Phys 328

Electronics for Scientists
PHYS-328
Brigitte Vachon
<Brigitte.Vachon@McGill.ca>
Department of Physics, McGill University
December 7, 2015
2015 Vachon
Contents
1 A brief history of electronics
2 A few useful reminders about impedance

12
2.1 Input impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Output impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.3 Signal transmission and impedance matching . . . . . . . . . . . . . . . . . . . . . . 14
3 Diode
3.1 Diode circuits . . . . . . . . . .
3.1.1 Voltage divider . . . . . .
3.1.2 Voltage regulator . . . . .
3.1.3 Limiter (clipper) circuits .
3.1.4 Variable limiter circuit . .
3.1.5 Clamp circuit . . . . . . .
3.1.6 Half-wave rectifier circuit
3.1.7 Full-wave bridge rectifier .
3.2 Different types of diodes . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
24
24
25
26
27
28
30
31
34
CONTENTS
3.2.1
3.2.2
3.2.3
CONTENTS
Light-emitting diode (LED) . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Photodiode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Zener diode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4 Transistors
4.1 Bipolar junction transistors (BJTs) . . . . . . . . . .
4.1.1 Theory . . . . . . . . . . . . . . . . . . . . .
4.1.2 Transistor biasing . . . . . . . . . . . . . . .
4.1.3 Intrinsic resistance . . . . . . . . . . . . . . .
4.1.4 Common-collector (emitter-follower) amplifier
4.1.5 Common-emitter amplifier . . . . . . . . . . .
4.1.6 Differential amplifier . . . . . . . . . . . . . .
4.2 Field effect transistors (FET) . . . . . . . . . . . . .
4.2.1 JFET . . . . . . . . . . . . . . . . . . . . . .
4.2.2 MOSFET . . . . . . . . . . . . . . . . . . . .
4.2.3 Theory . . . . . . . . . . . . . . . . . . . . .
4.2.4 Transconductance . . . . . . . . . . . . . . .
4.2.5 Current source circuit . . . . . . . . . . . . .
4.2.6 Transistor biasing . . . . . . . . . . . . . . .
4.2.7 Common-source amplifier . . . . . . . . . . .
4.2.8 Source-follower amplifier . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
39
42
45
50
54
57
61
66
74
75
76
77
82
84
87
88
91
5 Operational amplifiers
93
5.1 Characteristics of ideal op-amps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Brigitte VACHON
December 7, 2015
CONTENTS
5.2 Comparator . . . . . . . . . . .
5.3 Feedback . . . . . . . . . . . . .
5.3.1 Effect of negative feedback
5.4 Op-amp Golden rules . . . . .
5.5 Buffer . . . . . . . . . . . . . . .
5.6 Inverting amplifier . . . . . . . .
5.7 Non-inverting amplifier . . . . .
5.8 Summing amplifier . . . . . . . .
5.9 Integrator . . . . . . . . . . . .
5.10 Current source . . . . . . . . . .
5.11 Characteristics of real op-amps .
5.12 Oscillators . . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
on voltage gain . . .
on input impedance .
on output impedance
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
6 Digital electronics
6.1 Number codes in digital electronics .
6.1.1 Binary . . . . . . . . . . . .
6.1.2 Hexadecimal . . . . . . . . .
6.1.3 Binary Coded Decimal (BCD)
6.1.4 Gray code . . . . . . . . . .
6.1.5 ASCII . . . . . . . . . . . . .
6.2 Parallel and Serial Bit Transmission .
6.3 Logic functions . . . . . . . . . . . .
Brigitte VACHON
CONTENTS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
December 7, 2015
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
97
98
99
100
101
102
102
103
104
105
106
107
108
111
.
.
.
.
.
.
.
.
113
. 116
. 116
. 117
. 118
. 121
. 121
. 123
. 124
3
CONTENTS
CONTENTS
6.3.1 Example: Enable/Disable control . . . . .

6.4 Combinational Logic and Boolean Algebra . . . .
6.4.1 NAND and NOR gate universality . . . .
6.5 Karnaugh Maps . . . . . . . . . . . . . . . . . .
6.6 Sequential Logic . . . . . . . . . . . . . . . . . .
6.6.1 SR flip-flops . . . . . . . . . . . . . . . .
6.6.2 Level-triggered (or gated) SR flip-flop . . .
6.6.3 Edge-triggered SR flip-flops . . . . . . . .
6.6.4 Pulsed-triggered SR flip-flops (Master-Slave
6.6.5 Symbolic representation . . . . . . . . . .
6.6.6 D-type flip-flop (Gated D latch) . . . . . .
6.6.7 JK flip-flop . . . . . . . . . . . . . . . . .
6.6.8 Summary of flip-flop types . . . . . . . . .
6.6.9 Timing considerations . . . . . . . . . . .
6.6.10 Asynchronous (ripple) counters . . . . . .
6.6.11 Synchronous counters . . . . . . . . . . .
6.6.12 Memory Registers . . . . . . . . . . . . .
6.6.13 Shift registers . . . . . . . . . . . . . . . .
7 Analog/Digital interfacing
7.1 Sampling . . . . . . . . . . . . . . .
7.2 Nyquist(-Shannon) sampling theorem
7.3 Aliasing . . . . . . . . . . . . . . . .
7.3.1 Anti-aliasing filter . . . . . .
Brigitte VACHON
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
December 7, 2015
.
.
.
.
.
.
.
.
. .
. .
. .
. .
. .
. .
. .
. .
SR
. .
. .
. .
. .
. .
. .
. .
. .
. .
.
.
.
.
.
.
.
.
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
flip-flop)
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
126
126
130
131
137
138
145
146
148
149
149
153
156
157
159
163
165
166
.
.
.
.
169
. 170
. 171
. 172
. 175
4
CONTENTS
7.4
7.5
7.6
7.7
Sample and Hold . . . . . . . . . . . . .

Analogue-to-Digital Conversion . . . . .
Resolution . . . . . . . . . . . . . . . .
Analog to Digital Conversion Methods .
7.7.1 Staircase ADC . . . . . . . . . .
7.7.2 Successive Approximation ADC .
7.7.3 Parallel-Encoded (Flash) ADC .
7.8 Digital to Analog Conversion . . . . . .
7.8.1 Binary weighted resistor method .
7.8.2 R/2R Ladder DAC . . . . . . . .
8 Memory Devices
8.1 Memory Organization . . . . . . . .
8.2 Read-Only Memory (ROM) . . . . .
8.2.1 MROM . . . . . . . . . . . .
8.2.2 Progammable ROM (PROM)
8.2.3 Erasable Programmable ROM
8.2.4 Flash memory . . . . . . . .
8.3 Random Access Memory (RAM) . .
9 (If
9.1
9.2
9.3
you are interested) IC,

ASIC . . . . . . . . . . .
FPGA . . . . . . . . . .
Microprocessor . . . . . .
Brigitte VACHON
CONTENTS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
191
. 192
. 196
. 197
. 198
. 199
. 200
. 201
microprocessor...
. . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . .
203
. 203
. 212
. 216
. . . . . .
. . . . . .
. . . . . .
. . . . . .
(EPROM)
. . . . . .
. . . . . .
ASIC, FPGA,
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
December 7, 2015
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
176
177
179
180
180
182
183
185
187
188
CONTENTS
CONTENTS
A Grounding issues
218
A.1 Ground loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
A.2 AC ground . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
B Semiconductor technology
222
C Current in semiconductors
230
D Band gap theory
240
E Stability of amplifiers using feedback
253
Brigitte VACHON
December 7, 2015
Chapter 1
A brief history of electronics
1870-1910: Cathode rays, X rays and the electron (Hittorf, Crookes, Braun[Nobel],
Roentgen[Nobel], JJ Thomson[Nobel], Millikan[Nobel])
1885-1910:Wireless telegraphy (Hertz, Branly, Lodge, Marconi[Nobel])
1904-1932: Vacuum Tubes (Edison, Fleming, Coolidge, Schottky, Langmuir[Nobel])
1900-1927: Radio (Fessenden, Armstrong, Hazeltine, Black)
1920-1930: Television (Baird, Farnsworth, Zworykin)
1920-1940: Radar (Watson-Watt, Loomis, Rabi[Nobel], Alvarez[Nobel], Purcell[Nobel], Dicke)
1900-1930: Electrons, waves and the birth of solid state physics (Boltzmann,
Planck[Nobel], de Broglie, Schrodinger, Fermi[Nobel], Dirac[Nobel], Bloch[Nobel], Peierls, Wilson)
1934-1948: Transistors (Southworth, Ohl, Lark-Horovitz, Shockley[Nobel], Brattain[Nobel],
Bardeen [2Nobel])
7
CHAPTER 1. HISTORY
1958: Integrated circuits (Kilby[Nobel])
Brigitte VACHON
December 7, 2015
CHAPTER 1. HISTORY
(source: http://adoxoblog.wordpress.com/2013/02/27/colossus/)
(source: http://www.maximumpc.com/article/features/12_things_you_didn%E2%80%99t_know_about_eniac)
(source: http://www.computersciencelab.com/ComputerHistory/HistoryPt4.htm)
The die from an Intel 8742, an 8-bit microcontroller that includes a CPU running at 12 MHz, 128
bytes of RAM, 2048 bytes of EPROM, and I/O in the same chip.(source: wikipedia)
Brigitte VACHON
December 7, 2015
CHAPTER 1. HISTORY
Your typical electronic devices rely largely on integrated circuits and semi-conductor devices.
(source: http://www.sonyclassics.com/itmightgetloud/)
(source: Apple)
What is a semiconductor?
Material that has electrical conductivity between that of a conductor and that of an insulator.
What is a semiconductor device?
An electrical component that relies on the properties of semiconductor (ex. silicon).
Brigitte VACHON
December 7, 2015
10
CHAPTER 1. HISTORY
In the context of scientific research, electronics often custom designed and built for specific applications and needs.
Typical Data AcQuisition (DAQ) system
sensor
Signal conditioning
(ex. filter, pre-amp)
Processing
stage (ex. ADC,
microprocessor)
Thin Gap Chamber readout for the ATLAS detector at the CERN LHC.
Detector testing facility at McGill University.(source: B.Vachon)
Brigitte VACHON
December 7, 2015
11
Chapter 2
A few useful reminders about impedance
Impedance: Measure of the opposition that a circuit presents to a current when a voltage is applied.
An extension of the concept of resistance to AC circuits. (source: Wikipedia)
Since the concept of impedance in an AC circuit takes into account effects of inductance and capacitance, impedance is a frequency dependent quantity. Effects of inductance and capacitance
generally more pronounced at high frequency.
(source: P.Scherz and S.Monk, Practical Electronics for Inventors)
12
CHAPTER 2. IMPEDANCE
2.1
2.1. INPUT IMPEDANCE
Input impedance
Impedance seen looking into the input of a circuit or device.

Small input impedance Large current can be drawn into the input device
(Ex. An audio speaker has an input impedance of typically just a few ohms to draw lots of currents to drive voice coil)
Large input impedance Small current drawn into input device

(Ex. Audio hi-fi pre-amplifier. Device with large input impedance doesnt cause a drop in the source voltage.)
2.2
Output impedance
Impedance looking back into the device or circuit.

Small output impedance Large current can be drawn out of the device
(Ex. DC power supply in the lab with output impedance of order milli-ohms. Can supply large current without suffering from a drop
in voltage.)
Large output impedance Small current drawn out of device before voltage at the output drops
(Ex. Electric guitar, high-Z microphone)
Brigitte VACHON
December 7, 2015
13
2.3
Electronics 2.3.
for Scientists
SIGNAL
TRANSMISSION AND IMPEDANCE MATCHING
Signal transmission and impedance matching
In terms of efficient signal transmission, you generally want Zout Zin .
Device A
Device B
Zout
Zin
Videal
Rule of thumb: Input impedance should be at least 10 times larger than output impedance in order
to avoid the Device B overloading Device A.
Small output impedance typically a good thing since it allows to source a greater current while
avoiding excessive power lost to resistive heating.
We will see that one of the useful characteristics of operational amplifiers (Op amps) is their large
input impedance and low output impedance.
Brigitte VACHON
December 7, 2015
14
2.3. IMPEDANCE MATCHING
Impedance of transmission line
Signals will typically travel between two

devices over a transmission line.
It is important to understand that transmission lines have their own intrinsic
impedance.
Z0 < ZL
Zo > ZL
Z0 = ZL
Brigitte VACHON
December 7, 2015
15
2.3. IMPEDANCE MATCHING
Example of commonly used cables and their impedance.

Coaxial cable
Twisted-pair cable
Ribbon cable
(source: wikipedia)
(source: wikipedia)
(source: wikipedia)
Usage: Transmission line for

radio frequency (analog) signals. Usage: Telephone, ethernet
Pros: EM field carrying the signal only exists in between inner
and outer conductors. Protect
signal from external source of
EM interference.
Impedance: 50 for RG-58
Brigitte VACHON
Usage: Digital system, design for

low-level voltages.
Pros: EM interference into or

coming from the cable minimized Pros:
Flexible and multiple
by twisted geometry. EM field simultaneous parallel signals.
cancellation.
Impedance: 110-130
for
Impedance: 100 for ethernet 26AWG wire, 0.050 spacing and
cable
PVC insulation
December 7, 2015
16
Chapter 3
Diode
A diode is a two-terminal, passive (i.e. no built-in power source), non-linear 1 device.
A diode is essentially an electrically controlled one-way gate.
Usually made from semiconductors but can also be a vacuum tube device.
We will only consider here semiconductor diodes.
Thevenin and Norton laws do not apply to circuits containing non-linear devices like diodes!
17
CHAPTER 3. DIODE
Diodes are useful devices that can be used for many applications such as:
power conversion (rectifiers)
ex. automotive alternator, ACDC conversion
Over-voltage protection
ex. diodes on connection pins of an IC
logic gates
temperature measuring
ionizing radiation detectors
voltage dropper
etc.
(source: http://en.wikipedia.org/wiki/Diode)
Circuit Symbol:
Brigitte VACHON
anode
December 7, 2015
cathode
18
CHAPTER 3. DIODE
Doped semiconductor materials
n-type
p-type
(source: T.Floyd, Electronic Devices (Conventional Current Version))
Pentavalent impurity atom in a silicon

crystal structure. The extra electron from
the antimony (Sb) atom becomes a free
electron.
Trivalent impurity atom in a silicon crystal

structure. A boron (B) impurity atom is
shown.
See appendices B, C and D for more information.

Brigitte VACHON
December 7, 2015
19
CHAPTER 3. DIODE
External electrical field supplied by battery

forces electrons and holes to move to the center and recombine, thereby creating a current
through the device. [* Diagram missing a limiting resistor, why?]
Brigitte VACHON
The external battery voltage attracts electrons and holes away from the p-n junction,
thereby creating a depletion region. No current can flow through the diode.
December 7, 2015
20
CHAPTER 3. DIODE
Ideal diode model: Diode acts as short-circuit (wire) in one direction and as open circuit in other
direction. In other words, it only conducts current in one direction, with a very small forward voltage
drop.
(source: http://www.engineersgarage.com/tutorials/diodes)
Brigitte VACHON
December 7, 2015
21
CHAPTER 3. DIODE
Practical diode model: For many applications and circuit analysis a simple working model of a
diode is sufficient. In this working model, the diode is assumed to be a perfect diode and the average
forward voltage drop is approximately VD 0.6 V.
Brigitte VACHON
December 7, 2015
22
CHAPTER 3. DIODE
Real diode model:

forward bias turn-on voltage 0.6 V
forward current eV
reverse current small 1A
reverse breakdown voltage > 75 V (for IN914)
(source: http://www.engineersgarage.com/tutorials/diodes)
Diode (Ebers-Moll or Shockley) equation for voltage greater than breakdown voltage.
I = Io(eV /VT 1) where,
VT = kT /e (thermal voltage) is a characteristic of each diode (e.g. VT 50 mV for Silicon.)
I0 is characteristic of each diode and temperature DEPENDENT (e.g. near 200 C, I0 increases by
9% by degree).
Brigitte VACHON
December 7, 2015
23
CHAPTER 3. DIODE
3.1
Diode circuits
3.1.1
Voltage divider
3.1. DIODE CIRCUITS
Example: You have a 9 V battery but need lower voltage to power a device.
V0
0.6 V
0.6 V
0.6 V
Similar result could be achieved with resistor(s), but the use of diode(s) typically reduces the amount
of power wasted due to resistive heating and will supply a regulated voltage that is sensitive to current
variations.
Brigitte VACHON
December 7, 2015
24
CHAPTER 3. DIODE
3.1.2
3.1. DIODE CIRCUITS
Voltage regulator
The series resistor is used to set the desired output current.

Series resistor and diodes must have proper rating for amount of current drawn.
Brigitte VACHON
December 7, 2015
25
CHAPTER 3. DIODE
3.1.3
3.1. DIODE CIRCUITS
Limiter (clipper) circuits
Type of circuits used to cap voltages to preset limits. Useful as voltage protection circuitry or
signal conditioning.
Brigitte VACHON
December 7, 2015
26
CHAPTER 3. DIODE
3.1.4
3.1. DIODE CIRCUITS
Variable limiter circuit
What output voltage do you expect to see?

Brigitte VACHON
December 7, 2015
27
CHAPTER 3. DIODE
3.1.5
3.1. DIODE CIRCUITS
Clamp circuit
When Vin < 0.6 V, diode conducts and charge capacitor to VC = Vpeak 0.6.
Capacitor cannot discharge
Vout = Vin + VC
Brigitte VACHON
December 7, 2015
28
CHAPTER 3. DIODE
3.1. DIODE CIRCUITS
Building a DC power supply

All active electronic devices require a source of constand DC supplied either by a battery or a DC
power supply. Therefore, the DC power supply is one of the most common circuit!
Brigitte VACHON
December 7, 2015
29
CHAPTER 3. DIODE
3.1.6
3.1. DIODE CIRCUITS
Half-wave rectifier circuit
Rectifier circuits: Making alternating voltage uni-directional (first step in building a circuit that
transforms ACDC.)
This is called a half-wave rectifier circuit since it only provides a positive voltage for half the period
of the wave.
Brigitte VACHON
December 7, 2015
30
CHAPTER 3. DIODE
3.1.7
3.1. DIODE CIRCUITS
Full-wave bridge rectifier
This is an example of a common circuit called a full-wave bridge rectifier.

The circuit acts as if it is taking the absolute value of the alternating voltage (minus the voltage
drop of two diodes).
Brigitte VACHON
December 7, 2015
31
CHAPTER 3. DIODE
3.1. DIODE CIRCUITS
A full-wave bridge rectifier circuit is more often represented as follow:
(source: http://www.electronics-tutorials.ws/diode/diode_6.html)
Brigitte VACHON
December 7, 2015
32
CHAPTER 3. DIODE
3.1. DIODE CIRCUITS
If you add a capacitor in parallel with the load, the capacitor smooths out the ripples of the full-wave
rectifier output thereby providing essentially a DC voltage from an AC input. Voila!
Time constant is given by RLC.

When input voltage is -1.2 V < Vin <
1.2 V, then
Vout = (Vinpeak 1.2)et/RLC
.
To obtain mostly constant DC output
voltage, RLC T /2 (where T is the
period of the input voltage), in which
case,
Vout
(Vinpeak
T
...
1.2) 1
2RLC
Pre-made rectifier circuits are available in different

sizes/packages. =
Brigitte VACHON
December 7, 2015
33
CHAPTER 3. DIODE
3.2
3.2. DIFFERENT TYPES OF DIODES
Different types of diodes
In addition to the ordinary diode that we have been looking at until now, there exists other types
of diodes, here are a few you may come across as some point or another.
3.2.1
Light-emitting diode (LED)
Circuit symbol:
It is possible for an electron to jump

from the conduction band (higher
energy level) to the valence band (lower
energy level) mostly populated by holes.
LEDs are made using semiconductor
materials allowing the release of energy
in the form of a photon.
Need to package the diode to allow the
photons produced to escape.
Colour of emitted light is determined by
the band gap energy of the semiconductor materials.
(source: http://en.wikipedia.org/wiki/Light-emitting_diode)
Brigitte VACHON
December 7, 2015
34
CHAPTER 3. DIODE
3.2.2
Photodiode
Circuit symbol:
Here, the inverse process occurs. Incoming photons are absorbed in the p-n junction and the energy
deposited create additional electron-hole pairs in the depletion region. Think of it as a photoelectric effect taking place in the depletion region of the p-n junction.
The number of electron-hole pairs created is directly proportional to the intensity of light (i.e. number of photons).
Note that all p-n junction are sensitive to EM radiaton (UV, x-rays, -rays). In order to avoid
the malfunctioning of semiconductor devices due to induced photocurrent, electronic devices (ex.
transistors, ICs) are encapsulated in an opaque casing.
There are two principal modes of operation of a photodiode:
Photovoltaic mode: Photodiode is used with zero bias. Light creating electron-hole pairs
produces a current. (ex. solar cell)
Photoconductive mode: Photodiode reversed biased creating a larger depletion region. Incoming photons create electron-hole pairs which results in a reverse current flow which is directly
proportional to the illuminance2. (ex. optocoupler)
2
Perceived power of light incident on a surface, per unit area
Brigitte VACHON
December 7, 2015
35
CHAPTER 3. DIODE
Example of application: Solar cells
** Should say antireflective coating

Brigitte VACHON
December 7, 2015
36
CHAPTER 3. DIODE
Example of application: Data transfer over optical fiber
Optical fibers can transport data over much longer distances with minimal losses
Much higher bandwidth of data can be sent over optical fibers
Provide electrical isolation between two different electrical circuits
low noise
Reading out the

Semiconductor
Tracker
(source: atlas.ch)
Brigitte VACHON
December 7, 2015
37
CHAPTER 3. DIODE
3.2.3
Zener diode
Circuit symbol:
Zener diodes are similar to normal diodes but have smaller breakdown voltage, Vb, which is meant
to be attained (i.e. not bad or dammaging point to exceed in the operation of the diode).
These diodes are intended to be used in reverse bias mode.
Zener diodes are manufactured to have extremely sharp I-V curve breakdown. They are particularly
well-suited for the production of a reference voltage point and as voltage stabilizer removing ripple
from a low quality DC voltage.
Brigitte VACHON
December 7, 2015
38
Chapter 4
Transistors
Transistors are essential ingredients of virtually every electronic circuit!
Invented by Shockley, Bardeen and Brattain in 1947. Nobel prize in Physics in 1956.
(source: Bell Labs)
39
CHAPTER 4. TRANSISTORS
Transistors are three-terminal active components (i.e. need to supply voltage/current to make
them function).
(source: http://www.mikroe.com/old/books/keu/04.htm)
They can be thought of as being similar to a water faucet. While a knob on a faucet controls the
flow of water, a voltage and/or current applied to a control lead on a transistor controls the current flowing through the other two leads. In order words, the amount of current flowing through a
transistor is proportional to a control signal.
Ok, not very exciting up to now, is it? We are just more or less talking about an electrically controlled current faucet where the output current is proportional to the controlling input signal. But
heres what makes transistors so famous: The output power of a transistor can be higher
than the controlling input power, which means that you can use a transistor to
amplify a signal!
Brigitte VACHON
December 7, 2015
40
The two main general applications of transistors can be summarized as:

(current) control device
amplifier
There are two main families of transistors:

Bipolar junction transistors (BJTs)
control signal is a current
higher voltage gain in amplifier circuits than FETs.
Field-effect transistors (FETs)
control signal is a voltage (essentially zero current)
high input impedance ( 1014 !) since they draw very little current
typically used in integrated circuits
Brigitte VACHON
December 7, 2015
41
4.1
4.1. BIPOLAR JUNCTION TRANSISTORS (BJTS)
Bipolar junction transistors (BJTs)
The bipolar junction transistor consists in two pn junctions.
Brigitte VACHON
December 7, 2015
42
If zero volt is applied to the base, no

electrons can flow from the emitter to the
base; this is just like the p-n junction we
have seen in a diode.
If a positive voltage of at least 0.6 V is applied at the base, the holes in the base are
pushed toward the middle of the p-n junction and electrons attracted. Recombination
of electron-holes can occur and a current can
flow from the base to the emitter.
Also see nice video: https://www.youtube.com/watch?v=jVyyHfsAfyA

Brigitte VACHON
December 7, 2015
43
Bipolar junction transistors can come in either npn or pnp configuration.
Brigitte VACHON
December 7, 2015
44
4.1.1
Theory
Now lets define different voltages and currents of a bipolar junction transistor in oder to be able to
describe its operation more quantitatively.
Brigitte VACHON
December 7, 2015
45
Lets now go into more details of how a npn1 bipolar junction transistors functions in normal operation.
For current to flow through transistor, voltage at collector must be greater than voltage at the
emitter, i.e. VC > VE.
The base-to-emitter voltage exhibits a voltage drop of 0.6 V, therefore VB must be at least
0.6 V greater than VE. It also means that in this case, you get VBE 0.6 V
By Kirchoffs current law: IE = IC + IB .
By Kirchoffs voltage law: VCE = VBE + VCB .
The most important feature of a transistor is described by the following formula: IC = hF E IB = IB .
Here, refers to the current gain. Typical values of are around 100.
Using the previous formula, you can show that in fact IE = ( + 1)IB, and for typical values of
this means IE IC .
Note: Every transistor has its own unique value which depends on temperature and changes
in VCE. Therefore, do not design circuits where the behaviour directly depends on the precise
value of !
1
polarities are reversed for a pnp transistor
Brigitte VACHON
December 7, 2015
46
Exercise: Given the values below and = 100, find VE,IE, IB, IC and VC.
Brigitte VACHON
December 7, 2015
47
Characteristic curves of a transistor are used to identify the working points of a transistor.
Find IC
assuming
VCE=0
(source: http://www.electronics-tutorials.ws/transistor/tran_2.html)
Brigitte VACHON
December 7, 2015
48
(Linear) active region: Region of operation where the collector current is roughly proportional to
the base current. [Transistor fully ON, without output current control]
Saturation region: Region of operation where collector current is no longer proportional to the base
current, the transistor acts like a closed switch from collector to emitter.
cutoff region: Region where the base current approaches zero resulting in a zero collector current.
[Transistor fully OFF]
Note that it is important to remember that a transistor has maximum allowed values of IC, IB and
VCE not too be exceeded, otherwise you burn the transistor. Transistors also have a maximum power
rating given by the maximum product of VCE and IC.
The desired operation point of a transistor is set by biasing the transistor, i.e.
setting the DC voltages of the transistors terminals.
Brigitte VACHON
December 7, 2015
49
4.1.2
Transistor biasing
To keep the operating point of a transistor in the linear active region for use as an amplifier, one
needs to carefully set the DC voltages of the transistors terminals.
Brigitte VACHON
December 7, 2015
50
Brigitte VACHON
December 7, 2015
51
Here is the common method of setting the constant DC operating conditions of a transistor: the
quiescent point or Q-point.
Use a voltage-divider circuit to set the DC Base
voltage:
2
VB = VCC R1R+R
2
The following steps introduces some useful rules of
thumb:
1. Choose the DC value of your output in order to allow for the largest possible symmetric output signal
swing without clipping the signal. Use the rule of
thumb of setting the quiescent output voltage as
Vout(E or C) 1/2VCC.
2. Choose the quiescent current you want flowing
through your transistor. Typical values for IC
IE are of the order of mA (e.g. 1 to 10 mA). This
in turn will give you an approximate target value
for IB = IC /.
3. Pick a value of resistor(s) RE or C that satisfy the
quiescent output voltage and collector current you
choose. [Relative values of RC and RE when appropriate are also chosen for desired gain]
Brigitte VACHON
December 7, 2015
52
4. Choose values for R1 and R2 such that I1 I2 >

10IB , and that you forward bias the Base-Emitter
junction (i.e. VBE > 0.6 V). In other words, you
want to make the impedance of the voltage-divider
circuit small compared to the rest of the transistor
circuit it drives.
5. The capacitor C is a decoupling capacitor allowing
the AC input signal to be at ground or any other
ambient level. Choose C (which forms a high-pass
filter with the base impedance in parallel with R1
and R2) such that the input signal frequencies of
interest get through.
Brigitte VACHON
December 7, 2015
53
4.1.3
Intrinsic resistance
One feature of bipolar junction transistor is the intrinsic (typically small) resistance present within
the emitter junction region.
Consider the base-emitter junction as a simple diode. The current as function of voltage across the
diode can be described by the Ebers-Moll equation:
VT = kTe ' 25 mV at 200C

IS = Saturation current
IC = IS eVBE/VT 1 ' IE
This intrinsic emitter resistance can be defined as

VBE
re

IE VB
By differentiating the Ebers-Moll equation, one gets:
re =
Brigitte VACHON
kT 1
e IC
25
IC (mA)
December 7, 2015
54
Now, imagine fixing the emitter voltage and look at how much the base voltage change in response
to a change in base current. This defines the intrinsic resistance rBE to the emitter as seen at the
base, i.e.
25
kT 1
VBE
=

IB VE
e IC
IC(mA)

rBE
rBE = re
The base-emitter junction appears to have a low resistance when viewed from the emitter end, but
a higher resistance (by a factor of ) when viewed from the base.
Note: This effective resistance rBE may have to be taken into account when you choose values of R1
and R2 to bias your transistor.
Brigitte VACHON
December 7, 2015
55
Before introducing different amplifier circuits, let me introduce some quantities that we will want
to be able to calculate for each circuit. In general, what we want to know about an amplifier is:
Voltage gain: Gain = a =
Vout (RL =)
Vin
Current gain: Gain = g =
Iout
Iin
(although not as useful since Iout depends on load (RL)).
Transconductance: Ratio of change in transistor output current to changes in control voltage.

[To be discussed later]
Input impedance: Zin =
Vin
Iin
Output impedance: Zout =
Brigitte VACHON
Vout (RL =)
Iout (RL =0)
December 7, 2015
56
4.1.4
Common-collector (emitter-follower) amplifier
This circuit shows what is called a common-collector amplifier (or also often referred to as emitterfollower). The main feature of this circuit is that the output is almost a mirror image of the input
(i.e. the output follows the input). It has no voltage gain! (...and you dare call this an amplifier?)
Ve
Brigitte VACHON
December 7, 2015
57
Is this a useless circuit? Nope.

This kind of circuit has a much larger input impedance than output impedance, which means that
it has a much larger output current (IE ) compared to its input current (IB ), i.e. the emitter follower
has current gain!
This circuit makes it possible to drive a load, requiring much less power from the signal source than
would otherwise be needed if you drove the load directly from the signal source. Brilliant!
Well also see that emitter-followers are good impedance buffers, i.e. they have a large input
impedance and small output impedance.
What is the voltage gain of this amplifier?
Brigitte VACHON
December 7, 2015
58
What is the input impedance as seen directly at the base of the transistor (i.e. not including
the voltage divider stage)?
To answer this question, we need to figure out by how much IB changes as a result of a change
in VB .
Therefore, the input impedance at the base of the transistor is Zin =
VB
IB
= (RE + re)
What is the input impedance of this circuit amplifier (i.e. including the voltage divider stage) ?
[Lets ignore the decoupling capacitor for simplicity]
Brigitte VACHON
December 7, 2015
59
What is the output impedance of the amplifier (i.e. with the load removed, looking back into
the emitter output)?
Since, Vout = VB 0.6 then
vout = vB
and also, since VB = Vin IB RB , then for fixed Vin
vB = IB RB
Putting these two equations together you get:
vout = IB RB =
IE
RB
But what is RB ? This is the effective resistance at the base of the amplifier. Ignoring the bypass
capacitor and any output impedance from the AC input signal generator, then
RB = R1 // R2
The output impedance of this emitter-follower amplifier is therefore

v

RB

out

=
Zout =
.

IE
Strictly speaking, Zout should also include RE in parallel but typically RB / dominates.
The effect of resistors attached to the base is reduced by significantly lowering the output
impedance of this amplifier circuit. Thats a good thing!
Brigitte VACHON
December 7, 2015
60
4.1.5
Common-emitter amplifier
The following circuit shows what is called a common-emitter amplifier. It is called a commonemitter because both the base and emitter leads are common (grounded through resistors). The
main feature of this circuit is that it provides a voltage gain.
(source:
T.Floyd, Electronic Devices (Conventional Current Version))
Brigitte VACHON
December 7, 2015
61
What is the voltage gain of this amplifier?

For small changes in voltage at the base vB we get correspondingly small changes at the emitter vE = vB (the base and emitter have a common ground through resistors).
These cause changes in the emitter current: iE =
vE
RE
and therefore changes in the collector current: iC = iE

The quiescent point of the circuit is such that: Vout = VCC IC RC
However, in order to calculate the voltage gain, what we are interested in is the change in output
voltage due to a change in the input voltage:
RC
RC
= vB R
vout = iC RC = iE RC = vE R
E
Finally, we find that the voltage gain of the circuit is: Gain =
vout
vB
RC
= R
Note the negative sign!! This means the output is inverted. Why? If Vin increases, then IC
increases and, by Ohms law, VC (= Vout) then decreases, voila!
Brigitte VACHON
December 7, 2015
62
But now, wait a minute! Does that mean that if RE = 0 that the gain of the amplifier is infinite?
No.
What we have forgotten here is to take into account the intrinsic resistance within the emitter
junction region of the transistor. Looking into the base of the transistor, this small resistance
re appears in series with the resistance RE .
C
The actual voltage gain of the amplifier is therefore Gain = R R+r
e
E
If RE = 0, then it looks like you get a much higher voltage gain out of your amplifier than if
you actually put a resistance RE in the circuit. Why would you put a non-zero RE ?
Well, the value of the intrinsic resistance re is quite unstable and depends for example on temperature. Therefore, to avoid unstable amplifier gain, a resistor RE is placed at the emitter at
the expense of loss in voltage gain.
But the story is not over yet, theres a trick you should know about. If you put a capacitor in
parallel with RE , then a high-frequency output signal will by-pass RE making it effectively
disappear. And therefore, for high frequency input signals the gain is dictated by re and no
longer by RE , thereby recovering high voltage amplifier gain but with temperature stability.
Neat eh?
Brigitte VACHON
December 7, 2015
63
Another measure of the performance of an amplifier is the concept of transconductance.

Transconductance (gm) is defined as the ratio of output current to input voltage
and has units of ampere per volt, or siemens (S).
This will be a concept particularly useful to describe FET amplifiers as we will see later.
For BJTs, the transconductance is defined as:
gm =
IC
VB
And re-using the work weve done earlier, we find (taking into account the intrinsic emitter
resistance):
1
gm =
RE + re
Brigitte VACHON
December 7, 2015
64
What is the input impedance of the amplifier ?
What is the output impedance?
Brigitte VACHON
December 7, 2015
65
4.1.6
Differential amplifier
Circuit that amplifies the difference between two signals and rejects any common components.
Long-tailed pair differential amplifier circuit

(source: http://macao.communications.museum/eng/exhibition/secondfloor/moreinfo/2_17_1_OpAmpLab.html)
Brigitte VACHON
December 7, 2015
66
Looking at the quiescent state of the long-tailed pair differential amplifier circuit.
Brigitte VACHON
December 7, 2015
67
Mode of operation: Single-ended differential input
Brigitte VACHON
December 7, 2015
68
Show that the differential gain is:

out = RC
Gdif f = vv1v
2R
2
E
Lets only worry about incremental changes in

voltage such that well ignore the base-emitter constant 0.6 V drop.
v1 = REi1 + R1(i1 + i2)
v2 = REi2 + R1(i1 + i2)
Solving for i2:
i2 =
For RE R1, this reduces to

i2 =
vout = RCi2 =
Brigitte VACHON
(4.1)
(v2 v1)
2RE
Then we find
Gdiff =
(v2 v1)R1 + v2RE

2R1RE + RE2
RC
(v2 v1)
2RE
vout
v1 v2
RC
= 2R
E
December 7, 2015
69
Mode of operation: Double-ended differential input
Brigitte VACHON
December 7, 2015
70
Mode of operation: Common-mode inputs
Brigitte VACHON
December 7, 2015
71
Show that the common mode gain is:

RC
=
GCM = vvout
2
R
1,2
1 +RE
To calculate the common-mode gain, lets assume
that the two inputs are identical v1 = v2.
Then, from equation 4.1
i2 =
v2RE
v2
=
2R1RE + RE2 2R1 + RE
And the output voltage is given by

vout = RCi2 = v2
RC
2R1 + RE
The common mode gain is then

GCM =
Brigitte VACHON
vout
v1,2
RC for R R
C
= 2RR
'
E
1
2R1
1 +RE
December 7, 2015
72
Ideally, you want your differential amplifier to provide a high gain for your desired signals and zero
gain for common-mode signals.
A differential amplifier performance is typically expressed in terms of its Common-mode rejection ratio (CMRR).
The CMRR is defined as

CMRR =
Gdiff
GCM
R1 for R R
R
E
1
E
In general you want to make R1 large (i.e. a long tail) to have a high CMRR.
Brigitte VACHON
December 7, 2015
73
4.2
4.2. FIELD EFFECT TRANSISTORS (FET)
Field effect transistors (FET)
FET is the name of a very large family of transistors that include Junction Field Effects Transistors
(JFET) and Metal Oxide Semi-conductor Field Effect Transistor (MOSFET).
General characteristics:
Unlike BJTs, FETs are voltage-controlled devices (they require essentially no current).
Extremely large input impedance
Generally lower gain than BJTs
Susceptible to dammages from static electricity (especially MOSFETs)
Easy to build in small size for use in integrated circuits (ICs)
Large manufacturing spread in FET parameters (larger than BJTs)
Brigitte VACHON
December 7, 2015
74
4.2.1
JFET
n-channel JFET
D
p-channel JFET
D
G
S
Here is how a typical n-channel JFET works:
Without any gate voltage, current flows freely

through transistor.
Brigitte VACHON
A negative voltage applied to the gate (compared to the source voltage) creates a depletion region which chokes the flow of electrons thereby limiting the current flowing
through the transistor. Gate-source always reverse-biased.
December 7, 2015
75
4.2.2
MOSFET
There are several different types of MOSFETs and going through the detailed of them all is beyond
the scope of this course. For completeness, here are some of the MOSFET symbols you may come
across.
Enhancement MOSFETs
n-channel
p-channel
D
D
G
Depletions MOSFETs
n-channel
p-channel
D
D
G
S
G
S
G
S
To give you a general idea, here is how a typical (n-channel enhancement) MOSFET works:
(source: http://volga.eng.yale.edu/index.php/main/semiconductors)
No gate voltage applied, natural distribution

of holes shown throughout the p-doped body.
Brigitte VACHON
Positive gate voltage repels all holes and actually attracts electrons (from source and/or
drain) into the channel. A current between
the Drain and the Source can now flow.
December 7, 2015
76
4.2.3
Theory
Since both JFETs and MOSFETs have similar general behaviour, for the rest of the course, well
focus more or less exclusively on JFETs, although much of the circuits we will study and build could
also be built using MOSFETs.
We will first start by defining the following voltages and currents
Brigitte VACHON
December 7, 2015
77
The current ID flowing through the transistor varies as function of the voltage difference between
the drain and the source (VDS = VD -VS).
A maximum current will flow through a JFET when the gate voltage is identical to the source voltage
(i.e. VGS = VG - VS = 0 V).
This maximum drain current (IDSS) varies for different FETs and is given in the transistors data
table . Typical values range from about 1 mA to 1 A.
Brigitte VACHON
December 7, 2015
78
The cutoff voltage (VGS(off)) is the value of VGS that makes ID zero. Typical values range from -0.5
to -10 V. This value is normally given in the data table of your transistor.
Ohmic region: When VDS is small, current flowing through the transistor is proportional to VDS.
The transistor behaves like voltage-controlled resistor.
ID =
Brigitte VACHON
IDSS
VGS(off)2
2 VGS VGS(off) VDS VDS

December 7, 2015
(4.2)
79
Active region: The current flowing through the transistor (ID) is dictated by the voltage difference
between the gate and the source (i.e. VGS), and is mostly independent of VDS. This occurs because
the channel between the drain and source is so narrow that raising VDS does not increase ID.
ID =
IDSS
VGS(off)2
VGS VGS(off)
2
(4.3)
Brigitte VACHON
December 7, 2015
80
Exercise: If VGS(off) =-4 V and IDSS =12 mA, find the value of ID when VGS is set to -2 V and
0 V. Assume the JFET is in the active region. Indicate approximately where these operating points
would like on a typical I-V characteristic curve of a FET.
Brigitte VACHON
December 7, 2015
81
4.2.4
Transconductance
Since the gate current of a FET is zero, the natural gain parameter for FETs is transconductance.
We can even say that FETs are transconductance devices, i.e. a change in input voltage results in
a change in output current.
For a FET, transconductance is defined as:
ID
gm =

VGS VDS

Differentiating Equation 4.3 we find:

IDSS
ID
=2
VGS VGS(off)
gm(VGS) =

VGS VDS
VGS(off)2

Lets define the transconductance of a FET with its gate grounded such that VGS =0:
IDSS
gm(0) = 2
VGS(off)
The general transconductance can then be written as:
1
gm(VGS) = gm(0)
VGS
VGS(off)
or using Equation 4.3 as

gm(ID) = gm(0)
Brigitte VACHON
v
u
u
u
t
December 7, 2015
ID
IDSS
82
The value of transconductance changes with ID. If you change the value of the gate-source
voltage VGS , this in turn changes the current flowing through the transistor ID which in turn affects
the value of transconductance of the transistor.
You can think of transconductance as a measure of the amplification of a transistor.
You can also equivalently think of the transconductance as being related to a small internal resistance within the drain-source channel with value 1/gm, which therefore contributes to the output
impedance of the transistor. [Note: This is analogous to the value re for BJTs]
Brigitte VACHON
December 7, 2015
83
4.2.5
Current source circuit
What is the current produced by this current source (assuming the transistor is operated in the
active region)?
But this is not a perfect current source:

Current is not completely constant (slight slope of ID vs VDS curves)
Only works for VDS larger than the cutoff voltage VGS(off).
Large variations in current IDSS due to manufacturing.
Current not adjustable
Brigitte VACHON
December 7, 2015
84
Lets add a resistor RS and see how this improves the design of our
current source:
Again, assuming the transistor is operated in the active region:
VS IDRS = 0
and therefore,
VGS = VG VS = IDRS
What we find is that the value of VGS is inversely proportional to the
current flowing through the transistor. RS provides feedback to
the circuit!
[As ID increases, VGS increases in magnitude so we go to lower spot
on ID vs VDS plane, and ID decreases - an equilibrium is established]
(source: P.Scherz and S.Monk,

Practical Electronics for Inventors)
What is the current ID once you add a load resistance ? We have two equations:
ID =
IDSS
VGS(off)2
VGS VGS(off)
ID =
2
VGS
RS
With two equations and two unknowns, you can solve for ID.
Brigitte VACHON
December 7, 2015
85
Practical example, a light dimmer:
Brigitte VACHON
December 7, 2015
86
4.2.6
Transistor biasing
Just like for BJT, to keep the operating point of a transistor in the intended region of operation, one
needs to carefully set the DC voltages of the transistors terminals.
To operate a JFET, you need to reverse-bias the gate-source junction, which requires a negative gate
voltage VG as compared to the source voltage VS .
The most common type of JFET bias circuit is called self-bias.
What is the purpose of RG?
1. Pick ID = IDSS /2 to allow maximum swings
in drain current.
2. Calculate the required VGS using Equation 4.3.
3. Calculate the required value of RS (for a given
ID and VGS ).
4. Choose RD such that VD = VDD /2 to allow
maximum swings in output voltage.
5. Arbitrarily choose RG to be large to prevent
loading previous stage of the circuit.
Brigitte VACHON
December 7, 2015
87
4.2.7
Common-source amplifier
This circuit is equivalent to the BJT common-emitter amplifier, and provides voltage gain.
Brigitte VACHON
December 7, 2015
88
To determine the Q-point of the amplifier circuit you need to find ID and VDS:
In the active region, the characteristic curve can be expressed as:
2
IDSS
ID =
VGS VGS(off)
VGS(off)2
Also, without any input signals (i.e. DC state),
VGS = IDRS
Solve for ID. Then once you have ID you can find VD, VS and VDS.
Brigitte VACHON
December 7, 2015
89
Show that the gain of a common-source amplifier circuit (that does not have a C2 capacitor) is
gmRD
Gain = 1+
g R
m
If you add a capacitor C2 to the circuit, then Gain =gmRD. Do you know why?
What is the input impedance of this amplifier circuit?
Brigitte VACHON
December 7, 2015
90
4.2.8
Source-follower amplifier
This circuit is equivalent to the bipolar emitter-follower amplifier, it provides current but no voltage
gain; the output voltage follows the input voltage (i.e. approximately unity voltage gain).
Brigitte VACHON
December 7, 2015
91
S
Show that the gain of this amplifier is Gain = R R
+1/
gm , and that indeed Vout Vin.
S
Is the output in phase or 180 degrees out-of-phase with the input signal?
What is the output impedance of this amplifer?
Brigitte VACHON
December 7, 2015
92
Chapter 5
Operational amplifiers
Operational amplifiers (op-amps) are high performance differential amplifiers packaged in an integrated device.
large (open-loop) gain (at least at low frequencies)
large input impedance
Op-amps were originally conceived for doing mathematical operations in analog computers. Their
use has been partially superceeded by digital computers but still used extensively in analog circuits.
93
(source: https://www.sparkfun.com/products/9456)
CHAPTER 5. OPERATIONAL AMPLIFIERS
A typical op-amp is made up of three types of amplifier circuits: a differential amplifier, a voltage
amplifier, and a push-pull amplifier.
Differential amplifier: Amplifies the difference between the two input signals.
Voltage amplifier: Provides additional voltage gain.
push-pull amplifier: Provides a low output impedance and a high current driving capability (i.e.
high power capability).
Brigitte VACHON
December 7, 2015
94
An op-amp is an integrated device that contains several transistors, resistors and capacitors. You
dont need to know the internal circuitry of an op-amp but heres what a typical one could look like
Brigitte VACHON
December 7, 2015
95
5.1
5.1. CHARACTERISTICS OF IDEAL OP-AMPS
Characteristics of ideal op-amps
The output of an ideal op-amp is the difference between the non-inverting input (V+) and the
inverting input (V), amplified by the open-loop gain Ao.
Vout = Ao(V+ V)
Infinite voltage gain

Infinite bandwidth
Infinite input impedance
In real life, the output is however limited by the

supplied positive and negative saturation voltages (+Vs, Vs).
Zero output impedance

Brigitte VACHON
December 7, 2015
96
5.2
5.2. COMPARATOR
Comparator
A comparator circuit compares an input signal to a reference voltage, and indicates by a two-state
output whether the input signal is lower or higher than the reference voltage.
This kind of circuit is useful to know when an input analog signal is larger than some predefined
value (i.e. when a signal passes a threshold).
Brigitte VACHON
December 7, 2015
97
5.3
5.3. FEEDBACK
Feedback
Feedback: Take some of the output and mix it with the input.
Positive feedback: Add some fraction of the output to the input. Usually not desirable (ex. microphone in front of a speaker), unless you want to make an oscillator circuit.
Negative feedback: Subtract a fraction of the output from the input. Obtain more predictable and
stable operation.
Feedback can make an amplifier circuit less dependent on actual features of the amplifier
(Harold S. Black, 1928) [See Appendix E]
Op-amps almost always used with
feedback, and mostly using negative
feedback.
Lowers gain of the circuit, but...
Improves linearity (gain stability)
Increases bandwith of amplifier
Can increase input impedance and
lower output impedance of a circuit
Brigitte VACHON
December 7, 2015
98
5.3.1
5.3. FEEDBACK
Effect of negative feedback on voltage gain
Using negative feedback, the output

voltage is
Vout = A0(Vin f Vout)
Therefore the closed-loop voltage gain,
ACL, is
ACL =
A0
Vout
=
Vin
1 + f A0
Negative feedback provides stabilization of the voltage gain of an amplifier against dependence on
parameters such as temperature, frequency, etc.
Heres an illustration of this gain stabilization. Lets assume a feedback network that feeds back a
fraction of f=0.1 of the output voltage back to one of the amplifier input. The the closed-loop gain
of the circuit with feedback corresponding to different values of the open-loop gain is shown in the
table below:
A0 ACL = A0/(1 + f A0)
1,000
90.9
5,000
98.3
10,000
99.0
20,000
99.5
Brigitte VACHON
December 7, 2015
99
5.3.2
5.3. FEEDBACK
Effect of negative feedback on input impedance
Assume there exists a small voltage difference between the

two inputs, which implies a non-zero input current and an
input impedance which is not infinite. In this case,
Vin = Vd + Vf = Vd + f Vout
We have seen that with negative feedback
ACL =
A0
Vout
=
Vin
1 + f A0
and therefore,
(source: T.Floyd, Electronic Devices (Conventional Current

Version))
A0
Vin = Vd + f
Vin
1 + f A0
Now replacing Vd by IinZin (where Zin corresponds to the impedance of the amplifier), we find
Zin(CL) =
Vin
= (1 + f A0)Zin
Iin
Therefore, the use of negative feedback increases the effective input impedance of the circuit (which
is typically a good thing!).
Brigitte VACHON
December 7, 2015
100
5.3.3
5.3. FEEDBACK
Effect of negative feedback on output impedance
Using Kirchhoffs voltage law

Vout = A0Vd ZoutIout = A0(Vin Vf ) ZoutIout
and,
Vout = A0(Vin f Vout) ZoutIout
Re-arranging the terms,
Vout
(source: T.Floyd, Electronic Devices (Conventional Current
A0Vin ZoutIout
=
(1 + A0f )
Version))
The effective output impedance of the circuit is defined as

Zout(CL)
Vout
Zout
=
=

Iout Vin (1 + A0f )

Therefore, the use of negative feedback decreases the effective output impedance of the circuit (which
is typically a good thing!).
Brigitte VACHON
December 7, 2015
101
5.4
5.4. OP-AMP GOLDEN RULES
Op-amp Golden rules
Simple but effective approximation, useful for circuit analysis:

1. For an ideal op-amp, the inputs draw (or source) no current.
2. With negative feedback, op-amp will do whatever it can to always keep the voltages at the input
terminals identical, i.e. feedback network is automatically fed back as much current/voltage as
necessary to keep the voltages at the input terminals identical.
5.5
Buffer
Looking at the circuit you can see that V = Vout.
By the Golden rule (2), V+ = V and therefore
Vin = Vout.
The closed-loop gain of this amplifier configuration is therefore Gain = VVin = 1

out
The use of feedback increases the input impedance and decreases the output impedance of the circuit
(as compared to the open-loop input/output impedance of op-amp).
Brigitte VACHON
December 7, 2015
102
5.6
5.6. INVERTING AMPLIFIER
Inverting amplifier
What is the closed-loop gain of this op-amp
circuit ?
By golden rule (2), V = 0 V (virtual
ground)
Voltage across R2 is -Vout.
Voltage across R1 is Vin.
By golden rule (1), no current flows into
in = I = Vout
the op-amp I1 = V
2
R
R
1
2
The gain is therefore, Gain = R
R1
What about the input impedance of this

circuit?
Circuit looks like a resistor (R1) to ground.
Brigitte VACHON
Therefore Zin= R1.
December 7, 2015
103
5.7
5.7. NON-INVERTING AMPLIFIER
Non-inverting amplifier
What is the closed-loop gain of this op-amp

circuit ?
What about the input impedance of this circuit?

Brigitte VACHON
December 7, 2015
104
5.8
5.8. SUMMING AMPLIFIER
Summing amplifier
What is the output Vout of this op-amp circuit ?
Practical examples:
Audio mixer
D/A conversion
Adding DC offset to AC signal
Note that the input voltages (V1, V2, V3) can be
positive or negative, such that this circuit can
also do subtraction.
Brigitte VACHON
December 7, 2015
105
5.9
5.9. INTEGRATOR
Integrator
What is the output Vout of this op-amp circuit ?
By rule (2), V = 0 V , thus IC = C dVdtout
Also, we have IR =
Vin
R
By rule (1), IR = IC
Putting everything together: Vin = RIC = RC dVdtout
[Recall: For a capacitor Q=CV and
dV
I = dQ
dt = C dt ]
Integrating to solve for Vout,

Vout
Practical examples:
analogue computer
charge amplifier
Brigitte VACHON
1
=
RC
Vin(t)dt + const
Main problem: In real life, the input is not exactly zero

(ex. stray currents), and as a result, the circuit integrates until the output reaches the maximum/minimum
possible voltage given to the opamp.
Solution: Put a large resistor (ex. O(1M)) in parallel
with C to limit the integration time (and have negligible
effect on output waveform).
December 7, 2015
106
5.10
5.10. CURRENT SOURCE
Current source
What is the current flowing through the load?
Brigitte VACHON
December 7, 2015
107
5.11
5.11. CHARACTERISTICS OF REAL OP-AMPS
Characteristics of real op-amps
Ao 104 to 106
Zin 106 1012
Zout 10 1000
Input current nA to pA
Ao =
Zin =
Zout = 0
Input current = 0
Brigitte VACHON
December 7, 2015
108
Input offset voltage: For V+ V = 0, Vout6= 0. Amount of voltage that needs to be applied to
one of the inputs to zero the output.
Input offset current: Difference between I+ and I.
Common-mode input range: Range of input voltages for which the op-amp is guaranteed to
work, typically given by range of supplied external voltages or fraction thereof. If range not respected, possible results include saturation, gain sign change, circuit dammage, etc.
Differential input voltage range: Limits on voltage difference between the two inputs. If limit
exceeded (ex. > 0.5 V), can cause dammage.
Slew rate: Maximum rate of change of the output voltage (i.e. cant change output voltage instantaneously). Units: [V/s].
Slew rate becomes critical when dealing with high frequency input signals. (ex. op-amp 741 has 0.5
V/s, op-amp HA2539 has 600 V/s.
Frequency response: The gain of an op-amp depends on frequency of the input signal. Roll-off
as function of frequency deliberately designed into op-amp to avoid oscillation at high frequencies.
Brigitte VACHON
December 7, 2015
109
Frequency at which gain=1 (0 dB) is defined as

unity gain frequency, fT .
Frequency at which
power drops by 1/2 (voltage
gain drops by 1/ 2) is defined as the 3 dB point
or breakover frequency fB .
With negative feedback, the flat part of
the curve is extended, i.e. closed-loop
gain is lower but stable over larger
frequency range.
Frequency roll-off typically -6 dB/octave,

i.e amplitude halves for each doubling of frequency
(-20 dB per decade)
For typical op-amps, roll-off is fixed, such that

product of gain and bandwidth is a constant. For
example,
fB(OL)A(OL) = fB(CL)A(CL)
and,
Decibel definition:

A1
Amplitude: dB = 20 log10 A2

P1
Power: dB = 10 log10 P2 since P A2
Brigitte VACHON
fB(CL)A(CL) = fT 1
Thus, gain-bandwidth product is always equal to
fT = fB(CL)A(CL)
December 7, 2015
110
5.12
5.12. OSCILLATORS
Oscillators
Device that produces a repetitive waveform that can be used to drive other circuits. There is an
oscillator in essentially all electronic devices.
In digital electronics, oscillators producing square waves are called clocks and are used to push
bits of information through the digital circuit.
Heres an example of a simple square wave RC relaxation oscillator. Note that this is the first (and
only time) well use positive feedback with an op-amp!
Brigitte VACHON
December 7, 2015
111
5.12. OSCILLATORS
The non-inverting input V+ defines the threshold

voltage VT at which the op-amp flips from positive
to negative maximum voltage values.
If V1 is below VT , the op-amp outputs +15 V,
charging up the capacitor through R1 with a
time constant R1C.
When the voltage across the capacitor becomes
higher than VT = +7.5 V , then the op-amp
outputs -15 V, allowing the capacitor to discharge
and setting VT = 7.5 V
And the cycle repeats!
Brigitte VACHON
December 7, 2015
112
Chapter 6
Digital electronics
Digital electronics used in most modern data processing.
Less sensitive to noise and circuit parameters (ex. temperature dependence of components).
Conceptually easier than analog electronics; you can build circuits like LEGO construction.
113
CHAPTER 6. DIGITAL ELECTRONICS
In digital electronics, unlike analog electronics, there are only two voltage states present at any point
within the circuit: high or low.
These two states can represent one bit of a number, on/off state, true/false, a signal has passed a
given threshold, etc.
Logic levels (high/low) usually given by voltage (but can be current), and usually positive true
(but can also be negative true).
Brigitte VACHON
December 7, 2015
114
There are different conventions of what a high/low signal is, this depends on the logic family you
use. For example,
0 0 to 0.8 V
1 2 to 5.0 V
*TTL
(transistor-transistor logic)
0 0.1 to 1.0 V
1 3.5 to 5.0 V
*CMOS
(Complementary MOSFET)
NIM
(Nuclear instrumentation module logic)
ECL
(emitter coupled logic)
0 0 V
1 -0.8 to -1.0 V (current through 50 )
0 -1.8 V
1 -0.9 V
(source: Floyd, Digital Fundamentals)
Brigitte VACHON
December 7, 2015
115
6.1
Electronics for Scientists 6.1.
NUMBER CODES IN DIGITAL ELECTRONICS
Number codes in digital electronics
In everyday life, we use a decimal representation of numbers. A number is represented by successive

powers of 10. For example,
123.410 = 1 102 + 2 101 + 3 100 + 4 101
Since information in digital electronics is encoded as two distinct voltage states, other types of
number representation are more natural than the decimal system.
6.1.1
Binary
There are 10 kinds of people in the world - those that use binary and those that dont
Binary numbers are composed of only two binary digits, 0 and 1, which are called bits (for Binary
digIT).
A binary number of 8 bits is called a byte.
A number in the binary representation is expressed as successive powers of 2. For example,
11012 = 1 23 + 1 22 + 0 21 + 1 20 = 8 + 4 + 1 = 1310
The highest-order bit (leftmost bit) is called the most significant bit (MSB). The lowest-order bit
(rightmost bit) is called the least significant bit (LSB).
Brigitte VACHON
December 7, 2015
116
Exercise: Write the decimal number 3110 in binary form.
6.1.2
Hexadecimal
Another useful number representation in digital electronics. Here, there are 16 allowable digits
(0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F) and numbers are expressed as successive powers of 16. For example,
2D516 = 2 162 + D(= 1310) 161 + 5 160 =
(Decimal)
The hexadecimal number representation is particularly convenient since it is straight forward to

translate to/from a binary number representation. Each hexadecimal digit corresponds to 4 bits.
Brigitte VACHON
December 7, 2015
117
Microprocessors systems nowdays typically use 8-bit, 16-bit, 32-bit or 64-bit words. Large binary
words can therefore easily be broken down into a small number of hexadecimal digits of more
manageable size. For example, a 32-bit word can be simply expressed as 8 hex digits.
6.1.3
Binary Coded Decimal (BCD)
System where each decimal digit is separately encoded into a 4-bit binary number. For example,
12310 = |0001
0010
0011
{z }
| {z }
| {z }
1
Brigitte VACHON
December 7, 2015
118
This is somewhat of a wasteful number representation in terms of resource usage since it uses 4
bits for each digit even though 4 bits can cover 0-15 values.
Having said that, this is a very convenient representation for passing on information to a digital
(decimal) number display.
A seven-segment display used to display a decimal number.

(source: Wikipedia)
A 4-bit BCD code driving a seven-segment display.

(source: http://www.electronics-tutorials.ws/combination/comb_6.html)
Brigitte VACHON
December 7, 2015
119
Brigitte VACHON
December 7, 2015
120
6.1.4
Gray code
Binary number system where only one bit changes between

successive numerical values. This avoids, for example,
undefined transitional state when more than one bits
change not perfectly simultaneously.
This is a good system that prevents errors and its property
can easily be used for error correction. This is a natural
numerical code to use for counting. Also use in Karnaugh
maps (more on this later).
Design rule: Change the single LSB that produced a new
state.
6.1.5
Decimal
0
1
2
3
4
5
6
7
8
9
BCD
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001
Gray
0000
0001
0011
0010
0110
0111
0101
0100
1100
1101
ASCII
ASCII stands for American Standard Code for Information Interchange. This is a standard alphanumeric system used to transmit information (numbers, letters, symbols, commands) between
computers and peripherals.
The ASCII code consists of 128 unique 7-bit words (an additional bit is typically added to make it
compatible with 8-bit systems). Here are some example of ASCII characters:
Brigitte VACHON
December 7, 2015
121
Brigitte VACHON
December 7, 2015
122
6.2
Electronics for Scientists6.2.
PARALLEL AND SERIAL BIT TRANSMISSION
Parallel and Serial Bit Transmission
Clock: Most digital circuits require a clock signal (i.e. series of high and low pulses at a fixed
precise frequency). As we will see later, the clock signal is used to push binary digits (bits) through
the digital circuit.
Serial bit transmission: Bits are transmitted on the same electrical conductor, one after the
other. One bit is sent at each clock cycle (either on the rising of falling edge of the clock, depending
on the design of the digital circuit). Here, you need to know ahead of time how many bits are being
sent or alternatively how to recognize the start/end of a number. [Ex. transmission over USB,
FireWire, ethernet cable, etc.]
Example: Raw data from ATLAS muon chamber electronics
frame cnt
z
}|
header
{z
}|
data
board id
{z
}|
{z
}|
data
{z
}|
data
{z
}|
data
{z
}|
end of data
{ z
}|
00000018 004c4e42 00000000 cc7caaf9 d08ba3de d490443f d88b4386 fafafafa

00000000004c4e420000000021ff4b7524f7c4ec2912e39e2cb5521c30b9b4823492e8c4388b0477fafafafa
00000001004c4e4200000000b4b983bfb8b284d9bcb0a35afafafafa
00000001004c4e420000000034ea43fd38ee454d3ceb6b0440f4c6ba44fe641dfafafafa
00000002004c4e42000000002cdbc45c30d306c734d5c4bafafafafa
........
Parallel bit transmission: Each bit is sent simultaneously on a separate electrical conductors.
[Give example]
Brigitte VACHON
December 7, 2015
123
6.3. LOGIC FUNCTIONS
6.3
Logic functions
Combine digital inputs from two or more sources and produce a digital output that depends on the
state of those inputs.
Use devices (typically ICs) called gates that give combinational logic.
This allows you to perform Boolean algebra.
Brigitte VACHON
December 7, 2015
124
6.3. LOGIC FUNCTIONS
Brigitte VACHON
December 7, 2015
125
6.3.1
Electronics for 6.4.

Scientists
COMBINATIONAL
LOGIC AND BOOLEAN ALGEBRA
Example: Enable/Disable control
Here is an example use of one single logic gate. This gate is used to control the passage of an
incoming logic signal (ex. a clock signal).
6.4
Combinational Logic and Boolean Algebra
Combinational logic consists in combining logic gates together to perform complex functions.
Use rules and symbols of boolean algebra (i.e. algebra defined with variables that can either be
true or false) to design complex combinational logic circuits.
Brigitte VACHON
December 7, 2015
126

Scientists
COMBINATIONAL
Brigitte VACHON
December 7, 2015
127

Scientists
COMBINATIONAL
Identities 18 and 19 are called De Morgans

Theorem (extremely useful in combinational logic circuit design!)
Bubble pushing: Trick to design equivalent

logic functions (based on De Morgans theorem)
Change an AND gate into and OR gate
Add inversion bubbles on inputs/output
where there were originally none and
vice-versa
Brigitte VACHON
December 7, 2015
128

Scientists
COMBINATIONAL
Example: Find Boolean expression for the following circuit and use logic identities to implement the
same function with a smaller number of logic gates.
Brigitte VACHON
December 7, 2015
129
6.4.1

Scientists
COMBINATIONAL
NAND and NOR gate universality
Brigitte VACHON
December 7, 2015
130
6.5
6.5. KARNAUGH MAPS
Karnaugh Maps
We have seen how a Boolean expression can be simplified to reduce the number of logic gates that
are used to implement a digital circuit. This can be done quite quickly and easily for logic functions
with up to two inputs.
For more complex logic functions, simplifying by-hand a complex Boolean expression is time consuming and not necessarily an easy task.
For circuits with three to four inputs, one can use Karnaugh maps to convert any truth table
into the simplest possible Boolean expression for your complex digital logic function.
For circuits with more than four inputs, one will typically use specialized computer programs
that take in a truth table or Boolean expression and convert it into a simpler Boolean expression
that can be implemented with logic gates.
To see how to construct and use a Karnaugh map, lets look at a specific example.
Brigitte VACHON
December 7, 2015
131
6.5. KARNAUGH MAPS
Example: Design a digital circuit for a car alarm. The car alarm should be turned on only when the
key is in the ignition and either the drivers door is open or his/her seatbelt is not fastened.
A: Door (0=open, 1=closed)
B: Key (0=not in ignition, 1=in ignition)
C: Belt (0=unfastened, 1=fastened)
Q: Alarm (0=off, 1=on)
1. Write truth table describing the logic function you want to implement.
A
0
0
0
0
1
1
1
1
Brigitte VACHON
B
0
0
1
1
0
0
1
1
C
0
1
0
1
0
1
0
1
December 7, 2015
Q
0
0
1
1
0
0
1
0
132
6.5. KARNAUGH MAPS
2. Translate truth table into Karnaugh map.

Arrange input variables along two axes
Write two variables combinations in such a way that only one digit changes at a time.
A
0
0
0
0
1
1
1
1
B
0
0
1
1
0
0
1
1
C
0
1
0
1
0
1
0
1
Q
0
0
1
1
0
0
1
0
BC
00 01 11 10
0 0 0 1 1
A
1 0 0 0 1
3. Identify adjacent cells containing 1s in groups of 1, 2, 4, 8 or 16 cells (i.e powers of 2s).
Brigitte VACHON
December 7, 2015
133
6.5. KARNAUGH MAPS
4. For each group, identify the variables that are constant or a common denominator within the
group, i.e. the variables that define the outcome of all cells of the group.
AB
BC
5. Just make an OR of all expression describing all the groups you have identified. Voila!
+ B C
Q = AB
A
Q
B
C
Brigitte VACHON
December 7, 2015
134
6.5. KARNAUGH MAPS
Exercise: Design a simple logic circuit that implements the following truth table.
A
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
B
0
0
0
0
1
1
1
1
0
0
0
0
1
1
1
1
C
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
D
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
Q
0
1
0
1
0
1
1
1
0
1
0
1
0
1
0
1
CD
00 01 11 10
00
01
AB
11
10
Answer:
Brigitte VACHON
December 7, 2015
135
6.5. KARNAUGH MAPS
Some more notes about Karnaugh maps:

Larger groups give simpler logic.
To form groups of identical cells, you can wrap around at top/bottom and left/right edges of
the table.
If it is more convenient, you can alternatively group 0s. In this case you would need to find the
logic for being member of these groups, make an OR of all the groups expressions, and apply
inversion to the final resulting boolean expression.
Sometimes you can have an X instead of a 0 or 1 in a Karnaugh map (X = dont care). Use
X as 0 or 1 to achieve largest grouping.
Remember that best practical solution may not come from a Karnaugh map (ex. if you already
have spare gates on a chip already in use).
Brigitte VACHON
December 7, 2015
136
6.6
6.6. SEQUENTIAL LOGIC
Sequential Logic
Up to now, our digital electronics experience has only involved combinational logic gates which have
the property of immediately changing their output as a results of a change in input.
In order to provide memory to a circuit, the circuit needs to be able to latch (i.e. hold-on) to data
(i.e. bits) at a particular time.
Digital electronics circuits which have memory fall into the category of sequential logic. The word
sequential refers to the fact that to store/retrieve data involves a sequence of steps.
A clock signal is often used to push bits of information (i.e. 1s and 0s) through a sequential
circuit, and also provides an absolute time reference for all the sequences of steps involved in the
digital circuit.
Asynchronous circuit element: Output changes as soon as the input values changes.
Synchronous circuit element: Output changes as a result of changes on its inputs, only in synch
with a clock signal.
Brigitte VACHON
December 7, 2015
137
6.6.1
SR flip-flops
The most basic building block of all circuits storing data is a Set-Reset flip-flop, also known as SR
latch.
A SR flip-flop is a two-input one-bit memory bistable device.
Heres a simple flip-flop implemented using two NOR gates.
If input labeled S goes high, this sets the device output to 1.

If input labeled R goes high, this resets the device output to 0
If both S=R=0, the circuit holds the previous output state (it has memory!).
Brigitte VACHON
December 7, 2015
138
Transition table
condition
S R Q Q
Brigitte VACHON
December 7, 2015
139
condition
S R Q Q
Brigitte VACHON
December 7, 2015
140
Can also implement simple SR flip-flop using two NAND gates.

Q Q
condition
S R
Brigitte VACHON
December 7, 2015
141
Q Q condition
S R
Brigitte VACHON
December 7, 2015
142
Logic device where putting set input HIGH gives an output Q=1 is called active high.
Example: SR flip-flop implemented using NOR gates.
Logic device where putting set input LOW gives an output Q=1 is called active low.
Example: SR flip-flop implemented using NAND gates.
flip-flops
Logic symbols for an SR and SR
(source: T.L. Floyd, Digital Fundamentals)
Brigitte VACHON
December 7, 2015
143
Example use of flip-flop: Switch debouncer circuit.
Brigitte VACHON
December 7, 2015
144
6.6.2
Level-triggered (or gated) SR flip-flop
This is the first synchronous device we will study. The inputs of the flip-flop are enabled/disabled
by an external control signal (ex. clock). Only when the clock is HIGH are the inputs sampled, and
the resulting output is held unchanged while the clock is LOW.

Brigitte VACHON
December 7, 2015
145
6.6.3
Edge-triggered SR flip-flops
An edge-triggered SR flip-flop only samples the inputs during the edge of a clock signal. The circuit
can be designed to respond either to the positive () or the negative () edge of the clock signal.
pulse
transition
detector
Example of a pulse transition detector (rising edge)
Brigitte VACHON
December 7, 2015
146
CLK
pulse
transition
detector
S
0
1
0
1
R Qn
Qn
mode
0 Qn1 Qn1
hold
0
1
0
SET
1
0
1
RESET
1
1
1
indeterminate
Brigitte VACHON
December 7, 2015
147
6.6.4
Pulsed-triggered SR flip-flops (Master-Slave SR flip-flop)
A two-stage device designed to avoid race conditions (see later).
Master enabled by CLK, Slave enabled by CLK.
CLK HIGH: Master produces Y,Y according to inputs S,R. Slave keeps old values of Q,Q.
are updated according to values of Y,Y.
CLK LOW: States Y,Y remain the same but Q,Q
Brigitte VACHON
December 7, 2015
148
6.6.5
Symbolic representation
6.6.6
D-type flip-flop (Gated D latch)
A D-type flip-flop (D for Data) is a single input device. The circuit of a D-type flip-flop consists
in a SR flip-flop where you tie the S and R inputs together with an inverter such that R = S always.
This configuration avoids ever having the flip-flop be in an indetermined state.
This type of flip-flop also does not have a hold state, it only has a set and reset condition.
Brigitte VACHON
December 7, 2015
149
Brigitte VACHON
December 7, 2015
150
Just as weve seen before for a basic SR flip-flop, D-type flip-flop also comes with the following
variations:
Level-triggered (clocked) D-type flip-flop
Edge-triggered D-type flip-flop
Pulse-triggered D-type flip-flop (Master-Slave D-type flip-flop)

Brigitte VACHON
December 7, 2015
151
Example use of D-type flip-flop:

Brigitte VACHON
December 7, 2015
152
6.6.7
JK flip-flop
A (edge-triggered) JK flip-flop is similar to a SR flip-flop with SJ, and RK.

set: (J=1, K=0)
reset: (J=0, K=1)
hold: (J=0, K=0)
The main difference is that it also has a condition:
toggle: (J=1, K=1)

Brigitte VACHON
December 7, 2015
153
There also exists a pulse-triggered JK flip-flop

(Master-Slave JK flip-flop)

Brigitte VACHON
December 7, 2015
154
Example use of JK flip-flop
A JK flip-flop used to divide by two

the frequency of a clock signal.
A JK flip-flop used to divide by four

the frequency of a clock signal.
Brigitte VACHON
December 7, 2015
155
6.6.8
Summary of flip-flop types
Since it can easily get confusing, lets now summarize the different types of flip-flops we have seen
so far:
SR flip-flop: Two inputs. Can be in set, reset, hold or not used states.
asynchronous or default
Level-triggered (clocked or gated)
Edge-triggered
Pulsed-triggered (Master-Slave)
D-type flip-flop: Single input. No indeterminate nor hold states, only set and reset
states possible.
Level-triggered (clocked or gated)
Edge-triggered
Pulse-triggered (Master-Slave)
JK flip-flop: Two inputs. Allowed states are set, reset, hold and toggle.
Edge-triggered
Pulse-triggered (Master-Slave)
Brigitte VACHON
December 7, 2015
156
6.6.9
Timing considerations
In the design of digital electronics circuits, one needs to avoid possible synchronization problems
that could results in an undetermined outcome. This is called a race condition.
To avoid race conditions, one often builds in a hold time where inputs have to stay in a state for a
minimum time after the clock to define the output.
Race conditions can arise due to:
propagation time of signals
delays in logic function
Brigitte VACHON
December 7, 2015
157
Brigitte VACHON
December 7, 2015
158
6.6.10
Asynchronous (ripple) counters
Can link together JK flip-flops to make a circuit that counts how many clock pulses have occured.
Use the data output from the previous element to clock the next element.
Recall that J=K=1 corresponds to toggle mode.
This is called a ripple counter because the effect of the clock signal moves down the chain of
JK flip-flops.
The example circuit on the next slide is called a modulus-16 counter because it returns zeros
after 16 clock pulses. To make a modulus-2n counter, you need n flip-flops.
Brigitte VACHON
December 7, 2015
159
Brigitte VACHON
December 7, 2015
160
If you need a modulus-N counter where N 6= 2n, you can build in some logic that will reset all the
flip-flops to zero when you get to that number.
Brigitte VACHON
December 7, 2015
161
Brigitte VACHON
December 7, 2015
162
6.6.11
Synchronous counters
The last output stages of an asynchronous (ripple) counters take time to react to the original
incoming clock signal due to propagation delays.
Can instead construct a synchronous counter where all flip-flops are driven at the same time
by a common clock signal.
Use AND gates to change flip-flop states from hold to toggle.
Brigitte VACHON
December 7, 2015
163
Brigitte VACHON
December 7, 2015
164
6.6.12
Memory Registers
Simple circuit designed to hold a state even if input changes.

Use D-type flip-flop to store the bit information you want to hold on to.
Brigitte VACHON
December 7, 2015
165
6.6.13
Shift registers
A shift register is made by connecting a series of (D-type) flip-flops such that each Q output
drives the next D input, and all clock inputs are driven simultaneously.
The idea is that at each clock pulse, the pattern of 0s and 1s in the register shifts to the right
(with data D entering from the left).
Shift registers are useful for conversion of parallel data to serial data and vice-versa, and can
also be used as a way to queue information.
Serial-in/serial-out shift register
CLK Q0 Q1 Q2 Q3
1
1 0 0 0
2
0 1 0 0
3
0 0 1 0
4
1 0 0 1
5
1 1 0 0
0 1 1 0
6
7
1 0 1 1
.....
Brigitte VACHON
December 7, 2015
166
Serial-in/parallel-out shift register
Brigitte VACHON
December 7, 2015
167
Parallel-in/serial-out shift register
This particular use of a shift register requires additional logic to load the parallel data into the
D-type flip-flops before enabling the serial reading out of the shift register.
Brigitte VACHON
December 7, 2015
168
Chapter 7
Analog/Digital interfacing
Almost any electronic device relies on the ability to convert an analog signal into a digital one, and
vice-versa. [Give examples of where ADC and DAC are used in everyday life.]
Typical digital signal processing system
169
CHAPTER 7. ANALOG/DIGITAL INTERFACING Electronics for Scientists
7.1
7.1. SAMPLING
Sampling
Process of taking a number of discrete values at points along the waveform. (the more samples you
take the more accurately you can define a waveform).
Brigitte VACHON
December 7, 2015
170
CHAPTER 7. ANALOG/DIGITAL INTERFACING Electronics for Scientists 7.2. NYQUIST(-SHANNON) SAMPLING THEOREM
7.2
Nyquist(-Shannon) sampling theorem
[A qualitative explanation of the sampling theorem:

http://www.youtube.com/watch?v=PlVnW2kv4tE]
The Nyquist-Shannon sampling theorem provides a prescription for the minimum sampling rate
needed for the sampled function to represent correctly the analog signal (i.e. no information loss and
original signal can be reconstructed completely).
Qualitatively:
The sampling frequency should be at least twice the highest frequency contained in the signal.
Mathematically:
fs 2fc
fs = sampling frequency
fc = highest frequency contained in the signal
Note that Nyquist sampling theorem only applies to bandlimited signals (signals with a finite maximum frequency). In real life, due to bandwidth limits of microphones, instruments, amplifiers, etc.
an analog pulse from such instruments will have a maximum frequency in the Fourier frequency
domain, and thus the Nyquist sampling theorem applies.
The Nyquist frequency is defined to be half the chosen sampling rate (whether or not your chosen
sampling rate satisfies the Nyquist-Shannon sampling criterion defined above).
Brigitte VACHON
December 7, 2015
171
7.3
7.3. ALIASING
Aliasing
Aliasing: Frequency not present in an original signal but that nevertheless could match the sampled
points. This occurs when a signal is undersampled.
Amplitude (dB)
falias(N ) = |f N fs|
fsignal =
900 Hz
fsamplingl =
1000 Hz
(source: http:
falias1 =
100 Hz
//www.dataq.com/blog/faqs/really-need-know-sample-rate/)
falias2 =
1100 Hz
falias3 =
2100 Hz
falias3 =
3100 Hz
f(Hz)
A 900 Hz waveform (black) produces an

aliased, 100 Hz waveform (gray) when
under-sampled at 1000 Hz.
Example: Wagon wheel effect in a film:

http://en.wikipedia.org/wiki/Wagon-wheel_effect
Example: Audio aliasing:
https://en.wikipedia.org/wiki/Aliasing
Brigitte VACHON
December 7, 2015
172
7.3. ALIASING
You can experiment with the phenomena of aliasing using an applet at http://www2.egr.uh.
edu/~glover/applets/Sampling/Sampling.html
Left: No aliasing
Right: Aliasing
Note that when you go from digital to analog conversion, the analog signal is reconstructed using
only frequencies less than the Nyquist freqency which is defined as half the sampling rate.
Brigitte VACHON
December 7, 2015
173
7.3. ALIASING
To avoid undesirable effects of aliasing, you can oversample your analog signal.
Example: Audio CDs have a sampling rate of 44100 Hz. The Nyquist frequency is therefore 22050 Hz.
If recorded sounds have frequency components higher than 22050 Hz, the aliased frequency images
of the sound will overlap partly with the original sound frequency spectrum, thereby distorting the
output signal produced from the sampled version of the original audio signal.
One way to avoid this distortion is by oversampling the original audio signal at much higher sample
rate than the range of human hearing (20 kHz) demands, thereby ensuring that any aliasing effects
occur at supersonic frequencies (frequencies above those of audible sound for humans). For example,
using a 64x oversampling converter, means that you are sampling an audio signal at 44.1 kHz x 64
2.8 MHz (!). The Nyquist frequency here would therefore be 1.4 MHz, a much higher frequency
than what can actually be recorded by a microphone.
Brigitte VACHON
December 7, 2015
174
7.3.1
7.3. ALIASING
Anti-aliasing filter
Note that the Nyquist-Shannon criteria is to have a sampling rate twice the highest frequency actually in the signal, and not the highest frequency of interest.
One might decide that they dont need the higher frequencies in an original analog signal. These
higher frequencies should be removed prior to signal sampling to avoid aliasing. The filters that do
that are called anti-aliasing filters.
Example: Human speech has components about 10 kHz but it is still easy to understand speech
only considering frequencies of the voice less than 3.4 kHz. Therefore, you can filter out frequency
components of the speech above 3.4 kHz and then sample the resulting analog signal at a rate equal
or greater than 3.4 kHz x 2 = 6.8 kHz. This would satisfy the Nyquist sampling theorem and ensure
that your original (filtered) analog input can be reconstructed completely and without any aliasing.
Brigitte VACHON
December 7, 2015
175
7.4
7.4. SAMPLE AND HOLD
Sample and Hold
After filtering and sampling, the sampled voltage levels must be kept constant until the next sample
occurs. This step in the signal processing is important to allow the analog-to-digital converter (ADC)
circuit to have time to process the sampled value.
Brigitte VACHON
December 7, 2015
176
7.5
7.5. ANALOGUE-TO-DIGITAL CONVERSION
Analogue-to-Digital Conversion
Converting an analog signal into a series of binary number.
Brigitte VACHON
December 7, 2015
177
7.5. ANALOGUE-TO-DIGITAL CONVERSION
Ideal transfer function for an ADC
(source: Texas Instruments, Understanding Data Converters, Application Report, SLAA013, 1995, www.ti.com/lit/an/slaa013/slaa013.pdf)
Brigitte VACHON
December 7, 2015
178
7.6
7.6. RESOLUTION
Resolution
Resolution: Step size or quantization levels. The graininess of your digital representation.
The previous example shows that a 4-bit converter does not have a sufficient resolution to make the
analog signal appear as a continous (smooth) function.
To improve the resolution of the signal conversion, you use a larger bit-number, ex. 6-bit, 8-bit,
10-bit, etc.
Example: If your converter has a resolution of 8-bit, then this defines 28 = 256 analog steps (quantization levels). Now if your converter uses a maximum voltage scale of 0 to 15 V, then each analog
step is of (15 V / 256) = 0.058 V
Using high resolution converter is more expensive.
The cost of reducing the reference voltage is a loss of input dynamic range.
Brigitte VACHON
December 7, 2015
179

7.7. ANALOG TO DIGITAL CONVERSION METHODS
7.7
Analog to Digital Conversion Methods
Converting an analog signal into a series of binary number.

There are several different kinds of methods use to do this A/D conversion, well only see three.
7.7.1
Staircase ADC
Simple approach but slow

Need to start digitization process from zero for each conversion
Rate of ramp-up limited by settling time of DAC and comparator.
n-bit conversion could take up to 2n settling time where settling times of a few microseconds are
typical.
Brigitte VACHON
December 7, 2015
180

Digital
output
Brigitte VACHON
December 7, 2015
181

7.7.2
Successive Approximation ADC
The value of each bit of a binary output is found one at a time, starting with the MSB first.
1. Start conversion by setting MSB (D7) bit high and all other lines low.
2. DAC returns half of maximum dynamic scale (+5 V here)
3. Comparator compares DAC value with input analog voltage
4. If analog input greater than +5 V, SAR keeps D7 line high, otherwise puts D7 line low.
Brigitte VACHON
December 7, 2015
182

5. At next clock pulse, next bit (D6) is tried (i.e. put high)
6. etc.
Note that this is a faster (10 to 300 s) search algorithm than with the counter (staircase) method
above.
What happens if the analog input changes during A/D conversion? Errors in the ADC output! (i.e.
you would not get a binary number representative of the analog input value at the beginning of the
conversion process)
7.7.3
Parallel-Encoded (Flash) ADC
Fastest ADC since it converts analog signal in a single step.

Can easily do 1 Giga-samples per second.
Input analog signal fed simultaneously to one input of each comparator.
Other comparator input connected to n equally spaced reference voltages.
Brigitte VACHON
December 7, 2015
183

The priority encoder is enabled by a pulse on the EN input. The frequency of the enabled pulse
consists of the sampling frequency.
The (octal-to-binary) priority encoder transforms the value of the input lines interpreted as an octal
input into a 3-bit binary output.
Brigitte VACHON
December 7, 2015
184
7.8
7.8. DIGITAL TO ANALOG CONVERSION
Digital to Analog Conversion
Converting series of binary numbers into analog voltages.
Note that one will usually use a low-pass filter at the output of the DAC in order to smooth off the
square edges due to the quantization of the digital input. This is called a reconstruction filter.
Brigitte VACHON
December 7, 2015
185
Ideal transfer function for a DAC
(source: Texas Instruments, Understanding Data Converters, Application Report, SLAA013, 1995, www.ti.com/lit/an/slaa013/slaa013.pdf)
Brigitte VACHON
December 7, 2015
186
7.8.1
Binary weighted resistor method
Switches are usually FETs

Each binary word of data (n bits long) results in open
or closed switches.
Brigitte VACHON
December 7, 2015
187
Note that each resistor is different (R, 2R, 4R, ...). For converting a 10-bit word, you need 10 resistors with a range of values from R to 512R. The different-valued resistors have different temperature
coefficients such that a calibration of this DAC is very temperature-dependent.
In practice, this approach only useful for converting binary words with a small number of bits, for
higher resolution use another technique called a R/2R ladder DAC.
7.8.2
R/2R Ladder DAC
Pros:
Only need two resistor values: R and 2R (and RF for the feedback)
Can easily add steps to the resistor ladder to improve resolution
Brigitte VACHON
December 7, 2015
188
Current drawn through any of the switches is always the same, regardless of whether each switch is
up/down. This is because in both cases the current flows directly to ground.
Total current supplied by Vref is therefore also constant.
Brigitte VACHON
December 7, 2015
189
Diagram showing how the R/2R circuit can be reduced to deduce the fraction of the total current
flowing through each switch.
Exercise: Given R=10 k, RF = 20 k and Vref = +5 V, if switches S3, S2, S1 and S0 are put in
position to represent the binary number 0101 (510), calculate what will be the output voltage?
Brigitte VACHON
December 7, 2015
190
Chapter 8
Memory Devices
Memory devices allow you to store binary data on a temporary or permanent basis.
Different storage technologies exist: Semiconductor, magnetic, optical, paper. [Give examples
for each category]
We will only discuss here semiconductor memory devices.
191
CHAPTER 8. MEMORY DEVICES
8.1
8.1. MEMORY ORGANIZATION
Memory Organization
Memory devices are typically arranged in an array of individual memory cells, each representing
one bit (0 or 1).
Each bit can be uniquely identified by specifying a row and column number.
(source: http://www.zeepedia.com/read.php?memory_organization_capacity_density_signals_and_basic_operations_read_write_address_data_
signals_digital_logic_design&b=9&c=39)
Brigitte VACHON
December 7, 2015
192
Location of unit of data in memory is called a memory address.
An n-bit memory address can uniquely identify 2n locations.

10-bit address can uniquely identify 210 = 1, 024 locations (binary definition of Kilo)
20-bit address can uniquely identify 220 = 1, 048, 576 locations (binary definition of Mega)
32-bit address can uniquely identify 232 = 4, 294, 967, 296 locations (binary definition of 4G).
Brigitte VACHON
December 7, 2015
193
Memory capacity identified by number of units it can store times the unit size.
Example:
A memory array of 8 rows x 8 columns can store 8 different Bytes = 8 Byte memory.
Rows are selected using a 3-bit address.
A memory array of 64 rows x 1 column can store 64 different bits = 64-bit memory.
A memory array of 1024 rows x 8 columns can store 1024 different Bytes = 1 kilo-Byte memory
(1 KB).
Brigitte VACHON
December 7, 2015
194
Basic memory operations
(source: http://www.zeepedia.com/read.php?memory_organization_capacity_density_signals_and_basic_operations_read_write_address_data_
signals_digital_logic_design&b=9&c=39)
Brigitte VACHON
December 7, 2015
195
8.2. READ-ONLY MEMORY (ROM)
The are two different families of semiconductor memories: Read-only memory (ROM) and Random
Access memory (RAM)
Semiconductor
Memory devices
8.2
Non-volatile memories
Read-Only Memory
(ROM)
Volatile memories
Random Access Memory
(RAM)
MROM
PROM
EPROM
EEPROM
Flash
Etc...
SRAM
DRAM
... and lots of other flavours..
Read-Only Memory (ROM)
ROM devices are a type of memory called non-volatile which means that they do not lose stored
data when power is removed from the device. ROMs are used to store data permanently (or
semipermanently). Typically used to store data used repeatedly in a system (e.g. initialization,
look-up tables, conversions).
Typically used for read-only operations and not to write to after the initial programming. There
exists however ROM-like memory devices on which you can erase and write (relatively slowly)
new data to memory (ex. Flash).
Brigitte VACHON
December 7, 2015
196
8.2.1
MROM
A Mask ROM (MROM) is a device permanently programmed by the manufacturer.
Somewhat expensive to produce ROM (initial mask), so only cost effective when producting a large
number of devices.
Example: Commonly used in computer to store system operating instructions and data to decode
keyboard instructions.
Brigitte VACHON
December 7, 2015
197
8.2.2
Progammable ROM (PROM)
Main difference with a MROM is that PROM come from the manufacturer unprogrammed and are
custom programmed by (typically) engineers to meet the users needs. Just like MROM, once a
PROM is programmed, the memory values cannot be changed.
Programming a PROM is done by burning or leaving intact a memory link.
Brigitte VACHON
December 7, 2015
198
8.2.3
Erasable Programmable ROM (EPROM)
Memory cells made of specialized MOSFET transistors with an additional floating gate (electrically
insulated from control gate and drain-to-source channel). Floating gate can store electrons, thereby
affecting normal operation of gate.
Programming one transistor: Apply high-voltage
pulse (e.g. 12 V) to the control gate with
drain-source biased. This sets up an electric field.
Electrons are attracted to the surface of channel
and some with enough energy tunnel through
insulating layer to get trapped in floating gate.
Reading one transistor: Apply voltage on the
control gate.
Electrons on floating gate, if
present, will inhibit voltage from turning on
the transistor. When no (little) current flows, this
corresponds to a 0.
Erasing: Remove all electrons stored on floating

gates using UV light through window on the IC.
UV light provides enough energy to electrons to
tunnel out of floating gate.
Example: Used to store microcontrollers main program.
Brigitte VACHON
December 7, 2015
199
8.2.4
Flash memory
ROM-like device that can nevertheless be erased and re-programmed (although it is a relatively slow
process).
Also uses transistors with floating gate, similar to EPROM.
Instead of using UV light to erase the memory (remove electrons on floating gate), it uses a large
voltage in opposite polarity than what is used to program the memory cell.
Example of flash memory found in a USB key.

(source: Wikipedia)
Example: Used for mass storage devices, digital camera, music players, cell phones, etc.
Brigitte VACHON
December 7, 2015
200
8.3
8.3. RANDOM ACCESS MEMORY (RAM)
Random Access Memory (RAM)
These devices are a type of memory called volatile which means that they lose stored data when
power is removed from the device.
Typically used for short term data storage.
There are a large number of different types of RAM, but let me just mention two of the basic types:
Static RAM (SRAM) and Dynamic RAM (DRAM).
DRAM
SRAM
Memory cell consists of one capacitor and

one transistor.
Memory cells made of flip-flops.

A bit written in memory stays there until it is overwritten or until the power is
turned off.
Brigitte VACHON
December 7, 2015
The bit written to memory disappears

within milliseconds (due to capacitor discharge). In order to store a bit in memory for longer, you need to keep refreshing that memory cell with the bit value
you want it to have.
201
8.3. RANDOM ACCESS MEMORY (RAM)
Can read from SRAM much faster than from DRAM.

DRAM can store more data than SRAM for the same physical size (DRAM cell much simpler).
SRAM used for relatively small amounts of read/write memory, and DRAM typically used
for larger amounts of read/write memory ( MB).
SRAM need less power to operate than DRAM (which needs constant refeshing), and more
energy efficient.
SRAM: Typically found in Application Specific IC (ASIC), Pocket calculators, cache memory for
computer CPU.
DRAM: Computer memory modules (ex. DIMMs)
Brigitte VACHON
December 7, 2015
202
Chapter 9
(If you are interested) IC, ASIC, FPGA,
microprocessor...
[Note on the final exam]
An integrated circuit (IC) consists in electronic circuits built on a small silicon plate (chip).
An Application Specific Integrated Circuit (ASIC) is an integrated circuit customized for a particular use as opposed to a general purpose.
9.1
ASIC
The following slides explain the different ASIC production steps as described on the Nobel website.
(source: http://www.nobelprize.org/educational/physics/integrated_circuit/history/)
203
CHAPTER 9. (IF YOU ARE INTERESTED) IC, ASIC,

Electronics
FPGA,forMICROPROCESSOR...
Scientists
(source: http://www.icsense.com/icdesign-chipdesign)
Brigitte VACHON
9.1. ASIC
(source: http://www.scienceasart.org/layout-integrated-circuit)
December 7, 2015
204

Electronics
Scientists
(source: Wikipedia)
Brigitte VACHON
9.1. ASIC
(source: http://www.tf.uni-kiel.de/matwis/amat/elmat_en/kap_6/illustr/i6_1_2.html)
December 7, 2015
205

Electronics
Scientists
(source: http://apcmag.com/picture-gallery-how-a-chip-is-made.htm)
9.1. ASIC
(source:
http://www.newtechpress.net/2014/04/01/economics-and-vengeance-wreak-havoc-with-wafer-market/)
Brigitte VACHON
December 7, 2015
206

Electronics
Scientists
Brigitte VACHON
December 7, 2015
9.1. ASIC
207

Electronics
Scientists
Brigitte VACHON
December 7, 2015
9.1. ASIC
208

Electronics
Scientists
Brigitte VACHON
December 7, 2015
9.1. ASIC
209

Electronics
Scientists
(source: Wikipedia)
Brigitte VACHON
9.1. ASIC
(source: http://www.rulestoolsfools.com/?p=803)
December 7, 2015
210

Electronics
Scientists
9.1. ASIC
Ball Grid Array (BGA) packaging

(source: http://electronics.stackexchange.com/questions/70292/how-are-bga-dies-constructed)
(source: http://www.pcmag.com/encyclopedia/term/38577/bga)
(source: http://www.radio-electronics.com/info/data/smt/smd-bga-ball-grid-array-package.php)
(source: https://learn.sparkfun.com/tutorials/integrated-circuits/ic-packages)
Surface Mount packaging

Brigitte VACHON
December 7, 2015
211

Electronics
Scientists
9.2
9.2. FPGA
FPGA
A Field Programmable Gate Array (FPGA) is a complex integrated circuit containing a very large
number of logic blocks that can be configured after manufacturing.
(source: Wikipedia)
(source: Wikipedia)
Pros: Reconfigurable, low cost

Cons: Typically slower than ASIC and consume more power
Brigitte VACHON
December 7, 2015
212

Electronics
Scientists
9.2. FPGA
Schematic diagram of the three different types of components making up an FPGA.
(source: http://www.ni.com/white-paper/6983/en/)
Brigitte VACHON
December 7, 2015
213

Electronics
Scientists
9.2. FPGA
Place-and-route on an FPGA.
You are in control of designing the circuit by configuring the FPGA connections as you wish.
(source: http://www.eecg.toronto.edu/ vaughn/vpr/e64.html)

Brigitte VACHON
(source: http://www.eecg.toronto.edu/ vaughn/vpr/e64.html)

December 7, 2015
214

Electronics
Scientists
9.2. FPGA
FPGAs are configured using a Hardware Description Language (HDL). The FPGA configuration is
called firmware (as opposed to hardware or software).
Programming in HDL is very different than

programming in software languages (ex. C++,
java, python, etc.).
FPGA circuit can do parallel operations!
FPGA must be configured every time you power
it up (configuration held in RAM, but many
FPGA also have on-board flash memory where
firmware versions can be kept in memory).
An ASIC is designed such that each pin is for a
specific feature, in an FPGA, you decide what you
want to use a pin for.
(source: Martin Schoeberl, The FPGA Hello World Example,

www.jopdesign.com/cyclone/hello world.pdf)
Brigitte VACHON
December 7, 2015
215

Electronics
Scientists
9.3
9.3. MICROPROCESSOR
Microprocessor
An IC containing an entire central processing unit(CPU). It consists of a programmable device

taking as input digital data, performing user-specific instructions, stored in its memory, on this data,
and providing a digital output.
With a microprocessor, you have control over the software (the circuit on the chip is fixed as opposed
to an FPGA).
Microcontroller is programmed using language typically closely related to C programming language.
An Arduino Uno board with a microprocessor mounted on it (long black rectangular IC)
(source: http://arduino.cc/en/Main/arduinoBoardUno)
Brigitte VACHON
December 7, 2015
216

Electronics
Scientists
9.3. MICROPROCESSOR
Demo: Rolling the dice [Thanks to Mark Orchard-Webb!]
Brigitte VACHON
December 7, 2015
217
Appendix A
Grounding issues
Proper grounding is a major issue in the design of complex and large-scale electronic systems.
The word ground is typically used to define a reference voltage level of zero potential.
Earth ground: Connection terminated at a rod (or pipe) driven into the earth.
Signal ground: Common voltage reference point or return line of a signal.

Can be at a different potential than earth ground.
Frame or Chassis
218
APPENDIX A. GROUNDING ISSUES
Simple example of a common floating return (current return line not necessarily tied to earth
ground).
Brigitte VACHON
December 7, 2015
219
A.1
A.1. GROUND LOOPS
Ground loops
Noise in a circuit is often due to poor grounding design which can lead to ground loops. Ground
loops occur when there is more than one return path for the ground current.
Unwanted current flow due to potential difference between ground points caused by non-zero
impedance in ground lines.
Second return path can be thought of as acting like a big antenna which can pick up EM fields
inducing a loop current.
Heres an example of an audio

system where you may hear a
hum noise
(source: http://www.prosoundweb.com/article/preventing_hum_and_rfi_in_your_studio/P2/)
A typical sign that a ground loop exists in your circuit, or that a ground is missing, is the presence
of induced power line (60 Hz) noise in the circuit.
To avoid ground loops, use a single-point ground.
Brigitte VACHON
December 7, 2015
220
A.2. AC GROUND
Electronic systems containing both analog and digital circuits should have the analog and digital
grounds kept separate, and eventually connected together. This will minimize noise from the digital
circuit sometimes generating current spikes as the circuit changes state. Analog circuit can also
generate current spikes which could then affect the digital circuitry if the grounds are not kept
separate.
A.2
AC ground
A constant DC point is an AC ground and is equivalent to an ordinary ground for a signal voltage.
Brigitte VACHON
December 7, 2015
221
Appendix B
Semiconductor technology
The following content is an extract from the book: P.Scherz and S.Monk, Practical Electronics for
Inventors, McGraw-Hill, 3rd Edition.
222
Appendix C
Current in semiconductors
The following content is an extract from the book: Thomas L. Floyd, Electronic Devices (Conventional Current Version) Floyd, Basic Electronics for scientists and engineers, Prentice Hall,
2012.
230
C URRENT
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
C URRENT
Si
Si
Si
Si
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11
FIGURE 1 1 1
Covalent bonds in a silicon crystal.
Si
Si
S EMICONDUCTORS
IN
SECTION 12
CHECKUP
13
Si
Si
IN
Si
Si
What is the basic difference between conductors and insulators?

How do semiconductors differ from conductors and insulators?
How many valence electrons does a conductor such as copper have?
How many valence electrons does a semiconductor have?
Name three of the best conductive materials.
What is the most widely used semiconductive material?
Why does a semiconductor have fewer free electrons than a conductor?
How are covalent bonds formed?
What is meant by the term intrinsic?
What is a crystal?
S EMICONDUCTORS
The way a material conducts electrical current is important in understanding how
electronic devices operate. You cant really understand the operation of a device such
as a diode or transistor without knowing something about current in semiconductors.
After completing this section, you should be able to
Describe how current is produced in a semiconductor

Discuss conduction electrons and holes
Explain an electron-hole pair
Discuss recombination
Explain electron and hole current
As you have learned, the electrons of an atom can exist only within prescribed energy
bands. Each shell around the nucleus corresponds to a certain energy band and is separated
from adjacent shells by band gaps, in which no electrons can exist. Figure 112 shows the
energy band diagram for an unexcited (no external energy such as heat) atom in a pure silicon crystal. This condition occurs only at a temperature of absolute 0 Kelvin.
12
"
I NTRODUCTION
TO
E LECTRONICS
F IG U RE 112
Energy band diagram for an unexcited

atom in a pure (intrinsic) silicon
crystal. There are no electrons in the
conduction band.
Energy
Conduction band
Band gap
Valence band (shell 3)
Second band (shell 2)
First band (shell 1)
Nucleus
Conduction Electrons and Holes

An intrinsic (pure) silicon crystal at room temperature has sufficient heat (thermal) energy
for some valence electrons to jump the gap from the valence band into the conduction band,
becoming free electrons. Free electrons are also called conduction electrons. This is illustrated in the energy diagram of Figure 113(a) and in the bonding diagram of Figure 113(b).
"
F IG U RE 113
Creation of electron-hole pairs in a

silicon crystal. Electrons in the conduction band are free electrons.
Energy
Heat
energy
Band gap
Valence
band
+4
Free
electron
Conduction
band
Free
electron
Hole
Heat
energy
Hole
+4
Electron-hole pair
(a) Energy diagram
(b) Bonding diagram
When an electron jumps to the conduction band, a vacancy is left in the valence band
within the crystal. This vacancy is called a hole. For every electron raised to the conduction band by external energy, there is one hole left in the valence band, creating what is
called an electron-hole pair. Recombination occurs when a conduction-band electron
loses energy and falls back into a hole in the valence band.
To summarize, a piece of intrinsic silicon at room temperature has, at any instant, a
number of conduction-band (free) electrons that are unattached to any atom and are essentially drifting randomly throughout the material. There is also an equal number of holes in
the valence band created when these electrons jump into the conduction band. This is illustrated in Figure 114.
C URRENT
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Generation of an
electron-hole pair
Si
Si
Si
Si
IN
S EMICONDUCTORS
13
FIGUR E 1 14
Electron-hole pairs in a silicon crystal.

Free electrons are being generated
continuously while some recombine
with holes.
Recombination of
an electron with
a hole
Si
Heat energy
Electron and Hole Current

When a voltage is applied across a piece of intrinsic silicon, as shown in Figure 1-15, the thermally generated free electrons in the conduction band, which are free to move randomly in
the crystal structure, are now easily attracted toward the positive end. This movement of free
electrons is one type of current in a semiconductive material and is called electron current.
Si Si Si Si Si
Si Si Si Si Si
Si Si Si Si Si
Another type of current occurs in the valence band, where the holes created by the free
electrons exist. Electrons remaining in the valence band are still attached to their atoms
and are not free to move randomly in the crystal structure as are the free electrons.
However, a valence electron can move into a nearby hole with little change in its energy
level, thus leaving another hole where it came from. Effectively the hole has moved from
one place to another in the crystal structure, as illustrated in Figure 116. Although current
in the valence band is produced by valence electrons, it is called hole current to distinguish
it from electron current in the conduction band.
As you have seen, conduction in semiconductors is considered to be either the movement of free electrons in the conduction band or the movement of holes in the valence
band, which is actually the movement of valence electrons to nearby atoms, creating hole
current in the opposite direction.
It is interesting to contrast the two types of charge movement in a semiconductor with
the charge movement in a metallic conductor, such as copper. Copper atoms form a different type of crystal in which the atoms are not covalently bonded to each other but consist
of a sea of positive ion cores, which are atoms stripped of their valence electrons. The
valence electrons are attracted to the positive ions, keeping the positive ions together and
forming the metallic bond. The valence electrons do not belong to a given atom, but to the
crystal as a whole. Since the valence electrons in copper are free to move, the application
of a voltage results in current. There is only one type of currentthe movement of free
electronsbecause there are no holes in the metallic crystal structure.
FIGUR E 1 15
Electron current in intrinsic silicon is

produced by the movement of thermally generated free electrons.
14
"
I NTRODUCTION
TO
E LECTRONICS
F IG U RE 116
5 A valence electron moves 3 A valence electron moves

into 4th hole and leaves
into 2nd hole and leaves
a 5th hole.
a 3rd hole.
Hole current in intrinsic silicon.
6 A valence electron moves

into 5th hole and leaves
a 6th hole.

into 3rd hole and leaves
a 4th hole.
Si
1 A free electron
leaves hole in
valence shell.
into 1st hole and leaves
a 2nd hole.
Si
Si
When a valence electron moves left to right to fill a hole while leaving another hole behind, the hole
has effectively moved from right to left. Gray arrows indicate effective movement of a hole.
SECTION 13
CHECKUP
14
N-T Y P E
1.
2.
3.
4.
AND
Are free electrons in the valence band or in the conduction band?

Which electrons are responsible for electron current in silicon?
What is a hole?
At what energy level does hole current occur?
P-T Y PE S EMICONDUCTORS
Semiconductive materials do not conduct current well and are of limited value in their
intrinsic state. This is because of the limited number of free electrons in the conduction
band and holes in the valence band. Intrinsic silicon (or germanium) must be modified by
increasing the number of free electrons or holes to increase its conductivity and make it
useful in electronic devices. This is done by adding impurities to the intrinsic material.
Two types of extrinsic (impure) semiconductive materials, n-type and p-type, are the key
building blocks for most types of electronic devices.
Describe the properties of n-type and p-type semiconductors

Define doping
Explain how n-type semiconductors are formed
Describe a majority carrier and minority carrier in n-type material
Explain how p-type semiconductors are formed
Describe a majority carrier and minority carrier in p-type material
Since semiconductors are generally poor conductors, their conductivity can be drastically increased by the controlled addition of impurities to the intrinsic (pure) semiconductive
material. This process, called doping, increases the number of current carriers (electrons
or holes). The two categories of impurities are n-type and p-type.
N-Type Semiconductor
To increase the number of conduction-band electrons in intrinsic silicon, pentavalent impurity atoms are added. These are atoms with five valence electrons such as arsenic (As),
phosphorus (P), bismuth (Bi), and antimony (Sb).
N-T YPE
AND
P-T YPE S EMICONDUCTORS
15
As illustrated in Figure 117, each pentavalent atom (antimony, in this case) forms covalent bonds with four adjacent silicon atoms. Four of the antimony atoms valence electrons are used to form the covalent bonds with silicon atoms, leaving one extra electron.
This extra electron becomes a conduction electron because it is not involved in bonding.
Because the pentavalent atom gives up an electron, it is often called a donor atom. The
number of conduction electrons can be carefully controlled by the number of impurity
atoms added to the silicon. A conduction electron created by this doping process does not
leave a hole in the valence band because it is in excess of the number required to fill the
valence band.
Si
Si
Sb
Free (conduction) electron

from Sb atom
Si
Si
Majority and Minority Carriers Since most of the current carriers are electrons, silicon
(or germanium) doped with pentavalent atoms is an n-type semiconductor (the n stands for
the negative charge on an electron). The electrons are called the majority carriers in
n-type material. Although the majority of current carriers in n-type material are electrons,
there are also a few holes that are created when electron-hole pairs are thermally generated. These holes are not produced by the addition of the pentavalent impurity atoms.
Holes in an n-type material are called minority carriers.
P-Type Semiconductor
To increase the number of holes in intrinsic silicon, trivalent impurity atoms are added.
These are atoms with three valence electrons such as boron (B), indium (In), and gallium
(Ga). As illustrated in Figure 118, each trivalent atom (boron, in this case) forms covalent
bonds with four adjacent silicon atoms. All three of the boron atoms valence electrons are
used in the covalent bonds; and, since four electrons are required, a hole results when each
trivalent atom is added. Because the trivalent atom can take an electron, it is often referred
to as an acceptor atom. The number of holes can be carefully controlled by the number of
trivalent impurity atoms added to the silicon. A hole created by this doping process is not
accompanied by a conduction (free) electron.
Majority and Minority Carriers Since most of the current carriers are holes, silicon (or
germanium) doped with trivalent atoms is called a p-type semiconductor. The holes are the
majority carriers in p-type material. Although the majority of current carriers in p-type
material are holes, there are also a few conduction-band electrons that are created when
electron-hole pairs are thermally generated. These conduction-band electrons are not produced by the addition of the trivalent impurity atoms. Conduction-band electrons in p-type
material are the minority carriers.
FIGUR E 1 17
Pentavalent impurity atom in a silicon crystal structure. An antimony

(Sb) impurity atom is shown in the
center. The extra electron from the
Sb atom becomes a free electron.
16
I NTRODUCTION
TO
E LECTRONICS
"
FIG UR E 1 1 8
Trivalent impurity atom in a silicon

crystal structure. A boron (B) impurity atom is shown in the center.
Si
Hole from B atom
Si
Si
Si
SECTION 14
CHECKUP
15
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Define doping.
What is the difference between a pentavalent atom and a trivalent atom?
What are other names for the pentavalent and trivalent atoms?
How is an n-type semiconductor formed?
How is a p-type semiconductor formed?
What is the majority carrier in an n-type semiconductor?
What is the majority carrier in a p-type semiconductor?
By what process are the majority carriers produced?
By what process are the minority carriers produced?
What is the difference between intrinsic and extrinsic semiconductors?
T HE PN J UNCTION
When you take a block of silicon and dope part of it with a trivalent impurity and the other
part with a pentavalent impurity, a boundary called the pn junction is formed between the
resulting p-type and n-type portions. The pn junction is the basis for diodes, certain transistors, solar cells, and other devices, as you will learn later.
Describe how a pn junction is formed

Discuss diffusion across a pn junction
Explain the formation of the depletion region
Define barrier potential and discuss its significance
potential in silicon and germanium
Discuss energy diagrams
Define energy hill
State the values of barrier
A p-type material consists of silicon atoms and trivalent impurity atoms such as boron.
The boron atom adds a hole when it bonds with the silicon atoms. However, since the number of protons and the number of electrons are equal throughout the material, there is no
net charge in the material and so it is neutral.
T HE PN J UN C T I O N
17
An n-type silicon material consists of silicon atoms and pentavalent impurity atoms such as
antimony. As you have seen, an impurity atom releases an electron when it bonds with four
silicon atoms. Since there is still an equal number of protons and electrons (including the free
electrons) throughout the material, there is no net charge in the material and so it is neutral.
If a piece of intrinsic silicon is doped so that part is n-type and the other part is p-type,
a pn junction forms at the boundary between the two regions and a diode is created, as
indicated in Figure 119(a). The p region has many holes (majority carriers) from the
impurity atoms and only a few thermally generated free electrons (minority carriers). The
n region has many free electrons (majority carriers) from the impurity atoms and only a
few thermally generated holes (minority carriers).
pn junction
p region
n region
Depletion region
p region
n region
Barrier
potential
(a) The basic silicon structure at the instant of junction formation
showing only the majority and minority carriers. Free electrons
in the n region near the pn junction begin to diffuse across the
junction and fall into holes near the junction in the p region.
(b) For every electron that diffuses across the junction and
combines with a hole, a positive charge is left in the n region
and a negative charge is created in the p region, forming a
barrier potential. This action continues until the voltage of
the barrier repels further diffusion. The blue arrows between
the positive and negative charges in the depletion region
represent the electric field.
F IGU RE 119
Formation of the depletion region. The width of the depletion region is exaggerated for illustration
purposes.
Formation of the Depletion Region

The free electrons in the n region are randomly drifting in all directions. At the instant of
the pn junction formation, the free electrons near the junction in the n region begin to diffuse across the junction into the p region where they combine with holes near the junction,
as shown in Figure 119(b).
Before the pn junction is formed, recall that there are as many electrons as protons in
the n-type material, making the material neutral in terms of net charge. The same is true for
the p-type material.
When the pn junction is formed, the n region loses free electrons as they diffuse across
the junction. This creates a layer of positive charges (pentavalent ions) near the junction.
As the electrons move across the junction, the p region loses holes as the electrons and
holes combine. This creates a layer of negative charges (trivalent ions) near the junction.
These two layers of positive and negative charges form the depletion region, as shown in
Figure 119(b). The term depletion refers to the fact that the region near the pn junction is
depleted of charge carriers (electrons and holes) due to diffusion across the junction. Keep
in mind that the depletion region is formed very quickly and is very thin compared to the n
region and p region.
After the initial surge of free electrons across the pn junction, the depletion region has
expanded to a point where equilibrium is established and there is no further diffusion of
HISTORY NOTE
After the invention of the light bulb,
Edison continued to experiment and
in 1883 found that he could detect
electrons flowing through the
vacuum from the lighted filament to
a metal plate mounted inside the
bulb. This discovery became known
as the Edison effect.
An English physicist, John
Fleming, took up where Edison left
off and found that the Edison effect
could also be used to detect radio
waves and convert them to electrical
signals. He went on to develop a
two-element vacuum tube called the
Fleming valve, later known as the
diode. Modern pn junction devices
are an outgrowth of this.
18
I NTRODUCTION
TO
E LECTRONICS
HISTORY NOTE
Russell Ohl, working at Bell Labs
in 1940, stumbled on the
semiconductor pn junction. Ohl
was working with a silicon sample
that had an accidental crack down
its middle. He was using an
ohmmeter to test the electrical
resistance of the sample when he
noted that when the sample was
exposed to light, the current that
flowed between the two sides of
the crack made a significant jump.
This discovery was fundamental to
the work of the team that invented
the transistor in 1947.
electrons across the junction. This occurs as follows. As electrons continue to diffuse
across the junction, more and more positive and negative charges are created near the junction as the depletion region is formed. A point is reached where the total negative charge in
the depletion region repels any further diffusion of electrons (negatively charged particles)
into the p region (like charges repel) and the diffusion stops. In other words, the depletion
region acts as a barrier to the further movement of electrons across the junction.
Barrier Potential Any time there is a positive charge and a negative charge near each
other, there is a force acting on the charges as described by Coulombs law. In the depletion region there are many positive charges and many negative charges on opposite sides of the pn
junction. The forces between the opposite charges form an electric field, as illustrated in
Figure 119(b) by the blue arrows between the positive charges and the negative charges. This
electric field is a barrier to the free electrons in the n region, and energy must be expended to
move an electron through the electric field. That is, external energy must be applied to get the
electrons to move across the barrier of the electric field in the depletion region.
The potential difference of the electric field across the depletion region is the amount of
voltage required to move electrons through the electric field. This potential difference is
called the barrier potential and is expressed in volts. Stated another way, a certain
amount of voltage equal to the barrier potential and with the proper polarity must be applied across a pn junction before electrons will begin to flow across the junction. You will
learn more about this when we discuss biasing in Chapter 2.
The barrier potential of a pn junction depends on several factors, including the type of
semiconductive material, the amount of doping, and the temperature. The typical barrier
potential is approximately 0.7 V for silicon and 0.3 V for germanium at 25C. Because germanium devices are not widely used, silicon will be used throughout the rest of the book.
Energy Diagrams of the PN Junction and Depletion Region

The valence and conduction bands in an n-type material are at slightly lower energy levels
than the valence and conduction bands in a p-type material. Recall that p-type material has
trivalent impurities and n-type material has pentavalent impurities. The trivalent impurities
exert lower forces on the outer-shell electrons than the pentavalent impurities. The lower
forces in p-type materials mean that the electron orbits are slightly larger and hence have
greater energy than the electron orbits in the n-type materials.
An energy diagram for a pn junction at the instant of formation is shown in Figure
120(a). As you can see, the valence and conduction bands in the n region are at lower energy levels than those in the p region, but there is a significant amount of overlapping.
The free electrons in the n region that occupy the upper part of the conduction band in
terms of their energy can easily diffuse across the junction (they do not have to gain additional energy) and temporarily become free electrons in the lower part of the p-region conduction band. After crossing the junction, the electrons quickly lose energy and fall into
the holes in the p-region valence band as indicated in Figure 1-20(a).
As the diffusion continues, the depletion region begins to form and the energy level of
the n-region conduction band decreases. The decrease in the energy level of the conduction
band in the n region is due to the loss of the higher-energy electrons that have diffused
across the junction to the p region. Soon, there are no electrons left in the n-region conduction band with enough energy to get across the junction to the p-region conduction band, as
indicated by the alignment of the top of the n-region conduction band and the bottom of the
p-region conduction band in Figure 120(b). At this point, the junction is at equilibrium;
and the depletion region is complete because diffusion has ceased. There is an energy gradiant across the depletion region which acts as an energy hill that an n-region electron
must climb to get to the p region.
Notice that as the energy level of the n-region conduction band has shifted downward,
the energy level of the valence band has also shifted downward. It still takes the same
amount of energy for a valence electron to become a free electron. In other words, the energy gap between the valence band and the conduction band remains the same.
S UMMARY
Energy
Minority carriers
Majority carriers
Conduction
band
Valence
band
Valence
band
Majority carriers
p region
Minority carriers
pn junction
n region
(a) At the instant of junction formation

#
19
Energy
Conduction
band
p region
pn junction
and depletion
region
n region
(b) At equilibrium
F IGU RE 120
Energy diagrams illustrating the formation of the pn junction and depletion region.
SECTION 15
CHECKUP
1.
2.
3.
4.
5.
6.
What is a pn junction?
Explain diffusion.
Describe the depletion region.
Explain what the barrier potential is and how it is created.
What is the typical value of the barrier potential for a silicon diode?
What is the typical value of the barrier potential for a germanium diode?
SUMMARY
Section 11
According to the classical Bohr model, the atom is viewed as having a planetary-type structure
with electrons orbiting at various distances around the central nucleus.

According to the quantum model, electrons do not exist in precise circular orbits as particles as
in the Bohr model. The electrons can be waves or particles and precise location at any time is
uncertain.
The nucleus of an atom consists of protons and neutrons. The protons have a positive charge and
the neutrons are uncharged. The number of protons is the atomic number of the atom.
Electrons have a negative charge and orbit around the nucleus at distances that depend on their
energy level. An atom has discrete bands of energy called shells in which the electrons orbit.
Atomic structure allows a certain maximum number of electrons in each shell. In their natural
state, all atoms are neutral because they have an equal number of protons and electrons.
The outermost shell or band of an atom is called the valence band, and electrons that orbit in
this band are called valence electrons. These electrons have the highest energy of all those in the
atom. If a valence electron acquires enough energy from an outside source such as heat, it can
jump out of the valence band and break away from its atom.
Section 12
Insulating materials have very few free electrons and do not conduct current at all under normal
circumstances.
Materials that are conductors have a large number of free electrons and conduct current very well.
Semiconductive materials fall in between conductors and insulators in their ability to conduct
current.
Semiconductor atoms have four valence electrons. Silicon is the most widely used semiconduc-
tive material.
Appendix D
Band gap theory
The following content is an extract from the book: D.Eggleston, Basic Electronics for scientists and
engineers, Cambridge University Press, 2011.
240
Appendix E
Stability of amplifiers using feedback
The following is a re-print of H.S. Blacks original paper describing the invention of negative feedback
as a mean to stabilize amplifiers and also provide larger operational bandwidth.
H.S. Black, Stabilized feed-back amplifiers, Proc. IEEE, vol. 72, no. 6, pp. 716-722, June 1984.
253
Stabilized Feed-Back Amplifiers
BY
H. S.BLACK
MEMBER ALEE
Due to advances in vacuum-tube development and amplifier

technique, it now is possible to secure any desired amplification of the electrical waves used in the communication field.
When manyamplifiers are worked in tandem,however, it
becomes difficult to keep the over-all circuit efficiency constant, variations in battery potentials andcurrents, small when
considered individually, adding up to produce serious transmission changes for the over-all circuit. Furthermore, although
it has remarkably linear properties, when the modem vacuum
tube amplifieris used to handle a numberof carrier telephone
channels,extraneousfrequenciesaregenerated whichcause
interference between the channels. To keep this interference
withinproperbounds involvesserioussacrifice of effective
amplifiercapacityorthe
use of apush-pullarrangement
which, while giving some increasein capacity, addsto maintenance difficulty.
However,bybuilding
an amplifierwhosegain
is made
deliberately, say40 decibels higherthan necessary (l0,OOO fold
excess on energy basis) and then feeding the output back to
the input in such a way as to throw away the excess gain, it
has been found possible to effect extraordinary improvement
in constancy of amplification and freedom from non-linearity.
By employing this feed-back principle, amplifiers have been
builtand usedwhosegainvariedless
than 0.01 db with a
change in platevoltagefrom 240 to 260 voltsandwhose
modulation products were 75 db below the signal output at
full load. For an amplifier of conventional design and comparable size this change in plate voltage would have produced
about 0.7 db variation while the modulation products would
have been only 35 db down; in other words, 40 db reduction
in modulation products was effected. (On an energy basis the
reduction was 10,OOO fold.)
Full text of apaper recommended for publication by the ALEE.
committee on communication, andscheduled for discussion atthe
A.I.EE winter convention, Jan. 23-26, 1934. Manuscript submitted
March 28, 1933;released for publication December 4, 1933. Not
published in pamphlet form.
Reprinted from Nectrical hgineering (The Official Monthly Journal and Transactions of the A.I.E.E.), vol. 53, pp. 114-120, Jan. 1934.
Copyright, 1934, by the American Institute of Electrical Engineers.
716
Stabilized feed-back possesses other advantages indreduced delay and delay distortion,reduced noise disturbance
from the power supply circuits and various other feature best
appreciated by practical designersof amplifiers.
It is f a r from a simple proposition to employ feed-back in
this way because of the very special control requiredof phase
shifts in theamplifier
and feed-backcircuits, not only
throughout the useful frequency band but for a wide range of
frequencies above and below this band. Unless these relations
are maintained, singing will o c c u , usually at frequencies
outside the useful range. Once having achieved adesign,
however, in which proper phase relations are secured, experience has demonstrated that the performance obtained is perfectly reliable.
The carrier-incable system dealt with in a recent Institute
paper (Carrier in Cables by A. B. Clark and B. W.Kendall.
A.I.EE 'TUNS., k c . 1933, p. 10%) involves many amplifiers
in tandem with many telephone channels passing through each
amplifier and constitutes, therefore, an ideal field for application of this feed-back principle. A field trial of this system
was made at Morristown, New Jersey, in which 70 ofthese
amplifiers were operated in tandem. The results of this trial
werehighly satisfactoryanddemonstrated conclusively the
correctness of the theory and the practicabilityof its commercial application.
CIRCUIT
ARRANGEMENT
In theamplifier of Fig. 1, aportion of the output is
returned to the input to produce feed-back action. The upper
branch, called the p circuit, is represented as containing active
elements such as an amplifier while the lower branch, called
the 8-circuit, is shownas a passive network, The
way a voltage
is modified after once traversing each circuitis denoted p and
/3, respectively, and the product, pp, represents how a voltage
ismodifiedaftermakingasingle
journey aroundamplifier
and feed-back circuits. Both p and 8 are complex quantities,
functions of frequency, and in the generalized concept either
or both may be greater or lessin absolute value than unity: (p
is not used in the sense that it is used sometimes, namely, to
denote the amplification constant of a particular tube, but as
the complex ratio of the output to the input voltage of the
amplifier circuit).
Fig. 2 shows an arrangement convenient forsome purposes
where,byusingbalancedbridges
in the input and output
circuits, interaction between the circuits that connect to the
input and output is avoided. Thereby feed-back action and
amplifier impedances are made independentof the properties
of circuits connectedto the amplifier.
GENERAL
EQUATION
In Fig. 1, 8 is zero without feed-back and a signal voltage,
e,, applied to the input of the p-circuit produces an output
voltage. This is made up of what is wanted,the amplified
PROCEEDINGS OF THE I E E E , VOL. 72,NO. 6, JUNE1984
CHANGE IN
GAINDUE TO FEED-BACK
From eq 2,theamplificationwithfeed-back
equals the
amplificationwithoutfeed-backdivided
by (1 - a/?). The
is tochange
effect of addmgfeed-back,therefore,usually
the gain of the amplifier and this change will be expressed as
where Gc- is a3 change in gaindue to feed-back. As a

quantitative measureof the effectof feed-back l w i l l
1 - PB
be used and the feed-back referred to as positive feed-buck or

1
negative feed-back according as the absolute value of 1- ILB
Fig. 1. Amplifii system with feed-b&

e. signal input voltage
p . propagation of amplifier circuit
p e . signal output voltage without feed-back
n. noise output voltage without feed-back
d( E ) . distortion outputvoltage without feed-back
8. propagation of feed-back circuit
E. s@
output voltage with feed-back
N . noise output voltage with feed-back
D . distortion output voltage with feed-back
Iheoutputvoltagewithfeed-backisE+N+Dandisthesumofpe+n
+ d( E ) , the value without feed-back plus
p p [ E + N + Dl due to feed-back.
E+N+D=pe+n+d(E)+pB(E+N+D]
[E+N+DKl-pB)-pe+n+d(E)
E+N+D-&+n + d o
1-pB
1-pB
1-18
If
s 1, E = =
Under this conditiontheamplification is indepen-
is greater or less than unity. Positive feed-back increases the

gain of the amplifier; negative feed-back reduces it. The term
cases where the
feed-back is notlimitedmerelytothose
absolute value of -is other than unity.
From pB
1 - PB
= Ip&@
and (3), it may be shown that
10-GCFo= 1- 2lpplcos@
(4)
ljbB12
which is theequationfora
familyof concentric circles of
radius 10-GcF/lo aboutthepoint
1, 0. Fig. 3 is apolar
dent of p but does depend upon 8. Consequently the over-all characteristic

will be controlled by the feed-back circuit which may include equaIizers or
other corrective networks.
Fig. 2. C i t of a negative feed-back amplifier
signal, E,, and componentsthatare

not wanted,namely,
noise and distortion designatedNo and Do and assumed to be
generated within the amplifier. It is further assumed that the
noise is independent of the signal and the distortionof modulation a function on& of the signal output. Using the notation
of Fig. 1, the output without feed-back may be written as:
E,
+ No + Do = peo + n + d ( E , )
(1)
where zero subscripts referto conditions without feed-back.

With feed-back, B is not zero and the input to the p-circuit
becomes e , + @ ( E + N + D ) . The output is E + N + D
and is equal to p [ e , + B ( E + N + D ) ] + n + dE or
Fig. 3. Tbe vector field of
p/3
See caption for Fig. 4

The complex quantity p 8 represents the ratio by which the amplifier and
feed-back (or more generally p and 8 ) modify a voltage in a single trip
around the closed path
First, there is a set of boundary curves indicated by letters which giveeither
limiting or sigdicant values of lppl and
Second there is a family of curves in which db changein gain due to
feed-back is the parameter.
e.
Conditions in which gain and modulation are unaffectedby feed-back.
B. Constant amplification ratio against small variations in 1/31,
Constant change in gain, - against variations in IpI and 181. Stable

1
1- PPI
In the output signal, noise andmodulationaredividedby

(1 - p b ) , and assuming 1
1 - pBI > 1, all are reduced.
BLACK: STABILIZED FEED-BACK AMPLIFIERS
phase shift through the amplifier against variation in @p.

The boundary on which the stability of amplification is unaffected by
feed-back.
71 7
of course a complex ratio. It will bedesignated A, and

referred to as the amplification with feed-back.
To consider the effect of feed-back upon stability of amplification, the stability will be viewed as the ratio of a change,
&AF, to A, where &AF is due to a change either in p or @ and
theeffectsmaybederivedby
assuming thevariations are
small.
Fib3
C.
D. IGPI = I
E CP 90". Improvement in gain stability corrrsponds to twice db re
F.
G.
H.
I.
J.
diagram of the vector field of pB = IpBpB. Using rectangular

instead of polar coordinates, Fig. 4 corresponds to Fig. 3 and
may be regarded as a diagram of the field of pB where the
parameter is db change in gain due to feed-back. From these
diagrams all of the essential propertiesof feed-back action can
be obtained such as change in amplification, effect on linearity, changein stability dueto variations in various partsof the
system, reduction of noise,etc. Certain significant boundaries
have been designated similarlyon both figures.
For example, boundaryA is the locusof zero change in gain
due to feed-back. Along this parametric contour line where
theabsolutemagnitude of amplification is notchangedby
feed-back action, values of lpBl range from zero to 2 and the
phase shift, CP aroundthe ampliFier andfeed-backcircuits
equal
1pB[/2 a d , therefore, lies betww -90 deg and
+ 90 deg. For all conditions insideor above this boundary, the
gain with feed-back is increased; outside or below, the gain is
decreased.
STABILITY
From eq 2, -is the amplified signal with feed-back
1 - PB
and - therefore, is an index of the amplification. It is
1 - PB'
718
IfpB~l,itisseenthat~orthe~-circuitisstab~by
an amount corresponding to the reduction in amplification
and the effect of introducing a gain or loss in the p-circuit is
to produce no material change in the over-all amplificationof
the system; the stability of amplification as affected by B or
theB-circuit is neitherappreciablyimprovednordegraded
since increasing the loss in the &circuit raises the gain of the
amplifier by anamount almost corresponding to theloss
introduced and viceversa. If both p and B are varied and the
variations sufficiently small, the effect is the same as if each
were changed separately and the two results then combined.
In certain practical applicationsof amplifiers it is the change
in gain or ammeter or voltmeter reading at the output that is a
measure of the stability rather than the complex ratio previously treated. The conditions surroundmg gain stability may
be examined by considering the absolute value of Ap. This is
shorn as fonows:
Let (db)represent the gain in decibels
to A,.
Then
o
c
To get the absolute value of the amplification:

PB = IaBW
The stability of amplification which is proportional to the

gain stability is given by
Acuriousfacttobenotedfromeqllisthatitispossibleto
choose a value of pB (namely, lpBl= set@) so thatthe
numerator of the nght hand term vanishes. This meam that
the gain stabilityis perfect, assuming differential variations in
Ipl. Referring to Figs. 3 and 4, contour C is thelocus of
IpBI = sec @ and it includes all amplifierswhose gain is
PROCEEDINGS O F T H E IEEE, VOL. 72, NO. 6, JUNE 1964
unaffected by small variationsin Jpl.In this way it is possible

even to stabilize an amplifier whose feed-back is positive, i.e.,
feed-backmaybeutilized to raisethegain of anamplifier
and, at the same time, the gain stability with feed-back need
not be degraded but on the contrary may be improved. If a
similar procedure is followed with an amplifier whose feedback is negative, the gain stability theoretically will be perfect
and independent of the reductions in gain due to feed-back.
Over too wide a frequency band practical difficultieswill limit
the improvements possible by these methods.
With negative feed-back, gain stability always is improved
by an amount at least as great as corresponds to the reduction
in gainandgenerallymore;
with positivefeed-back,gain
stability never is degraded by more than would correspond to
the increase in gain and under appropriate conditions, assuming the variations are not too great is as good or much better
thanwithoutfeed-back.Withpositivefeed-back,thevariations in p or 8 must not be permitted to become sufficiently
great as to cause the amplifier to sing or give rise to instability
as defined in the section devoted to the conditions for avoiding singing.
the output. Denoting this amplification by u and uo, respectively, it can be shown that the relation between the 2 noise
ratios is (uo/u)(l - p8). This is called the noise index.
If noise is introduced in the power supply circuitsof the last
tube, ao/a = 1 and the noise index is (1 - p8). As a result of
this relation less expensive power supply filters are possiblein
the last stage.
MODULATION
8-Circuit Equalization. Referring to eq 2, the output voltage E approaches - eo/8 as 1 - p 8 = -a8 and equals it in
absolute value if cos@ = -where p 8 = IpBp3. Under
21PBI
these circumstances increasing the loss in the 8-circuit 1 db
raisesthegain
of theamplifier 1 db, and vice-versa, thus
giving any gain-frequency characteristic for which a like lossfrequency characteristic can be inserted in the 8-circuit. This
procedure has been termed 8-circuit equalization. It possesses
other advantages and properties which are beyond the xope
of this paper.
To determine the effect of feed-back action upon modulationproduced in theamplifiercircuit,

it is convenient to
assume that the output of undistorted signal is made the same
withandwithoutfeed-backand
that acomparisonthen is
made of the difference in modulation with and without feedback. Therefore,withfeed-back,the
input is changed to
e = e,(l - p/3) and, referring to eq 2, the output voltage is
.be," andthegeneratedmodulation, d ( E ) , assumes its value
d( E )
4EO)
withoutfeed-back, d(Eo), and -becomes 1 -PB
1 - ILB
whichis - This relationship is approximatebecause
1 - P8'
the voltage at the input without feed-back is free from distortion and with feed-back it is not and, hence, the assumption
that the modulation is a function only of the signal output used
in deriving eq 2 is not necessarily justified.
"
From the relationship D
Do
it is to be concluded
1 - PB'
that modulationwithfeed-back will bereduceddecibelfor
decibel as the effect of feed-back action causes an arbitrary db
reduction in the gain of the amplifier; i.e., when the feed-back
is negative. With positive feed-back the opposite
is true, the
modulation being increased by an amount corresponding to
the increase in amplification.
If modulation in the 8-circuit is a factor, it can be shown
that usually in its effect on the output the modulation levelat
theoutputduetononlinearity
of the/3-circuitisapproxip8
multiplied by the modulation generated
in the
mately 1 - P8
/?-circuit acting alone and without feed-back.
ADDITIONAL,
EFFECTS
Noise. A criterion of the worth of a reduction in noise is

thereduction in signal-tcmoise ratio at theoutput of an
amplifier. Assuming that the amount of noise introduced is
the same in2 systems, for example, with and without feed-back,
respectively,and that thesignaloutputsarethe
same, a
comparison of the signal-tcmoise ratios will be affected by the
amplification between the place at which the noise enters and
BLACK:
In the
A , 0 is theover-allphaseshift
expression A , = 1
-Pal
withfeed-back,and
it can beshown that the phase shift
through the amplifer with feed-back may be made to apprwch
the phase shift through the 8-circuit p h 180 deg. The effect of
phase shift in the 8-circuit is not reduced correspondingly. It
will be recalled that in reducing the change in phase shift with
is the slopeof the phase shift
frequency, envelope delay, which
with respect to the angular velocity, o = 2 vf, also is reduced.
The delay distortion likewise is reduced because a measure of
delay distortion at a particular frequency is thedifference
betweentheenvelopedelay at that frequency and the least
envelope delay in the band.
Phase Shift, Envelope Delay,DelayDistortion.
AVOIDSINGING
Having considered the theory up to this point, experimental evidence was readilyacquired to demonstrate that p8
might assume large values, 10 to lO,OOO, provided CP was not
at the same time zero. However, one noticeable feature about
thefield of p 8 (Figs. 3 and 4) is that it implies that even
though the phase shift is zero and the absolute value of p 8
exceeds unity, self-oscillations or singing will not result. This
may or may not be true. Whenfirst thinking about this matter
it was suspected that owing to practical nonlinearity, singing
would result whenever the gain around the closed loop equaled
or exceeded the loss and simultaneously the phase shift was
zero; i.e., p 8 = Ip8I + J o 2 1. Results of experiments, however,seemed to indicate somethingmorewasinvolvedand
these matters were described to H. Nyquist who developed a
more general criterion for freedom from instability applicable
to an amplifier having linear positive constants. (For a complete description of the criterion for stability and instability
andexactlywhatismeantbyenclosingthepoint
(l,O),
referenceshouldbemade
to RegenerationTheory, by H.
Nyquist, Bell System Technical J m l , v. XI, July 1932, p.
126-47.)
To use this criterion, plot p 8 (the modulus and argument

varywithfrequency)and
its complexconjugate in polar
coijrdinates for all values of frequency from 0 to + m, If the
resultingloop
or loops do notenclosethepoint
(1,O)
the system will be stable, otherwise not. The envelope of the
transient response of astableamplifieralwaysdiesaway
719
exponentially with time; that of an unstable amplifier in all

physicaTly maliable cases incnaseswith time. characteristics
AandBinFig.5areresultsofmeasurementson2different
amplifiers; the amplifier having p/3 characteristic denoted A
was stable, theother unstable.
. The number of stages of amplification that can be used in a
single amplifier is not significant except in so far as it affects
the question of avoidmg singing. Amplifiers with considerable
negativefeed-backhave
been tested where thenumber of
stages ranged from1to 5, inclusive. In every case the feed-back
path was from the output of the last tube to the input of the
first tube.
EXPERIMENTAL &SULTS
Figs. 6 and 7 show how the gain-frequency and modulation
characteristics of the 3-stage impedance coupled amplifier of
Fig. 2 are improvedbynegativefeed-back.
In Fig. 7 the
improvement in harmonics is not equal exactly to the decibel
reduction in gain. Fig. 8 shows measurements on a different
amplifier in which harmonics are reduced as negative feed-back

is incmsed, deciil for deciil over a 6Sdb range.
That the gain with frequency practically is independent of
small variations in Ipl is shown by Fig. 9.This is a chammistic of the Morristown amplifier,described in thepaper by
Clark and Kendall referred to previously, qvhichmeets the
severe requirements imposed upon a repeater amplifia for use
in cable carriersyse
tms.Designed to amphfy frequencies from
4 kc to 40 kc the maximumchange in gain due to variations in
plate voltage does not exceed 7/10,000 db per volt and at 20
kc the change is only 1/20,000 db per volt. This illustrates
that for small changes in lpl, the ratio of the stability without
feed-back to the stability with feed-back, called the
stability
and gain stability is imindex, approaches
1 - IPBlrnQ
proved at least as much as the gain is reduced and usually
1
more,andistheoreticallyperfectifcosQ= IPBI
720
PROCEEDINGS OF THEIEEE, VOL. 72, NO. 6, JUNE1984
09 REDVCTW H o*w W C TO FEEDBKK
Fig. 8. Improvement of harmonin with feed-back

One exampleof another amplifier in which, with 60-db feed-back, harmonic
currents in the output are only 1 thousandth and their energy 1 millionth of
the values without feed-back
Yw*LpoI1yIDo(p
Fig. 10. Gain-load characteri$ic with and without feed-back for

a low
level a m p l i f i e r desigwd to ampiify frequencies from 3.5 to 50 kc
FRCWENCV
OB C M f f i E
-.2
IN
CAIN
RILOCYCLCS
WITHOUT CCEDBACK
*P
r4
Fig. 11. Phase shift delay, and delay distortion with and without
feed-back for
a single tube voice frequency amplifier
Fig. 9. Repesentative gain stability of a single amplifier as determined by measuring @ feed-amplifiers in tandem at Morristom, New Jersey
The upper figure shows the absolute value of the stability index. It can be
seen that between 20 and 25 kc the improvement in stability is more than
1,OOO to 1 yet the reduction in gain was less than 35 db
The lower figure showschange in gain of the feed-backamplifierwith

changes in the plate battery voltage and the corresponding changes in gain
without feed-back. At some frequencies the change in gain is of the same
sign as without feed-back and at others it is of opposite sign and it can be
seen that near 23 kc the stability must be perfect
BLACK: STABILIZED FEED-BACK AMPLIFIERS
In Fig. 10 is indicated the effectiveness with which the gain

of a feed-back amplifier can be made independent of variations in input amplitude practically up to the overload point
of the amplifier. These measurementswere made on a 3-stage
amplifier designedto work from 3.3 kc to 50 kc.
As shown in Fig. 11, the negative feed-back may be used to
improve phase shift and reduce delay and delay
distortion.
Thesemeasurements weremade on an experimental 1-tube
amplifier, 35-8,500 cycles, feeding back around the low side
windings of the input and output transformers.
In Fig. 12 is given the gain-frequency characteristic of an
amplifier with and without feed-back when
in the fl-circuit
there is an equalizerdesigned to makethegain-frequency
characteristic oftheamplifier
with feed-back of thesame
shape as theloss-frequencycharacteristic
of anonloaded
telephone cable.
721
mmcvcmmaaum
Fig. 12. Cain-frequency cbaraaeristic of an amplifier with an
equnlizerintheB-cimlit
This was designed to have a gain frequency characteristic with feed-back of

the same shape as the loss frequencycharacteristic of a nonloaded telephone cable
CONCLUSION
The feed-back amplifier dealt within this paper was developed primarily with requirements in mind for a cable carrier
telephone system, involving many amplifiers in tandem with
many telephone channels passing
through each amplifier. Most
of the examples of feed-back amplifier performance naturally
have been drawnfromamplifiersdesignedfor
this field of
operation. In this field,vacuum tube amplifiersnormally
722
possessinggoodcharacteristicswithrespect
to stability,and
freedomfrom distortion are made to possess superlatively
good characteristics by application of the feed-back principle.
However, certain types of amplifiers, in which economy has
been secured by sacrificing performance characteristics, particularly as regards distortion, can be made to possess improved characteristics by the application of feed-back. Discussion of these amplifiers is beyond the scope of this paper.
PROCEEDINGS OF THE IEEE, VOL. 72,NO. 6, JUNE1984

Phys 328

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Phys 328

Caricato da

Copyright:

Formati disponibili

Electronics for Scientists

2 A few useful reminders about impedance

Electronics for Scientists

Light-emitting diode (LED) . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Electronics for Scientists

Electronics for Scientists

6.3.1 Example: Enable/Disable control . . . . .

Sample and Hold . . . . . . . . . . . . .

you are interested) IC,

Electronics for Scientists

Electronics for Scientists

D Band gap theory

E Stability of amplifiers using feedback

Electronics for Scientists

1958: Integrated circuits (Kilby[Nobel])

Electronics for Scientists

Electronics for Scientists

Electronics for Scientists

(source: P.Scherz and S.Monk, Practical Electronics for Inventors)

Electronics for Scientists

2.1. INPUT IMPEDANCE

Impedance seen looking into the input of a circuit or device.

Large input impedance Small current drawn into input device

Impedance looking back into the device or circuit.

TRANSMISSION AND IMPEDANCE MATCHING

Signal transmission and impedance matching

In terms of efficient signal transmission, you generally want Zout  Zin .

Electronics for Scientists

2.3. IMPEDANCE MATCHING

Impedance of transmission line

Signals will typically travel between two

(source: P.Scherz and S.Monk, Practical Electronics for Inventors)

Electronics for Scientists

2.3. IMPEDANCE MATCHING

Example of commonly used cables and their impedance.

Usage: Transmission line for

Usage: Digital system, design for

Pros: EM interference into or

Electronics for Scientists

Electronics for Scientists

Doped semiconductor materials

(source: T.Floyd, Electronic Devices (Conventional Current Version))

Pentavalent impurity atom in a silicon

Trivalent impurity atom in a silicon crystal

See appendices B, C and D for more information.

Electronics for Scientists

(source: P.Scherz and S.Monk, Practical Electronics for Inventors)

External electrical field supplied by battery

Electronics for Scientists

Electronics for Scientists

(source: T.Floyd, Electronic Devices (Conventional Current Version))

Electronics for Scientists

Real diode model:

Electronics for Scientists

3.1. DIODE CIRCUITS

Electronics for Scientists

3.1. DIODE CIRCUITS

The series resistor is used to set the desired output current.

Electronics for Scientists

3.1. DIODE CIRCUITS

Limiter (clipper) circuits

Electronics for Scientists

3.1. DIODE CIRCUITS

In terms of efficient signal transmission, you generally want Zout Zin .