additional rent increases as the incident luminous flux
electron-hole pairs are generated which result in increases. For a constant level of illumination, the
an increased reverse saturation current. This cur- reverse current increases only slightly with increas-
rent is approximately proportional to the intensity ing reverse bias indicating that the reverse current
of the light. The current vs voltage characteristic is largely independent of the reverse-bias voltage.
of the photodiode in the dark state is essentially It can also be seen from the curve that reverse
that of a conventional diode as shown in curve current flows even as the reverse bias is reduced
1 of Fig. 14.7. Upon illumination however the to zero and then to a small forward bias up to about
characteristic shifts downward to curve 2 and 0.3 V. Forward bias above this value causes the
then to curve 3 as light intensity increases. reverse current to fall, and it goes to zero when the
These curves do not pass through the origin since forward bias is approximately 0.5 V. “Dark cur-
rent” is the reverse saturation current of the device
in the absence of incident light.
The short-circuit current has a very linear rela-
tionship with incident light intensity. A typical
reverse (short-circuit) current vs illumination
characteristic for this device is shown in
Fig. 14.8. Here luminous flux is the total amount
of light emitted or received by a surface, and its
unit of measurement is the lumen (lm) where
Fig. 14.6 Symbol of photodiode
CURRENT
1000
Revers e Curren t (μA)
100
SATURATION
CURRENT
0 VOLTAGE 10
1 Voc
Voc’
Isc 1
2
Isc’
3 0.1
INCREASING 10¹ 10² 10³ 10⁴
LIGHT LEVEL Illuminance (lx)
Fig. 14.7 Current vs voltage characteristics for a Fig. 14.8 Typical reverse current vs illuminance for
photodiode photodiode
Photodiode 539
0.6
0.5
ResisƟvity (A/W)
0.4
0.3
0.2
0.1
0
300 400 500 600 700 800 900 1000 1100
Wavelength (nm)
Ic (mA)
1
E=1000 Lx
E=750 Lx
0.5
E=500 Lx
E=250 Lx
0 1 2 3 4 5
Fig. 14.13 Circuit for Example 14.4 VCE (Volts)
one side of a doorway and an opto-transistor Fig. 14.20. This device provides low noise and a
receiver Tr1 positioned on the other side. Resistor continuously variable resistance that varies from
R1 ¼ 220 Ω sets the current through these leds. 15 Ω to 50 MΩ in response to a current input. Its
Both are positioned such that radiation from the characteristics are typically an isolation voltage of
leds falls on the phototransistor thereby activating 2500 Vrms, an input current of 10 mA and an
it. The current flow turns on the green LED D2 in output voltage of 30 V. Applications of this
the emitter circuit of the phototransistor. This pulls device include audio limiting and compression,
the base of Tr2 low, and this transistor turns off the automatic gain control, circuit isolation and logic
red LED D3 in its emitter circuit. The potentiometer interfacing, SCR and triac drivers and noiseless
VR1 ¼ 2 k is adjusted to ensure that Tr2 is off under switching.
normal conditions. If the infrared radiation is One example shown in Fig. 14.21 is a remote
interrupted by entry through the doorway, then gain control system. Here the light-dependent
the opto-transistor turns off causing the green resistor is used as the feedback resistor in an
LED to go off. The base of Tr2 goes high causing inverting amplifier. This enables the remote con-
Tr2 to turn on and switching on the red LED. trol of the gain of the circuit using the associated
Resistor R2 ¼ 1 k limits the current in the red light-emitting diode. The light-dependent resistor
LED. A buzzer may be included between the emit- is activated by light from the LED, and then the
ter and ground of Tr2 to provide an audible alarm. reduced resistance of the light-dependent resistor
causes the magnitude of the output voltage of the
There is also the photodarlington which is a amplifier to be reduced. Another application
Darlington pair that is photosensitive. In such a example is the switching circuit of Fig. 14.22. In
device, the first transistor is photosensitive, and this circuit the light-dependent resistor is included
its emitter is connected to the base of the second in a potential divider arrangement with another
transistor. The resulting gain is therefore higher resistor R that is connected to a 5-volt supply.
than the normal phototransistor, but the response When a logic signal is applied to the LED, light
of the photodarlington is slower. from the LED falls on the LDR, and then the
resistance of the LDR falls and hence so does
the voltage at the junction of these two resistors.
14.4 Opto-isolator
This voltage change causes the logic gate to 14.4.2 Photofet Opto-isolator
change state.
Another opto-isolator is one with a photofet at the
Exercise 14.1 output as shown in circuit Fig. 14.23. An example
Explore the use of the photoresistor opto-coupler of this is the H11F series from Fairchild Semi-
in the design of (i) a voltage-controlled Sallen- conductor. Each device contains a gallium-
Key low-pass filter and (ii) a voltage-controlled aluminium arsenide infrared-emitting diode cou-
Wien bridge oscillator. In both cases, two pled to a bi-directional photosensitive FET that
photoresistor opto-couplers would have to be responds as a light-dependent resistor. The
used to simultaneously vary two resistors in arrangement functions as an isolated FET with a
these systems. resistance variation from 100 Ω to 300 MΩ. Its
characteristics include a maximum input diode
forward current of 60 mA, isolation resistance of
greater than 100 Gohms, continuous detector cur-
rent of 100 mA and a breakdown voltage of
30 V (Fig. 14.23).
The circuit element can function as a variable
resistor as it has an extremely linear relationship
between diode current and circuit resistance. It
can also be used as an electronic switch. The
bi-directional photofet and the optical isolation
provided by the photodiode make this circuit
element a very versatile solid-state relay. Two
Fig. 14.22 Logic interface applications are shown in Fig. 14.24. They are
both variable attenuators that allow complete iso-
lation of the control signal.
This is the photoconductive mode. This current an input voltage signal Vi applied at the
flow through a resistor is easily detected as a non-inverting input of op-amp A drives the IR
voltage. This opto-coupler responds very quickly LED such that a reverse current flows through
to changes in the LED current and is useful where diode D1 that develops a voltage across R1
opto-coupling with high speed is required. which, because of the feedback is equal to Vi.
Diode D2 is similarly connected to a separate
Example 14.6 supply Vcc2 ¼ Vcc1 with the output connected to
Outline a system in which the photodiode opto- resistor R2 ¼ R1 which is itself connected to a
coupler is used in a linear opto-isolator system. voltage follower op-amp B. The currents in
diodes D1 and D2 are equal, and therefore the
Solution One IC that is designed for linear signal voltage developed across R2 is equal to Vi
isolation is the IL300 linear opto-coupler from resulting in the voltage Vo ¼ Vi.
Vishay. It comprises an infrared (IR) LED D3
irradiating two photodiodes D1 and D2 in one
package. Photodiode D1 is used to generate a
14.4.4 Phototransistor Opto-isolator
feedback control signal that is used to establish
the IR LED drive current necessary to produce a
The opto-isolator using a photo transistor is
linear signal output from photodiode D2. A sim-
shown in Fig. 14.27. The light from the LED
ple implementation is shown in Fig. 14.26. Here
activates the phototransistor and causes it to con-
op-amp A is connected in a non-inverting config-
duct. An example of such a device is the 4N25
uration and its output drives the IR LED D3.
from Vishay Semiconductors. Opto-isolators
Resistor R3 assists in limiting the value of this
using a photodarlington are also available. The
current. The cathode of photodiode D1 is
connected to a reverse-bias supply Vcc1, and the
anode is connected to feedback resistor R1. Thus,
4N32 is an example. Such circuits are ideal for opto-coupler LED. When the transistor is
interfacing one family of logic circuits to another switched on by the microcontroller, the LED
or for isolating an analog or digital system from radiates and the opto-transistor is activated. This
the physical world. The current to the LED must causes the output of the opto-transistor to go low,
as usual be limited by an external resistor in series and this signal level is coupled to a separate
with the LED. The LED can be driven from an circuit that can be completely electrically isolated
AC source providing a diode is connected in from the microcontroller.
series with the LED. Such a diode will protect
the LED from reverse DC voltages. The current Example 14.7
through the phototransistor can be converted to a Use the phototransistor opto-coupler to interface
voltage by connecting a resistor to either the two analog audio systems.
collector or the emitter as shown in Fig. 14.16.
A typical application is shown in Fig. 14.28. Solution One approach to isolating two audio
Here, the digital output of a microcontroller is systems is shown in Fig. 14.29. Here a quiescent
coupled to the base of a transistor that drives the current is established through the LED of the
FORWARD CONDUCTION
REGION
HOLDING CURRENT IH
NO GATE CURRENT
REVERSE BLOCKING
V
REGION
FORWARD BLOCKING
BREAKDOWN
AVALANCHE
REGION
FORWARD BREAK-
REVERSE BREAK-OVER VOLTAGE OVER VOLTAGE
REVERSE BLOCKING
V
REGION FORWARD BLOCKING
REGION
BREAKDOWN
AVALANCHE
This minimum value is that which enables the holding current value. In the on state, the forward
device to remain conducting even after the voltage across the thyristor is typically around
removal of the gate current. Cessation of the 1 volt.
gate current before the holding current level is It is possible to turn on a thyristor by applying
attained will cause the thyristor to turn off again. a forward voltage whose rate of change dV/dt
Once the thyristor is turned on, the current must exceeds some critical value specified by the man-
exceed the holding current in order to sustain the ufacturer. In this turn-on mode, the voltage across
thyristor in the conducting state. In such the device does not have to exceed the breakover
circumstances the device remains in the voltage for switch on to occur. At this critical
conducting state independently of the gate value of dV/dt, the rapid change of voltage across
which no longer exercises any control. The thy- the SCR results in a flow of current through the
ristor current may vary but must not fall below the internal junction capacitances. In particular a flow
550 14 Special Devices
Static Switching
SCR static switching circuits use either a constant
or varying DC signal for SCR turn-on for both
DC and AC power control. One simple approach
is the use of manual switching and a DC bias
supply to turn on the device as shown in
Fig. 14.36. The closure of normally off switch
SW1 delivers a voltage to the gate of the SCR via Fig. 14.38 Static switching of SCR using normally
the resistor RG thereby turning it on. Normally on closed switch
switch SW2 is used to interrupt the current to the
SCR in order to turn it off. The DC bias voltage gate of the SCR. Several switches can then be
may be removed and resistor RG returned to the placed in series so that triggering can again take
supply voltage as shown in Fig. 14.37. Several place at several locations.
switches may be added in parallel with SW1 so
that the SCR can be turned on from many Example 14.8
locations. Switch SW1 can be moved to a gate- A thyristor has IG ¼ 500 μA, VGC ¼ 0.7 V,
ground position where it must be normally closed VAC ¼ 0.2 V and IH ¼ 10 mA. Using VTRIG ¼ 3 V
as shown in Fig. 14.38. When opened, a trigger- and VCC ¼ 24 V, determine the resistor RG in
ing signal is applied through resistor RG to the order to trigger the thyristor into conduction and
Silicon-Controlled Rectifier 551
the resulting anode current for a load RL ¼ 100 Ω exceed the peak reverse voltage to which it will
in the circuit of Fig. 14.37. be subjected. Diode D1 prevents the gate cathode
junction of the SCR from being subjected to
Solution From Kirchoff’s voltage law, reverse voltages. Again, the normally open switch
VTRIG ¼ IGRG + VGC from which SW1 may be replaced by a normally closed
RG ¼ (VTRIG VGC)/IG ¼ (3 0.7)/500 μA ¼ 4.6 switch between the gate and the cathode.
k. The anode current IA is found from
VCC ¼ IARL + VAC. Hence IA ¼ (VCC VAC)/ Phase Control Switching
RL ¼ (24 0.2)/100 ¼ 238 mA. The basic system can be modified to introduce
phase control as shown in Fig. 14.41. Here poten-
A simple static switch for AC power control to tiometer VR1 in conjunction with resistors R1 and
a load is shown in Fig. 14.39. It is a basic half- R2 allows the adjustment of the triggering voltage
wave circuit that is triggered by a signal through to the SCR to switch the SCR at a specified point
resistor R1 to the gate of the SCR when SW1 is in the signal cycle. This is shown in Fig. 14.42
closed. The SCR conducts for one half-cycle and
turns off for the other half as the waveform goes
through zero. The load waveform is therefore a
half-wave rectified AC as shown in Fig. 14.40.
The reverse voltage rating of the SCR must
Fig. 14.39 Static AC power switch Fig. 14.41 Phase control switching of SCR
AC
Input
Load
Current
t
Load
Current
Example 14.9
For an operating frequency of 60 Hz, find the
capacitor value required to achieve a phase lag
of 50o with a resistor of 12 k.
Fig. 14.43 Extended phase control switching of SCR
Solution From Eq. (14.2),
where phase adjustment from 0o to 90o can be tan 50o ¼ 12 103 2 π 60 C. Therefore
effected corresponding to SCR conduction for C ¼ 1210tan 50o
¼ 0:26 μF.
3
2π60
180o to 90o and the shaded area represents non-
conduction. For phase adjustment from 0o to 180o
Example 14.10
corresponding to SCR conduction for 180–0o, the
Design a static switch to control power to an AC
circuit must be further modified by replacing R2
load through the action of an SCR being powered
by a capacitor C1 and introducing another diode
by a 15 vAC supply.
D2 as shown in Fig. 14.43. The capacitor
introduces delay in the triggering of the SCR as
Solution The basic system is shown in
C1 charges via VR1 and R1 to the triggering volt-
Fig. 14.39. It comprises an SCR in series with a
age level. Since the voltage across the capacitor in
load with the arrangement being connected to the
a series RC network lags the driving AC voltage
mains supply. With the switch open, no activating
by an angle θ given by
pulse can be applied to the gate, and therefore the
1 R SCR remains off. If the switch is closed on the
θ ¼ tan ð14:2Þ positive half-cycle, a current pulse is supplied to
XC
Silicon-Controlled Rectifier 553
Load
Current
the gate through the resistor causing it to turn on Therefore, only unidirectional flow occurs
and thereby connect the load to the mains supply. through the load and SCR. As the rectified AC
The device will turn off during the negative half- voltage rises across the resistor-potentiometer
cycle of the supply but will turn on again on the potential divider, the potentiometer is adjusted
positive half-cycle. The result is that the supply to so that triggering of the SCR occurs at an appro-
the load is a half-wave supply. The diode ensures priate point in the AC cycle thereby realizing
unidirectional gate current flow. The value of the phase control. The resulting load waveform is
resistor R1 is calculated using the peak supply shown in Fig. 14.46. The potentiometer enables
voltage say 4 volts at which turn occurs and the adjustment of the firing angle from 0o to 90o for
gate current value required to trigger the thyristor. each half cycle. For phase adjustment from 0o to
For the 2N3669, the trigger current has a maxi- 180o corresponding to SCR conduction for
mum value of 40 mA giving R ¼ 4010 4
3 ¼ 100 Ω. 180–0o, the circuit must be modified by replacing
This value of gate resistor allows for the unknown R2 by a capacitor C1 as shown in Fig. 14.47. The
load resistance which will reduce the gate current capacitor introduces delay in the triggering of the
below the 40 mA value used in the calculation. SCR as C1 charges via VR1 and R1 to the trigger-
ing voltage level. The resulting load waveform is
Full-Wave Control shown in Fig. 14.48.
The circuit in Fig. 14.45 is designed to control AC
power to a DC load using an SCR. The load and A simple application of this system is a tem-
the SCR are connected after the diode bridge. perature controller as shown in Fig. 14.49. The
554 14 Special Devices
Load
Current
AC
Input
Load
Current
Fig. 14.48 Waveform resulting from extended phase control of AC power to DC load
Silicon-Controlled Rectifier 555
diode bridge arrangement produces unidirectional potential divider, the potentiometer is adjusted
current flow through the thyristor, while the resis- so that triggering of the SCR occurs at a specified
tor, capacitor and thermostat together regulate the point in the AC cycle thereby effecting phase
gate drive to the device. At low temperatures control. The effect on the load waveform is
when the thermostat is open, for each half-cycle shown in Fig. 14.51. Using the potentiometer,
of the supply voltage, the capacitor is charged the firing angle can be varied from 0o to 90o for
through the resistor to a potential that triggers each half cycle. Again, for phase adjustment from
the SCR, thereby allowing current flow to the 0o to 180o corresponding to SCR conduction for
heater. As the ambient temperature rises, the ther- 180–0o, the circuit must be modified by replacing
mostat closes at a specific temperature and short R2 by a capacitor C1.
circuits the gate-cathode terminals as a result of
which the SCR cannot be triggered. No power is
then delivered to the heater and the temperature 14.5.2 Gate Turn-Off Switch
falls.
The full-wave circuit in Fig. 14.50 is designed The symbol for the gate turn-off switch is shown
to control AC power to an AC load using an SCR. in Fig. 14.52. This device is effectively an SCR
The load is connected before the diode bridge, that can be turned off (and on) at the gate-cathode
while the SCR is connected after the bridge. junction. This feature is very useful in many
Therefore, only unidirectional flow occurs power-switching applications. This device can
through the SCR. As the full-wave rectified AC both be turned on and off at the gate which results
voltage rises across the resistor-potentiometer in a simplification of the associated circuitry.
AC
Input
Load
Current
connected to both p-type and n-type material as and this enables gate-T1 voltage turn-on to be of
shown. This enables conventional current flow in either polarity. Like the thyristor, the current
either direction, i.e., from T1(+) to T2() via p1, through the triac must exceed the latching current
n3, p2 and n2 and from T2(+) to T1() via p2, n3, before the cessation of the gate current if the
p1 and n1. The gate terminal G is also connected device is to stay on. When turned on, the device
to both semiconductor types namely p1 and n4, current must not fall below the holding current.
The static characteristic of the triac is shown in
Fig. 14.57. Again, similar to the thyristor, the
T1 T1 forward breakover voltage falls as the gate trigger
current is increased. The device operates in either
Gate n4 n1
the first quadrant when terminal T1 is positive
Gate
p1 relative to terminal T2 and in the third quadrant
n3
p2 when T2 is positive relative to T1.
The triac like the SCR will be turned on if the
n2 rate of change of the terminal voltage exceeds a
T2 specified value. A snubber circuit can again be
T2 used to prevent turn-on by sharp voltage changes.
Fig. 14.56 Triac construction
MINIMUM HOLDING
BREAKOVER VOLTAGE CURRENT GATE TRIGGERING RANGE
V
BREAKOVER VOLTAGE
MINIMUM HOLDING
CURRENT
Fig. 14.58 Static AC Switch Fig. 14.59 Triac phase control circuit
AC
Input
Load
Current
Load
Current
varying the gate resistor value and hence the time R
in the half-cycle when the triac switches on. By θ ¼ tan1 ð14:2Þ
XC
adding a capacitor as shown in Fig. 14.61, the
resulting RC phase-shifting network enables the where θ is the phase angle of the capacitor voltage
increase of the firing angle to almost 180 as in degrees, XC is the capacitor reactance and R is
shown in Fig. 14.62. the sum of R1 and VR1, Eq. (14.2) can be used to
The capacitor introduces delay in the trigger- approximately determine component values.
ing of the triac as C1 charges via R1 and VR1 to the When driving inductive loads, a snubber cir-
triggering voltage level. Since the voltage across cuit consisting of a resistor (100 Ω) in series with
a capacitor in an RC network lags the input AC a capacitor (0.1 μF) placed across the terminals of
voltage by an angle θ given by the triac may be used to protect against the effects
of rapid increases in terminal voltage.
560 14 Special Devices
Holding Current
Switching Current
Reverse Breakdown
Voltage
V
VMN Break-Over
Voltage
fans or pumps can be switched, but a snubber holding current IH. Once the device begins to
circuit may be necessary. conduct, it will continue until the anode current
is reduced to a value less than the holding current.
This device is generally designed for unidirec-
14.7 Shockley Diode tional operation and is often used in SCR trigger-
ing circuits.
The Shockley diode in Fig. 14.65 is essentially an One simple application is in a relaxation oscil-
SCR without the gate terminal. Its characteristic lator as shown in Fig. 14.67a. Closure of the
therefore corresponds to that of the SCR with zero switch results in the charging of the capacitor
gate current as shown in Fig. 14.66. From this and increase of the voltage across the capacitor.
characteristic it can be seen that for a voltage When the voltage reaches the breakover voltage
applied across the anode and the cathode, little of the Shockley diode, the diode conducts thereby
or no current flows. In the case of a reverse discharging the capacitor and the capacitor volt-
voltage, a value exceeding the reverse breakdown age falls. Once the diode current drops below the
voltage of the device (typically 10–20 volts) holding current of the diode, the diode turns off
causes the device to turn on. It is not intended and the capacitor begins to charge again. The
that the device operates in this mode. When the capacitor waveform is shown in Fig. 14.67b.
forward voltage is increased beyond the The rising waveform can be made linear by
breakover voltage VBR, current flow increases using a constant current source to charge the
rapidly and the device turns on when this current capacitor.
exceeds the latching value. The switching current
at this point is IS. The terminal voltage drops Example 14.12
significantly to a voltage VMN, and conduction Design a relaxation oscillator to produce a linear
continues providing the current exceeds the waveform with a slope of 10 mV/μs. Use a
562 14 Special Devices
Fig. 14.67 Shockley diode. (a) Relaxation oscillator and (b) output waveform
3
and therefore C1 ¼ I= dV
dt ¼ 10 A=104 Vs1 ¼
0:1 μF: A rail-to-rail op-amp such as the
LMC6482 is used in the voltage follower mode Fig. 14.68 Circuit for Example 14.12
to buffer the capacitor voltage and deliver an
output voltage. The peak output voltage is
VBR ¼ 12 V with a minimum value less than
VMN ¼ 1 V.
14.8 Diac
Holding
Break-Over Current
Voltage
Break-Over
V
Holding Voltage
Current
voltage to a value less than 1 V. This is shown in to its former value. While the triggering action of
Fig. 14.76. the UJT involves the channel region close to B1
The UJT is now triggered (sometimes referred that is represented by resistor RB1, the channel
to as the ON state), and the maximum current region close to B2 that is represented by resistor
flows through the channel. When the emitter volt- RB2 is largely unaffected by this action and the
age falls to a sufficiently low value, the diode associated emitter current flow. The emitter volt-
turns off and the effective resistance of RB1 rises age that results in the forward-biasing of the pn
junction and the flow of an emitter current is
usually designated VP and is given by
V P ¼ ηV BB þ 0:7 ð14:6Þ
Example 14.15
A UJT with η ¼ 0.75 is powered by a 10-volt
supply. If the emitter is at zero potential, deter-
mine the reverse bias across the pn junction.
Fig. 14.75 UJT operating configuration
VE (V)
Cut-Off Negative Resistance Saturation
Region Region Region
VP
VV
I E (mA)
IP IV
Solution From Eq. (14.5), triggered and the capacitor discharges through the
VRB1 ¼ ηVBB ¼ 0.75 10 ¼ 7.5 V. Hence the emitter through RB1 and R2 with time constant
reverse bias across the pn junction is 7.5 V. (RB1 + R2)C1 that is usually much smaller than
Unijunction transistors are available in power R1C1. Therefore, the charge time is usually much
ratings up to about 500 mW with voltage ratings larger than the discharge time. When the capaci-
up to about 60 V and peak currents ranging tor voltage falls to VE ¼ VV, the diode turns off,
between 50 mA and 2 A. A very common appli- the UJT returns to its original state, capacitor
cation of this device is in a relaxation oscillator. charging again commences, and the cycle is
The basic system using the widely available repeated. The waveform at the emitter and at B1
2N2646 UJT is shown in Fig. 14.77. are shown in Fig. 14.78. The former results from
the capacitor charge and is an approximate saw-
When the switch is closed, the capacitor tooth wave. The latter signal results from the
charges exponentially through resistor R1 with a current flow through resistor R2 during the dis-
time constant R1C1. As a result, the voltage at the charge process and is a short duration pulse.
emitter rises. When VE ¼ VP, the unijunction is The peak value of the voltage at the emitter is
VP, while the peak value of the voltage at B1 is
given by
R2
V B1 ðpk Þ ¼ V ð14:7Þ
R2 þ R3 þ RBB ðmnÞ BB
giving
0:9
f ¼ 1=R1 C ln ð14:10Þ
1η
Example 14.16
Using the 2N2926 unijunction transistor, design a
relaxation oscillator with a frequency of oscilla-
tion of 1 kHz and operating from a 9-volt supply.
VE (V)
Cut-Off Negative Resistance Saturation
Region Region Region
VP
VV
I E (mA)
IP IV
nonconducting state. When the anode voltage VA when the PUT is off, the output voltage Vo is
exceeds the gate voltage by 0.7V such that its zero. When the device is triggered on, the rapid
value equals VP where rise in current causes a sudden rise in Vo to a value
that can be controlled.
V P ¼ ηV BB þ 0:7 ð14:13Þ
A relaxation oscillator can also be
the anode-gate junction becomes forward-biased implemented using a PUT such as the 2N6027.
and an anode current IA begins to flow. This value The basic system is shown in Fig. 14.82. When
of anode voltage is called the peak voltage VP as power is applied, the capacitor charges exponen-
shown on the anode characteristic in Fig. 14.81. tially through resistor R4 with a time constant
The PUT now rapidly turns on and the anode R4C. As a result, the voltage at the anode rises.
voltage falls as the current increases down to the When VA ¼ VP, the PUT is triggered and the
valley point. This corresponds to negative resis- capacitor discharges through the PUT with time
tance over this portion of the characteristic. Simi- constant R3C that is usually much smaller than
lar to the UJT, at the peak point the peak emitter R4C. As a result, the time to charge is usually
voltage VP is associated with a peak emitter cur- much larger than the time to discharge. When the
rent IP, while at the valley point, the minimum capacitor voltage falls to VA ¼ VV, the pn junction
emitter voltage VV is associated with a minimum becomes reverse-biased and the PUT turns off. It
emitter current IV. External resistors R1 and R2 in then returns to its original state at which time
the PUT correspond to the internal resistors RB1 capacitor charging again commences and the
and RB2 in the UJT. By adjusting these resistor cycle is repeated. The waveforms at the anode
values, the PUT can be programmed to have and cathode are shown in Fig. 14.83. The former
different values of VP and IP. The output is usu- results from the charging and discharging of the
ally taken at the cathode with the current IA devel- capacitor and is an approximate sawtooth wave.
oping a voltage drop across resistor R3. Thus, The latter signal results from the current flow
570 14 Special Devices
V BB V V
t 1 ¼ R4 C ln
V BB V P
V BB
¼ R4 C ln
V BB ηV BB
1
¼ R4 C ln ð14:14Þ
1η
1
f ¼ 1=R4 C ln ð14:16Þ
1η
Example 14.17
Using the 2N6027 PUT, design a relaxation oscil-
lator with a frequency of oscillation of 1 kHz and
operating from a 9-volt supply.
1=R4 C ln 1
¼ 1=R4 C ln 1
’ 1=R4 C: calculation. The battery is charged during this
1η 10:63
part of the cycle. The transistor Tr1 is used to
This gives C ¼ 1/R4f ¼ 1/100 103 103 ¼ 0.01 stop the charging process when the battery is fully
charged. It is switched on when the voltage at its
μF. A small resistor R3 ¼ 100 Ω enables a pulse
base exceeds the turn-on value. The battery volt-
voltage out while not affecting VP.
age at which this occurs is set by the potential
divider comprising R3 ¼ 8.2 k, potentiometer
Finally, the close relationship between the UJT
VR1 ¼ 1 k and R4 ¼ 1 k in conjunction with the
and the PUT is demonstrated in Fig. 14.84 where
diode D2. The turn-on voltage is the sum of the
the UJT symbol and the UJT equivalent circuit
transistor base-emitter voltage and the diode volt-
are compared with a PUT drawn as a UJT.
age giving 1.4 V. The maximum battery voltage
VB at which the transistor turns on occurs when
the potentiometer wiper is at the lower position
14.11 Applications and is given by 1:4 1 k ¼ ð1þ1þ8:2Þ k . This gives
VB
AC power to a heater. Potentiometer VR1 along heater as the supply voltage to the triac goes
with the NTC (negative temperature coefficient) through zero. If, however, at the end of the cycle
thermistor R2 and resistor R3 sets a voltage at the the temperature has not exceeded the set point,
trigger input pin 2. In the steady state with a then the voltage at pin 2 will be less than Vcc/3
temperature above a set point such that the resis- and the 555 keeps the triac on until the set point
tance of the thermistor results in a potential at pin temperature is exceeded after which turn-off
2 which is above the threshold voltage of Vcc/3, occurs. At the specified temperature, the thermis-
the timing capacitor C1 is clamped at ground tor must have a resistance R2 such that
potential by the internal transistor connected to VR1 + R2 ¼ 2R3. This corresponds to a voltage
pin 7, and the output at pin 3 is low thereby at pin 2 of Vcc/3. Using a thermistor with a base
ensuring that the triac is off. If the temperature resistance of 5 k (at 25oC) along with VR1 ¼ 2k
falls below the specified value, the resistance of and R3 ¼ 3.3 k allows variation of the set point by
the thermistor increases resulting in the voltage at adjustment of the potentiometer. Decreasing VR1
the trigger input falling below Vcc/3. This starts results in a higher potential at pin 2 and therefore
the timing cycle and the output goes high, turning a lower set point temperature. Increasing VR1
on the triac and power to the heater. If as a result towards its maximum value reduces the voltage
the temperature rises above the set point and the at pin 2 and hence increases the set point temper-
thermistor resistance falls causing the voltage at ature. The cycle time T is set by R1 and C1 based
pin 2 to rise above Vcc/3 before the end of the on T ¼ 1.1R1C1. For R1 ¼ 1 M and C1 ¼ 10 μF,
timing cycle, then the output goes low at the end we get T ¼ 1.1 106 10 106 ¼ 11 s. Using
of the cycle thereby switching off power to the a 2N6071 triac which has a gate trigger current
Applications 573
system is shown in Fig. 14.91. Here a 555 timer is op-amp which is a rail-rail device so that it can
connected as an astable multivibrator. Capacitor be operated from a single-ended supply while
C1 is charged via resistor R1 and diode D1 and transmitting low voltages to its output. When the
discharges via resistor R2 and diode D2. By stairstep voltage across C3 reaches a threshold
making R2 appropriately larger than R1, the out- value, the unijunction Tr2 fires and discharges
put of the 555 is a set of short duration pulses. this capacitor. This discharge produces a voltage
These pulses charge capacitor C3 through VR1 pulse across resistor R4 which turns on transistor
and R3. These pulses cause the voltage across C3 Tr1 and resets the 555. The cycle then starts again
to increase in steps thereby producing a stairstep with pulses from the 555 delivering charge to
waveform. This waveform is buffered by the capacitor C3 and building a stairstep of voltages.
576 14 Special Devices
Potentiometer VR1 sets the charge current into the induced current can be converted to a useable
capacitor and therefore varies the size of the steps output voltage.
while VR2 sets the voltage across the UJT and 5. Using the photoresistor opto-isolator along
therefore sets the number of steps before dis- with an op-amp, design a remote gain control
charge occurs. amplifier system.
6. Outline the approach by which the photofet
Ideas for Exploration (i) Use this stairstep gen- opto-coupler can be used in the design of (i) a
erator to develop a transistor curve tracer. See the voltage-controlled Sallen-Key low-pass filter
article “Transistor and FET Curve Tracer” by and (ii) a voltage-controlled Wien bridge
Daniel Metzger, Electronics World, Vol. 86, No. oscillator.
2, p52, August 1971, where a system that 7. A microcontroller with a 0–5 V output is used
employs a stairstep generator is discussed. to activate another digital system from which
it must be isolated. Use the configuration
shown in Fig. 14.93 to realize such a system
with VCC1 ¼ VCC2 ¼ 5 V.
Problems
8. A thyristor used in a static power switch has
IG ¼ 750 μA, VGC ¼ 0.7 V, VAC ¼ 0.4 V and
1. Design a system using a photocell and an
IH ¼ 15 mA. Using VTRIG ¼ 4 V and
op-amp as a comparator to switch on a
VCC ¼ 36 V, determine the resistor RG in
motor when ambient light exceeds a certain
order to trigger the thyristor into conduction
level.
and the resulting anode current for a load
2. Design a system using a photocell and a
RL ¼ 700 Ω in the circuit of Fig. 14.37.
transistor to switch on an alarm when ambi-
9. For an operating frequency of 300 Hz, find
ent light rises above a specified level. Indi-
the capacitor value required to achieve a
cate the changes needed for the system to
phase lag of 75o with a resistor of 56 k.
indicate the absence rather than the presence
10. Describe the operation of a static triac switch
of light.
for controlling AC power to a load. Show
3. Design a light metre to cover the ranges
how phase control can be introduced to
10–10,000 lx using the photo diode in the
allow control over the full half cycle.
zero bias mode shown in Fig. 14.92 and a
11. Design a system to remotely switch on a large
1 mA milliameter.
(8 A) heating load using the phototriac and a
4. Explain the operation of the phototransistor
triac.
and indicate one method by which the light-
12. Using a Shockley diode having VS ¼ 15 V,
VH ¼ 1.3 V and IH ¼ 5 mA, design a
Fig. 14.92 Circuit for question 3 Fig. 14.93 Circuit for question 7
Bibliography 577
half-wave rectifier (see Half-wave rectifier) Enhancement MOSFET common drain amplifier, 153, 154
Zener diode regulators, 28 (see Zener diode regulators) Enhancement MOSFET common gate amplifier, 154
Diode specifications, 12 Enhancement MOSFET common source amplifier, 151,
Diode tester, 33, 34, 82 153
Diode types Enhancement-type MOSFET
LED, 14 channel conductivity, 108
photodiodes, 15, 16 drain characteristics, 108
PIN diode, 16 gate-source voltage, 108
power diodes, 14 mutual characteristics, 108, 109
Schottky diode, 16 n-channel, 106, 107
signal diodes, 14 p-channel, 109, 110
varactor diodes, 14 positive voltage, 107
Zener (see Zener diodes) reverse-biased, 107
Diodes applications symbols, 109, 110
battery charger, 31, 32 Enhancement-type MOSFET common drain amplifier,
blown fuse indicator, 33 114
crystal radio, 36, 37 Enhancement-type MOSFET common gate amplifier,
diode tester, 33, 34 112–115
emergency phone-line light, 34 Equivalent amplifier circuit, 222
lead-acid battery monitor, 34 Equivalent circuit, 224, 227
light meter, 32, 33 Error voltage, 259
phone alert, 34 Extrinsic (impure) semiconductors
polarity indicator, 32 crystal lattice, 5
research project, 36 doping, 5
TL431A, 36 n-type material, 6
zener diode regulator using voltage doubler, 35, 36 p-type material, 6, 7
Direct coupled amplifier, 168–170
Direct coupled ring-of-three, 238 F
Direct coupling, 167 Fairchild Semiconductor, 544
Discharge transistor, 513 Feedback
Discrete voltage regulator, 402 instability, 245
Dominant pole compensation, 273, 274 linearity and frequency response, 245
DONOR atom, 6 negative, 245
Doping, 5, 6 principle, 245
Drain-source voltage, 91 Feedback amplifiers
Drift current, 5 bandwidth increase, negative feedback, 252, 253
Drift velocity, 5 basic system, 245, 246
Dry soil indicator, 82, 83 characteristics, 247
Duty cycle modulation, 505, 506 classification
Duty cycles, 515 current amplifier, 246, 247
transconductance amplifier, 246
E trans-resistance amplifier, 247
Electromagnetic field detector, 199 voltage amplifier, 245, 246
Electron flows, 53 gain stabilization (see Gain stabilization, negative
Electronic fuse, 78 feedback)
Electronic switch, 544 harmonic distortion, 253
Emergency phone-line light, 33, 34 input resistance
Emitter amplifier, 166, 173 current-series feedback, 255
Emitter follower, 269, 270 current-shunt feedback, 255
Emitter follower configuration, 542 voltage/current sampling, 254
Emitter follower oscillator, 479 voltage-series feedback, 254
Energy levels/shells voltage-shunt feedback, 254, 255
bands overlap, 3 output resistance
conduction electrons, 3 current-series feedback, 257
inter-atomic/inter-molecular bonds, 3 current-shunt feedback, 257
letters and numbers, 2 output current sampling, 256
orbiting electrons, 2 voltage-series feedback, 256
sublevels, 2 voltage-shunt feedback, 256, 257
valence electrons, 3 stability and compensation
584 Index
Triangular wave generator (cont.) Variable transistor voltage regulator, 393, 394
frequency control, 505, 506 Variable wide-band filter, 451
half-cycle period, 502 VBE multiplier, 184, 185
non-inverting node, 504 Video amplifier, 282, 283
output amplitude, 504 Video frequency amplification, 282
output waveforms, 503, 505 Virtual earth, 291
resistor, 502 Vishay semiconductors, 545
sawtooth, 506, 507 Voltage amplifiers, 245, 246
TL082, 503 closed-loop bandwidth, 266, 267
transistors, 501 feedback factor, 266
variation, 503, 504 feedback type, 266
VCOs (see Voltage-controlled oscillators (VCOs)) GBP, 267
Triangular waves, 493 input/output impedance, 268
Triggering methods low-frequency closed-loop gain, 266
phase control switching, 558, 559 maximum feedback, 267
static AC switch, 558 open-loop characteristics, 267, 268
Twin-T notch filter, 443 operational, 265, 266
Twin-T oscillator, 488 requirements, 265
Two BJT amplifier, 164 transfer function, 266
Two phase control triac circuits, 562 Voltage-controlled oscillators (VCOs)
Two-stage amplifier, 163–165, 167 bandpass filter, 510
Two-stage common emitter amplifier, 165, 166 bipolar transistor, 508
Two-transistor configuration, 265 characteristic equation, 510
Two-transistor current source, 182 commercial, 510
controlling signal, 507
U frequency, 510
Un-bypassed emitter resistance, 258 multiplier, 510
Unijunction transistor (UJT) PSPICE output waveforms, 510
capacitor charges, 567 sinewave shaper exploits, 508
characteristics, 565, 566 sinusoidal, 508, 509
charge time, 567 square/triangular waveform, 508
equivalent circuit, 565, 571 variable frequency oscillator, 510, 511
interbase resistance, 565 Voltage-controlled resistor, 194–195
Intrinsic Standoff Ratio, 565 Voltage-controlled Sallen-Key low-pass filter, 544
negative resistance, 565 Voltage-controlled Wien bridge oscillator, 544
operating configuration, 565, 566 Voltage divider biasing, 61–64, 99, 100
output waveforms, 567 Voltage divider biasing technique, 66, 67, 69
physical composition, 565 Voltage divider-biased common drain amplifier, 114
programmable (see Programmable UJT) Voltage doubler, 526, 527
p-type material, 565 Voltage follower, 294, 295
relaxation oscillator, 567 Voltage regulators
reverse bias, 566 definition, 380
symbol, 565, 571 ripple and regulation, 381
Universal filter applications shunt, 381
all-pass notch, 450, 451 Zener diode, 382, 383
research project 1, 451 Voltage sampling, 258
research project 2, 452 Voltage transfer characteristics, 52
second-order speech filters, 450 Voltage-series feedback amplifier, 248, 254, 256, 258–260
variable wide-band filter, 451 Voltage-shunt feedback amplifier, 248, 249, 254–257,
Universal filter system, 451 263–265
Universal motor speed controller, 573
Unsymmetrical waveform, 497, 498 W
Waveform generators, 493
V Wide-band Wien bridge oscillator, 482
Valence electrons, 1, 4 Widlar current mirror, 183
Valence shell, 1 Wien bridge oscillator circuit, 481
Varactor diodes, 14 Wien notch filter, 444
Variable frequency band-pass filter, 451 Wilson current mirror, 183
Variable frequency oscillator, 510, 511 Window comparator, 523
Index 595
Y power rating, 13
Y-parameters, 145 reverse current, 13
symbol, 12, 13
Z temperature coefficient, 13
Zener diode regulator using voltage doubler, 35, 36 voltages, 14
Zener diode regulators, 28 Zener region, 12, 13
breakdown voltage, 28 Zener diode tester, 525, 526
DC voltage output, 28 Zener effect, 11
fixed supply voltage/variable load, 28, 29 Zener power rating, 13
steep diode characteristic, 28 ZENER region, 12
variable supply voltage/fixed load, 29, 30 Zener voltage, 14, 498
variable supply voltage/load, 30, 31 Zero bias, 7, 8
Zener diodes, 12, 497, 499 Zero gate current, 562
doping levels, 13
dynamic resistance, 13