Semiconductors and Transistors - Alexander Schure

ELECTRONIC TECHNOLOGY SERIES
SEMICONDUCTORS
and TRANSISTORS
''
4/•• • 4J' •• .-· :•,·~·> .
.... ,,.. ..,·;,·'
•',,
• · ,,
. -:~:: \:· .. ~:
''
. ,· ~.\:.
•'
.~ / ··1 '•
., '
··:,,_·
:•
a publication
SEMICONDUCTORS
AND
TRANSISTORS
Edited by
Alexander Schure, Ph.D., Ed. D.
JOHN F. RIDER PUBLISHER, INC.,
NEW YORK
Copyright February 1961 by
JOHN F. RIDER PUBLISHER, INC.

All rights reserved. This book or parts thereof may not be reproduced
in any form or in any language without permission of the publisher.
Library of Congress Catalog Card No. 60-150 24
Printed in the United States of America
PREFACE
Semiconductors and the transistors derived from these materials

had a limited use in the 1940's. They are now prominently used
in components manufactured for electronic communication equip-
ment.
The circuits and design considerations pertaining to semiconduc-
tors and transistors are a vital part of electronic theories and
practice. It is necessary that those concerned with electronics pos-
sess a working knowledge of their essential relationships to this
important, comparatively new, aspect.
The intent of this book is to present, discuss, and evaluate the
important ideas relating to an understanding of the theory and
characteristics of these devices. Thus, the book begins with funda-
mental presentations and concepts of semiconductors. It then leads
into the assembly of a series of semiconductors into various types
of transistors, presents basic concepts, and gives an understanding
of transistor characteristics, as well as a development of circuit
operations. The mathematical analyses are simple; but, the treat-
ments are sufficiently extensive to permit the reader to develop
full comprehension of the pertinent theory. To insure this aim,
adequate information is given relating to broad concepts and
designed for ready use. Detailed descriptions of selected major
topics are presented. And, through presentation of practical
situations and problems, the reader is given an opportunity to
apply the principles he has learned.
Specific attention is given to atomic structure, the quantum
theory, conductors, insulators, and semiconductors. Conduction by
holes, semiconductors with impurities, and the semiconductor rec-
tifier are also detailed. The p-n junction under equilibrium con-
ditions, the p-n junction in reverse and forward bias, diode
types (including photodiodes) , and an extension of p-n junction
theory into the transistor receive specific attention. Also discussed
is a comparison of the basic transistor and vacuum tube circuits,
transistor symbols, characteristics, analysis and design, drawing the
iv PREFACE
load line, establishing the Q point for the common base and
common collector amplier, regions of operation, static and dynamic
quantities, circuit theorems, and hybrid parameters. Finally, h
parameters and elements of design relating to voltage, current,
and power gain are analyzed. A foundation is thereby provided
upon which more advanced concepts can be built.
Grateful acknowledgment is made to the staff of the New York
Institute of Technology for its assistance in the preparation of the
manuscript of this book.
February 1961
New York, N. Y.
A.S.
CONTENTS
Chapter Page
1 Introduction to the Semiconductors ....... .

The Atom • The Quantum Theory • Atoms in
Metals • A Conductor • An Insulator • The Semi-
conductor • Conduction by Holes • Semiconductors
with Impurities • Review Questions
2 The P-N Junction Diode 23

The Semiconductor Rectifier • The P-n Junction
Diode Under Equilibrium Conditions • P-n Junction
Diode in Reverse Bias • P-n Junction Diode in
Forward Bias • Zener Breakdown Voltage • Diode
Types • Junction Capacitance • The Tunnel Diode •
Photoresistors and Photodiodes • Review Questions
Introduction to the Transistor ..... 45

Comparison of Transistors to Vacuum Tubes • Com-
parison of Basic Transistor and Vacuum Tube
Circuits • Review Questions
4 The Transistor as a Circuit Element 63

Transistor Characteristics • Common Emitter Char-
acteristics • Drawing the Load Line • Establishing
the Proper Emitter Bias • Establishing the Q Point
for the Common Base Amplifier • Establishing the
Q Point for the Common Collector Amplifier •
Finding Rn• Distortion and Regions of Operation •
Active Region • Saturation Region • Cutoff Region •
Interrelation of Q Point, V00 , and Regions of
Operation • Review Questions
vi CONTENTS
Chapter Page
5 Small Signal Analysis ................. . 103

Static and Dynamic Quantities • Equivalent Circuits
-"Black Boxes"• Circuit Theorems• Hybrid Param-
eters • Graphical Determination of Hybrid Param-
eters • h Parameters for the Common Base and
Common Collector Circuits • Variation of h Param-
eters with Operating Conditions • h Parameter
Equivalent Circuits • Source Resistance • Input Re-
sistance • Voltage Gain • Current Gain • Power
Gain • Isolation Between Input and Output •
Review Questions
Index .......................... . 136

Chapter 1
INTRODUCTION TO THE SEMICONDUCTORS
Up to World War II, a material was considered either a con-

ductor or an insulator of electricity, or it varied in values of either.
As an outgrowth of crystal rectifier research during the 1940s, we
began to understand the processes of a new type of conductive
medium, the semiconductor. A semiconductor has a conductivity
value somewhere between that of an insulator and a metallic con-
ductor. The identifying characteristic of a semiconductor is that
elements called holes (as well as electrons) are responsible for
electronic conduction. (In a normal conductor, electrons are the
sole carriers of electric current.) Three or more pieces of semi-
conductor material fastened together in a special way form a
TRANSISTOR. Thus, the transistor is born after the process of semi-
conduction.
1. The Atom
The atomic theory of matter was one of the most important

contributions to physical science. In 1805, John Dalton advanced
the idea that all matter was made up of smaller particles called
atoms (from the Greek word atomos, indivisible) . Today, his
theory is an accepted fact. An atom is the smallest particle of an
element that still possesses the properties of that element.
What does an atom look like? Where did it come from? Why
does it behave as it does? Through the years, scientists have
explored the mystery surrounding the atom. Today, our atomic
1
2 SEMICONDUCTORS & TRANSISTORS
theory not only explains many of the atom's physical character-

istics, but can predict what will happen when two or more atoms
come into contact with one another.
Every atom consists of a nucleus and one or more electrons.
Electrons spin around the nucleus of an atom in much the same
manner as the planets of our solar system orbit the sun. Figure 1
illustrates the simplest of atomS-the hydrogen atom. Here, the
nucleus is not an entity in itself, but is a term used only to desig-
nate the central part of an atom. Essentially, the nucleus contains
one or more subparticles called protons. Just as the sun is larger
than any of the planets, the proton is larger than the electron
(about 1800 times larger). In the simple hydrogen atom, one elec-
tron orbits one proton. In larger atoms, many electrons orbit about
a nucleus holding many protons.
PROTON
Fig. 1. The hydrogen atom.
ELECTRON
Actually, a comparison of an atom with our solar system is quite

weak, since there are several glaring differences. Whereas our solar
system is millions of miles in diameter, the atom has a diameter
in the order of an Angstrom unit. 1 For instance, if an ordinary
marble were expanded to equal the size of the earth, its constituent
atoms would be about the size of ping pong balls; or, a marble
contains about a billion billion ( I 024 ) atoms.
The motion of our solar system is governed by the attractive
force of gravity. This force exists between all particles of matter.
Gravitational forces also exist between all particles of an atom,
but are eclipsed by a much stronger force-electrostatic force.
Governing this force is the familiar Coulomb's Law: like electri-
cally charged bodies repel one another, and unlike charged bodies
attract one another. The electron is a negatively charged body;
the proton is a positively charged body. The electron and proton
have the same magnitude, but opposite types of charges. Two elec-
trons, or two protons, tend to repel each other. A proton and an
I The Angstrom unit, symbolized by A, denotes a length of 0.0000000001 meter

(or, I x 10-10 meter).
INTRODUCTION TO THE SEMICONDUCTORS 3
Fig. 2. Shells of an atom.
8
UCLEUS
1st SHELL
( 2 ELECTRONS)
MAXIMUM
2nd SHELL
( 8 ELECTRONS)
MAXIMUM 3rd SHELL
( 18 ELECTRONS)
MAXIMUM
electron attract one another. For convenience, we designate both

the electron and proton as having one unit of electronic charge.
An atom having the same number of protons in its nucleus as
it has electrons outside is electrically neutral, with a net electronic
charge of zero. This is the natural case with all atoms, and is
true for the hydrogen atom of Fig. I. If an atom gains or loses
electrons, it is then called an ion. Ions are atoms having an excess
or deficiency of electrons orbiting about the nucleus. For instance,
if an atom loses two electrons, it is designated an ion with a +2
charge (or a valence of +2). If an atom has two extra electrons,
then it becomes a negative ion with a valence of -2.
Do electrons ever hit one another when so many of them spin
around the nucleus at once? NO, because no two electrons have
exactly the same orbit. If we think of the orbit of two electrons
as being different from one another, when their respective orbital
radii are different, then it may be said that no two electrons travel
at exactly the same distance from the nucleus. The orbital radius
may also be expressed in terms of energy. Since an electron far
from the nucleus has a larger potential energy than an electron
close to the nucleus, we may also say that no two electrons ever
have the same energy. Although all of the electrons of an atom
have different energies, or orbits, they tend to crowd together in
energy bands called shells. There is a much greater distance be-
tween the electrons of any two shells than between the electrons
of a given shell (see Fig. 2) .
Figure 3 schematically illustrates some common neutral atoms.
This is a two-dimensional representation, assuming all the elec-
trons travel around the nucleus in a flat plain. The dotted lines
indicate different shells. (Within each shell, the electrons indicated

have minute energy differences and, therefore, slightly different
orbital radii.) Notice that the heavier atoms have more electrons
and more protons. The number of protons in a neutral atom (or
the number of electrons) is called the atomic number. Starting
with Hydrogen, atomic number l, the numbers extend to 92 (ex-
cluding the artificially created elements of the atomic age). If
there were room to draw all 92 elements, one would notice that,
/' '
IC\\
~
...
0}
__,, /a-----.._ ,,,.,-::_-"'
/
I O I //✓,,, --',\ \
HYDROGEN
~ - - / I IQ.._) ~
~ ~---V / / /
HELIUM \
..--
\\.___-------,,,.-,/
---
"- SILICON
GERMANIUM
Fig. 3. Schematic representation of some neutra I atoms.
as the atomic number increases, the number of electrons in any

given shell do not increase continually. The first shell allows only
two electrons. With atomic number 3 (Lithium), the extra elec-
tron begins a new shell, which has a maximum number of eight.
After that, the third shell must be filled with eighteen electrons,
and so on. It must be emphasized that, although each shell may
contain only a certain maximum number of electrons, there are
very many unused permissible orbits within each shell.
It was mentioned previously that the forces governing the action
of an atom are electrostatic in nature. This is only partly true. An
electron stays in an orbit because the electrostatic force of attrac-
tion to the nucleus exactly balances the centrifugal force that tends
to throw the electron out of the atom. This is a very delicate
balance. The velocity of the electron must be just right, because
the centrifugal force varies as the square of the velocity. If the
electron gains or loses any energy, this balance cannot be main-
tained; and the electron will move to a different orbit. The elec-
tron may not, however, jump to an occupied orbit or to a filled
shell. Only discrete energy levels are allowed.
2. The Quantum Theory
The amount of energy an electron can absorb is proportional

to how far it jumps from one orbit to another and, since only
certain jumps are allowed, the electron may absorb only discrete
amounts of energy. The reverse is also true. If an electron _jumps
from an outer orbit to one slightly nearer the nucleus, it will give
up a discrete amount of energy. To reverse this action, i.e., to make
the electron jump back to an outer orbit, would necessitate the
electron absorbing exactly the same amount of energy. These bun-
dles of energy are called quanta of energ;y. Energy may be supplied
to an atom by several means.
Energy is usually given off by an atom in the form of light. All
visible light,' in fact, all electromagnetic radiation (radio waves,
heat waves, X rays, etc.) is the result of electrons in atoms _jump-
ing from higher to lower energy levels. This, in essence, is the
Quantum Theory of Light. This theory plays a vital role in the
action of semiconducting devices.
These bundles, or quanta of energy being radiated,. are often
called photons. (A photon is not energy, pure and simple.) Photons
exist at the boundary line between mass and energy, possessing
some of the features of each. The argument as to exactly what a
photon is has raged for fifty years. However, we do know that if
an electron at a higher orbit corresponding to an energy level W 1
jumps down to a lower orbit corresponding to an energy level W :!•
the emitted photon will have a wavelength (.\.) in Angstrom units
of:
12,400
.\.=~~~A (I)
W1-W2
where W 1 and W 2 are in electron volts 2
This establishes a direct relationship between the wavelength of
the emitted photon and the difference in energy between the two
orbits in question. Einstein's equation E =
MC2 (where E =
energy, M = mass, and C =
speed of light) told us that there is
an equivalence between mass and energy. Consequently, the photon
has a mass in addition to a specific energy. The shorter the wave-
length the higher the energy, and the greater the mass of the
photon.
2 The electron volt is a unit of energy, just as the kilowatt hour is a unit of
energy. It represents the amount of energy gained by an electron accelerated
by a potential difference of one volt.
LOWER ENERGY
LEVEL
PHOTON EMITTED FltJ. 4. Absorption and emla•

sion of photons.
PHOTON ABSORBED
When a photon has precisely the correct amount of energy

(corresponding to a specific wavelength) and hits an orbiting elec-
tron of an atom, it may be absorbed by the electron-and, in the
process, impart all of its energy to that electron. The electron, in
turn, will jump to a higher permissible energy level, where the
difference in energy between the old and the new orbit equals the
energy of the photon, in accordance with Equation (1).
A completely reversible action may take place within an atom.
If an electron falls to a lower permissible energy level, a photon
is emitted. If a photon of the correct energy (i.e., wavelength)
hits an electron, it will knock the electron up to a higher energy
state. It is even possible that this electron might fall back to its
original orbit and re-emit a photon of the same wavelength, as
shown in :Hg. 4. However, if a photon of other than the mrrect
or permissible energy hits an electron of an atom, it will pass right
through-without affecting the atom and without being itself
affected.
Is it possible to impart enough energy to an electron to knock
it out of the atom? YES. If, by photon bombardment or some other
means, an electron is given enough energy, it can jump right out
of the atom. This process is called ionization, because an atom-
minus the electron-is called an ion.
It takes energy to ionize an atom by removing an electron that
is orbiting near the nucleus; whereas it takes relatively little energy
to ionize an atom by removing the loosely-held outer electrons
(called valence electrons). In this instance, the discrete energy
level rule does not apply; i.e., a photon need contain only enough
energy to make the electron leave the atom. If it has more than
this minimum energy, the departing electron will retain the addi-
tional energy in the form of velocity.
Photon bombardment is not the only way to ionize, or excite
an atom. Ordinary heat is sufficient. Or, if another high speed
electron from somewhere happens to hit an orbiting electron, the
orbiting electron may be raised to a higher energy level-0r even
be knocked out of the atom. The only restriction is that the im-
pinging electron (or any other source of energy) must possess just
the right amount of energy to excite an orbiting electron to a
higher energy state, or must have the minimum amount of energy
to ionize the atom.
3. Atoms In Metals
Thus far in our study of the atom, we have concentrated on its

behavior in an isolated state. We will now discuss how a number
of similar and dissimilar atoms behave together. As previously
mentioned, all matter is composed of one or more types of atoms.
A material composed of only one type of atom is called an element.
Iron, lead, hydrogen, etc., are common examples of elements.
When atoms of different elements are combined, the resulting
material is called a compound. For instance, the compound water
is made up of two atoms of the element hydrogen and one atom
of the element oxygen. (The familiar symbol H!!O designates its
atomic ingredients.) Yet the compound water bears little resem-
blance to its ingredients of oxygen and hydrogen. Another example
is common table salt. It is composed of one atom each of the ele-
ments sodium and chlorine. Alone, each of these elements is a
deadly poison; when combined to form salt, they become essential
to life. Although there are only 92 elements, the number of com-
pounds in existence is literally in the millions.
It is often said that all matter has three states: gaseous, liquid,
and solid. When the different atoms of the material are separated
by a large distance, compared to the radii of their outer orbits,
then the attractive forces tending to keep them together (more
will be said later) are rather weak, and the atoms drift apart to
form a gas. When atoms are separated by a distance comparable
to their atomic radii, about 1 A, the various atoms slip and slide
about, and the substance is in the liquid state. However, when
atoms are so close together that the outer orbits of adjacent atoms
overlap, the attractive forces between the atoms become large. The
material is then called a solid, wherein each atom is tightly bound
in a particular position. It is this solid state which interests us.
Fig. 5. The sharing of elec•

trons.
Fig. 6. Geometric represen-

tation of a crystal structure.
CRYSTAL STRUCTURE OF GERMANIUM
A strange thing happens when atoms are so dose together. The

encircling electron of one atom has difficulty distinguishing the
orbit of its own atom from the overlapping orbit of the adjacent
atom. And, the electron spends as much time circling about its
atom as it does the other. This is termed sharing; i.e., the two
atoms "share" an electron from each other's outer orbit. (See
Fig. 5.) Sharing of electrons results in the greater strength of the
solid structure.
The force of attraction between two atoms sharing one of their
outer, or valence, electrons is called a rovalent force. A covalent,
or electron pair bond, exists between the two atoms. This type of
bond is so important, so nearly always present in substances, that
Professor G. N. Lewis of the University of California (1875-1946),
wh<\ first explained its electronic structure, called it THE chemical

bond. The process of sharing of electrons is not the same as that
of ionization. The valence electrons are neither lost nor gained
but shared.
Crystals are the formation of numbers of atoms joined in covalent
bonds. A crystal is a body that is composed of an often-repeated
geometric figure. The crystal structure of germanium is illustrated
in Fig. 6. This is a simple cubic crystal structure. There are many,
more complicated structures. All crystals are merely repetitions of
a basic structure, called the crystal lattice. Figure 6 is a simple
cubic lattice. Externally, a crystal takes on the appearance of its
lattice. Iron Pyrite (fool's gold) has a cubic lattice and, therefore,
the entire crystal is cubic. In Fig. 6, each circle symbolizes an
atom, while the bars joining them indicate the covalent forces,
or bonds of attraction.
Figure 7 illustrates a two-dimensional view of a crystal composed
of what is called the diamond lattice structure. Imagine that you
are close to the crystal and can see the outer valence electrons of
each atom. (The electrons belonging to the inner shells of each
atom are tightly bound to their parent nuclei and are of no inter-
est to us.) We can now examine many interesting properties of
metals.
Each large circle in Fig. 7 represents the nucleus and bound
electrons (i.e., all but the outer valence electrons of each atom).
In this particular crystal, each atom has four valence electrons.
Each valence electron has joined in a covalent bond with its neigh-
boring atom. The dotted lines indicate this covalent bond and
show that the valence electrons are "shared" by each neighboring
atom. Note the geometrical symmetry that exists throughout the
entire structure. Every valence electron of every atom is accounted
for and is tied with an adjacent atom in a covalent bond. What
would happen if each of the atoms of our hypothetical metal had
five, instead of four, valence electrons? The extra electron associ-
ated with each atom obviously could not fit in the existing system
of covalent bonds; there would be no room in the scheme. Exactly
what does happen depends upon the temperature.
As was said, heat supplies energy to atoms. Absolute zero
(-460°F) is the lowest possible temperature. At this point, there
is no heat whatsoever. According to classical theory, the electrons
of any atom possess no energy at this temperature. The Quantum
Theory, disproved this belief. Electrons at absolute zero do pro-
duce energy. These extra valence electrons for each atom are
loosely bound, because they are not tied in covalent bonds. Con-
sequently, even at absolute zero, some of these extra electrons may
have just enough energy to leave their parent atoms and wander
Fig. 7. A two-dimensional
view of II crystal.
aimlessly throughout the metal, and not be captured by any other

nuclei. These wandering electrons are called free electrons and the
atoms that are left are called ions, because they lack a valence
electron. This happens at absolute zero. However, it is not a very
practical picture, because a temperature at absolute zero has never
been created. Remember-at any temperature, a proportion of the
electrons will have all ranges of energies. And, at absolute zero,
although most of them will have very low energies, there will always
be some electrons with enough energy to escape.
At higher temperatures, a greater proportion of these extra elec-
trons will have enough energy to leave their parent atom and
wander. The effect is not unlike making popcorn. When heat is
applied, all of the corn is about the same temperature; but only
some of the corn pops. When more heat is applied, a greater pro-
portion of the corn pops. It is futile, however, to talk about just
when a given kernel will pop. For the same reason, we cannot talk
about the energy of an individual electron, but only of the energy
that most of the electrons probably have. So we say that if the
temperature is high enough, nearly all of these extra electrons
will probably have enough energy to become free electrons.a
3 It is worthwhile to note that if the metal gets hot enough (about 2000°C) the
extra electrons will have so much energy that some of them will not only
become free electrons but will leave the surface of the metal. This is how
thermionic emission from a hot cathode is accomplished in a vacuum tube.
T FINISH
PATH OF ELECTRON'S
r,;,., RANDOM MOTION IN A
\!::,/ CONDUCTOR
I
Fig. 8. Motion of an elec- I
tron: (A) With no applied
field; (B) With an applied
field.
@ i
START
+
NET DRIFT IN
DIRECTION TO RIGH
,___ _ _ _ _--I SOURCE OF

VOLTAGE _+_ _ _ _ __,
It is these extra electrons that are present in all metallic con-

ductors. The-y are called free electrons because they are free to
drift in any direction. Of course, in the absence of an externally
applied force, each free electron will travel just so far before it
bumps into another atom or ion. Since the direction of travel is
changed during each collision, the net travel in any given direction
is zero, as shown in Fig. 8 (A). Each point where the direction of
travel is abruptly changed represents a collision. The distance the
electron travels between collisions is called the mean free path.
4. A Conductor
If our piece of metal were connected across the terminals of a

battery [Fig. 8 (B) ], an electric field would then be set up across
this intervening space. Under these conditions, the free electrons
would slowly drift toward the positive terminal, as indicated in
Fig. 9. These moving electrons constitute an electric current. In
short, this is the process of electronic conduction in a metal. If
the battery voltage is increased, the electric field is increased, and
the electrons are attracted to the positive terminal with a greater
force. The current will then increase. Of course, the free electrons
can only reach the positive terminal by an indirect path, since they
are continually colliding with other atoms.
Note that superimposed upon the random movement is a steady
drift in the direction indicated. Because numerous collisions im-
pe<le the flow of free electrons, it is said that the metal has resist-
ance. The ability of free electrons to move through this maze of
impeding atoms is called the mobility of the electrons. There is a
direct relationship between the mobility, the net drift velocity in
a given direction, and the applied electric field. Thus if V4 is the
net drift velocity of the free electron in meters per second, f is the
,_ ..
I I I I I I
I I I : I I
..,.) -- •io,.
_, - .... _____
--.ie'.),.
__ - _____ ••✓
••1 -- eiO,•
- ----
...._
,..,___ '•)
I I II I I
I I I I I I F19. 9. Motion of free elec-
I I I I I I_.
trons towards positive ter-
I I I I I I
~ ·~ ~ + minals.
==~,o•~~==~o•~----_-_!,•~~-=
'•1 - ti).........
1 I I I I,.,I
I I
I I
:
I I
I I
I 1
I
-- i1Q1 i- - - -.,o.---- -.,o,e--
__ ,.,
•!1
_,
1 I
•!1
I I
,.,
•!1 ----
;:..:;..- - ,., ~ - _, ..._ --
I I
I I I I I I
' I l I I I
applied electric field in volts per meter, and µ. is the mobility, we

have the relationship:
(2)
The minus sign appears because the electrons always travel in the
opposite direction of the applied field.
Is atomic theory justified by Ohm's Law? We know that a metal
has resistance and that current through the metal is proportional
to the voltage across it. How does the atomic theory justify the fact
that different metals have different resistances? Remember, our
hypothetical metal had but one extra electron per atom. Another
material could have been used that had more than one free elec-
tron to contribute to the flow of current. One, two, and sometimes,
three, electrons (depending on the metal) per atom fall into this
"extra" classification (i.e., they are not tied in covalent bonds with
other atoms) and hence can become free electrons. With all these
free electrons, the material is a better conductor. Therefore, various
metals have different resistances, if they have different numbers of
unbonded valence electrons.
What would happen if the metal is now heated? Empirical
observations prove that the resistances of all conductors increase
with temperature. At first glance atomic theory might suggest that
the resistance of a metal should decrease as the temperature is

raised, because the added heat would probably liberate more of
these "extra" electrons. This is true; but this effect is completely
overshadowed by another. When heat is applied, the relatively
massive and stable nuclei begin to vibrate back and forth in a
random manner (although they never drift out of their assigned
positions in the crystal structure). When this vibration occurs, the
free electrons moving toward the positive electrode collide more
easily with the nuclei than would if the nuclei were stable. Of
course, free electrons collide with nuclei (or ions) almost all of
the time-with or without heat. 4 But when the temperature is
raised and the nuclei vibrate, the incidence of collision becomes so
much greater that the net velocity drift towards one electrode is
lessened. The conductive current decreases, equivalent to an in-
crease in resistance. This positive temperature coefficient (a rise
of resistance with a rise in temperature) is an identifying charac-
teristic of all conductors. In more technical terms when the metal
is heated, the mobility of the free electrons decreases at a much
faster rate than the increase of thermally-generated free electrons.
5. An Insulator
We now know that a material's resistance is governed, to a large

extent, by the number of free electrons that are available for con-
duction at a given temperature in a given volume of the material.
If a material has all of its valence electrons tide in covalent bonds,
as was shown in Fig. 7, there will be no free electrons for conduc-
tion and it will be a perfect insulator. There is, of course, no such
thing as a perfect insulator. Some impurities in a material are
always present that are not tied in interatomic bonds and, there-
fore, become a source of a limited number of free electrons. A
good practical insulator has about I 07 free electrons per cubic
meter; a good conductor has 1028 free electrons per cubic meter.
Thus, a good insulator has only one thousand billion billionth
(1/1021) the number of free electrons that a good conductor has.
In order to present a simplified explanation of some processes
of solid state physics, a highly stylized situation was depicted.
Actually, metals are rarely a single crystal, but many crystal lat-
tices bound in an arbitrary or fixed way. Nevertheless, the remarks
previously made regarding electronic conduction are true. Yet, the
neat crystal of Fig. 7 is not truly representative of a conductor.
4 It is well known that a conductor becomes hot when sufficient current passes
through it. This is a direct result of its many electron-ion collisions,
SEMICONDUCTORS & TRANSISTORS
Usually there are so many free electrons present that no heed is

given to individual ions. But, in the modern concept, a metallic
conductor is visualized only as a periodic three-dimensional array
of tightly-bound ions, with a "sea" or swarm of free electrons
moving at random throughout the structure.
6. The Semiconductor
A semiconductor falls into a special classification all its own. It

is neither an insulator nor a conductor, but can be made to appear
as either.
Let us again visualize a metallic crystalline solid (see Fig. 7) .
We know that when heat is applied to the crystal, the atoms are
energized, and vibrate back and forth. This is called thermal agi-
tation. If the atoms of the material have the special feature of
forming very loose covalent bonds with one another, then the
energy acquired by the atoms through heat might be sufficient
to break some of these bonds and create free electrons for
conduction. Many materials have this property; they are called
semiconductors. Some of the commercially important semiconduc-
tors are germanium, silicon, and copper oxide. (Unless otherwise
specified, germanium will be used to illustrate all aspects of semi-
conductor theory.) Semiconductors may be thought of as existing
in two broad categories: those with, and those without, impurities.
Semiconductors having no impurities are called intrinsic semicon·
ductors.
Figure 7 is an intrinsic semiconductor, because the entire struc-
ture is a homogeneous mass of interlocking atoms; there are no
impurities present. If the temperature of this crystal is low enough,
the material will not conduct current (i.e., it will be a perfect
insulator) . It will not conduct because the valence electrons of all
the atoms are bonded to adjacent atoms. The atomic picture is
exactly the same as that of a physical insulator. The difference,
however, between a physical insulator and an intrinsic semicon-
ductor is that at moderate temperature (room temperatures) ,
some of the weakly-bonded semiconductor electrons break away and
become free electrons. With an insulator, the covalent bonds are
much stronger, and a moderate amount of heat would not supply
enough energy to break the covalent bonds. This is probably the
only atomic distinction between the two ma.terials.
It is even possible to heat an insulator to such a high temperature
that some of the covalent bonds break, resulting in an accompany-
ing liberation of free electrons. Under these conditions, an insulator
will become an intrinsic semiconductor. This is why some insula-
tors cannot be used at high temperatures.
Heat need not be the only source of energy used to break

covalent bonds in a semiconductor. If a photon of sufficient energy
strikes a valence electron, it will break the bond. What will happen?
Let us assume a photon has just broken one covalent bond and
liberated one electron. A vacant place is now left in the location
from which the electron was just ejected. Perhaps the most impor-
tant single concept in semiconductor electronics is that this vacancy,
produced by the ejected electron, acts as if it were itself a positive
I I I I I I
I I I I I I
--
,.,
I I
- ~!' 0- ·~---!'
,.,
I I I
~ - - - 6 -- - - o ,.-- - -
\~)
I
___ ',! ___!J 1!._ __ _

,., ,., t'•)
I I I I
I I
I I
I
I
I
I I I
I I
Fig. 10. Holes and electrons 1
I I : I I
•~ ,~ l~
~~'()•~=-=- -=~)0•!.-=--==
in an intrinsic semiconductor.
1
=-=--=- ~1() i"__~
J
,.,I HOLE---1
~''1 \ (•11
I I I
I I I I FREE I I
I I I I ELECTRON 1 I
-----
- --~' o-----
,.,
------
,.,
•~____ !iO•~
,.,
____ !l o-----
,.,
•J 1
(i)
'-!._ __ _
I I I I I 1
I I I I I I
I I I I I I
electron. This vacancy is called a hole and it (along with the free
electron simultaneously created) is free to move about in the
crystal structure.
When an electric field is set up across a section of the crystal
(see Fig. 10) , the electron will drift to the positive side of the
field, and the hole to the negative side. Conduction by electrons
is called electron oonduction. Conduction by holes is called deficit
conduction. That the two types of conduction may simultaneously
--and even separately-talle place within a semiconductor is the
essence of transistor them 1•.
7. Condudion by Holes
Holes may serve as conductors of electric current. If (as in Fig.

10) a covalent bond of an intrinsic semiconductor has just been
broken and a hole created, it is relatively easy for a bonded elec-
tron in one of the surrounding atoms to break away and fall into
the vacant place represented by the hole. Remember, a hole is
equivalent to a positively charged body and, therefore, exerts an
attractive force on other negative particles in the immediate vicin-
lI Ii II
leJ leJ lel
----•10-
;,,
,.-----.,o-,.------
____ .... _____ _, ------- 0-,.----
1 -----
(i) ti\ 1i,
1 : I I I I
!!
l!l
il!)! !!
l.!J Fig. 11. Motion of electrons
_____!,IO l!_____!l O 1!______!1 O•!___ _
.J"6-e---lo1 e 1i,
N~~ I : I I I
HOLE I I
I I
I I
I I
I I
I I
I I I I I I
l!, l!I l!I
-----.10 ,.-
-----
,., 1i1
______ ..,,,1O<•----
----•...,,,,,10 (.------•
.._ ____
,.,
-----
II :I II II II lI
I I I I I I
ELECTRON MOVING TO RIGHT TO
FILL HOLE CREATES NEW HOLE.
THIS IS EQUIVALENT TO HOLE
MOVING TO LEFT.
ity. In Fig. 11, it can be seen that an electron from an adjacent

atom has broken its bond and filled the original hole. In the
process, however, a new hole is created in the location from which
the electron came.
In this manner, a hole can jump from atom to atom. A hole
traveling in one direction is created by an electron traveling in the
opposite direction. It is said that a hole current in one direction
is equivalent to an electron current in the opposite direction. This
is a very important concept. Holes do not exist; they only represent
the absence of electrons.
In Fig. 10, an electron and hole (called an electron-hole pair)
have simultaneously been created by the action of photon bom-
bardment. If given sufficient time, we will see the drifting electron
undoubtedly fall back into the hole. The free electron and hole
have effectively disappeared and ceased to exist. The time it takes
for this to happen is called the average lifetime of the electron-hole
pair. The average lifetime of germanium and silicon is approxi-
mately 100 microseconds. After this time, annihilation of the elec-
tron-hole pair takes place due to this recombination.
As previously stated, under the influence of an electric field, the
holes will drift to the negative side of the field, the electrons to
the positive side. When a hole approaches the negative plate (see
Fig. 12) , a free electron jumps out of this plate and fills the hole.
Remember that the plate can be considered a wire from a battery
connected to one end of the semiconductor material. Since it is a
conductor, this plate has a ready supply of free electrons. Thus,
I
I I
l• J
•1,...,,---
oc;- -
1 I
: I
,.J -- -
Fig. 12. Mpvement of holes
---... '-.;;_,,,
OIQ(•
( .......
1
,-
I
--
+
I I
le)
•(•'1
0(•--__ _
I I
FREE ELECTRON FROM METALLIC ELECTRODE
FILLS IN HOLE
for every hole attracted to the negative plate, an electron enters

the semiconductor. Holes can never travel in the conducting plate
or wire. Although positive holes and negative electrons move in
opposite directions within the semiconductor crystal, the electric
current flows in one direction-the direction of electron flow.
Let us summarize some salient information about intrinsic, or
pure, semiconductors.
In an intrinsic semiconductor, a hole can only be

created when a free electron is ejected from a covalent
bond. Thus, the concentration of holes always equals
the concentration of free electrons. At room tempera-
tures, thermal agitation always creates new electron-
hole pairs while, at the same time, other electron-hole
pairs disappear because of recombination. The genera-
tion and recombination of electron-hole pairs results
in an equilibrium between the two processes, and a
definite electron-hole concentration will exist for a
given temperature.
Holes have a lower mobility than free electrons.
Holes must travel across a semiconductor in jumps
from one atom to another, but an electron may travel
a more direct path, even considering the many elec-
tron-ion collisions.
When a conductive metal is heated, the mobility of
the free electrons decreases and the production of free
electrons increases. The decrease in mobility, however,
overshadows the latter effect, resulting in an overall
increase in resistance, as the temperature is raised. As
the temperature of a semiconductor is raised, the mo-

bility decreases as well, and the rate of production of
electron-hole pairs increases. With a semiconductor,
however, the increased production of electron-hole
pairs is much greater than the decrease in mobility.
Consequently, an intrinsic semiconductor's resistance
decreases with an increase in temperature. In other
words, a conductor has a positive temperature coeffi-
cient and a semiconductor has a negative temperature
coefficient. For reasons described previously, an insu-
lator, like a semiconductor, has a negative temperature
coefficient.
8. Semlcondudors with Impurities
Intrinsic semiconductors are not very useful, because electrons

and holes are always produced in equal numbers. Practical use of
semiconductor devices is possible only because of the addition of
minute quantities of impurities to the crystal structure. By virtue
of these impurities, we are able to make either electrons or holes
the predominant carrier of electric current.
A basic requirement in adding these impurities is that the
atoms of the impurity material must be almost like the intrinsic
semiconductor atoms. When this condition is met, it is possible
to fit some of the impurity atoms in the semiconductor crystal
lattice replacing some of the intrinsic semiconductor atoms in
the process. This is easily done by mixing a minute quantity of
the impurity with a batch of molten semiconductor material.
When the crystal starts to form, as the mass is cooled, the impurity
enters into a covalent bond with the surrounding semiconductor
atoms.
What happens with the addition of one type of impurity is
shown in Fig. 13. Here, an impurity atom (having 5 valence
electrons) is surrounded by 5 germanium atoms, each having 4
valence electrons. Note that one of the impurity's valence electrons
is extra; that is, it is not bonded to any other atom. For this one
extra electron, we may say that it most probably will (even at
room temperatures) possess enough energy to break its bond with
the nucleus and become a free electron. However, note that in
this case, unlike the intrinsic semiconductor, a hole was not
created with the liberation of the electron. In addition, it must
be remembered that the overall crystal structure has not been
altered by the liberation of this extra electron. The ion that is
left after the liberation, "looks" to the crystal almost like another
germanium atom, insofar as its ability to help hold the crystal

together is concerned.
Since, in this instance, a free electron has been donated to
the crystal, it is said that the impurity atom is a donor atom.
Typical donor elements that have five valence electrons are
phosphorus, arsenic, and antimony. When an electric field is
set up across a crystal, most of the current will be carried by
I I I I I I
I I I I I I
===~ 0,~-=-=-_-_!10
I I I I I I
'!}
1
'!J '!'
I!~~= ~>O I!_-_~
Fig. 13. Addition of im• Ii) ,e\ FREE 1i1
purity with atoms of 5 I I I I ELECTRONS' I
valence electrons to an in- I I IFROM DONOR I
trinsic semiconductor with I I I I /ATOMS I I
I I I I I I
atoms of 4 valence electrons.
--- '!' '!'• '!' ,--
____--!' 0- '.!--- ---!~_
____ 0 ,~------
____!,Q'~--
,.,
I I DONOR
/·(•'
I I
'•'
I l
:
,.,
I
I
I
I
..
ATOM I
I
I
I
I
,
I
,.,
=------~•~=---_-_!lQ•~=------!lQ•!=-=
I
I
I
I
I
I
GERMANIUM 1•) (•I [•1

~M I I I I I I
I I I I I I
negative free electrons. So, the crystal is called an n-type semi-

conductor. Of course, not all the current is carried this way.
There are still thermally-generated electron-hole pairs from the
millions of other germanium atoms. The free electrons contributed
by the donor atoms, however, far exceed the number of holes.
Hence the electrons in an n-type semiconductor are called the
majority carriers (of electric current), while the holes are called
the minority-carriers. Electrons and holes are called mobile charges,
as opposed to the immobile ions.
If an impurity material were added to an intrinsic germanium
crystal whose atoms had one less valence electron than the ger-
manium atom, namely three, a situation would result as illustrated
in Fig. 14. The fact that this impurity has a missing place-a
hole-where an electron should be is analogous to the situation
that resulted when an electron-hole pair was created in an intrinsic
semiconductor. An electron from a neighboring germanium atom
may easily break its covalent bond and jump into this hole. We
say the hole accepts an electron; thus this impurity atom is called
an acceptor atom. Note that the material will now have a pre-
dominance of positive hole current carriers. So, the crystal is
Fig. 14. Addition of im-

purity with atoms of 3
valence electrons to an in-
trinsic semiconductor.
called a p-type semiconductor. Again, because of therm.al agitation,

some electron-hole pairs will be present. Of course, the holes will
far outnumber the free electrons, and so the holes are called
majority carriers, and the electrons, minority carriers. Typical
acceptor impurities are gallium, indium, and boron.
The process of adding impurity atoms to a semiconductor is
called doping. It is a very critical process, calling for the utmost
control over the amount of impurity added. It is not hard to
visualize what would happen if an overabundance of impurities
were added to a semiconductor. For example, if intrinsic germanium
were to be overdoped with n-type material, so many free electrons
from the donor atoms would permeate the crystal, that the device
would undoubtedly act as a conductor.
It is interesting to note that, when an intrinsic semiconductor
is doped with donor atoms, not only do the number of free
electrons increase (at a given temperature), but the number of
holes, which normally would be present because of thermal agi-
tation, decreases. This happens because, with so many free elec-
trons available, the recombination rate is higher. In other words
as soon as a hole is created from the thermal agitation of a
germanium atom, there is always nearby a free electron to fill in
the hole. This recombination results in the death of the hole
and the free electron, and the re-establishment of a covalent
bond. But, since there are so many more free electrons than holes,
the decrease in the number of electrons is hardly noticeable. Of
course, the opposite reasoning applies when intrinsic germanium
is doped with p-type impurities: the number of holes increases, and
the number of electrons decreases.
Even a minute quantity of impurity added to an intrinsic semi-

conductor exerts an influence over its conductivity. In an intrinsic
germanium crystal there are 4.52 x 1022 atoms per cubic centimeter.
If to this are added 2.2 x 1014 impurity atoms per cubic centimeter,
the chemical difference caused by the addition of the impurity
could hardly be detected, since only one impurity atom is added
for every 200 million germanium atoms. However, if each of these
impurity atoms contributes a free electron, the conductivity of
the crystal would increase by a factor of, approximately, seven.
At room temperatures, it can be assumed that nearly all of
the impurity atoms of a doped semiconductor will ionize and,
therefore, liberate holes or electrons for conduction. At these
temperatures, the number of holes or electrons created because
of the addition of the impurity, far exceed the number of thermally-
generated electron-hole pairs. If, however, the temperature is
raised high enough, and the number of electron-hole pairs created
far exceeds the number of impurity holes or electrons, the total
number of electrons will approximately equal the total number
of holes. For example, if the temperature of the doped germanium
crystal referred to above is raised enough, every germanium atom
would be ionized. In fact, every covalent bond would probably
be broken, liberating four electron-hole pairs per atom, or about
20 x 1022 electrons and holes per cubic centimeter. Whereas, at
room temperature, the electrons contributed by the donor atoms
fa.I' exceeded the number of holes thermally generated, now essen-
tially the opposite is happening. In fact, there are so many more
electron-hole pairs thermally that the additional electrons con-
tributed by the donor atoms hardly affect the ratio of the number
of holes to the number of electrons. In the example cited, the
new ratio of holes to electrons is approximately 1/1.000001. Con-
sequently, at high enough temperatures, the number of holes will
roughly equal the number of electrons-regardless of the type or
amount of doping. This equality of holes and free electrons was
previously attributed to an intrinsic semiconductor. Consequently,
a doped semiconductor is said to have "gone intrinsic" at high
temperatures. This effect places a definite upper temperature limit
on a semiconductor's operation. Practically, germanium semi-
conductor devices are usable up to about 85°C, whereas silicon
devices will operate up to, and sometimes over, 200°C. The
reason for this difference (in the upper operating temperature
between the two materials) is that it takes about 1.25 electron
volts of energy to break a silicon covalent bond, but only 0.75
electron volt to break a germanium covalent bond. Therefore, at
high temperatures-where a large amount of thermal energy
is supplied to the crystal-more germanium covalent bonds are

likely to be broken than silicon covalent bonds.
In addition to a high temperature limitation, n-and p-type
materials have a low operational temperature limit. When the
temperature is so low that the thermal energy supplied to the
impurity atoms is insufficient for their ionization, the material
effectively becomes an intrinsic semiconductor. This effect is so
important that for low temperature operation, heaters are often
supplied to semiconductor devices.
9. Review Questions
I. If oxygen has an atomic number 16, how many protons does

it have in its nucleus? How many electrons are there in the
neutral oxygen atom?
2. What is a photon? How is our daily life dependent upon it?
3. If a photon has 12 electron volts (ev) of energy, and it strikes
an electron in an atom whose ionization potential (energy
needed for ionization) is only IO electron volts, does the
photon ionize the atom? If so, where do the extra 2 electron
volts go?
4. Do electrons have any energy at absolute zero? How do the
classical and modern theories differ on this point?
5. Why do conductors have positive temperature coefficients, and
semiconductors negative temperature coefficients?
6. Although holes and electrons move in opposite directions
under the influence of an electric field, why does this result
in a current in one direction?
7. What are the conditions under which a doped semiconductor
appears to be an intrinsic semiconductor? Does this effect
vary among different semiconductor materials?
8. What happens to a hole if it collides with a free electron?
9. Is it possible to know the exact energy of any given electron
at a particular time?
10. What would be the outcome if an intrinsic semiconductor
crystal were doped simultaneously with donor and acceptor
atoms?
Chapter 2
THE P-N JUNCTION DIODE
10. The Semiconductor Rectifier
Alone, p- or n-type semiconductor crystals conduct current in

either direction. However, when a p- and n-type crystal are placed
together, the combination exhibits the important property of
rectification; that is, the property of conducting current in one
direction only. Because the property of rectification is dependent
upon the junction, or interface between the p- and n-crystals, semi-
conductor rectifiers are often called p-n junction diodes.
11. The P-n Junction Diode Under Equilibrium Conditions
Let us place a piece of hypothetical n- and p-type germanium

material together. The doped semiconductor materials are sche-
matically illustrated before contact by Fig. 15 (A) . (For the present,
we will not consider thermally generated electron-hole pairs from
germanium atoms. Hence, only holes, electrons, and impurity ions
are shown.) Before ionization, each neutral donor atom in the
n material still possessed its extra electron. Each atom may also
be represented as a donor ion with a + I charge, its extra electron
possessing a - I charge. The net charge on each atom is still zero,
even though we show the ion and electron separately.
Similarly, in the p material, each acceptor atom can be shown
as an ion with a - I charge, its hole with a + I charge. This
23
methcxl of illustration is advantageous because now a neutral and

an ionized atom may be treated with the same symbols. In Fig. 15A,
some of the donor atoms are ionized (due to thermal energy)
, j.f 'f
'f • '' '© 'f f©
P-TYPE GERMANIUM N-TYPE GERMANIUM
§> ~ §> 9-... 9-. _c:;_9 f f @•

(Bl (§1 ~ '? q_ 9.... _c;_c:; @• f 'I
Fig. 15. A. P-type germa-
~ (§1 ~ 0o....., 0o-.. ...,.9 -~ EB• EB
• 'fl nium and :,-type germa-
p N nium; B. The p-n junction
Vo diode. C. Charge density
I
I distribution. D. Electrostatic
DIFFUSION BIAS: potential variations.
I POTENTIAL I
I I
I I
(Cl
DISTANCE I DISTANCE FROM

FROM JUNCTION I JUNCTION IN
IN P SIDE I N SIDE
I
I
I
I
I
I I
MINORITY CARRIER HOLE MAY EASILY
ROLL DOWN POTEN TIAL HILL 7
(D) 1
I I J.tQ__,,____....
__
MAJORITY CARRIEJ HOLE

MUST HAVE ADDITIONAL
t
ENERGY TO ROLL UP
POTENTIAL HILL I
"'--m
and the electrons that have been released are freely moving about.
In the p-type material, the reverse is happening: some holes have
been liberated and are roaming around the crystal.
THE P-N JUNCTION DIODE 25
Figure 15A is a highly stylized illustration and does not portray

the exact atomic picture of either crystal. However, this analysis
allows a meaningful explanation of the p-n junction action. For
instance, almost all the impurity atoms are ionized, even at room
temperatures. Yet in the illustration, only a few electrons and
holes are shown separated from their respective parent atoms
(ions) . This is a seemingly incorrect representation. However,
although almost all the impurity atoms are ionized, both crystals
are homogeneous (or even) in charge density. That is, in the
n-type crystal, the number of free electrons in any given area of
the crystal is about the same. Assuming the donor ions are evenly
distributed throughout the crystal (they are), then in any given
region of the crystal the net electrical charge is zero; i.e., the
total negative charge of the free electrons in this region effectively
cancels the total positive charge of the donor ions in the same
region. Therefore, the crystal has a neutral charge throughout its
structure, and we illustrated it as essentially containing neutralized
donor ions, or just plain neutral atoms. Physically, this assump-
tion may be questioned; electrically, it is true.
Because the thermally agitated free electrons wander about,
there will always be some variation in charge density. At some
points in the crystal there will be momentary concentrations of
free electrons. To account for this slight concentration of negative
charge, some of the electrons in Fig. 15A are shown separated
from their parent donor atoms.
In the p-type crystal (Fig. 15A), we have also simplified an
involved process. Here, however, holes were shown to move in
the space between donor atoms. This, too, is justified. Holes may
skip from germanium atom to germanium atom, and these atoms
are not shown. A hole does not roam around the crystal exactly
as a free electron does, because a hole is not a physical entity.
A hole is only the absence of an electron in a covalent bond.
But, when we view large sections of the crystal at once, there are
so many holes, and the atoms are so close together, that little
error is introduced when we assume that holes can move freely.
It may be said (using the same logic we applied to the n-type
crystal) that the charge density everywhere is approximately zero.
Or, as with free electrons in the n-material, the concentration of
positive holes in any given region is about equal to the concen-
tration of negative acceptor ions. The small variations in this
net zero charge density are indicated by the presence of a few
holes, separated from their parent atoms.
What would happen if the n- and p-type crystals were placed
together, without an external electrical connection? On the right
side of the junction between the two materials (see Fig. 15B)
there is a concentration of free electrons; on the left side, a
concentration of holes. If the device has been fabricated correctly,
the junction between the two materials does not represent a dis-
continuity of the atomic crystal structure and the two materials
together represent one homogeneous figure. Consequently, since
there is a high concentration of free electrons on one side of this
junction, and an extremely small concentration of free electrons
on the other side of the junction, the free electrons will tend to
drift across the junction. This type of thermal drifting is called
diffusion and is not ordinarily encountered in conductors.
Electrons tend to diffuse, or drift from points of high con-
centration. to points of lower concentration, in the same way as
the odor from an uncapped bottle of ammonia diffuses throughout
a room free of moving air current. Electron diffusion acros!! the
junction occurs because this is the only direction that the thermally
agitated electrons may take without almost immediately colliding
with other free electrons. Much as water seeks its own level,
these electrons (if left to themselves) will spill across the junction
and create an even concentration throughout the crystal. The
holes, too, will diffuse across the junction from left to right.
When an electron diffuses across the junction, it leaves its
positive ion and deposits its own negative charge to the p material.
Similarly, a hole diffusing across the junction adds its positive
charge to the n material and leaves a negative ion. The electrons
and holes that scatter across the junction do not diffuse very deeply
into their respective materials. An electron crossing the junction
almost immediately combines with a hole that (up to that point)
was neutralizing a negative acceptor ion, assuming the hole had
not already diffused across the junction. If it had already diffused,
the electron would probably combine with some other hole in
the immediate vicinity. A similar argument applies to hole
diffusing to the n material. A hole may meet an electron while
crossing the junction. The collision results in the inevitable
annihilation of both carriers.
Before this electron-hole recombination, the charge density was
zero in each side of the crystal. Now the ions near the junction in
both materials have been stripped of their neutralizing holes or
electrons and, therefore, possess a charge. These ions are called
uncovered ions. The charge density distribution across the diode
(see Fig. 15C) is no longer zero. Note that in the immediate vicinity
of the rows of uncovered ions, the charge density is at a maximum.
It is negative in the p material and positive in the n material.
There is a smooth transition from negative to positive charge
density across the junction region. Of course, the rest of the

crystal has a neutral, or zero charge density.
As soon as ions are uncovered in the vicinity of the junction,
an electric field is created between these oppositely charged ions.
An electric field always exists between any two points of charge
density variation. Visualize a fictitious generator across the junc-
tion region, with a potential V 0 representing the net potential
drop across the field. Since the electric field is created because of
the diffusion of holes and electrons across the junction, the voltage
existing across the field is called the diffusion bias potential. On
most diodes, the diffusion bias potential is about 0.2 to 0.3 volt.
The electric field, although created by electron-hole diffusion, has
a repelling effect on the further diffusion of these electrons or holes
across the junction. Obviously, the positive ions in the n material
will repel any positive holes attempting to cross the junction.
Also, the negative ions in the p material constitute an effective
barrier against further migration of free electrons across the
junction. Therefore, the space between the donor and acceptor
uncovered ions is called the barrier, or space charge region, and its
width is denoted by the symbol W. This space charge region is
about I0- 4 cm wide.
Figure 15D illustrates the electrostatic potential variations across
the device. Any hole attempting to cross the barrier region must
travel up an electrostatic hill of a height equal to the diffusion
bias, V0 • In simpler terms, the positive hole must travel toward
the uncovered positive donor ion IN SPITE of the repellent force
existing between these two positively charged bodies. The "hill"
terminology is imaginary and helps get across the idea of a hole
climbing a hill against this symbolic gravity, which is caused by
the repellent force. It cannot do this unless an additional force
is applied.
Similarly, an electron must travel up a potential hill of the
same magnitude when crossing to the p side. Since the electron is
negative, Fig. 15D more correctly illustrates the picture of a
"hill," when it is visualized upside down.
Thermally generated electron-hole pairs from the germanium
atoms in the diode will not produce a net current flow across the
junction. In the n material, electrons that are thermally generated
from the breaking of germanium covalent bonds act just as other
free electrons in this material. Most of the holes that are thermally
generated will immediately disappear by recombination with the
free electrons that are present. The few holes that escape annihila-
tion diffuse over to the junction and fall down the potential hill,
as shown in Fig. 15D. This hole current is precisely balanced by
another hole current going in the opposite direction, which results

when a few holes (majority carriers in the p side) have enough
thennal energy to climb the potential hill. The height of the
potential barrier automatically adjusts itself to insure that this
happens. Consequently, since both hole currents have the same
magnitude and travel in the opposite direction, the net hole
current is zero.
In the p side, those thermally generated electron minority car-
riers that escape recombination fall down the potential hill to
the n side. Similarly, this electron current is exactly balanced by
another (resulting from some electrons in the n side having
enough thennal energy to climb the potential hill and go to the
p side) and, therefore, the net electron current is also zero.
The above could have been deduced by simple logic. Without
an external electrical connection to the crystal, there could never
be a continuous current flow across the junction. Electric current
must flow in a loop from the generator to the load, and back
again. Therefore, it is obvious that, for whatever the reason, the
net electron and hole currents across the junction must be zero.
12. P-n Junction Diode in Reverse Bias
To have the property of rectification, the p-n junction diode

must conduct current in one direction only. Let us first consider
how the diode becomes practically an insulator current flow.
In Fig. 16A, our p-n diode has been connected to a battery,
with the positive battery terminal connected to the end of the
n material and the negative battery terminal connected to the
end of the p material. (We assume that the ends of the diode
are solidly connected to a metallic end plate of negligible resist-
ance.) This method of connection-positive battery terminal to
n material and negative battery terminal to p material-is called
reverse, or blocking bias. We shall show that when the diode is
reverse biased, it has a maximum resistance to the flow of current.
With the application of reverse bias, a field is created between
the end plates. This field is in a direction such that the positive
plate attracts the negative free electrons in the n material, and the
negative plate attracts the positive holes in the p material. With
this crowding of the majority carriers at either end of the diode
(as shown in Fig. 16A), the region of uncovered ions increases
in width. Figure 16B shows how, by broadening the area of high
charge density, this broadening of the space charge, or barrier
region, is accomplished.
Fig. 16. A. P-n junction di-

ode reverse biased. 8.
_ 1,11 :+
Charge density distribution. 0 I
C. Electrostatic potential 0°0 0 0 o:o 0 0 0•0
••
0
variations.
.:!:
0 000
0 0 010
I
0
••••• :!:
0• 0•
o<to~ 0
I ••••
••
0000~
O O0
0 0 o:o
I
0 0
<t·.".
(Al
p ~ ~--r
I
I
I
I
(Bl
REVERSE
BIAS
(Cl POTENTIAL
HILL•V11+Vo
A comparison of the polarity of this electric field (between the

end plates of the diode) with the polarity of the natural diffusion
bias field of Fig. 15B shows that they are in the same direction
and aid each other. In other words, the natural diffusion bias,
V 0 , across the junction has been effectively increased by the amount
of the reverse bias battery potential E. The new potential hill
created is shown in Fig. I6C. The hill is now much higher than
before the application of reverse bias. Because of this high hill,
the majority carrier holes in the p side will probably never have
enough thermal energy to climb, or surmount, the barrier. More-
over, the electrons in the n side will not be able to climb this
hill and diffuse to the p side. Consequently, even with the appli-
cation of a slight amount of reverse bias, conduction by majority
carriers cannot take place across the junction. Only 0.1 or 0.2
volt of reverse bias is needed to almost completely stop this

majority carrier migration across the junction.
If the action of a diode were based entirely on the impurities
contained in the material, then a reverse-biased diode would be
a perfect insulator. However, at anything but very low tempera-
tures, electron-hole pairs are still generated. The majority carriers
created in this way (i.e., the electrons thermally generated in the
n side, and the holes similarly produced in the p side) will behave
just as the other majority carriers do. They do not contribute to
a conductive current.
The thermally generated minority carriers, however, present a
problem. For instance, thermally generated minority carrier holes
in the n material will be repelled, instead of attracted to the
positive plate. They will diffuse to the junction and fall down
the potential hill. Of course, recombination will annihilate many
of these holes long before they ever reach the junction. The fact
that the potential is much higher than before the application of
reverse bias is unimportant. A hole will roll down the hill, so to
speak, regardless of its height. The holes that escape recombination
and cross the junction to the p side migrate to the negative plate.
The free electrons from this conductive plate will combine with
the holes. Hence, thermally generated holes in the n material
give rise to an electron current, through the connecting wire
from the battery.
This analysis applies equally to the thermally generated elec-
trons in the p material. The electrons left after recombination
will cross the junction by falling down the potential hill. Then
they will drift over to, and enter, the positive conductive plate,
thus causing an electron current in the wire. Minority carrier
generation, in both n and p materials, causes an electron current
in the wire that moves in the same direction. (See Fig. 15A.)
Essentially, for every hole that combines with a free electron from
the left end plate of the p material, a free electron enters the
right hand plate of the n material.
Now if, starting from zero, in Fig. 16A, the reverse bias battery
voltage were increased, this small reverse bias current would
advance to a certain point, and then stop. The limiting case
would be reached, when all of the thermally generated minority
carriers contributed to the conduction current. Since the amount
of minority carrier generation is wholly dependent upon tempera~
ture, if the bias voltage were increased beyond that point, there
would be no further increase in current. For this reason, the
reverse bias current is called the reverse saturation current, and
is designated by the symbol 10 • 10 reaches its maximum value at
a reverse bias voltage of 0.1 or 0.2 volt. (This effect is analogous

to the saturation of plate current in a vacuum tube. At the point
where the plate voltage is so high that all of the electrons ther-
mally emitted from the cathode are attracted to the plate, a
______+--1:11111---------,
+ ggggo;o't't't't
--gg ggo:o~<;><t't..::..
ggggoio<tc;>c;>c;>
(Al p _J l l N
Fig. 17. A. P-n junction di·
ode forward biased. 8,
Charge density distribution.
C. Electrostatic potential
variations.
(Bl
Vo+ Va•
REVERSE BIAS
(C)
further increase of plate voltage cannot produce more electrons.

Hence, the current is the same with increasing plate voltage
beyond the point of saturation.)
The performance of a rectifier depends greatly upon its ability
to retard the flow of current in one direction. In a vacuum diode,
this reverse current is zero, because electrons never travel from
plate to cathode. In a semiconductor, however, we have reverse
saturation current. At room temperatures, so few electron-hole
pairs are generated from the breaking of covalent germanium
bonds that the reverse saturation current is in the order of a
few microamperes. For all practical purposes, the diode may be
considered an insulator. Unfortunately, 10 varies radically with
temperatures. (/ 0 for germanium, increases about 11% per degree
centrigrade in temperature.) While at room temperatures 10 is of

little consequence, it may become serious at high temperatures.
Stability problems in high temperature transistor operation are
caused by the heat of this current.
13. P-n Junction Diode in Forward Bias
When the battery across our diode is reversed (the positive

terminal is connected to the end of the p material and the negative
terminal is connected to the n material) , the diode is said to
be forward biased. Biased in this direction (see Fig. 17A) , the
field across the diode is of a polarity such as to repel the majority
carriers. That is, the free electrons in the n side are repelled
from the negative end plate and forced toward the junction.
Morever, the holes in the material will be repelled from the
positive end plate and will crowd together near the junction. It
is easy to see now that the width of the region of uncovered ions
is much less. Essentially, the application of forward bias lowers
the height of the potential barrier. When Fig. 17A is compared
with Fig. 17B, it is seen that the polarity of the applied forward
bias is opposing the natural diffusion bias potential, V 0 • This is
shown in Fig. l 7C, wherein the potential barrier height is illus-
trated for the forward, reverse, and unbiased condition. Using our
previous analysis for the reversed and unbiased diode, the lower-
ing of the potential hill implies that many holes in the p material
now have enough thermal energy to climb the low potential
hill. Also, the electrons in the n material will rush over to the
p side. The small potential barrier offers little opposition to this
boiling mass of mobile charges.
As the electrons cross the junction and enter the p material,
they meet holes going in the opposite direction, towards the n
material. Some of these electrons will disappear due to recombina-
tion. Others will penetrate quite deeply into the p material,
before recombining with the many holes always present on this
side of the junction. Eventually, due to recombination, all of the
free electrons injected into the p material will disappear. At a
distance far from the junction in the p material, the current is
composed entirely of holes. (The relationship between total, hole,
and electron current across a p-n diode is shown in Fig. 18.)
Notice that in the n material, the holes that rush across the
junction immediately start to recombine and, therefore, the hole
current in the n material gradually decreases until it disappears
entirely. Since doping symmetry exists in this diode (each material

has been equally doped), the same situation results in the p
material. We can conclude that, at great distances from the junc-
tion (a. large distance compared with W), the forward current
will be carried by the appropriate majority carriers. The process
whereby majority carriers from one type of material cross the
junction and move into a different material, where they become
minority carriers, is called minority carrier injection. It is a very
important principle in the action of transistors.
Let us trace the path of current around the circuit. Starting
from the positive electrode of the battery, free electrons drift
p N
,___ _ _ _ _+-1l!i11-
_ _ _ _ _ _.....,
Fig. 11. Current car-
rier in the p region
is by holes; In the n
region by electrons. TOTAL CURRENT
HOLE CURRENT
fN P SIDE
DISTANCE FROM DISTANCE FROM

JUNCTION ON P SIDE JUNCTION ON N SIDE
through the conductive wire and enter the conductive end plate
of the p material, as shown in Fig. 18. The free electrons then
leave the end plate and enter the semiconductor, where, almost
instantly, they combine with holes and disappear. On the right
hand side of the diode, the majority carrier free electrons enter
the conductive plate and proceed down the wire toward the
battery. The electrons (through an electrochemical action) move
through the battery. This completes the circuit. The current
entering the diode must be exactly the same as the current leaving
the diode. So, the total current must be the same at all points in
the diode (Fig. 18) where if the hole and electron currents are lin-
early added at any point, they always total the same overall current.
If the forward bias potential directly opposes the natural dif-

fusion bias and tends to reduce the barrier potential-"Would the
resistance of the diode be reduced to zero if the forward bias
were adjusted to equal this diffusion bias? And wouldn't this zero
resistance make the forward current extremely large?"
For two reasons the answer to both questions is NO. First, the
semiconductor and end plate materials themselves have a certain
resistance. Second, with any appreciable forward current through
the diode, a voltage drop will occur across this resistance, and the
full bias potential will not be reflected across the junction. Hence,
when the external forward bias potential reaches a magnitude com-
1/1
0..
~
<(
0
0::
2
~
,os Fig. 19. Typical current volt•

10 4 age characteristics of p-n
junction diode showing
10 3
Zener breakdown.
I02
10
+
0.1 0.2 0.3 VOLTS
AVALANCHE OR
ZENER BREAKDOWN
POINT
parable to the diffusion bias potential, V 0 (about 0.l to 0.3 volt),

the forward current becomes limited by the above-mentioned
resistances. This can be seen from the volt-ampere characteristics
of the p-n diode of Fig. 19. Here, as the forward bias increases
from zero to about 0.l volt, the forward current increases at an
exponential rate. When about 0.2 volt is reached, the increase
in current becomes a linear function (straight line) of the forward
bias. The straight portion is caused by the resistance of the
material. It cannot be assumed that the current may become
arbitrarily large. Each manufactured diode has a maximum per-
missible forward current, which is based on the ability of the
device to dissipate the heat generated in its resistance. A good
practical rectifier diode has a back-to-forward resistance of I 00,000: 1.
In the discussion of forward bias, we have so far neglected a

consideration of the reverse saturation current. Now even if this
current did exist, we could not properly call it by the same name,
for the diode is not reverse-biased. Regardless of the bias situation,
however, the electron-hole pairs are still generated; they are
dependent only upon the temperature. As with reverse bias, these
thermally generated holes in the p side act like holes already
present. Electrons thermally generated in the n side act the same
as the other majority carriers. The minority carriers that are
thermally generated will be attracted to the end plates, but will
probably never get there. On the way, they will combine with the
ever-present majority carriers. The forward current of a diode is
usually thousands of times greater than this minute thermal
current and, therefore, electron-hole pairs generated during for-
ward conduction are of little consequence.
14. Zener Breakdown Voltage
It was shown previously that the reverse saturation current would

be constant with an increase of reverse bias. This is not strictly
true. As the bias is increased, the thermally-generated electrons
in the p material rush to the junction faster and faster. They are
accelerated by the field that exists across the region. If the reverse
voltage is made large enough, these electrons may collide with
other bound electrons and knock them out of their covalent
bonds. These newly created electrons may gain enough speed to
knock still more electrons out of covalent bonds. Each new elec-
tron created in this method may liberate many more electrons.
The cumulative process is called avalanche multiplication, and
results in a very large reverse current. In this state, the diode is
said to be in avalanche breakdown. (See Fig. 19.)
Even if (owing to lack of sufficient velocity of the thermal elec-
trons) avalanche breakdown did not occur, the electric field
across the junction can become large enough, with a high reverse
voltage, to directly rupture the germanium covalent bonds. The
plethora of new free electrons created in this way amounts to
the same high reverse current as with avalanche multiplication.
Although this reverse voltage breakdown may occur from two
distinct causes, this phenomenon is usually referred to as the
Zener breakdown, after the man who first explained how covalent
bonds could be broken with a high electron field. The voltage at
which breakdown occurs is termed the Zener voltage, or the
Zener breakdown voltage.
It is significant to note that once the reverse voltage has been

reduced below the Zener value, the diode will recover; that is,
the reverse current will once again consist only of the reverse satu-
ration current. Of course, if the current during breakdown is too
great, the junction may he permanently damaged. It has been found
that, during manufacture, it is possible to control the magnitude
of the Zener voltage. Special "Zener" diodes are manufactured as
voltage regulators, having breakdown voltages varying from about
IO to over 400 volts. The diode is valuable as a voltage regulator
because, during breakdown, the reverse voltage is nearly constant
for a wide variation of current. This can he seen from the almost
vertical aspect of the breakdown part of the diode volt-ampere
characteristic in Fig. 19.
15. Diode Types
There are over 2000 different types of semiconductor diodes

currently available. They vary in size, power handling capacity,
Zener breakdown voltages, forward and reverse current, tempera-
ture limitations, package dimensions and style, etc. One can always
consult a diode catalogue for specific needs; however, a general
knowledge of some of the characteristics of the most popular
diode types is advantageous.
The point contact diode is about the oldest type of semiconductor
device. The earliest radio receivers were "crystal sets." The diode
in this radio was a piece of galena (lead sulfide) crystal. The
body of the crystal served as one connection, and a "cat's whisker,"
or fine wire probe, served as the other. To make the diode work,
the cat's whisker was carefully placed against the surface of the
crystal until a spot with the proper impurities was found. Both the
diode and the receiver were so simple that home radio receivers
were constructed by the thousands, as soon as the first radio
programs went on the air. Since there was no way of knowing
the precise location of the impurity atoms in the crystal structure
(in fact, nobody even knew how it worked) , exasperating hours
were often spent searching the face of the crystal with the cat's
whisker for the most sensitive spot (where the station seemed
to he received the loudest).
Today, the construction of point contact diodes is a fine art.
Figure 20 (A) shows a point contact diode. The cat's whisker
is still present, but is now permanently attached to the most
sensitive spot. To create a junction, a surge of current is passed
through the cat's whisker that has been attached to an n-type
germanium slab. During the short period of current flow, a small
SMALL "ISLAND"
OF P-TYPE
GERMANIUM
CA T'S WHISKER
,✓-_
P-N JUNCTION
N-TYPE GERMANIUM GROWN JUNCTION DIODE ALLOY OR FUSED

JUNCTION DIODE
POINT CONTACT
(Al (Bl (Cl
TOP LEAD
WELDED SEAL
GLASS
11• ,--tlmt--... ALUMINUM-SILICON
16 JUNCTION
BASE LEAD
POWER DIODE
(RECTIFIER) SILICON
(0)
SILICON POWER DIODE
(El
Fig. 20. Most popular diode types: (A) point contact; (B) grown junction;
(C) alloy or fused junction; (0) power rectifier; (E) silicon power.
dot or "island" of p-type germanium is found where the wire

makes contact with the crystal. Thus, a p-n junction is formed
within an extremely small area. Typical forward currents of 4 to
11 ma with a forward bias as small as I volt are achievable. The
dicxle is able to withstand reverse voltages of about 250 volts.
Germanium point contact dicxles are used for relatively low
power applications such as video second detectors, computer flip-
flops, and low-level power supplies. Silicon point contact diodes
are often used for microwave mixers. However, they are more
fragile than the germanium kind and will stand only a few volts
in the reverse direction. Strictly speaking, a point contact diode
is a junction dicxle, because a junction exists. However, because
the junction is so small, it is usually referred to merely as a point
contact dicxle. (The term junction diode usually refers to other
types of dicxles.)
A grown junction diode (Fig. 20 (B) ) is prepared by adding
to pure molten (intrinsic) germanium an impurity of the proper
kind to make it either a p or n type. With the temperature con-

trolled precisely, a crystal starts to form, or grow about a tiny
seed crystal. As the doped crystal starts to form, it is slowly with-
drawn from the molten mass of germanium, called the melt.
At the proper point, the melt is doped with a different kind of
impurity, so that the crystal grown thereafter is of the opposite
type.
An alloy or fused junction diode (Fig. 20 (C) ) is prepared by
placing a small dot or pellet of acceptor impurity, such as iridium,
on one surface of a wafer of n-type germanium. When this com-
bination is fired to the proper temperature, the iridium fuses into
the germanium wafer. Thus a p-n junction is formed between the
iridium pellet and the germanium wafer.
In the diffused junction diode, a p-n junction is formed by
exposing a piece of n-type germanium or silicon to a gaseous
p-type gas. The gas diffuses through the crystal lattice and,
where the diffusion stops, a junction is formed.
POWER DIODES refer to diodes used in rectifiers delivering an
appreciable amount of power to a load. Figure 20 (D) is a
picture of a typical stud-mounted power diode that can carry an
average forward current of more than 5 amperes. Fused and dif-
fused construction are used. Typical units can withstand inverse
peak voltages up to one thousand volts. Much of the compactness
of these diodes is due to the fact that with such a low forward
resistance, little power is dissipated in the unit. Hence, large
heat radiating fins are not needed.
For high temperature operation (above 85°C) silicon power
diodes are used. Figure 20 (E) shows an actual size cross section
of a silicon junction power diode. Typical diodes with this case
size can handle 750 ma at temperatures up to 200°C.
16. Junction Capacitance
The barrier, or space charge region, W, in Figs. 15C, 16A, and

17 A is important in still another respect. If a given reverse bias is
applied to a diode, a certain space region will result from a given
width of uncovered donor and acceptor ions. If the reverse bias is
increased slightly by an amount ~ E, then the width of the un-
covered ions grows slightly larger by an amount ~ W. It does this
because, with an increase in reverse bias, the end conducting
plates of the diode have a slightly greater attraction for the
majority carriers. Therefore, they remove these mobile charges
from the extremities of the space charge region. This can be
seen in Fig. 21, where the outer row of holes and electrons are
.----_-11ill + - t:.EI-'+_ _ ____,
e e e e e'®
I
® @ @ @
Fig. 21. The effect of barrier

e e e e 61® @ ® ® @
L
capacitance. e e e e el@ @ ® @ ®
~OLD
W A:TER
REVERSE BIAS
INCREASES BY
AMOUNT t:.E
removed from their respective ions, and diffuse toward the ends
of the crystal. Of course, this movement of charges is only momen-
tary; the current stops as soon as the space charge width has
adjusted to the new, higher value of reverse bias. It happens almost
instantaneously. The time is limited, principally, by the resistance
of the semiconductor material.
This phenomenon is precisely what happens within a capacitor,
under the action of a change in applied voltage. For example, if
the applied voltage across a capacitor is constant, no current
flows. If, suddenly, the voltage increases a slight amount, there
is a momentary surge of current, as electrons are removed from
one plate and deposited on the other plate. Any device that
exhibits this property must be called a capacitor. In a semi-
conductor diode, this capacitance of reverse bias is called the
barrier capacitance. The barrier capacitance is proportional to the
junction cross-sectional area, and inversely proportional to the
space charge width W. Hence, as the reverse bias is increased,
W increases, and the capacitance decreases.
All semiconductor diodes possess this barrier capacitance in
some degree. Effectively, the capacitance is in parallel with the
diode. Thus, during reverse bias at high radio frequencies, current
is likely to flow through this capacitance, rather than being stopped
by the high back resistance of the diode. This capacitance, then,
places a high-frequency limitation on the use of the diode as a
detector, or in some other application involving radio frequencies.
The junction area of a point contact diode, however, is very
small, and so its barrier capacitance is in the order of a fraction
of a micromicrofarad (µ.µ./) . This is compared to a typical grown
junction diode with a cross-sectional area of l square millimeter,
and an accompanying barrier capacitance of 5-50 µ.µ.f. This low
barrier capacitance of the point contact diode is the reason they
are used almost exclusively in the uh£ and microwave region.

Normally, barrier capacitance is a hindrance. Certain diodes,
however, have been produced that take advantage of this capaci-
tance. In fact, they are actually used as voltage variable diode
capacitors. In manufacture, certain physical and chemical features
affecting this capacitance are maximized, so as to create a diode
having as great a change in capacitance, for a given voltage charge,
as possible. Figure 22 (A) is a graph of a typical voltage variable
diode capacitor. Notice that the capacitance decreases as the
reverse bias increases. It is this control of capacitance that makes
the diode useful as a replacement for a reactance tube in automatic
160
.,. 140
=:
~ 120
l&I
~ 100
~
o 80
J Fig. 22. (A) Graph of a typi-
cal voltage variable diode
~
~ 60
/v capacitor. (B) Schematic rep-
resentation.
V
- -
a: /
~ 40
~
~ 20
0
-16 -14 -12 -10 -8 -6 -4 -2 0
(A) REVERSE BIAS IN VOLTS
frequency control circuits, and similar devices. It has a greater

change or capacitance for a given change in applied voltage than
a reactance tube. And, the power necessary to make it operate is
infinitesimal, compared to the heater and plate power required
by a conventional reactance tube.
The voltage variable diode capacitor (marketed under the
name of Varicap, among others) is not a perfect device. The
resistance of the semiconductor material places an effective high-
frequency limit on its use. Figure 22 (B) is a schematic repre-
sentation of the device. The power losses in the series resistance,
R 8 , place an upper limit of several hundred megacycles on its use.
In early 1959, Microwave Associates, Inc. began production of a
much improved version of this diode capacitor, called the Variactor.
It had a series resistance approximately 100 times less than

previously manufactured capacitor diodes. The upper frequency
limit of this diode is still to be determined, but it is thought to
be many thousands of megacycles.
Because of its nonlinear capacity vs voltage relationship and
low loss, a new kind of amplifier became possible: the parametric
amplifier. The principal advantage of this kind of amplifier (other
than the fact that it is a solid state device operating in the micro-
wave region) is that because of its low R., it contributes very
little noise to the signal it is amplifying. This low noise parametric
amplifier, which features the Variactor diode as the amplifying
element, is now being used as an r-f amplifier in long distance radar
installations, radio astronomy, and the like.
17. The Tunnel Diode
Another semiconductor diode recently introduced is the Esaki

Tunnel Diode. The phenomenon of electrons traveling within
certain energy levels in an especially prepared p-n junction was
first observed by Leo Esaki of Japan. Since this tunneling effect
became known, active research has been devoted towards improv-
ing the diode's constructional techniques, as well as understanding
the theory of operation. At this writing, tunnel diodes no larger
tlran a normal p-n junction diode are being used in experimental
oscillators, amplifiers, detectors, etc., up to the 10,000-mc range.
The startling performance of the tunnel diode on almost no
power, low noise, and eventually, low cost, makes it a prime can-
didate for the replacement of the transistor in many fields.
18. Photoresistors and Photodiodes
When radiation falls on an intrinsic semiconductor crystal, its

conductivity increases. This was emphasized earlier, when it was
shown how high energy photons from bombarding light will
break some of the germanium covalent bonds, and create electron-
hole pairs. The resistance of the material decreases, as the number
of free electrons and holes increases. Thus, an intrinsic semi-
conductor is really a light-sensitive resistor; more appropriately,
a photoresistor. The dark resistance, or the resistance of the crystal
in the absence of light, is in the neighborhood of four-five thousand
ohms. The resistance may drop to below one thousand ohms with a
strong light. Several manufacturers make photoresistors, although
(owing to their low dark resistance) their popularity is not as
great as the photodiode.
When light falls on one side of a p-n junction diode, new

electron-hole pairs are generated along with the thermally generated
electron-hole pairs. If the diode is under open-circuit conditions,
then the light-generated minority carriers will diffuse across the
junction. In this instance, light is acting as a minority-carrier
injector. However, the net current in the diode under open-circuit
conditions must always remain zero. Consequently, to cancel this
minority current, majority carriers must flow across the junction
in the opposite direction and at the same rate. The majority
carriers must overcome the potential hill created by the diffusion
bias, before they can diffuse across the junction. Hence, the barrier
potential automatically lowers itself to allow sufficient majority
carriers to cross the junction and cancel the light-induced minority
current. Fundamentally, the barrier height has been reduced as
a result of the light. A voltage will appear across the diode's
terminals exactly equal to the amount by which the potential
barrier has been reduced. If the external circuit is closed, then
this external voltage will cause a current to flow. This process,
which is the generation of electric power from the energy of
incident light, is called the photovoltaic efject. The device is called
a photovoltaic cell; more popularly, a sun or solar battery. Under
short-circuit conditions, the cell will cause a current of up to 50 ,.,,a
per millilumen. These sun batteries are currently being used to
power uh£ transmitters in America's Explorer satellites, as well
as in portable radios, and many other devices.
When reverse bias is applied to a p-n junction diode exposed
to light, the resulting electron-hole pairs generated increase the
reverse current. Figure 23A shows the response characteristics of a
typical photodiode, and Fig. 23B shows an overall and cross-
sectional photodiode. Notice that a lens is provided to insure
that entering light is concentrated on a small area of the crystal.
So that a minimum amount of light-created electron-hole pairs
recombine, as they diffuse to the junction after generation, the
light must be shown on a spot very near the junction. The sensi-
tivity of the device therefore decreases, as the light is shown on
the crystal at points farther from the junction. 1
Unfortunately, the sensitivity of a photodiode is largely depend-
ent on the operating temperature. Both the light and dark currents
1 Although photodiodes are constructed to maximize their light-sensitive char-

acteristics, ALL semiconductor diodes are light-sensitive. The performance of a
diode as a rectifier may be seriously impaired if light is allowed to enter the
junction region. If the diode has a transparent encasing (many do), special
precautions must be taken to insure that the unit is not exposed to light.
THE P-N JUNCTION DIODE
400
'\.. 12.5ML
'
(/)
"'o: 300 "\.. 9.4ML
Fig. 23A. Average response

characteristics of a photo-
"':;;
Q,
<(
~ 200
' ~00,000 OHMS
'
LOAD LINE ------
u 6.2ML
diode. :i
100 "\..
' "\
3.1 ML
DARK
20 40 60 80 100 120
VOLTS
Fig. 23B. Two views of the

t=f ~1/~~NS PLASTIC
p-n junction photodiode. • 111.:----- 8

7
LIii 1
11
LEADS
LfP-N JUNCTION DIODE
increase with a temperature increase. However, the light current

increases at a greater rate than the dark current. Hence, the diode
is more sensitive at higher operating temperatures. When the
application requires a constant dark current over a wide tem-
perature range, balanced bridge circuits are used to cancel out
this dark current. Photodiodes are used in almost every instance
where conventional photocells are used. These include headlight
dimmers, motion picture sound systems, computer punch-card
optical readout system, and many others.
19. Review Questions
l. What is the difference between diffusion and random drifting

of thermally agitated electrons?
2. Describe the process that occurs when a piece of n and p
crystals are placed together.
3. What is meant by uncovered ions? How are uncovered ions
important in forming the space charge region?
4. Why is the word saturation contained in the term "reverse
saturation current"?
5. If a semiconductor rectifier is to be used in a plate voltage

supply for a radio, what are some of the important character-
istics of the diode you must know before installing it?
6. What is the difference between avalanche and Zener break-
down? Is a diode likely to be damaged if this occurs?
7. What is meant by junction capacitance, and how does it
occur?
8. Why are photodiodes more popular than photoresistors?
9. How are the characteristics of a semiconductor diode rectifier
altered if light is allowed to enter the junction region?
IO. What is the optimum position for light to shine on a photo-
diode? Why?
1I. Explain the theoretical difference between a photoresistor,
photodiode, and photovoltaic cell?
Chapter 3
INTRODUCTION TO THE TRANSISTOR
There are many ways to describe the action of a transistor.

Here, the subject will be approached in terms of the material
studied in Chapter 2. A theoretical description of a transistor can
be viewed as merely an extension of p-n junction theory.
In Chapter 2, a reversed bias diode was almost an insulator to
the flow of electric current. If any germanium covalent bonds
were broken, due to thermal energy, the minority carriers thus
liberated (electrons in the p side and holes in the n side) were
instantly attracted to the junction by the strong electric field
existing in this region. They fell down the potential hill, even-
tually entered the end plates, and contributed to a conductive
current. Although this reverse saturation current is of little value
(it is chiefly responsible for lowering the back resistance of the
diode), the fact that a largely unused potential hill exists across
the junction is intriguing. We have a hill, but nothing to roll down
the hill. If we could obtain electrons and/or holes to roll down
this hill, easily and inexpensively, then a useful device might be
created.
The problem is somewhat analogous to one faced by Dr. Lee
DeForest. He was confronted with a vacuum tube diode that
conducted current quite efficiently in one direction. He was seek-
ing a method by which he could easily control this flow of elec-
trons from cathode to plate. His solution was to insert a wire
mesh (called a grid) between the cathode and plate. When he
45
FORWARD REVERSE
BIASED BIASED
DIODE DIODE
,----A--. ....._
/. ...... /
I ----------- 1'- '-
Fig. 24. Pictorial diagram of
a junction transistor shown
I N rp \ N
\ as a forward and reverse
\ J biased junction diode with
\/
..:~VEE
+
" -- -
.......
---
/ .......
-- ---
/
/ +
Vee-==.:.
common p material.
applied a small potential to this grid, he found that the intensity

of the electron stream could be controlled. It took very little
voltage variation on the grid to create a rather large voltage varia-
tion across a plate load resistor: an amplifier was created. To
accomplish approximately the same thing with a semiconductor
(i.e., to create an amplifier) , a method must be found to produce
a supply of holes or electrons. If we could easily vary this supply,
then the reverse current could be varied.
One method of creating electrons or holes is to shine a light
on the junction region of a reversed bias diode. The intensity of
the light could be varied in accordance with the information to
be amplified. This is in essence a photodiode. A photodiode is
excellent for its specific purpose. However, to require a light
(with its accompanying lenses and power supply) to be an integral
part of a semiconductor amplifier would not be advantageous.
Another way of controlling the reverse current is to vary the
temperature of the diode. Unfortunately, due to the mass of the
object, there would be a rather long delay between the time of
application of heat and the time when the semiconductor reached
the desired new temperature. This sluggishness is called thermal
capacity. What is needed is a method instantaneous in action.
Consequently, an all-electronic means of genera.ting holes or elec-
trons, and then injecting them into the junction region of a
reversed bias diode.
A process called minority carrier injection takes place within
a forward bias diode. (See Chapter 2.) In this process, the thermal
energy given to electrons in the n material is sufficient to cause
them to climb a relatively small potential hill. When they cross
the junction to the p material, they become minority carriers.
This is precisely the process we are looking for. The question
remaining is: Can a forward bias diode be attached to a reversed
bias diode in a way such as to insure that minority carriers
created in this forward bias diode can be injected into the
_junction region of the reversed bias diode? In practice, this is done
INTRODUCTION TO THE TRANSISTOR 47
quite simply (see Fig. 24) by attaching both diodes to a common

element. The resultant device is called a junction transistor,
for it is made up of two junction diodes having a common p
material. It is also called an n-p-n transistor, because the piece of
p material is sandwiched between two pieces of n material.
The left-hand junction of this transistor forms a forward bias
diode. When in forward bias (as evidenced by the polarity of the
EMITTER COLLECTOR
Fig. 25. Schematic
representation of Fig.
24. +
- Vee
BASE
battery voltage, VEE) , minority carrier electrons are injected into

the p region. For this reason, the left-hand forward bias junction
is called the emitter. The right-hand junction is in reverse bias.
If it were by itself-unattached to the rest of the transistor-the
only reverse current would be the reverse saturation current. The
strong electric field that exists across this junction, however, will
attract or collect any minority carriers that are injected into this
region by the emitter. Because the right-hand junction attracts
minority carriers, it is called the collector. The region between the
emitter and collector is called the base. Schematically, this tran-
sistor is illustrated in Fig. 25. The arrow on the emitter pointing
away from the base indicates that this is an n-p-n transistor. It
points in the direction that positive current flows when the emitter
is in forward bias. Hence, in forward bias, electrons flow in the
opposite direction of the arrow. 1
So as to correlate the many things that are happening within an
n-p-n transistor, the various electron and hole currents will be
dealt with in detail.
The same n-p-n transistor is depicted in Fig. 26, where the
emitter is shown in forward bias. Consistent with previous p-n
junction theory, minority carriers will be injected across the
emitter-base junction into the p material. Since it is this injection
of minority carriers (electrons) into the base region that is the
1 The direction of positive current (flowing from plus (+) to minus (-) in 1he
external circuit) is purely an engineering convention. In this book, the <lirec-
tion of electron flow will be taken as the direction of the current.
sole purpose of the emitter, the efficiency of the system could be

improved by making the free electrons carry the greatest propor-
tion of the forward current in the vicinity of the emitter-base
junction-instead of the 1: 1 ratio between holes and electrons
that normally exists in any symmetrically doped forward bias
diode (see Fig. 18). This is done by doping the n material more
heavily than the p material. It insures that on the right side of
the emitter-base junction, almost all of the forward current will be
carried by electrons. This is a very desirable situation because
CONDUCTOR END PLAtE
( ELECTRONS ENTER N-MATERIAL
FROM WIRE) ( ELECTRONS DIFFUSING
·..... :• • • :• ....
P / N ACROSS BASE)
\
• .......
•. :.· ...
N
• • •••••••• •..•··-~.
.. ·..........
•:•.... :.......
•• • •
•·.•:-~•
• ·................
• •• • ••• •:-::: . •••••••
• • • •••••• •• ••••• • • • • • •• • • • • ••• •
-. . .• .• .•..•.•.•..•.:••••
...
•
••••• •• • •••••• • •
••• •••• •• • • • • • • •••••
••
.. .
• .. . . . . . .. .. . ..
• • • • • • • • •• • • ••••••• ··=:-:•:::: ..··
•••••••
., : ·=· •• ••••••• • • • • ••••• •
~
~
• • • ._( SMALL NUMBER OF

I
- BASE CURRENT'-
( 2%) ,,
ELECTRONS REACHING
BASE TERMINAL)
:!:_-
~ VEE
+
Yee
-
EMITTER CURRENT j COLLECTOR CURRENT
-- --
_ ( IOOo/.) (98%)
-
( NOTE: CURRENT IN WIRE IS ELECTRON CURRENT)
Fig. 26. The chief current carriers in an n-p-n junction transistor are electrons.
(with reference to Fig. 18) the electron current in the p material

decreases exponentially as the distance from the junction increases,
due to the recombination of these free electrons with the majority
carrier holes. Therefore, the greater the concentration of electrons
at the emitter junction, the greater will be the concentration of
electrons at any nominal distance from this junction.
Now if these electrons were uninfluenced by the collector, they
would proceed through the base material to the base terminal
connection in normal junction diode manner. However, with a
reversed bias diode in the same region, any injected minority
carrier electrons that happen to drift, by diffusion to the right,
and actually reach the collector will fall down a large potential
hill. Consequently, two things tend to happen at once. The
minority carrier electrons injected in the p material diffuse down

to the base terminal connection (as would happen in any forward
bias diode) while at the same time they diffuse to the right
towards the collector. They diffuse toward the collector because
there is a reduced concentration of electrons in this vicinity,
i.e., any electrons reaching the collector will immediately fall
down the potential hill leaving the base side of the collector-base
junction deficient in this type of mobile carriers. This diffusion
of electrons to the collector is to be maximized in a successful
junction transistor.
Unfortunately, although the base p material is not as heavily
doped as the emitter material, millions of holes are still present.
Some of these majority carrier holes are bound to combine with
the injected electrons, as they diffuse across the base region to the
collector. Whether there will be any electrons left to actually
reach the collector is largely a function of the distance they have
to travel through the base material. To reduce the number of
electron-hole recombinations, and at the same time reduce the
likelihood of too many electrons diffusing down to the base
terminal, the width of the base wafer is made exceedingly small
(about one-thousandth of an inch thick) with respect to its
cross-sectional area. This practical solution assures that most of
the injected minority carriers actually reach the collector (about
98%). The fraction of the emitter current transferred to the
collector circuit is called the current gain.
The reader may ask: Can a transistor be operated backwards,
with the collector and emitter interchanged? This can be done.
The operation, however, would be at greatly reduced efficiency,
since the transistor is designed primarily as a one-way device. For
instance, the area of the collector junction is usually larger than
the area of the emitter junction. The collector is made larger so
as to offset the spreading effect of the free electrons as they cross
the base region. If the collector were smaller in cross-sectional
area, some of the free electrons might hit the surface of the base
material or, more likely, diffuse to the base electrode connection.
In addition, the emitter resistance is often made less than the
collector resistance, to lessen the input power needed to create
a given emitter current. For these reasons, transistors are usually
manufactured asymmetrically. (Symmetrical transistors, with equal
emitter and collector junction areas are made, but only for
specialized tasks. An example of one application for a symmetrical
transistor would be a transistor phase detector. Symmetrical tran-
sistors do not operate at maximum efficiency in either direction.)
y -11 +
N
I
Ip
I
DIFFUSION BIAS POTENTIAL
-11 +
N C
I I
1-----1
I BASE I
~
I I
I Fig. 27. Electrostatic poten-
tia I variation across n-p-n
O EMITTER: ICOLLECTOR junction transistor. (A) un-
@ I I biased; (B) forward biased

I emitter.
I
I I
-v I -.-1
I
i_.§PACE CHARGE
WIDTH W
--!EFFECTIVE!-.- I
BASE
I WIDTH : I
1,- - - - -, I UNBIASED
1----,
/ I YTRANSISTOR
0 I
1 BIASED
EMITTER REVERSE TRANSISTOR
BIAS POTENTIAL I
I COLLECTOR
I REVERSE
I BIAS
I POTENTIAL
®
+v _:_1
The electrostatic potential vanat1ons across an unbiased n-p-n
junction transistor are shown in Fig. 27. For the unbiased case,
the up and then down potential hill is merely the potential hill
of two unbiased p-n junction diodes placed back to back (this
can be checked by referring back to Figs. 16A, 16B, and 16C).
As was true in the discussion of diffusion bias, the potential
barriers adjust themselves so that no net current will flow across
either junction. If forward bias is placed across the emitter (Fig.
27), the emitter-base barrier potential hill will be lowered by the
amount of the forward bias. With the application of reversed
bias to the collector, the collector-base barrier potential hill will
be raised by the amount of the reversed bias. From Fig. 27, it can
be seen that electrons easily climb the small emitter potential hill,
diffuse across the base region, and then fall down the high collector
potential hill. Once the free electrons reach the collector and
fall down the hill, they diffuse over to the collector terminal, and
enter the connecting wire. Proceeding through the collector and
emitter bias batteries, they enter the n semiconductor material,
and the circuit is complete. The current flowing in the common,
or base, lead in Fig. 26 is caused by several things. (1) The
holes that combine with some of the injected electrons as they
diffuse across the base region must be supplied from the base
terminal (this is done by electrons entering the base terminal
from the base p material). (2) The forward current of the emitter
diode will contain some holes from the p region and these also
must be created by electrons entering the base terminal from the
base material. (3) The reverse saturation current (1 00 ) of the
reversed bias collector diode must also flow through this same
base terminal lead. (This latter effect is so small as to be con-
sidered negligible at room temperature.)
Since the current entering the emitter is essentially equal to
the current leaving the collector, one may believe that amplifica-
tion cannot be achieved. This, however, is not true. The amplifi-
cation property of a transistor depends on the fact that the
emitter and collector circuits have vastly different resistances. In
effect, current is transferred from the low-resistant emitter circuit
to the high-resistant collector circuit. The word transistor itself is
a contraction of the words transfer-resistor.
As a simple example of how the n-p-n transistor acts as an actual
amplifier, let us examine Fig. 28. Here, a signal source, V., has
been connected to the emitter, in series with the emitter bias source.
A load resistance, RL, has been connected in series with the col-
lector bias supply. Connected in this manner, with the base
N P N I
I N-P-N
I
I I I
I I
1---:--1-1
EMITTER 13ARRIER
POTENTIAL WITH
I
Vee
,1
NO Sl~NAL I
I
Fig. 28. The n·p-n transistor as an amplifier, with zero signal.

terminal common to both emitter and collector bias batteries,

the circuit is referred to as a common, or grounded, base amplifier.
When the signal voltage is zero, the number of free electrons
crossing the emitter junction and entering the base region is
determined solely by the emitter-base bias, as depicted in Fig. 28,
which, for convenience, is a repetition of Fig. 27 (B). Now, assume
that the signal voltage is increased slightly in the negative direc-
tion. Since the signal source is in series with the emitter bias
battery, a deflection in the negative direction will add to the
emitter-base forward bias battery potential, causing the emitter
N p N
N-P-N
- + - +
Fig. 29. The n-p-n transistor as an amplifier, with negative-going signal.
potential hill to be reduced further. This permits a greater num-

ber of free electrons to enter the base region (Fig. 29) , slightly
increasing the emitter current. Since most of the electrons entering
the base region will diffuse to the collector circuit and through
the load resistance, RL, this increase in emitter current, due to
the signal fluctuation, will produce an approximate equal current
increase in the load resistance.
If the signal voltage moves in the positive direction the same
amount that it moved in the negative direction, the emitter
potential hill will be increased by the amount of this deviation,
since the signal voltage now subtracts from, instead of adds to, the
emitter bias potential. This increase in the height of the emitter
potential hill decreases the diffusion of free electrons to the
base region, and there will be less current flow in the load resist-
ance (Fig. 30) . It can be seen now that the fluctuations of emitter
current due to the signal voltage changes are essentially reproduced
in the collector circuit. In the low-resistant emitter circuit, these
signal induced current changes produce a very small voltage
change. However, in the collector circuit, where the load resistance

is relatively high, the same current changes produce a large voltage
variation. Hence, voltage amplification is achieved in the device.
For a more quantitative treatment of the same phenomenon,
let us assume that the signal voltage increases in the negative
N p N
I
I
I N-P-N
I
,--f-1
I
I I
Vee I
I
HILL WITH
I
NO SIGNAL
Fig. 30. The n·p-n transistor as an amplifier, with positive-going signal.
direction by an amount ~ V. (pronounced delta V1) • The forward-

biased emitter represents a low resistance, r., so that this slight
increase in emitter voltage will produce a correspondingly large
increase in emitter current of:
(l)
The amount of this increase of emitter current that reaches the

collector is dependent upon the current gain of the transistor.
If, for instance, 98% of the free electrons injected into the base
region always reach the collector, then 98% of this increase in
emitter current would also reach the collector. Hence, the amount
of ~ IE reaching the collector (and producing a change in col-
lector current, ~ le) is equal to the current gain X ~ IE. The
current gain for a common base circuit is called alpha (a) . Then:
(2)
This change in collector current will produce a corresponding

change in the voltage across the load resistance (called the output
voltage, V0 ) equal to:
(3)
The voltage amplification of the amplifier (A,,) is defined as

the ratio of the change in output voltage to the change in input
voltage. Therefore:
A - A.Eo (4)
V - A.E1
Expressed in tenns of the load and emitter resistances:

Av = A.E 0 = aA.IERL = aRL (5)
A. E 1 A. !Er~ r.
Since a is approximately 1, the voltage amplification is essentially
equal to the ratio of RLfr•. With a load resistance of, say, 3000
ohms, and a typical emitter resistance of about- 30 ohms, the volt-
age amplification is 3000/30 100. =
A power gain is also realized. The change in power dissipated
in the load resistance (power output, P 0 ) due to the change in
collector current is:
(6)
The change in input power dissipated in the emitter forward
resistance (input power, Pi) is:
P1 = A.IE2 r. (7)
The power gain (Ap) is the ratio of the two:
(8)
The details of the grounded base amplifier will be dealt with

in Chapter 4. This example was given only to demonstrate the
principle of power and voltage amplification occurring when a
current is transferred from 'a low-resistance circuit to a higher
resistance circuit. In practice, other factors must be considered
before a transistor amplifier may be designed.
20. Comparison of Transistors to Vacuum Tubes
Let us examine some of the similarities and dissimilarities

between the transistor and the vacuum tube. Knowing their re-
lationship, we can often call upon our previous knowledge of
vacuum tubes, when designing transistor circuits.
Both transistors and vacuum tubes are designed to amplify a
signal. Both the transistor and the triode vacuum tube have three
basic elements. In fact, the transistor is often referred to as a
semiconductor triode. This external similarity of the two devices
TRANSISTOR VERSUS VACUUM TUBE
PLATE
N
Fig. 31. Comparing BASE GRID
transistor and vacuum
tube elements.
N
CATHODE
suggests that the emitter, collector, and base of a transistor have

an analogous relationship to the cathode, plate, and grid, respec-
tively, of a triode vacuum tube (Fig. 31).
The current carriers in a vacuum tube (electrons) are liberated
by the cathode. The equivalent job in a transistor is performed
by the emitter; it emits current carriers. Hence, the cathode in a
vacuum tube is analogous to the emitter in a transistor.
Current carriers in a vacuum tube travel through the vacuum
tube to the plate, and then to the external circuit. In a transistor,
current carriers travel across the base region, and to the collector.
The plate of a vacuum tube is analogous to the collector in a
transistor.
The electrons of a vacuum tube must flow through the grid
before reaching the plate. In a transistor, the current carriers
must flow through the base section before reaching the collector.
The grid is analogous to the base. This relationship is further
appreciated when we consider the effect of an input signal. In a
vacuum tube, the signal varies the voltage between the cathode
and grid. In a transistor, the signal varies the voltage between
the emitter and base.
The transistor is essentially a CURRENT-OPERATED DEVICE; that
is, differences in current flow control its operation. The common
base current amplification factor, a, is the ratio of a change
in collector current produced by a change in emitter current. The
vacuum tube is a VOLTAGE-OPERATED DEVICE; electrostatic potential
variations between its elements control its operation. The voltage
amplification factor p. is the electron tube equivalent to a. It is
the ratio of a change in plate voltage produced by a change in
grid voltage.
Along with the above analogous relationships, in a vacuum
tube, the current carriers move through a vacuum; in a transistor,
they move through a solid.
INPUT
GROUNDED BASE
GROUNDED GRID
(Al
OUTPUT
INPUT INPUT
GROUNDED EMITTER GROUNDED CATHODE
(Bl
o---e---t,.,...__ _-11-• ,-- +

OUTPUT
GROUNDED COLLECTOR GROUNDED PLATE

(EMITTER FOLLOWER) ( CATHODE FOLLOWER)
(C)
Fig. 32. Comparison between the transistor and accompanying tube configurations.
(A) Grounded Base vs Grounded Grid; (B) Grounded Emitter vs Grounded
Cathode; (C) Grounded Collector vs Grounded Plate.
21. Comparison of Basic Transistor and Vacuum Tube Circuits
This analogy of a transistor's elements to those of a vacuum

tube suggests that a grounded base amplifier is similar to the
grounded grid vacuum tube amplifier, as shown in Fig. 32 (A) .2
2 To further illustrate the analogy, in the illustration the input signal to the
transistor amplifier is applied across a resistor. This is done to allow the signal
to be fed through a capacitor and hence prevent bias current from flowing
through the signal source, which is not shown. The signal-induced voltage
variations across this resistor will add and subtract from the emitter bias in
the same way as a signal placed directly in series with the emitter bias battery.
In the grounded, or common, grid circuit, the cathode current

equals the plate current; in the grounded, or common, base cir-
cuit, the emitter current approximately equals the collector cur-
rent. (Voltage and power gains for the common base amplifier are
in the order of 100 and 400.) The input resistance of both
grounded base and grounded grid amplifiers ranges from 30 to
about 300 ohms. The output resistance is about 100,000 to 500,000
ohms, for both circuits. There is no signal polarity inversion in
either circuit; i.e., when the input signal goes positive, the output
signal also goes positive.
The grounded base configuration is not the only method of
connecting a transistor as an amplifier. Figure 32 (B) shows a
grounded, or common emitter, amplifier. (It derives its name from
the fact that the emitter is common to both the input and output.)
Similar to the common base amplifier, the signal in this circuit
is also applied between the emitter and base, except that in this
instance the collector current does not flow through the input
circuit. Both the tube and transistor are capable of current gains,
and there is a signal polarity inversion between input and output.
The grounded emitter circuit has a higher input resistance
(300-1,000 ohms) and a lower output resistance (5-50,000 ohms)
than the grounded base amplifier. For these reasons, the grounded
emitter is the most popular general transistor amplifier circuit.
(Current gains much greater than one are achievable. The reason
for this will be dealt with in Chapter 4.)
The grounded collector circuit of Fig. 32 (C) is similar to the
grounded plate, or cathode follower, vacuum tube circuit. The
input resistance is much higher than the other two configurations.
The output resistance is very low. This circuit is primarily used
for impedance matching between two circuits. There is no signal
polarity inversion between input and output. For this reason,
the grounded collector circuit is often called an emitter follower,
signifying its similarity to the cathode follower. Both circuits
have a very high current and power gain, with a voltage gain of
less than one.
We have concerned ourselves with the n-p-n transistor. If holes
may act as positive electrons, is not a p-n-p transistor possible? It
is not only possible, but it is more popular than the n-p-n tran-
sistor. The two types of transistors are practically identical in
their outward performance, except that holes are now the major
current carriers, and have a slightly lower mobility than free
electrons. Thus, a separate theoretical discussion of p-n-p tran-
sistors is unnecessary.
A p-n-p common base transistor amplifier is shown in Fig. 33.
Notice that the symbol has the emitter arrow pointing towards
the base. (The emitter arrow points in the direction that positive
current will flow when the emitter is forward biased. Of course
the arrow points in the opposite direction of electron flow.) In
addition, the polarity of the bias batteries is opposite to that in
the n-p-n transistor. It is important to immediately recognize the
schematic differences between n-p-n and p-n-p transistors. Failure
to do this will probably result in connecting the bias batteries
backwards. A transistor is a physically rugged device; however,
INPUT OUTPUT
Fig. 33. A p-n-p common

base transistor amplifier.
----------+
it is very sens1t1ve to electrical overloads, etc. Connecting the

collector batteries backwards will result in ruining the device.
Become familiar with the n-p-n and p-n-p versions of the three
basic amplifier configurations. These circuits are presented in
Fig. 34, along with their accompanying bias battery polarities.
Transistors and diodes are classified according to the same
rules. Transistors may be called grown junction or diffused junc-
tion or diffused base or surface barrier or point contact, etc.
Junction formation is the governing rule.
The grown junction transistor is produced by the technique
described in connection with the grown p-n diode. Grown junction
transistors are characterized by a rather low noise figure and
relatively stable characteristics.
The construction of the diffused or alloy junction transistor is
analogous to the alloy diode. Two small dots of indium are
placed on either side of a thin wafer of n-type germanium. The
entire structure is heated to about 500°C, for a short time (Fig.
35). Under the influence of heat, the indium dots dissolve into
the germanium wafer (this action is similar to soldering action),
changing the n-type germanium to p-type germanium. Notice
that the collector dot is larger than the emitter dot. This is to
counteract the spreading effect of the holes as they cross the base
region on their way to the collector. If it is larger than the emitter,
it will collect a greater percentage of this divergent flow of holes.
COMMON BASE COMMON EMITTER COMMON COLLECTOR

N-P-N TRANSISTOR CIRCUITS
COMMON BASE COMMON EMITTER COMMON COLLECTOR

P-N-P TRANSISTOR CIRCUITS
Flt• M. Beale transistor configuration; (A) n-p-n tranabtor drculfll

(B) p-n-p tran1l1tor circuits.
N-TYPE GERMANIUM
/
--------....
IRIDIUM \
I I
I I
I ,..__, I
COLLECTOR
I ~Lr-:r::P !
P-TYPE
GERMANIUM
I BASE I
I COLLECTOR EMITTER
I (CONSTRUCTION) :
I,. o.4so• .. 1
Flt. 15. DlffvMd (alloyed) junction trallli1tor.

A diffused base transistor does not have an analogous diode type.

To manufacture the transistor, a wafer of p-type germanium 0.003
inch thick is heated to the proper temperature. Under the action
of heat, a layer of n-type impurity is caused to diffuse a short
distance into the germanium slab (Fig. 36), thus forming the
base region. A thin strip of p-type germanium, forming the emitter,
BASE
N-TYPE F19. 36. Diffused base junc-

IMPURITY tion transistor.
P-TYPE
GERMANIUM
COLLECTOR
is evaporated and alloyed atop this n-impurity slab. The collector

becomes the original p-type germanium wafer. Double gold leads
are welded to the emitter and base elements. This type of con-
struction possesses several inherent advantages. It allows operation
at an extremely high frequency. The 2N509 and 2N5J7 today
are in commercial production, operating in the 1000 me region-
something considered fantastic a short time ago. Although fabri-
cated of germanium, this transistor can withstand a temperature
up to 100°C. It has a very low collector barrier capacitance and
thus helps to provide good high-frequency characteristics. The dif-
fused base transistor was used in the telemetering transmitters of
both the Vanguard and Explorer satellites, on a frequency of about
106 me. This is but one striking example of how the transistor has
surpassed the vacuum tube in certain specialized tasks requiring
reliability, ruggedness, and efficiency.
The Philco surface barrier transistor is another type of high
frequency transistor. It utilizes the junction principle for its
operation, but does not use p- and n-type germanium. It utilizes
only n-type germanium. The emitter and collector are electroplated
on opposite sides of an n-type germanium wafer. The emitter and
collector materials do not diffuse into the n material (as in the
diffused junction transistor) , but remain on the surface only.
ELECTROPLATED ELECTROPLATED
EMITTER COLLECTOR
Fig. 37. Surface barrier junc•
tion transistor.
Figure 37 illustrates the details of the surface barrier transistor.

Because of a rather complex energy phenomenon that exists at
the surface of the germanium, a high negative electric field is
built up here, repelling free electrons attempting to enter the
surface region. It is this surface of the germanium crystal that is
referred to as the "barrier." One metal contact to the crystal is
made positive; it reduces this negative barrier; the contact serves
as the emitter. The other metal contact is made negative with
respect to the main crystal; this serves as the collector. The main
wafer of n-type germanium serves as the base. The entire structure
is an n-p-n transistor configuration. Its high frequency performance
is due to the extremely thin layer of germanium separating the
collector from the emitter. This transistor is used as a video am-
plifier, an oscillator, etc., in the vhf region.
The point contact transistor was the first type of transistor
available. The method of manufacture is quite similar to that of
making a single point-contact diode. Two phosphor-bronze wires
are placed against a wafer of n-type germanium. A surge of current
is passed through the two wires, thus forming two islands of p-type
germanium directly beneath the point where the wires touch the
germanium. Although a point contact transistor may serve in
certain circuits where two other types of transistors would other-
wise be needed (such as a multivibrator) , because of its poor
reliability and unstable characteristics, it is rarely used today.
Most transistor manufacturers have discontinued production of
point contact transistors. For this reason, and because point con-
tact transistor theory is not completely understood, we will deal
exclusively with the various types of junction transistors.
I. Why is the base region of a grown junction transistor rather

narrow with respect to the width of the emitter or collector
sections?
2. Why is the effective base region narrower with a reversed bias
collector junction than with an unbiased collector junction?
3. Why does a common base transistor have a voltage and power
gain when the current gain is less than unity?
4. Where does the word transistor originate?
5. What three vacuum tube circuits are analogous to the common
base, common emitter, and common collector transistor circuits
respectively?
6. Why could not temperature be used as a minority carrier
injector into the base region?
7. What is a symmetrical transistor? Are normal transistors asym-
metrical? If so, why?
Chapter 4
THE TRANSISTOR AS A CIRCUIT ELEMENT
The analysis and design of a transistor circuit is a logical step-

by-step procedure. We will develop a simple method by which
a transistor amplifier may be designed-knowing the basic charac-
teristics of the transistor to be used and the specifications of the
proposed amplifier. Here, we will deal mostly with transistor char-
acteristics and the problems involved in establishing the quiescent
point of operation for the different configurations. Since we will
be dealing with many types of voltages and currents, an explanation
of the various symbols to be used and the rules governing their
use is in order.
Basically, any voltage or current, etc., may be denoted by a
letter and a subscript. If the letter is of the upper case variety,
it indicates a de or an nns value. If it is a lower case letter, it
represents an instantaneous value. Thus I, JI, and P represent
nns or de values of current, voltage, and power, respectively.
Instantaneous values of the same quantities: i, v, and p.
The subscript has a variety of meanings. If voltage is being
measured, the subscript usually consists of two letten. The fint
letter denotes the electrode at which the voltage is measured, and
the second subscript letter denotes the reference electrode with
respect to which the measurement is made. De and instantaneous
total values are represented by upper case subscripts, and instan-
taneous variations from the no-signal value (called the quiescent
value) are represented by lower case letten. Bias supply voltages
are indicated by a repeated upper case subscript.
63
TABLE 1
COMMON TRANSISTOR SYMBOLS
V EB = The emitter bias supply voltage.

V cc The collector bias supply voltage.
VcE The quiescent voltage between collector and emitter.
VCB = The quiescent voltage between collector and base.
V 00 The rms signal voltage, or the rms variation from the
quiescent value.
vcB = The instantaneous total value of collector to base voltage.
v 00 The instantaneous variation from the quiescent value
of the collector to base voltage.
NOTE: The last four symbols are shown in Fig. 38, which repre-
sents the voltage variations from collector to base of a
hypothetical transistor.
Figure 39 shows the polarity convention for the n-p-n and p-n-p
common base configuration. Here, whether the transistor is an
n-p-n or p-n-p device, the emitter and collector voltages are always
DEFINED as being positive with respect to the base. Hence, if (as
actually happens) the emitter is negative with respect to the
base, as in an n-p-n transistor, the negative sign will accompany
this voltage, because it is opposite to the convention. Remember
to use the base as a reference. Then the actual voltages on the other
elements will have the correct and logical polarity.
It is now known thait current in a conductor is actually the
flow of electrons. This was not known to early circuit theorists.
Hence, a convention was adopted that electric current flows from
positive to negative. It was visualized that there was a rise in
potential within a battery (from minus to plus) and a fall of
potential outside the battery (from plus to minus). Early circuit
theory was developed with this convention in mind. When the
electron theory was developed, it was too late to change the old
convention. Hence, the old concept that current flows from + to -
is still in use. It has often been argued that the convention should
be changed. The argument centers about what we mean by current
flow.
Basically, current flow is the flow of electrical charges-be they
minus electrons, or positive ions or holes. In a vacuum tube,
THE TRANSISTOR AS A CIRCUIT ELEMENT 65
Vcb(RMS SIGNAL VALUE)
LLJ
C)
c(
1-
..J
0 Vea veb ( INSTANTANEOUS VARIATION FROM
> (D-C QUIESCENT QUIESCENT VALUE)
VALUE)
Vea
( INSTANTANEOUS VALUE
OF TOTAL CURRENT)
0 TIME
Fig. 38. Voltage variations from c<>llector to base of a hypothetical transistor.
P-N-P N-P-N
---le
Vea
Fig. 39. Polarity convention for n-p-n and p-n•p common base configurations.
electrons are the main carriers of current and so the old con-
vention of current flowing from the plate to cathode might be
considered incorrect. However, there are gas tubes where positive
ions do travel from plate to cathode. In transistors, neither the
electron nor the positive current theory is completely correct.
Both positive holes and negative electrons are the carriers of
electricity, depending on whether the transistor is of the n-p-n
or p-n-p variety.
Because almost all transistor literature (and even the schematic
symbol of a transistor) has been developed with positive current
theory in mind, this convention will be applied here. Even if the
reader is aware of the electron theory, confusion will not arise.
No matter which convention is followed, the SAME results are
arrived at. In Fig. 39, this cun-ent convention is applied to the
two types of transistors. The emitter arrow in each case always
points in the direction of positive current flow, when the junction

is in forward bias. Notice also that the currents are defined as
being positive for both configurations, when they flow into the
transistor. Now, under actual bias conditions (shown in Fig. 40),
the emitter and collector are not always positive with respect to
the base, and the currents do not always flow into the transistor.
P-N-P
---
-le
N-P-N
- le
+
Vea
+
Pit, 40. N-p-n and p-n-p common ba.. configurations under actual bl111 conditions.
When the direction of current is opposite to the defined direction

into the transistor, a negative sign is assigned to it. Thus the p-n-p
transistor has a positive emitter current and bias, and a negative
collector current and bias. The n-p-n transistor (Fig. 40) has a
negative emitter current and bias, and a positive collector current
and bias.
23. Transistor Characteristics
In Chapter 3, it was noted that even when no signal was applied

to the common base amplifier, an emitter and collector current
flowed. The presence of this constant bias current is necessary
for the proper operation of a transistor, just as the presence of
plate and grid bias voltages are necessary for the successful opera-
tion of a vacuum tube amplifier. This no signal, or quiescent bias
current in a transistor presents an average current, which the
signal current either adds to or subtracts from. It is important
that this quiescent current is of the proper magnitude, so that
equal positive and negative excursions of the input signal produce
symmetrical variations of the output current about an average
value. An incorrect quiescent bias current may cause one or more
forms of distortion in the output signal.
A constant quiescent bias current flowing in the collector will
10
9
I
I I
2N35 N-P-N
COMMON BASE _
30•c -
I,:•-8ma.
8
I
-7
7 I
/!.Vee
•'
-6
6 I
le -5
5
I
4 -4
I
3
-3
I
2
Ill/A -2
I
IHI/IA -I
0
I ~6/11/t lo
-2 0 2 4 6 8 10 12 14 16 18 20 22 24
COLLECTOR-TO-BASE VOLTAGE (Veal IN VOLTS
Fig. 41. Static collector characteristics for the common base configuration of a
2N35 n-p-n transistor.
cause a collector-to-base voltage drop. If the signal current in the

emitter circuit causes the collector current to vary above and
below its quiescent value, then the collector-to-base voltage will
also vary above and below its quiescent value, as shown in Figs.
28, 29, and 30.
Unfortunately, the resistance of the emitter and collector junc-
tions are not always constant. Hence, Ohm's Law cannot always
be used to express the relationship between the various currents
and voltages in a transistor. For this and other reasons, this infor-
mation is usually supplied by the manufacturer in graphical form.
The two most commonly used graphs of transistor characteristics
deal with the collector and emitter respectively. Of these, the
collector graph is the most popular and useful. Figure 41 shows
a graph of the static collector characteristics for the common
base configuration of a 2N35 n-p-n junction transistor. The graph
consists of a series of curves. Each curve shows, with the emitter
current held constant, the collector current variation as the
collector-to-base voltage is changed. The curves are obtained quite
simply, using the test apparatus shown with the circuit of Fig. 42.
After the collector current is observed and plotted for variations
in the collector-to-base voltage, and for a given value of emitter
Fig. 42. Test circuit required

to obtain characteristics of
Fig. 41.
Vee===
22.5
V
current, the emitter current is changed, and the procedure is re-

peated. Different curves are shown for increments of 2 ma of emitter
current. The entire collection is called a family of collector
curves, with the emitter current as a parameter (a value of emitter
current set for each curve) . The family of curves shows what
value of emitter current must exist to produce any combination
of VeB and le.
The minus sign in front of each emitter current line in the
graph exists because in an n-p-n transistor, lB flows out of the
transistor.
Many interesting facts come to light with the aid of this graph.
For instance, a rough estimate· of the dynamic output resistance
of the transistor may be obtained. This is the resistance that the
load "sees" when looking back into the transistor between the
emitter and the base leads. It is analogous to the dynamic plate
resistance may be obtained by dividing a certain change in the
resistance may be obtained by dividing a certain change in the
collector-to-base voltage by the accompanying change of collector
current, while the emitter current is held constant. This is the
slope of each le = VeB line. For example, in Fig. 41, the 6 ma
emitter line moves upward about 0.2 ma :as the collector-to-emitter
voltage varies from 0 to 20 volts. The output resistance (R 0 ) is
then:
20volts
Ro=
AVcB
Ale 0.2ma
= 100,000 ohms (9)
As previously stated, the output resistance of the common

base configuration is quite high. This permits a high load resist-
ance. The voltage gain is roughly proportional to the load resist-
ance divided by the emitter resistance. Notice that the output
impedance or resistance is quite constant over a wide variation
of V OB and le, as evidenced by the fact that the slope or tilt of
each base current line is about the same. (It is important to
state here that the dynamic R 0 is not the resistance that would be
measured by connecting a voltmeter across the collector base
junction and an ammeter in series with the plate. This volt-

ammeter resistance is the de static resistance. The slope of the line
is the ac dynamic resistance. Chapter 5 contains a more detailed
discussion of the differences between these two resistances.)
Equation (2) gave us a relationship relating the change in
collector current, the change in emitter current, and the current
gain of the common base.
(10)
Solving this equation for a, yields a definition for this quantity:
~le
a= - - (11)
~IE
V cB constant
a is taken with a constant collector-to-base voltage. Its official
name is the forward short-circuit current gain for the common
base configuration. 1 To obtain this information from the common
base collector characteristics, it is noted that for a constant
collector-to-base voltage of 2 volts, the collector current varies
from O to 6 ma, while the emitter current only varies from O to
5.95 ma. Thus:
_ ~1 0 _ 5.95ma = _
995 (12)
a - ~IE - 6.00ma -
(The symbol :::: means approximately equal to)
a is not related to the external circuit. It applies only to the
common base configuration. The value obtained from this graph
(as obtained from any graph) is not exact; the manufacturer
specifies a more reliable value. a for the 2N35 is given as 0.98;
typical values for a vary between .90 and .99. (Again, this is the
dynamic a, of concern when the amplification of ac signals are
involved. When dealing with de switching circuits, the de, or
static a is useful.)
Although it cannot be detected from the graph, there is a
minute collector current flowing even when the emitter current is
zero. Since, when the emitter current is zero, no minority carriers
are injected into the base region, this current can only come from
thermally-generated minority carriers. It is the familiar reverse
saturation current; but, this time residing between the collector-
base junction. It is called 100 to denote that it is the collector
t The word "short circuit" only refers to the ac value; de voltages and currents
are present.
reverse saturation current. At room temperature (30°C), 100 is

only a few microamperes, and so can be neglected with respect to
the much larger collector current. However, it is highly dependent
on the temperature-approximately doubling in value for every
I0°C rise in temperature. Where 100 is typically only 5 p.a. at room
temperature, it will have increased to about 650 µ.a at I00°C.
Since I 00 is often of a magnitude that cannot be ignored, the
expression relating the collector current with the emitter current
must be modified slightly to include this term. Of course, thermally-
generated minority carriers in the base region "look" to the col-
lector, just as injected minority carriers do. Hence, these two
currents add in the collector circuit. Equation (2) must now be
modified to read:
(13)
It is apparent that large heat-induced increase in 100 might
disrupt the circuit operation, although the common base is not as
sensitive to these changes in 100 as other configurations. The
I
I
2N35 N-P-N
COMMON EMITTER
30°c
10
i.---
\~
-
-- --~ 32...-- L---
a
E J.11.n
8
6v=:L---
V
~
i..----
----- - ~
i--
100
80
~
i--
-
4i.---
ble 60
v--
40
2
V
V 20
~
0
0 20 24
4 8 12 16
COLLECTOR•TO·EMITTER VOLTAGE (Vcil IN VOLTS
Fit,. 43. Static collector or output characteristics for the common emitter
configuration of a 2N35 n•p-n transistor.
common base static collector characteristics must be taken at a

given temperature, and this information must be presented along
with the graph. Transistor characteristic graphs are usually shown
for a temperature of 30°C.
20K
Fig. 44. Test circuit required
to obtain characteristics of
Fig. 43.
+ +
-=.Vu Vee _;;;
_-15V
24. Common Emitter Characteristics
Figure 43 shows the static collector or output characteristics of

the common emitter configuration. The curves were obtained in
the same manner as they were for the common base, using only
the common emitter test circuit of Fig. 44. In this instance,
collector current is plotted for variations in collector-t~mitter
voltage, with the base current as a parameter. Notice that all of
the base current curves merge into the same vertical line near
VoE = 0. This graph is similar to a pentode vacuum tube's static
plate characteristics.
It is immediately noticed from Fig. 43 that the slope of each
line is greater than the common base. This signifies that the
dynamic output resistance is less. With the common emitter, the
output resistance is approximately the resistance betwen the col-
lector and the emitter. It can be seen that along the 120 pa base
current line, the collector current rises 0.6 ma for a variation of
collector-to-emitter of 2 to 8 volts. Hence the output resistance is:
AVcE 7 volts
Ro=
Ale
= . ma
06
= ll,600ohms (14)
As with the common base configuration, the output resistance

is relatively constant with different values of emitter and collector
currents. (Later we will make use of this very important fact.)
The base current in a transistor is equal to the difference between
the collector and emitter currents. If this base current could be
varied, it would result in a change in the collector current. Since
the base current represents the difference between the collector
and emitter currents, a change in this difference current must

result in a change in either the collector or emitter current, or
both. But since the emitter current is the source of the collector
current, both the emitter and collector change. In the common base
configuration, this fact held little significance, since the emitter
and not the base current controlled the collector current. In the
common emitter, however, this is not the case. The signal is
applied between the base and emitter in such a way as to prcxluce
changes in the base current. Since the majority of transistor cir-
cuits in practice employ this common emitter configuration, a
further study of common emitter current gain is in order.
Neglecting 100 for the moment, the relationship between the
change in emitter current and the change in collector current in
the common base amplifier was found to be:
Ale= aAIE (15)
The base current is always equal to the difference between the
emitter and collector current. Hence, the change in base current
must be equal to the difference between the change in the emitter
and collector currents.
(16)
Substituting a a IE from Equation (15) for A le in Equation (16),
we have:
A IB = A IE - a A IE (17)
Alu = A IE (1 - a)
Solving this for A IE, we find that:
Alu
(18)
1-a
But Equation (15) still holds for any configuration. Hence, we
substitute (~ IB/1 - a) in Equation (18) for ~ IE in Equation
(15) :
Ale= aAIE = a ( ~ ) = ( - a - ) AIB (19)

I-a 1-a
Therefore, the change in collector current is equal to the change

in base current X the quantity (a/1 - a) . The term (a/1 - a) is
the current gain for the common emitter, corresponding to the a
found in the common base circuit. It is denoted by the symbols
a 0 B, aE and /3. The last symbol (pronounced beta) is by far the
THE TRAN~STOR AS A CIRCUIT ELEMENT 73
most popular term and will be used in this book. In terms of a,

the common base current gain, /3 is defined as:
a
/3 = -----,,----
1-a
(20)
The values of a and/or f3 are given by the manufacturer. With

the aid of Equation (20) a conversion of either value may be
made. As opposed to a, whose value is never greater than 1 in a
junction transistor, /3 may assume values ranging from IO to 100,
depending upon the transistor. As can be seen from Equation (20) ,
the closer that a approaches l, the greater /3 becomes. The 2N35
has an a of 0.98, and so /3 equals (0.98/1 - 0.98) = 49.
When a is near l, a very slight variation in a has a greatly
magnified effect on 13. Unfortunately, the state of the transistor
art has not progressed to the point where a may be held constant
during manufacture. One can expect a certain variability in a
between different transistors of the same type, even ones made
by the same manufacturer. This slight a variability gives rise to
much greater changes in /3, and so the common emitter character-
istic curves are subject to a rather wide variation in practice.
From the common base relationship A 10 = a A IE, we previously
obtained an expression for a:
Ale
a=
AIE
Similarly, for the common base:
(a)
hence:
/3 = (b) (21)
With the aid of Equation (21), /3 may be obtained directly

from the common emitter collector static characteristics of Fig. 43.
For a collector-to-emitter voltage of IO volts, a base current varia-
tion from 60 to 160 p.a produces a collector current variation from
3 to 7.5 ma. Hence f3 is found to be:
A le 4.95 ma - 1.2 ma 3.25 X I0- 3 amperes

47 22
/3 = A 1B = 100 p.a - 20 p.a = 80 X I0- 6 amperes = ( )
This value agrees rather nicely with the calculated value of 49

for the 2N35 using Equation (20).
The affect of 100 upon the collector current is much greater

than for the common base. Its effect may be appreciated from the
following consideration: l 00 for any configuration Hows across the
collector-base junction. The electron current resulting from 100
must enter or leave the base terminal, depending on whether the
device is a p-n-p or an n-p-n transistor. In either case, the base
current is varied by this reverse saturation current. This change
in the base current is tantamount to a minute signal and, there-
fore, gives rise to a collector current of le = Ieo (/3) = Ieo
[a/1 -a] in accordance with part (a) of Equation (21). However,
the reverse saturation current Hows not only in the base but also in
the collector. Hence, the total collector current due to 100 is the
sum of two terms: the magnified 100 of the base plus the 100
actually flowing in the collector.
~a)
le= leo+leo ( 1
le=I (1- l~a)

00
lo= loo ( !=: - l~a)
= Ieo ( I ~a ) =
0
/~ a (23)
We know that the total collector current is the sum of the

normal or quiescent collector current plus the collector reverse
saturation current. Adding equations (a) of Equations (21) and
(23) , we obtain an expression for the total collector current for
the common emitter.
(a)
1-a
le = IB ( I ~a ) +
lco
I-a
(b) (24)
Since 100 is multiplied by the factor (1/1 - a) when it reaches

the collector, the circuit is much more sensitive to temperature
variations than the common base, where l 00 is not magnified at
all when it reaches the collector. For instance, with a = 0.98 as
with the 2NJ5, the quantity (1/1 - a) is equal to 50. Thus the
effect on the collector current due to 100 is about fifty times as
great for the common emitter as for the common base.
The collector current for the curve of IB = 0 µ.a represents this
magnification of 100 in the collector circuit. Here, 100 is about
Fig. 45. A practical common-

emitter voltage amplifier em-
ploying the 2N35 transistor. --
OUTPUT
CIRCUIT
+
RL
3.9K t
OUTPUT
-=-
B
Vee
: - 22.5V
!
0.3 ma, whereas it was only a few microamperes in the common
base configuration of the same transistor. This magnification of
100 in the common emitter circuit gives rise to special problems
of circuit stability when operating at high temperatures. These
problems will be dealt with in Chapter 5.
25. Drawing the Load line
The static collector characteristics of the common emitter cir-

cuit of Fig. 43 remain the same regardless of the external circuit.
By itself, this graph only refers to the voltage directly across, and
the currents flowing through, the transistor terminals. With an
actual amplifier, we must combine this data with data regarding
the external circuit, so that the operation of the amplifier may be
graphically portrayed.
In Fig. 45, a more practical common emitter voltage amplifier
employing the 2N35 transistor is shown. The point A and B
separate the transistor proper from the external output circuit.
It is known from the common base characteristics how the internal
circuit of the transistor amplifier (viz., the transistor proper)
reacts to different voltages and currents. We wish to determine
how it will react when the external circuit is connected at points
A and B. This can be done graphically by determining the
appropriate current and voltage existing across the collector-base
terminals (A and B) when these terminals are shorted and open-
circuited. These conditions will give us two points of interest
on the collector characteristics graph. Between these two points
a line called a load line will be drawn. Many operational properties
of the amplifier may be explained with the aid of this load line.
To do this coherently and logically, so that we may gain as
much information as possible, the problem will be approached
in a simple mathematical manner. In Fig. 46, the same transistor
circuit is shown. For the moment, we are interested only in the
collector output circuit and will not concern ourselves with the
O-C BLOCKING, A-C

COUPLING CAPACITOR
Fig. 46. Circuit of Fig. 45,

with addition of d-c block-
INPUT ing coupling capacitor.
SIGNAL
base circuit. The collector current flows in the loop indicated

in the figure. It will cause a voltage drop across the collector-
emitter terminals and the load resistance. The sum of these two
voltage drops must add up to the collector bias supply voltage.
This is Kirchhoff's famous voltage law: the sum of all of the voltage
drops around any closed loop is zero. Taken in the direction of
positive current travel, and assuming that all voltage drops
external of the source are positive and the voltage rise in the
source is negative (negative drop), we proceed as follows around
the loop:
(rise in battery)
+ (drop across load resistance)
+ (drop across emitter-base terminals) =0 (25)
or - Vee+ VcE + V (across RL) =0 (26)

or, when the resistive voltage drops are expressed in terms of I X R:
- V00 + V CE+ IcRL = 0. (27)
le = (28)
Equation (27) is merely an expression equating all of the volt-

age drops in the collector loop to zero. Now, if we solve it for 10 ,
with the stipulation that VCE =
0, this will correspond to a short-
circuit condition where the current is limited purely by RL.
Thus:
(29)
(The vertical line in Equation (29) serves to separate I O from

the conditions under which it is being measured.)
This short-circuit collector current is located on the common
2N35 N·P-N
COMMON EMITTER
(30°C)
Cl
E
~8f---+----b...-=-+---l--------,=-~+-----1--
u
!:!
0 4 8 12 16 20 24
COLLECTOR-TO·EMITTER VOLTAGE (VcEllN VOLTS
Fig. 47. Common emitter characteristics.
emitter characteristics of Fig. 47 above the zero collector-to-emitter

voltage point.
If the transistor were open-circuited in such a way as to stop
current flowing from emitter to collector, then no current would
flow in the circuit. If the collector current is reduced to zero,
this corresponds to an open circuit of the transistor terminals.
It may be expressed mathematically by solving Equation (27)
for VoB•
(30)
and letting le = 0
VeE = Vee- (0) RL = Vee (31)
This point (VOE = V00) is also located on the base line of the
graph opposite the zero collector current point. Now connect a
line between the two points just located. This line is the load line.
Here we have two graphs, dealing with two entirely different
subjects, superimposed on one another. One is a graph of the
collector characteristics, which has nothing to do with the external
circuit. The other straight line graph is governed entirely by
the external circuit and would be the same irrespective of the
kind of transistor used. The combination of these two graphs
gives us the solution of our problem. We may now specify the
action of the entire amplifier. The important conclusion to be
gained from the above is that the operation of the amplifier
must always be on the load line. For any given value of base
current, only those combinations of V c111 and 10 that represent
points on the load line are attainable. In effect, this only means
that between the open and short circuit conditions on the load
line, there lie intermediate values of resistance which must con-
form to Ohm's Law and, therefore, must lie in a line between the
two end points. The graphical construction of a load line is a very
important operation in transistor work.
Since the operation of the transistor must always be on the
load line, the job of the circuit designer is to pick a quiescent
point of operation. The signal will make the collector current
vary above and below this operating point, hence it is usually
advantageous to place the quiescent point (Point Q in Fig. 47)
in the middle of the load line--or at least in the middle of the
base current curves which the load line crosses. In our example,
this point is taken where the 60 µ:a base current line crosses the
load line. Using the Q point and the load line, we can graphically
illustrate the operation of the amplifier.
The signal is applied to the base. The quiescent base current
is 60 µ:a. We will assume that the signal causes an increase in
base current of 40µ.a. Since we know that the transistor must
always operate along the load line, we know that the resultant
change in collector current from its quiescent value of 3.1 ma will
be found at the intersection of the load line with the new base
current value of 100 µ.a. This increase is 1.5 ma. The amplitude of
the voltage change across the transistor represents the output sig-
nal, and this is also read directly off the graph. (NOTE: Since the
Vcc remains constant, the change in V0111 must equal the change of
voltage across RL. This change in V 0111, representing the difference
between the quiescent V0111 and the new V0111 , is found on the
V 0111 axis directly below the intersection of the load line with the
100 µ:a base current line.
2N35 N-P-N
COMMON EMITTER
... u
;: 6 rl'--=--1---~,.i....--====--4----
z
ILi
;.~=~~~lt~~~:j~;:~:~~:::~~~==~~~i~~~~~~~~=~~-~
~2~~=il~~t~~i~i~=i~=t~~~-~-~
0
u
0 8 12 16 I 20 24
I I
COLLECTOR TO EMITTER VOLTAGE (Veil IN VOLTS
I I
I I
I I
Fig. 48. Common emitter characteristics with a-c signal applied.
It is clear that the transistor is acting as a variable resistor.

When the signal increases the base current, the collector-to-
emitter resistance decreases and V OB decreases. The opposite hap-
pens when the signal causes the base current to decrease. In Fig. 48,
the same amplifier is depicted with a sine wave signal applied to
the base. The 60 µ.a quiescent base current line is used as the zero
reference point for the input signal. This corresponds to a
quiescent output voltage (VaB) of 10.2 volts. The plus and
minus 20 µ.a base current variations due to the signal cause the
collector-to-emitter voltage to vary - 5.7 volts and + 7.0 volts,
respectively, from the quiescent value of 10.2 volts. Notice that
although the input signal was perfectly sinusoidal, the output
voltage is slightly distorted. There is some variation of the spacing
of the base current lines although it is somewhat exaggerated in
this example. If the output were taken directly across the collectm-
emitter terminals, there would always be the quiescent voltage
present. This is not desirable. It is the job of the amplifier to
amplify a weak varying signal, not to contribute a large de

component to it. For this reason, the ac component-the fluctua-
tions in V 0 ~are passed through the coupling capacitor of Fig.
46 and on to the next circuit. The capacitor blocks the de com-
ponent, and so is often called a blocking capacitor. Our primary
interest is not in the quiescent voltage or current, but only in the
signal variations.
The current amplification of this common emitter amplifier
may be determined graphically as the ratio of the total swing of
the collector current over the total swing of the base current.
Ai =~ = 4.60 ma - 1.35 ma
~ IB 100 µ.a - 20 µ.a
= 3.35 X I0- 8 amperes = 42 (32)

80 X I0- 6 amperes
Here Ai refers to the operational current gain; i.e., the current gain
with the external circuit connected. {:J refers to the short circuit
current gain. f3 for the 2N35 common emitter is 47, as previously
measured. Comparing A 1 with [3, confirms the interesting fact
that the greatest current gain achievable by an amplifier is the
short-circuit gain. In an actual amplifier, the short-circuit current
gain is only approached, not reached.
Normally, if the voltage gain of the amplifier in Fig. 46 were
to be computed, we would take the ratio of the change of the
voltage across the load resistance to the change in the signal
voltage. Remember that the collector characteristic curves (Fig.
43) relate base current (and not base-to-emitter voltage) to other
transistor voltages and currents. A mere knowledge of the change
in base-to-emitter voltage (~ VBB) would be sufficient to calculate
the change in collector current and the base-to-emitter voltage.
lationship between the base current and the base-to-emitter voltage.
Unfortunately, the emitter-base (unlike the collector-base) junc-
tion resistance varies substantially with large changes of voltage
across the junction. Conseqently, voltage and current do not have
a linear relationship. Since in our example of Fig. 46, the signal
source is connected directly to the base, we can expect that any
signal voltage change will be reflected across the base-emitter
junction. If the signal is sinusoidal, the base-to-emitter voltage
must also change sinusoidally, BUT THE BASE CURRENT WILL NOT
CHANGE SINUSOIDALLY. Because this is a non-linear V-1 relation-
ship, it will undoubtedly be somewhat distorted, the exact amount
being dependent upon the signal amplitude. Since the usual· func-
tion of the device is to amplify a signal voltage faithfully, without

adding any components of its own, this distortion may hinder its
usefullness as an amplifier.
With the aid of a graph of the input characteristics relating
VBB to I B (not shown) , we can set the bias level to produce the
correct quiescent base current of 60 µa. This can be done because
the quiescent bias is a static condition and regardless of the junc-
r--,MEGOHM 7
I ~
Fig. 49. A constant current
source.
I J.!-
:r
-:-IV
i
L ____ _J
CURRENT SOURCE
:
I
0 J LOAD
RESISTANCE
tion resistance, the correct operating point can always be obtained,

using the load line method described in connection with the col-
lector circuit. However, the signal current passing through the
base must be sinusoidal. This cannot be done if the signal voltage
is used as the controlling medium in Fig. 46. Irrespective of the
operating point selected, the resistance of the junction will vary
when the signal causes the voltage across the junction to change
above or below the quiescent condition. This change in emitter
junction resistance causes a change in the slope of the static input
characteri&tics (Fig. 47) . Since we have stated that a transistor
is a current-operated device, it would seem desirable to make the
sinusoidal signal voltage produce sinusoidal current variations
across the emitter-base junction. To do this, we need to make the
signal a current, rather than a voltage, source.
Everyone is familiar with voltage generators. A perfect voltage
generator is a voltage source that never changes its voltage with a
varying load. In practice, large generators or batteries approach
this state. A current generator, however, is a device that delivers
the same current to the load without regard to the load resistance.
This device is not usually encountered in electronics. A current
generator may be simulated by placing a voltage source in series
with a very large resistance. as shown in Fig. 49. In Fig. 49, the
voltage source has a potential of l volt and the series resistance
is 1 megohm. If the external terminals were shorted-correspond-
ing to a load of 0 ohms--the current flowing through the short
circuit would be I =E/R =1/1,000,000 =
I p.a. Now, if the short
circuit were replaced by a 1-ohm resistor, the current would be
1/1,000,001 (approximately 0.999999 µ;i). This is still 1 µ;i, for

all practical purposes. Even if the load resistance were raised to
100 ohms, the current would still be 1/1,000,100 =0.9999 µ.a. A
practical current source is created from a voltage generator in
series with a high resistance.
If this current source represents the signal source, the input
current (the base current) will be independent of the resistance
of the emitter-base junction. This is shown in Fig. 50, where
the signal has been coupled to the base through a capacitor and
a high resistance, R,r The capacitor in this case only serves the
Fig. 50. A common emitter

amplifler.
purpose of keeping the bias voltage out of the signal source. It

is considered a short circuit to the signal voltage alternating com-
ponent. R,, makes the base signal current independent of the
emitter-base junction resistance. The bias current is also fed
through a high resistance RB, RB serves the same purpose to the
bias voltage source, as R,, serves to the signal voltage source-
namely, to make the quiescent bias current independent of the
emitter-base junction resistance. Note that the signal is fed to the
emitter-base junction resistance and RB in parallel. This in no
way alters the previous discussion. Since RB is so much larger
than the emitter-base resistance, it may be neglected insofar as the
signal is concerned.
It is not to be construed from the above that R,, is always a
physical resistance. R,, usually represents the effective output
resistance of the previous stage. In Chapter 3, it was mentioned
that the output resistance or impedance of the common emitter
and common base transistor configurations was much larger than
their input resistances. Insofar as distortion is concerned, this is
a most desirable situation, for it means that the output of each
amplification stage will act as a current source for the subsequent
stage. Where the input signal is extremely small (such as from a
dynamic microphone, etc.), the signal excursions to either side of
THE TRANSISTOR AS A CIRCUIT aEMENT 83
the Q point are reduced to such an extent that the variations in

input resistance are insignificant. Hence, the necessity of a high
input resistance of the signal source is not as stringent, and one
transistor may be transformer-matched to another to achieve
maximum power transfer. Chapter 5 deals with this type of small
signal amplifier.
The common emitter amplifier of Fig. 50 is a more realistic circuit
than the simplified circuits previously illustrated. Actually, there
is a further improvement that seems rather obvious. The base and
collector bias batteries have a common connection with the same
polarity. Hence, there is no reason to be content with two bias
fitt. 51. A practical version

of • common emitter am-
plifier.
batteries. The circuit is readily modified to that of Fig. 51. Cal-

culation of the voltage gain of this amplifier is now much easier.
The voltage gain may be practically stated as the ratio of the
change in voltage across RLJ produced by a change in the signal
voltage. It is realized that with the large series resistance R 11, the
voltage gain will decrease; but, with the choice of a lower voltage
gain, or excessive distortion, the former is the more desirable.
26. Establishing the Proper Emitter Blas
In the common emitter circuit of Fig. 52 (a redrawn version of

Fig. 51), the quiescent base current is independent of the quiescent
collector current. The method of base biasing in Fig. 52 is, there-
fore, called fixed bias. The collector load line is the same as in
the circuit of Fig. 46 (as shown in the graph of Fig. 48) because R 1,
and V 00 are the same. The problem at hand is to select the proper
bias resistor, RB, to yield the required 60 µ.a.
Applying Kirchhoff's voltage law to the base loop, we proceed
by equating the voltage drops around this loop to zero, as shown
in Fig. 52.
(SS)
Fig. 52. A common emitter

circuit.
or
(34)
The previous discussion involving a current source, stated that the

quiescent base current will be independent of the base-emitter
voltage drop, providing RB is large. With this in mind, we can
logically neglect VBE in the above equation. (In all future cal-
culations, this will be done.) Considering that V BB is only a few
millivolts, as compared with Vcc, 22.5 volts, little error is made.
- V cc+ IBRB =0 (35)
RB= Vee (36)

IB
In this example, the quiescent IB has been selected as 60 µ,a.
Hence:
22.5 volts
3.75 X I0+ 5 ohms
60 X I0- 8 amps
375,000 ohms (37)
This establishes the quiescent point for the fixed bias common
emitter amplifier.
Summary of Procedure for Determining the Quiescent Point for
the Common Emitter Fixed-Bias Amplifier
I. For a given load resistance, write the collector
voltage loop equation. Solve for V CB with I c =0;
solve for 10 with V cB = 0. This yields the open
and short-circuit voltage and current respectively.
Locate these points on the collector characteristics.
Draw the load line between them.
2. Select the operating points as approximately the

middle of the region of equally spaced base current
lines. This Q point establishes the base current
needed.
3. To select the required base resistor, RB, to yield
this base current, write Kirchhoff's voltage law
equation around the base loop, neglecting the
base-emitter junction voltage drop. Solve for RB.
27. Establishing the Q Point for the Common Base Amplifier
The procedure for establishing the Q point for the common

base circuit is approximately the same as for the common emitter
amplifier. The general method evolved for the common emitter
will be applied to the 2N35 common base amplifier of Fig. 53.
Cc 2N35
Fig. 53. 2N35 common base

amplifier circuit.
'- ,JI
, ...,:,r =+ Vee
--: 22.5V
Assuming that the same value of load resistance (3900 ohms)

will be used, we determine the equation for the load line by
writing Kirchhoff's voltage law equation around the collector
loop, as indicated by the dotted line. It may be argued that the
collector current does not actually travel from the collector to
the base and back through V00 and, hence, Kirchhoff's law does
not apply. This is not true! A knowledge of the actual path of
collector current is not necessary to write these equations, as
long as a loop is formed where current could flow, if conditions
were right. Kirchhoff's voltage law states only that the sum of the
voltage drops around this closed loop equals zero.
In Fig. 53, the voltage drops around the collector loop are:
(38)
Solving this for V 0B with the condition 10 =0, we determine the
open circuit voltage between the collector-base terminals. Thus:
(39)
Now solving Equation (38) for I 0 , with VoB =

0, establishes the
collector current with the collector-base tenninals shorted.
Voe
Io= - - (40)
RL
Note that the open and short circuit equations are identical for
the common base and common emitter circuits. In practice, for
either circuit, it is not necessary to write down the complete
10
2N35 N·P-N
9 COMMON BASE
•-8ma.
8
i -7
!: 7
~ 6
I-
z
a:: 5
l&I
a::
::,
u 4
~
u 3
l&I
:I
0
2
u
0 2 14 6 8 WI~ ~ ~ ~ 20 22 24
COLLECTOR-TO-BASE
I _ _ _ __, VOLTAGE IN VOLTS I
I I
,...
I OUTPUT YOLTAGE--......,!OI
I
I
I
fi9. 54. Common base collector characteristics.
voltage loop expressions-just apply Equations (39) and (40).

These points are located on the appropriate collector characteristic
graph as indicated in Fig. 54. Since the values of V00 and RL
are the same in this and the previous example, the points are
the same, resulting in identical load lines.
Once the collector load line has been drawn, the quiescent
point may be chosen. We will choose the Q point at Ia 3 ma. =
This emitter current may be established in a manner similar to
that previously used for the base current of the common emitter
amplifier. With VEB = 22.5 volts, we again write the expression
for the voltage drops around the emitter loop as shown by the
solid line in Fig. 53. Again, we assume that the emitter current
flows from the base to emitter, through the emitter bias battery,
and back to the base again. This reasoning is the same as that
for the path for collector current. As long as we account for all
of the voltage drops around the closed emitter-base loop, our
assumptions are justified. The voltage loop equation is:
- VEE+ VEB +IERE =0 (41)
Applying the same method to the common emitter circuit, we
neglect the few millivolts of drop across the emitter-base junction
and the equation becomes:
-VEE+IERE =0 (42)
or
- -
RE - VEE
- (43)
IE
Substituting the value for the quiescent ls and V1111 of the circuit,
Rs becomes:
22.5
R:,: 3 X 10- 8
= 7.5 X 108 = 7500ohms (44)
This establishes the quiescent point for the common base amplifier.
28. Establishing the Q Point for the Common Collector Amplifier
As with the vacuum tube cathode follower, establishing the

quiescent point for the common collector (emitter follower), is
rather involved, compared to the other two configurations. The
emitter follower has an emitter resistor that is common to both
the input and output circuits. Hence, the collector and base
circuits are now dependent upon one another; in the other con-
figurations, they were independent. With the common base and
common emitter circuits, the input bias current could be specified
without regard to the collector circuit-as evidenced by Equations
(43) and (46). Now, however, because of this common resistor,
the collector current will be present in the analogous equation
for the common collector circuit. Consequently, there are two
equations and two unknowns which must be solved simultaneously,
before the Q point may be determined. A rather straightforward
graphical method has been devised.
In Fig. 55 (A) , a typical emitter follower circuit is shown.

This circuit has the primary function of impedance matching, as
does the cathode follower. Figure 55 (B) shows a somewhat more
practical version of the same circuit, utilizing a common bias
battery. The collector load line may be obtained in the normal
Fig. 55. Common emitter

follower circuits: (A) typical;
(Al (B) practica I.
.
2N35
I
I
I +
= Vee
-
I
J
(Bl
way, by wntmg the expression for the voltage drops around the
collector loop. The path of the collector current for this equation
is shown by the solid line. Using the same rules developed for the
other two configurations, we may write:
(45)
Here, we are presented with the fact that both the base current
and the collector current flow through the same load resistance.
Each current contributes to the voltage drop across this resistance,
as evidenced by the l 0 R1, and InR1, terms in Equation (45). This
might cause trouble were it not for the fact that the magnitude
of the base and collector currents are so different. Whereas the
base current is in microamperes, the collector current is in milli-
amperes. For all practical purposes, then, this minute base current
can be neglected with respect to the much larger collector current.
The base current can little add or detract from the voltage drop
already established by the much larger collector current. Con-
sequently, as an engineering approximation, we are justified in
neglecting the 18 RL term in Equation (45). Hence:
- Vcc+ IcRL + VCE =0 (46)
This is the same equation obtained for the collector load line
in the common base and common emitter configuration. Skipping
the intermediate steps and solving directly for the short circuit
current and open circuit voltage, we have:
= Vcc = 22.5 volts (47)

le= 0
_ Vee 22.5volts
le - Rr, 5.77 ma (48)
3900ohms
VcE =0
The two points are located on the common emitter collector
characteristics of Fig. 56, and the load line is drawn in the usual
manner. (The emitter follower is but a modified common emitter
amplifier. Hence, the common emitter characteristics are used
for both configurations.)
The next step is to write down the voltage drops around the
base loop (the path of the base current is indicated by the dotted
line in Fig. 55 (B) ) .
- Vee+ IBRL + VBE + IBRB + IcRL =0 (49)
Again, it is seen that both the collector and base currents flow
through the load resistance, RL. Now, however, we are very much
interested in the small base current, and so its presence cannot be
neglected with respect to the collector current. Solving the equa-
tion for 18 and neglecting (as usual) the minute V BE• we obtain:
- Vee+ IB (RL + RB) + IcRL =0 (50)
(51)
In this last expression, we know Vcc, RL, and R 8 because they

were given. I 8 and / c are the only unknowns in the expression.
Picking a value for / c will yield a corresponding value for / 8 •
However, this is the function of the load line equation. If this is
2N3!1 N·P·N
COMMON EMITTER
I a1----i1----i1-:::::,,~=----i~--+~.-,,::c.+---+---+----1
~
4 8 12 16 20 24
COLLECTOR TO EMITTER VOLTAGE (Vc1 l IN VOLTS
fie. 56. Common emitter collector char• cteri1tlca.
true, then no new information is gained by writing the base loop

expression. It will be presently shown that Equation (51) is but
a re-arranged version of Equation (45). To do this, we will have
to perform some mathematical gymnastics.
First, V cs is always the algebraic sum of V CB + V BB,
or
(52)
When (as usual) we neglect the minute base-to-emitter voltage

drop, Equation (52) becomes:
VoE = Vos (55)
Second, write the voltage drop equation around the loop in

Fig. 55 (B) containing R 8 and defined by the line containing X's.
- Vos+ lsRB =0 (54)

or
(55)
But, we now know that VCB = Vca from Equation (53) . Hence,
V08 may be substituted for VCB in Equation (55) , yielding:
VOE = IeRe (56)
This is a very important substitution and will be referred to
again. Returning to Equation (45) and substituting IBRB for VoB,
- Vco+ lcRL + leRL + IeR8 =0 (57)
which, when solved for I B, is identical to the load line Equation
(51). Hence, no new information has been ~ined by writing the
base loop expression in the form of Equation (51).
This problem may be eliminated by substituting in Equation
(51) for 10 , its value as defined by the load line Equation (46).
Therefore, Equation (51) now becomes:
In this expression, only I B and Vca are unknown. When simplified,

this expression becomes:
RL+RB
Vco-Vcc+VOE VcE
--~-~--
RL + Re
= -=-----==---
RL+Re
or
(59)
When values are substituted for our example, Re = 100,000 ohms
and RL = 3900 ohms, in this latest expression, we have:
VcE = le (103,900ohms) (60)
Equations (59) and (60) define what is called a bias line. The
equations indicate that V08 and /B are related in a certain way.
If values are picked for I B and the resulting Vca is evaluated in
each case, several points may be located on the collector character-
istics, and the bias line drawn between the points. The point
where the bias line intersects the load line, represents the simul-
taneous solution of Equations (46) and (51), and indicates the
quiescent point of operation. In the example at hand, three points
will he selected by choosing three convenient values of IB (values
that base current lines have been drawn for) . The results are:
VOE IB
(in volts) (inµ;,.)
4.18 40
6.19 60 (61)
10.30 100
The bias line resulting from these points intersects the load line
at IB ~ 78 µ;,., 10 ~ 3.75 ma, and V 08 ~ 7.9 volts. This is the
quiescent point of operation for this circuit. The steps in obtaining
the Q point may he summarized as follows:
I. Write the collector voltage loop equation and draw
the load line equation in the usual way, using
Equations (47) and (48).
2. Write the base voltage loop equation and solve
for IB. Substitute in this equation the value for
I O obtained from the load line equation. This
always reduces to Equation (59) .
3. Insert circuit constants in Equation (59).
4. Select several convenient base currents and solve
for the corresponding V aE• Plot these points so
that the bias line crosses the load line. The inter-
section of the load and bias lines defines the
Q point.
29. Finding RB
The previous discussion of the common collector circuit covers

the procedure for finding the Q point, when the circuit con-
stants (RL, Rn, and V0 0 ) are given. The practical problem, from
a design standpoint, is how to find the correct RB once the load
line has been drawn. (Factors governing the selection of RL and
V 00 will shortly be discussed.)
Once the load line is drawn, to find the appropriate RB, it
must first he decided where the Q point is to he located. For
clarity, it will he assumed that the Q point has been chosen at
the same point as in the solution of the previous problem (illus-
trated in Fig. 56) . If we assume that this is to be the Q point,
we may also assume the values of IB and V 08 corresponding to
that point on the collector characteristics. In this instance they
are 78 µ,a and 7.9 volts, respectively. With this information, Equa-
tion (59) may be solved directly for RB, since RB is the only
unknown in the equation.
VcE = IB (RL+RB)
VcE
IB
VcE
RB - - - - RL (62)
IB
7.9volts
RB = --==---
78 p,a
- 3.9 X 1()3 ohms
RB = 101 X 1()3 - 3.9 X 1()3 = 97,100 ohms (63)

This is within 3% of the value originally selected for RB, and is
excellent for a graphical solution. In fact, with a required RB
of 97,100 ohms, a 100,000 ohm-resistor would probably be used
as the closest standard value available.
This method of selecting RB from a specified Q point is general
and may even be used when there is a separate base bias battery,
as shown in Fig. 55 (A) . Of course, in this case, V BB would be
present in Equation (51), instead of V00 • Hence, it will not
cancel with the V 00 term introduced in Equation (58) , and the
simplified expression of Equation (59) will contain this V BB
term. Other than this, the procedure is exactly the same as out-
lined for the single bias battery case. (See problem 7 at the end
of this chapter.)
30. Distortion and Regions of Operation
The object of a transistor amplifier is to convert variations of

the input signal current into variations of the output collector
current. The input and output current in any successful amplifier
must be, as nearly as possible, exact replicas of one another-
other than the fact that they may not be of the same amplitude.
In the common emitter circuit, the output current is larger
than the input current. In the common base circuit, they are
approximately equal. In either case, the essential job of the ampli-
fier is to transfer current from a low resistant input circuit to a
high resistant output circuit and, in the process, not contribute
any signal of its own.
The ability of the collector current to faithfully reproduce the
shape of the input signal current (for any configuration) depends
upon the extent that the base or emitter current lines (depending
.e
!
.!I
...z
l&J
a:
a:
::,
u
a:
...u
0
l&J
..J
..J
0
u
Flt, 57, Saturation region of a common base amplifier.
upon the type of amplifier be. .

Thi~ is a very important con:J co~s1dered). are evenly spaced.
realized in practice. erauon and is never completely
31. Active Region
Examination
current of Figs~es48 an_d ~4 shows that the base current and
emitter lines
is. the region in whidi ti{e~::~ nar~ almost evenly. spaced. This
with the above discussion) is thne is drawn and (m accordance
amplification with the least d' e_ only region where successful
the active region Th. . h 1stor~1on can take place. It is caJled
. . is is t e region that h be
previous discussions of th as en used for all
a mpl'fi
I ers. The active re io
e common
. base an d common emitter
e~itter is forward biase3 a:l\~haracterized_ by the fact that the
this region the collector and e c_ollector is reversed biased. In
Equations (13) and ( ) N emitter current are related by
24 . ormally, the quiescent point will be
taken as right in the middle of this region as shown in Fig. 57.

From the figure, it can be seen that, as the emitter current increases
in increments of 1 ma from the Q point, the collector increases
in approximate increments of 1 ma. As long as this happens, the
amplifier is operating in the active region and distortion is at a
minimum. Theoretically, then, fJ or a (depending on the con-
figuration) should remain constant throughout the active region.
32. Saturation Region
If the emitter current in the common base amplifier of the

above discussion increases beyond a point of about 5.9 ma, the
emitter current lines tend to crowd together in the saturation
region of Fig. 57. Examination of the figure shows that this crowd-
ing takes place to the left of V08 = 0. Here, the collector voltage
is of the opposite polarity and produces a positive instead of a
negative bias. This region in which both the emitter and collector
are forward biased is called the saturation region. It is indicated
on the graph to the left of the V cB = 0 point. Assuming that the
Q point remains at the point indicated on the graph, this satura-
tion condition will exist for part of the input signal cycle, if the
input signal current is large enough to cause the emitter current
to rise above a critical value. The fact that V cB will reverse and
cause a forward collector bias even when V co remains unchanged
in magnitude and polarity becomes apparent from examination
of the collector voltage loop equation (Equation 38) . (This equa•
tion is for the common base amplifier of Fig. 53, for which the
load line of Fig. 57 has been drawn.)
(64)
Solving this for Ven:
Ven = Voo-loRL (65)
From Equation (65), it is noted that V 00 and 10 have an opposite
effect on VcB• Now Vco and RL of the relation are always constant,
but if 10 increases sufficiently, the right hand side of the equation
will become zero. This is the point that separates the saturation
region from the active region. If the value of 10 increases still
further, the effective value of V cB becomes negative, and the
transistor goes into the saturation region. The region of negative
V cB is shown in Fig. 57.
Solving Equation (65) for l c with the condition that V CB = 0,
we find the critical collector current where saturation begins:
Io\ (66)
!Sat
As long as the collector current exceeds this critical value, the
transistor will remain in saturation. Of course, the collector cur-
rent does not vary spontaneously by itself, but is controlled by
the emitter current. The critical value of emitter current at which
saturation begins may be estimated from Equation (65) by sub-
stituting alE [from Equation (15) ] 1 for 10 , hence:
(67)
Setting Ven 0 as the point where saturation starts:
(68)
or
(69)
When the transistor enters the saturation region, bottoming

is said to have taken place, since the emitter current lines abruptly
drop to the bottom of the characteristics in this region. The most
obvious evidence that the transistor has bottomed, or entered
the saturation region, is that large changes in emitter current
produce almost no change in collector current. In Fig. 21, for
example, a change of emitter current from - 6 to - 8 ma (along
the load line) only produces a change of about 0.3 ma collector
current.
If a signal is applied to the transistor that drives it to satura-
tion, that portion of the input signal that exceeds the critical
saturation current will be clipped and will not appear in the out-
put signal. This is demonstrated in the graph of Fig. 22 by an
input signal that causes the emitter current to exceed the critical
value [as established by Equation (69) ]. Here R =
3900 ohms,
Vca =
22.5 volts, and a =
0.98, for the 2N35 common base ampli-
fier of Fig. 53. Applying Equation (69) , we find that analytically,
the critical emitter current is:
22.5
IE (Sat) = (0.98) (3900 ohms)
5.9ma (70)
1 Ico is considered negligible and for the common base, 10 ""' aIE.
From Fig. 57, it can be seen that saturation, or bottoming, begins

at slightly less than 6 ma, about 5.9 ma. At this point the 5.9 ma
emitter current line (if one were shown) crosses both the load
line and where V OB =
0. As can be seen, the entire part of the
signal that exceeds this critical emitter current-namely that from
I 1
2N35 N-P-N
COMMON BASE OUTPUT
(COLLECTOR) CHARACTERISTICS
Is•-8ma
8
I
.~
0
E -7
~
u6
...
Vee
.,, CJRL
r, I
-6
I I
t--
zUJ
II:
~ 4
' , , ...
......
...
I
\
INPUT\ SIGNAL /- 4 _
-5
1 0
E
I'
\, "'
(.)
II: I I / -'3 in
0
t--
(.)
Ill/I I
I ............. I I
UJ 2 -2 --
I
.J I .......... I
.J
0
(.)
I '-... \ J -I
-
I ......... I
I I N_t•O
0
-2111/fi
,,,,, IO
I
2 4 I
8 10 12 14 16 1a
1
20 22 24
I
I
----;::::±;-;:::-:::----~I
OUTPUT SIGNAL
I
I
Fig. 58. Common base collector characteristics with applied a-c signal driving
transistor to saturation.
approximately 6 to 7 ma-is clipped off and is not present in

the output current. The clipped voltage swing across the load
resistance is shown on a vertical axis at the bottom of the graph.
If the transistor is to be used as a linear amplifier stage, extreme
care must be exercised to insure that the input signal does not
rise to a value such as to drive the transistor to saturation, or
into the nonlinear region, where the output is not an amplified
version of the input.
33. Cutoff Region
The cutoff region of a transistor is defined as the region where

both the collector and emitter are in reverse bias. It is shown in
Fig. 57. The existence of this region is obvious when it is
remembered that only when the emitter is in forward bias can
minority carriers be injected into the base material. With no
I 1
I I 2N35N-P-N
COMMON BASE
OUTPUT (COLLECTOR) CHARACTERISTICS
I I
l1 •-8ma
8
I -7
;;
~
-6
6
...
z i"--- ""-
I
-5
----
' , ... \
I
a 4
I
I
.... I
-4
~
u
I I -3
LIi
I . . r-----. Q POINT / INPUTISIGNAL
~
-2
[J
I
2
I~
\'ClI
u I
I I r--..... -I
I I .... ~ • 0
I I
0
20 22T
2 I
I
4 6 8 10 12
~ 16 18
24
I I
k::-
I
OUTPUT SIGNAL
.I I
I
I I
~
I I
I
I
L I
fltl. 59. Common baa collector cher11cterl1tic1-ditferent quiescent point.
minority carriers being injected, the only collector current is the

reverse saturation current. If the input signal current causes the
emitter voltage to go to zero, or into reverse bias, that portion of
the signal present in the cutoff region will be clipped from the
output signal, as it is in the llaturation region.
In Fig. 59, an input signal is applied to the same common base
amplifier, except that this time the stage is operating at a lower
Q point (referring to the load line starting at 22.5 volts). The

part of the signal that drives the emitter into cutoff is clipped
from the output signal.
34. Interrelation of Q Point, Vee, and Regions of Operation

From the previous discussion, it should be evident that there
are several reasons why the Q point is usually established in the
middle of the active region. In both Figs. 58 and 59, the input
signal is of the same peak-to-peak (p·p) amplitude, 5.9 ma. In
the first example, the Q point was established at IB = 4.0 ma,
and the top part of the input signal is seen to drive the transistor
into saturation. In the second example, the bottom part of the
same 5.9 ma peak-to-peak signal causes the transistor to enter
the cutoff region when the Q point is established at IB = 2.0 ma.
Obviously then, picking a Q point too near either the cutoff or
saturation end of the load line will result in a clipped and dis-
torted output signal. Of course, the input signal may be kept
quite small, thus allowing a greater margin in the selection of
the Q point, without danger of clipping.
If in Fig. 58, the Q point had been selected at a point slightly
lower than 4.0 ma, the 5.9 ma peak-to-peak signal would just fit
in the active region-neither driving the circuit to cutoff nor
saturation. This is the extreme, or limiting case, where the peak-
to-peak emitter signal just equals the value Vcc/a RL. To avoid
clipping of either the negative or positive peaks, the operating
point must be located at a point of IE = Vcc/2aRL or, exactly
in the middle of the active region. As has been mentioned, it is
not a good practice to use an input signal this large, as temperature
variations, battery drain, etc., will cause the Q point to shift
somewhat. In this limiting case, any drift of Q point will cause
clipping, due to saturation or cutoff.
Then, it is advantageous to use as large a V00 as possible, if
large signals are to be handled without distortion. This so-called
dynamic range of the amplifier is a direct function of the value
of V 0 0 . To illustrate the concept, the same load line has been
drawn for three values of V 00 in Fig. 60. Notice that for the left-
hand load line (corresponding to V 00 = 2 volts) , the input signal
of 1 ma peak-to-peak nearly approaches saturation and cutoff.
This is because a small V 00 was chosen for the value of load line
resistance. For the middle load line with a V 0 0 of 8 volts, the
amplifier will handle a peak-to-peak input signal of 4 ma without
overloading (i.e., reaching saturation or cutoff). For the load line
using V00 = I 4 volts, an input peak-to-peak signal of 7 ma can
be applied. In each case, the load line corresponds to an RL of

2000 ohms. Thus, although the input signal handling capacity,
the dynamic range, of the amplifier has been increased by increas-
ing the value of V cc, the amplification has not increased. In each
case, the ratio of input current to output voltage is 2.
In Fig. 61, the same amplifier is operated with a constant V00,
but with three different load line resistances. As can be seen,
as the resistance of the load line increases, the amplification
increases, and the dynamic range decreases. As the resistance of
the load line decreases, the amplification decreases, and the dynamic
range increases. Thus, although the slope of the load line governs
the amplification, it also affects the dynamic range.
The larger the V 0 0 used, the larger the input signal that can
be handled without distortion. There is, however, a limit on the
tolerable value of V cc• If it is too large and the transistor happens
to go into cutoff momentarily, then V0B = V 0 0 [from Equation
I I I COMMON
2N35 N-P-N
BASE
I I
OUTPUT (COLLECTOR) CHARACTERISTICS I
le=-8ma
8
0
E -7
~
~6 V'
~ I -6
~L•2K
'
1--
z
l&I ~
-5
a:
a:
B4 '\ -4
a:
~
,1 '\
"'-RL•2K/ \~ -- - -- ,... --- ,r- - -3
lil I
"'
-2
::l 2
8 I~~•2K '\. \ \ I -I
0
~,''(f\, ~,~ J I"
-I • 0 ' 2 4 6 I 8 10 12 14 16
\. )
18 20 22
IE•10
24
~ j I I I
k:::::::
rt:,
1
I I
II II
I
I
I
k4
I OUTPUT SIGNALS
I
I
I
I I I
L ~
Fig. 60. Th& dynamic rang& of a common base amplifier.

(38) ]. For this reason, V 00 is not usually chosen greater than the
maximum tolerable reverse voltage across the collector-base junc-
tion, unless provisions in the signal amplitude and/or bias cir-
cuitry are made, to prevent the stage from ever entering the
cutoff region. If this should occur, the ensuing Zener breakdown
of the collector junction will possibly ruin the transistor. Of
course, if a very large RL is used, the current will not damage
the device.
Up to this point in our study of transistor circuits, we have
not considered the effect of the load into which the signal is fed.
We have only considered signal voltage variations existing acro6s
the collector load resistance. The existence of a load connected
to RL through the coupling capacitor or transformer often has
the effect of reducing the effective load presented to the transistor.
Note that an excessive increase in the magnitude of RL is not the
answer to the problem of securing greater amplification.
l 2N35 N-P-N I
COMMON BASE OUTPUT
I
(COLLECTOR) CHARACTERISTICS
It•-Bmo
I
~ -7
I\_,. RL • 1.5K I
-6
~ 6 I
IA.I \ -5
0:
0: " -- I
:)
(.)
0: 4
1' ~ ,....,,,
C- INPUT -4
"'-: il-RL•3KI l I
0
IK
I rQ[jl
I- -3
(.)
' ~ 11
JI
IA.I
I
.J
.J
8 2
"NJ !-- ~- -7\ 8-INPUT -2
,1 ~"~ 11.-~
I
Rc•l2K
-# l•Iii "-. -I
r1 I " ' ~ ~ A-INPUT 7
! 11 I i ; I ~-o
0
-I
:,,,, I
I 2 4--.J 61 I B 10 12 14 16 18 20 22 24
Fig. 61. Three different load resistances applied to the common base amplifier
of Fig. 60.
1. Why is the reverse saturation current more of a problem with

the common emitter configuration than with the common
base configuration?
2. How is it possible to observe the difference in the output
resistance between the common base and common emitter
amplifiers by examination of their static collector character-
istics?
3. What restrictions does the load line place on the operation
of an amplifier?
4. What is the advantage of having a high output resistance of
a signal source connected to the input of a transistor
amplifier?
5. Explain the difference between V BE, V be, vBE, and vbe·
6. For what purpose is the bias line used?
7. With reference to Fig. 55 (A), assume that the Q point is to
be established at 18 = 40 ma. Calculate the required RB
if V88 = 5 volts, Vee = 22.5 volts and RL = 3900 ohms.
Draw the load and bias lines on one of the common emitter
output characteristics.
8. For the common base amplifier of Fig. 53, RL = 3300 ohms.
=
The Q point is to be at IE 4 ma.
(a) Find the proper RE, assuming VEE and Vcc are IO and
20 volts, respectively. (b) What peak amplitude of a sinu-
soidal input current is allowed before driving the device into
saturation? (c) Cutoff? (d) Which is reached first as the
amplitude of the input signal is slowly increased from zero?
9. Why is it often advisable to limit Vcc to a value not greater
than V cB1 Under what conditions is it permissible?
IO. If it was known that the input signal would always cause
operation in the active region, could Vcc be safely made
greater than VcB?
Chapter 5
SMALL SIGNAL ANALYSIS
In Chapter 4, the graphical characteristics of a transistor were

described. It was shown that as the input signal varied the bias
point, the oollector current varied, and amplification was achieved.
This method of analysis is usually used where large signals--
causing large shifts from the quiescent point-are involved, such
as in a power amplifier. When the input signal is very small, it
is difficult, if not impossible, to graphically measure the shift of
the Q point.
A new method of analysis is presented in this chapter. It makes
use of a technique of substituting an equivalent electrical circuit
in place of the transistor. The equivalent circuit, once developed,
can be used to simulate the transistor in nearly all aspects of per·
formance. This equivalent transistor circuit is developed with the
assumption in mind that the characteristics of the transistor (a
input and output impedance, etc.) do not change as the input
signal moves the Q point back and forth on the load line. The
assumption is accurate to a good approximation, when the input
signal is small. Hence, the name small signal analysis. (Unless
otherwise stated, only small signals will be dealt with here.)
36. Static and Dynamic Quantities
Consider the static input characteristics of a common emitter

junction transistor, shown in Fig. 62. The input resistance of the
103
transistor's emitter-base junction may be measured at any point

on the curve, by merely observing the ratio of emitter-to-base
voltage to the base current. This is called the static or de input
resistance. It exists only at the point of measurement. For a dif-
ferent base current, a different static input resistance is observed.
Resistance on any graph possessing voltage and current coordi-
nates is the slope of the curve, or line. The slope is equal to the
ratio between the vertical and horizontal travel of the line. Thus,
in Fig. 62, the slope of the dotted line extending from zero to
point A is R = 0.3 volts/190 µ.a = 1500 ohms. This is approxi-
mately the de resistance (as measured by the volt-ammeter method)
across the emitter-base junction. The slope, or tilt of the dotted
line would become greater or less, as the junction resistance cor-
respondingly became greater or less.
Assume that in Fig. 62, a given quiescent base current has been
established at point A. What is the resistance to a small increase
in base current? If the input signal swings the point on the curve
from A to B, the increase in emitter-base voltage is only 0.01 volt
for an increase in base current of 35 µ.a. The resistance for this
increase is only 285 ohms. Clearly, this is different from the de
resistance of 1500 ohms. It is this 285-ohm dynamic resistance
that will be presented to an ac signal applied to the junction.
VI
~> 0.4
.,.
~
~
Lt.I 0.3
(!)
<
!::i
g
a:
Lt.I
I-
!::
:IE
Lt.I
6 0.1
I-
I
Lt.I
VI
<
CD /
/
0
50 100 150 200 300 350
BASE CURRENT (I 8 ) IN ,.a250
Fig. 62. Typical transistor static input characteristics.

SMALL SIGNAL ANALYSIS 105
Now establishing the fact that the dynamic resistance changes

for different values of quiescent base current, how is the dynamic
resistance measured accurately, if it is continually changing? The
problem is solved practically by drawing a small right triangle
under the curve, as indicated in the illustration. Only when the
hypotenuse of that triangle (always a straight line) lies on the
'kl
-
ID
>
II.I
(!)
ct II.I
~.J
oct
>0
II)
0::0 0.232
........
1-o
!::z 0.225
::Ef
'tlx
~II.I
I
II.I
II)
ct
m
65 70 75 80 85 90
BASE CURRENT (I 1 ) IN ~a EXPANDED SCALE
Flt• 63. Expanded portion of Fig. 62.
curve are we able to state the dynamic resistance. This required

that the triangle be exceedingly small, so that the curve "looks"
like a straight line at that point. It can be seen that this is not
strictly true for the triangles drawn. In Fig. 63, the small shaded
section has been expanded. The portion of the total curve repre-
sented in this shaded portion is so small that it is nearly straight.
The hypotenuse of the right triangle drawn under this portion of
the curve lies on the curve nearly perfectly, and the dynamic
resistance can be stated with some confidence. Too, if the input
signal does not drive the operating point outside the shaded
section, it can be considered that the input resistance is constant
over this section, for the slope is constant. What has been said
about the input resistance remaining constant for small changes
in input signal applies to the other parameters of a transistor.
Under the condition that a transistor's characteristics remain
constant during a small signal swing about the operating point,
these characteristics may be measured and a simple electrical
circuit, an equivalent circuit may be constructed, possessing the
same constant characteristics that were measured. This equivalent

transistor circuit may then be substituted for the transistor in the
external circuit. Once this has been done, an analytical, rather
than graphical analysis of the system is possible. Small signal
,-- - -------7 r----,

I I
I ci ci ci I I I
Ig g ~ ~ I I 200 I
1-,___._~,___._____, I I I
L ________ _J L ___ _J
CIRCUIT EQUIVALENT CIRCUIT

(Al
EQUIVALENT CIRCUITS
(Bl
Fig. 64. Equivalent circuits: (A) simple example; (B) complex example.
analysis of a transistor circuit is quicker and more direct, and

brings to light many facets of circuit performance not evident
from a graphical approach.
37. Equivalent Circults-''Black Boxes"
An equivalent circuit is a circuit that electrically represents the

performance of some device. About the simplest example of an
equivalent circuit is illustrated in Fig. 64 (A) . Here, a group of
five 100-ohm resistors are connected in parallel. The total resist-
ance measured would be 20 ohms. Thus, an equivalent electrical
circuit for these resistors would be one 20-ohm resistor. A dotted
SMAU SIGNAL ANALYSIS 107
line has been drawn around the resistors in each case to denote
that, to the world outside THE BOXES ARE EQUAL. Indeed, if one
excludes the difference in power dissipation between the boxes,
it is impossible to tell from an external measurement how many
resistors are present. Only a presence of 20 ohms resistance is
sensed. Thus, from only an external measurement of what is in
the box, one may reproduce the electrical equivalence. If a
capacitor or inductor or resistance, or any combination or number
of these were in the box, they could only present one impedance
to the output terminals. Once this terminal impedance has been
determined by an external measurement, then a single resistor in
series, or parallel with, the proper reactance could replace a
very complicated circuit. This is symbolized in Fig. 64 (B) . When
there are sources of voltage or current involved, several electrical
theorems must be employed to develop an equivalent circuit.
38. Circuit Theorems
Two fundamental theorems of modern circuit theory are

Thevenin's and Norton's theorems. The use of Thevenin's theorem
may be illustrated by Fig. 65 (A) . Here, a dry cell has been con-
nected to a 9-ohm external resistance. The cell has an open circuit
voltage of IO.O volts and an external resistance of l ohm. Thevenin's
theorem says that if the imperfect battery is represented by a
perfect voltage source1 of IO volts in series with the battery's
internal resistance, then viewed from the terminals, an exact
equivalence exists. The example may be so trivial that the sig-
nificance of the theorem may be overlooked. What is important
is that at no time is it necessary to open the battery to determine
its characteristics. All that is necessary is to measure the resistance
from the external terminals, along with the open-circuit terminal
voltage.
In Fig. 65 (B) , the same battery is connected to a somewhat
more complicated circuit. What is desired is to replace the battery
and the two resistors with a single voltage source and a single
resistor. To begin to simplify this circuit, we could represent the
1 A perfect or constant voltage source is one that maintains a constant voltage

across its terminals regardless of the current flowing-even under short circuit
conditions, where an infinite current flows. This theoretical representation is
only realized in practice with a battery of infinite capacitance, but it is neces-
sary to use with Thevenin's theorem. A voltage source is considered to have
no resistance, only voltage.
r---------
1
II
I 9.0.
I
I
I
IL_________ _JI
CIRCUIT
(Al
r---------7
I
I
I 20.9.n
I
I
I
I _________ _j
L
CO~PLETE
THEVENIN'$
EQUIVALENT
CIRCUIT
(Bl
Fis,. 65. Thevenin's equivalent of active circuits: (A) simple; (B) complex.
battery by its Thevenin equivalent. This is done inside the circle,

as indicated. Next, we measure the open-circuit voltage existing
at the output terminals of the network. Since with open circuit
conditions at the output, no current will flow through the 20-ohm
resistor, the open-circuit voltage exists across the 9-ohm resistor. The
battery current is then: I = E (R 1 +R2) 10 (1+9) = I ampere. The
voltage drop across the 9-ohm resistor is then: E IR =
IX9 = =
9 volts. Therefore, 9 volts also exists at the output terminals. This
is the voltage source for the Thevenin's equivalent circuit. The
external resistance of the network (the resistance presented to
the output terminals) may be measured in practice, by a number
of methods. It may be calculated by noting that the 20-ohm
resistor is in series with the parallel combination of I and 9 ohms.

The parallel combination reduces to 0.9 ohm; thus, the total out-
put resistance is 0.9 + 20 = 20.9 ohms. Then, the Thevenin's
equivalent may be represented as a perfect voltage source of 9
volts in series with a resistance of 20.9 ohms. It is impossible to
tell from an external measurement that the two circuits are
different. This is why the box made by a dotted line is often called
a black box. It denotes that we are not interested in whether the
equivalent circuit approximates the original circuit layout, but
only whether it reacts on the output terminals in the same way.
Norton's Theorem may also be used to constrtict an equivalent
circuit, for the same networks as illustrated for Thevenin's
Theorem. If the internal resistance (as measured across the out-
put terminals) is connected in parallel with a perfect, or constant,
current source2 equal to the current flowing across the shorted
output terminals of the network, an exact equivalent circuit is
formed. In Fig. 66 (A), ~he same battery used for Thevenin's
Theorem is shown. The current source shown in the equivalent
circuit equals the current flowing when the 9-ohm resistor is
shorted. The arrow denotes the direction of current flow. It should
be in the direction to produce the same polarity voltage across the
terminals as with the original circuit. 1h the second example, the
short circuit (Fig. 66 (B) ) current may be calculated from the
original circuit (Fig. 55 (B) ) , but assuming that Thevenin's
Theorem is correct, we may short the terminals of the Thevenin's
equivalent circuit (Fig. 55 (B)) and obtain: I = 9 volts/20.9
ohms= 0.43 ampere. In both the Norton and Thevenin equivalent
circuits, the open-circuit voltage at the terminals is 9 volts. If
any external load is connected to these terminals, the resulting
voltage-current flow, power dissipation, etc.-will be the same
for both equivalent circuits, as well as the original circuit.
To sum up the construction for the Thevenin's and Norton's
equivalent circuits, assume that an unknown active network within
a black box is given. Assume that the open circuit terminal voltage
and short circuit current are measurable, and denoted as V0 and I.
2 A perfect current source is a device that has a constant current flowing through
it at all times, regardless of the external circuit conditions. If the load is open-
circuited across a current source, an infinite voltage exists. This theoretical
representation is only partially realized in practice with a high voltage source
in series with a very large resistance. (See Chapter 4 for a more detailed dis-
cussion.) A constant current source is considered to have an infinite resistance,
and a voltage determined by the I x R of the external circuit.
----,I
I
1n I gn
I
CIRCUIT NORTON'S EQUIVALENT CIRCUIT
r
I
----- 7
20il
I
I 9il
I
I
L _____ _J
CIRCUIT NORTON'S EQUIVALENT CIRCUIT
Fig. 66. Norton's equivalent of active circuits: (A) simple; (8) complex.
respectively. Then, the Thevenin's equivalent is a voltage source

equal to V0 volts in series with a resistance equal to V 0 /I,. The
Norton's equivalent is a current source equal to 18 in parallel with
a resistance equal to V0 I,. This relationship is shown in Fig. 67.
Kirchhoff's voltage law is again needed: The algebraic sum of
the voltage drops around any closed loop is zero. Figure 68 gives
some examples of circuits and their related voltage law equations.
These theorems are absolutely essential in representing any
active electronic circuit, for in many cases, the internal complexity
of the device is so great (i.e., as in a transistor) that specifying
a precise circuit is impossible. What is possible, however, is to
construct an electronic equivalent circuit that will duplicate the
device's action at its output terminals.
A transistor has two input and two output terminals. It may be
represented by a black box having an input and output set of
terminals. This is the standard representation of a two terminal-
pair network, and is shown in Fig. 69. All two terminal-pair net-
works may be analyzed, knowing only 4 quantities-the voltage
across the input and output terminals, and the currents flowing
in or out of these terminals. The complexity of the internal cir-
ACTIVE
NETWORK
CIRCUIT THEVENIN'S NORTON'S

EQUIVALENT EQUIVALENT
Fig. 67. Norton's vs Thevenin's equivalent of an active circuit.
cuitry of the box is only evident when the voltage and current
relationships at its terminals are known. Once these relationships
are known, it is possible to construct an equivalent circuit that
will duplicate these terminal volt-ampere relationships.
IA
+,,_-_- -_,,,n,"---
R1
Fig. 68. Applications of

Kirchhoff's Law.
VA•IAR 1 +IAR 2 V1 •V5 +I 1R 1

v.-r.<R1+R2l
CONVENTION: TREAT VOLTAGE RISES (FROM-TO+)
AS NEGATIVE,AND VOLTAGE FALLS
(+TO-) AS POSITIVE.
It is known that a transistor has a current amplification, and

an input and output impedance. It is also known that putting a
voltage across the output terminals of a transistor will cause a
voltage to exist across the input terminals. This latter fact was
demonstrated in the introduction to the transistor, where it was
shown that a transistor is not strictly a one-way device-there
exists some reverse current amplification. In fact, bilateral, or sym-
Fig. 69. 2 terminal-pair net•

work.
--+
--
-
V1 BLACK BOX
ACTIVE NETWORK
--.:-_
V2
---
+
metrical, transistors are available where the current gain is the

same in both directions. This is mentioned to point out that
(unlike vacuum tubes) the input terminals are not completely
isolated from the output terminals. This reverse current ampli-
fication will create a voltage drop across the input terminals
when a voltage is placed across the output terminals. This effect
of output voltage on input voltage is called the reverse voltage
amplification, or just voltage feedback ratio.
39. Hybrid Parameters
All of the above parameters manifest themselves in the voltage-

current relationship of the transistor, when represented as a two
terminal-pair network. Many mathematical equations may be
written expressing this relationship. Since any equations written
should be capable of laboratory verification, only a few have
reached any degree of popularity, it being rather difficult to per-
form certain tests on a transistor, such as open-circuiting the
output terminals. A set of equations that has become universally
accepted combines the best features of both open-and short-circuit
tests of a transistor's input and output terminals. These equations
are called hybrid equations, for they are formed by combining
the best feaetures of two other sets of equations. The defining
hybrid equations are:
V1 = h11i1 + h12V2 (71)
i2 = h21i1 + h22V2 (72)
Equations (71) and (72) are written in terms of the input and
output terminal voltage-current relationships. The unknown quan-
tities, the h's are called hybrid parameters.
Solving Equation (71) for hw and letting v, = 0:
hu = Tv Iv2 = o = acoutput
the input resistance with the
terminals shorted for
(v, = 0).
(73)
Solving Equation (71) for hu with i 1 = 0:

= ratio of the output voltage (74)
that is fed back to the input,
with the input terminals open
circuited (i1 =
0)
Solving Equation (72) for hu and letting v 2 = 0:

= current gain, a, with output (75)
terminals shorted for ac
(v, = 0)
Solving Equation (72) for hu and letting i 1 = 0:
= output conductance with in- (76)
put terminals open circuited
for ac (i 1 0)=
These four hybrid parameters define the input and output im-
pedance, and the current or voltage gain (under short- or open-
circuit conditions) in both directions. It is important to reiterate
that these parameters refer only to small signal ac values, as
evidenced by the use of small letters. Also the terms "open and
shorted" apply only to the ac values, and it is assumed that the de
conditions necessary to establish the correct bias conditions are
present.
It has become standard practice to use the terminology:
11 = i = input
12 = r reverse transfer
21 =f forward transfer
22 = o = output
In addition, to differentiate between the three configurations:
b = common base
e = common emitter
c = common collector
Adding these two subscripts to each h parameter, instead of the
numbers, the h parameters for the three configurations may be
written as shown in Table 5-1.
Small ac, or instantaneous, signals are denoted by small case
letters. They are equivalent to small changes in the quiescent
value, which are denoted by the symbol ~- Hence the h parameters
for, say, the common emitter configuration may be defined as:
hle- - -vbe
lb
.- I V.,.,
=
=0
~ VBE
~ IB
I
VCE = Constant
= input
dance
impe-
(77)
h re -~1
- Vee ic =0 IB = Constant
= voltage feed-
back ratio or
reverse open
circuit voltage
gain (78)
= Forward small
Constant signal short-
circuit current
gain, /3 (79)
a Io I = Output con-
a VCE IB = Constant ductance (80)
The h parameters for the other two configurations may be defined
in a similar way.
TABLE 2
h PARAMETER CONFIGURATIONS
=
h 1b h 11 =
input resistance, common base
Common hrb = h 12 =
reverse voltage gain, common base
Base hfb = h 21 = forward current gain, common base
{ h,,b= h 22 =
output conductance, common base
h = h 11
10 = input resist,mce, common emitter
Common hre = h12 = reverse voltage gain, common emitter
Emitter hte = h 21 = forward current gain, common emitter
{
h,,. = h 22 = output conductance, common emitter
hie = h = input resistance, common collector

11
Common hrc = h = reverse voltage gain, common collector
12
Collector
{
hrc = h = fiorward current gain, common collector
21
h 00 = h 22 = output conductance, common collector
40. Graphical Determination of Hybrid Parameters
Defining the h parameters in terms of small changes of average

value of voltages or currents, as in Equations (77) to (80) , sug-
gests that a graphical measurement of these quantities is possible-
for the graphical static input and output characteristics should
contain every combination of tests performable on the transistor,
when treated as a two-terminal-pair network..
Figure 70 is a graph of the static input characteristics (V8a vs 18 ).

The procedure for obtaining the input h parameters (he. and hre)
from this graph is essentially that of measuring the quantities in
Equations (77) and (78), under the restriction that the appropri-
ate quantity, or parameter, is held constant. In Fig. 70, the small-
signal input resistance of the transistor is (as before) merely the
slope of the V8 .-I8 curve. As was mentioned, the slope, or tilt,
of this line (V ca =
7 volts) is a function of the dependent on
0.5
C/1
'0::i ,......
I
I
I 1 •C
VCE"l4.0
; 0.4 I
I
! I
I
:
111
a:
l&I
I-
t-
0.3
Vc1:_~--
i
l&I
~ 0.1
il,
~
m
0 50 100 150 200 250 300 350
BASE CURRENT U1l IN ,-a
l'ifl, 70. Typical static Input characteristics, for measurement of he. and hre• of
junction transistor.
the base current. Consequently, it is necessary to establish a de

operating point, before the slope is measured. For this example,
a point of 150 µ.a base current has been picked. As before, the
slope is measured by drawing a small triangle under the VcB line,
thus measuring the vertical and horizontal travel of the line. If
the VcB line is rather curved at the point where the triangle is
drawn, a small triangle is necessary, for it is assumed that the
hypotenuse of this triangle-formed by the V CB line-is straight.
Thus, the error in this approximation is less as the triangle
becomes smaller. The compromise between drawing a small tri-
angle and the ability to measure it is within the student's discretion.
AVBE IVcE = 7 volts = (0.28-0.24) volts

ll IB (108-105) µ.a
0.04
75 X I0- 6
= 530ohms (81)
Notice that the input resistance h,e becomes greater (greater

slope), as the base current decreases. Of course, under small-signal
conditions, where the variation of base current during the signal
swing is minute, this variation is not a problem. Only with this in
mind, can any figure be given for the input impedance.
hre, the reverse voltage amplification with the input terminals
open circuited for ac, is measured on the same input characteristic
graph. At the same value of VoB (or as close as is possible), the
ratio of a change in V oB to an accompanying change in V BB is
measured at a constant ls of 150 µ.a, corresponding to the same
Q point. Here, the small change in V oB is measured between the
14.0 and 7.0 V0B lines on most input characteristic graphs vary
widely in their spacing, tending to crowd together for high values
of collector to emitter voltages. Therefore, adjacent V oB lines,
or lines as close together without incurring measurement difficul-
ties, should be chosen.
h = ll VBEI IB = lOOma = (0.288 - 0.266) volts 0.022

-7-
re A VcE (14- 7) volts
::!c 3.15 X 10-3 (82)
Notice that hre is a dimensionless quantity. Due to the large varia-
tion of V OB line spacing, hre may vary considerably with the operat-
ing point. hre may vary in a typical transistor from 6 X I0- 2 to
1 X I0- 3•
To obtain the remaining two h parameters, we must consult
the static output characteristics for the same transistor (Fig. 71).
Carefully keeping the same operating point as in the graph of the
input characteristics, we perform the same measurements, but
with different parameters.
h fe __ Alo \VcE = 7.0volts (4.5-2.9) ma 1.6 X I0- 3

= ~5~0-x.......,..1-=-0-----=-
A IB = ( 150-100) µ.a 6
160 X 10+1
= 50 ::!, 32 (83)
Typical values for this small signal current gain (fl) are 20 to
over 100. It is important to stress that this is the small signal ac
current gain, as opposed to the de current gain. The two p's are
denoted as h10 and HFB, respectively.
I Ia•C
0 ~
E 8
~
I
I
I
200 ,.a
I
~ I
I
lz
Ill
6
II:
II:
:::>
u
~ 4
Ill
...J
...J
8 2
5 10 15 20
COLLECTOR-TO-EMITTER VOLTAGE (VcEl IN VOLTS
Fig. 71. Typical static output characteristics, for measurement of h/e and hoe•
of junction transistor.
h 0 e, the output conductance of the transistor with the input

terminals open circuited for ac, is measured on, the output char-
acteristic graph in the same way as the input resistance (h.,,) is
measured on the input characteristic graph. It is the slope of the
base current line. Keeping the same de operating point of 150 ma
of base current and a collector to emitter voltage of 7 volts, we
construct a small triangle under the base current line, to measure
the horizontal and vertical change. Thus:
h 06 =~1~ VcE IB = 150ma

(4.8 - 4.4) ma
(9.!> - 5.5) volts
= 0.4
4.0
X I0- 8 = l X I0- 4 mhos
= IOOµmhos (85)
This is actually a rather low output resistance for the common
emitter configuration. Typical values of output resistances vary
from 20,000 ohms to 100,000 ohms, corresponding to an output con-
ductance, h 0 e, of l 0-50 µmhos.
This graphical determination of the small signal h parameters
has been illustrated with the common emitter circuit. Since the
output curves of this configuration are more tilted, determination
of h 16 and h 06 is more accurate. h parameters obtained in this way

are sufficiently accurate for most design work. The method of
obtaining the common base h parameters from the input and
output characteristics is carried on in exactly the same way.
41. h Parameters for the Common Base

and Common Colledor Circuits
Although the h parameters may be obtained graphically, often

the input characteristics (sometimes the output characteristics
also) are not given on the manufacturer's data sheet for the
transistor. In such cases, the h parameters will often be given
along with the conditions under which they were measured (i.e.,
operating point, frequency, etc.). Today, with the advent of high-
frequency transistors, several sets of h parameters may be given,
one for ea.ch frequency. The low-frequency measurement is usually
made at I kc.
Manufacturers usually give the h parameter values for one con-
figuration, usually the common base or common emitter. If this
is not the configuration of interest, a conversion must be made
from the h parameter values of one configuration to ahother.
Table 4 gives the hybrid parameter conversion formulae from
one configuration to another.
h parameter values for a typical transistor in the three con-
figurations are shown in Table 3. Notice that in the table, the
forward short circuit current gain is negative for the common base
and common collector. This will be explained in the subsequent
section on h parameter equivalent circuits. Assuming that these
values were originally measured with the common emitter circuit,
the table may be checked with the aid of the conversion formulae
of Table 3. Calculating the common base h parameters with
the aid of this table, we have:
Given: h1e = 2000 ohms
h .. = 600 X 10-s
hie = 49
hoe= 20 µmhos
hie 2 X 103 2 X 103
h1b = I +h,e I +49 = 5 X 101 = 0.4 x I()2 ohms
= 40 ohms (85)
(2 X 103) (20 X }0-6)
-h,e = I +49
-600 X I0- 6
(400 X IO-•)
= 50
-6 X IO-•
= 8 X IO-• - 6 X I0- 4
= 2 X I0- 4
(86)
-49 -49
~= = 1+49 = 5 0 = - 0·98 (87)
20 X I0-6 20X I0- 6

I +49 = 50
= 0.4 X 10-6 m.hos
= 0.4 µ.mhos (88)
TABLE 4
h PARAMETER VALUES FOR A TYPICAL TRANSISTOR
IN THE THREE CONFIGURATIONS
Common Emitter Common Base Common Collector

h 1e = 2000 ohms h 1b =40ohms h 1c = 2000 ohms
hre = 6 X IO-• hrb =2 X IO-• h,e = I
hre = =
49 fJ htb =- 0.98 =- a hre =
-50
hoe = 30 µ.mhos h,,b =0.4 µ.mhos h..c = 30 µ.mhos
Manufacturers have had a field day with h parameter notations.

Although the notation convention used in this book is by far the
most popular, a knowledge of the other notations used is neces-
sary to quickly comprehend manufacturer's specifications. Table
5 equates the notations used in this book with others commonly
used.
42. Variation of h Parameters with Operating Conditions
It is important to reiterate that a set of given or measured h

parameters are only valid for one set of operating conditions.
The extent to which they vary with different operating condi-
tions, varies with the parameter and the actual transistor. Many
manufacturers give graph and tables to indicate the extent of h
parameter variation with operating conditions. Figure 72 is a
page from a typical manufacturer's data sheet. It shows the varia-
tions of h parameters for temperature, emitter current, collector
voltage, and junction temperature variations. Notice that the h
parameter variations are normalized to the values found at a
TABLE 3
HYBRID PARAMETER CONVERSION FORMULAS
To convert from:
CBtoCE CEtoCB
hib hie
hie= h1b =
I +hn, I +hte
hibhob -h,b hie hoe

h,e = h,b = -h,e
I +hte I +hte
-htb -hte
hte = htb =
I +htb 1 +hre
hob hoe
hoe= hob=
I +htb 1 +hre
CBtoCC CEtoCC
h1b
hie= hie = hie
1 +htb
1-hibhob
hrc = h,c = I - hre
I +htb
~ I ~ 1
-1
hrc = hrc = - (1 + hre)
1 +htb
hob hoc= hoe

hoc=
1 +htb
SMAU SIGNAL ANALYSIS 121
T A B L E 3 (continued)
HYBRID PARAMETER CONVERSION FORMULAS
To convert from:
CCtoCE CCtoCB
hre = 1-hre hrb = hre- -1
hre =- (1 + hre) hfb =
standard operating condition. The technique of normalization

involves dividing all the readings by the standard value (i.e., the
value at the standard operating conditions). To find the correct
set of h parameters for a given transistor and its operating con-
ditions, it is first necessary to find the published values for the
standard conditions. Then, consult the graph (if any) , to find
the percentage variation at the new operating conditions.
43. h Parameter Equivalent Circuits
Now that we have defined a set of h parameters conforming to

open- and short-circuit tests on the input and output terminals
of our black box, we must construct an equivalent circuit that
is capable of reproducing the results of these tests. This will be
accomplished if the circuit is designed to conform to the input and
output defining h parameter equations.
Earlier the common base circuit was usually used by manu-
facturers as a reference configuration, with all information given
in terms of it. Today, the trend is towards using the common
emitter citcuit since it is the most often used configuration.
To construct the C.E. (common emitter) equivalent circuit,
we start with the input defining h parameter equation.
TAB LE 5
HYBRID NOTATIONS IN CURRENT USAGE
= hbb = h11e = I/rue

Common
Emitter
r hre
hre
hbe = h12e = Ube = Ure
hbe = h21e = a cb
h..., = heh = h22e = l/r22e
= hee = h11b I/rm
Common
Base
r hrb =
hfb =
h.,b
hec
hce
= h12b = Uec = Urb
=~lb= ace
hcc = h22b = l/r22b
= hbb = hue= l/r11c
Common
Collector
r hrc hbe = h12c = Ube = Ure

hrc = heh= h21c = aeb
h.,c hce = h22c = l/r22c
Vi = h1ei1 + hreV2
Where v1 = vbe, v2 = Vee, and i1 = ib (89)
This equation relates input voltage to the input current and out-
put voltage. Certain conclusions may be drawn, when it is re-
written in more definitive form:
v1 (input signal voltage) = h 1e (ohms) X i 1 (input current)
+ hre (pure number) X v2 (output voltage) .
This equation must obviously conform to Kirchhoff's voltage law. 8
Since we know that v1 is the signal input voltage (a voltage rise) ,
the other side of the equation must be in a voltage fall. The first
term on the right side of this equation is the voltage drop caused
by i 1 flowing through the input resistance, h,e, The second term is
a voltage source which is dependent upon the output voltage.
From this, an input circuit may be drawn, as shown in Fig. 73 (A).
a See Chapter 4, for a detailed explanation of this law.

TYPICAL COMMON EMITTER CHARACTERISTICS

TYPE STIO
100,.. j
10
9
! I ;
600,.a
~ 8
: I I
I 500,.a I I
-~ 400,.a r!ia s7
6
150-C 25°C/
I
:,,.,VjLQ
i I
4
:
I
--1- 200,,..
I l9•100jL0
I011,a
OO I 2 3 4 5 6 7 8 9 10 II 12 13 14 15
I~
8 0
,,.
/
I
I/
I'
/
i
-ss•c
I
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
COLLECTOR-TO-EMITTER VOLTAGE, VCE (VOLTS) INPUT VOLTAGE, VBE (VOLTS)
SMALL SIGNAL CHARACTERISTICS VS. EMITTER SMALL SIGNAL CHARACTERISTICS VS. COLLECTOR
CURRENT Vc•6 VOLTS, TJ•25°C VOLTAGE IE•-lma.,TJ•25°C
10 10
5
- htb
'\.
h,;-~
' \. / -~ :::::~
--
2
hrb ' /_ h,b h,b
,_h;b
_h,,
0.5
=hft.,-
-hob
.......
' hf
h~
- hf•
-
h,;;c
hrb
hob-
0.2
I
0.1
I 0.1
-0.1 -0.2 -0.5 -I -2 -5 -10 I 2 5!02050
EMITTER CURRENT,ltlma) COLLECTOR VOLTAGE, Ve (VOLTS)
SMALL SIGNAL CHARACTERISTICS VS. TEMPERATURE COLLECTOR CUTOFF CURRENT VS. JUNCTION
Vc8 •6 VOLTS,1£•-lma TEMPERATURE Vc9"6 VOLTS, 1£•0
0 10 J} 10
/
i;; ~ 5
/
IE , z 2
5 ~ ~
i
I
h..:;- a o.s
./
/k ~
a2i -~~
0.2 /
;
u
2
/_ /
~
htb - h,b- >-- ~hlb- 1--h,, -
I 4 t;
~ .005
.02
.01
./
/
/
0.5
h,, I 0
.002
./
u .0010
-50 25 50 o IOO 150 25 50 75 100 125 150
JUNCTION TEMPERATURE, lj (°C) JUNCTION TEMPERATURE, lj (°Cl
Fig. 72. Typical data sheet.

r - - - . - -,
BASE I --11
I
1 - - - - --..!
i ~COLLECTOR
I ~ l2
I
+ I
hie I : ~ I fho,v2I
+
v, I I I Vz
I h11 i.1
r
h,.v0
I
I
I
I
I
I
ho, I
I i- L!_}
RL
+
EMITTER L - - - _J _ _J EMITTER
L ---
(A) (8)
(C)
t
Vz
t
L _____ - - - __ _J
(0)
t
Vz
L _ _ _ _ _ _ _ _ _ _J
(E)
Fl9. 73. (A) Common emitter input equivalent circuit; (8) Common emitter
equivalent circuit; (C) Complete common h parameter emitter equivalent
circuit; (0) Complete h parameter common base equivalent circuit; (E) Complete
h parameter common collector equivalent circuit.
The de bias circuitry has been omitted, as this is a small signal

circuit only. If Kirchhoff's voltage law equation is applied to
this input circuit, Equation (89) is evolved.
Notice that the feedback voltage source is of the same polarity
as the signal input voltage and thus opposes the flow of signal
current, i1 • This feedback voltage process is called negative feed,
back, since the output signal is fed back in opposition to the
input voltage. This can be demonstrated by rewriting the input
equation as:
(90)
The h parameter equation for the output is:
i2 = hteit + h,,.,v2 (91)
This may also be written in a more explanatory form:
i2 (output current) = hre (pure number) X i 1 (input current)
+ h 00 (output conductance) X v2 (output voltage)
The output equivalent circuit is shown in Fig. 73 (B). i, is the
output (load) current. The first term on the right hand side,
h1eiv is the amount of current generated in the current source.
It is dependent upon the input current and the forward short
circuit current gain. The second term, h0 ev,, is also a current
(from Ohm's law). Then Equation (91) conforms to Kirchhoff's
current law and says that i, splits at the node indicated, part
going down hoe and having a value of h 0 e,, and part going through
the current generator.
Putting the input and output equivalent circuits together, as
in Fig. 73 (C), we now have the complete common emitter equiva-
lent circuit. Notice that the output current generator "forces"
current up from the bottom to the top of the load resistance.
Then the load voltage (V nL) is opposite in phase to the input
voltage. This 180° phase shift is a property of the common emitter
circuit, just as with the common cathode tube circuit. The C.E.
is the only one of the three configurations that has this phase
reversing property. This phase reversal can also be deduced from
the fact that the actual load voltage is opposite in polarity to the
assumed output voltage (v,) polarity. The output current, i1 ,
which flows through the load, has the assumed direction. Con-
sequently, there is no current phase reversal between the input
and output.
One of the most useful aspects of the h parameter equivalent
circuit is that each configuration has the same equivalent circuit,
if the correct subscript is inserted for each parameter. In Figs.
73 (D) and (E) , the complete equivalent circuit is shown for the
common base and common collector circuits, respectively. The
signal source and load resistance are shown with each circuit. The
Thevenin's equivalent of the signal source is also shown.
~
A B
I
I
R
L
.
I
------, Rg I
- -1Ro
I I
Rg I I
I
I
L-------------~ I
I
I
IA 1B
Fig. 74. A generalized amplifier equivalent circuit.
It is evident that in all three equivalent circuits, the output

current generator is pointed in the same direction, even though
there is no voltage phase reversal with the common base and
common collector. To reconcile this inconsistency, the current
generators for the common 'base and common collector circuits
must be turned around. This is done quite effectively by defining
h1b and h1c as being negative. Therefore, although the current
source is drawn downward (to be consistent with the C.E. circuit),
the current really flows upward, around, and down through the
load resistance, in the opposite direction assumed for i 2 • Con-
sequently, the current undergoes a 180° phase shift, and there is
no voltage phase reversal.
The h parameters are defined and measured under short- or
open-circuit conditions, whereas in an operational amplifier stage,
the input and output terminals are never short-circuited. The h11
parameter, for instance, does not represent the operating input
resistance, R;. The same may be said for the other h parameters.
Consequently, when operating (in circuit) parameters are to be
determined, they must be calculated from the equivalent circuit
under operating conditions. Since, but for the proper subscript,
all three /, parameter equivalent circuits are identical, the formulae
for Ri (input resistance), R 0 (output resistance), Av (voltage

gain), A, (current gain), and A 11 (power gain) will be the same
for all three configurations. (The derivation of these formulae
by the simultaneous solution of several sets of equations will not
materially add to the present study of transistors and, therefore,
they are given here without proof.)
In Fig. 74, a generalized amplifier equivalent circuit is shown
The h parameter formulae to follow are in reference to this
generalized circuit. To obtain the specific formulas for the C.E.,
C.B., or C.C. configurations, insert the proper h parameter sub-
script in the formula, as indicated in Table 2.
44. Source Resistance
In the generalized amplifier of Fig. 74, the source output

resistance (Ru) is the resistance "seen" when "looking" to the left
of the line A-A. This will be the same whether the Thevenin's or
Norton's equivalent of the signal source is shown.
45. Input Resistance
The input resistance of the amplifier (R 1) is the resistance

"seen" by the source "looking" to the right of the line A-A with
the amplifier load connected.
R1 = ~ (92)
11
Where ~ = h 11 h22 - h 1 2h21
46. Output Resistance
The output resistance (R 0 ) of the amplifier stage is defined as

the resistance seen by the load looking to the left of the line B-B
with the signal source connected.
(93)
47. Voltage Gain
The voltage gain (Av) is the ratio of the output voltage to

the input voltage, when the transistor is in the circuit with the
source ancl load connected.
A = ~ = - h21RL (94)
v v1 h 11 +t:..RL
NOTE: A negative A., denotes a voltage phase reversal.
(A)
Fig. 75. (A) Common emitter
circuit; (B) Common base cir-
cuit; (C) Common collector
circuit.
+
(Bl
(Cl
48. Current Gain
The current gain (A;) is the ratio of the output (load) current
to the input current with the source and load connected.
A1 =~= +h21 (95)

i1 1 + i½:iRL
NOTE: A negative A, denotes a current phase reversal.
49. Power Gain
The power gain (A,,) is the ratio of the power delivered to

load, to the power delivered to the input of the amplifier. Since
the input power, P, = v 1 i11 ·the output power, P 0 = v,i,. Then:
A, ,-- ~ -- V2i2
. -- ~ X ~
. -- IA I IAI IV (96)
P1 V1l1 V1 11
The vertical bars denote that A, and A,, should be treated as +

regardless of signs, so that A,, is always positive. A,, may be
obtained directly by:
(97)
As an example of how these formulae are applied, let us cal-

culate R,, R 0 , A,,, A,, and A,, for a one-stage amplifier, in each of
the three configurations. We will use the transistor whose h
parameters are listed in Table 4. Again, bias voltages, etc. are
not considered here, as it is assumed that they are correct. For
the common emitter circuit of Fig. 75 (A) ., we have:
h 1p= 2000 ohms
h,e = 6 X 10- 4
hre = 49
hoe = 3 X 10- 11
The manufacturer recommends Rs = 500 ohms, RL = 20,000
ohms. Then:
a =h1ehoe - hreh,e
= (2 X 103) (3 X 10- (6 X 10-
11) - 4) (49)
= 6 X 10- 2 - 2.94 X 10- 2
= 3.06 X J0- 2
2 X 103 + (3.06 X 10- 2) (2 X 104)

= 1 + (30 X 10-6) (2 X J04)
20 X 102 + 6.12 X 102

= l +60 X I0- 2
26.12 X 102
1.6
= 16.3 X 102
= 1630ohms (98)
Ro = -=-h_l~•~+_R---=g'----
hoe Rg + A
2Xl03 +5Xl02
(30 X 10- 6) (5 X 102) + (3.06 X }0-2)
2.5 X 103
1.5 X 10- 2 + 3.06 X I0- 2
2.5 X 103
4.56 X 10- 2
= 0.55 X 10 5
= 55,000 ohms (99)
(From Equation (98), h 1e + A RL = 26.12 X 102 )

- (49) (2 X 104)
26.12 X 102
- 3.75 X 102 =- 375 (100)
(from Equation (98) , 1 + h 0 • Rr, = 1.6)

+49
~ - = +30.5 (101)
AP = jAvj jA1j = (375) (30.5) = 11,400 (102)
Using the same procedure for the common base circuit of Fig.
75 (B) , the manufacturer recommends Rg 100 ohms, RL =
60,000 =
ohms. The h parameters for this configuration from Table 4 are:
h 1b = 40ohms
h,b = 2 X 10- 4
hfb = - 0.98
hob = 0.4 µ.mhos
A = h1bhob - hfbhrb
= (40) (0.4 X 10- (-0.98)

6) - (2 X 10- 4)
= 16 X 10- + 1.96 X 10-4
6
= 0.16 X 10- 4 + 1.96 X 10-4

= 2.12 X 10- 4
h1b+.1RL
Then: R1 = 1 + hob RL
40 + (2.12 X 10- 4) (6 X 104)
=
1 + (0.4 X 10- 6 ) (6 X 104)
40 + 12.72 52.72
= 1 + 2.4 X 10-2
1 + 0.024
52.72
= 1.024
= 51 ohms (103)
h1b+ Rg
Ro=
hobR,+A
40+ 100
(0.4 X 10- 6 ) (1 X 102) + (2.12 X 10- 4)
140
0.4 X 10- 4 + 2.12 X 10- 4
140
= = 57.5 X 104
2.42 X 10- 4
= 575,000 ohms (104)
[from Equation (103), h 1b + A RL = 52.72]

- (- 0.98) (6 X 104)
- 52.72
+ 5.9 X 104 = 1.14 X 103 = 1,140 (105)
5.272 X 101
hfb
[from Equation (103), 1 + hob RL = 1.024]

-0.98
= 1.024 = - 0·95
(106)
AP = !Av! IA1I = (1140) (0.95) = 1080 (107)

For the CC circuit of Fig. 75 (C), using recommended values of
Rg = 20,000 ohms and RL = 500 ohms, we again calculate the
operating parameters using the C.C. h parameters from Table 4.
hie = 2000 ohms
hrc =1
htc = -50
hoc = 30 p.IDhos
A. = h1choc - hrch,c
= (2 X 103) (30 X I0- 6) - (- 50) (I)
= 60 X 10- 3 + 50
= 0.06 + 50 = 50.06 :::::: 50.1
h1c+ A. RL
I+ l\,c RL
+
2 X 108 (50.1) (5 X 102)
l + (30 X 10- 6 ) (5 X 102)
2X 1()-'J + 250.5 X 102
1 + 150 X 10- 4
2 X 103 + 25.05 X 103
1 + 0.015
25.25 X 103
:::::: 25.25Kohm (108)
1.015
Ro = ----=-h_l--=c+,---R~'- 2 X 103 + 20 X 103

hoc R, + A. (30 X 10- 6) (20 X 103) + 50.1
22 X 103 22 X 103
600 X 10- 3 + 50.1 0.6 + 50.1
22 X l03
= 50.7
= 0.435 X 103 = 435 ohm (109)
[from Equation (108), h 1c + A. RL = 25.25 X 103)

- (- 50) (500) + (5 X 101) (5 X 102)
25.25 X 103 25.25 X103
25 X 1()3
25.25 X 1()3 = 0.98 (110)
A1 = +hrc
1 +hocRL
[from Equation (108), I+ hoc RL = 1.015]
-50
1.015 = -49.2 (111)
Ap = IAvl IA,I = (0.98) (49.2)

= 48.2 (112)
The results of these calculations are given in Table 6. From
this table, certain generalized comments may be made with regard
to the three configurations.
The common emitter circuit has a high (-) voltage and a
high (+) current gain. It is the only configuration with a voltage
and current gain greater than unity. It has the highest power
gain of the three circuits. It is the only configuration that can
achieve a power gain in a multistage r-c coupled amplifier. For
these reasons, together with the fact tha.t only one bias battery
is required, it is the most popular configuration.
The common base circuit has a (+) voltage and (-) current
gain. It has the highest voltage gain of the three circuits, while
having a current gain less than unity. The common base circuit
has the lowest input and highest output resistance of the three
configurations. Hence, it is often used to match a low to high
impedance.
The common collector circuit (emitter follower circuit) has
a (+) voltage and (-) current gain. It has the highest current
gain of the three circuits, while having a voltage gain less than
unity. It has the highest input and lowest output resistance of
the three configurations. Therefore, it is often used to match a
high to low impedance.
50. Isolation Between Input and Output
Examination of Equations (92) and (93) shows that the input

resistance of the transistor is affected by the value of the load
resistance, and the output resistance is affected by the value of the
source resistance. These phenomena have been mentioned previ-
ously, but now we have a quantitative expression dealing with it.
It is very important, for unlike a vacuum tube, the input and
output are not isolated. The load must be specified before the
TABLE 6
TABULATION OF RESULTS OF EQUATIONS (98) TO (112)
Item Common Emitter Common Base Common Collector

R, l630ohms 51 ohms 25,250ohms
Ro 55,000ohms 575,000ohms 435ohms
Ao -375 + 1,140 +0.98
A, +30 -0.95 -49.2
A, ll,400 1080 48.2
TABLE 7
INPUT AND OUTPUT RESISTANCE OF TYPICAL TRANSISTOR

WHEN SOURCE AND LOAD RESISTANCE
EQUAL ZERO AND INFINITY OHMS.
Quantity& R, R, Ro Ro
Condition RL =0 RL = 00 Rg =0 Rg = 00
Reduced A hu I
Formula
= hu -~ --- -~-
A
Common
2000ohms IOOOohms 65,000ohms 33,000ohms
Emitter
Common
40ohms 550ohms 190,000 ohms 2.5megs
Base
Common
2000ohms l.6megs 40ohms 33,000ohms
Collector
input resistance is known, and the source resistance must be

specified before the output resistance is known. Equations (92)
and (93) have been solved for the condition that the source and
load resistance is both zero and infinity. The results are given in
Table 7. This table may be used as a quick rule-of-thumb

guide for evaluating the input and output resistance under the
conditions mentioned. It is noted that the common collector
configuration has the least isolation between input and output.
I. Define the essential differences between large and small signal

analysis.
2. Why is it that hu is + f3 but - a?
3. Verify the common collector h parameters listed in Table
4, using the formulae of Table 3.
4. Which transistor configuration would you use for the greatest
voltage gain, the greatest current gain, the greatest power
gain? Why?
5. Why is the common emitter configuration the most popular?
Is it always the best circuit to use? Under what conditions
would you use the other circuits?
6. With the common emitter h parameters developed in Equa-
tions (81) to (84), calculate the voltage, current, power gain,
and the input and output resistance in the C.E. configuration
shown in Fig. 75 (A). Use Rg = 1000 ohms, RL = 15,000 ohms.
7. Convert the h parameters referred to in question 6 to common
base h parameters. Calculate the quantities asked for in ques-
tion 6 for an Rg= 200 ohms, Rr.= 50,000 ohms.
8. Convert the h parameters referred to in question 6 to com-
mon collector h parameters. Calculate the quantities asked for
in question 6 for an R 0= 50,000 ohms, Rr.= 100 ohms.
9. Short circuit the output and calculate the input resistance for
the configurations referred to in questions 6, 7, and 8. Use the
abbreviated formulas of Table 7.
I 0. Is it possible to use the h parameter equivalent circuit at
high frequencies? At other operating points than the param-
eters were measured at? How does one convert from the
parameters of one operating point to the parameters of
another operating point?
136 INDEX
Absolute zero, 9 characteristics, 71, 72, 7!1, 74

Acceptoratom, 19 characteristic curve, 75
Ac dynamic resistance, 69 fixed-bias, amplifier, 84
Active region, 94, 95, 98 voltage amplifier, 75
Alloy, !18 Comparison oi basic transistor and
Angstrom, 2 vacuum tube circuits, 56
Angstrom unit, 5 Compounds, 7
Antimony, 19 Computer flip-flops, !17
Arsenic, 19 Conduction by holes, 15, 16, 17, 18
Asymmetrical transistor, 49 Conductive current, 45
Atom, 1, 2, !I, 4 Conductor, II, 12, Ill
Atoms in metals, 7, 8, 9, 10, 11 Constant bias current, 66
Atomic ingredients, 7 Constant characteristics, I 06
Atomic number, 4 Covalent bonds, 9
Avalanche breakdown, !15 Covalent force, 8
Avalanche multiplication, !15 Coulomb's Law, 2
Back-to-forward resistance, !14 Crystal lattice, 9
Barrier capacitance, !19 Crystal structure, Ill
Barrier region, 27 Current:
Base voltage loop equation, 92 amplification, 80
Base wafer, 49 carriers, 55
Bias supplies voltage, 6!1 gain, 49, 5!1, 128
Bias voltage, 82 operated devioe, 55
Blocking bias, 29 source, ll0
Boron, 20 Cutoff region, 97
Cathode follower, 88 D-c input resistance, 104
Centrifugal force, 4 D-c static resistance, 68
Charge density, 24 Deficit conduction, 15
Chemical difference, 21 Diamond lattice, 9
Circuit theorem, 107, 108, 109, ll0, ll l Difference current, 72
Collector, 47 Diffused base transistor, 58, 59
base junction, 49, 74 Diffused junction, 26
current, 69 diode, !18
load line, 88 transistor, 58
Collector-to-base current, 67 Diffusion, 25
voltage, 68, 69 Diffusion bias potential, 27
Collector-to-emitter voltage, 71, 79 Diode types, !16, !17, !18
Collector voltage loop equation, 92 Distortion, 82
Common base, Ill! Distortion and region of operation.
amplifier, 72, 85, 96 9!1, 94
characteristics, 75 Donor atom, 19, 21
configuration, 64, 69 Doping, 20
current gain, 7!1 Dynamic output resistance, 68, 71
parameter, ll8 plate resistance, 68
transistor amplifier, 58 quantities, 10!1, 104, I05. 106
Common collector, I Ill range, 100
amplifier, 87 resistance, l 04
circuit, I 18 Einstein's equation, 5
Common emitter, 9!1, 11!1 Ejected electron, 15
amplifier, 57, 82 Electric current, I I
INDEX 137
Electric field, 11 parameter, 112, 11!1, 114
Electron conduction, 15 conversion formulas, 121, 122
Electron current, 16, 48 Hydrogen, 4
Electron-hole diffusion, 27 Impurity atoms, 21, 24
pair, 16 Indium, 20, 58
recombination, 49 Input and output resistance of typical
collision, I 7 transistors, 1!14
Electron orbit, !I Input resistance, 127
Electron sharing, 8 Input signal, 105
Electrons, 2 Instantaneous total value, 6!1
Electrostatic force, 2, 4 lnsulatcr, l!I, 14
Electrostatic hill, 27 Interatomic bonds, l!I
Element, 7 Interrelation of Q point, V00, and
Emitters, 47 region of operation, 97, 98, 99,
Emitter, base, 52 100, 101
bias, 8!1, 84, 85 Intrinsic germanium, 20
current, 68, 70 Intrinsic semiconductor, 14, 17
follower, 57, 88 Ionization, 6
resistance, 68 Iron pyrite, 9
Emitter-base junction, 47 Isolation between input and output
Equivalent circuits, 106, 107 l!l!I
Explorer satellites, 42 Junction:
Finding R 8 , 92, 9!1 capacitance, !18, !19, 40, 41
Fixed bias, 8!1 diode, !17
Forward bias, !II, !12, !IS, !14, !15 region, 26
Forward current, !14 transistor, 47
Forward short-circuit current gain, Kirchhoff's voltage law, 76, 85
69, 125 Lead sulfide, !16
Free electrons, 10, 11 Light, 5
Fused junction diode, !18 Linear amplifier, 97
Galena crystal, !16 Load line, 75
Gallium, 20 Load resistance, 68
Gas tubes, 65 Low-resistance circuit, 54
Germanium, 9, 14, 16 Majority carriers, 19, !10
point contact diodes, !17 Mean free path, 11
wafer, 59 Microwave mixers, !17
Graphical analysis, 106 Minority carrier injection, !1!1, 46
Graphical determination of hybrid Minority carriers, 19, 45
parameters, 114, 115, 116, 117, 118 Mobile charge, 19
Graphical static input, 114 Mobility of the electrons, 12
Grounded base amplifiers, 52, 56 Natural diffusion bias, 29
Grounded collector, 57 Negative barrier, 61
Grown junction diode, !17 Negative electrons, 17, 65
Grown junction transistor, 58 Negative feedback, 125
Headlight dimmer, 4!1 Negative t,emperature coefficient, 18
High-resistance circuit, 54 Neutral atoms, !I
Hole current, 16 Net hole current, 28
H parameter equivalent circuit, 121, N-impurity slab, 60
122, 125, 126, 127 Nonlinear capacity, 41
Hybrid notation. 122 Nonlinear region, 97
138 INDEX
Norton's equivalent circuit, 109 Recombination, 20

Norton's theorem, 107 Reverse bias, 28, 29, 30, 31
N-p-n transistor, 47 Reverse-biased diode, 30
N-p-n transistor configuration, 61 Reverse current, 46
N-type semiconductor, 19 Reverse saturation current, 31, 35, 47,
Nucleus, 2 51, 70, 74
Ohm's law, 12 Reverse voltage amplification, 112, 116
Operating point, 105 Review questions, 22, 44-44, 62, 102, 1!15
Output: Saturation region, 95, 96, 97
characteristics, 114 Semiconductor, I, 14, 15
conductance, 117 rectifier, 23
current generator, 126, 127 triode, 54
equivalent circuit, 125 Semiconductors with impurities, 18-22
resistance, 127 Shells, !I
Parameter, 68 Short-circuit collector current, 77
values, typical transistor, 19 Silicon, 16
Parametric amplifier, 41 Silicon point contact diodes, 37
Parent atoms, 24 Silicon power diodes, 38
Phosphorus, 19 Small signal amplifier, 87
Photodiode, 41, 42, 43, 46 Small signal analysis, I 03
Photon, 5 Solar battery, 42
Photon bombardment, 6, 16 Solid state physics, 13
Photoresistors, 41, 42, 43 Source resistance, 127
Photovoltaic cell, 42 Space charge, 39
Physical insulator, 14 Static collector characteristics, 67, 71
P-n junction diodes, 2!1, 28-!15 Static input junction resistance, 81
Point contact diode, !16 Static quantities, 103, 104, 105, 106
Point contact transistor, 58, 61 Surface barrier transistor, 58, 61
Positive: Symmetrical transistor, 49
current, 47 Uncovered ions, 26
electrode, 13- Vacuum tube diode, 45
holes, I 7, 65 Valence electrons, 3, 6, 9, 18
ions, 65 Variable resistor, 79
temperature coefficient, 13, 18 Variactor diodes, 41
terminal, 11 Variation of h parameter, 119, 120
Positively charged body, 15 Varicap, 40
Potential energy, 3 Video secopd detector, 37
Potential hill, 27 Volt-ampere characteristics, !14
Power gain, 129 Voltage amplification, 5!1
Power supplies, 37 Voltage feedback ratio, 112
Protons, 2 Voltage gain, 127
P-type crystal, 24, 25 Voltage generator, 81
Q point, 85, 87 Voltage-operated device, 55
Quanta of energy, 5 Voltage regulator, 36
Quantum theory, 5, 6, 7 Voltage variable diode capacitor, 40
Quiescent base current, 78, 83 Wavelength, 6
Quiescent bias current, 66, 81 Zener breakdown voltage, 35, 36, 101
Quiescent output voltage, 79 Zener diodes, 36
Quiescent point of operation, 78, 84 Zero collector-to-emitter voltage, 77
Random movement, 11

Semiconductors and Transistors - Alexander Schure

Uploaded by

Copyright:

Available Formats

Semiconductors and Transistors - Alexander Schure

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Semiconductors and Transistors - Alexander Schure

Uploaded by

Copyright:

Available Formats

ELECTRONIC TECHNOLOGY SERIES

Alexander Schure, Ph.D., Ed. D.

JOHN F. RIDER PUBLISHER, INC.,

JOHN F. RIDER PUBLISHER, INC.

Semiconductors and the transistors derived from these materials

1 Introduction to the Semiconductors ....... .

2 The P-N Junction Diode 23

Introduction to the Transistor ..... 45

4 The Transistor as a Circuit Element 63

5 Small Signal Analysis ................. . 103

Index .......................... . 136

INTRODUCTION TO THE SEMICONDUCTORS

Up to World War II, a material was considered either a con-

The atomic theory of matter was one of the most important

theory not only explains many of the atom's physical character-

Fig. 1. The hydrogen atom.

Actually, a comparison of an atom with our solar system is quite

I The Angstrom unit, symbolized by A, denotes a length of 0.0000000001 meter

Fig. 2. Shells of an atom.

electron attract one another. For convenience, we designate both

indicate different shells. (Within each shell, the electrons indicated

Fig. 3. Schematic representation of some neutra I atoms.

as the atomic number increases, the number of electrons in any

2. The Quantum Theory

The amount of energy an electron can absorb is proportional

PHOTON EMITTED FltJ. 4. Absorption and emla•

When a photon has precisely the correct amount of energy

Thus far in our study of the atom, we have concentrated on its

Fig. 5. The sharing of elec•

Fig. 6. Geometric represen-

CRYSTAL STRUCTURE OF GERMANIUM

A strange thing happens when atoms are so dose together. The

wh<\ first explained its electronic structure, called it THE chemical

aimlessly throughout the metal, and not be captured by any other

,___ _ _ _ _--I SOURCE OF

It is these extra electrons that are present in all metallic con-

If our piece of metal were connected across the terminals of a

applied electric field in volts per meter, and µ. is the mobility, we

the resistance of a metal should decrease as the temperature is

We now know that a material's resistance is governed, to a large

Usually there are so many free electrons present that no heed is

A semiconductor falls into a special classification all its own. It

Heat need not be the only source of energy used to break

___ ',! ___!J 1!._ __ _

Holes may serve as conductors of electric current. If (as in Fig.

ity. In Fig. 11, it can be seen that an electron from an adjacent

for every hole attracted to the negative plate, an electron enters

In an intrinsic semiconductor, a hole can only be

the temperature of a semiconductor is raised, the mo-

8. Semlcondudors with Impurities

Intrinsic semiconductors are not very useful, because electrons

germanium atom, insofar as its ability to help hold the crystal

GERMANIUM 1•) (•I [•1

negative free electrons. So, the crystal is called an n-type semi-

Fig. 14. Addition of im-

called a p-type semiconductor. Again, because of therm.al agitation,

Even a minute quantity of impurity added to an intrinsic semi-

is supplied to the crystal-more germanium covalent bonds are

I. If oxygen has an atomic number 16, how many protons does

THE P-N JUNCTION DIODE

_ ',! _!J 1!._ __ _