Unit 57 - Electronics

Analogue
Contents

1 The Transistor 1
1.1 Transistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.2 Importance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.3 Simplified operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.4 Comparison with vacuum tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.1.5 Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.1.6 Part numbering standards / specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.1.7 Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.1.8 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.1.9 Directory of external websites with datasheets . . . . . . . . . . . . . . . . . . . . . . . . 12
1.1.10 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.1.11 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.1.12 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2 Bipolar junction transistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.2.2 Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.2.3 Regions of operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.2.4 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.2.5 Theory and modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.2.6 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
1.2.7 Vulnerabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
1.2.8 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1.2.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1.2.10 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
1.3 Field-effect transistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

i
ii CONTENTS

1.3.1 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
1.3.2 Basic information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
1.3.3 More about terminals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
1.3.4 FET operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
1.3.5 Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
1.3.6 Types of field-effect transistors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
1.3.7 Advantages of FET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
1.3.8 Disadvantages of FET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
1.3.9 Uses of FET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
1.3.10 See also FET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
1.3.11 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
1.3.12 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

2 Semiconductor Materials 34
2.1 Silicon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.1.1 Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.1.2 Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.1.3 Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.1.4 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.1.5 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.1.6 Biological role . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.1.7 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.1.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.1.9 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.1.10 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.2 Germanium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.2.1 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.2.2 Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.2.3 Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.2.4 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.2.5 Precautions for chemically reactive germanium compounds . . . . . . . . . . . . . . . . . 51
2.2.6 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.2.7 Footnotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.2.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.2.9 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2.3 Gallium arsenide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
CONTENTS iii

2.3.1 Preparation and chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

2.3.2 Comparison with silicon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.3.3 Other applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.3.4 Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.3.5 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.3.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
2.3.7 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

3 Applications 63
3.1 Voltage-controlled oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
3.1.1 Types of VCO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
3.1.2 Control of frequency in VCOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.1.3 Voltage-controlled crystal oscillators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.1.4 VCO design and circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.1.5 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.1.6 Voltage-controlled crystal oscillator as a clock generator . . . . . . . . . . . . . . . . . . . 66
3.1.7 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.1.8 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.1.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.1.10 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.2 Frequency-shift keying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.2.1 Implementations of FSK Modems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.2.2 Other forms of FSK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.2.3 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.2.4 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.2.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.2.6 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.3 Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.3.1 Figures of merit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.3.2 Amplifier types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.3.3 Classification of amplifier stages and systems . . . . . . . . . . . . . . . . . . . . . . . . 73
3.3.4 Power amplifier classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.3.5 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.3.6 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.3.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.3.8 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
iv CONTENTS

4 Background Theory 89
4.1 Electron hole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.1.1 Solid-state physics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.1.2 Holes in quantum chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.1.3 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.1.4 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.2 P–n junction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.2.1 Properties of a p–n junction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.2.2 Equilibrium (zero bias) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.2.3 Forward bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
4.2.4 Reverse bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.2.5 Governing Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.2.7 Non-rectifying junctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.2.8 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.2.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.2.10 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.2.11 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.3 Bipolar transistor biasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.3.1 Bias circuit requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.3.2 Types of bias circuit for Class A amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.3.3 Class B and AB amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4.3.4 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.3.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.3.6 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.3.7 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

5 Common Integrated Circuits 104

5.1 555 timer IC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
5.1.1 Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
5.1.2 Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5.1.3 Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5.1.4 Example applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
5.1.5 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
5.1.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
5.1.7 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
CONTENTS v

5.1.8 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

5.2 Operational amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.2.1 Circuit notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.2.2 Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.2.3 Op-amp characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.2.4 Internal circuitry of 741-type op-amp . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5.2.5 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.2.6 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.2.7 Historical timeline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.2.8 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
5.2.9 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
5.2.10 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
5.2.11 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.2.12 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.3 Phase-locked loop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.3.1 Practical analogies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
5.3.2 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
5.3.3 Structure and function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5.3.4 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
5.3.5 Block diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.3.6 Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.3.7 Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.3.8 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
5.3.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
5.3.10 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
5.4 Voltage regulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
5.4.1 Measures of regulator quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
5.4.2 Electronic voltage regulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5.4.3 Electromechanical regulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.4.4 Automatic voltage regulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
5.4.5 AC voltage stabilizers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
5.4.6 DC voltage stabilizers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
5.4.7 Active regulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
5.4.8 Example linear regulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
5.4.9 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
vi CONTENTS

5.4.10 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

5.4.11 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
5.5 Comparator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
5.5.1 Differential Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
5.5.2 Op-amp voltage comparator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
5.5.3 Working . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
5.5.4 Key specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
5.5.5 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
5.5.6 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
5.5.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
5.5.8 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

6 Sensors 151
6.1 Thermistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
6.1.1 Basic operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
6.1.2 Steinhart–Hart equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
6.1.3 B or β parameter equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
6.1.4 Conduction model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
6.1.5 Self-heating effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
6.1.6 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
6.1.7 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
6.1.8 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
6.1.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
6.1.10 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
6.2 Photodiode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
6.2.1 Principle of operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
6.2.2 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
6.2.3 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
6.2.4 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
6.2.5 Photodiode array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
6.2.6 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
6.2.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
6.2.8 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
6.3 Photoresistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
6.3.1 Design considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
6.3.2 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
CONTENTS vii

6.3.3 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

6.3.4 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
6.3.5 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
6.4 Analogue switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
6.4.1 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

7 The Decibel 163

7.1 Decibel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
7.1.1 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
7.1.2 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
7.1.3 Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
7.1.4 Advantages and disadvantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
7.1.5 Uses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
7.1.6 Suffixes and reference values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
7.1.7 Related units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
7.1.8 Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
7.1.9 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
7.1.10 Notes and references . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
7.1.11 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.2 Noise (electronics) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.2.1 Noise types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.2.2 Coupled noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
7.2.3 Quantification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
7.2.4 Dither . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.2.5 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.2.6 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.2.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.2.8 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.2.9 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.3 Switched capacitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.3.1 The switched capacitor resistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.3.2 The Parasitic Sensitive Integrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
7.3.3 The Parasitic Insensitive Integrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
7.3.4 The Multiplying Digital to Analog Converter . . . . . . . . . . . . . . . . . . . . . . . . 180
7.3.5 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
7.3.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
viii CONTENTS

7.4 H bridge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

7.4.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
7.4.2 Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
7.4.3 Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
7.4.4 Operation as an inverter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
7.4.5 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
7.4.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
7.4.7 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
7.5 Hall effect sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
7.5.1 Hall probe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
7.5.2 Working principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
7.5.3 Materials for Hall effect sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
7.5.4 Signal processing and interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
7.5.5 Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
7.5.6 Disadvantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
7.5.7 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
7.5.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
7.5.9 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186

8 Filters 187
8.1 Low-pass filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
8.1.1 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
8.1.2 Ideal and real filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
8.1.3 Continuous-time low-pass filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
8.1.4 Electronic low-pass filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
8.1.5 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
8.1.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
8.1.7 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
8.2 High-pass filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
8.2.1 First-order continuous-time implementation . . . . . . . . . . . . . . . . . . . . . . . . . 192
8.2.2 Discrete-time realization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
8.2.3 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
8.2.4 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
8.2.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
8.2.6 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
8.3 Band-pass filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
CONTENTS ix

8.3.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196

8.3.2 Q-factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
8.3.3 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
8.3.4 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
8.3.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
8.3.6 External links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

9 Text and image sources, contributors, and licenses 198

9.1 Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
9.2 Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
9.3 Content license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
Chapter 1

The Transistor

1.1 Transistor changes the current through another pair of terminals.

For other uses, see Transistor (disambiguation).
A transistor is a semiconductor device used to amplify signal.
Because the controlled (output) power can be higher than
the controlling (input) power, a transistor can amplify a
Today, some transistors are packaged individu-
ally, but many more are found embedded in integrated
circuits.
The transistor is the fundamental building block of mod-
ern electronic devices, and is ubiquitous in modern elec-
tronic systems. Following its development in 1947 by
American physicists John Bardeen, Walter Brattain, and
William Shockley, the transistor revolutionized the field
of electronics, and paved the way for smaller and cheaper
radios, calculators, and computers, among other things.
The transistor is on the list of IEEE milestones in elec-
tronics, and the inventors were jointly awarded the 1956
Nobel Prize in Physics for their achievement.

1.1.1 History

Main article: History of the transistor

Assorted discrete transistors. Packages in order from top to bot-
tom: TO-3, TO-126, TO-92, SOT-23
and switch electronic signals and electrical power. It is
composed of semiconductor material with at least three
terminals for connection to an external circuit. A voltage
or current applied to one pair of the transistor’s terminals
The thermionic triode, a vacuum tube invented in 1907,
propelled the electronics age forward, enabling amplified
radio technology and long-distance telephony. The tri-
ode, however, was a fragile device that consumed a lot
of power. Physicist Julius Edgar Lilienfeld filed a patent
for a field-effect transistor (FET) in Canada in 1925,
which was intended to be a solid-state replacement for
the triode.[1][2] Lilienfeld also filed identical patents in the
United States in 1926[3] and 1928.[4][5] However, Lilien-
feld did not publish any research articles about his devices
nor did his patents cite any specific examples of a working
prototype. Because the production of high-quality semi-
conductor materials was still decades away, Lilienfeld’s
solid-state amplifier ideas would not have found practi-
cal use in the 1920s and 1930s, even if such a device

1
2 CHAPTER 1. THE TRANSISTOR

sresistance.[9][10][11] According to Lillian Hoddeson and

Vicki Daitch, authors of a biography of John Bardeen,
Shockley had proposed that Bell Labs’ first patent for a
transistor should be based on the field-effect and that he
be named as the inventor. Having unearthed Lilienfeld’s
patents that went into obscurity years earlier, lawyers at
Bell Labs advised against Shockley’s proposal because
the idea of a field-effect transistor that used an electric
field as a “grid” was not new. Instead, what Bardeen,
Brattain, and Shockley invented in 1947 was the first
point-contact transistor.[6] In acknowledgement of this
accomplishment, Shockley, Bardeen, and Brattain were
jointly awarded the 1956 Nobel Prize in Physics “for their
researches on semiconductors and their discovery of the
transistor effect.”[12]
In 1948, the point-contact transistor was independently
A replica of the first working transistor. invented by German physicists Herbert Mataré and
Heinrich Welker while working at the Compagnie des
Freins et Signaux, a Westinghouse subsidiary located
had been built.[6] In 1934, German inventor Oskar Heil in Paris. Mataré had previous experience in develop-
patented a similar device.[7] ing crystal rectifiers from silicon and germanium in the
German radar effort during World War II. Using this
knowledge, he began researching the phenomenon of
extquotedblinterference extquotedbl in 1947. By wit-
nessing currents flowing through point-contacts, similar
to what Bardeen and Brattain had accomplished earlier in
December 1947, Mataré by June 1948, was able to pro-
duce consistent results by using samples of germanium
produced by Welker. Realizing that Bell Labs’ scientists
had already invented the transistor before them, the com-
pany rushed to get its “transistron” into production for
amplified use in France’s telephone network.[13]
The first high-frequency transistor was the surface-barrier
germanium transistor developed by Philco in 1953, ca-
pable of operating up to 60 MHz.[14] These were made
by etching depressions into an N-type germanium base
John Bardeen, William Shockley and Walter Brattain at Bell from both sides with jets of Indium(III) sulfate until it
Labs, 1948. was a few ten-thousandths of an inch thick. Indium elec-
troplated into the depressions formed the collector and
From November 17, 1947 to December 23, 1947, John emitter.[15][16] The first all-transistor car radio, which was
Bardeen and Walter Brattain at AT&T's Bell Labs in the produced in 1955 by Chrysler and Philco, used these
United States, performed experiments and observed that transistors in its circuitry and also they were the first suit-
when two gold point contacts were applied to a crystal of able for high-speed computers.[17][18][19][20]
germanium, a signal was produced with the output power The first working silicon transistor was developed at Bell
greater than the input.[8] Solid State Physics Group leader Labs on January 26, 1954 by Morris Tanenbaum.[21] The
William Shockley saw the potential in this, and over the first commercial silicon transistor was produced by Texas
next few months worked to greatly expand the knowl- Instruments in 1954.[22] This was the work of Gordon
edge of semiconductors. The term transistor was coined Teal, an expert in growing crystals of high purity, who
by John R. Pierce as a portmanteau of the term tran-
1.1. TRANSISTOR 3

A Darlington transistor opened up so the actual transistor chip

(the small square) can be seen inside. A Darlington transistor
is effectively two transistors on the same chip. One transistor is
much larger than the other, but both are large in comparison to
transistors in large-scale integration because this particular ex-
ample is intended for power applications.

Philco surface-barrier transistor developed and produced in built in 2002 ... for [each] man, woman, and child on
1953 Earth.”[29]
The transistor’s low cost, flexibility, and reliability have
made it a ubiquitous device. Transistorized mechatronic
had previously worked at Bell Labs.[23] The first MOS
circuits have replaced electromechanical devices in con-
transistor actually built was by Kahng and Atalla at Bell
trolling appliances and machinery. It is often easier and
Labs in 1960.[24]
cheaper to use a standard microcontroller and write a
computer program to carry out a control function than
to design an equivalent mechanical control function.
1.1.2 Importance

The transistor is the key active component in practically
all modern electronics. Many consider it to be one of the
greatest inventions of the 20th century.[25] Its importance
in today’s society rests on its ability to be mass-produced
using a highly automated process (semiconductor device
fabrication) that achieves astonishingly low per-transistor
costs. The invention of the first transistor at Bell Labs
was named an IEEE Milestone in 2009.[26]
Although several companies each produce over a billion
individually packaged (known as discrete) transistors ev-
ery year,[27] the vast majority of transistors are now pro-
duced in integrated circuits (often shortened to IC, mi-
crochips or simply chips), along with diodes, resistors,
capacitors and other electronic components, to produce
complete electronic circuits. A logic gate consists of up
to about twenty transistors whereas an advanced micro-
processor, as of 2009, can use as many as 3 billion transis-
tors (MOSFETs).[28] “About 60 million transistors were
1.1.3 Simplified operation
The essential usefulness of a transistor comes from its
ability to use a small signal applied between one pair of
its terminals to control a much larger signal at another pair
of terminals. This property is called gain. It can produce
a stronger output signal, a voltage or current, that is pro-
portional to a weaker input signal; that is, it can act as
an amplifier. Alternatively, the transistor can be used to
turn current on or off in a circuit as an electrically con-
trolled switch, where the amount of current is determined
by other circuit elements.
There are two types of transistors, which have slight dif-
ferences in how they are used in a circuit. A bipolar tran-
sistor has terminals labeled base, collector, and emitter.
A small current at the base terminal (that is, flowing be-
tween the base and the emitter) can control or switch a
much larger current between the collector and emitter
4 CHAPTER 1. THE TRANSISTOR

VCC
IBE +6V

VOUT 1k
collector
ICE
base
VIN
emitter

BJT used as an electronic switch, in grounded-emitter configura-

tion.

cause current is flowing from collector to emitter freely.

A simple circuit diagram to show the labels of a n–p–n bipolar When saturated, the switch is said to be on.[30]
transistor.
Providing sufficient base drive current is a key problem in
the use of bipolar transistors as switches. The transistor
terminals. For a field-effect transistor, the terminals are provides current gain, allowing a relatively large current
labeled gate, source, and drain, and a voltage at the gate in the collector to be switched by a much smaller cur-
can control a current between source and drain. rent into the base terminal. The ratio of these currents
varies depending on the type of transistor, and even for a
The image to the right represents a typical bipolar tran-
particular type, varies depending on the collector current.
sistor in a circuit. Charge will flow between emitter and
In the example light-switch circuit shown, the resistor is
collector terminals depending on the current in the base.
chosen to provide enough base current to ensure the tran-
Because internally the base and emitter connections be-
sistor will be saturated.
have like a semiconductor diode, a voltage drop develops
between base and emitter while the base current exists. In any switching circuit, values of input voltage would be
The amount of this voltage depends on the material the chosen such that the output is either completely off,[31] or
transistor is made from, and is referred to as VBE. completely on. The transistor is acting as a switch, and
this type of operation is common in digital circuits where
only “on” and “off” values are relevant.
Transistor as a switch
Transistor as an amplifier
Transistors are commonly used as electronic switches,
both for high-power applications such as switched-mode
The common-emitter amplifier is designed so that a small
power supplies and for low-power applications such as
logic gates. change in voltage (Vᵢ ) changes the small current through
the base of the transistor; the transistor’s current amplifi-
In a grounded-emitter transistor circuit, such as the light- cation combined with the properties of the circuit mean
switch circuit shown, as the base voltage rises, the emitter that small swings in Vᵢ produce large changes in Vₒᵤ .
and collector currents rise exponentially. The collector
voltage drops because of reduced resistance from collec- Various configurations of single transistor amplifier are
tor to emitter. If the voltage difference between the col- possible, with some providing current gain, some voltage
lector and emitter were zero (or near zero), the collector gain, and some both.
current would be limited only by the load resistance (light From mobile phones to televisions, vast numbers of prod-
bulb) and the supply voltage. This is called saturation be- ucts include amplifiers for sound reproduction, radio
1.1. TRANSISTOR 5

to a simple electrical heating element, much like a

V+ light bulb filament.
• Small size and minimal weight, allowing the devel-
opment of miniaturized electronic devices.

RC • Low operating voltages compatible with batteries of

R1 Vout only a few cells.
• No warm-up period for cathode heaters required af-
C
Vin B
Cout ter power application.
• Lower power dissipation and generally greater en-
ergy efficiency.
Cin E • Higher reliability and greater physical ruggedness.
• Extremely long life. Some transistorized devices
R2 RE CE have been in service for more than 50 years.
• Complementary devices available, facilitating the
design of complementary-symmetry circuits, some-
thing not possible with vacuum tubes.
• Greatly reduced sensitivity to mechanical shock
Amplifier circuit, common-emitter configuration with a voltage-
and vibration, thus reducing the problem of
divider bias circuit.
microphonics in sensitive applications, such as au-
dio.
transmission, and signal processing. The first discrete-
transistor audio amplifiers barely supplied a few hun-
dred milliwatts, but power and audio fidelity gradually Limitations
increased as better transistors became available and am-
plifier architecture evolved. • Silicon transistors can age and fail.[32]

Modern transistor audio amplifiers of up to a few hundred
watts are common and relatively inexpensive.
1.1.4 Comparison with vacuum tubes
Prior to the development of transistors, vacuum (elec-
tron) tubes (or in the UK “thermionic valves” or just
“valves”) were the main active components in electronic
equipment.
Advantages
The key advantages that have allowed transistors to re-
place their vacuum tube predecessors in most applica-
tions are
• No power consumption by a cathode heater; the
characteristic orange glow of vacuum tubes is due
• High-power, high-frequency operation, such as that
used in over-the-air television broadcasting, is better
achieved in vacuum tubes due to improved electron
mobility in a vacuum.
• Solid-state devices are more vulnerable to
electrostatic discharge in handling and opera-
tion
• A vacuum tube momentarily overloaded will just get
a little hotter; solid-state devices have less mass to
absorb the heat due to overloads, in proportion to
their rating
• Sensitivity to radiation and cosmic rays (special
radiation-hardened chips are used for spacecraft de-
vices).
• Vacuum tubes create a distortion, the so-called tube
sound, that some people find to be more tolerable to
the ear.[33]
6 CHAPTER 1. THE TRANSISTOR

1.1.5 Types Bipolar transistors are so named because they conduct

by using both majority and minority carriers. The bipo-
BJT and JFET symbols lar junction transistor, the first type of transistor to be
JFET and IGFET symbols mass-produced, is a combination of two junction diodes,
and is formed of either a thin layer of p-type semicon-
ductor sandwiched between two n-type semiconductors
Transistors are categorized by
(an n–p–n transistor), or a thin layer of n-type semicon-
ductor sandwiched between two p-type semiconductors
• Semiconductor material (date first used): the (a p–n–p transistor). This construction produces two p–n
metalloids germanium (1947) and silicon (1954)— junctions: a base–emitter junction and a base–collector
in amorphous, polycrystalline and monocrystalline junction, separated by a thin region of semiconductor
form; the compounds gallium arsenide (1966) and known as the base region (two junction diodes wired to-
silicon carbide (1997), the alloy silicon-germanium gether without sharing an intervening semiconducting re-
(1989), the allotrope of carbon graphene (research gion will not make a transistor).
ongoing since 2004), etc.—see Semiconductor ma-
terial BJTs have three terminals, corresponding to the three
layers of semiconductor—an emitter, a base, and a col-
• Structure: BJT, JFET, IGFET (MOSFET), lector. They are useful in amplifiers because the cur-
insulated-gate bipolar transistor, “other types” rents at the emitter and collector are controllable by a
relatively small base current.”[35] In an n–p–n transistor
• Electrical polarity (positive and negative): n–p–n, operating in the active region, the emitter–base junction
p–n–p (BJTs); n-channel, p-channel (FETs) is forward biased (electrons and holes recombine at the
junction), and electrons are injected into the base re-
• Maximum power rating: low, medium, high gion. Because the base is narrow, most of these elec-
trons will diffuse into the reverse-biased (electrons and
• Maximum operating frequency: low, medium, high, holes are formed at, and move away from the junction)
radio (RF), microwave frequency (the maximum ef- base–collector junction and be swept into the collector;
fective frequency of a transistor is denoted by the perhaps one-hundredth of the electrons will recombine in
term fT , an abbreviation for transition frequency— the base, which is the dominant mechanism in the base
the frequency of transition is the frequency at which current. By controlling the number of electrons that can
the transistor yields unity gain) leave the base, the number of electrons entering the col-
lector can be controlled.[35] Collector current is approx-
• Application: switch, general purpose, audio, high imately β (common-emitter current gain) times the base
voltage, super-beta, matched pair current. It is typically greater than 100 for small-signal
transistors but can be smaller in transistors designed for
• Physical packaging: through-hole metal, through-
high-power applications.
hole plastic, surface mount, ball grid array, power
modules—see Packaging Unlike the field-effect transistor (see below), the BJT is a
low–input-impedance device. Also, as the base–emitter
• Amplification factor h ₑ, βF (transistor beta)[34] or voltage (Vbe) is increased the base–emitter current and
g (transconductance). hence the collector–emitter current (Ice) increase expo-
nentially according to the Shockley diode model and the
Thus, a particular transistor may be described as silicon, Ebers-Moll model. Because of this exponential relation-
surface-mount, BJT, n–p–n, low-power, high-frequency ship, the BJT has a higher transconductance than the
switch. FET.
Bipolar transistors can be made to conduct by exposure
to light, because absorption of photons in the base re-
Bipolar junction transistor (BJT) gion generates a photocurrent that acts as a base cur-
rent; the collector current is approximately β times the
Main article: Bipolar junction transistor photocurrent. Devices designed for this purpose have
1.1. TRANSISTOR 7

a transparent window in the package and are called metal–semiconductor junction. These, and the HEMTs
phototransistors. (high-electron-mobility transistors, or HFETs), in which
a two-dimensional electron gas with very high carrier mo-
bility is used for charge transport, are especially suitable
Field-effect transistor (FET) for use at very high frequencies (microwave frequencies;
several GHz).
Main articles: Field-effect transistor, MOSFET and
FETs are further divided into depletion-mode and
JFET
enhancement-mode types, depending on whether the
channel is turned on or off with zero gate-to-source volt-
The field-effect transistor, sometimes called a unipolar age. For enhancement mode, the channel is off at zero
transistor, uses either electrons (in n-channel FET) or bias, and a gate potential can “enhance” the conduction.
holes (in p-channel FET) for conduction. The four termi- For the depletion mode, the channel is on at zero bias, and
nals of the FET are named source, gate, drain, and body a gate potential (of the opposite polarity) can “deplete”
(substrate). On most FETs, the body is connected to the the channel, reducing conduction. For either mode, a
source inside the package, and this will be assumed for more positive gate voltage corresponds to a higher current
the following description. for n-channel devices and a lower current for p-channel
In a FET, the drain-to-source current flows via a conduct- devices. Nearly all JFETs are depletion-mode because
ing channel that connects the source region to the drain the diode junctions would forward bias and conduct if
region. The conductivity is varied by the electric field that they were enhancement-mode devices; most IGFETs are
is produced when a voltage is applied between the gate enhancement-mode types.
and source terminals; hence the current flowing between
the drain and source is controlled by the voltage applied Usage of bipolar and field-effect transistors
between the gate and source. As the gate–source volt-
age (Vgs) is increased, the drain–source current (Ids) in- The bipolar junction transistor (BJT) was the most com-
creases exponentially for Vgs below threshold, and then at monly used transistor in the 1960s and 70s. Even after
a roughly quadratic rate ( Ids ∝ (Vgs −VT )2 ) (where VT MOSFETs became widely available, the BJT remained
is the threshold voltage at which drain current begins)[36]the transistor of choice for many analog circuits such as
in the extquotedblspace-charge-limited extquotedbl re- amplifiers because of their greater linearity and ease of
gion above threshold. A quadratic behavior is not ob- manufacture. In integrated circuits, the desirable prop-
served in modern devices, for example, at the 65 nm tech- erties of MOSFETs allowed them to capture nearly all
nology node.[37] market share for digital circuits. Discrete MOSFETs can
For low noise at narrow bandwidth the higher input resis- be applied in transistor applications, including analog cir-
tance of the FET is advantageous. cuits, voltage regulators, amplifiers, power transmitters
and motor drivers.
FETs are divided into two families: junction FET (JFET)
and insulated gate FET (IGFET). The IGFET is more
commonly known as a metal–oxide–semiconductor FET Other transistor types
(MOSFET), reflecting its original construction from lay-
ers of metal (the gate), oxide (the insulation), and semi- For early bipolar transistors, see Bipolar junction tran-
conductor. Unlike IGFETs, the JFET gate forms a p–n sistor#Bipolar transistors.
diode with the channel which lies between the source and
drain. Functionally, this makes the n-channel JFET the
solid-state equivalent of the vacuum tube triode which, • Bipolar junction transistor
similarly, forms a diode between its grid and cathode.
Also, both devices operate in the depletion mode, they --- Heterojunction bipolar transistor, up to sev-
both have a high input impedance, and they both conduct eral hundred GHz, common in modern ultra-
current under the control of an input voltage. fast and RF circuits
Metal–semiconductor FETs (MESFETs) are JFETs in --- Schottky transistor
which the reverse biased p–n junction is replaced by a --- Avalanche transistor
8 CHAPTER 1. THE TRANSISTOR
Transistor symbol drawn on Portuguese pavement in the
University of Aveiro.
--- Darlington transistors are two BJTs connected
together to provide a high current gain equal
to the product of the current gains of the two
transistors.
--- Insulated-gate bipolar transistors (IGBTs) use
a medium-power IGFET, similarly connected
to a power BJT, to give a high input
impedance. Power diodes are often connected
between certain terminals depending on spe-
cific use. IGBTs are particularly suitable for
heavy-duty industrial applications. The Asea
Brown Boveri (ABB) 5SNA2400E170100 il-
lustrates just how far power semiconduc-
tor technology has advanced.[38] Intended for
three-phase power supplies, this device houses
three n–p–n IGBTs in a case measuring 38 by
140 by 190 mm and weighing 1.5 kg. Each
IGBT is rated at 1,700 volts and can handle
2,400 amperes.
--- Photo transistor
--- Multiple-emitter transistor, used in transistor–
transistor logic
--- Multiple-base transistor, used to amplify very-
low-level signals in noisy environments such as
the pickup of a record player or radio front
ends. Effectively, it is a very large number
of transistors in parallel where, at the output,
the signal is added constructively, but random
noise is added only stochastically.[39]
• Field-effect transistor
--- Carbon nanotube field-effect transistor (CN-
FET)
--- JFET, where the gate is insulated by a reverse-
biased p–n junction
--- MESFET, similar to JFET with a Schottky
junction instead of a p–n junction
∗ High-electron-mobility transistor
(HEMT, HFET, MODFET)
--- MOSFET, where the gate is insulated by a
shallow layer of insulator
--- Inverted-T field-effect transistor (ITFET)
--- FinFET, source/drain region shapes fins on the
silicon surface.
--- FREDFET, fast-reverse epitaxial diode field-
effect transistor
--- Thin-film transistor, in LCDs.
--- Organic field-effect transistor (OFET), in
which the semiconductor is an organic com-
pound
--- Ballistic transistor
--- Floating-gate transistor, for non-volatile stor-
age.
--- FETs used to sense environment
∗ Ion-sensitive field effect transistor (IF-
SET), to measure ion concentrations in
solution.
∗ EOSFET, electrolyte-oxide-
semiconductor field-effect transistor
(Neurochip)
∗ DNAFET, deoxyribonucleic acid field-
effect transistor
• Diffusion transistor, formed by diffusing dopants
into semiconductor substrate; can be both BJT and
FET
• Unijunction transistors can be used as simple pulse
generators. They comprise a main body of either P-
type or N-type semiconductor with ohmic contacts
at each end (terminals Base1 and Base2). A junction
with the opposite semiconductor type is formed at
a point along the length of the body for the third
terminal (Emitter).
• Single-electron transistors (SET) consist of a gate
island between two tunneling junctions. The tun-
neling current is controlled by a voltage applied to
the gate through a capacitor.[40]
1.1. TRANSISTOR 9

• Nanofluidic transistor, controls the movement standards; in each the alphanumeric prefix provides clues
of ions through sub-microscopic, water-filled to type of the device.
channels.[41]

• Multigate devices
--- Tetrode transistor
--- Pentode transistor
--- Trigate transistors (Prototype by Intel)
--- Dual-gate FETs have a single channel with
two gates in cascode; a configuration opti-
mized for high-frequency amplifiers, mixers,
and oscillators.
• Junctionless nanowire transistor (JNT), developed
at Tyndall National Institute in Ireland, was the first
transistor successfully fabricated without junctions.
(Even MOSFETs have junctions, although its gate is
electrically insulated from the region the gate con-
trols.) Junctions are difficult and expensive to fabri-
cate, and, because they are a significant source of
current leakage, they waste significant power and
generate significant waste heat. Eliminating them
held the promise of cheaper and denser microchips.
The JNT uses a simple nanowire of silicon sur-
rounded by an electrically isolated “wedding ring”
that acts to gate the flow of electrons through the
wire. This method has been described as akin to
squeezing a garden hose to gate the flow of wa-
ter through the hose. The nanowire is heavily n-
doped, making it an excellent conductor. Crucially
the gate, comprising silicon, is heavily p-doped; and
its presence depletes the underlying silicon nanowire
thereby preventing carrier flow past the gate.
Japanese Industrial Standard (JIS)
The JIS-C-7012 specification for transistor part numbers
starts with “2S”,[43] e.g. 2SD965, but sometimes the “2S”
prefix is not marked on the package – a 2SD965 might
only be marked “D965 extquotedbl; a 2SC1815 might be
listed by a supplier as simply “C1815”. This series some-
times has suffixes (such as “R”, “O”, “BL”... standing for
“Red”, “Orange”, “Blue” etc.) to denote variants, such as
tighter hFE (gain) groupings.
European Electronic Component Manufacturers As-
sociation (EECA)
The Pro Electron standard, the European Electronic
Component Manufacturers Association part numbering
scheme, begins with two letters: the first gives the semi-
conductor type (A for germanium, B for silicon, and C
for materials like GaAs); the second letter denotes the
intended use (A for diode, C for general-purpose tran-
sistor, etc.). A 3-digit sequence number (or one letter
then 2 digits, for industrial types) follows. With early de-
vices this indicated the case type. Suffixes may be used,
with a letter (e.g. “C” often means high hFE, such as in:
BC549C[44] ) or other codes may follow to show gain (e.g.
BC327-25) or voltage rating (e.g. BUK854-800A[45] ).
The more common prefixes are:

• Vacuum-channel transistor: In 2012, NASA and the

National Nanofab Center in South Korea were re-
ported to have built a prototype vacuum-channel Joint Electron Devices Engineering Council
transistor in only 150 nanometers in size, can be (JEDEC)
manufactured cheaply using standard silicon semi-
conductor processing, can operate at high speeds The JEDEC EIA370 transistor device numbers usually
even in hostile environments, and could consume start with “2N”, indicating a three-terminal device (dual-
just as much power as a standard transistor.[42] gate field-effect transistors are four-terminal devices, so
begin with 3N), then a 2, 3 or 4-digit sequential num-
ber with no significance as to device properties (although
1.1.6 Part numbering standards / specifi- early devices with low numbers tend to be germanium).
cations For example 2N3055 is a silicon n–p–n power transistor,
2N1301 is a p–n–p germanium switching transistor. A
The types of some transistors can be parsed from the part letter suffix (such as “A”) is sometimes used to indicate
number. There are three major semiconductor naming a newer variant, but rarely gain groupings.
10 CHAPTER 1. THE TRANSISTOR

Proprietary Semiconductor material

Manufacturers of devices may have their own proprietary The first BJTs were made from germanium (Ge). Silicon
numbering system, for example CK722. Since devices (Si) types currently predominate but certain advanced
are second-sourced, a manufacturer’s prefix (like “MPF” microwave and high-performance versions now employ
in MPF102, which originally would denote a Motorola the compound semiconductor material gallium arsenide
FET) now is an unreliable indicator of who made the (GaAs) and the semiconductor alloy silicon germanium
device. Some proprietary naming schemes adopt parts (SiGe). Single element semiconductor material (Ge and
of other naming schemes, for example a PN2222A is a Si) is described as elemental.
(possibly Fairchild Semiconductor) 2N2222A in a plas- Rough parameters for the most common semiconductor
tic case (but a PN108 is a plastic version of a BC108, not
materials used to make transistors are given in the table to
a 2N108, while the PN100 is unrelated to other xx100 the right; these parameters will vary with increase in tem-
devices). perature, electric field, impurity level, strain, and sundry
Military part numbers sometimes are assigned their own other factors.
codes, such as the British Military CV Naming System. The junction forward voltage is the voltage applied to the
Manufacturers buying large numbers of similar parts may emitter–base junction of a BJT in order to make the base
have them supplied with “house numbers”, identifying conduct a specified current. The current increases ex-
a particular purchasing specification and not necessar- ponentially as the junction forward voltage is increased.
ily a device with a standardized registered number. For The values given in the table are typical for a current of
example, an HP part 1854,0053 is a (JEDEC) 2N2218 1 mA (the same values apply to semiconductor diodes).
transistor[46][47] which is also assigned the CV number: The lower the junction forward voltage the better, as this
CV7763[48] means that less power is required to “drive” the transis-
tor. The junction forward voltage for a given current de-
creases with increase in temperature. For a typical silicon
junction the change is −2.1 mV/°C.[49] In some circuits
Naming problems special compensating elements (sensistors) must be used
to compensate for such changes.
With so many independent naming schemes, and the ab- The density of mobile carriers in the channel of a MOS-
breviation of part numbers when printed on the devices, FET is a function of the electric field forming the chan-
ambiguity sometimes occurs. For example two different nel and of various other phenomena such as the impurity
devices may be marked “J176” (one the J176 low-power level in the channel. Some impurities, called dopants, are
Junction FET, the other the higher-powered MOSFET introduced deliberately in making a MOSFET, to control
2SJ176). the MOSFET electrical behavior.
As older “through-hole” transistors are given surface- The electron mobility and hole mobility columns show the
mount packaged counterparts, they tend to be assigned average speed that electrons and holes diffuse through the
many different part numbers because manufacturers have semiconductor material with an electric field of 1 volt per
their own systems to cope with the variety in pinout ar- meter applied across the material. In general, the higher
rangements and options for dual or matched n–p–n+p–n– the electron mobility the faster the transistor can operate.
p devices in one pack. So even when the original device The table indicates that Ge is a better material than Si in
(such as a 2N3904) may have been assigned by a stan- this respect. However, Ge has four major shortcomings
dards authority, and well known by engineers over the compared to silicon and gallium arsenide:
years, the new versions are far from standardized in their
naming. • Its maximum temperature is limited;

• it has relatively high leakage current;

• it cannot withstand high voltages;

1.1.7 Construction
• it is less suitable for fabricating integrated circuits.
1.1. TRANSISTOR 11

Because the electron mobility is higher than the hole mo- through-hole (or leaded), and surface-mount, also known
bility for all semiconductor materials, a given bipolar n– as surface-mount device (SMD). The ball grid array
p–n transistor tends to be swifter than an equivalent p– (BGA) is the latest surface-mount package (currently
n–p transistor. GaAs has the highest electron mobility of only for large integrated circuits). It has solder “balls”
the three semiconductors. It is for this reason that GaAs on the underside in place of leads. Because they are
is used in high-frequency applications. A relatively re- smaller and have shorter interconnections, SMDs have
cent FET development, the high-electron-mobility tran- better high-frequency characteristics but lower power rat-
sistor (HEMT), has a heterostructure (junction between ing.
different semiconductor materials) of aluminium gal- Transistor packages are made of glass, metal, ceramic, or
lium arsenide (AlGaAs)-gallium arsenide (GaAs) which plastic. The package often dictates the power rating and
has twice the electron mobility of a GaAs-metal barrier frequency characteristics. Power transistors have larger
junction. Because of their high speed and low noise, packages that can be clamped to heat sinks for enhanced
HEMTs are used in satellite receivers working at frequen- cooling. Additionally, most power transistors have the
cies around 12 GHz. HEMTs based on gallium nitride collector or drain physically connected to the metal en-
and aluminium gallium nitride (AlGaN/GaN HEMTs) closure. At the other extreme, some surface-mount mi-
provide a still higher electron mobility and are being de- crowave transistors are as small as grains of sand.
veloped for various applications.
Often a given transistor type is available in several pack-
Max. junction temperature values represent a cross ages. Transistor packages are mainly standardized, but
section taken from various manufacturers’ data sheets. the assignment of a transistor’s functions to the terminals
This temperature should not be exceeded or the transistoris not: other transistor types can assign other functions
may be damaged. to the package’s terminals. Even for the same transis-
Al–Si junction refers to the high-speed (aluminum– tor type the terminal assignment can vary (normally indi-
silicon) metal–semiconductor barrier diode, commonly cated by a suffix letter to the part number, q.e. BC212L
known as a Schottky diode. This is included in the ta- and BC212K).
ble because some silicon power IGFETs have a parasitic
reverse Schottky diode formed between the source and
drain as part of the fabrication process. This diode can Flexible transistors Researchers have made several
be a nuisance, but sometimes it is used in the circuit. kinds of flexible transistors, including organic field-effect
transistors.[50][51][52] Flexible transistors are useful in
some kinds of flexible displays and other flexible elec-
Packaging tronics.

See also: Semiconductor package and Chip carrier

Discrete transistors are individually packaged transis- 1.1.8 See also

• Band gap

• Digital electronics

• Moore’s law

• Semiconductor device modeling

• Transistor count

• Transistor model
Assorted discrete transistors
• Transresistance
tors. Transistors come in many different semiconductor
packages (see image). The two main categories are • Very-large-scale integration
12 CHAPTER 1. THE TRANSISTOR

1.1.9 Directory of external websites with 1.1.10 References

datasheets
• 2N3904/2N3906, BC182/BC212 and
BC546/BC556: Ubiquitous, BJT, general-purpose,
low-power, complementary pairs. They have plas-
tic cases and cost roughly ten cents U.S. in small
quantities, making them popular with hobbyists.
• AF107: Germanium, 0.5 watt, 250 MHz p–n–p
BJT.
• BFP183: Low-power, 8 GHz microwave n–p–n
BJT.
• LM394: “supermatch pair”, with two n–p–n BJTs
on a single substrate.
• 2N2219A/2N2905A: BJT, general purpose,
medium power, complementary pair. With metal
cases they are rated at about one watt.
• 2N3055/MJ2955: For years, the n–p–n 2N3055 has
been the “standard” power transistor. Its comple-
ment, the p–n–p MJ2955 arrived later. These 1
MHz, 15 A, 60 V, 115 W BJTs are used in audio-
power amplifiers, power supplies, and control.
[1] Vardalas, John, Twists and Turns in the Development of
the Transistor IEEE-USA Today’s Engineer, May 2003.
[2] Lilienfeld, Julius Edgar, “Method and apparatus for con-
trolling electric current” U.S. Patent 1,745,175 1930-01-
28 (filed in Canada 1925-10-22, in US 1926-10-08).
[3] “Method And Apparatus For Controlling Electric Cur-
rents”. United States Patent and Trademark Office.
[4] “Amplifier For Electric Currents”. United States Patent
and Trademark Office.
[5] “Device For Controlling Electric Current”. United States
Patent and Trademark Office.
[6] “Twists and Turns in the Development of the Transistor”.
Institute of Electrical and Electronics Engineers, Inc.
[7] Heil, Oskar, “Improvements in or relating to electrical
amplifiers and other control arrangements and devices”,
Patent No. GB439457, European Patent Office, filed in
Great Britain 1934-03-02, published 1935-12-06 (origi-
nally filed in Germany 1934-03-02).
[8] “November 17 – December 23, 1947: Invention of the
First Transistor”. American Physical Society.

• 2SC3281/2SA1302: Made by Toshiba, these BJTs [9] Bell Laboratories (1983). S. Millman, ed. A History of
have low-distortion characteristics and are used in Engineering and Science in the Bell System, Physical Sci-
high-power audio amplifiers. They have been widely ence (1925-1980). AT&T Bell Laboratories. p. 102.
counterfeited .
[10] David Bodanis (2005). Electric Universe. Crown Publish-
• BU508: n–p–n, 1500 V power BJT. Designed for ers, New York. ISBN 0-7394-5670-9.
television horizontal deflection, its high voltage ca-
[11] “transistor”. American Heritage Dictionary (3rd ed.).
pability also makes it suitable for use in ignition sys- Boston: Houghton Mifflin. 1992.
tems.
[12] “The Nobel Prize in Physics 1956”.
• MJ11012/MJ11015: 30 A, 120 V, 200 W, high
power Darlington complementary pair BJTs. Used [13] “1948 - The European Transistor Invention”. Computer
in audio amplifiers, control, and power switching. History Museum.

• 2N5457/2N5460: JFET (depletion mode), general [14] Proceeding of the IRE, Dec 1953, Author: W.E. Bradley
purpose, low power, complementary pair. - Philco Corp.,Research Division, Volume 41 issue 12,
pages 1702-1706
• BSP296/BSP171: IGFET (enhancement mode),
medium power, near complementary pair. Used for [15] Wall Street Journal, Dec 04 1953, page 4, Article “Philco
Claims Its Transistor Outperforms Others Now In Use”
logic level conversion and driving power transistors
in amplifiers. [16] Electronics magazine, January 1954, Article “Electro-
plated Transistors Announced”
• IRF3710/IRF5210: IGFET (enhancement mode),
40 A, 100 V, 200 W, near complementary pair. [17] Wall Street Journal, “Chrysler Promises Car Radio With
For high-power amplifiers and power switches, es- Transistors Instead of Tubes in '56”, April 28th 1955,
pecially in automobiles. page 1
1.1. TRANSISTOR
[18] Los Angeles Times, May 08, 1955, page A20, Article:
“Chrysler Announces New Transistor Radio”
[19] Philco TechRep Division Bulletin, May–June 1955, Vol-
ume 5 Number 3, page 28
[20] Article” Some Recollections of the Philco Transac S-
2000”, Author: Saul Rosen - Purdue University Computer
Science Dept., June 1991, page 2
[21] IEEE Spectrum, The Lost History of the Transistor, Au-
thor: Michael Riordan, May 2004, pp 48-49
[22] J. Chelikowski, “Introduction: Silicon in all its Forms”,
Silicon: evolution and future of a technology (Editors: P.
Siffert, E. F. Krimmel), p.1, Springer, 2004 ISBN 3-540-
40546-1.
[23] Grant McFarland, Microprocessor design: a practical
guide from design planning to manufacturing, p.10,
McGraw-Hill Professional, 2006 ISBN 0-07-145951-0.
[24] W. Heywang, K. H. Zaininger, “Silicon: The Semi-
conductor Material”, Silicon: evolution and future of a
technology (Editors: P. Siffert, E. F. Krimmel), p.36,
Springer, 2004 ISBN 3-540-40546-1.
[25] Robert W. Price (2004). Roadmap to Entrepreneurial Suc-
cess. AMACOM Div American Mgmt Assn. p. 42.
ISBN 978-0-8144-7190-6.
[26] extquotedblMilestones:Invention of the First Transistor at
Bell Telephone Laboratories, Inc., 1947”. IEEE Global
History Network. IEEE. Retrieved 3 August 2011.
[27] FETs/MOSFETs: Smaller apps push up surface-mount
supply
[28] extquotedblATI and Nvidia face off.” Oct 7, 2009. Re-
trieved on Feb 2, 2011.
[29] Jim Turley. “The Two Percent Solution” 2002.
[30] Kaplan, Daniel (2003). Hands-On Electronics. New York:
Cambridge University Press. pp. 47–54, 60–61. ISBN
978-0-511-07668-8.
[31] apart from a small value due to leakage currents
[32] John Keane and Chris H. Kim, “Transistor Aging,” IEEE
Spectrum (web feature), April 25, 2011.
[33] van der Veen, M. (2005). “Universal system and output
transformer for valve amplifiers”. 118th AES Convention,
Barcelona, Spain.
[34] “Transistor Example”. 071003 bcae1.com
13
[35] Streetman, Ben (1992). Solid State Electronic Devices.
Englewood Cliffs, NJ: Prentice-Hall. pp. 301–305. ISBN
0-13-822023-9.
[36] Horowitz, Paul; Winfield Hill (1989). The Art of Elec-
tronics (2nd ed.). Cambridge University Press. p. 115.
ISBN 0-521-37095-7.
[37] W. M. C. Sansen (2006). Analog design essentials. New
York ; Berlin: Springer. p. §0152, p. 28. ISBN 0-387-
25746-2.
[38] “IGBT Module 5SNA 2400E170100” (PDF). Retrieved
2012-06-30.
[39] Zhong Yuan Chang, Willy M. C. Sansen, Low-Noise
Wide-Band Amplifiers in Bipolar and CMOS Technologies,
page 31, Springer, 1991 ISBN 0792390962.
[40] “Single Electron Transistors”. Snow.stanford.edu. Re-
trieved 2012-06-30.
[41] Sanders, Robert (2005-06-28). “Nanofluidic transistor,
the basis of future chemical processors”. Berkeley.edu.
Retrieved 2012-06-30.
[42] The return of the vacuum tube?
[43] “Clive TEC Transistors Japanese Industrial Standards”.
Clivetec.0catch.com. Retrieved 2012-06-30.
[44] “Datasheet for BC549, with A,B and C gain groupings”
(PDF). Retrieved 2012-06-30.
[45] “Datasheet for BUK854-800A (800volt IGBT) extquot-
edbl (PDF). Retrieved 2012-06-30.
[46] “Richard Freeman’s HP Part numbers Crossreference”.
Hpmuseum.org. Retrieved 2012-06-30.
[47] Transistor–Diode Cross Reference – H.P. Part Numbers
to JEDEC (pdf)
[48] “CV Device Cross-reference by Andy Lake”. Qsl.net.
Retrieved 2012-06-30.
[49] A.S. Sedra and K.C. Smith (2004). Microelectronic cir-
cuits (Fifth ed.). New York: Oxford University Press. pp.
397 and Figure 5.17. ISBN 0-19-514251-9.
[50] Jhonathan P. Rojas, Galo A. Torres Sevilla, and Muham-
mad M. Hussain. “Can We Build a Truly High Perfor-
mance Computer Which is Flexible and Transparent? ex-
tquotedbl.
[51] Kan Zhang, Jung-Hun Seo1, Weidong Zhou and Zhen-
qiang Ma. “Fast flexible electronics using transferrable
silicon nanomembranes”. 2012.
[52] Lisa Zyga. “Carbon nanotube transistors could lead to in-
expensive, flexible electronics”. 2011.
14 CHAPTER 1. THE TRANSISTOR

1.1.11 Further reading • BBC: Building the digital age photo history of tran-
sistors
• Amos S W & James M R (1999). Principles of
Transistor Circuits. Butterworth-Heinemann. ISBN • The Bell Systems Memorial on Transistors
0-7506-4427-3.
• IEEE Global History Network, The Transistor and
• Bacon, W. Stevenson (1968). “The Transistor’s Portable Electronics. All about the history of tran-
20th Anniversary: How Germanium And A Bit of sistors and integrated circuits.
Wire Changed The World”. Bonnier Corp.: Popu-
lar Science, retrieved from Google Books 2009-03- • Transistorized. Historical and technical information
22 (Bonnier Corporation) 192 (6): 80–84. ISSN from the Public Broadcasting Service
0161-7370.
• This Month in Physics History: November 17 to De-
• Horowitz, Paul & Hill, Winfield (1989). The Art of cember 23, 1947: Invention of the First Transistor.
Electronics. Cambridge University Press. ISBN 0- From the American Physical Society
521-37095-7.
• 50 Years of the Transistor. From Science Friday, De-
• Riordan, Michael & Hoddeson, Lillian (1998). cember 12, 1997
Crystal Fire. W.W Norton & Company Limited.
ISBN 0-393-31851-6. The invention of the tran- • Charts showing many characteristics and giving di-
sistor & the birth of the information age rect access to most datasheets for 2N, 2SA, 2SB.
2SC, 2SD, 2SH-K, and other numbers.
• Warnes, Lionel (1998). Analogue and Digital Elec-
tronics. Macmillan Press Ltd. ISBN 0-333-65820- • Common transistor pinouts
5.
• Large table of transistor characteristics
• “Herbert F. Mataré, An Inventor of the Transistor
has his moment”. The New York Times. 24 Febru-
ary 2003. 1.2 Bipolar junction transistor
• Michael Riordan (2005). “How Europe Missed
the Transistor”. IEEE Spectrum 42 (11): 52–57. “Junction transistor” redirects here. For other uses, see
doi:10.1109/MSPEC.2005.1526906. Junction transistor (disambiguation).
BJT redirects here. For the Japanese language profi-
• C. D. Renmore (1980). Silicon Chips and You. ciency test, see Business Japanese Proficiency Test.
ISBN 0-8253-0022-3. Schematic symbols for
PNP- and NPN-type
• Wiley-IEEE Press. Complete Guide to Semiconduc- BJTs.
tor Devices, 2nd Edition.
A bipolar junction transistor (BJT or bipolar tran-
1.1.12 External links sistor) is a type of transistor that relies on the contact of
two types of semiconductor for its operation. BJTs can be
• The CK722 Museum. Website devoted to the “classic” used as amplifiers, switches, or in oscillators. BJTs can
hobbyist germanium transistor be found either as individual discrete components, or in
large numbers as parts of integrated circuits.
• Jerry Russell’s Transistor Cross Reference Database.
Bipolar transistors are so named because their opera-
• The DatasheetArchive. Searchable database of tran- tion involves both electrons and holes. These two kinds
sistor specifications and datasheets. of charge carriers are characteristic of the two kinds
of doped semiconductor material; electrons are majority
• The Transistor Educational content from Nobel- charge carriers in n-type semiconductors, whereas holes
prize.org are majority charge carriers in p-type semiconductors. In
1.2. BIPOLAR JUNCTION TRANSISTOR 15

contrast, unipolar transistors such as the field-effect tran- The electrons in the base are called minority carriers be-
sistors have only one kind of charge carrier. cause the base is doped p-type, which makes holes the
Charge flow in a BJT is due to diffusion of charge car- majority carrier in the base.
riers across a junction between two regions of different To minimize the percentage of carriers that recombine
charge concentrations. The regions of a BJT are called before reaching the collector–base junction, the transis-
emitter, collector, and base.[note 1] A discrete transistor has tor’s base region must be thin enough that carriers can
three leads for connection to these regions. Typically, diffuse across it in much less time than the semiconduc-
the emitter region is heavily doped compared to the other tor’s minority carrier lifetime. In particular, the thickness
two layers, whereas the majority charge carrier concen- of the base must be much less than the diffusion length
trations in base and collector layers are about the same. of the electrons. The collector–base junction is reverse-
By design, most of the BJT collector current is due to the biased, and so little electron injection occurs from the
flow of charges injected from a high-concentration emit- collector to the base, but electrons that diffuse through
ter into the base where there are minority carriers that the base towards the collector are swept into the col-
diffuse toward the collector, and so BJTs are classified as lector by the electric field in the depletion region of the
minority-carrier devices. collector–base junction. The thin shared base and asym-
metric collector–emitter doping is what differentiates a
bipolar transistor from two separate and oppositely bi-
1.2.1 Introduction ased diodes connected in series.

n++ p n+

E iEn electrons iC
C
Voltage, current, and charge control
iE
iEp
iE holes iC
recombination
iB1 iB2 The collector–emitter current can be viewed as being
controlled by the base–emitter current (current control),
B
vBE vCB
or by the base–emitter voltage (voltage control). These
iB
views are related by the current–voltage relation of the
base–emitter junction, which is just the usual exponen-
tial current–voltage curve of a p-n junction (diode).[1]
NPN BJT with forward-biased E–B junction and reverse-biased
B–C junction
The physical explanation for collector current is the
amount of minority carriers in the base region.[1][2][3]
BJTs come in two types, or polarities, known as PNP Due to low level injection (in which there are much
and NPN based on the doping types of the three main fewer excess carriers than normal majority carriers) the
ambipolar transport rates (in which the excess majority
terminal regions. An NPN transistor comprises two
semiconductor junctions that share a thin p-doped anode and minority carriers flow at the same rate) is in effect
determined by the excess minority carriers.
region, and a PNP transistor comprises two semiconduc-
tor junctions that share a thin n-doped cathode region. Detailed transistor models of transistor action, such as
In typical operation, the base–emitter junction is forward the Gummel–Poon model, account for the distribution
biased, which means that the p-doped side of the junc- of this charge[4]explicitly to explain transistor behaviour
tion is at a more positive potential than the n-doped side, more exactly. The charge-control view easily handles
and the base–collector junction is reverse biased. In an phototransistors, where minority carriers in the base re-
NPN transistor, when positive bias is applied to the base– gion are created by the absorption of photons, and han-
emitter junction, the equilibrium is disturbed between dles the dynamics of turn-off, or recovery time, which
the thermally generated carriers and the repelling electric depends on charge in the base region recombining. How-
field of the n-doped emitter depletion region. This allows ever, because base charge is not a signal that is visible at
thermally excited electrons to inject from the emitter into the terminals, the current- and voltage-control views are
the base region. These electrons diffuse through the base generally used in circuit design and analysis.
from the region of high concentration near the emitter to- In analog circuit design, the current-control view is some-
wards the region of low concentration near the collector. times used because it is approximately linear. That is, the
16 CHAPTER 1. THE TRANSISTOR

collector current is approximately βF times the base cur- than unity due to recombination of charge carriers as they
rent. Some basic circuits can be designed by assuming cross the base region. Alpha and beta are more precisely
that the emitter–base voltage is approximately constant, related by the following identities (NPN transistor):
and that collector current is beta times the base current.
However, to accurately and reliably design production
BJT circuits, the voltage-control (for example, Ebers– IC
αF =
Moll) model is required.[1] The voltage-control model re- IE
quires an exponential function to be taken into account, IC
but when it is linearized such that the transistor can be βF =
IB
modelled as a transconductance, as in the Ebers–Moll αF βF
model, design for circuits such as differential amplifiers βF = ⇐⇒ αF =
1 − αF βF + 1
again becomes a mostly linear problem, so the voltage-
control view is often preferred. For translinear circuits, in
which the exponential I–V curve is key to the operation, 1.2.2 Structure
the transistors are usually modelled as voltage controlled
with transconductance proportional to collector current.
In general, transistor level circuit design is performed us- E B C
ing SPICE or a comparable analog circuit simulator, so
model complexity is usually not of much concern to the
designer. n
p
Turn-on, turn-off, and storage delay

The Bipolar transistor exhibits a few delay characteris-

n
tics when turning on and off. Most transistors, and espe-
cially power transistors, exhibit long base-storage times Simplified cross section of a planar NPN bipolar junction tran-
that limit maximum frequency of operation in switching sistor
applications. One method for reducing this storage time
is by using a Baker clamp. A BJT consists of three differently doped semiconductor
regions, the emitter region, the base region and the collec-
tor region. These regions are, respectively, p type, n type
Transistor parameters: alpha (α) and beta (β) and p type in a PNP transistor, and n type, p type and n
type in an NPN transistor. Each semiconductor region
The proportion of electrons able to cross the base and is connected to a terminal, appropriately labeled: emitter
reach the collector is a measure of the BJT efficiency. (E), base (B) and collector (C).
The heavy doping of the emitter region and light dop-
ing of the base region causes many more electrons to be The base is physically located between the emitter and the
injected from the emitter into the base than holes to be collector and is made from lightly doped, high resistiv-
injected from the base into the emitter. The common- ity material. The collector surrounds the emitter region,
emitter current gain is represented by βF or hFE; it is ap- making it almost impossible for the electrons injected
proximately the ratio of the DC collector current to the into the base region to escape without being collected,
DC base current in forward-active region. It is typically thus making the resulting value of α very close to unity,
greater than 100 for small-signal transistors but can be and so, giving the transistor a large β. A cross section
smaller in transistors designed for high-power applica- view of a BJT indicates that the collector–base junction
tions. Another important parameter is the common-base has a much larger area than the emitter–base junction.
current gain, αF. The common-base current gain is ap- The bipolar junction transistor, unlike other transistors, is
proximately the gain of current from emitter to collec- usually not a symmetrical device. This means that inter-
tor in the forward-active region. This ratio usually has changing the collector and the emitter makes the transis-
a value close to unity; between 0.98 and 0.998. It is less tor leave the forward active mode and start to operate in
1.2. BIPOLAR JUNCTION TRANSISTOR 17

reverse mode. Because the transistor’s internal structure for very high speed applications (see HBT, below).
is usually optimized for forward-mode operation, inter-
changing the collector and the emitter makes the values
of α and β in reverse operation much smaller than those
in forward operation; often the α of the reverse mode
NPN
is lower than 0.5. The lack of symmetry is primarily
due to the doping ratios of the emitter and the collec-
tor. The emitter is heavily doped, while the collector is
lightly doped, allowing a large reverse bias voltage to be

C
applied before the collector–base junction breaks down.
The collector–base junction is reverse biased in normal
operation. The reason the emitter is heavily doped is to
increase the emitter injection efficiency: the ratio of car-

B
riers injected by the emitter to those injected by the base.
For high current gain, most of the carriers injected into
the emitter–base junction must come from the emitter.

E
The symbol of an NPN BJT. The symbol is extquotedblnot
pointing in.”

NPN is one of the two types of bipolar transistors, con-

Die of a KSY34 high-frequency NPN transistor, base and emitter sisting of a layer of P-doped semiconductor (the “base”)
connected via bonded wires
between two N-doped layers. A small current entering
the base is amplified to produce a large collector and
The low-performance “lateral” bipolar transistors some- emitter current. That is, when there is a positive potential
times used in CMOS processes are sometimes designed
difference measured from the emitter of an NPN transis-
symmetrically, that is, with no difference between for- tor to its base (i.e., when the base is high relative to the
ward and backward operation.
emitter) as well as positive potential difference measured
Small changes in the voltage applied across the base– from the base to the collector, the transistor becomes ac-
emitter terminals causes the current that flows between tive. In this “on” state, current flows between the collec-
the emitter and the collector to change significantly. This tor and emitter of the transistor. Most of the current is
effect can be used to amplify the input voltage or cur- carried by electrons moving from emitter to collector as
rent. BJTs can be thought of as voltage-controlled current minority carriers in the P-type base region. To allow for
sources, but are more simply characterized as current- greater current and faster operation, most bipolar tran-
controlled current sources, or current amplifiers, due to sistors used today are NPN because electron mobility is
the low impedance at the base. higher than hole mobility.
Early transistors were made from germanium but most A mnemonic device for the NPN transistor symbol is ex-
modern BJTs are made from silicon. A significant minor- tquotedblnot pointing in extquotedbl, based on the arrows
ity are also now made from gallium arsenide, especially in the symbol and the letters in the name.[5]
18 CHAPTER 1. THE TRANSISTOR

Δφn ΔφG

E n
p

B Δφp
n

Bands in graded heterojunction NPN bipolar transistor. Barri-

C
ers indicated for electrons to move from emitter to base, and for
holes to be injected backward from base to emitter; Also, grad-
ing of bandgap in base assists electron transport in base region;
Light colors indicate depleted regions

The symbol of a PNP BJT. The symbol extquotedblpoints in rier for holes to inject backward from the base into the
proudly.”
emitter, denoted in the figure as Δφ , to be made large,
while the barrier for electrons to inject into the base Δφ
PNP is made low. This barrier arrangement helps reduce mi-
nority carrier injection from the base when the emitter-
base junction is under forward bias, and thus reduces base
The other type of BJT is the PNP, consisting of a layer of
current and increases emitter injection efficiency.
N-doped semiconductor between two layers of P-doped
material. A small current leaving the base is amplified The improved injection of carriers into the base allows
in the collector output. That is, a PNP transistor is “on” the base to have a higher doping level, resulting in lower
when its base is pulled low relative to the emitter. resistance to access the base electrode. In the more tra-
ditional BJT, also referred to as homojunction BJT, the
The arrows in the NPN and PNP transistor symbols are
efficiency of carrier injection from the emitter to the base
on the emitter legs and point in the direction of the
is primarily determined by the doping ratio between the
conventional current flow when the device is in forward
emitter and base, which means the base must be lightly
active mode.
doped to obtain high injection efficiency, making its re-
A mnemonic device for the PNP transistor symbol is ex- sistance relatively high. In addition, higher doping in the
tquotedblpointing in (proudly/permanently) extquotedbl, base can improve figures of merit like the Early voltage
based on the arrows in the symbol and the letters in the by lessening base narrowing.
name.[6]
The grading of composition in the base, for example,
by progressively increasing the amount of germanium in
Heterojunction bipolar transistor a SiGe transistor, causes a gradient in bandgap in the
neutral base, denoted in the figure by ΔφG, providing a
The heterojunction bipolar transistor (HBT) is an im- “built-in” field that assists electron transport across the
provement of the BJT that can handle signals of very high base. That drift component of transport aids the normal
frequencies up to several hundred GHz. It is common in diffusive transport, increasing the frequency response of
modern ultrafast circuits, mostly RF systems.[7][8] Het- the transistor by shortening the transit time across the
erojunction transistors have different semiconductors for base.
the elements of the transistor. Usually the emitter is com- Two commonly used HBTs are silicon–germanium and
posed of a larger bandgap material than the base. The fig- aluminum gallium arsenide, though a wide variety of
ure shows that this difference in bandgap allows the bar- semiconductors may be used for the HBT structure. HBT
1.2. BIPOLAR JUNCTION TRANSISTOR 19

structures are usually grown by epitaxy techniques like charged carriers flowing from emitter to collector).
MOCVD and MBE. This mode corresponds to a logical “on”, or a closed
switch.

1.2.3 Regions of operation • Cutoff: In cutoff, biasing conditions opposite

of saturation (both junctions reverse biased) are
present. There is very little current, which corre-
sponds to a logical “off”, or an open switch.
• Avalanche breakdown region

The modes of operation can be described in terms of the

applied voltages (this description applies to NPN transis-
tors; polarities are reversed for PNP transistors):

• Forward-active: base higher than emitter, collector

higher than base (in this mode the collector current
is proportional to base current by βF ).
The relationship between IC , UCE and IB .
• Saturation: base higher than emitter, but collector is
Bipolar transistors have five distinct regions of operation, not higher than base.
defined by BJT junction biases.
• Cut-Off: base lower than emitter, but collector is
higher than base. It means the transistor is not let-
• Forward-active (or simply, active): The base– ting conventional current go through from collector
emitter junction is forward biased and the base– to emitter.
collector junction is reverse biased. Most bipo-
lar transistors are designed to afford the greatest • Reverse-active: base lower than emitter, collector
common-emitter current gain, βF, in forward-active lower than base: reverse conventional current goes
mode. If this is the case, the collector–emitter cur- through transistor.
rent is approximately proportional to the base cur-
rent, but many times larger, for small base current In terms of junction biasing: ('reverse biased base–
variations. collector junction' means Vbc < 0 for NPN, opposite for
• Reverse-active (or inverse-active or inverted): By PNP)
reversing the biasing conditions of the forward- Although these regions are well defined for sufficiently
active region, a bipolar transistor goes into reverse- large applied voltage, they overlap somewhat for small
active mode. In this mode, the emitter and collec- (less than a few hundred millivolts) biases. For example,
tor regions switch roles. Because most BJTs are de- in the typical grounded-emitter configuration of an NPN
signed to maximize current gain in forward-active BJT used as a pulldown switch in digital logic, the “off”
mode, the βF in inverted mode is several times state never involves a reverse-biased junction because the
smaller (2–3 times for the ordinary germanium tran- base voltage never goes below ground; nevertheless the
sistor). This transistor mode is seldom used, usu- forward bias is close enough to zero that essentially no
ally being considered only for failsafe conditions and current flows, so this end of the forward active region can
some types of bipolar logic. The reverse bias break- be regarded as the cutoff region.
down voltage to the base may be an order of magni-
tude lower in this region.
Active-mode NPN transistors in circuits
• Saturation: With both junctions forward-biased, a
BJT is in saturation mode and facilitates high current The diagram shows a schematic representation of an
conduction from the emitter to the collector (or the NPN transistor connected to two voltage sources. To
other direction in the case of NPN, with negatively make the transistor conduct appreciable current (on the
20 CHAPTER 1. THE TRANSISTOR

act value (for example see op-amp). The value of this

gain for DC signals is referred to as hFE , and the value
of this gain for small signals is referred to as hfe . That is,
IC when a small change in the currents occurs, and sufficient
time has passed for the new condition to reach a steady
state hfe is the ratio of the change in collector current to
n C the change in base current. The symbol β is used for both
hFE and hfe .[9]
VCE IB The emitter current is related to VBE exponentially. At
p B
room temperature, an increase in VBE by approximately
60 mV increases the emitter current by a factor of 10.
Because the base current is approximately proportional to
VBE the collector and emitter currents, they vary in the same
n E way.

Active-mode PNP transistors in circuits

IE

Structure and use of NPN transistor. Arrow according to IE

schematic.
VEB
order of 1 mA) from C to E, VBE must be above a min-
p E
imum value sometimes referred to as the cut-in voltage.
The cut-in voltage is usually about 650 mV for silicon VCE
BJTs at room temperature but can be different depending n B
on the type of transistor and its biasing. This applied volt-
age causes the lower P-N junction to 'turn on', allowing a
IB
flow of electrons from the emitter into the base. In active
mode, the electric field existing between base and col- p C
lector (caused by VCE) will cause the majority of these
electrons to cross the upper P-N junction into the collec-
tor to form the collector current IC. The remainder of the
electrons recombine with holes, the majority carriers in IC
the base, making a current through the base connection to
form the base current, IB. As shown in the diagram, the
emitter current, IE, is the total transistor current, which
is the sum of the other terminal currents, (i.e., IE = IB + Structure and use of PNP transistor.
IC).
In the diagram, the arrows representing current point in The diagram shows a schematic representation of a PNP
the direction of conventional current – the flow of elec- transistor connected to two voltage sources. To make the
trons is in the opposite direction of the arrows because transistor conduct appreciable current (on the order of 1
electrons carry negative electric charge. In active mode, mA) from E to C, VEB must be above a minimum value
the ratio of the collector current to the base current is sometimes referred to as the cut-in voltage. The cut-in
called the DC current gain. This gain is usually 100 or voltage is usually about 650 mV for silicon BJTs at room
more, but robust circuit designs do not depend on the ex- temperature but can be different depending on the type of
1.2. BIPOLAR JUNCTION TRANSISTOR
transistor and its biasing. This applied voltage causes the
upper P-N junction to 'turn-on' allowing a flow of holes
from the emitter into the base. In active mode, the elec-
tric field existing between the emitter and the collector
(caused by VCE ) causes the majority of these holes to
cross the lower p-n junction into the collector to form
the collector current IC . The remainder of the holes re-
combine with electrons, the majority carriers in the base,
making a current through the base connection to form the
base current, IB . As shown in the diagram, the emitter
current, IE , is the total transistor current, which is the
sum of the other terminal currents (i.e., IE = IB + IC).
In the diagram, the arrows representing current point in
the direction of conventional current – the flow of holes
is in the same direction of the arrows because holes carry Bipolar transistors
positive electric charge. In active mode, the ratio of the
collector current to the base current is called the DC cur- • Point-contact transistor – first transistor ever con-
rent gain. This gain is usually 100 or more, but robust
circuit designs do not depend on the exact value. The
value of this gain for DC signals is referred to as hFE ,
and the value of this gain for AC signals is referred to
as hfe . However, when there is no particular frequency
range of interest, the symbol β is used.
• Junction transistors
It should also be noted that the emitter current is related
to VEB exponentially. At room temperature, an increase
in VEB by approximately 60 mV increases the emitter cur-
rent by a factor of 10. Because the base current is approx-
imately proportional to the collector and emitter currents,
they vary in the same way.
1.2.4 History
The bipolar point-contact transistor was invented in De-
cember 1947 at the Bell Telephone Laboratories by
John Bardeen and Walter Brattain under the direction of
William Shockley. The junction version known as the
bipolar junction transistor, invented by Shockley in 1948,
enjoyed three decades as the device of choice in the de-
sign of discrete and integrated circuits. Nowadays, the
use of the BJT has declined in favor of CMOS technology
in the design of digital integrated circuits. The incidental
low performance BJTs inherent in CMOS ICs, however,
are often utilized as bandgap voltage reference, silicon
bandgap temperature sensor and to handle electrostatic
discharge.
21
Germanium transistors
The germanium transistor was more common in the
1950s and 1960s, and while it exhibits a lower “cut off”
voltage, typically around 0.2 V, making it more suitable
for some applications, it also has a greater tendency to
exhibit thermal runaway.
Early manufacturing techniques
Various methods of manufacturing bipolar transistors
were developed.[10]
structed (December 1947), a bipolar transistor, lim-
ited commercial use due to high cost and noise.
--- Tetrode point-contact transistor – Point-
contact transistor having two emitters. It be-
came obsolete in the middle 1950s.
--- Grown-junction transistor – first bipolar junc-
tion transistor made.[11] Invented by William
Shockley at Bell Labs. Invented on June 23,
1948.[12] Patent filed on June 26, 1948.
--- Alloy-junction transistor – emitter and collec-
tor alloy beads fused to base. Developed at
General Electric and RCA[13] in 1951.
∗ Micro-alloy transistor (MAT) – high
speed type of alloy junction transistor.
Developed at Philco.[14]
∗ Micro-alloy diffused transistor (MADT)
– high speed type of alloy junction tran-
sistor, speedier than MAT, a diffused-
base transistor. Developed at Philco.
∗ Post-alloy diffused transistor (PADT) –
high speed type of alloy junction transis-
tor, speedier than MAT, a diffused-base
transistor. Developed at Philips.
--- Tetrode transistor – high speed variant of
grown-junction transistor[15] or alloy junction
transistor[16] with two connections to base.
--- Surface-barrier transistor – high speed metal
barrier junction transistor. Developed at
Philco[17] in 1953.[18]
22 CHAPTER 1. THE TRANSISTOR

--- Drift-field transistor – high speed bipolar Emitter diffusion

Base drift
Collector
junction transistor. Invented by Herbert Kroe-
mer[19][20] at the Central Bureau of Telecom-
munications Technology of the German Postal
Service, in 1953. Ec
--- Spacistor – circa 1957. Ef

--- Diffusion transistor – modern type bipolar diffusion

junction transistor. Prototypes[21] developed n-type

depletion depletion
n-type
Ev
region p-type region
at Bell Labs in 1954.
∗ Diffused-base transistor – first implemen- Band diagram for NPN transistor in active mode, showing injec-
tation of diffusion transistor. tion of electrons from emitter to base, and their overshoot into
∗ Mesa transistor – Developed at Texas In- the collector.
struments in 1957.
∗ Planar transistor – the bipolar junc- Both types of BJT function by letting a small current input
tion transistor that made mass-produced to the base control an amplified output from the collec-
monolithic integrated circuits possible. tor. The result is that the transistor makes a good switch
Developed by Dr. Jean Hoerni[22] at that is controlled by its base input. The BJT also makes a
Fairchild in 1959. good amplifier, since it can multiply a weak input signal
--- Epitaxial transistor – a bipolar junction tran- to about 100 times its original strength. Networks of tran-
sistor made using vapor phase deposition. See sistors are used to make powerful amplifiers with many
epitaxy. Allows very precise control of doping different applications. In the discussion below, focus is
levels and gradients. on the NPN bipolar transistor. In the NPN transistor in
what is called active mode, the base–emitter voltage VBE
and collector–base voltage VCB are positive, forward bi-
1.2.5 Theory and modeling asing the emitter–base junction and reverse-biasing the
collector–base junction. In the active mode of operation,
electrons are injected from the forward biased n-type
Emitter Base Collector emitter region into the p-type base where they diffuse as
n-type
depletion
region p-type
depletion
region n-type minority carriers to the reverse-biased n-type collector
and are swept away by the electric field in the reverse-
biased collector–base junction. For a figure describing
Ec forward and reverse bias, see semiconductor diodes.
Ef

Large-signal models
Ev
In 1954 Jewell James Ebers and John L. Moll introduced
their mathematical model of transistor currents:[23]
Band diagram for NPN transistor at equilibrium.
Ebers–Moll model The DC emitter and collector cur-
Transistors can be thought of as two diodes (P–N junc- rents in active mode are well modeled by an approxima-
tions) sharing a common region that minority carriers can tion to the Ebers–Moll model:
move through. A PNP BJT will function like two diodes
that share an N-type cathode region, and the NPN like
two diodes sharing a P-type anode region. Connecting ( VBE )
two diodes with wires will not make a transistor, since IE = IES e VT − 1
minority carriers will not be able to get from one P–N
junction to the other through the wire. IC = α F IE
1.2. BIPOLAR JUNCTION TRANSISTOR 23

αR ICD αF IED αR ICD αF IED

IE IC IE IC
E C E C
IED ICD IED ICD
IB IB
B B
Ebers–Moll Model for an NPN transistor.[24] * IB, IC, IE: base, Approximated Ebers–Moll Model for an NPN transistor in the
collector and emitter currents * ICD, IED: collector and emit- forward active mode. The collector diode is reverse-biased so
ter diode currents * αF, αR: forward and reverse common-base ICD is virtually zero. Most of the emitter diode current (αF is
current gains nearly 1) is drawn from the collector, providing the amplification
of the base current.

αR ICD αF IED
• αF is the common base forward short circuit current
gain (0.98 to 0.998)

• IES is the reverse saturation current of the base–

IE IC emitter diode (on the order of 10−15 to 10−12 am-
peres)
E C • VBE is the base–emitter voltage
IED ICD
IB • Dn is the diffusion constant for electrons in the p-
type base
B
• W is the base width
Ebers–Moll Model for a PNP transistor.

The α and forward β parameters are as described previ-

IB = (1 − αF ) IE ously. A reverse β is sometimes included in the model.

The base internal current is mainly by diffusion (see The unapproximated Ebers–Moll equations used to de-
Fick’s law) and scribe the three currents in any operating region are given
below. These equations are based on the transport model
for a bipolar junction transistor.[25]
qDn nbo VVEB
Jn (base) = e T
W ( VBE VBC ) IS ( VVBC )
iC = IS e VT − e VT − e T −1
where βR

IS ( VVBE ) IS ( VVBC )
• VT is the thermal voltage kT /q (approximately 26 iB = e T −1 + e T −1
mV at 300 K ≈ room temperature). βF βR
( VBE VBC ) IS ( VVBE )
• IE is the emitter current iE = IS e VT − e VT + e T −1
βF
• IC is the collector current where
24 CHAPTER 1. THE TRANSISTOR

• iC is the collector current Base-width modulation Main article: Early Effect

• iB is the base current

As the collector–base voltage ( VCB = VCE −VBE ) varies,
• iE is the emitter current the collector–base depletion region varies in size. An in-
crease in the collector–base voltage, for example, causes
• βF is the forward common emitter current gain (20 a greater reverse bias across the collector–base junction,
to 500) increasing the collector–base depletion region width, and
decreasing the width of the base. This variation in base
• βR is the reverse common emitter current gain (0 to width often is called the extquotedblEarly effect extquot-
20) edbl after its discoverer James M. Early.

• IS is the reverse saturation current (on the order of Narrowing of the base width has two consequences:
10−15 to 10−12 amperes)
• There is a lesser chance for recombination within
• VT is the thermal voltage (approximately 26 mV at the “smaller” base region.
300 K ≈ room temperature).
• The charge gradient is increased across the base,
• VBE is the base–emitter voltage and consequently, the current of minority carriers
injected across the emitter junction increases.
• VBC is the base–collector voltage

Both factors increase the collector or “output” current of

the transistor in response to an increase in the collector–
base voltage.
VCE1 In the forward-active region, the Early effect modifies the
collector current ( iC ) and the forward common emitter

B
current gain ( βF ) as given by:
VBE
( )
n p n VCE
vBE
iC = IS e VT 1+
VA
E C ( )
Weff βF = βF 0 1+
VCB
VA
ΔWeff
VA
E C ro =
IC
n p n where:

VBE B depletion • VCE is the collector–emitter voltage

regions • VA is the Early voltage (15 V to 150 V)

• βF 0 is forward common-emitter current gain when

VCE2 VCB = 0 V

Top: NPN base width for low collector-base reverse bias; Bottom: • ro is the output impedance
narrower NPN base width for large collector-base reverse bias.
Hashed regions are depleted regions. • IC is the collector current
1.2. BIPOLAR JUNCTION TRANSISTOR 25

Punchthrough When the base–collector volt- 1 ii io 2

age reaches a certain (device specific) value, the
base–collector depletion region boundary meets the hix
base–emitter depletion region boundary. When in this
state the transistor effectively has no base. The device
thus loses all gain when in this state.
Vi hox Vo
hrxVo hfxii
Gummel–Poon charge-control model The Gummel–
Poon model[26] is a detailed charge-controlled model of 3
BJT dynamics, which has been adopted and elaborated by
others to explain transistor dynamics in greater detail than
Generalized h-parameter model of an NPN BJT.
the terminal-based models typically do . This model also Replace x with e, b or c for CE, CB and CC topologies respec-
includes the dependence of transistor β -values upon the tively.
direct current levels in the transistor, which are assumed
current-independent in the Ebers–Moll model.[27]
and output voltages. This two-port network is particu-
larly suited to BJTs as it lends itself easily to the analysis
Small-signal models of circuit behaviour, and may be used to develop further
accurate models. As shown, the term “x” in the model
rb'c represents a different BJT lead depending on the topol-
ogy used. For common-emitter mode the various sym-
rbb i C bols take on the specific values as:
B i b B' c

Cc • x = 'e' because it is a common-emitter topology

rb'e • Terminal 1 = Base
gce
Ce gmvb'e • Terminal 2 = Collector
• Terminal 3 = Emitter
E
• iᵢ = Base current (i )
Hybrid-pi model
• iₒ = Collector current (i )
hybrid-pi model Main article: hybrid-pi model
• Vᵢ = Base-to-emitter voltage (VBE)

The hybrid-pi model is a popular circuit model used for • Vₒ = Collector-to-emitter voltage (VCE)
analyzing the small signal behavior of bipolar junction
and field effect transistors. Sometimes it is also called and the h-parameters are given by:
Giacoletto model because it was introduced by L.J. Gia-
coletto in 1969. The model can be quite accurate for low- • hᵢₓ = hᵢₑ – The input impedance of the transistor
frequency circuits and can easily be adapted for higher (corresponding to the base resistance r ᵢ).
frequency circuits with the addition of appropriate inter-
electrode capacitances and other parasitic elements. • hᵣₓ = hᵣₑ – Represents the dependence of the tran-
sistor’s IB–VBE curve on the value of VCE. It is
usually very small and is often neglected (assumed
h-parameter model Another model commonly used to be zero).
to analyze BJT circuits is the extquotedblh-parameter ex-
tquotedbl model, closely related to the hybrid-pi model • h ₓ = h ₑ – The current-gain of the transistor. This
and the y-parameter two-port, but using input current and parameter is often specified as hFE or the DC
output voltage as independent variables, rather than input current-gain (βDC) in datasheets.
26 CHAPTER 1. THE TRANSISTOR

• hₒₓ = 1/hₒₑ – The output impedance of transistor. • Common emitter

The parameter hₒₑ usually corresponds to the output
admittance of the bipolar transistor and has to be • Common base
inverted to convert it to an impedance.
• Common collector
As shown, the h-parameters have lower-case subscripts
and hence signify AC conditions or analyses. For DC Temperature sensors
conditions they are specified in upper-case. For the CE
topology, an approximate h-parameter model is com-
Main article: Silicon bandgap temperature sensor
monly used which further simplifies the circuit analysis.
For this the hₒₑ and hᵣₑ parameters are neglected (that is,
they are set to infinity and zero, respectively). It should Because of the known temperature and current depen-
also be noted that the h-parameter model as shown is dence of the forward-biased base–emitter junction volt-
suited to low-frequency, small-signal analysis. For high- age, the BJT can be used to measure temperature by sub-
frequency analyses the inter-electrode capacitances that tracting two voltages at two different bias currents in a
are important at high frequencies must be added. known ratio .

Etymology of hFE The 'h' refers to its being an h- Logarithmic converters

parameter, a set of parameters named for their origin in a
hybrid equivalent circuit model. 'F' is from forward cur-
rent amplification also called the current gain. 'E' refers
to the transistor operating in a common emitter (CE) con-
figuration. Capital letters used in the subscript indicate
that hFE refers to a direct current circuit.
Because base–emitter voltage varies as the log of the
base–emitter and collector–emitter currents, a BJT can
also be used to compute logarithms and anti-logarithms.
A diode can also perform these nonlinear functions but
the transistor provides more circuit flexibility.

1.2.6 Applications 1.2.7 Vulnerabilities

The BJT remains a device that excels in some appli- Exposure of the transistor to ionizing radiation causes
cations, such as discrete circuit design, due to the very radiation damage. Radiation causes a buildup of 'de-
wide selection of BJT types available, and because of its fects’ in the base region that act as recombination cen-
high transconductance and output resistance compared to ters. The resulting reduction in minority carrier lifetime
MOSFETs. The BJT is also the choice for demanding causes gradual loss of gain of the transistor.
analog circuits, especially for very-high-frequency appli-
cations, such as radio-frequency circuits for wireless sys-Power BJTs are subject to a failure mode called
tems. Bipolar transistors can be combined with MOS- secondary breakdown, in which excessive current and
FETs in an integrated circuit by using a BiCMOS process normal imperfections in the silicon die cause portions of
of wafer fabrication to create circuits that take advantagethe silicon inside the device to become disproportionately
of the application strengths of both types of transistor. hotter than the others. The doped silicon has a negative
temperature coefficient, meaning that it conducts more
current at higher temperatures. Thus, the hottest part of
Amplifiers the die conducts the most current, causing its conductivity
to increase, which then causes it to become progressively
Main article: Electronic amplifier hotter again, until the device fails internally. The thermal
runaway process associated with secondary breakdown,
The transistor parameters α and β characterizes the once triggered, occurs almost instantly and may catas-
current gain of the BJT. It is this gain that allow BJTs trophically damage the transistor package.
to be used as the building blocks of electronic amplifiers. If the emitter-base junction is reverse biased into
The three main BJT amplifier topologies are avalanche or Zener mode and current flows for a short
1.2. BIPOLAR JUNCTION TRANSISTOR
period of time, the current gain of the BJT will be per- [10] Third case study – the solid state advent (PDF)
manently degraded.
1.2.8 See also
• Bipolar transistor biasing
• Gummel plot
• Technology CAD (TCAD)
1.2.9 References
[1] See point-contact transistor for the historical origin of
these names.
[1] Paul Horowitz and Winfield Hill (1989). The Art of Elec-
tronics (2nd ed.). Cambridge University Press. ISBN
978-0-521-37095-0.
[2] Juin Jei Liou and Jiann S. Yuan (1998). Semiconductor
Device Physics and Simulation. Springer. ISBN 0-306-
45724-5.
[3] General Electric (1962). Transistor Manual (6th ed.). p.
12. “If the principle of space charge neutrality is used in
the analysis of the transistor, it is evident that the collector
current is controlled by means of the positive charge (hole
concentration) in the base region. ... When a transistor
is used at higher frequencies, the fundamental limitation
is the time it takes the carriers to diffuse across the base
region...” (same in 4th and 5th editions)
[4] Paolo Antognetti and Giuseppe Massobrio (1993).
Semiconductor Device Modeling with Spice. McGraw–Hill
Professional. ISBN 0-07-134955-3.
[5] Alphonse J. Sistino (1996). Essentials of electronic cir-
cuitry. CRC Press. p. 64. ISBN 978-0-8247-9693-8.
[6] Alphonse J. Sistino (1996). Essentials of electronic cir-
cuitry. CRC Press. p. 102. ISBN 978-0-8247-9693-8.
[7] D.V. Morgan, Robin H. Williams (Editors) (1991).
Physics and Technology of Heterojunction Devices. Lon-
don: Institution of Electrical Engineers (Peter Peregrinus
Ltd.). ISBN 0-86341-204-1.
[8] Peter Ashburn (2003). SiGe Heterojunction Bipolar Tran-
sistors. New York: Wiley. Chapter 10. ISBN 0-470-
84838-3.
[9] Paul Horowitz and Winfield Hill (1989). The Art of Elec-
tronics (2nd ed.). Cambridge University Press. pp. 62–
66. ISBN 978-0-521-37095-0.
27
[11] Transistor Museum, Historic Transistor Photo Gallery,
Bell Labs Type M1752
[12] Morris, Peter Robin (1990). “4.2”. A History of the World
Semiconductor Industry. IEE History of Technology Se-
ries 12. London: Peter Peregrinus Ltd. p. 29. ISBN
0-86341-227-0.
[13] Transistor Museum, Historic Transistor Photo Gallery,
RCA TA153
[14] High Speed Switching Transistor Handbook (2nd ed.).
Motorola. 1963. p. 17.
[15] Transistor Museum, Historic Transistor Photo Gallery,
Western Electric 3N22
[16] The Tetrode Power Transistor PDF
[17] Transistor Museum, Historic Transistor Photo Gallery,
Philco A01
[18] Transistor Museum, Historic Transistor Photo Gallery,
Surface Barrier Transistor
[19] Herb’s Bipolar Transistors IEEE Transactions on Electron
Devices, vol. 48, no. 11, November 2001 PDF
[20] Influence of Mobility and Lifetime Variations on Drift-
Field Effects in Silicon-Junction Devices PDF
[21] Transistor Museum, Historic Transistor Photo Gallery,
Bell Labs Prototype Diffused Base Triode
[22] Transistor Museum, Historic Transistor Photo Gallery,
Fairchild 2N1613
[23] J.J. Ebers and J.L Moll (1954) “Large-signal behavior of
junction transistors”, Proceedings of the Institute of Radio
Engineers, 42 (12) : 1761–1772.
[24] Adel S. Sedra and Kenneth C. Smith (1987). Microelec-
tronic Circuits, second ed. p. 903. ISBN 0-03-007328-6.
[25] A.S. Sedra and K.C. Smith (2004). Microelectronic Cir-
cuits (5th ed.). New York: Oxford. Eqs. 4.103–4.110, p.
305. ISBN 0-19-514251-9.
[26] H. K. Gummel and R. C. Poon, “An integral charge con-
trol model of bipolar transistors,” Bell Syst. Tech. J., vol.
49, pp. 827–852, May–June 1970
[27] A.S. Sedra and K.C. Smith (2004). Microelectronic Cir-
cuits (5th ed.). New York: Oxford. p. 509. ISBN 0-19-
514251-9.
28 CHAPTER 1. THE TRANSISTOR

1.2.10 External links

• Simulation of a BJT in the Common Emitter Circuit

• Lessons In Electric Circuits – Bipolar Junction

Transistors (Note: this site shows current as a flow
of electrons, rather than the convention of showing
it as a flow of holes)

• EncycloBEAMia – Bipolar Junction Transistor

• Characteristic curves

• ENGI 242/ELEC 222: BJT Small Signal Models

• Transistor Museum, Historic Transistor Timeline

• ECE 327: Transistor Basics – Summarizes simple

Ebers–Moll model of a bipolar transistor and gives
several common BJT circuits.

• ECE 327: Procedures for Output Filtering Lab –

Section 4 (“Power Amplifier”) discusses design of
a BJT-Sziklai-pair-based class-AB current driver in High-power N-channel field-effect transistor
detail.

• BJT Operation description for undergraduate and The field-effect transistor was first patented by Julius
first year graduate students to describe the basic Edgar Lilienfeld in 1926 and by Oskar Heil in 1934, but
principles of operation of Bipolar Junction Transis- practical semiconducting devices (the JFET) were devel-
tor. oped only much later after the transistor effect was ob-
served and explained by the team of William Shockley at
Bell Labs in 1947, immediately after the 20-year patent
1.3 Field-effect transistor period eventually expired. The MOSFET, which largely
superseded the JFET and had a more profound effect on
electronic development, was invented by Dawon Kahng
“FET” redirects here. For other uses, see FET (disam-
and Martin Atalla in 1960.[1]
biguation).
The field-effect transistor (FET) is a transistor that
uses an electric field to control the shape and hence the 1.3.2 Basic information
conductivity of a channel of one type of charge carrier in
a semiconductor material. FETs are unipolar transistors
See also: Charge carrier § Majority and minority carriers
as they involve single-carrier-type operation. The concept
of the FET predates the bipolar junction transistor (BJT),
though it was not physically implemented until after BJTs FETs can be majority-charge-carrier devices, in which
due to the limitations of semiconductor materials and the the current is carried predominantly by majority carriers,
relative ease of manufacturing BJTs compared to FETs or minority-charge-carrier devices, in which the current
at the time. is mainly due to a flow of minority carriers.[2] The device
consists of an active channel through which charge carri-
ers, electrons or holes, flow from the source to the drain.
1.3.1 History Source and drain terminal conductors are connected to
the semiconductor through ohmic contacts. The conduc-
Main article: History of the transistor tivity of the channel is a function of the potential applied
across the gate and source terminals.
1.3. FIELD-EFFECT TRANSISTOR
The FET’s three terminals are:[3]
• Source (S), through which the carriers enter the
channel. Conventionally, current entering the chan-
nel at S is designated by IS.
• Drain (D), through which the carriers leave the
channel. Conventionally, current entering the chan-
nel at D is designated by ID. Drain-to-source voltage
is VDS.
• Gate (G), the terminal that modulates the channel
conductivity. By applying voltage to G, one can con-
trol ID.
1.3.3 More about terminals
Gate
Oxide
Source Drain
n+ x n+
p L
Body
Cross section of an n-type MOSFET
All FETs have source, drain, and gate terminals that cor-
respond roughly to the emitter, collector, and base of
BJTs. Most FETs have a fourth terminal called the body,
base, bulk, or substrate. This fourth terminal serves to
bias the transistor into operation; it is rare to make non-
trivial use of the body terminal in circuit designs, but its
presence is important when setting up the physical layout
of an integrated circuit. The size of the gate, length L
in the diagram, is the distance between source and drain.
The width is the extension of the transistor, in the direc-
tion perpendicular to the cross section in the diagram.
Typically the width is much larger than the length of the
gate. A gate length of 1 µm limits the upper frequency to
about 5 GHz, 0.2 µm to about 30 GHz.
The names of the terminals refer to their functions. The
gate terminal may be thought of as controlling the open-
ing and closing of a physical gate. This gate permits elec-
trons to flow through or blocks their passage by creating
29
or eliminating a channel between the source and drain.
Electrons flow from the source terminal towards the drain
terminal is influenced by an applied voltage. The body
simply refers to the bulk of the semiconductor in which
the gate, source and drain lie. Usually the body terminal
is connected to the highest or lowest voltage within the
circuit, depending on the type of the FET. The body ter-
minal and the source terminal are sometimes connected
together since the source is often connected to the high-
est or lowest voltage within the circuit, although there are
several uses of FETs which do not have such a configu-
ration, such as transmission gates and cascode circuits.
1.3.4 FET operation
See also: Field effect (semiconductor)
The FET controls the flow of electrons (or electron holes)
I DS I DS
VDS VGS - VP
=
SAT
Linear region Saturation region
VGS0 = 0
VGS1 < VGS0
Saturation
VGS2 < VGS1
region
VGS3 < VGS2
Channel
VGS off
VP
Linear
region Channel off VDS
VGS4 < VVp
I–V characteristics and output plot of a JFET n-channel transis-
tor.
from the source to drain by affecting the size and shape of
a “conductive channel” created and influenced by voltage
(or lack of voltage) applied across the gate and source ter-
minals. (For simplicity, this discussion assumes that the
body and source are connected.) This conductive channel
is the “stream” through which electrons flow from source
to drain.
n-channel
In an n-channel depletion-mode device, a negative gate-
to-source voltage causes a depletion region to expand in
width and encroach on the channel from the sides, nar-
rowing the channel. If the depletion region expands to
completely close the channel, the resistance of the chan-
nel from source to drain becomes large, and the FET is ef-
fectively turned off like a switch. This is called pinch-off,
and the voltage at which it occurs is called the pinch-off
voltage. Conversely, a positive gate-to-source voltage in-
30
creases the channel size and allows electrons to flow eas- to be part of the ohmic or linear region, even where drain
ily. current is not approximately linear with drain voltage.
In an n-channel enhancement-mode device, a conductive
channel does not exist naturally within the transistor, and
a positive gate-to-source voltage is necessary to create
one. The positive voltage attracts free-floating electrons
within the body towards the gate, forming a conductive
channel. But first, enough electrons must be attracted
near the gate to counter the dopant ions added to the body
of the FET; this forms a region with no mobile carriers
called a depletion region, and the voltage at which this
occurs is referred to as the threshold voltage of the FET.
Further gate-to-source voltage increase will attract even
more electrons towards the gate which are able to create
a conductive channel from source to drain; this process is
called inversion.
p-channel
In a p-channel depletion-mode device, a positive volt-
age from gate to body creates a depletion layer by
forcing the positively charged holes away from the
gate-insulator/semiconductor interface, leaving exposed
a carrier-free region of immobile, negatively charged ac-
ceptor ions..
Operation
For either enhancement- or depletion-mode devices, at
drain-to-source voltages much less than gate-to-source
voltages, changing the gate voltage will alter the channel
resistance, and drain current will be proportional to drain
voltage (referenced to source voltage). In this mode the
FET operates like a variable resistor and the FET is said
to be operating in a linear mode or ohmic mode.[4][5]
If drain-to-source voltage is increased, this creates a sig-
nificant asymmetrical change in the shape of the chan-
nel due to a gradient of voltage potential from source
to drain. The shape of the inversion region becomes
“pinched-off” near the drain end of the channel. If drain-
to-source voltage is increased further, the pinch-off point
of the channel begins to move away from the drain to-
wards the source. The FET is said to be in satura-
tion mode;[6] although some authors refer to it as active
mode, for a better analogy with bipolar transistor operat-
ing regions.[7][8] The saturation mode, or the region be-
tween ohmic and saturation, is used when amplification is
needed. The in-between region is sometimes considered
CHAPTER 1. THE TRANSISTOR
Even though the conductive channel formed by gate-to-
source voltage no longer connects source to drain during
saturation mode, carriers are not blocked from flowing.
Considering again an n-channel enhancement-mode de-
vice, a depletion region exists in the p-type body, sur-
rounding the conductive channel and drain and source re-
gions. The electrons which comprise the channel are free
to move out of the channel through the depletion region
if attracted to the drain by drain-to-source voltage. The
depletion region is free of carriers and has a resistance
similar to silicon. Any increase of the drain-to-source
voltage will increase the distance from drain to the pinch-
off point, increasing the resistance of the depletion re-
gion in proportion to the drain-to-source voltage applied.
This proportional change causes the drain-to-source cur-
rent to remain relatively fixed, independent of changes
to the drain-to-source voltage, quite unlike its ohmic be-
havior in the linear mode of operation. Thus, in satura-
tion mode, the FET behaves as a constant-current source
rather than as a resistor, and can effectively be used as
a voltage amplifier. In this case, the gate-to-source volt-
age determines the level of constant current through the
channel.
1.3.5 Composition
The FET can be constructed from a number of semicon-
ductors, with silicon being by far the most common. Most
FETs are made with conventional bulk semiconductor
processing techniques, using a single crystal semiconduc-
tor wafer as the active region, or channel.
Among the more unusual body materials are amorphous
silicon, polycrystalline silicon or other amorphous semi-
conductors in thin-film transistors or organic field-effect
transistors (OFETs) that are based on organic semicon-
ductors; often, OFET gate insulators and electrodes are
made of organic materials, as well. Such FETs are
manufactured using a variety of materials such as sili-
con carbide (SiC), gallium arsenide (GaAs), gallium ni-
tride (GaN), and indium gallium arsenide (InGaAs). In
June 2011, IBM announced that it had successfully used
graphene-based FETs in an integrated circuit.[9][10] These
transistors are capable of about 2.23 GHz cutoff fre-
quency, much higher than standard silicon FETs.[11]
1.3. FIELD-EFFECT TRANSISTOR
1.3.6 Types of field-effect transistors
Depletion-type FETs under typical voltages: JFET, poly-silicon
MOSFET, double-gate MOSFET, metal-gate MOSFET, MESFET.
Depletion
Electrons
Holes
Metal
Insulator
Top: source, bottom: drain, left: gate, right: bulk. Voltages that
lead to channel formation are not shown.
The channel of a FET is doped to produce either an
n-type semiconductor or a p-type semiconductor. The
drain and source may be doped of opposite type to
the channel, in the case of enhancement mode FETs,
or doped of similar type to the channel as in deple-
tion mode FETs. Field-effect transistors are also distin-
guished by the method of insulation between channel and
gate. Types of FETs include:
• The JFET (junction field-effect transistor) uses a re-
verse biased p–n junction to separate the gate from
the body.
• The MOSFET (metal–oxide–semiconductor field-
effect transistor) utilizes an insulator (typically
SiO2 ) between the gate and the body.
31
• The DGMOSFET (dual-gate MOSFET) is a FET
with two insulated gates.
• The DEPFET is a FET formed in a fully depleted
substrate and acts as a sensor, amplifier and memory
node at the same time. It can be used as an image
(photon) sensor.
• The FREDFET (fast-reverse or fast-recovery epi-
taxial diode FET) is a specialized FET designed to
provide a very fast recovery (turn-off) of the body
diode.
• The HIGFET (heterostructure insulated gate field-
effect transistor) is now used mainly in research.
• The MODFET (modulation-doped field-effect
transistor) uses a quantum well structure formed by
graded doping of the active region.
• The TFET (tunnel field-effect transistor) is based on
band-to-band tunneling.[12]
• The IGBT (insulated-gate bipolar transistor) is a de-
vice for power control. It has a structure akin to a
MOSFET coupled with a bipolar-like main conduc-
tion channel. These are commonly used for the 200–
3000 V drain-to-source voltage range of operation.
Power MOSFETs are still the device of choice for
drain-to-source voltages of 1 to 200 V.
• The HEMT (high-electron-mobility transistor),
also called a HFET (heterostructure FET), can be
made using bandgap engineering in a ternary semi-
conductor such as AlGaAs. The fully depleted
wide-band-gap material forms the isolation between
gate and body.
• The ISFET (ion-sensitive field-effect transistor) can
be used to measure ion concentrations in a solution;
when the ion concentration (such as H+ , see pH elec-
trode) changes, the current through the transistor
will change accordingly.
• The BioFET (Biologically sensitive field-effect
transistor) is a class of sensors/biosensors based
on ISFET technology which are utilized to de-
tect charged molecules; when a charged molecule
is present, changes in the electrostatic field at
the BioFET surface result in a measurable change
in current through the transistor. These include
EnFETs, ImmunoFETs, GenFETs, DNAFETs,
CPFETs, BeetleFETs, and FETs based on ion-
channels/protein binding. [13]
32 CHAPTER 1. THE TRANSISTOR

• The MESFET (metal–semiconductor field-effect 1.3.8 Disadvantages of FET

transistor) substitutes the p–n junction of the JFET
with a Schottky barrier; and is used in GaAs and It has a relatively low gain-bandwidth product com-
other III-V semiconductor materials. pared to a BJT. The MOSFET has a drawback of being
very susceptible to overload voltages, thus requiring spe-
• The NOMFET is a nanoparticle organic memory cial handling during installation.[15] The fragile insulat-
field-effect transistor. ing layer of the MOSFET between the gate and channel
makes it vulnerable to electrostatic damage during han-
• The GNRFET (graphene nanoribbon field-effect dling. This is not usually a problem after the device has
transistor) uses a graphene nanoribbon for its chan- been installed in a properly designed circuit.
nel.
FETs often have a very low 'on' resistance and have a high
• The VeSFET (vertical-slit field-effect transistor) is 'off' resistance. However the intermediate resistances are
a square-shaped junctionless FET with a narrow slit significant, and so FETs can dissipate large amounts of
connecting the source and drain at opposite corners. power while switching. Thus efficiency can put a pre-
Two gates occupy the other corners, and control the mium on switching quickly, but this can cause transients
current through the slit. that can excite stray inductances and generate significant
voltages that can couple to the gate and cause uninten-
• The CNTFET (carbon nanotube field-effect transis- tional switching. FET circuits can therefore require very
tor). careful layout and can involve trades between switching
speed and power dissipation.
• The OFET (organic field-effect transistor) uses an
organic semiconductor in its channel.

• The DNAFET (DNA field-effect transistor) is a 1.3.9 Uses of FET

specialized FET that acts as a biosensor, by using
a gate made of single-strand DNA molecules to de- The most commonly used FET is the MOSFET. The
tect matching DNA strands. CMOS (complementary metal oxide semiconductor)
process technology is the basis for modern digital
• The QFET (quantum field effect transistor) takes integrated circuits. This process technology uses an
advantage of quantum tunneling to greatly increase arrangement where the (usually “enhancement-mode”)
the speed of transistor operation by eliminating the p-channel MOSFET and n-channel MOSFET are con-
traditional transistor’s area of electron conduction. nected in series such that when one is ON, the other is
OFF.
In FETs, electrons can flow in either direction through
1.3.7 Advantages of FET the channel when operated in the linear mode. The nam-
ing convention of drain terminal and source terminal is
The main advantage of the fet is its high input resistance, somewhat arbitrary, as the devices are typically (but not
on the order of 100 MΩ or more. Thus, it is a voltage- always) built symmetrically from source to drain. This
controlled device, and shows a high degree of isolation makes FETs suitable for switching analog signals between
between input and output. It is a unipolar device, de- paths (multiplexing). With this concept, one can con-
pending only on majority current flow.. Because base struct a solid-state mixing board, for example.
current noise will increase with shaping time,[14] a FET
typically produces less noise than a bipolar junction tran- A common use of the FET is as an amplifier. For exam-
sistor (BJT), and is thus found in noise sensitive electron- ple, due to its large input resistance and low output resis-
ics such as tuners and low-noise amplifiers for VHF and tance, it is effective as a buffer in common-drain (source
satellite receivers. It is relatively immune to radiation. It follower) configuration.
exhibits no offset voltage at zero drain current and hence IGBTs see application in switching internal combustion
makes an excellent signal chopper. It typically has better engine ignition coils, where fast switching and voltage
thermal stability than a BJT.[3] blocking capabilities are important.
1.3. FIELD-EFFECT TRANSISTOR 33

1.3.10 See also FET [11] http://physicsworld.com/cws/article/news/2012/dec/10/

flexible-graphene-transistor-sets-new-records
• Chemical field-effect transistor
[12] Ionescu, A. M.; Riel, H. (2011). “Tunnel field-effect tran-
• Quantum field effect transistor sistors as energy-efficient electronic switches”. Nature
479 (7373): 329–337. doi:10.1038/nature10679. PMID
• ISFET 22094693.

[13] Poghossianb, Arshak (2002). “Recent advances in biolog-

• MOSFET
ically sensitive field-effect transistors (BioFETs) extquot-
• FET amplifier edbl. Analyst 127: 1137–1151. doi:10.1039/B204444G.

[14] http://www-physics.lbl.gov/~{}spieler/physics_198_
notes/PDF/VIII-5-noise.pdf
1.3.11 References
[15] Allen Mottershead (2004). Electronic devices and circuits.
[1] http://www.computerhistory.org/semiconductor/ New Delhi: Prentice-Hall of India. ISBN 81-203-0124-
timeline/1960-MOS.html 2.

[2] Jacob Millman (1985). Electronic devices and circuits.

Singapore: McGraw-Hill International. p. 397. ISBN 1.3.12 External links
0-07-085505-6.
• Field Effect Transistor Applications
[3] Millman (1985). Electronic devices and circuits. Singa-
pore: McGraw-Hill. pp. 384–385. ISBN 0-07-085505- • PBS The Field Effect Transistor
6.
• Junction Field Effect Transistor
[4] C Galup-Montoro & Schneider MC (2007). MOS-
FET modeling for circuit analysis and design. Lon- • CMOS gate circuitry
don/Singapore: World Scientific. p. 83. ISBN 981-256-
810-7. • Winning the Battle Against Latchup in CMOS Ana-
log Switches
[5] Norbert R Malik (1995). Electronic circuits: analysis,
simulation, and design. Englewood Cliffs, NJ: Prentice • Nanotube FETs at IBM Research
Hall. pp. 315–316. ISBN 0-02-374910-5.
• Field Effect Transistors in Theory and Practice
[6] RR Spencer & Ghausi MS (2001). Microelec-
tronic circuits. Upper Saddle River NJ: Pearson • The Field Effect Transistor as a Voltage Controlled
Education/Prentice-Hall. p. 102. ISBN 0-201-36183-3. Resistor

[7] A. S. Sedra and K.C. Smith (2004). Microelectronic cir-

cuits (Fifth Edition ed.). New York: Oxford. p. 552.
ISBN 0-19-514251-9.

[8] PR Gray, PJ Hurst, SH Lewis & RG Meyer (2001). Anal-

ysis and design of analog integrated circuits (Fourth Edi-
tion ed.). New York: Wiley. pp. §1.5.2 p. 45. ISBN
0-471-32168-0.

[9] http://www.physorg.com/news/
2011-06-ibm-graphene-based-circuit.html

[10] Lin, Y.-M., Valdes-Garcia, A., Han, S.-J., Farmer, D.

B., Sun, Y, Wu, Y, Dimitrakopoulos, C., Grill, A,
Avouris, P, and Jenkins, K. A. (2011). “Wafer-Scale
Graphene Integrated Circuit”. Science 332: 1294–1297.
doi:10.1126/science.1204428. PMID 21659599.
Chapter 2

Semiconductor Materials

2.1 Silicon combined with silica sand and gravel, to make concrete.
Silicates are also in whiteware ceramics such as porcelain,
and in traditional quartz-based soda-lime glass and many
This article is about the chemical element. For other other specialty glasses. More modern silicon compounds
uses, see Silicon (disambiguation). such as silicon carbide form abrasives and high-strength
“Element 14” redirects here. For other uses, see Element ceramics. Silicon is the basis of the widely-used synthetic
14 (disambiguation). polymers called silicones.
Not to be confused with the silicon-containing synthetic
Elemental silicon also has a large impact on the modern
polymer silicone.
world economy. Although most free silicon is used in the
steel refining, aluminium-casting, and fine chemical in-
Silicon is a chemical element with symbol Si and atomic dustries (often to make fumed silica), the relatively small
number 14. It is a tetravalent metalloid, less reactive than portion of very highly purified silicon that is used in semi-
its chemical analog carbon, the nonmetal directly above it conductor electronics (< 10%) is perhaps even more crit-
in the periodic table, but more reactive than germanium, ical. Because of wide use of silicon in integrated circuits,
the metalloid directly below it in the table. Controversy the basis of most computers, a great deal of modern tech-
about silicon’s character dates to its discovery; it was first nology depends on it.
prepared and characterized in pure form in 1823. In
Silicon is an essential element in biology, although only
1808, it was given the name silicium (from Latin: silex,
tiny traces of it appear to be required by animals.[8] How-
hard stone or flint), with an -ium word-ending to suggest
ever, various sea sponges as well as microorganisms like
a metal, a name which the element retains in several non-
diatoms and radiolaria secrete skeletal structures made
English languages. However, its final English name, first
of silica. Silica is often deposited in plant tissues, such
suggested in 1817, reflects the more physically similar el-
as in the bark and wood of Chrysobalanaceae and the
ements carbon and boron.
silica cells and silicified trichomes of Cannabis sativa,
Silicon is the eighth most common element in the uni- horsetails and many grasses.[9]
verse by mass, but very rarely occurs as the pure free el-
ement in nature. It is most widely distributed in dusts,
sands, planetoids, and planets as various forms of silicon 2.1.1 Characteristics
dioxide (silica) or silicates. Over 90% of the Earth’s
crust is composed of silicate minerals, making silicon the Physical
second most abundant element in the Earth’s crust (about
28% by mass) after oxygen.[7] Further information: Monocrystalline silicon
Most silicon is used commercially without being sepa-
rated, and indeed often with little processing of com- Silicon is a solid at room temperature, with relatively high
pounds from nature. These include direct industrial melting and boiling points of 1414 and 3265 °C, respec-
building-use of clays, silica sand and stone. Silicate goes tively. It has a greater density in a liquid state than a solid
into Portland cement for mortar and stucco, and when state. It does not contract when it freezes like most sub-

34
2.1. SILICON 35

Silicon powder
Silicon crystallizes in a diamond cubic crystal structure

and form five or six bonds in a sometimes more labile

stances, but expands, similar to how ice is less dense than
silicate form. Tetra-valent silicon is relatively inert, but
water. With a relatively high thermal conductivity of 149
still reacts with halogens and dilute alkalis, but most acids
W·m−1 ·K−1 , silicon conducts heat well and as a result is
(except for some hyper-reactive combinations of nitric
not often used to insulate hot objects.
acid and hydrofluoric acid) have no known effect on it.
In its crystalline form, pure silicon has a gray color However, having four bonding electrons gives it, like car-
and a metallic luster. Like germanium, silicon is rather bon, many opportunities to combine with other elements
strong, very brittle, and prone to chipping. Silicon, like or compounds in the right circumstances.
carbon and germanium, crystallizes in a diamond cubic
crystal structure, with a lattice spacing of 0.5430710 nm
(5.430710 Å).[10] Isotopes
The outer electron orbital of silicon, like that of carbon,
has four valence electrons. The 1s, 2s, 2p and 3s subshells Main article: isotopes of silicon
are completely filled while the 3p subshell contains two
electrons out of a possible six. Naturally occurring silicon is composed of three sta-
Silicon is a semiconductor. It has a negative temperature ble isotopes, silicon-28, silicon-29, and silicon-30, with
coefficient of resistance, since the number of free charge silicon-28 being the most abundant (92% natural abun-
[12]
carriers increases with temperature. The electrical re- dance). Out of these, only silicon-29 is of use in NMR
[13]
sistance of single crystal silicon significantly changes and EPR spectroscopy. Twenty radioisotopes have
under the application of mechanical stress due to the been characterized, with the most stable being silicon-32
piezoresistive effect.[11] with a half-life of 170 years, and silicon-31 with a half-
life of 157.3 minutes.[12] All of the remaining radioactive
isotopes have half-lives that are less than seven seconds,
Chemical and the majority of these have half-lives that are less than
one tenth of a second.[12] Silicon does not have any known
[12]
Silicon is a metalloid, readily either donating or shar- nuclear isomers.
ing its four outer electrons, allowing for many forms of The isotopes of silicon range in mass number from 22
chemical bonding. Like carbon, it typically forms four to 44.[12] The most common decay mode of six isotopes
bonds. Unlike carbon, it can accept additional electrons with mass numbers lower than the most abundant stable
36
isotope, silicon-28, is β+, primarily forming aluminium
isotopes (13 protons) as decay products.[12] The most
common decay mode(s) for 16 isotopes with mass num-
bers higher than silicon-28 is β−, primarily forming phos-
phorus isotopes (15 protons) as decay products.[12]
Occurrence
Quartz crystal cluster from Tibet. The naturally occurring min-
eral is a network solid with the formula SiO2 .
See also: Silicate minerals
Measured by mass, silicon makes up 27.7% of the Earth’s
crust and is the second most abundant element in the
crust, with only oxygen having a greater abundance.[14]
Silicon is usually found in the form of complex silicate
minerals, and less often as silicon dioxide (silica, a ma-
jor component of common sand). Pure silicon crystals
are very rarely found in nature.
The silicate minerals—various minerals containing sili-
con, oxygen and reactive metals—account for 90% of
the mass of the Earth’s crust. This is due to the fact
that at the high temperatures characteristic of the forma-
tion of the inner solar system, silicon and oxygen have a
great affinity for each other, forming networks of silicon
and oxygen in chemical compounds of very low volatil-
ity. Since oxygen and silicon were the most common
non-gaseous and non-metallic elements in the debris from
CHAPTER 2. SEMICONDUCTOR MATERIALS
supernova dust which formed the protoplanetary disk in
the formation and evolution of the Solar System, they
formed many complex silicates which accreted into larger
rocky planetesimals that formed the terrestrial planets.
Here, the reduced silicate mineral matrix entrapped the
metals reactive enough to be oxidized (aluminium, cal-
cium, sodium, potassium and magnesium). After loss of
volatile gases, as well as carbon and sulfur via reaction
with hydrogen, this silicate mixture of elements formed
most of the Earth’s crust. These silicates were of rela-
tively low density with respect to iron, nickel, and other
metals non-reactive to oxygen and thus a residuum of un-
combined iron and nickel sank to the planet’s core, leav-
ing a thick mantle consisting mostly of magnesium and
iron silicates. These are thought to be mostly silicate per-
ovskites, followed in abundance by the magnesium/iron
oxide ferropericlase.[15]
Examples of silicate minerals in the crust include those
in the pyroxene, amphibole, mica, and feldspar groups.
These minerals occur in clay and various types of rock
such as granite and sandstone.
Silica occurs in minerals consisting of very pure sili-
con dioxide in different crystalline forms, quartz, agate
amethyst, rock crystal, chalcedony, flint, jasper, and
opal. The crystals have the empirical formula of sili-
con dioxide, but do not consist of separate silicon dioxide
molecules in the manner of solid carbon dioxide. Rather,
silica is structurally a network-solid consisting of silicon
and oxygen in three-dimensional crystals, like diamond.
Less pure silica forms the natural glass obsidian. Biogenic
silica occurs in the structure of diatoms, radiolaria and
siliceous sponges.
Silicon is also a principal component of many meteorites,
and is a component of tektites, a silicate mineral of pos-
sibly lunar origin, or (if Earth-derived) which has been
subjected to unusual temperatures and pressures, possi-
bly from meteorite strike.
2.1.2 Production
Alloys
Ferrosilicon, an iron-silicon alloy that contains varying
ratios of elemental silicon and iron, accounts for about
80% of the world’s production of elemental silicon, with
China, the leading supplier of elemental silicon, provid-
ing 4.6 million tonnes (or 2/3 of the world output) of
silicon, most of which is in the form of ferrosilicon. It
2.1. SILICON 37

of the following ways: SiO2 + C → SiO + CO or SiO + 2

C → SiC + CO. However, provided the concentration of
SiO2 is kept high, the silicon carbide can be eliminated
by the chemical reaction 2 SiC + SiO2 → 3 Si + 2 CO.
As noted above, metallurgical grade silicon “metal” has
its primary use in the aluminium casting industry to make
aluminium-silicon alloy parts. The remainder (about
45%) is used by the chemical industry, where it is pri-
marily employed to make fumed silica, with the rest used
in production of other fine chemicals such as silanes and
some types of silicones.[18]
As of September 2008, metallurgical grade silicon costs
Ferrosilicon alloy about US$1.45 per pound ($3.20/kg),[19] up from $0.77
per pound ($1.70/kg) in 2005.[20]

is followed by Russia (610,000 t), Norway (330,000 t),

Brazil (240,000 t) and the United States (170,000 t).[16] Electronic grade
Ferrosilicon is primarily used by the steel industry (see
below). The use of silicon in semiconductor devices demands a
much greater purity than afforded by metallurgical grade
Aluminium-silicon alloys (called silumin alloys) are heav-
silicon. Very pure silicon (>99.9%) can be extracted di-
ily used in the aluminium alloy casting industry, where
rectly from solid silica or other silicon compounds by
silicon is the single most important additive to aluminium
molten salt electrolysis.[21][22] This method, known as
to improve its casting properties. Since cast aluminium
early as 1854[23] (see also FFC Cambridge process), has
is widely used in the automobile industry, this use of sil-
the potential to directly produce solar-grade silicon with-
icon is thus the single largest industrial use (about 55%
out any carbon dioxide emission at much lower energy
of the total) of “metallurgical grade” pure silicon (as this
consumption.
purified silicon is added to pure aluminium, whereas fer-
rosilicon is never purified before being added to steel).[17] Solar grade silicon cannot be used for microelectronics.
To properly control the quantum mechanical properties,
the purity of the silicon must be very high. Bulk sili-
Metallurgical grade con wafers used at the beginning of the integrated cir-
cuit making process must first be refined to a purity of
Elemental silicon not alloyed with significant quantities 99.9999999% often referred to as “9N” for “9 nines”, a
of other elements, and usually > 95%, is often referred to process which requires repeated applications of refining
loosely as silicon metal. It makes up about 20% of the technology.
world total elemental silicon production, with less than 1
to 2% of total elemental silicon (5–10% of metallurgi- The majority of silicon crystals grown for device produc-
cal grade silicon) ever purified to higher grades for use in tion are produced by the Czochralski process, (CZ-Si) It
electronics. Metallurgical grade silicon is commercially was the cheapest method available. However, single crys-
prepared by the reaction of high-purity silica with wood, tals grown by the Czochralski process contain impurities
charcoal, and coal in an electric arc furnace using carbon because the crucible containing the melt often dissolves.
electrodes. At temperatures over 1,900 °C (3,450 °F), Historically, a number of methods have been used to pro-
the carbon in the aforementioned materials and the sili- duce ultra-high-purity silicon.
con undergo the chemical reaction SiO2 + 2 C → Si + 2 Early silicon purification techniques were based on the
CO. Liquid silicon collects in the bottom of the furnace, fact that if silicon is melted and re-solidified, the last parts
which is then drained and cooled. The silicon produced in of the mass to solidify contain most of the impurities.
this manner is called metallurgical grade silicon and is at The earliest method of silicon purification, first described
least 98% pure. Using this method, silicon carbide (SiC) in 1919 and used on a limited basis to make radar compo-
may also form from an excess of carbon in one or both nents during World War II, involved crushing metallurgi-
38
cal grade silicon and then partially dissolving the silicon lizes tribromosilane in place of trichlorosilane and fluid
powder in an acid. When crushed, the silicon cracked
so that the weaker impurity-rich regions were on the out-
side of the resulting grains of silicon. As a result, the
impurity-rich silicon was the first to be dissolved when
treated with acid, leaving behind a more pure product.
In zone melting, also called zone refining, the first silicon yet to be any major factories built using this process.
purification method to be widely used industrially, rods
of metallurgical grade silicon are heated to melt at one
end. Then, the heater is slowly moved down the length of
the rod, keeping a small length of the rod molten as the
silicon cools and re-solidifies behind it. Since most im-
purities tend to remain in the molten region rather than
re-solidify, when the process is complete, most of the im-
purities in the rod will have been moved into the end that
was the last to be melted. This end is then cut off and dis-
carded, and the process repeated if a still higher purity is
desired.[24]
At one time, DuPont produced ultra-pure silicon by re-
acting silicon tetrachloride with high-purity zinc vapors
at 950 °C, producing silicon by SiCl4 + 2 Zn → Si + 2
ZnCl2 . However, this technique was plagued with prac-
tical problems (such as the zinc chloride byproduct solid-
ifying and clogging lines) and was eventually abandoned
in favor of the Siemens process. In the Siemens process,
high-purity silicon rods are exposed to trichlorosilane at
1150 °C. The trichlorosilane gas decomposes and de-
posits additional silicon onto the rods, enlarging them be-
cause 2 HSiCl3 → Si + 2 HCl + SiCl4 . Silicon produced
from this and similar processes is called polycrystalline
silicon. Polycrystalline silicon typically has impurity lev-
els of less than one part per billion.[25][26][27]
In 2006 REC announced construction of a plant based on
fluidized bed (FB) technology using silane: 3 SiCl4 + Si
+ 2 H2 → 4 HSiCl3 , 4 HSiCl3 → 3 SiCl4 + SiH4 , SiH4
→ Si + 2 H2 .[28] The advantage of fluid bed technology
is that processes can be run continuously, yielding higher
yields than Siemens Process, which is a batch process.
Today, silicon is purified by converting it to a silicon
compound that can be more easily purified by distillation
than in its original state, and then converting that silicon
compound back into pure silicon. Trichlorosilane is the
silicon compound most commonly used as the interme-
diate, although silicon tetrachloride and silane are also
used. When these gases are blown over silicon at high
temperature, they decompose to high-purity silicon.
In addition, there is the Schumacher process, which uti-
CHAPTER 2. SEMICONDUCTOR MATERIALS
bed technology. It requires lower deposition tempera-
tures, lower capital costs to build facilities and operate,
no hazardous polymers nor explosive material, and pro-
duces no amorphous silicon dust waste, all of which are
drawbacks of the Siemens process.[29] However, there are
2.1.3 Compounds
• Silicon forms binary compounds called silicides
with many metallic elements whose properties range
from reactive compounds, e.g. magnesium silicide,
Mg2 Si through high melting refractory compounds
such as molybdenum disilicide, MoSi2 .[30]
• Silicon carbide, SiC (carborundum) is a hard, high
melting solid and a well known abrasive. It may also
be sintered into a type of high-strength ceramic used
in armor.
• Silane, SiH4 , is a pyrophoric gas with a simi-
lar tetrahedral structure to methane, CH4 . When
pure, it does not react with pure water or dilute
acids; however, even small amounts of alkali im-
purities from the laboratory glass can result in a
rapid hydrolysis.[31] There is a range of catenated
silicon hydrides that form a homologous series of
compounds, Si
nH
2n+2 where n = 2–8 (analogous to the alkanes).
These are all readily hydrolyzed and are thermally
unstable, particularly the heavier members.[32][33]
• Disilenes contain a silicon-silicon double bond
(analogous to the alkenes) and are generally highly
reactive requiring large substituent groups to stabi-
lize them.[34] A disilyne with a silicon-silicon triple
bond was first isolated in 2004; although as the com-
pound is non-linear, the bonding is dissimilar to that
in alkynes.[35]
• Tetrahalides, SiX4 , are formed with all the
halogens.[36] Silicon tetrachloride, for example, re-
acts with water, unlike its carbon analogue, carbon
tetrachloride.[37] Silicon dihalides are formed by the
high temperature reaction of tetrahalides and sili-
con; with a structure analogous to a carbene they are
reactive compounds. Silicon difluoride condenses to
form a polymeric compound, (SiF
2.1. SILICON 39

2) some are optically active when central chirality ex-

n.[33] ists. Long chain polymers containing a silicon back-
bone are known, such as polydimethysilylene (SiMe
• Silicon dioxide (silica) is a high melting solid with a 2)
number of crystal forms; the most familiar of which n.[44] Polycarbosilane, [(SiMe
is the mineral quartz. In crystalline quartz each sil- 2)
icon atom is surrounded by four oxygen atoms that 2CH
bridge to other silicon atoms to form a three di- 2]
mensional lattice (see below for the vitreous or glass n with a backbone containing a repeating -Si-Si-C
form of pure silica). [37] Silica is soluble in water at unit, is a precursor in the production of silicon car-
high temperatures forming a range of compounds bide fibers.[44]
called monosilicic acid, Si(OH)4 .[38]
• Under the right conditions monosilicic acid read-
ily polymerizes to form more complex silicic acids, 2.1.4 History
ranging from the simplest condensate, disilicic acid
(H6 Si2 O7 ) to linear, ribbon, layer and lattice struc- Attention was first drawn to silica as the possible oxide of
tures which form the basis of the many silicate a fundamental chemical element by Antoine Lavoisier,
minerals and are called polysilicic acids {Siₓ(OH)₄– in 1787.[45] After an attempt to isolate silicon in 1808,
₂ₓ} .[38] Sir Humphry Davy proposed the name “silicium” for sil-
icon, from the Latin silex, silicis for flint, flints, and adding
• Silica can be fused directly into glass form, as so- the extquotedbl-ium” ending because he believed it was a
called fused quartz, which contains no crystalline metal.[46] In 1811, Gay-Lussac and Thénard are thought
structure. With with oxides of other elements, the to have prepared impure amorphous silicon, through the
high temperature reaction of silicon dioxide can give heating of recently isolated potassium metal with silicon
a wide range of mixed glasses and glass-like network tetrafluoride, but they did not purify and characterize
solids with various properties.[39] Examples include the product, nor identify it as a new element.[47] Silicon
soda-lime glass, borosilicate glass and lead crystal was given its present name in 1817 by Scottish chemist
glass. Thomas Thomson. He retained part of Davy’s name
• Silicon sulfide, SiS2 , is a polymeric solid (unlike its but added extquotedbl-on” because he believed that sili-
carbon analogue the liquid CS2 ).[40] con was a nonmetal similar to boron and carbon.[48] In
1823, Berzelius prepared amorphous silicon using ap-
• Silicon forms a nitride, Si3 N4 which is a ceramic.[41] proximately the same method as Gay-Lussac (potassium
Silatranes, a group of tricyclic compounds contain- metal and potassium fluorosilicate), but purifying the
ing five-coordinate silicon, may have physiological product to a brown powder by repeatedly washing it.[49]
properties.[42] As a result he is usually given credit for the element’s
discovery.[50][51]
• Many transition metal complexes containing a
metal-silicon bond are now known, which include Silicon in its more common crystalline form was not
complexes containing SiH prepared until 31 years later, by Deville.[52][53] By
nX electrolyzing impure sodium-aluminium chloride con-
3−n ligands, SiX3 ligands, and Si(OR)3 ligands.[42] taining approximately 10% silicon, he was able to obtain
a slightly impure allotrope of silicon in 1854.[54] Later,
• Silicones are large group of polymeric compounds more cost-effective methods have been developed to iso-
with an (Si-O-Si) backbone. An example is the late silicon in several allotrope forms, the most recent be-
silicone oil PDMS (polydimethylsiloxane). These ing silicene.
polymers can be crosslinked to produce resins and
elastomers.[43] Because silicon is an important element in semiconduc-
tors and high-technology devices, many places in the
• Many organosilicon compounds are known which world bear its name. For example, Silicon Valley in
contain a silicon-carbon single bond. Many of these California, bears the element’s name since it is the base
are based on a central tetrahedral silicon atom, and for a number of computer technology-related industries.
40 CHAPTER 2. SEMICONDUCTOR MATERIALS

Other geographic locations with connections to the in- of the inner planets of the solar system make planetary
dustry have since been named after silicon as well. Ex- silicon compounds found there mostly silicates and sil-
amples include Silicon Forest in Oregon, Silicon Hills in ica. Free silicon, or compounds of silicon in which the
Austin, Texas, Silicon Saxony in Germany, Silicon Valley element is covalently attached to hydrogen, boron, or ele-
in India, Silicon Border in Mexicali, Mexico, Silicon Fen ments other than oxygen, are mostly artificially produced.
in Cambridge, England, Silicon Roundabout in London, They are described below.
Silicon Glen in Scotland, and Silicon Gorge in Bristol, Silicon compounds of more modern origin function as
England. high-technology abrasives and new high-strength ceram-
ics based upon silicon carbide. Silicon is a component of
some superalloys.
2.1.5 Applications
Alternating silicon-oxygen chains with hydrogen attached
Compounds to the remaining silicon bonds form the ubiquitous
silicon-based polymeric materials known as silicones.
Building materials. Most silicon is used industrially These compounds containing silicon-oxygen and oc-
without being separated into the element, and indeed of- casionally silicon-carbon bonds have the capability to
ten with comparatively little processing from natural oc- act as bonding intermediates between glass and or-
currence. Over 90% of the Earth’s crust is composed ganic compounds, and to form polymers with useful
of silicate minerals, which are compounds of silicon and properties such as impermeability to water, flexibility
oxygen, often with metallic ions when charged silicate and resistance to chemical attack. Silicones are often
anions require cations to balance charge. Many of these used in waterproofing treatments, molding compounds,
have direct commercial uses, such as clays, silica sand and mold-release agents, mechanical seals, high temperature
most kinds of building stone. Thus, the vast majority of greases and waxes, and caulking compounds. Silicone is
uses for silicon are as structural compounds, either as the also sometimes used in breast implants, contact lenses,
silicate minerals or silica (crude silicon dioxide). Sili- explosives and pyrotechnics.[56] Silly Putty was originally
cates are used in making Portland cement (made mostly made by adding boric acid to silicone oil.[57]
of calcium silicates) which is used in building mortar and
modern stucco, but more importantly, combined with sil- Alloys
ica sand, and gravel (usually containing silicate minerals
like granite), to make the concrete that is the basis of most Elemental silicon is added to molten cast iron as
of the very largest industrial building projects of the mod- ferrosilicon or silicocalcium alloys to improve perfor-
ern world. [55] mance in casting thin sections and to prevent the forma-
Ceramics and glass. Silica is used to make fire brick, a tion of cementite where exposed to outside air. The pres-
type of ceramic. Silicate minerals are also in whiteware ence of elemental silicon in molten iron acts as a sink for
ceramics, an important class of products usually con- oxygen, so that the steel carbon content, which must be
taining various types of fired clay minerals (natural alu- kept within narrow limits for each type of steel, can be
minium phyllosilicates). An example is porcelain which more closely controlled. Ferrosilicon production and use
is based on the silicate mineral kaolinite. Ceramics in- is a monitor of the steel industry, and although this form
clude art objects, and domestic, industrial and building of elemental silicon is grossly impure, it accounts for 80%
products. Traditional glass (silica-based soda-lime glass) of the world’s use of free silicon. Silicon is an important
also functions in many of the same ways, and is also used constituent of electrical steel, modifying its resistivity and
for windows and containers. In addition, specialty silica ferromagnetic properties.
based glass fibers are used for optical fiber, as well as The properties of silicon can be used to modify alloys
fiberglass used for structural support and insulation. with metals other than iron. “Metallurgical grade” sili-
Artificial silicon compounds. Very occasional elemen- con is silicon of 95–99% purity. About 55% of the world
tal silicon is found in nature, and also naturally-occurring consumption of metallurgical purity silicon goes for pro-
compounds of silicon and carbon (silicon carbide) or ni- duction of aluminium-silicon alloys (silumin alloys) for
trogen (silicon nitride) are found in stardust samples or aluminium part casts, mainly for use in the automotive
meteorites in presolar grains, but the oxidizing conditions industry. Silicon’s importance in aluminium casting is
2.1. SILICON 41

that a significantly high amount (12%) of silicon in alu- der the proper conditions. Silicon has become the most
minium forms a eutectic mixture which solidifies with popular material to build both high power semiconduc-
very little thermal contraction. This greatly reduces tear- tors and integrated circuits. The reason is that silicon is
ing and cracks formed from stress as casting alloys cool to the semiconductor that can withstand the highest temper-
solidity. Silicon also significantly improves the hardness atures and electrical powers without becoming dysfunc-
and thus wear-resistance of aluminium.[17][18] tional due to avalanche breakdown (a process in which an
electron avalanche is created by a chain reaction process
whereby heat produces free electrons and holes, which in
Electronics turn produce more current which produces more heat). In
addition, the insulating oxide of silicon is not soluble in
Main article: Semiconductor device fabrication water, which gives it an advantage over germanium (an
Since most elemental silicon produced remains as fer- element with similar properties which can also be used
rosilicon alloy, only a relatively small amount (20%) of in semiconductor devices) in certain type of fabrication
[58]
the elemental silicon produced is refined to metallurgical techniques.
grade purity (a total of 1.3–1.5 million metric tons/year). Monocrystalline silicon is expensive to produce, and is
The fraction of silicon metal which is further refined to usually only justified in production of integrated circuits,
semiconductor purity is estimated at only 15% of the where tiny crystal imperfections can interfere with tiny
world production of metallurgical grade silicon.[18] How- circuit paths. For other uses, other types of pure silicon
ever, the economic importance of this small very high- which do not exist as single crystals may be employed.
purity fraction (especially the ~ 5% which is processed to These include hydrogenated amorphous silicon and up-
monocrystalline silicon for use in integrated circuits) is graded metallurgical-grade silicon (UMG-Si) which are
disproportionately large. used in the production of low-cost, large-area electron-
Pure monocrystalline silicon is used to produce silicon ics in applications such as liquid crystal displays, and of
wafers used in the semiconductor industry, in electron- large-area, low-cost, thin-film solar cells. Such semi-
ics and in some high-cost and high-efficiency photovoltaic conductor grades of silicon which are either slightly less
applications. In terms of charge conduction, pure sili- pure than those used in integrated circuits, or which are
con is an intrinsic semiconductor which means that unlike produced in polycrystalline rather than monocrystalline
metals it conducts electron holes and electrons that may form, make up roughly similar amount of silicon as are
be released from atoms within the crystal by heat, and produced for the monocrystalline silicon semiconduc-
thus increase silicon’s electrical conductivity with higher tor industry, or 75,000 to 150,000 metric tons per year.
temperatures. Pure silicon has too low a conductivity However, production of such materials is growing more
(i.e., too high a resistivity) to be used as a circuit element quickly than silicon for the integrated circuit market. By
in electronics. In practice, pure silicon is doped with 2013 polycrystalline silicon production, used mostly in
small concentrations of certain other elements, a pro- solar cells, is projected to reach 200,000 metric tons per
cess that greatly increases its conductivity and adjusts its year, while monocrystalline semiconductor silicon pro-
electrical response by controlling the number and charge duction (used in computer microchips) remains below
[18]
(positive or negative) of activated carriers. Such control 50,000 tons/year.
is necessary for transistors, solar cells, semiconductor de-
tectors and other semiconductor devices, which are used
in the computer industry and other technical applications. 2.1.6 Biological role
For example, in silicon photonics, silicon can be used as
a continuous wave Raman laser medium to produce co- Although silicon is readily available in the form of
herent light, though it is ineffective as an everyday light silicates, very few organisms have a use for it. Diatoms,
source. radiolaria and siliceous sponges use biogenic silica as
In common integrated circuits, a wafer of monocrys- a structural material to construct skeletons. In more
talline silicon serves as a mechanical support for the cir- advanced plants, the silica phytoliths (opal phytoliths)
cuits, which are created by doping, and insulated from are rigid microscopic bodies occurring in the cell;
each other by thin layers of silicon oxide, an insulator that some plants, for example rice, need silicon for their
is easily produced by exposing the element to oxygen un- growth.[59][60][61] The possible biological potential of sili-
42
con as bioavailable orthosilicic acid and the potential ben-
eficial effects on human health has been reviewed.[62]
Silicon is needed for synthesis of elastin and collagen;
the aorta contains the highest quantity of elastin and
silicon.[63] Silicon is currently under consideration for el-
evation to the status of a “plant beneficial substance by
the Association of American Plant Food Control Offi-
cials (AAPFCO).”[64][65] Silicon has been shown in uni-
versity and field studies to improve plant cell wall strength
and structural integrity,[66] improve drought and frost re-
sistance, decrease lodging potential and boost the plant’s
natural pest and disease fighting systems.[67] Silicon has
also been shown to improve plant vigor and physiology by
improving root mass and density, and increasing above
ground plant biomass and crop yields.[66]
Hypothetical silicon-based lifeforms are the subject of
silicon biochemistry, by analogy with carbon-based life-
forms. Silicon, being below carbon in the periodic table,
is thought to have similar enough properties that would
make silicon-based life possible, but much different from
life as we know it.
2.1.7 See also
• Amorphous silicon
• Black silicon
• Covalent superconductors
• List of silicon producers
• Monocrystalline silicon
• Polycrystalline silicon
• Printed silicon electronics
• Silicon tombac
• Silicon Valley
• Silicon Wadi
• Transistor
2.1.8 References
[1] Ram, R. S. et al. (1998). “Fourier Transform Emis-
sion Spectroscopy of the A2D–X2P Transition of SiH and
SiD”. J. Mol. Spectr. 190: 341–352. PMID 9668026.
CHAPTER 2. SEMICONDUCTOR MATERIALS
[2] Physical Properties of Silicon. New Semiconductor Ma-
terials. Characteristics and Properties. Ioffe Institute
[3] Magnetic susceptibility of the elements and inorganic
compounds, in Lide, D. R., ed. (2005). CRC Handbook
of Chemistry and Physics (86th ed.). Boca Raton (FL):
CRC Press. ISBN 0-8493-0486-5.
[4] Hopcroft, Matthew A.; Nix, William D.; Kenny, Thomas
W. (2010). “What is the Young’s Modulus of Silicon?
extquotedbl. Journal of Microelectromechanical Systems
19 (2): 229. doi:10.1109/JMEMS.2009.2039697.
[5] Weeks, Mary Elvira (1932). “The discovery of the
elements: XII. Other elements isolated with the aid
of potassium and sodium: beryllium, boron, sili-
con, and aluminum”. Journal of Chemical Education
9 (8): 1386–1412. Bibcode:1932JChEd...9.1386W.
doi:10.1021/ed009p1386.
[6] Voronkov, M. G. (2007). “Silicon era”. Rus-
sian Journal of Applied Chemistry 80 (12): 2190.
doi:10.1134/S1070427207120397.
[7] Nave, R. Abundances of the Elements in the Earth’s Crust,
Georgia State University
[8] Nielsen, Forrest H. (1984). “Ultratrace Elements
in Nutrition”. Annual Review of Nutrition 4: 21–
41. doi:10.1146/annurev.nu.04.070184.000321. PMID
6087860.
[9] Cutter, Elizabeth G. (1978). Plant Anatomy. Part 1 Cells
and Tissues (2 ed.). London: Edward Arnold. ISBN 0
7131 2639 6.
[10] O'Mara, William C. (1990). Handbook of Semiconductor
Silicon Technology. William Andrew Inc. pp. 349–352.
ISBN 0-8155-1237-6.
[11] Hull, Robert (1999). Properties of crystalline silicon. p.
421. ISBN 978-0-85296-933-5.
[12] NNDC contributors (2008). Alejandro A. Sonzogni
(Database Manager), ed. “Chart of Nuclides”. Upton
(NY): National Nuclear Data Center, Brookhaven Na-
tional Laboratory. Retrieved 2008-09-13.
[13] Jerschow, Alexej. “Interactive NMR Frequency Map”.
New York University. Retrieved 2011-10-20.
[14] Geological Survey (U.S.) (1975). Geological Survey pro-
fessional paper.
[15] Anderson, Don L. (2007) New Theory of the Earth.
Cambridge University Press. ISBN 978-0-521-84959-3,
ISBN 0-521-84959-4
2.1. SILICON
[16] “Silicon Commodities Report 2011”. USGS. Retrieved
2011-10-20.
[17] Apelian, D. (2009) Aluminum Cast Alloys: Enabling
Tools for Improved Performance. North American Die
Casting Association, Wheeling, Illinois.
[18] Corathers, Lisa A. 2009 Minerals Yearbook. USGS
[19] “Metallurgical silicon could become a rare commodity –
just how quickly that happens depends to a certain extent
on the current financial crisis”. Photon International. Re-
trieved 2009-03-04.
[20] “Silicon”. usgs.gov. Retrieved 2008-02-20.
[21] Rao, Gopalakrishna M. (1980). “Electrowinning of Sil-
icon from K2 SiF6 -Molten Fluoride Systems”. Jour-
nal of the Electrochemical Society 127 (9): 1940.
doi:10.1149/1.2130041.
[22] De Mattei, Robert C. (1981). “Electrodeposition of Sil-
icon at Temperatures above Its Melting Point”. Jour-
nal of the Electrochemical Society 128 (8): 1712.
doi:10.1149/1.2127716.
[23] Deville, H. St. C. (1854). “Recherches sur les métaux, et
en particulier sur l'aluminium et sur une nouvelle forme
du silicium”. Ann. Chim. Phys. 43: 31.
[24] Siffert, Paul; Krimmel, E. F (2004). Silicon: Evolution
and future of a technology. p. 33. ISBN 978-3-540-
40546-7.
[25] Yasuda, Kouji; Saegusa, Kunio; Okabe, Toru H.
(2010). “Production of Solar-grade Silicon by
Halidothermic Reduction of Silicon Tetrachloride”.
Metallurgical and Materials Transactions B 42: 37.
Bibcode:2011MMTB...42...37Y. doi:10.1007/s11663-
010-9440-y.
[26] Yasuda, Kouji; Okabe, Toru H. (2010). “Solar-
grade silicon production by metallothermic reduction”.
JOM 62 (12): 94. Bibcode:2010JOM....62l..94Y.
doi:10.1007/s11837-010-0190-8.
[27] Van Der Linden, P. C.; De Jonge, J. (2010). “The prepa-
ration of pure silicon”. Recueil des Travaux Chimiques des
Pays-Bas 78 (12): 962. doi:10.1002/recl.19590781204.
[28] “Analyst silicon field trip”. hugin.info. March 28, 2007.
Retrieved 2008-02-20.
[29] High Purity Polysilicon – Schumacher Process. Peak Sun
Silicon. Retrieved on 2011-08-07.
[30] Greenwood 1997, pp. 335–337.
[31] Greenwood 1997, p. 339.
43
[32] Greenwood 1997, p. 337.
[33] Holleman, Arnold F.; Wiberg, Nils (2007). Lehrbuch
der anorganischen Chemie (102 ed.). Berlin: de Gruyter.
ISBN 3-11-017770-6.
[34] Stone, F. G.; West, Robert (1996) Multiply Bonded Main
Group Metals and Metalloids, Academic Press, ISBN 0-
12-031139-9, p. 255
[35] Sekiguchi, A; Kinjo, R; Ichinohe, M (2004). “A stable
compound containing a silicon-silicon triple bond”. Sci-
ence 305 (5691): 1755–7. Bibcode:2004Sci...305.1755S.
doi:10.1126/science.1102209. PMID 15375262.
[36] Greenwood 1997, pp. 340–341.
[37] Greenwood 1997, p. 342.
[38] Greenwood 1997, p. 346.
[39] Greenwood 1997, p. 344.
[40] Greenwood 1997, pp. 359–360.
[41] Greenwood 1997, p. 360.
[42] Lickiss, Paul D. (1994). Inorganic Compounds of Silicon,
in Encyclopedia of Inorganic Chemistry. John Wiley &
Sons. pp. 3770–3805. ISBN 0-471-93620-0.
[43] Greenwood 1997, pp. 364–365.
[44] Mark, James. E (2005). Inorganic polymers. Oxford Uni-
versity Press. pp. 200–245. ISBN 0-19-513119-3.
[45] In his table of the elements, Lavoisier listed five “salifiable
earths” (i.e., ores that could be made to react with acids to
produce salts (salis = salt, in Latin)): chaux (calcium ox-
ide), magnésie (magnesia, magnesium oxide), baryte (bar-
ium sulfate), alumine (alumina, aluminium oxide), and
silice (silica, silicon dioxide). About these “elements”,
Lavoisier speculates: “We are probably only acquainted
as yet with a part of the metallic substances existing in
nature, as all those which have a stronger affinity to oxy-
gen than carbon possesses, are incapable, hitherto, of be-
ing reduced to a metallic state, and consequently, being
only presented to our observation under the form of oxyds,
are confounded with earths. It is extremely probable that
barytes, which we have just now arranged with earths, is
in this situation; for in many experiments it exhibits prop-
erties nearly approaching to those of metallic bodies. It
is even possible that all the substances we call earths may
be only metallic oxyds, irreducible by any hitherto known
process.” – from page 218 of: Lavoisier with Robert Kerr,
trans., Elements of Chemistry, … , 4th ed. (Edinburgh,
Scotland: William Creech, 1799). (The original passage
appears in: Lavoisier, Traité Élémentaire de Chimie, …
(Paris, France: Cuchet, 1789), vol. 1, page 174.)
44
[46] Davy, Humphry (1808) “Electro chemical researches, on
the decomposition of the earths; with observations on
the metals obtained from the alkaline earths, and on the
amalgam procured from ammonia,” Philosophical Trans-
actions of the Royal Society [of London], 98 : 333–370.
On page 353 Davy coins the name “silicium” : “Had I
been so fortunate as to have obtained more certain evi-
dences on this subject, and to have procured the metal-
lic substances I was in search of, I should have proposed
for them the names of silicium [silicon], alumium [alu-
minium], zirconium, and glucium [beryllium].”
[47] Gay-Lussac and Thenard, Recherches physico-chimiques
… (Paris, France: Deterville, 1811), vol. 1, pages 313–
314 ; vol. 2, page 55–65.
[48] Thomas Thomson, A System of Chemistry in Four Vol-
umes, 5th ed. (London, England: Baldwin, Cradock, and
Joy, 1817), vol. 1. From page 252: “The base of silica
has been usually considered as a metal, and called silicium.
But as there is not the smallest evidence for its metallic
nature, and as it bears a close resemblance to boron and
carbon, it is better to class it along with these bodies, and
to give it the name of silicon.”
[49] See:
• Berzelius announced his discovery of silicon (“sili-
cium”) in: Berzelius, J. (presented: 1823 ; pub-
lished: 1824) “Undersökning af flusspatssyran
och dess märkvärdigaste föreningar” (Investiga-
tion of hydrofluoric acid and of its most notewor-
thy compounds), Kongliga Vetenskaps-Academiens
Handlingar [Proceedings of the Royal Science
Academy], 12 : 46–98. The isolation of silicon
and its characterization are detailed in the sec-
tion titled “Flussspatssyrad kisseljords sönderdel-
ning med kalium,” pages 46–68.
• The above article was reprinted in German in:
J. J. Berzelius (1824) extquotedblII. Untersuchun-
gen über Flussspathsäure und deren merkwürdig-
sten Verbindungen extquotedbl (II. Investigations
of hydrofluoric acid and its most noteworthy com-
pounds), Annalen der Physik, 77 : 169–230. The
isolation of silicon is detailed in the section ti-
tled: “Zersetzung der flussspaths. Kieselerde durch
Kalium” (Decomposition of silicate fluoride by
potassium), pages 204–210.
• The above article was reprinted in French in:
Berzelius (1824) “Décomposition du fluate de silice
par le potassium” (Decomposition of silica fluoride
by potassium), Annales de Chimie et de Physique,
27 : 337–359.
CHAPTER 2. SEMICONDUCTOR MATERIALS
• Reprinted in English in: Berzelius (1825) “On the
mode of obtaining silicium, and on the charac-
ters and properties of that substance,” Philosophical
Magazine, 65 (324) : 254–267.
[50] Weeks, Mary Elvira (1932). “The discovery of the
elements: XII. Other elements isolated with the aid
of potassium and sodium: beryllium, boron, sili-
con, and aluminum”. Journal of Chemical Education
9 (8): 1386–1412. Bibcode:1932JChEd...9.1386W.
doi:10.1021/ed009p1386.
[51] Voronkov, M. G. (2007). “Silicon era”. Rus-
sian Journal of Applied Chemistry 80 (12): 2190.
doi:10.1134/S1070427207120397.
[52] In 1854, Deville was trying to prepare aluminium metal
from aluminium chloride that was heavily contaminated
with silicon chloride. Deville used two methods to pre-
pare aluminium: heating aluminium chloride with sodium
metal in an inert atmosphere (of hydrogen); and melting
aluminum chloride with sodium chloride and then elec-
trolyzing the mixture. In both cases, pure silicon was pro-
duced: the silicon dissolved in the molten aluminium, but
crystallized upon cooling. Dissolving the crude aluminum
in hydrochloric acid revealed flakes of crystallized silicon.
See: Henri Sainte-Claire Deville (1854) “Note sur deux
procédés de préparation de l'aluminium et sur une nou-
velle forme du silicium” (Note on two procedures for the
preparation of aluminium and on a new form of silicon),
Comptes rendus, 39 : 321–326.
Subsequently Deville obtained crystalline silicon by heat-
ing the chloride or fluoride of silicon with sodium metal,
isolating the amorphous silicon, then melting the amor-
phous form with salt and heating the mixture until most
of the salt evaporated. See: H. Sainte-Claire Deville
(1855) “Du silicium et du titane” (On silicon and tita-
nium), Comptes rendus, 40 : 1034–1036.
[53] Information on silicon – history, thermodynamic, chemi-
cal, physical and electronic properties: Etacude.com. El-
ements.etacude.com. Retrieved on 2011-08-07.
[54] Silicon: History. Nautilus.fis.uc.pt. Retrieved on 2011-
08-07.
[55] Greenwood 1997, p. 356.
[56] Koch, E.C.; Clement, D. (2007). “Special Materials in
Pyrotechnics: VI. Silicon – An Old Fuel with New Per-
spectives”. Propellants, Explosives, Pyrotechnics 32 (3):
205. doi:10.1002/prep.200700021.
[57] Walsh, Tim (2005). “Silly Putty”. Timeless toys: clas-
sic toys and the playmakers who created them. Andrews
McMeel Publishing. ISBN 978-0-7407-5571-2.
2.2. GERMANIUM 45

[58] Semiconductors Without the Quantum Physics. Elec- • Mineral.Galleries.com – Silicon

tropaedia
• WebElements.com – Silicon
[59] Rahman, Atta-ur-. “Silicon”. Studies in Natural Products
Chemistry 35. p. 856. ISBN 978-0-444-53181-0. • CDC – NIOSH Pocket Guide to Chemical Hazards

[60] Exley, C (1998). “Silicon in life:A bioinorganic solution

to bioorganic essentiality”. Journal of Inorganic Biochem-
istry 69 (3): 139. doi:10.1016/S0162-0134(97)10010-1.
2.2 Germanium
[61] Epstein, Emanuel (1999). “SILICON”. Annual Re- Not to be confused with Geranium.
view of Plant Physiology and Plant Molecular Biology 50:
641–664. doi:10.1146/annurev.arplant.50.1.641. PMID
15012222. Germanium is a chemical element with symbol Ge and
atomic number 32. It is a lustrous, hard, grayish-white
[62] Martin, Keith R. (2013). “Chapter 14. Silicon: The metalloid in the carbon group, chemically similar to its
Health Benefi ts of a Metalloid”. In Astrid Sigel, Helmut
group neighbors tin and silicon. Purified germanium is a
Sigel and Roland K. O. Sigel. Interrelations between Es-
semiconductor, with an appearance most similar to ele-
sential Metal Ions and Human Diseases. Metal Ions in Life
Sciences 13. Springer. pp. 451–473. doi:10.1007/978- mental silicon. Like silicon, germanium naturally reacts
94-007-7500-8_14. and forms complexes with oxygen in nature. Unlike sil-
icon, it is too reactive to be found naturally on Earth in
[63] LOEPER J., LOEPER J., FRAGNY M. The physiologi- the free (native) state.
cal role of the silicon and its antiatheromatous action Bio-
chemistry of silicon and related problems. Nobel Fonda- Because very few minerals contain it in high concentra-
tion Symposium 40. Edited by Gerd BENDZ and Ing- tion, germanium was discovered comparatively late in
var LINDQVIST. Plenum Press. New York and London. the history of chemistry. Germanium ranks near fifti-
1978. ISBN 0-306-33710-X eth in relative abundance of the elements in the Earth’s
crust. In 1869, Dmitri Mendeleev predicted its existence
[64] “AAPFCO Board of Directors 2006 Mid-Year Meeting”. and some of its properties based on its position on his
Association of American Plant Food Control Officials.
periodic table and called the element ekasilicon. Nearly
Retrieved 2011-07-18.
two decades later, in 1886, Clemens Winkler found the
[65] Miranda, Stephen R.; Bruce Barker. “Silicon: Summary new element along with silver and sulfur, in a rare mineral
of Extraction Methods”. Harsco Minerals. August 4, called argyrodite. Although the new element somewhat
2009. Retrieved 2011-07-18. resembled arsenic and antimony in appearance, its com-
bining ratios in the new element’s compounds agreed with
[66] “Silicon nutrition in plants”. Plant Health Care,Inc.: 1. 12
Mendeleev’s predictions for a relative of silicon. Winkler
December 2000. Retrieved 2011-07-01.
named the element after his country, Germany. Today,
[67] Prakash, Dr. N.B. (2007). “Evaluation of the calcium sil- germanium is mined primarily from sphalerite (the pri-
icate as a source of silicon in aerobic and wet rice”. Uni- mary ore of zinc), though germanium is also recovered
versity of Agricultural Science Bangalore: 1. commercially from silver, lead, and copper ores.
Germanium “metal” (isolated germanium) is used as a
2.1.9 Bibliography semiconductor in transistors and various other electronic
devices. Historically the first decade of semiconduc-
• Greenwood, Norman N; Earnshaw, Alan (1997). tor electronics was based entirely on germanium. To-
Chemistry of the Elements (2 ed.). Oxford: day, however, its production for use in semiconductor
Butterworth-Heinemann. ISBN 0-08-037941-9. electronics is a small fraction (2%) of that of ultra-high
purity silicon, which has largely replaced it. Presently,
germanium’s major end uses are in fibre-optic systems,
2.1.10 External links infrared optics and in solar cell applications. Germa-
nium compounds are also used for polymerization cat-
• Silicon at The Periodic Table of Videos (University alysts and have most recently found use in the produc-
of Nottingham) tion of nanowires. This element forms a large number
46 CHAPTER 2. SEMICONDUCTOR MATERIALS

of organometallic compounds, such as tetraethylgermane, con” confirmed that it belonged in this place on the peri-
which are useful in organometallic chemistry. odic table.[6][13] With further material from 500 kg of ore
Germanium is not thought to be an essential element from the mines in Saxony, Winkler confirmed the chemi-
for any living organism. Some complexed organic ger- cal properties of the new element in 1887.[5][6][14] He also
manium compounds are being investigated as possible determined an atomic weight of 72.32 by analyzing pure
pharmaceuticals, though none have yet proven success- germanium tetrachloride (GeCl
ful. Similar to silicon and aluminum, natural germa- 4), while Lecoq de Boisbaudran deduced 72.3 by a
nium compounds tend to be insoluble in water, and comparison of the lines in the spark spectrum of the
[15]
thus have little oral toxicity. However, synthetic soluble element.
germanium salts are nephrotoxic, and synthetic chemi- Winkler was able to prepare several new compounds
cally reactive germanium compounds with halogens and of germanium, including its fluorides, chlorides,
hydrogen are irritants and toxins. sulfides, germanium dioxide, and tetraethylgermane
(Ge(C2 H5 )4 ), the first organogermane.[5] The physical
data from these compounds — which corresponded well
2.2.1 History with Mendeleev’s predictions — made the discovery an
important confirmation of Mendeleev’s idea of element
See also: History of the transistor periodicity. Here is a comparison between the prediction
and Winkler’s data:[5]

In his report on The Periodic Law of the Chemical El-
ements, in 1869, the Russian chemist Dmitri Ivanovich
Mendeleev predicted the existence of several unknown
chemical elements, including one that would fill a gap in
the carbon family in his Periodic Table of the Elements,
located between silicon and tin.[3] Because of its position
in his Periodic Table, Mendeleev called it ekasilicon (Es),
and he estimated its atomic weight as about 72.0.
In mid-1885, at a mine near Freiberg, Saxony, a new
mineral was discovered and named argyrodite, because of
its high silver content.[n 1] The chemist Clemens Winkler
analyzed this new mineral, which proved to be a com-
bination of silver, sulfur, and a new element. Winkler
was able to isolate this new element and found it some-
what similar to antimony, in 1886.[5][6] Before Winkler
published his results on the new element, he decided that
he would name his element neptunium, since the recent
discovery of planet Neptune in 1846 had been preceded
by mathematical predictions of its existence.[n 2] How-
ever, the name “neptunium” had already been given to
another proposed chemical element (though not the el-
ement that today bears the name neptunium, which was
discovered in 1940),[n 3] so instead, Winkler named the
new element germanium (from the Latin word, Germa-
nia, for Germany) in honor of his homeland.[6] Argy-
rodite proved empirically to be Ag8 GeS6 .
Until the late 1930s, germanium was thought to be a
poorly conducting metal.[16] Germanium did not become
economically significant until after 1945, when its prop-
erties as a semiconductor were recognized as being use-
ful in electronics. During World War II, small amounts
of germanium had begun to be used in some special
electronic devices, mostly diodes.[17][18] Its first major
use was the point-contact Schottky diodes for radar pulse
detection during the War.[16] The first silicon-germanium
alloys were obtained in 1955.[19] Before 1945, only a
few hundred kilograms of germanium were produced
in smelters each year, but by the end of the 1950s,
the annual worldwide production had reached 40 metric
tons.[20]
The development of the germanium transistor in 1948[21]
opened the door to countless applications of solid state
electronics.[22] From 1950 through the early 1970s, this
area provided an increasing market for germanium,
but then high-purity silicon began replacing germanium
in transistors, diodes, and rectifiers.[23] For example,
the company that became Fairchild Semiconductor was
founded in 1957 with the express purpose of produc-
ing silicon transistors. Silicon has superior electrical
properties, but it requires much greater purity, which
could not be commercially achieved in the early years of
semiconductor electronics.[24]

Because this new element showed some similarities with Meanwhile, the demand for germanium for use in
the elements arsenic and antimony, its proper place in the fiber optics communication networks, infrared night vi-
periodic table was under consideration, but its similari- sion systems, and polymerization catalysts increased
ties with Dmitri Mendeleev’s predicted element “ekasili- dramatically.[20] These end uses represented 85% of
2.2. GERMANIUM
worldwide germanium consumption in 2000.[23] The US
government even designated germanium as a strategic
and critical material, calling for a 146 ton (132 t) supply
in the national defense stockpile in 1987.[20]
Germanium differs from silicon in that the supply for
germanium is limited by the availability of exploitable
sources, while the supply of silicon is only limited by pro-
duction capacity since silicon comes from ordinary sand
or quartz. As a result, while silicon could be bought in
1998 for less than $10 per kg,[20] the price of 1 kg of
germanium was then almost $800.[20]
2.2.2 Characteristics
Under standard conditions germanium is a brittle, silvery-
white, semi-metallic element.[25] This form constitutes an
allotrope technically known as α-germanium, which has
a metallic luster and a diamond cubic crystal structure,
the same as diamond.[23] At pressures above 120 kbar, a
different allotrope known as β-germanium forms, which
has the same structure as β-tin.[26] Along with silicon,
gallium, bismuth, antimony, and water, it is one of the
few substances that expands as it solidifies (i.e. freezes)
from its molten state.[26]
Germanium is a semiconductor. Zone refining tech-
niques have led to the production of crystalline germa-
nium for semiconductors that has an impurity of only
one part in 1010 ,[27] making it one of the purest materials
ever obtained.[28] The first metallic material discovered
(in 2005) to become a superconductor in the presence of
an extremely strong electromagnetic field was an alloy of
germanium with uranium and rhodium.[29]
Pure germanium is known to spontaneously extrude very
long screw dislocations. They are one of the primary
reasons for the failure of older diodes and transistors
made from germanium; depending on what they even-
tually touch, they may lead to an electrical short.
Chemistry
Elemental germanium oxidizes slowly to GeO2 at 250
°C.[30] Germanium is insoluble in dilute acids and alkalis
but dissolves slowly in concentrated sulfuric acid and re-
acts violently with molten alkalis to produce germanates
([GeO
3]2−
). Germanium occurs mostly in the oxidation state +4 al-
47
though many compounds are known with the oxidation
state of +2.[31] Other oxidation states are rare, such as +3
found in compounds such as Ge2 Cl6 , and +3 and +1 ob-
served on the surface of oxides,[32] or negative oxidation
states in germanes, such as −4 in GeH
4. Germanium cluster anions (Zintl ions) such as
Ge4 2− , Ge9 4− , Ge9 2− , [(Ge9 )2 ]6− have been prepared
by the extraction from alloys containing alkali met-
als and germanium in liquid ammonia in the presence
of ethylenediamine or a cryptand.[31][33] The oxidation
states of the element in these ions are not integers—
similar to the ozonides O3 − .
Two oxides of germanium are known: germanium diox-
ide (GeO
2, germania) and germanium monoxide, (GeO).[26] The
dioxide, GeO2 can be obtained by roasting germanium
disulfide (GeS
2), and is a white powder that is only slightly soluble in
water but reacts with alkalis to form germanates.[26] The
monoxide, germanous oxide, can be obtained by the high
temperature reaction of GeO2 with Ge metal.[26] The
dioxide (and the related oxides and germanates) exhibits
the unusual property of having a high refractive index
for visible light, but transparency to infrared light.[34][35]
Bismuth germanate, Bi4 Ge3 O12 , (BGO) is used as a
scintillator.[36]
Binary compounds with other chalcogens are also known,
such as the disulfide (GeS
2), diselenide (GeSe
2), and the monosulfide (GeS), selenide (GeSe), and
telluride (GeTe).[31] GeS2 forms as a white precipitate
when hydrogen sulfide is passed through strongly acid
solutions containing Ge(IV).[31] The disulfide is appre-
ciably soluble in water and in solutions of caustic al-
kalis or alkaline sulfides. Nevertheless, it is not solu-
ble in acidic water, which allowed Winkler to discover
the element.[37] By heating the disulfide in a current of
hydrogen, the monosulfide (GeS) is formed, which sub-
limes in thin plates of a dark color and metallic luster,
and is soluble in solutions of the caustic alkalis.[26] Upon
melting with alkaline carbonates and sulfur, germanium
compounds form salts known as thiogermanates.[38]
Four tetrahalides are known. Under normal conditions
GeI4 is a solid, GeF4 a gas and the others volatile liq-
uids. For example, germanium tetrachloride, GeCl4 ,
is obtained as a colorless fuming liquid boiling at 83.1
°C by heating the metal with chlorine.[26] All the tetra-
halides are readily hydrolyzed to hydrated germanium
48 CHAPTER 2. SEMICONDUCTOR MATERIALS

dioxide.[26] GeCl4 is used in the production of organoger- Isotopes

manium compounds.[31] All four dihalides are known and
in contrast to the tetrahalides are polymeric solids.[31] Ad- Main article: Isotopes of germanium
ditionally Ge2 Cl6 and some higher compounds of for-
mula GenCl₂n₊₂ are known.[26] The unusual compound
Ge6 Cl16 has been prepared that contains the Ge5 Cl12 Germanium has five naturally occurring isotopes, 70Ge,
unit with a neopentane structure.[39] 72Ge, 73Ge, 74Ge, 76Ge. Of these, 76Ge is very slightly
radioactive, decaying by double beta decay with a half-
Germane (GeH4 ) is a compound similar in structure to life of 1.78×1021 years. 74Ge is the most common iso-
methane. Polygermanes—compounds that are similar to tope, having a natural abundance of approximately 36%.
alkanes—with formula GenH₂n₊₂ containing up to five 76Ge is the least common with a natural abundance of
germanium atoms are known.[31] The germanes are less approximately 7%.[44] When bombarded with alpha par-
volatile and less reactive than their corresponding silicon ticles, the isotope 72Ge will generate stable 77Se, releas-
analogues.[31] GeH4 reacts with alkali metals in liquid ing high energy electrons in the process.[45] Because of
ammonia to form white crystalline MGeH3 which con- this, it is used in combination with radon for nuclear bat-
tain the GeH3 − anion.[31] The germanium hydrohalides teries.[45]
with one, two and three halogen atoms are colorless re-
active liquids.[31] At least 27 radioisotopes have also been synthesized rang-
ing in atomic mass from 58 to 89. The most stable of
The first organogermanium compound was synthesized these is 68Ge, decaying by electron capture with a half-
by Winkler in 1887; the reaction of germanium tetrachlo- life of 270.95 d. The least stable is 60Ge with a half-life
ride with diethylzinc yielded tetraethylgermane (Ge(C of 30 ms. While most of germanium’s radioisotopes de-
2H cay by beta decay, 61Ge and 64Ge decay by β+ delayed
5) proton emission.[44] 84Ge through 87Ge isotopes also ex-
4).[5] Organogermanes of the type R4 Ge (where R is an hibit minor β− delayed neutron emission decay paths.[44]
alkyl) such as tetramethylgermane (Ge(CH
3)
4) and tetraethylgermane are accessed through the cheap-
est available germanium precursor germanium tetrachlo-
ride and alkyl nucleophiles. Organic germanium hydrides Occurrence
such as isobutylgermane ((CH
3) Germanium is created through stellar nucleosynthesis,
2CHCH mostly by the s-process in asymptotic giant branch stars.
2GeH The s-process is a slow neutron capture of lighter el-
3) were found to be less hazardous and may be ements inside pulsating red giant stars.[46] Germanium
used as a liquid substitute for toxic germane gas has been detected in the atmosphere of Jupiter[47] and
in semiconductor applications. Many germanium in some of the most distant stars.[48] Its abundance in the
reactive intermediates are known: germyl free radicals, Earth’s crust is approximately 1.6 ppm.[49] There are only
germylenes (similar to carbenes), and germynes (similar a few minerals like argyrodite, briartite, germanite, and
to carbynes).[40][41] The organogermanium compound 2- renierite that contain appreciable amounts of germanium,
carboxyethylgermasesquioxane was first reported in the but no mineable deposits exist for any of them.[23][50]
1970s, and for a while was used as a dietary supplement Some zinc-copper-lead ore bodies contain enough ger-
and thought to possibly have anti-tumor qualities.[42] manium that it can be extracted from the final ore
[49]
Using a ligand called Eind (1,1,3,3,5,5,7,7-octaethyl-s- concentrate. An unusual enrichment process causes a
hydrindacen-4-yl) germanium is able to form a double high content of germanium in some coal seams, which
bond with oxygen (germanone).[43] was discovered by Victor Moritz Goldschmidt during a
broad survey for germanium deposits.[51][52] The highest
concentration ever found was in the Hartley coal ash with
up to 1.6% of germanium.[51][52] The coal deposits near
Xilinhaote, Inner Mongolia, contain an estimated 1600
tonnes of germanium.[49]
2.2. GERMANIUM 49

2.2.3 Production oxide is reduced by the reaction with hydrogen to obtain

germanium suitable for the infrared optics or semicon-
About 118 tonnes of germanium was produced in 2011 ductor industry:
worldwide, mostly in China (80 t), Russia (5 t) and
United States (3 t).[23] Germanium is recovered as a by- GeO2 + 2 H2 → Ge + 2 H2 O
product from sphalerite zinc ores where it is concentrated
in amounts of up to 0.3%,[53] especially from sediment-
hosted, massive Zn–Pb–Cu(–Ba) deposits and carbonate- The germanium for steel production and other [56]
industrial
hosted Zn–Pb deposits. Figures for worldwide Ge re- processes is normally reduced using carbon:
serves are not available, but in the US it is estimated at
450 tonnes.[23] In 2007 35% of the demand was met by GeO2 + C → Ge + CO2
recycled germanium.[49]
While it is produced mainly from sphalerite, it is also
2.2.4 Applications
found in silver, lead, and copper ores. Another source
of germanium is fly ash of coal power plants which use
The major end uses for germanium in 2007, worldwide,
coal from certain coal deposits with a large concentration
were estimated to be: 35% for fiber-optic systems, 30%
of germanium. Russia and China used this as a source
infrared optics, 15% for polymerization catalysts, and
for germanium.[54] Russia’s deposits are located in the far
15% for electronics and solar electric applications.[23]
east of the country on Sakhalin Island. The coal mines
The remaining 5% went into other uses such as phos-
northeast of Vladivostok have also been used as a germa-
phors, metallurgy, and chemotherapy.[23]
nium source. The deposits in China are mainly located in
the lignite mines near Lincang, Yunnan; coal mines near
Xilinhaote, Inner Mongolia are also used.[49] Optics
The ore concentrates are mostly sulfidic; they are con-
verted to the oxides by heating under air, in a process The most notable physical characteristics of germania
known as roasting: (GeO2 ) are its high index of refraction and its low optical
dispersion. These make it especially useful for wide-
angle camera lenses, microscopy, and for the core part
GeS2 + 3 O2 → GeO2 + 2 SO2
of optical fibers.[57][58] It also replaced titania as the sil-
ica dopant for silica fiber, eliminating the need for sub-
Part of the germanium ends up in the dust produced dur- sequent heat treatment, which made the fibers brittle.[59]
ing this process, while the rest is converted to germanates At the end of 2002 the fiber optics industry accounted for
which are leached together with the zinc from the cin- 60% of the annual germanium use in the United States,
der by sulfuric acid. After neutralization only the zinc but this use accounts for less than 10% of world wide
stays in solution and the precipitate contains the germa- consumption.[58] GeSbTe is a phase change material used
nium and other metals. After reducing the amount of for its optic properties, such as in rewritable DVDs.[60]
zinc in the precipitate by the Waelz process, the residing
Waelz oxide is leached a second time. The dioxide is ob- Because germanium is transparent in the infrared it is
tained as precipitate and converted with chlorine gas or a very important infrared optical material, that can be
hydrochloric acid to germanium tetrachloride, which has readily cut and polished into lenses and windows. It
a low boiling point and can be distilled off:[54] is especially used as the front optic in thermal imag-
ing cameras working in the 8 to 14 micron wavelength
range for passive thermal imaging and for hot-spot de-
GeO2 + 4 HCl → GeCl4 + 2 H2 O
tection in military, night vision system in cars, and fire
GeO2 + 2 Cl2 → GeCl4 + O2 fighting applications.[56] It is therefore used in infrared
spectroscopes and other optical equipment which require
Germanium tetrachloride is either hydrolyzed to the ox- extremely sensitive infrared detectors.[58] The material
ide (GeO2 ) or purified by fractional distillation and then has a very high refractive index (4.0) and so needs to
hydrolyzed.[54] The highly pure GeO2 is now suitable for be anti-reflection coated. Particularly, a very hard spe-
the production of germanium glass. The pure germanium cial antireflection coating of diamond-like carbon (DLC),
50 CHAPTER 2. SEMICONDUCTOR MATERIALS

refractive index 2.0, is a good match and produces a cious metal alloys. In sterling silver alloys, for instance, it
diamond-hard surface that can withstand much environ- has been found to reduce firescale, increase tarnish resis-
mental rough treatment.[61][62] tance, and increase the alloy’s response to precipitation
hardening. A tarnish-proof sterling silver alloy, trade-
marked Argentium, contains 1.2% germanium.[23]
Electronics
High purity germanium single crystal detectors can pre-
Silicon-germanium alloys are rapidly becoming an im- cisely identify radiation sources—for example in airport
[69]
portant semiconductor material, for use in high-speed security. Germanium is useful for monochromators
integrated circuits. Circuits utilizing the properties of for beamlines used in single crystal neutron scattering
Si-SiGe junctions can be much faster than those using and synchrotron X-ray diffraction. The reflectivity has
silicon alone.[63] Silicon-germanium is beginning to re- advantages over silicon in neutron and high energy X-
[70]
place gallium arsenide (GaAs) in wireless communica- ray applications. Crystals of high purity germanium
tions devices. The SiGe chips, with high-speed prop- are used in detectors for gamma spectroscopy and the
[23]
[71]
erties, can be made with low-cost, well-established pro- search for dark matter. The slightly radioactive Ger-
duction techniques of the silicon chip industry. [23] manium 76, which decays only through double-beta de-
cay, is used to study that process (for example, in the on-
The recent rise in energy cost has improved the eco- going MAJORANA demonstrator experiment).
nomics of solar panels, a potential major new use of
germanium.[23] Germanium is the substrate of the wafers
for high-efficiency multijunction photovoltaic cells for Inorganic germanium and health hazard
space applications.
Because germanium and gallium arsenide have very sim- Inorganic germanium and organic germanium are differ-
ilar lattice constants, germanium substrates can be used ent chemical compounds of germanium and their proper-
to make gallium arsenide solar cells.[64] The Mars Ex- ties are different. Inorganic germanium will accumulate
ploration Rovers and several satellites use triple junction inside the body and will impose health hazards after con-
gallium arsenide on germanium cells.[65] sumed. Organic germanium is reported to be potentially
beneficial for health.[72]
Germanium-on-insulator substrates are seen as a poten-
tial replacement for silicon on miniaturized chips.[23] Germanium is not thought to be essential to the health
Other uses in electronics include phosphors in fluorescent of plants or animals. Germanium in the environment
lamps,[27] and germanium-base solid-state light-emitting has little or no health impact. This is primarily be-
diodes (LEDs).[23] Germanium transistors are still used in cause it usually occurs only as a trace element in ores
some effects pedals by musicians who wish to reproduce and carbonaceous materials, and is used in very small
the distinctive tonal character of the “fuzz extquotedbl- quantities that are not likely to be ingested, [23]
in its vari-
tone from the early rock and roll era, most notably the ous industrial and electronic applications. For similar
Dallas Arbiter Fuzz Face. [66] reasons, germanium in end-uses has little impact on the
environment as a biohazard. Some reactive intermedi-
ate compounds of germanium are poisonous (see precau-
Other uses tions, below).[73]
As early as 1922, doctors in the United States used
Germanium dioxide is also used in catalysts for the inorganic form of germanium to treat patients with
polymerization in the production of polyethylene tereph- anemia.[73] It was used in other forms of treatments,
thalate (PET).[67] The high brilliance of the produced such as a purported immune system booster, but its
polyester is especially used for PET bottles marketed in efficiency has been dubious. Its role in cancer treat-
Japan.[67] However, in the United States, no germanium is ments has been debated, with the American Cancer So-
used for polymerization catalysts.[23] Due to the similarity
ciety contending that no anticancer effects have been
between silica (SiO2 ) and germanium dioxide (GeO2 ), demonstrated.[74][75] U.S. Food and Drug Administration
the silica stationary phase in some gas chromatography research has concluded that inorganic germanium, when
columns can be replaced by GeO2 .[68] used as a nutritional supplement, “presents potential hu-
In recent years germanium has seen increasing use in pre- man health hazard extquotedbl.[42]
2.2. GERMANIUM 51

Certain germanium compounds are available in low dose [3] R. Hermann published claims in 1877 of his discovery of a
in the U.S. as nonprescription dietary “supplements” in new element beneath tantalum in the periodic table, which
oral capsules or tablets. Other germanium compounds he named neptunium, after the Greek god of the oceans
have been administered by alternative medical practi- and seas.[9][10] However this metal was later recognized to
tioners as non-FDA-allowed injectable solutions. Solu- be an alloy of the elements niobium and tantalum.[11] The
name extquotedblneptunium extquotedbl was much later
ble inorganic forms of germanium used at first, notably
given to the synthetic element one step past uranium in the
the citrate-lactate salt, led to a number of cases of renal Periodic Table, which was discovered by nuclear physics
dysfunction, hepatic steatosis and peripheral neuropathy researchers in 1940.[12]
in individuals using them on a chronic basis. Plasma
and urine germanium concentrations in these individu-
als, several of whom died, were several orders of mag- 2.2.8 References
nitude greater than endogenous levels. A more recent
organic form, beta-carboxyethylgermanium sesquioxide [1] Magnetic susceptibility of the elements and inorganic
(propagermanium), has not exhibited the same spectrum compounds, in Handbook of Chemistry and Physics 81st
of toxic effects.[76] edition, CRC press.

Certain compounds of germanium have low toxicity
to mammals, but have toxic effects against certain
bacteria.[25]
2.2.5 Precautions for chemically reactive
germanium compounds
Some of germanium’s artificially-produced compounds
are quite reactive and present an immediate hazard to hu-
man health on exposure. For example, germanium chlo-
ride and germane (GeH4 ) are a liquid and gas, respec-
tively, that can be very irritating to the eyes, skin, lungs,
and throat.[72]
[2] “Properties of Germanium”. Ioffe Institute.
[3] Kaji, Masanori (2002). “D. I. Mendeleev’s concept of
chemical elements and The Principles of Chemistry ex-
tquotedbl (PDF). Bulletin for the History of Chemistry 27
(1): 4–16. Retrieved 2008-08-20.
[4] (PDF) Argyrodite—Ag
8GeS
6 (Report). Mineral Data Publishing. http://www.
handbookofmineralogy.org/pdfs/argyrodite.pdf. Re-
trieved 2008-09-01.
[5] Winkler, Clemens (1887). “Mittheilungen über des Ger-
manium. Zweite Abhandlung”. J. Prak. Chemie (in Ger-
man) 36 (1): 177–209. doi:10.1002/prac.18870360119.
Retrieved 2008-08-20.

[6] Winkler, Clemens (1887). “Germanium, Ge, a New

2.2.6 See also Nonmetal Element”. Berichte der deutschen chemis-
chen Gesellschaft (in German) 19 (1): 210–211.
• Transistor doi:10.1002/cber.18860190156. English translation at
• Vitrain the Wayback Machine (archived December 7, 2008)

[7] Adams, J. C. (November 13, 1846). “Explanation of the

2.2.7 Footnotes
[1] From Greek, argyrodite means silver-containing.[4]
[2] Just as the existence of the new element had been pre-
dicted, the existence of the planet Neptune had been pre-
dicted in about 1843 by the two mathematicians John
Couch Adams and Urbain Le Verrier, using the calcu-
lation methods of celestial mechanics. They did this
in attempts to explain the fact that the planet Uranus,
upon very close observation, appeared to be being pulled
slightly out of position in the sky.[7] James Challis started
searching for it in July 1846, and he sighted this planet on
September 23, 1846.[8]
observed irregularities in the motion of Uranus, on the hy-
pothesis of disturbance by a more distant planet”. Monthly
Notices of the Royal Astronomical Society (Blackwell Pub-
lishing) 7: 149. Bibcode:1846MNRAS...7..149A.
[8] Challis, Rev. J. (November 13, 1846). “Account of ob-
servations at the Cambridge observatory for detecting the
planet exterior to Uranus”. Monthly Notices of the Royal
Astronomical Society (Blackwell Publishing) 7: 145–149.
Bibcode:1846MNRAS...7..145C.
[9] Sears, Robert (July 1877). “Scientific Miscellany”. The
Galaxy (Columbus, O[hio]: Siebert & Lilley) 24 (1): 131.
ISBN 0-665-50166-8. OCLC 16890343 243523661
77121148.
52
[10] “Editor’s Scientific Record”. Harper’s new monthly mag-
azine 55 (325): 152–153. June 1877.
[11] van der Krogt, Peter. “Elementymology & Elements Mul-
tidict: Niobium”. Retrieved 2008-08-20.
[12] Westgren, A. (1964). “The Nobel Prize in Chemistry
1951: presentation speech”. Nobel Lectures, Chemistry
1942–1962. Elsevier.
[13] “Germanium, a New Non-Metallic Element”. The Manu-
facturer and Builder: 181. 1887. Retrieved 2008-08-20.
[14] Brunck, O. (1886). “Obituary: Clemens Win-
kler”. Berichte der deutschen chemischen
Gesellschaft (in German) 39 (4): 4491–4548.
doi:10.1002/cber.190603904164.
[15] de Boisbaudran, M. Lecoq (1886). “Sur le poids atomique
du germanium”. Comptes rendus (in French) 103: 452.
Retrieved 2008-08-20.
[16] Haller, E. E. “Germanium: From Its Discovery to SiGe
Devices” (PDF). Department of Materials Science and En-
gineering, University of California, Berkeley, and Materi-
als Sciences Division, Lawrence Berkeley National Labo-
ratory, Berkeley,. Retrieved 2008-08-22.
[17] W. K. (1953-05-10). “Germanium for Electronic De-
vices”. NY Times. Retrieved 2008-08-22.
[18] “1941 – Semiconductor diode rectifiers serve in WW II”.
Computer History Museum. Retrieved 2008-08-22.
[19] “SiGe History”. University of Cambridge. Retrieved
2008-08-22.
[20] Halford, Bethany (2003). “Germanium”. Chemical & En-
gineering News (American Chemical Society). Retrieved
2008-08-22.
[21] Bardeen, J.; Brattain, W. H. (1948). “The Tran-
sistor, A Semi-Conductor Triode”. Physical Reviews
74 (2): 230–231. Bibcode:1948PhRv...74..230B.
doi:10.1103/PhysRev.74.230.
[22] “Electronics History 4 – Transistors”. National Academy
of Engineering. Retrieved 2008-08-22.
[23] U.S. Geological Survey (2008). “Germanium—Statistics
and Information”. U.S. Geological Survey, Mineral Com-
modity Summaries. Retrieved 2008-08-28. “Select 2008”
[24] Teal, Gordon K. (July 1976). “Single Crystals of Germa-
nium and Silicon-Basic to the Transistor and Integrated
Circuit”. IEEE Transactions on Electron Devices. ED-23
(7): 621–639. doi:10.1109/T-ED.1976.18464.
CHAPTER 2. SEMICONDUCTOR MATERIALS
[25] Emsley, John (2001). Nature’s Building Blocks. Oxford:
Oxford University Press. pp. 506–510. ISBN 0-19-
850341-5.
[26] Holleman, A. F.; Wiberg, E.; Wiberg, N. (2007).
Lehrbuch der Anorganischen Chemie, 102nd ed. de
Gruyter. ISBN 978-3-11-017770-1. OCLC 145623740
180963521 219549154.
[27] “Germanium”. Los Alamos National Laboratory. Re-
trieved 2008-08-28.
[28] Chardin, B. (2001). “Dark Matter: Direct Detection”. In
Binetruy, B. The Primordial Universe: 28 June – 23 July
1999. Springer. p. 308. ISBN 3-540-41046-5.
[29] Lévy, F.; Sheikin, I.; Grenier, B.; Huxley, Ad.
(August 2005). “Magnetic field-induced supercon-
ductivity in the ferromagnet URhGe”. Science 309
(5739): 1343–1346. Bibcode:2005Sci...309.1343L.
doi:10.1126/science.1115498. PMID 16123293.
[30] Tabet, N; Salim, Mushtaq A. (1998). “KRXPS study of
the oxidation of Ge(001) surface”. Applied Surface Sci-
ence 134 (1–4): 275. Bibcode:1998ApSS..134..275T.
doi:10.1016/S0169-4332(98)00251-7.
[31] Greenwood, Norman N.; Earnshaw, Alan (1997). Chem-
istry of the Elements (2nd ed.). Butterworth-Heinemann.
ISBN 0080379419.
[32] Tabet, N; Salim, M.A; Al-Oteibi, A.L (1999). “XPS
study of the growth kinetics of thin films obtained by
thermal oxidation of germanium substrates”. Journal of
Electron Spectroscopy and Related Phenomena. 101–103:
233. doi:10.1016/S0368-2048(98)00451-4.
[33] Xu, Li; Sevov, Slavi C. (1999). “Oxidative Coupling of
Deltahedral [Ge9 ]4− Zintl Ions”. J. Am. Chem. Soc. 121
(39): 9245–9246. doi:10.1021/ja992269s.
[34] Bayya, Shyam S.; Sanghera, Jasbinder S.; Aggarwal,
Ishwar D.; Wojcik, Joshua A. (2002). “Infrared
Transparent Germanate Glass-Ceramics”. Journal of
the American Ceramic Society 85 (12): 3114–3116.
doi:10.1111/j.1151-2916.2002.tb00594.x.
[35] Drugoveiko, O. P.; Evstrop'ev, K. K.; Kondrat'eva, B. S.;
Petrov, Yu. A.; Shevyakov, A. M. (1975). “Infrared re-
flectance and transmission spectra of germanium dioxide
and its hydrolysis products”. Journal of Applied Spec-
troscopy 22 (2): 191. Bibcode:1975JApSp..22..191D.
doi:10.1007/BF00614256.
[36] Lightstone, A. W.; McIntyre, R. J.; Lecomte, R.; Schmitt,
D. (1986). “A Bismuth Germanate-Avalanche Photodi-
ode Module Designed for Use in High Resolution Positron
2.2. GERMANIUM
Emission Tomography”. IEEE Transactions on Nuclear
Science 33 (1): 456–459. Bibcode:1986ITNS...33..456L.
doi:10.1109/TNS.1986.4337142.
[37] Johnson, Otto H. (1952). “Germanium and its In-
organic Compounds”. Chem. Rev. 3 (3): 431.
doi:10.1021/cr60160a002.
[38] Fröba, Michael; Oberender, Nadine (1997). “First syn-
thesis of mesostructured thiogermanates”. Chemical
Communications (18): 1729. doi:10.1039/a703634e.
[39] Beattie, I.R.; Jones, P.J.; Reid, G.; Webster, M.;
(1998). “The Crystal Structure and Raman Spectrum of
Ge5 Cl12 ·GeCl4 and the Vibrational Spectrum of Ge2 Cl6
extquotedbl. Inorg. Chem. 37 (23): 6032–6034.
doi:10.1021/ic9807341. PMID 11670739.
[40] Satge, Jacques (1984). “Reactive intermediates in
organogermanium chemistry”. Pure & Appl. Chem. 56
(1): 137–150. doi:10.1351/pac198456010137.
[41] Quane, Denis; Bottei, Rudolph S. (1963). “Organoger-
manium Chemistry”. Chemical Reviews 63 (4): 403–442.
doi:10.1021/cr60224a004.
[42] Tao, S. H.; Bolger, P. M. (June 1997). “Hazard As-
sessment of Germanium Supplements”. Regulatory
Toxicology and Pharmacology 25 (3): 211–219.
doi:10.1006/rtph.1997.1098. PMID 9237323.
[43] Broadwith, Phillip (25 March 2012). “Germanium-
oxygen double bond takes centre stage”. Chemistry World.
Retrieved 2014-05-15.
[44] Audi, G.; Bersillon, O.; Blachot, J.; Wapstra, A.H.
(2003). “Nubase2003 Evaluation of Nuclear and De-
cay Properties”. Nuclear Physics A (Atomic Mass Data
Center) 729 (1): 3–128. Bibcode:2003NuPhA.729....3A.
doi:10.1016/j.nuclphysa.2003.11.001.
[45] Perreault, Bruce A. “Alpha Fusion Electrical Energy
Valve”, US Patent 7800286, issued September 21, 2010.
PDF copy at the Wayback Machine (archived October 12,
2007).
[46] Sterling, N. C.; Dinerstein, Harriet L.; Bowers, Charles
W. (2002). “Discovery of Enhanced Germanium Abun-
dances in Planetary Nebulae with the Far Ultravio-
let Spectroscopic Explorer”. The Astrophysical Jour-
nal Letters 578 (1): L55–L58. arXiv:astro-ph/0208516.
Bibcode:2002ApJ...578L..55S. doi:10.1086/344473.
[47] Kunde, V.; Hanel, R.; Maguire, W.; Gautier, D.; Ba-
luteau, J. P.; Marten, A.; Chedin, A.; Husson, N.;
Scott, N. (1982). “The tropospheric gas composition of
Jupiter’s north equatorial belt /NH3 , PH3 , CH3 D, GeH4 ,
53
H2 O/ and the Jovian D/H isotopic ratio”. Astrophysical
Journal 263: 443–467. Bibcode:1982ApJ...263..443K.
doi:10.1086/160516.
[48] Cowan, John (2003-05-01). “Astronomy: El-
ements of surprise”. Nature 423 (29): 29.
Bibcode:2003Natur.423...29C. doi:10.1038/423029a.
PMID 12721614.
[49] Höll, R.; Kling, M.; Schroll, E. (2007). “Metallogenesis
of germanium—A review”. Ore Geology Reviews 30 (3–
4): 145–180. doi:10.1016/j.oregeorev.2005.07.034.
[50] Lifton, Jack (2007-04-26). “Byproducts II: Another Ger-
manium Rush? extquotedbl. Resource Investor.com. Re-
trieved 2008-09-09.
[51] Goldschmidt, V. M. (1930). “Ueber das Vorkommen des
Germaniums in Steinkohlen und Steinkohlenprodukten”.
Nachrichten von der Gesellschaft der Wissenschaften zu
Göttingen, Mathematisch-Physikalische Klasse: 141–167.
[52] Goldschmidt, V. M.; Peters, Cl. (1933). “Zur Geochemie
des Germaniums”. Nachrichten von der Gesellschaft der
Wissenschaften zu Göttingen, Mathematisch-Physikalische
Klasse: 141–167.
[53] Bernstein, L (1985). “Germanium geochemistry
and mineralogy”. Geochimica et Cosmochimica Acta
49 (11): 2409. Bibcode:1985GeCoA..49.2409B.
doi:10.1016/0016-7037(85)90241-8.
[54] Naumov, A. V. (2007). “World market of germanium and
its prospects”. Russian Journal of Non-Ferrous Metals 48
(4): 265–272. doi:10.3103/S1067821207040049.
[55] R.N. Soar (1977). “USGS Minerals Information”. U.S.
Geological Survey Mineral Commodity Summaries (U.S.
Geological Survey). January 2003, January 2004, January
2005, January 2006, January 2007,January 2010. ISBN
0-85934-039-2. OCLC 16437701.
[56] Moskalyk, R. R. (2004). “Review of germanium process-
ing worldwide”. Minerals Engineering 17 (3): 393–402.
doi:10.1016/j.mineng.2003.11.014.
[57] Rieke, G.H. (2007). “Infrared Detector Arrays for As-
tronomy”. Annual Review of Astronomy and Astro-
physics 45 (1): 77. Bibcode:2007ARA&A..45...77R.
doi:10.1146/annurev.astro.44.051905.092436.
[58] Brown, Jr., Robert D. (2000). “Germanium” (PDF). U.S.
Geological Survey. Retrieved 2008-09-22.
[59] “Chapter III: Optical Fiber For Communications”. Stan-
ford Research Institute. Retrieved 2008-08-22.
54 CHAPTER 2. SEMICONDUCTOR MATERIALS

[60] “Understanding Recordable & Rewritable DVD First Edi- (1996). “Optimization of Germanium for Neutron
tion” (PDF). Optical Storage Technology Association Diffractometers”. International Journal of Modern
(OSTA). Archived from the original on 2009-04-19. Re- Physics E 5 (1): 131. Bibcode:1996IJMPE...5..131A.
trieved 2008-09-22. doi:10.1142/S0218301396000062.

[61] Lettington, Alan H. (1998). “Applications of diamond-
like carbon thin films”. Carbon 36 (5–6): 555–560.
doi:10.1016/S0008-6223(98)00062-1.
[62] Gardos, Michael N.; Bonnie L. Soriano; Steven H.
Propst (1990). “Study on correlating rain erosion resis-
tance with sliding abrasion resistance of DLC on ger-
manium”. In Feldman, Albert; Holly, Sandor. Proc.
SPIE. SPIE Proceedings 1325 (Mechanical Properties):
99. doi:10.1117/12.22449.
[71] Diehl, R.; Prantzos, N; Vonballmoos, P (2006).
“Astrophysical constraints from gamma-ray spec-
troscopy”. Nuclear Physics A 777: 70. arXiv:astro-
ph/0502324. Bibcode:2006NuPhA.777...70D.
doi:10.1016/j.nuclphysa.2005.02.155.
[72] Gerber, G.B.; Léonard, A. (1997). “Mutagenicity,
carcinogenicity and teratogenicity of germanium com-
pounds”. Regulatory Toxicology and Pharmacology 387
(3): 141–146. doi:10.1016/S1383-5742(97)00034-3.

[63] Washio, K. (2003). “SiGe HBT and BiCMOS tech-
nologies for optical transmission and wireless commu-
nication systems”. IEEE Transactions on Electron De-
vices 50 (3): 656. Bibcode:2003ITED...50..656W.
doi:10.1109/TED.2003.810484.
[64] Bailey, Sheila G.; Raffaelle, Ryne; Emery, Keith (2002).
“Space and terrestrial photovoltaics: synergy and diver-
sity”. Progress in Photovoltaics Research and Applications
10 (6): 399. doi:10.1002/pip.446.
[65] Crisp, D.; Pathare, A.; Ewell, R. C. (2004). “The per-
formance of gallium arsenide/germanium solar cells at
the Martian surface”. Acta Astronautica 54 (2): 83–101.
Bibcode:2004AcAau..54...83C. doi:10.1016/S0094-
5765(02)00287-4.
[66] Szweda, Roy (2005). “Germanium phoenix”. III-Vs Re-
view 18 (7): 55. doi:10.1016/S0961-1290(05)71310-7.
[73] Brown Jr., Robert D. (PDF). Commodity
Survey:Germanium (Report). US Geological Surveys.
http://minerals.usgs.gov/minerals/pubs/commodity/
germanium/220798.pdf. Retrieved 2008-09-09.
[74] “Germanium”. American Cancer Society. Retrieved
2008-08-31.
[75] Slavik, Milan; Blanc, Oscar; Davis, Joan (1983).
“Spirogermanium: A new investigational drug of
novel structure and lack of bone marrow toxic-
ity”. Investigational New Drugs 1 (3): 225–234.
doi:10.1007/BF00208894. PMID 6678870.
[76] Baselt, R. (2008). Disposition of Toxic Drugs and Chem-
icals in Man (8 ed.). Foster City, CA: Biomedical Publi-
cations. pp. 693–694.

[67] Thiele, Ulrich K. (2001). “The Current Status of Catal- 2.2.9 External links
ysis and Catalyst Development for the Industrial Process
of Poly(ethylene terephthalate) Polycondensation”. Inter- • Germanium at The Periodic Table of Videos (Uni-
national Journal of Polymeric Materials 50 (3): 387–394. versity of Nottingham)
doi:10.1080/00914030108035115.

[68] Fang, Li; Kulkarni, Sameer; Alhooshani, Khalid; Ma-

lik, Abdul (2007). “Germania-Based, Sol-Gel Hybrid 2.3 Gallium arsenide
Organic-Inorganic Coatings for Capillary Microextrac-
tion and Gas Chromatography”. Anal. Chem. 79 (24): Gallium arsenide (GaAs) is a compound of the ele-
9441–9451. doi:10.1021/ac071056f. PMID 17994707. ments gallium and arsenic. It is a III-V direct bandgap
[69] Keyser, Ronald; Twomey, Timothy; Upp, Daniel. semiconductor with a zinc blende crystal structure. Gal-
“Performance of Light-Weight, Battery-Operated, High lium arsenide is used in the manufacture of devices such
Purity Germanium Detectors for Field Use” (PDF). as microwave frequency integrated circuits, monolithic
Oak Ridge Technical Enterprise Corporation (ORTEC). microwave integrated circuits, infrared light-emitting
Archived from the original on October 26, 2007. Re- diodes, laser diodes, solar cells and optical windows.[2]
trieved 2008-09-06.
GaAs is often used as a substrate material for the epitaxial
[70] Ahmed, F. U.; Yunus, S.M.; Kamal, I.; Begum, growth of other III-V semiconductors including: Indium
S.; Khan, Aysha A.; Ahsan, M.H.; Ahmad, A.A.Z. gallium arsenide, aluminum gallium arsenide and others.
2.3. GALLIUM ARSENIDE 55

2.3.1 Preparation and chemistry Oxidation of GaAs occurs in air and degrades perfor-
mance of the semiconductor. The surface can be passi-
In the compound, gallium has a +3 oxidation state. Gal- vated by depositing a cubic gallium(II) sulfide layer using
lium arsenide single crystals can be prepared by three in- a tert-butyl gallium sulfide compound such as (t
dustrial processes:[2] BuGaS)
7.[5]
• The vertical gradient freeze (VGF) process. Most
GaAs wafers are produced using this process.[3] Semi-insulating crystals
• Crystal growth using a horizontal zone furnace in the
If a GaAs boule is grown with excess arsenic present, it
Bridgman-Stockbarger technique, in which gallium
gets certain defects, in particular arsenic antisite defects
and arsenic vapors react, and free molecules deposit
(an arsenic atom at a gallium atom site within the crystal
on a seed crystal at the cooler end of the furnace.
lattice). The electronic properties of these defects (inter-
• Liquid encapsulated Czochralski (LEC) growth is acting with others) cause the Fermi level to be pinned to
used for producing high-purity single crystals that near the center of the bandgap, so that this GaAs crystal
can exhibit semi-insulating characteristics (see be- has very low concentration of electrons and holes. This
low). low carrier concentration is similar to an intrinsic (per-
fectly undoped) crystal, but much easier to achieve in
practice. These crystals are called “semi-insulating”, re-
Alternative methods for producing films of GaAs flecting their high resistivity of 107 –109 Ω·cm (which is
include:[2][4] quite high for a semiconductor, but still much lower than
a true insulator like glass).[6]
• VPE reaction of gaseous gallium metal and arsenic
trichloride:
Etching

2 Ga + 2 AsCl Wet etching of GaAs industrially uses an oxidizing agent

3 → 2 GaAs + 3 Cl such as hydrogen peroxide or bromine water,[7] and the
2 same strategy has been described in a patent relating to
processing scrap components containing GaAs where the
• MOCVD reaction of trimethylgallium and arsine: Ga3+
is complexed with a hydroxamic acid (“HA”), for
example:[8]
Ga(CH
3)
3 + AsH GaAs + H
3 → GaAs + 3 CH 2O
4 2 + “HA” → “GaA” complex + H
3AsO
4+4H
• Molecular beam epitaxy (MBE) of gallium and 2O
arsenic:
This reaction produces arsenic acid.
4 Ga + As
4 → 4 GaAs
2.3.2 Comparison with silicon
or GaAs advantages

2 Ga + As Some electronic properties of gallium arsenide are supe-

2 → 2 GaAs rior to those of silicon. It has a higher saturated electron
56 CHAPTER 2. SEMICONDUCTOR MATERIALS

velocity and higher electron mobility, allowing gallium (MOVPE). Because GaAs and AlAs have almost the
arsenide transistors to function at frequencies in excess of same lattice constant, the layers have very little induced
250 GHz. Unlike silicon junctions, GaAs devices are rel- strain, which allows them to be grown almost arbitrarily
atively insensitive to heat owing to their wider bandgap. thick. This allows for extremely high performance high
Also, GaAs devices tend to have less noise than silicon electron mobility, HEMT transistors and other quantum
devices, especially at high frequencies. This is a result well devices.
of higher carrier mobilities and lower resistive device
parasitics. These properties recommend GaAs circuitry
in mobile phones, satellite communications, microwave
Silicon advantages
point-to-point links and higher frequency radar systems.
It is used in the manufacture of Gunn diodes for genera-
tion of microwaves. Silicon has three major advantages over GaAs for inte-
grated circuit manufacture. First, silicon is abundant and
Another advantage of GaAs is that it has a direct band cheap to process. Si is highly abundant in the Earth’s
gap, which means that it can be used to absorb and emit crust, in the form of silicate minerals. The economy of
light efficiently. Silicon has an indirect bandgap and so scale available to the silicon industry has also reduced the
is relatively poor at emitting light. Nonetheless, advances adoption of GaAs.
silicon LEDs and lasers may be possible.
In addition, a Si crystal has an extremely stable structure
As a wide direct band gap material with resulting resis- mechanically and it can be grown to very large diameter
tance to radiation damage, GaAs is an excellent material boules and can be processed with very high yields. It is
for space electronics and optical windows in high power also a decent thermal conductor, thus enabling very dense
applications. packing of transistors that need to get rid of their heat of
Because of its wide bandgap, pure GaAs is highly resis- operation, all very desirable for design and manufacturing
tive. Combined with the high dielectric constant, this of very large ICs. Such good mechanical characteristics
property makes GaAs a very good electrical substrate and also makes it a suitable material for the rapidly developing
unlike Si provides natural isolation between devices and field of nanoelectronics.
circuits. This has made it an ideal material for microwave The second major advantage of Si is the existence of a
and millimeter wave integrated circuits, MMICs, where native oxide (silicon dioxide, SiO ), which is used as an
2
active and essential passive components can readily be insulator in electronic devices. Silicon dioxide can eas-
produced on a single slice of GaAs. ily be incorporated onto silicon circuits, and such lay-
One of the first GaAs microprocessors was developed in ers are adherent to the underlying Si. SiO2 is not only
the early 1980s by the RCA corporation and was con- a good insulator (with a band gap of 8.9 eV), but the Si-
sidered for the Star Wars program of the United States SiO2 interface can be easily engineered to have excellent
Department of Defense. Those processors were sev- electrical properties, most importantly low density of in-
eral times faster and several orders of magnitude more terface states. GaAs does not have a native oxide and
radiation hard than silicon counterparts, but they were does not easily support a stable adherent insulating layer.
rather expensive.[9] Other GaAs processors were im- Aluminum oxide (Al2 O3 ) has been extensively studied
plemented by the supercomputer vendors Cray Com- as a possible gate oxide for GaAs (and InGaAs). How-
puter Corporation, Convex, and Alliant in an attempt to ever, at this point the electrical properties of the inter-
stay ahead of the ever-improving CMOS microproces- faces aren't comparable to those of the Si-SiO2 interface.
sor. Cray eventually built one GaAs-based machine in The third, advantage of silicon is that it possesses a
the early 1990s, the Cray-3, but the effort was not ade- higher hole mobility compared to GaAs (500 versus 400
quately capitalized, and the company filed for bankruptcy cm2 V−1 s−1 ).[10] This high mobility allows the fabrica-
in 1995. tion of higher-speed P-channel field effect transistors,
Complex layered structures of gallium arsenide in com- which are required for CMOS logic. Because they lack a
bination with aluminium arsenide (AlAs) or the alloy fast CMOS structure, GaAs circuits must use logic styles
AlₓGa₁-ₓAs can be grown using molecular beam epi- which have much higher power consumption; this has
taxy (MBE) or using metalorganic vapor phase epitaxy made GaAs circuits less able to compete with silicon logic
circuits.
2.3. GALLIUM ARSENIDE 57

For manufacturing solar cells, silicon has relatively low
absorptivity for the sunlight meaning about 100 microm-
eters of Si is needed to absorb most sunlight. Such a layer
is relatively robust and easy to handle. In contrast, the ab-
sorptivity of GaAs is so high that only a few micrometers
of thickness are needed to absorb all of the light. Conse-
quently GaAs thin films must be supported on a substrate
material.[11]
Silicon is a pure element, avoiding the problems of stoi-
chiometric imbalance and thermal unmixing of GaAs.
Silicon has a nearly perfect lattice, impurity density is
very low and allows very small structures to be built (cur-
rently down to 16 nm[12] ). GaAs in contrast has a very
high impurity density, which makes it difficult to build
integrated circuits with small structures, so the 500 nm
process is a common process for GaAs.
sensitive to infrared radiation (QWIP).
GaAs diodes can be used for the detection of X-rays.[19]
Light-emission devices
GaAs has been used to produce (near-infrared) laser
diodes since 1962.[20]
2.3.4 Safety
The environment, health and safety aspects of gallium
arsenide sources (such as trimethylgallium and arsine)
and industrial hygiene monitoring studies of metalorganic
precursors have been reported.[21] California lists gallium
arsenide as a carcinogen.[22] However, there is no evi-
dence for a primary carcinogenic effect of GaAs.[23]

2.3.3 Other applications

2.3.5 See also
Solar cells and detectors
• Aluminium arsenide
Another important application of GaAs is for high ef-
ficiency solar cells. Gallium arsenide (GaAs) is also • Aluminium gallium arsenide
known as single-crystalline thin film and are high-cost • Arsine
high-efficiency solar cells.
In 1970, the first GaAs heterostructure solar cells were • Cadmium telluride
created by the team led by Zhores Alferov in the • Gallium antimonide
USSR.[13][14][15] In the early 1980s, the efficiency of the
best GaAs solar cells surpassed that of silicon solar cells, • Gallium arsenide phosphide
and in the 1990s GaAs solar cells took over from sili-
con as the cell type most commonly used for Photovoltaic • Gallium manganese arsenide
arrays for satellite applications. Later, dual- and triple-
• Gallium phosphide
junction solar cells based on GaAs with germanium and
indium gallium phosphide layers were developed as the • Gallium nitride
basis of a triple-junction solar cell, which held a record
efficiency of over 32% and can operate also with light as • Heterostructure emitter bipolar transistor
concentrated as 2,000 suns. This kind of solar cell pow-
ers the rovers Spirit and Opportunity, which are exploring • MESFET
Mars' surface. Also many solar cars utilize GaAs in solar • Indium arsenide
arrays.
GaAs-based devices hold the world record for the • Indium gallium arsenide
highest-efficiency single-junction solar cell at 28.8%.[16] • Indium phosphide
This high efficiency is attributed to the extreme high
quality GaAs epitaxial growth, surface passivation by the • Light-emitting diode
AlGaAs,[17] and the promotion of photon recycling by the
thin film design.[18] • Metal semiconductor field effect transistor

Complex designs of AlₓGa₁₋ₓAs-GaAs devices can be • MOVPE

58
• Multijunction
• Photomixing
• Trimethylgallium
2.3.6 References
[1] Refractive index of GaAs. Ioffe database
[2] Moss, S. J. and Ledwith, A. (1987). The Chemistry of the
Semiconductor Industry. Springer. ISBN 0-216-92005-1.
[3] Scheel, Hans J., and Tsuguo Fukuda. (2003). Crystal
Growth Technology. Wiley. ISBN 0471490598.
[4] Smart, Lesley and Moore, Elaine A. (2005). Solid State
Chemistry: An Introduction. CRC Press. ISBN 0-7487-
7516-1.
[5] “Chemical vapor deposition from single organometallic
precursors” A. R. Barron, M. B. Power, A. N. MacInnes,
A. F.Hepp, P. P. Jenkins U.S. Patent 5,300,320 (1994)
[6] Dopants and Defects in Semiconductors, by Matthew D.
McCluskey, Eugene E. Haller, pp. 41 and 66, ISBN 978-
1439831526
[7] Brozel, M. R. and Stillman, G. E. (1996). Properties of
Gallium Arsenide. IEEE Inspec. ISBN 0-85296-885-X.
[8] “Oxidative dissolution of gallium arsenide and separation
of gallium from arsenic” J. P. Coleman and B. F. Monzyk
U.S. Patent 4,759,917 (1988)
[9] Šilc, Von Jurij; Robič, Borut and Ungerer, Theo (1999).
Processor architecture: from dataflow to superscalar and
beyond. Springer. p. 34. ISBN 3-540-64798-8.
[10] Appendix G, Sze, S. M. (1985). Semiconductor Devices
Physics and Technology, John Wiley & Sons ISBN 0-471-
87424-8
[11] Single-Crystalline Thin Film. US Department of Energy
[12] Handy, Jim (17 July 2013) Micron NAND Reaches
16nm. thememoryguy.com
[13] Alferov, Zh. I., V. M. Andreev, M. B. Kagan, I. I. Pro-
tasov and V. G. Trofim, 1970, ‘‘Solar-energy convert-
ers based on p-n AlₓGa₁-ₓAs-GaAs heterojunctions,’’ Fiz.
Tekh. Poluprovodn. 4, 2378 (Sov. Phys. Semicond. 4,
2047 (1971))
[14] Nanotechnology in energy applications. im.isu.edu.tw. 16
November 2005 (in Chinese) p. 24
[15] Nobel Lecture by Zhores Alferov at nobelprize.org, p. 6
CHAPTER 2. SEMICONDUCTOR MATERIALS
[16] Yablonovitch, Eli; Miller, Owen D.; Kurtz, S. R.
(2012). “The opto-electronic physics that broke the
efficiency limit in solar cells”. 2012 38th IEEE
Photovoltaic Specialists Conference. p. 001556.
doi:10.1109/PVSC.2012.6317891. ISBN 978-1-4673-
0066-7.
[17] Schnitzer, I. et al. (1993). “Ultrahigh spontaneous
emission quantum efficiency, 99.7% internally and 72%
externally, from AlGaAs/GaAs/AlGaAs double het-
erostructures”. Applied Physics Letters 62 (2): 131.
doi:10.1063/1.109348.
[18] Wang, X. et al. (2013). “Design of GaAs So-
lar Cells Operating Close to the Shockley–Queisser
Limit”. IEEE Journal of Photovoltaics 3 (2): 737.
doi:10.1109/JPHOTOV.2013.2241594.
[19] Glasgow University report on CERN detector.
Ppewww.physics.gla.ac.uk. Retrieved on 2013-10-
16.
[20] Hall, Robert N.; Fenner, G. E.; Kingsley, J. D.; Soltys,
T. J. and Carlson, R. O. (1962). “Coherent Light
Emission From GaAs Junctions”. Physical Review Let-
ters 9 (9): 366–369. Bibcode:1962PhRvL...9..366H.
doi:10.1103/PhysRevLett.9.366.
[21] Shenai-Khatkhate, D V; Goyette, R; DiCarlo, R L and
Dripps, G (2004). “Environment, health and safety
issues for sources used in MOVPE growth of com-
pound semiconductors”. Journal of Crystal Growth
272 (1–4): 816–821. Bibcode:2004JCrGr.272..816S.
doi:10.1016/j.jcrysgro.2004.09.007.
[22] “Chemicals Listed Effective August 1, 2008 as Known
to the State of California to Cause Cancer or Reproduc-
tive Toxicity: gallium arsenide, hexafluoroacetone, ni-
trous oxide and vinyl cyclohexene dioxide”. OEHHA.
2008-08-01.
[23] Bomhard, E. M.; Gelbke, H.; Schenk, H.; Williams,
G. M.; Cohen, S. M. (2013). “Evaluation of
the carcinogenicity of gallium arsenide”. Crit-
ical Reviews in Toxicology 43 (5): 436–466.
doi:10.31.109/104084444.2013.792329/.
2.3.7 External links
• Case Studies in Environmental Medicine: Arsenic
Toxicity
• Physical properties of gallium arsenide (Ioffe Insti-
tute)
• Facts and figures on processing gallium arsenide
2.3. GALLIUM ARSENIDE 59

A polycrystalline silicon rod made by the Siemens process

PDMS – a silicone compound

Monocrystalline silicon ingot grown by the Czochralski process

60 CHAPTER 2. SEMICONDUCTOR MATERIALS

Germane is similar to methane.

Silicon wafer with mirror finish

Nucleophilic addition with an organogermanium compound.

Silica skeletons of radiolaria in false color.

Renierite
2.3. GALLIUM ARSENIDE 61

A PET bottle

A typical single-mode optical fiber. Germanium oxide is a

dopant of the core silica (Item 1).
1. Core 8 µm
2. Cladding 125 µm
3. Buffer 250 µm
4. Jacket 400 µm

A pristine 2-inch single crystal gallium arsenide wafer with a

(100) surface orientation. Purple features are a reflection of a
nitrile glove.
62 CHAPTER 2. SEMICONDUCTOR MATERIALS

GaAs Conduction band

3 T = 300 K

2
Energy (eV)

0.29 eV 0.38 eV

1
E g= 1.42 eV

-1
Valence band
L[111] Γ X[100]
Wave vector k
Band structure of GaAs. The direct gap of GaAs results in effi-
cient emission of infrared light at 1.424 eV (~870 nm).

High-efficiency, triple-junction gallium arsenide solar cells cov-

ering the MidSTAR-1 satellite
Chapter 3

Applications

3.1 Voltage-controlled oscillator
A voltage-controlled oscillator or VCO is an electronic
oscillator whose oscillation frequency is controlled by
a voltage input. The applied input voltage determines
the instantaneous oscillation frequency. Consequently,
modulating signals applied to control input may cause Relaxation oscillators can generate a sawtooth or tri-
frequency modulation (FM) or phase modulation (PM). angular waveform. They are commonly used in mono-
A VCO may also be part of a phase-locked loop.
A microwave (12-18 GHz) Voltage Controlled Oscillator
input controls the resonant frequency. A varactor diode’s
capacitance is controlled by the voltage across the diode.
Consequently, a varactor can be used to change the ca-
pacitance (and hence the frequency) of an LC tank. A
varactor can also change (“pull”) the resonant frequency
of a crystal resonator.
lithic integrated circuits (ICs). They can provide a wide
range of operational frequencies with a minimal number
of external components. Relaxation oscillator VCOs can
have three topologies: 1) grounded-capacitor VCOs, 2)
emitter-coupled VCOs, and 3) delay-based ring VCOs.
The first two of these types operate similarly. The time
spent in each state depends on the rate of charge or dis-
charge of a capacitor. The delay-based ring VCO op-
erates somewhat differently however. For this type, the
gain stages are connected in a ring. The output frequency
is then a function of the delay in each stage.
Harmonic oscillator VCOs have these advantages over re-
laxation oscillators.

3.1.1 Types of VCO • Frequency stability with respect to temperature,

VCOs can be generally categorized into two groups based
on the type of waveform produced: 1) harmonic oscilla-
tors, and 2) relaxation oscillators.
Linear or harmonic oscillators generate a sinusoidal
waveform. Harmonic oscillators in electronics usually
consist of a resonator with an amplifier that replaces the
resonator losses (to prevent the amplitude from decay-
ing) and isolates the resonator from the output (so the
load does not affect the resonator). Some examples of
harmonic oscillators are LC-tank oscillators and crystal
oscillators. In a voltage-controlled oscillator, a voltage
noise, and power supply is much better for harmonic
oscillator VCOs.
• They have good accuracy for frequency control since
the frequency is controlled by a crystal or tank cir-
cuit.
A disadvantage of harmonic oscillator VCOs is that they
cannot be easily implemented in monolithic ICs. Relax-
ation oscillator VCOs are better suited for this technol-
ogy. Relaxation VCOs are also tunable over a wider range
of frequencies.

63
64 CHAPTER 3. APPLICATIONS

3.1.2 Control of frequency in VCOs
• Where the oscillator drives equipment that may gen-
erate radio-frequency interference, adding a varying
voltage to its control input can disperse the interfer-
ence spectrum to make it less objectionable. See
spread spectrum clock.

Voltage-controlled oscillator schematic - audio

A voltage-controlled capacitor is one method of mak-

ing an LC oscillator vary its frequency in response to a
control voltage. Any reverse-biased semiconductor diode
displays a measure of voltage-dependent capacitance and
can be used to change the frequency of an oscillator by
varying a control voltage applied to the diode. Special-
purpose variable capacitance varactor diodes are avail-
able with well-characterized wide-ranging values of ca-
pacitance. Such devices are very convenient in the man-
ufacture of voltage-controlled oscillators[note 1] For low-
frequency VCOs, other methods of varying the frequency
(such as altering the charging rate of a capacitor by means
of a voltage controlled current source) are used. See
Function generator.
The frequency of a ring oscillator is controlled by vary- A 26 MHz TCVCXO.
ing either the supply voltage, the current available to each
inverter stage, or the capacitive loading on each stage. A temperature-compensated VCXO (TCVCXO) incor-
porates components that partially correct the depen-
dence on temperature of the resonant frequency of the
3.1.3 Voltage-controlled crystal oscillators crystal. A smaller range of voltage control then suf-
fices to stabilize the oscillator frequency in applications
A voltage-controlled crystal oscillator (VCXO) is used for where temperature varies, such as heat buildup inside a
fine adjustment of the operating frequency. The fre- transmitter.
quency of a voltage-controlled crystal oscillator can be
varied a few tens of parts per million (ppm), because the Placing the oscillator in a temperature-controlled “oven”
high Q factor of the crystals allows “pulling” over only a at a constant but higher-than-ambient temperature is an-
small range of frequencies. other way to stabilize oscillator frequency. High stability
crystal oscillator references often place the crystal in an
There are two reasons for using a VCXO: oven and use a voltage input for fine control.[1] The tem-
perature is selected to be the turnover temperature: the
• To adjust the output frequency to match (or perhaps temperature where small changes do not affect the res-
be some exact multiple of) an accurate external ref- onance. The control voltage can be used to occasionally
erence. adjust the reference frequency to a NIST source. Sophis-
3.1. VOLTAGE-CONTROLLED OSCILLATOR 65

ticated designs may also adjust the control voltage over VCO freq-domain equations
time to compensate for crystal aging.

Ftuning (s) = Ko · Vin (s)

VCO time-domain equations
Ftuning (s)
= Θout (s)
ftuning (t) = s
f∫quiescent + Ko · vin (t)
ftuning (t) dt = Analog applications such as frequency modulation and
θout (t) frequency-shift keying often need to control an oscillator
• fo is the quies- frequency with an input — a voltage-controlled oscillator
cent frequency (VCO). The functional relationship between the control
of the oscillator voltage and the output frequency may not be linear. Over
small ranges, the relationship is approximately linear, and
• Ko is called linear control theory can be used.
the oscillator
sensitivity, or There are devices called voltage-to-frequency converters
gain. Its units (VFC). These devices are often designed to be very linear
are hertz per over a wide range of input voltages.
volt.
• ftuning (t) 3.1.4 VCO design and circuits
is the sym-
bol for the Tuning range, tuning gain and phase noise are the impor-
time-domain tant characteristics of a VCO. Generally low phase noise
waveform is preferred in the VCO. The noise present in the control
that is the signal and the tuning gain affect the phase noise; high
VCO’s tunable noise or high tuning gain imply more phase noise. Other
frequency important elements that determine the phase noise are the
component. transistor’s flicker noise (1/f noise),[2] the output power
• θout (t) is the level, and the loaded Q of the resonator.[3] See Leeson’s
symbol for the equation. The low frequency flicker noise affects the
time-domain phase noise because the flicker noise is heterodyned to
waveform that the oscillator output frequency due to the active devices
is the VCO’s non-linear transfer function. The effect of flicker noise
output phase. can be reduced with negative feedback that linearizes the
transfer function (for example, emitter degeneration).
• vin (t) is the
time-domain Leeson’s expression[4] for single-sideband (SSB) phase
symbol of the noise in dBc/Hz (decibels relative to output level per
control (input) Hertz) is[5]
voltage of the
VCO; it is [ (( )2 )(
1 f0 fc
sometimes also L(fm ) = 10 log 2 2Ql fm +1 fm +
represented as )( )]
F kT
vtune (t) 1 Ps

where f 0 is the output frequency, Q is the

VCOs used at radio frequency have a more complex re- loaded Q, f is the offset from the output fre-
lationship than the simplistic one shown here. There can quency (Hz), f is the 1/f corner frequency,
be some non-linearity in the relationship. F is the noise factor of the amplifier, k is
66 CHAPTER 3. APPLICATIONS

Boltzmann’s constant, T is absolute tempera- range, linearity, and distortion are often most important
ture in Kelvins, and P is the oscillator output specs. Audio-frequency VCOs for use in musical con-
power. texts were largely superseded in the 1980s by their digital
counterparts, DCOs, due to their output stability in the
Commonly used VCO circuits are the Clapp and Colpitts face of temperature changes during operation. From the
oscillators. The more widely used oscillator of the two is 1990s on, pure software is the primary sound-generating
Colpitts and these oscillators are very similar in configu- method, but VCOs have become popular again often
ration. thanks to their imperfections.

VCOs generally have the lowest Q-factor of the used os-

cillators, and so suffer more jitter than the other types. 3.1.6 Voltage-controlled crystal oscillator
The jitter can be made low enough for many applications
as a clock generator
(such as driving an ASIC), in which case VCOs enjoy the
advantages of having no off-chip components (expensive)
A clock generator is an oscillator that provides a tim-
or on-chip inductors (low yields on generic CMOS pro-
ing signal to synchronize operations in digital circuits.
cesses). These oscillators also have larger tuning ranges
VCXO clock generators are used in many areas such as
than the other kinds, which improves yield and is some-
digital TV, modems, transmitters and computers. Design
times a feature of the end product (for instance, the dot
parameters for a VCXO clock generator are tuning volt-
clock on a graphics card which drives a wide range of
age range, center frequency, frequency tuning range and
monitors).
the timing jitter of the output signal. Jitter is a form of
phase noise that must be minimised in applications such
3.1.5 Applications as radio receivers, transmitters and measuring equipment.
The tuning range of a VCXO is typically a few parts per
VCOs are used in: million over a control voltage range of typically 0 to 3
volts. When a wider selection of clock frequencies is
• Function generators, needed the VCXO output can be passed through digi-
tal divider circuits to obtain lower frequency(ies) or be
• The production of electronic music, to generate fed to a PLL (Phase Locked Loop). ICs containing both
variable tones in synthesizers, a VCXO (for external crystal) and a PLL are available.
A typical application is to provide clock frequencies in a
• Phase-locked loops,
range from 12 kHz to 96 kHz to an audio digital to analog
• Frequency synthesizers used in communication converter.
equipment.

3.1.7 See also

Voltage-to-Frequency converters are voltage-controlled
oscillators, with a highly linear relation between applied
• VFO
voltage and frequency. They are used to convert a slow
analog signal (such as from a temperature transducer) • VCF
to a digital signal for transmission over a long distance,
since the frequency will not drift or be affected by noise. • VCA
VCOs may have sine and/or square wave outputs. Func-
tion generators are low-frequency oscillators which fea- • LFO
ture multiple waveforms, typically sine, square, and tri-
angle waves. Monolithic function generators are voltage- • modular synthesizer
controlled. Analog phase-locked loops typically contain
• Digitally controlled oscillator, DCO
VCOs. High-frequency VCOs are usually used in phase-
locked loops for radio receivers. Phase noise is the most • Numerically controlled oscillator, NCO
important specification for them. Low-frequency VCOs
are used in analog music synthesizers. For these, sweep • Phase-locked loop, PLL
3.2. FREQUENCY-SHIFT KEYING 67

3.1.8 Notes
[1] A voltage-controlled inductor would be in principle as
useful, but such devices are unsatisfactory at the frequen-
cies usually desired.

Data
3.1.9 References
[1] For example, an HP/Agilent 10811 reference oscillator

[2] Wideband VCO from Herley - General Microwave - “For

optimum performance, the active element used is a sil-
icon bipolar transistor. (This is in lieu of GaAs FETs Carrier
which typically exhibit 10-20 dB poorer phase noise per-
formance) extquotedbl

[3] Rhea, Randall W. (1997), Oscillator Design & Com-

puter Simulation (Second ed.), McGraw-Hill, ISBN 0-07-
052415-7

[4] Leeson, D. B. (February 1966), “A Simple Model of Modulated Signal

Feedback Oscillator Noise Spectrum”, Proceedings of the

IEEE 54 (2): 329–330, doi:10.1109/PROC.1966.4682 An example of binary FSK

[5] Rhea 1997, p. 115

FSK signal can be done using the Goertzel algorithm very

3.1.10 External links efficiently, even on low-power microcontrollers.[4]
• schematics

• Designing VCOs and Buffers Using the UPA family

of Dual Transistors
3.2.2 Other forms of FSK

3.2 Frequency-shift keying Minimum-shift keying

Frequency-shift keying (FSK) is a frequency modula- Main article: Minimum-shift keying

tion scheme in which digital information is transmitted
through discrete frequency changes of a carrier wave.[1]
The simplest FSK is binary FSK (BFSK). BFSK uses Minimum frequency-shift keying or minimum-shift key-
ing (MSK) is a particular spectrally efficient form of co-
a pair of discrete frequencies to transmit binary (0s and
herent FSK. In MSK, the difference between the higher
1s) information.[2] With this scheme, the “1” is called the
and lower frequency is identical to half the bit rate. Con-
mark frequency and the “0” is called the space frequency.
The time domain of an FSK modulated carrier is illus- sequently, the waveforms that represent a 0 and a 1 bit
trated in the figures to the right. differ by exactly half a carrier period. The maximum
frequency deviation is δ = 0.25 fm, where fm is the max-
imum modulating frequency. As a result, the modulation
3.2.1 Implementations of FSK Modems index m is 0.5. This is the smallest FSK modulation index
that can be chosen such that the waveforms for 0 and 1
Reference implementations of FSK modems exist and are are orthogonal. A variant of MSK called GMSK is used
documented in detail.[3] The demodulation of a binary in the GSM mobile phone standard.
68 CHAPTER 3. APPLICATIONS

Audio FSK cassettes . AFSK is still widely used in amateur radio, as

it allows data transmission through unmodified voiceband
Audio frequency-shift keying (AFSK) is a modulation equipment. Radio control gear uses FSK, but calls it FM
technique by which digital data is represented by changes and PPM instead.
in the frequency (pitch) of an audio tone, yielding an AFSK is also used in the United States' Emergency Alert
encoded signal suitable for transmission via radio or System to transmit warning information . It is used at
telephone. Normally, the transmitted audio alternates be- higher bitrates for Weathercopy used on Weatheradio by
tween two tones: one, the “mark”, represents a binary NOAA in the U.S.
one; the other, the “space”, represents a binary zero.
The CHU shortwave radio station in Ottawa, Canada
AFSK differs from regular frequency-shift keying in per- broadcasts an exclusive digital time signal encoded using
forming the modulation at baseband frequencies. In radio AFSK modulation.
applications, the AFSK-modulated signal normally is be-
ing used to modulate an RF carrier (using a conventional FSK is commonly used in Caller ID and remote metering
technique, such as AM or FM) for transmission. applications: see FSK standards for use in Caller ID and
remote metering for more details
AFSK is not always used for high-speed data commu-
nications, since it is far less efficient in both power and
bandwidth than most other modulation modes. In addi- 3.2.4 See also
tion to its simplicity, however, AFSK has the advantage
that encoded signals will pass through AC-coupled links, • Amplitude-shift keying (ASK)
including most equipment originally designed to carry
music or speech. • Continuous-phase frequency-shift keying (CPFSK)
AFSK is used in the U.S. based Emergency Alert System • Dual-tone multi-frequency (DTMF), another en-
to notify stations of the type of emergency, locations af- coding technique representing data by pairs of audio
fected, and the time of issue without actually hearing the frequencies
text of the alert.
• Frequency-change signaling

3.2.3 Applications • Multiple frequency-shift keying (MFSK)

• Orthogonal frequency division multiplexing

In 1910, Reginald Fessenden invented a two-tone method (OFDM)
of transmitting Morse code. Dots and dashes were differ-
ent tones of equal length.[5] The intent was to minimize • Phase-shift keying (PSK)
transmission time.
• Federal Standard 1037C
Some early CW transmitters employed an arc converter
that could not be conveniently keyed. Instead of turning • MIL-STD-188
the arc on and off, the key slightly changed the transmit-
ter frequency in a technique known as the compensation-
wave method.[6] The compensation-wave was not used at 3.2.5 References
the receiver. The method consumed a lot of bandwidth
and caused interference, so it was discouraged by 1921.[7] [1] Kennedy, G.; Davis, B. (1992). Electronic Communica-
tion Systems (4th ed.). McGraw-Hill International. ISBN
Most early telephone-line modems used audio frequency- 0-07-112672-4., p 509
shift keying (AFSK) to send and receive data at rates up
to about 1200 bits per second. The common Bell 103 [2] FSK: Signals and Demodulation (B. Watson)
and Bell 202 modems used this technique.[8] Even to- http://www.xn--sten-cpa.se/share/text/tektext/
digital-modulation/FSK_signals_demod.pdf
day, North American caller ID uses 1200 baud AFSK
in the form of the Bell 202 standard. Some early [3] Teaching DSP through the Practical Case Study of
microcomputers used a specific form of AFSK modu- an FSK Modem (TI) http://www.ti.com/lit/an/spra347/
lation, the Kansas City standard, to store data on audio spra347.pdf
3.3. AMPLIFIER 69

[4] FSK Modulation and Demodulation With the MSP430
Microcontroller (TI) http://www.ti.com/lit/an/slaa037/
slaa037.pdf
[5] Morse 1925, p. 44; Morse cites British patent 2,617/11.
[6] Bureau of Standards 1922, pp. 415–416
[7] Little 1921, p. 125
[8] Kennedy & Davis 1992, pp. 549–550
shape but with a larger amplitude. In this sense, an am-
plifier modulates the output of the power supply to make
the output signal stronger than the input signal.
The four basic types of electronic amplifiers are voltage
amplifiers, current amplifiers, transconductance ampli-
fiers, and transresistance amplifiers. A further distinction
is whether the output is a linear or nonlinear representa-
tion of the input. Amplifiers can also be categorized by
their physical placement in the signal chain.[1]

• Bureau of Standards (1922), The Principles Un- 3.3.1 Figures of merit

derlying Radio Communication (Second ed.), U.S.
Army Signal Corps, Radio Communications Pam- Main article: Amplifier figures of merit
phlet No. 40. Revised to April 24, 1921.
Amplifier quality is characterized by a list of specifica-
• Little, D. G. (April 1921), “Continuous Wave Radio
tions that includes:
Communication”, Electric Journal 18: 124–129

• Morse, A. H. (1925), Radio: Beam and Broadcast, • Gain, the ratio between the magnitude of output and
London: Ernest Benn Limited input signals

• Bandwidth, the width of the useful frequency range

3.2.6 External links • Efficiency, the ratio between the power of the out-
put and total power consumption
• dFSK: Distributed Frequency Shift Keying Modu-
lation in Dense Sensor Networks • Linearity, the degree of proportionality between
input and output

• Noise, a measure of undesired noise mixed into the

3.3 Amplifier output

• Output dynamic range, the ratio of the largest and

For other uses, see Amplifier (disambiguation).
the smallest useful output levels
An electronic amplifier, amplifier, or (informally)
• Slew rate, the maximum rate of change of the out-
put
R1 R3 R4 +V supply

Q3
• Rise time, settling time, ringing and overshoot
Q4
that characterize the step response
Input
D1
Output
C1 Q1 Q2 D2
• Stability, the ability to avoid self-oscillation
Q5
R7 R8 C2

R2 R5
R6
3.3.2 Amplifier types
0V (ground)

Amplifiers are described according to their input and out-

A practical bipolar transistor amplifier circuit put properties.[2] They exhibit the property of gain, or
multiplication factor that relates the magnitude of the
amp is an electronic device that increases the power of output signal to the input signal. The gain may be speci-
a signal. It does this by taking energy from a power sup- fied as the ratio of output voltage to input voltage (voltage
ply and controlling the output to match the input signal gain), output power to input power (power gain), or some
70 CHAPTER 3. APPLICATIONS

combination of current, voltage, and power. In many and other instrument amplifiers. The essential compo-
cases, with input and output in the same unit, gain is unit- nents include active devices, such as vacuum tubes or
less (though often expressed in decibels (dB)). transistors. A brief introduction to the many types of
The four basic types of amplifiers are as follows:[1] electronic amplifiers follows.

• Voltage amplifier – This is the most common type of Power amplifier

amplifier. An input voltage is amplified to a larger
output voltage. The amplifier’s input impedance is The term power amplifier is a relative term with respect
high and the output impedance is low. to the amount of power delivered to the load and/or pro-
vided by the power supply circuit. In general the power
• Current amplifier – This amplifier changes an input amplifier is the last 'amplifier' or actual circuit in a sig-
current to a larger output current. The amplifier’s nal chain (the output stage) and is the amplifier stage that
input impedance is low and the output impedance is requires attention to power efficiency. Efficiency con-
high. siderations lead to the various classes of power amplifier
based on the biasing of the output transistors or tubes:
• Transconductance amplifier – This amplifier re-
see power amplifier classes.
sponds to a changing input voltage by delivering a
related changing output current.
Power amplifiers by application
• Transresistance amplifier – This amplifier responds
to a changing input current by delivering a related
• Audio power amplifiers
changing output voltage. Other names for the device
are transimpedance amplifier and current-to-voltage • RF power amplifier, such as for transmitter final
converter. stages (see also: Linear amplifier).

• Servo motor controllers, where linearity is not im-

In practice the power gain of an amplifier will depend on
portant.
the source and load impedances used as well as the in-
herent voltage/current gain; while an RF amplifier may • Piezoelectric audio amplifier includes a DC-to-
have its impedances optimized for power transfer, au- DC converter to generate the high voltage output re-
dio and instrumentation amplifiers are normally designed quired to drive piezoelectric speakers.[3]
with their input and output impedances optimized for
least loading and highest signal integrity. An amplifier
that is said to have a gain of 20 dB might have a voltage Power amplifier circuits Power amplifier circuits in-
gain of ten times and an available power gain of much clude the following types:
more than 20 dB (power ratio of 100), yet actually be
delivering a much lower power gain if, for example, the • Vacuum tube/valve, hybrid or transistor power am-
input is from a 600 ohm microphone and the output is plifiers
connected to a 47 kilohm input socket for a power am-
plifier. • Push-pull output or single-ended output stages
In most cases an amplifier will be linear; that is, the gain is
constant for any normal level of input and output signal. Vacuum-tube (valve) amplifiers
If the gain is not linear, e.g., clipping of the signal, the
output signal will be distorted. There are however cases Main article: Valve amplifier
where variable gain is useful. Exponential gain amplifiers
are used in certain signal processing applications.[1] According to Symons, while semiconductor amplifiers
There are many differing types of electronic amplifiers have largely displaced valve amplifiers for low power ap-
used in areas such as: radio and television transmitters plications, valve amplifiers are much more cost effec-
and receivers, high-fidelity (“hi-fi”) stereo equipment, tive in high power applications such as “radar, counter-
microcomputers and other digital equipment, and guitar measures equipment, or communications equipment” (p.
3.3. AMPLIFIER
An ECC83 tube glowing inside a preamp
56). Many microwave amplifiers are specially designed
valves, such as the klystron, gyrotron, traveling wave tube,
and crossed-field amplifier, and these microwave valves
provide much greater single-device power output at mi-
crowave frequencies than solid-state devices (p. 59).[4]
Valves/tube amplifiers also have niche uses in other areas,
such as
• electric guitar amplification
• in Russian military aircraft, for their EMP tolerance
A fully differential amplifier is a solid state integrated cir-
• niche audio for their sound qualities (recording, and cuit amplifier that uses external feedback to control its
71
audiophile equipment)
Transistor amplifiers
See also: Transistor, Bipolar junction transistor, Field-
effect transistor, JFET and MOSFET
The essential role of this active element is to magnify an
input signal to yield a significantly larger output signal.
The amount of magnification (the “forward gain”) is de-
termined by the external circuit design as well as the ac-
tive device.
Many common active devices in transistor amplifiers are
bipolar junction transistors (BJTs) and metal oxide semi-
conductor field-effect transistors (MOSFETs).
Applications are numerous, some common examples are
audio amplifiers in a home stereo or PA system, RF high
power generation for semiconductor equipment, to RF
and Microwave applications such as radio transmitters.
Transistor-based amplifier can be realized using various
configurations: for example with a bipolar junction tran-
sistor we can realize common base, common collector
or common emitter amplifier; using a MOSFET we can
realize common gate, common source or common drain
amplifier. Each configuration has different characteristic
(gain, impedance...).
Operational amplifiers (op-amps)
Main articles: Operational amplifier and Instrumentation
amplifier
An operational amplifier is an amplifier circuit with very
high open loop gain and differential inputs that employs
external feedback to control its transfer function, or gain.
Though the term today commonly applies to integrated
circuits, the original operational amplifier design used
valves.
Fully differential amplifiers
Main article: Fully differential amplifier
72
An LM741 general purpose op-amp
transfer function or gain. It is similar to the operational tunnel diode amplifier.
amplifier, but also has differential output pins. These are
usually constructed using BJTs or FETs.
Video amplifiers
These deal with video signals and have varying band-
widths depending on whether the video signal is for
SDTV, EDTV, HDTV 720p or 1080i/p etc.. The speci-
fication of the bandwidth itself depends on what kind of
filter is used—and at which point (−1 dB or −3 dB for
example) the bandwidth is measured. Certain require-
ments for step response and overshoot are necessary for
an acceptable TV image.
Oscilloscope vertical amplifiers
These deal with video signals that drive an oscilloscope
display tube, and can have bandwidths of about 500 MHz.
The specifications on step response, rise time, overshoot,
and aberrations can make designing these amplifiers dif-
ficult. One of the pioneers in high bandwidth vertical am-
plifiers was the Tektronix company.
Distributed amplifiers
Main article: Distributed Amplifier
CHAPTER 3. APPLICATIONS
These use transmission lines to temporally split the sig-
nal and amplify each portion separately to achieve higher
bandwidth than possible from a single amplifier. The out-
puts of each stage are combined in the output transmis-
sion line. This type of amplifier was commonly used on
oscilloscopes as the final vertical amplifier. The trans-
mission lines were often housed inside the display tube
glass envelope.
Switched mode amplifiers
These nonlinear amplifiers have much higher efficiencies
than linear amps, and are used where the power saving
justifies the extra complexity.
Negative resistance devices
Negative resistances can be used as amplifiers, such as the
Microwave amplifiers
Travelling wave tube amplifiers Main article:
Traveling wave tube
Traveling wave tube amplifiers (TWTAs) are used for
high power amplification at low microwave frequencies.
They typically can amplify across a broad spectrum of
frequencies; however, they are usually not as tunable as
klystrons.
Klystrons Main article: Klystron
Klystrons are specialized linear-beam vacuum-devices,
designed to provide high power, widely tunable amplifi-
cation of millimetre and sub-millimetre waves. Klystrons
are designed for large scale operations and despite having
a narrower bandwidth than TWTAs, they have the advan-
tage of coherently amplifying a reference signal so its out-
put may be precisely controlled in amplitude, frequency
and phase.
Musical instrument amplifiers
Main article: Instrument amplifier
3.3. AMPLIFIER 73

An audio power amplifier is usually used to amplify sig-

nals such as music or speech. Several factors are espe-
cially important in the selection of musical instrument
amplifiers (such as guitar amplifiers) and other audio am-
plifiers (although the whole of the sound system – compo-
nents such as microphones to loudspeakers – affect these
parameters):

• Frequency response – not just the frequency range

but the requirement that the signal level varies so
little across the audible frequency range that the hu-
man ear notices no variation. A typical specification The four types of dependent source—control variable on left,
for audio amplifiers may be 20 Hz to 20 kHz +/- 0.5 output variable on right
dB.
• Power output – the power level obtainable with little Each type of amplifier in its ideal form has an ideal in-
distortion, to obtain a sufficiently loud sound pres- put and output resistance that is the same as that of the
sure level from the loudspeakers. corresponding dependent source:[5]
• Low distortion – all amplifiers and transducers dis- In practice the ideal impedances are only approximated.
tort to some extent. They cannot be perfectly lin- For any particular circuit, a small-signal analysis is often
ear, but aim to pass signals without affecting the used to find the impedance actually achieved. A small-
harmonic content of the sound more than the human signal AC test current Ix is applied to the input or output
ear can tolerate. That tolerance of distortion, and node, all external sources are set to AC zero, and the cor-
indeed the possibility that some “warmth” or sec- responding alternating voltage Vx across the test current
ond harmonic distortion (Tube sound) improves the source determines the impedance seen at that node as R
“musicality” of the sound, are subjects of great de- = Vx / Ix.
bate. Amplifiers designed to attach to a transmission line at
input and/or output, especially RF amplifiers, do not fit
into this classification approach. Rather than dealing with
3.3.3 Classification of amplifier stages and voltage or current individually, they ideally couple with
systems an input and/or output impedance matched to the trans-
mission line impedance, that is, match ratios of voltage
Many alternative classifications address different aspects to current. Many real RF amplifiers come close to this
of amplifier designs, and they all express some particular ideal. Although, for a given appropriate source and load
perspective relating the design parameters to the objec- impedance, RF amplifiers can be characterized as ampli-
tives of the circuit. Amplifier design is always a compro- fying voltage or current, they fundamentally are amplify-
mise of numerous factors, such as cost, power consump- ing power.[6]
tion, real-world device imperfections, and a multitude of
performance specifications. Below are several different
approaches to classification:
Common terminal

Input and output variables
Electronic amplifiers use one variable presented as either
a current and voltage. Either current or voltage can be
used as input and either as output, leading to four types
of amplifiers. In idealized form they are represented by
each of the four types of dependent source used in linear
analysis, as shown in the figure, namely:
One set of classifications for amplifiers is based on which
device terminal is common to both the input and the out-
put circuit. In the case of bipolar junction transistors,
the three classes are common emitter, common base, and
common collector. For field-effect transistors, the cor-
responding configurations are common source, common
gate, and common drain; for triode vacuum devices, com-
mon cathode, common grid, and common plate. The com-
74
mon emitter (or common source, or common cathode
etc.) is most often configured to provide amplification
of a voltage applied between base and emitter, and the
output signal taken between collector and emitter will be
inverted, relative to the input. The common collector
arrangement applies the input voltage between base and
collector, and to take the output voltage between emit-
ter and collector. This results in negative feedback, and
the output voltage will tend to 'follow' the input voltage
(this arrangement is also used as the input presents a high
impedance and does not load the signal source, although
the voltage amplification will be less than 1 (unity)); the
common-collector circuit is therefore better known as an
emitter follower, source follower, or cathode follower.
Unilateral or bilateral
Other amplifiers may be classified by their function or
output characteristics. These functional descriptions usu-
When an amplifier has an output that exhibits no feed- ally apply to complete amplifier systems or sub-systems
back to its input side, it is called 'unilateral'. The input and rarely to individual stages.
impedance of a unilateral amplifier is independent of the
load, and the output impedance is independent of the sig-
nal source impedance.
If feedback connects part of the output back to the input
of the amplifier it is called a 'bilateral' amplifier. The in-
put impedance of a bilateral amplifier is dependent upon
the load, and the output impedance is dependent upon the
signal source impedance.
All amplifiers are bilateral to some degree; however they
may often be modeled as unilateral under operating con-
ditions where feedback is small enough to neglect for
most purposes, simplifying analysis (see the common
base article for an example).
Negative feedback is often applied deliberately to tailor
amplifier behavior. Some feedback, which may be posi-
tive or negative, is unavoidable and often undesirable, in-
troduced, for example, by parasitic elements such as the
inherent capacitance between input and output of a de-
vice such as a transistor and capacitative coupling due to
external wiring. Excessive frequency-dependent positive
feedback may cause what is intended/expected to be an
amplifier to become an oscillator.
Linear unilateral and bilateral amplifiers can be repre-
sented as two-port networks.
CHAPTER 3. APPLICATIONS
Inverting or non-inverting
Another way to classify amplifiers is by the phase rela-
tionship of the input signal to the output signal. An 'in-
verting' amplifier produces an output 180 degrees out of
phase with the input signal (that is, a polarity inversion
or mirror image of the input as seen on an oscilloscope).
A 'non-inverting' amplifier maintains the phase of the in-
put signal waveforms. An emitter follower is a type of
non-inverting amplifier, indicating that the signal at the
emitter of a transistor is following (that is, matching with
unity gain but perhaps an offset) the input signal. Volt-
age follower is also non inverting type of amplifier having
unity gain.
This description can apply to a single stage of an ampli-
fier, or to a complete amplifier system.
Function
• A servo amplifier indicates an integrated feedback
loop to actively control the output at some desired
level. A DC servo indicates use at frequencies down
to DC levels, where the rapid fluctuations of an audio
or RF signal do not occur. These are often used in
mechanical actuators, or devices such as DC motors
that must maintain a constant speed or torque. An
AC servo amp can do this for some ac motors.
• A linear amplifier responds to different frequency
components independently, and does not generate
harmonic distortion or Intermodulation distortion.
No amplifier can provide perfect linearity (even
the most linear amplifier has some nonlinearities,
since the amplifying devices—transistors or vacuum
tubes—follow nonlinear power laws such as square-
laws and rely on circuitry techniques to reduce those
effects).
• A nonlinear amplifier generates significant distor-
tion and so changes the harmonic content; there are
situations where this is useful. Amplifier circuits in-
tentionally providing a non-linear transfer function
include:
3.3. AMPLIFIER
--- a device like a Silicon Controlled Rectifier or
a transistor used as a switch may be employed
to turn either fully ON or OFF a load such as
a lamp based on a threshold in a continuously
variable input.
--- a non-linear amplifier in an analog computer
or true RMS converter for example can pro-
vide a special transfer function, such as loga-
rithmic or square-law.
--- a Class C RF amplifier may be chosen be-
cause it can be very efficient, but will be
non-linear; following such an amplifier with a
extquotedbltank extquotedbl tuned circuit can
reduce unwanted harmonics (distortion) suf-
ficiently to be useful in transmitters, or some
desired harmonic may be selected by setting
the resonant frequency of the tuned circuit to a
higher frequency rather than fundamental fre-
quency in frequency multiplier circuits.
--- Automatic gain control circuits require an
amplifier’s gain be controlled by the time-
averaged amplitude so that the output ampli-
tude varies little when weak stations are be-
ing received. The non-linearities are assumed
to be arranged so the relatively small signal
amplitude suffers from little distortion (cross-
channel interference or intermodulation) yet
is still modulated by the relatively large gain-
control DC voltage.
--- AM detector circuits that use amplification
such as Anode-bend detectors, Precision rec-
tifiers and Infinite impedance detectors (so ex-
cluding unamplified detectors such as Cat’s-
whisker detectors), as well as peak detector
circuits, rely on changes in amplification based
on the signal's instantaneous amplitude to de-
rive a direct current from an alternating cur-
rent input.
--- Operational amplifier comparator and detec-
tor circuits.
• A wideband amplifier has a precise amplification
factor over a wide frequency range, and is often used
to boost signals for relay in communications sys-
tems. A narrowband amp amplifies a specific nar-
row range of frequencies, to the exclusion of other
frequencies.
• An RF amplifier amplifies signals in the radio fre-
quency range of the electromagnetic spectrum, and
75
is often used to increase the sensitivity of a receiver
or the output power of a transmitter.[7]
• An audio amplifier amplifies audio frequencies.
This category subdivides into small signal amplifica-
tion, and power amps that are optimised to driving
speakers, sometimes with multiple amps grouped
together as separate or bridgeable channels to ac-
commodate different audio reproduction require-
ments. Frequently used terms within audio ampli-
fiers include:
--- Preamplifier (preamp), which may include a
phono preamp with RIAA equalization, or
tape head preamps with CCIR equalisation fil-
ters. They may include filters or tone control
circuitry.
--- Power amplifier (normally drives
loudspeakers), headphone amplifiers, and
public address amplifiers.
--- Stereo amplifiers imply two channels of out-
put (left and right), though the term sim-
ply means “solid” sound (referring to three-
dimensional)—so quadraphonic stereo was
used for amplifiers with four channels. 5.1
and 7.1 systems refer to Home theatre systems
with 5 or 7 normal spacial channels, plus a
subwoofer channel.
• Buffer amplifiers, which may include emitter follow-
ers, provide a high impedance input for a device
(perhaps another amplifier, or perhaps an energy-
hungry load such as lights) that would otherwise
draw too much current from the source. Line drivers
are a type of buffer that feeds long or interference-
prone interconnect cables, possibly with differential
outputs through twisted pair cables.
• A special type of amplifier - originally used in analog
computers - is widely used in measuring instruments
for signal processing, and many other uses. These
are called operational amplifiers or op-amps. The
“operational” name is because this type of amplifier
can be used in circuits that perform mathematical
algorithmic functions, or “operations” on input sig-
nals to obtain specific types of output signals. Mod-
ern op-amps are usually provided as integrated cir-
cuits, rather than constructed from discrete compo-
nents. A typical modern op-amp has differential in-
puts (one “inverting”, one “non-inverting”) and one
76 CHAPTER 3. APPLICATIONS

output. An idealised op-amp has the following char- vacuum tube days when the anode (output) voltage
acteristics: was at greater than several hundred volts and the
grid (input) voltage at a few volts minus. So they
--- Infinite input impedance (so it does not load were only used if the gain was specified down to
the circuitry at its input) DC (e.g., in an oscilloscope). In the context of
--- Zero output impedance modern electronics developers are encouraged to
use directly coupled amplifiers whenever possible.
--- Infinite gain
--- Zero propagation delay
Frequency range
The performance of an op-amp with these characteristics
Depending on the frequency range and other properties
is entirely defined by the (usually passive) components
amplifiers are designed according to different principles.
that form a negative feedback loop around it. The ampli-
fier itself does not effect the output. All real-world op-
• Frequency ranges down to DC are only used when
amps fall short of the idealised specification above—but
this property is needed. DC amplification leads to
some modern components have remarkable performance
specific complications that are avoided if possible;
and come close in some respects.
DC-blocking capacitors are added to remove DC
and sub-sonic frequencies from audio amplifiers.
Interstage coupling method • Depending on the frequency range specified differ-
ent design principles must be used. Up to the MHz
See also: multistage amplifiers range only “discrete” properties need be considered;
e.g., a terminal has an input impedance.
Amplifiers are sometimes classified by the coupling
• As soon as any connection within the circuit gets
method of the signal at the input, output, or between
longer than perhaps 1% of the wavelength of the
stages. Different types of these include:
highest specified frequency (e.g., at 100 MHz the
wavelength is 3 m, so the critical connection length
Resistive-capacitive (RC) coupled amplifier, using a is approx. 3 cm) design properties radically change.
network of resistors and capacitors For example, a specified length and width of a
By design these amplifiers cannot amplify DC sig-
PCB trace can be used as a selective or impedance-
nals as the capacitors block the DC component of
matching entity.
the input signal. RC-coupled amplifiers were used
very often in circuits with vacuum tubes or discrete • Above a few hundred MHz, it gets difficult to use
transistors. In the days of the integrated circuit a discrete elements, especially inductors. In most
few more transistors on a chip are much cheaper cases, PCB traces of very closely defined shapes are
and smaller than a capacitor. used instead.
Inductive-capacitive (LC) coupled amplifier, using a
The frequency range handled by an amplifier might be
network of inductors and capacitors
This kind of amplifier is most often used in selective specified in terms of bandwidth (normally implying a re-
radio-frequency circuits. sponse that is 3 dB down when the frequency reaches the
specified bandwidth), or by specifying a frequency re-
Transformer coupled amplifier, using a transformer sponse that is within a certain number of decibels be-
to match impedances or to decouple parts of the cir- tween a lower and an upper frequency (e.g. “20 Hz to 20
cuits kHz plus or minus 1 dB”).
Quite often LC-coupled and transformer-coupled
amplifiers cannot be distinguished as a transformer
is some kind of inductor. 3.3.4 Power amplifier classes
Direct coupled amplifier, using no impedance and
bias matching components Power amplifier circuits (output stages) are classified as
This class of amplifier was very uncommon in the A, B, AB and C for analog designs, and class D and E for
3.3. AMPLIFIER 77

switching designs based on the proportion of each input
cycle (conduction angle), during which an amplifying de-
vice is passing current. The image of the conduction an-
gle is derived from amplifying a sinusoidal signal. If the
device is always on, the conducting angle is 360°. If it is
on for only half of each cycle, the angle is 180°. The angle
of flow is closely related to the amplifier power efficiency.
The various classes are introduced below, followed by a
more detailed discussion under their individual headings
further down.
In the illustrations below, a bipolar junction transistor is
shown as the amplifying device. However the same at- Class A
tributes are found with MOSFETs or vacuum tubes.
are mainly used for specialized applications, such
as very high-power units. Also, class-E and class-F
amplifiers are commonly described in literature for
radio-frequency applications where efficiency of the
traditional classes is important, yet several aspects
deviate substantially from their ideal values. These
classes use harmonic tuning of their output networks
to achieve higher efficiency and can be considered a
subset of class C due to their conduction-angle char-
acteristics.

Conduction angle classes

Class A 100% of the input signal is used (conduction

angle Θ = 360°). The active element remains
conducting[8] all of the time.

Class B 50% of the input signal is used (Θ = 180°); the

active element carries current half of each cycle, and
is turned off for the other half.

Class AB Class AB is intermediate between class A and

B, the two active elements conduct more than half Class-A amplifier
of the time
Amplifying devices operating in class A conduct over the
Class C Less than 50% of the input signal is used (con-
entire range of the input cycle. A class-A amplifier is dis-
duction angle Θ < 180°).
tinguished by the output stage devices being biased for
class A operation. Subclass A2 is sometimes used to refer
A “Class D” amplifier uses some form of pulse-width to vacuum-tube class-A stages where the grid is allowed
modulation to control the output devices; the conduction to be driven slightly positive on signal peaks, resulting in
angle of each device is no longer related directly to the slightly more power than normal class A (A1; where the
input signal but instead varies in pulse width. These are grid is always negative[9] ), but this incurs a higher distor-
sometimes called “digital” amplifiers because the output tion level.
device is switched fully on or off, and not carrying current
proportional to the signal amplitude.
Advantages of class-A amplifiers
Additional classes There are several other amplifier • Class-A designs are simpler than other classes; for
classes, although they are mainly variations of the example class -AB and -B designs require two con-
previous classes. For example, class-G and class- nected devices in the circuit (push–pull output),
H amplifiers are marked by variation of the sup- each to handle one half of the waveform; class A
ply rails (in discrete steps or in a continuous fash- can use a single device (single-ended).
ion, respectively) following the input signal. Wasted
heat on the output devices can be reduced as ex- • The amplifying element is biased so the device is
cess voltage is kept to a minimum. The amplifier always conducting, the quiescent (small-signal) col-
that is fed with these rails itself can be of any class. lector current (for transistors; drain current for FETs
These kinds of amplifiers are more complex, and or anode/plate current for vacuum tubes) is close
78 CHAPTER 3. APPLICATIONS

to the most linear portion of its transconductance
curve.
• Because the device is never 'off' there is no “turn
on” time, no problems with charge storage, and gen-
erally better high frequency performance and feed-
back loop stability (and usually fewer high-order
harmonics).
• The point at which the device comes closest to be-
ing 'off' is not at 'zero signal', so the problems of
crossover distortion associated with class-AB and -
B designs is avoided.
• Best for low signal levels of radio receivers due to
low distortion.
• Single-ended output stages have an asymmetrical
transfer function, meaning that even order harmon-
ics in the created distortion tend not to be canceled
(as they are in push–pull output stages); for tubes,
or FETs, most of the distortion is second-order har-
monics, from the square law transfer characteristic,
which to some produces a “warmer” and more pleas-
ant sound.[11][12]
• For those who prefer low distortion figures, the use
of tubes with class A (generating little odd-harmonic
distortion, as mentioned above) together with sym-
metrical circuits (such as push–pull output stages, or
balanced low-level stages) results in the cancellation
of most of the even distortion harmonics, hence the
removal of most of the distortion.

Disadvantage of class-A amplifiers • Historically, valve amplifiers often used a class-A

power amplifier simply because valves are large and
• Class-A amplifiers are inefficient. A theoretical ef- expensive; many class-A designs use only a single
ficiency of 50% is obtainable with transformer out- device.
put coupling and only 25% with capacitive coupling,
unless deliberate use of nonlinearities is made (such Transistors are much cheaper, and so more elaborate de-
as in square-law output stages). In a power am- signs that give greater efficiency but use more parts are
plifier, this not only wastes power and limits oper- still cost-effective. A classic application for a pair of
ation with batteries, but increases operating costs class-A devices is the long-tailed pair, which is excep-
and requires higher-rated output devices. Ineffi- tionally linear, and forms the basis of many more com-
ciency comes from the standing current that must plex circuits, including many audio amplifiers and almost
be roughly half the maximum output current, and all op-amps.
a large part of the power supply voltage is present Class-A amplifiers are often used in output stages of high
across the output device at low signal levels. If high
quality op-amps (although the accuracy of the bias in low
output power is needed from a class-A circuit, the cost op-amps such as the 741 may result in class A or class
power supply and accompanying heat becomes sig- AB or class B, varying from device to device or with tem-
nificant. For every watt delivered to the load, the perature). They are sometimes used as medium-power,
amplifier itself, at best, uses an extra watt. For high
low-efficiency, and high-cost audio power amplifiers. The
power amplifiers this means very large and expen- power consumption is unrelated to the output power. At
sive power supplies and heat sinks. idle (no input), the power consumption is essentially the
same as at high output volume. The result is low effi-
Class-A power amplifier designs have largely been su- ciency and high heat dissipation.
perseded by more efficient designs, though they remain
popular with some hobbyists, mostly for their simplic-
ity. There is a market for expensive high fidelity class- Class B
A amps considered a “cult item” amongst audiophiles[10]
mainly for their absence of crossover distortion and re- Class-B amplifiers only amplify half of the input wave cy-
duced odd-harmonic and high-order harmonic distortion. cle, thus creating a large amount of distortion, but their
efficiency is greatly improved and is much better than
class A. Class-B amplifiers are also favoured in battery-
Single-ended and triode class-A amplifiers Some operated devices, such as transistor radios. Class B has a
hobbyists who prefer class-A amplifiers also prefer the maximum theoretical efficiency of π/4. (≈ 78.5%) This
use of thermionic valve (or “tube”) designs instead of is because the amplifying element is switched off alto-
transistors, for several reasons: gether half of the time, and so cannot dissipate power. A
3.3. AMPLIFIER
Class-B amplifier
Class-B push–pull amplifier
single class-B element is rarely found in practice, though
it has been used for driving the loudspeaker in the early
IBM Personal Computers with beeps, and it can be used
in RF power amplifier where the distortion levels are less
important. However, class C is more commonly used for
this.
A practical circuit using class-B elements is the push–pull
stage, such as the very simplified complementary pair ar-
rangement shown below. Here, complementary or quasi-
complementary devices are each used for amplifying the
opposite halves of the input signal, which is then recom-
bined at the output. This arrangement gives excellent ef-
ficiency, but can suffer from the drawback that there is a
small mismatch in the cross-over region – at the “joins”
between the two halves of the signal, as one output de-
vice has to take over supplying power exactly as the other
finishes. This is called crossover distortion. An improve-
ment is to bias the devices so they are not completely off
when they're not in use. This approach is called class AB
operation.
79
Class AB
Class AB is widely considered a good compromise for
amplifiers, since much of the time the music signal is
quiet enough that the signal stays in the “class A” region,
where it is amplified with good fidelity, and by defini-
tion if passing out of this region, is large enough that the
distortion products typical of class B are relatively small.
The crossover distortion can be reduced further by using
negative feedback.
In class-AB operation, each device operates the same way
as in class B over half the waveform, but also conducts
a small amount on the other half. As a result, the re-
gion where both devices simultaneously are nearly off
(the “dead zone”) is reduced. The result is that when
the waveforms from the two devices are combined, the
crossover is greatly minimised or eliminated altogether.
The exact choice of quiescent current (the standing cur-
rent through both devices when there is no signal) makes
a large difference to the level of distortion (and to the
risk of thermal runaway, that may damage the devices);
often the bias voltage applied to set this quiescent cur-
rent has to be adjusted with the temperature of the output
transistors (for example in the circuit at the beginning of
the article the diodes would be mounted physically close
to the output transistors, and chosen to have a matched
temperature coefficient). Another approach (often used
as well as thermally tracking bias voltages) is to include
small value resistors in series with the emitters.
Class AB sacrifices some efficiency over class B in favor
of linearity, thus is less efficient (below 78.5% for full-
amplitude sinewaves in transistor amplifiers, typically;
much less is common in class-AB vacuum-tube ampli-
fiers). It is typically much more efficient than class A.
Sometimes a numeral is added for vacuum-tube stages.
If the grid voltage is always negative with respect to the
cathode the class is AB1 . If the grid is allowed to go
slightly positive (hence drawing grid current, adding more
distortion, but giving slightly higher output power) on sig-
nal peaks the class is AB2 .
Class C
Class-C amplifiers conduct less than 50% of the input
signal and the distortion at the output is high, but high
efficiencies (up to 90%) are possible. The usual applica-
tion for class-C amplifiers is in RF transmitters operating
at a single fixed carrier frequency, where the distortion
80 CHAPTER 3. APPLICATIONS

input signal. Power can be coupled to a load by trans-

Class-C amplifier
is controlled by a tuned load on the amplifier. The input
signal is used to switch the active device causing pulses
of current to flow through a tuned circuit forming part of
the load.
The class-C amplifier has two modes of operation: tuned
and untuned.[13] The diagram shows a waveform from
a simple class-C circuit without the tuned load. This is
called untuned operation, and the analysis of the wave-
forms shows the massive distortion that appears in the sig-
nal. When the proper load (e.g., an inductive-capacitive
filter plus a load resistor) is used, two things happen. The
first is that the output’s bias level is clamped with the av-
erage output voltage equal to the supply voltage. This is
why tuned operation is sometimes called a clamper. This
former action with a secondary coil wound on the induc-
tor. The average voltage at the drain is then equal to the
supply voltage, and the signal voltage appearing across
the tuned circuit varies from near zero to near twice the
supply voltage during the rf cycle. The input circuit is bi-
ased so that the active element (e.g. transistor) conducts
for only a fraction of the RF cycle, usually one third (120
degrees) or less.[14]
The active element conducts only while the drain volt-
age is passing through its minimum. By this means,
power dissipation in the active device is minimised, and
efficiency increased. Ideally, the active element would
pass only an instantaneous current pulse while the volt-
age across it is zero: it then dissipates no power and 100%
efficiency is achieved. However practical devices have a
limit to the peak current they can pass, and the pulse must
therefore be widened, to around 120 degrees, to obtain a
reasonable amount of power, and the efficiency is then
60-70%.[14]
Class D
Main article: Class D amplifier
In the class-D amplifier the active devices (transistors)
Input

allows the waveform to be restored to its proper shape

despite the amplifier having only a one-polarity supply.
C
This is directly related to the second phenomenon: the Low-pass filter

waveform on the center frequency becomes less distorted. Switching controller

and output stage
Triangular wave generator
The residual distortion is dependent upon the bandwidth
of the tuned load, with the center frequency seeing very Block diagram of a basic switching or PWM (class-D) amplifier.
little distortion, but greater attenuation the farther from
the tuned frequency that the signal gets. function as electronic switches instead of linear gain de-
The tuned circuit resonates at one frequency, the fixed vices; they are either on or off. The analog signal is con-
carrier frequency, and so the unwanted frequencies are verted to a stream of pulses that represents the signal by
suppressed, and the wanted full signal (sine wave) is ex- pulse width modulation, pulse density modulation, delta-
tracted by the tuned load. The signal bandwidth of the sigma modulation or a related modulation technique be-
amplifier is limited by the Q-factor of the tuned circuit fore being applied to the amplifier. The time average
but this is not a serious limitation. Any residual harmon- power value of the pulses is directly proportional to the
ics can be removed using a further filter. analog signal, so after amplification the signal can be con-
In practical class-C amplifiers a tuned load is invariably verted back to an analog signal by a passive low-pass fil-
used. In one common arrangement the resistor shown in ter.
the circuit above is replaced with a parallel-tuned circuit The purpose of the output filter is to smooth the pulse
consisting of an inductor and capacitor in parallel, whose stream to an analog signal, removing the high frequency
components are chosen to resonate the frequency of the spectral components of the pulses. The frequency of the
3.3. AMPLIFIER 81

lated signal. Switching power supplies have even been

modified into crude class-D amplifiers (although typically
these can only reproduce low-frequencies with an accept-
able level of accuracy).
High quality class-D audio power amplifiers have now ap-
peared on the market. These designs have been said to ri-
val traditional AB amplifiers in terms of quality. An early
use of class-D amplifiers was high-power subwoofer am-
plifiers in cars. Because subwoofers are generally limited
to a bandwidth of no higher than 150 Hz, the switching
speed for the amplifier does not have to be as high as for
a full range amplifier, allowing simpler designs. Class-D
amplifiers for driving subwoofers are relatively inexpen-
Boss Audio class-D mono amplifier with a low pass filter for sive in comparison to class-AB amplifiers.
powering subwoofers
The letter D used to designate this amplifier class is sim-
ply the next letter after C and, although occasionally used
as such, does not stand for digital. Class-D and class-E
output pulses is typically ten or more times the highest amplifiers are sometimes mistakenly described as “digi-
frequency in the input signal to be amplified, so that the tal” because the output waveform superficially resembles
filter can adequately reduce the unwanted harmonics, re- a pulse-train of digital symbols, but a class-D amplifier
producing an accurate reproduction of the input. merely converts an input waveform into a continuously
The main advantage of a class-D amplifier is power ef- pulse-width modulated analog signal. (A digital wave-
ficiency. Because the output pulses have a fixed am- form would be pulse-code modulated.)
plitude, the switching elements (usually MOSFETs, but
valves (vacuum tubes) and bipolar transistors were once Additional classes
used) are switched either completely on or completely off,
rather than operated in linear mode. A MOSFET oper- Class E The class-E/F amplifier is a highly efficient
ates with the lowest resistance when fully on and thus (ex- switching power amplifier, typically used at such high fre-
cluding when fully off) has the lowest power dissipation quencies that the switching time becomes comparable to
when in that condition. Compared to an equivalent class- the duty time. As said in the class-D amplifier, the tran-
AB device, a class-D amplifier’s lower losses permit the sistor is connected via a serial LC circuit to the load, and
use of a smaller heat sink for the MOSFETs while also re- connected via a large L (inductor) to the supply voltage.
ducing the amount of input power required, allowing for
The supply voltage is connected to ground via a large ca-
a lower-capacity power supply design. Therefore, class-D pacitor to prevent any RF signals leaking into the supply.
amplifiers are typically smaller than an equivalent class-
The class-E amplifier adds a C (capacitor) between the
AB amplifier. transistor and ground and uses a defined L1 to connect to
Another advantage of the class-D amplifier is that it can the supply voltage.
operate from a digital signal source without requiring a The following description ignores DC, which can be
digital-to-analog converter (DAC) to convert the signal added easily afterwards. The above-mentioned C and L
to analog form first. If the signal source is in digital form, are in effect a parallel LC circuit to ground. When the
such as in a digital media player or computer sound card, transistor is on, it pushes through the serial LC circuit into
the digital circuitry can convert the binary digital signal the load and some current begins to flow to the parallel
directly to a pulse width modulation signal to be applied LC circuit to ground. Then the serial LC circuit swings
to the amplifier, simplifying the circuitry considerably. back and compensates the current into the parallel LC
Class-D amplifiers have been widely used to control circuit. At this point the current through the transistor is
motors, but they are now also used as power amplifiers, zero and it is switched off. Both LC circuits are now filled
with some extra circuitry to allow analogue to be con- with energy in C and L0 . The whole circuit performs a
verted to a much higher frequency pulse width modu- damped oscillation. The damping by the load has been
82 CHAPTER 3. APPLICATIONS

+Vcc in 1975.[15] Some earlier reports on this operating class

have been published in Russian.

L1
L L0 C0
Class F In push–pull amplifiers and in CMOS, the
even harmonics of both transistors just cancel. Experi-
T1 C R L ment shows that a square wave can be generated by those
amplifiers. Theoretically square waves consist of odd
harmonics only. In a class-D amplifier, the output fil-
ter blocks all harmonics; i.e., the harmonics see an open
load. So even small currents in the harmonics suffice to
Class-E amplifier generate a voltage square wave. The current is in phase
with the voltage applied to the filter, but the voltage across
the transistors is out of phase. Therefore, there is a min-
imal overlap between current through the transistors and
adjusted so that some time later the energy from the Ls voltage across the transistors. The sharper the edges, the
is gone into the load, but the energy in both C0 peaks at lower the overlap.
the original value to in turn restore the original voltage so
that the voltage across the transistor is zero again and it While in class D, transistors and the load exist as two sep-
can be switched on. arate modules, class F admits imperfections like the par-
asitics of the transistor and tries to optimise the global
With load, frequency, and duty cycle (0.5) as given pa- system to have a high impedance at the harmonics. Of
rameters and the constraint that the voltage is not only course there has to be a finite voltage across the tran-
restored, but peaks at the original voltage, the four pa- sistor to push the current across the on-state resistance.
rameters (L, L0 , C and C0 ) are determined. The class-E Because the combined current through both transistors is
amplifier takes the finite on resistance into account and mostly in the first harmonic, it looks like a sine. That
tries to make the current touch the bottom at zero. This means that in the middle of the square the maximum of
means that the voltage and the current at the transistor current has to flow, so it may make sense to have a dip
are symmetric with respect to time. The Fourier trans- in the square or in other words to allow some overswing
form allows an elegant formulation to generate the com- of the voltage square wave. A class-F load network by
plicated LC networks and says that the first harmonic is definition has to transmit below a cutoff frequency and
passed into the load, all even harmonics are shorted and reflect above.
all higher odd harmonics are open.
Any frequency lying below the cutoff and having its sec-
Class E uses a significant amount of second-harmonic
ond harmonic above the cutoff can be amplified, that is
voltage. The second harmonic can be used to reduce the an octave bandwidth. On the other hand, an inductive-
overlap with edges with finite sharpness. For this to work,
capacitive series circuit with a large inductance and a tun-
energy on the second harmonic has to flow from the load able capacitance may be simpler to implement. By reduc-
into the transistor, and no source for this is visible in the
ing the duty cycle below 0.5, the output amplitude can
circuit diagram. In reality, the impedance is mostly re- be modulated. The voltage square waveform degrades,
active and the only reason for it is that class E is a class
but any overheating is compensated by the lower overall
F (see below) amplifier with a much simplified load net- power flowing. Any load mismatch behind the filter can
work and thus has to deal with imperfections. only act on the first harmonic current waveform, clearly
In many amateur simulations of class-E amplifiers, sharp only a purely resistive load makes sense, then the lower
current edges are assumed nullifying the very motivation the resistance, the higher the current.
for class E and measurements near the transit frequency Class F can be driven by sine or by a square wave, for
of the transistors show very symmetric curves, which look a sine the input can be tuned by an inductor to increase
much similar to class-F simulations. gain. If class F is implemented with a single transistor,
The class-E amplifier was invented in 1972 by Nathan O. the filter is complicated to short the even harmonics. All
Sokal and Alan D. Sokal, and details were first published previous designs use sharp edges to minimise the overlap.
3.3. AMPLIFIER 83

U (V) Ampli class G

+ Vss

+ Vs

0
t

- Vs

- Vss

U (V) Ampli class H

+ Vss

+ Vs

0
t

- Vs

- Vss

Basic schematic of a class-H configuration

increases. The terms “class G” and “class H” are used

Rail voltage modulation
Classes G and H There are a variety of amplifier de-
signs that enhance class-AB output stages with more ef-
ficient techniques to achieve greater efficiencies with low
distortion. These designs are common in large audio am-
plifiers since the heatsinks and power transformers would
be prohibitively large (and costly) without the efficiency
interchangeably to refer to different designs, varying in
definition from one manufacturer or paper to another.
Class-G amplifiers (which use “rail switching” to de-
crease power consumption and increase efficiency) are
more efficient than class-AB amplifiers. These ampli-
fiers provide several power rails at different voltages and
switch between them as the signal output approaches each
level. Thus, the amplifier increases efficiency by reduc-
ing the wasted power at the output transistors. Class-G
amplifiers are more efficient than class AB but less ef-
ficient when compared to class D, however, they do not
have the electromagnetic interference effects of class D.
84
Class-H amplifiers take the idea of class G one step fur-
ther creating an infinitely variable supply rail. This is
done by modulating the supply rails so that the rails are
only a few volts larger than the output signal at any given
time. The output stage operates at its maximum effi-
ciency all the time. Switched-mode power supplies can
be used to create the tracking rails. Significant effi-
ciency gains can be achieved but with the drawback of
more complicated supply design and reduced THD per-
formance. In common designs, a voltage drop of about
10V is maintained over the output transistors in Class H
circuits. The picture above shows positive supply voltage
of the output stage and the voltage at the speaker output.
The boost of the supply voltage is shown for a real music
signal.
The voltage signal shown is thus a larger version of the
input, but has been changed in sign (inverted) by the
amplification. Other arrangements of amplifying device
are possible, but that given (that is, common emitter,
common source or common cathode) is the easiest to un-
derstand and employ in practice. If the amplifying el-
ement is linear, the output is a faithful copy of the in-
put, only larger and inverted. In practice, transistors are
not linear, and the output only approximates the input.
nonlinearity from any of several sources is the origin of
distortion within an amplifier. The class of amplifier (A,
B, AB or C) depends on how the amplifying device is
biased. The diagrams omit the bias circuits for clarity.
Any real amplifier is an imperfect realization of an ideal
amplifier. An important limitation of a real amplifier is
that the output it generates is ultimately limited by the
power available from the power supply. An amplifier sat-
urates and clips the output if the input signal becomes
too large for the amplifier to reproduce or exceeds oper-
ational limits for the device.
Doherty amplifiers The Doherty, a hybrid configura-
tion, is currently receiving renewed attention. It was in-
vented in 1934 by William H. Doherty for Bell Laborato-
ries—whose sister company, Western Electric, manufac-
tured radio transmitters. The Doherty amplifier consists
of a class-B primary or carrier stages in parallel with a
class-C auxiliary or peak stage. The input signal splits to
drive the two amplifiers, and a combining network sums
the two output signals. Phase shifting networks are used
in inputs and outputs. During periods of low signal level,
the class-B amplifier efficiently operates on the signal and
the class-C amplifier is cutoff and consumes little power.
CHAPTER 3. APPLICATIONS
During periods of high signal level, the class-B amplifier
delivers its maximum power and the class-C amplifier de-
livers up to its maximum power. The efficiency of previ-
ous AM transmitter designs was proportional to modula-
tion but, with average modulation typically around 20%,
transmitters were limited to less than 50% efficiency. In
Doherty’s design, even with zero modulation, a transmit-
ter could achieve at least 60% efficiency.[16]
As a successor to Western Electric for broadcast trans-
mitters, the Doherty concept was considerably refined by
Continental Electronics Manufacturing Company of Dal-
las, TX. Perhaps, the ultimate refinement was the screen-
grid modulation scheme invented by Joseph B. Sainton.
The Sainton amplifier consists of a class-C primary or
carrier stage in parallel with a class-C auxiliary or peak
stage. The stages are split and combined through 90-
degree phase shifting networks as in the Doherty ampli-
fier. The unmodulated radio frequency carrier is applied
to the control grids of both tubes. Carrier modulation is
applied to the screen grids of both tubes. The bias point
of the carrier and peak tubes is different, and is estab-
lished such that the peak tube is cutoff when modulation
is absent (and the amplifier is producing rated unmodu-
lated carrier power) whereas both tubes contribute twice
the rated carrier power during 100% modulation (as four
times the carrier power is required to achieve 100% mod-
ulation). As both tubes operate in class C, a significant
improvement in efficiency is thereby achieved in the final
stage. In addition, as the tetrode carrier and peak tubes
require very little drive power, a significant improvement
in efficiency within the driver stage is achieved as well
(317C, et al.).[17] The released version of the Sainton
amplifier employs a cathode-follower modulator, not a
push–pull modulator. Previous Continental Electronics
designs, by James O. Weldon and others, retained most
of the characteristics of the Doherty amplifier but added
screen-grid modulation of the driver (317B, et al.).
The Doherty amplifier remains in use in very-high-power
AM transmitters, but for lower-power AM transmitters,
vacuum-tube amplifiers in general were eclipsed in the
1980s by arrays of solid-state amplifiers, which could be
switched on and off with much finer granularity in re-
sponse to the requirements of the input audio. However,
interest in the Doherty configuration has been revived
by cellular-telephone and wireless-Internet applications
where the sum of several constant envelope users creates
an aggregate AM result. The main challenge of the Do-
herty amplifier for digital transmission modes is in align-
ing the two stages and getting the class-C amplifier to turn
3.3. AMPLIFIER 85

on and off very quickly.

R1 R3 R4 +V supply

Recently, Doherty amplifiers have found widespread use Q3

in cellular base station transmitters for GHz frequencies. Q4

Implementations for transmitters in mobile devices have Input

D1
Output
also been demonstrated. C1 Q1 Q2 D2
Q5
R7 R8 C2

R2 R5
3.3.5 Implementation R6

0V (ground)

Amplifiers are implemented using active elements of dif-

ferent kinds: A practical amplifier circuit

• The first active elements were relays. They were for

example used in transcontinental telegraph lines: a
weak current was used to switch the voltage of a bat- typical (though substantially simplified) design as found
tery to the outgoing line. in modern amplifiers, with a class-AB push–pull output
stage, and uses some overall negative feedback. Bipolar
• For transmitting audio, carbon microphones were
transistors are shown, but this design would also be real-
used as the active element. This was used to modu-
izable with FETs or valves.
late a radio-frequency source in one of the first AM
audio transmissions, by Reginald Fessenden on Dec. The input signal is coupled through capacitor C1 to the
24, 1906.[18] base of transistor Q1. The capacitor allows the AC sig-
nal to pass, but blocks the DC bias voltage established
• Amplifiers used vacuum tubes exclusively until the by resistors R1 and R2 so that any preceding circuit is
1960s. Today, tubes are used for specialist audio ap- not affected by it. Q1 and Q2 form a differential ampli-
plications such as guitar amplifiers and audiophile fier (an amplifier that multiplies the difference between
amplifiers. Many broadcast transmitters still use two inputs by some constant), in an arrangement known
vacuum tubes. as a long-tailed pair. This arrangement is used to conve-
• In the 1960s, the transistor started to take over. niently allow the use of negative feedback, which is fed
These days, discrete transistors are still used in high- from the output to Q2 via R7 and R8.
power amplifiers and in specialist audio devices. The negative feedback into the difference amplifier al-
lows the amplifier to compare the input to the actual out-
• Beginning in the 1970s, more and more transistors
put. The amplified signal from Q1 is directly fed to the
were connected on a single chip therefore creating
second stage, Q3, which is a common emitter stage that
the integrated circuit. A large number of ampli-
provides further amplification of the signal and the DC
fiers commercially available today are based on in-
bias for the output stages, Q4 and Q5. R6 provides the
tegrated circuits.
load for Q3 (A better design would probably use some
form of active load here, such as a constant-current sink).
For special purposes, other active elements have been
So far, all of the amplifier is operating in class A. The out-
used. For example, in the early days of the satellite com-
put pair are arranged in class-AB push–pull, also called
munication, parametric amplifiers were used. The core
a complementary pair. They provide the majority of
circuit was a diode whose capacity was changed by an
the current amplification (while consuming low quiescent
RF signal created locally. Under certain conditions, this
current) and directly drive the load, connected via DC-
RF signal provided energy that was modulated by the ex-
blocking capacitor C2. The diodes D1 and D2 provide
tremely weak satellite signal received at the earth station.
a small amount of constant voltage bias for the output
pair, just biasing them into the conducting state so that
Amplifier circuit crossover distortion is minimized. That is, the diodes
push the output stage firmly into class-AB mode (assum-
The practical amplifier circuit to the right could be the ing that the base-emitter drop of the output transistors is
basis for a moderate-power audio amplifier. It features a reduced by heat dissipation).
86 CHAPTER 3. APPLICATIONS

This design is simple, but a good basis for a practical de-
sign because it automatically stabilises its operating point,
since feedback internally operates from DC up through
the audio range and beyond. Further circuit elements
would probably be found in a real design that would roll
off the frequency response above the needed range to pre-
vent the possibility of unwanted oscillation. Also, the use
of fixed diode bias as shown here can cause problems if
the diodes are not both electrically and thermally matched
to the output transistors – if the output transistors turn
on too much, they can easily overheat and destroy them-
selves, as the full current from the power supply is not
limited at this stage.
A common solution to help stabilise the output devices is
to include some emitter resistors, typically an ohm or so.
Calculating the values of the circuit’s resistors and capac-
itors is done based on the components employed and the
intended use of the amp.
For the basics of radio frequency amplifiers using valves,
see Valved RF amplifiers.
output needed - e.g. short-time or continuous, and
dynamic range required - e.g. recorded program or
live
• In the case of high-powered audio applications re-
quiring long cables to the load - e.g. cinemas and
shipping centres - instead of using heavy gauge ca-
bles it may be more efficient to connect to the load
at line output voltage with matching transformers at
source and loads.
• To prevent instability and/or overheating, care is
need to ensure solid state amplifiers are ade-
quately loaded. Most have a rated minimum load
impedance.
• All amplifiers generate heat through electrical
losses. This heat must be dissipated via natural or
forced air cooling. Heat can damage or reduce ser-
vice life of electronic components. Consideration
should be given to the heating effects of or upon ad-
jacent equipment.

Notes on implementation Different methods of supplying power result in many dif-

ferent methods of bias. Bias is a technique by which the
Real world amplifiers are imperfect. active devices are set up to operate in a particular region,
or by which the DC component of the output signal is
• One consequence is that the power supply itself may set to the midpoint between the maximum voltages avail-
influence the output, and must itself be considered able from the power supply. Most amplifiers use several
when designing the amplifier devices at each stage; they are typically matched in spec-
ifications except for polarity. Matched inverted polarity
• a power amplifier is effectively an input signal
devices are called complementary pairs. Class-A ampli-
controlled power regulator - regulating the power
fiers generally use only one device, unless the power sup-
sourced from the power supply or mains to the am-
ply is set to provide both positive and negative voltages,
plifier’s load. The power output from a power am-
in which case a dual device symmetrical design may be
plifier cannot exceed the power input to it.
used. Class-C amplifiers, by definition, use a single po-
• The amplifier circuit has an “open loop” perfor- larity supply.
mance, that can be described by various parame- Amplifiers often have multiple stages in cascade to in-
ters (gain, slew rate, output impedance, distortion, crease gain. Each stage of these designs may be a dif-
bandwidth, signal to noise ratio, etc.) ferent type of amp to suit the needs of that stage. For
• Many modern amplifiers use negative feedback instance, the first stage might be a class-A stage, feed-
techniques to hold the gain at the desired value and ing a class-AB push–pull second stage, which then drives
to reduce distortion. Negative loop feedback has the a class-G final output stage, taking advantage of the
intended effect of electrically damping loudspeaker strengths of each type, while minimizing their weak-
motion, thereby damping the mechanical dynamic nesses.
performance of the loudspeaker.
• When assessing rated amplifier power output it is 3.3.6 See also
useful to consider the load to be applied, the form
of signal - i.e. speech or music, duration of power • Charge transfer amplifier
3.3. AMPLIFIER 87

• Distributed amplifier [8] RCA Receiving Tube Manual, RC-14 (1940) p 12

• Faithful amplification [9] ARRL Handbook, 1968; page 65

• Guitar amplifier [10] Jerry Del Colliano (20 February 2012), Pass Labs
XA30.5 Class-A Stereo Amp Reviewed, Home Theater
• Instrument amplifier Review, Luxury Publishing Group Inc.
• Instrumentation amplifier [11] Ask the Doctors: Tube vs. Solid-State Harmonics
• Low noise amplifier [12] Volume cranked up in amp debate
• Negative feedback amplifier
[13] A.P. Malvino, Electronic Principles (2nd Ed.1979. ISBN
• Operational amplifier 0-07-039867-4) p.299.

• Optical amplifier [14] Electronic and Radio Engineering, R.P.Terman, McGraw

Hill, 1964
• Power added efficiency
[15] N. O. Sokal and A. D. Sokal, “Class E – A New Class
• Programmable gain amplifier of High-Efficiency Tuned Single-Ended Switching Power
Amplifiers”, IEEE Journal of Solid-State Circuits, vol. SC-
• RF power amplifier 10, pp. 168–176, June 1975. HVK
• Valve audio amplifier [16] US patent 2210028, William H. Doherty, “Amplifier”, is-
sued 1940-08-06, assigned to Bell Telephone Laborato-
ries
3.3.7 References
[17] US patent 3314034, Joseph B. Sainton, “High Efficiency
[1] Patronis, Gene (1987). “Amplifiers”. In Glen Ballou. Amplifier and Push–Pull Modulator”, issued 1967-04-11,
Handbook for Sound Engineers: The New Audio Cyclo- assigned to Continental Electronics Manufacturing Com-
pedia. Howard W. Sams & Co. p. 493. ISBN 0-672- pany
21983-2.
[18] Lee, Thomas (2004). The Design of CMOS Radio-
[2] Robert Boylestad and Louis Nashelsky (1996). Electronic Frequency Integrated Circuits. New York, NY: Cambridge
Devices and Circuit Theory, 7th Edition. Prentice Hall University Press. p. 8. ISBN 978-0-521-83539-8.
College Division. ISBN 978-0-13-375734-7.

[3] • Mark Cherry, Maxim Engineering journal, volume

62, Amplifier Considerations in Ceramic Speaker
Applications, p.3, accessed 2012-10-01
[4] Robert S. Symons (1998). “Tubes: Still vital af-
ter all these years”. IEEE Spectrum 35 (4): 52–63.
doi:10.1109/6.666962.
[5] It is a curiosity to note that this table is a “Zwicky box”;
in particular, it encompasses all possibilities. See Fritz
Zwicky.
[6] John Everett (1992). Vsats: Very Small Aperture Termi-
nals. IET. ISBN 0-86341-200-9.
[7] Roy, Apratim; Rashid, S. M. S. (5 June 2012). “A
power efficient bandwidth regulation technique for a
low-noise high-gain RF wideband amplifier”. Central
European Journal of Engineering 2 (3): 383–391.
Bibcode:2012CEJE....2..383R. doi:10.2478/s13531-
012-0009-1.
3.3.8 External links
• Rane audio’s guide to amplifier classes
• Design and analysis of a basic class D amplifier
• Conversion: distortion factor to distortion attenua-
tion and THD
• An alternate topology called the grounded bridge
amplifier - pdf
• Contains an explanation of different amplifier
classes - pdf
• Reinventing the power amplifier - pdf
• Anatomy of the power amplifier, including informa-
tion about classes
88 CHAPTER 3. APPLICATIONS

• Tons of Tones - Site explaining non linear distortion

stages in Amplifier Models
• Class D audio amplifiers, white paper - pdf

• Class E Radio Transmitters - Tutorials, Schematics,

Examples, and Construction Details
Chapter 4

Background Theory
-
4.1 Electron hole Electrons: 1
Protons: 2
An electron hole is the conceptual and mathematical Neutrons: 2
Electron Hole
opposite of an electron, useful in the study of physics,
-
chemistry, and electronic engineering. The concept de-
scribes the lack of an electron at a position where one
could exist in an atom or atomic lattice. It is differ-
ent from the positron, which is an actual particle of Charge: +1
antimatter. He +
The electron hole was introduced into calculations for the
following situations:

When an electron leaves a helium atom, it leaves an electron hole

• If an electron is excited into a higher state it leaves a
in its place. This causes the helium atom to become positively
hole in its old state. This meaning is used in Auger
charged.
electron spectroscopy (and other x-ray techniques),
in computational chemistry, and to explain the low
electron-electron scattering-rate in crystals (metals, in an otherwise full bottle of water.[2]
semiconductors).

• In crystals, electronic band structure calculations

lead to an effective mass for the electrons, which
typically is negative at the top of a band. The neg-
ative mass is an unintuitive concept,[1] and in these
situations a more familiar picture is found by con-
sidering a positive charge with a positive mass.
4.1.1 Solid-state physics
In solid-state physics, an electron hole (usually referred
to simply as a hole) is the absence of an electron from an
otherwise full valence band. A hole is essentially a way
to conceptualize the interactions of the electrons within a
nearly full system, which is missing just a few electrons.
In some ways, the behavior of a hole within a semicon-
ductor crystal lattice is comparable to that of the bubble
Simplified analogy: Empty seat in an auditorium
Hole conduction in a valence band can be explained by
the following analogy. Imagine a row of people seated in
an auditorium, where there are no spare chairs. Someone
in the middle of the row wants to leave, so he jumps over
the back of the seat into an empty row, and walks out.
The empty row is analogous to the conduction band, and
the person walking out is analogous to a free electron.
Now imagine someone else comes along and wants to sit
down. The empty row has a poor view; so he does not
want to sit there. Instead, a person in the crowded row
moves into the empty seat the first person left behind.
The empty seat moves one spot closer to the edge and
the person waiting to sit down. The next person follows,
and the next, et cetera. One could say that the empty seat

89
90
moves towards the edge of the row. Once the empty seat The analogy above is quite simplified, and cannot explain
reaches the edge, the new person can sit down. why holes create an opposite effect to electrons in the Hall
In the process everyone in the row has moved along. effect and Seebeck effect. A more precise and detailed
If those people were negatively charged (like electrons), explanation follows.[3]
this movement would constitute conduction. If the seats
themselves were positively charged, then only the vacant
seat would be positive. This is a very simple model of
how hole conduction works.
In reality, due to the crystal structure properties, the hole
is not localized to a single position as described in the
previous example. Rather, the hole spans an area in the
crystal lattice covering many hundreds of unit cells. This
is equivalent to being unable to tell which broken bond
corresponds to the “missing” electron.
Instead of analyzing the movement of an empty state
in the valence band as the movement of many separate
electrons, a single equivalent imaginary particle called a
“hole” is considered. In an applied electric field, the elec-
trons move in one direction, corresponding to the hole
moving in the other. If a hole associates itself with a neu-
tral atom, that atom loses an electron and becomes posi-
tive. Therefore the hole is taken to have positive charge
of +e, precisely the opposite of the electron charge.
Detailed picture: A hole is the absence of a negative-
mass electron
A semiconductor electronic band structure (right) includes the
dispersion relation of each band, i.e. the energy of an electron E
as a function of the electron’s wavevector k. The “unfilled band”
is the semiconductor’s conduction band; it curves upward indi-
cating positive effective mass. The “filled band” is the semicon-
ductor’s valence band; it curves downward indicating negative
effective mass.
CHAPTER 4. BACKGROUND THEORY
• The dispersion relation determines how electrons re-
spond to forces (via the concept of effective mass).[3]
A dispersion relation is the relationship between
wavevector (k-vector) and energy in a band, part of the
electronic band structure. In quantum mechanics, the
electrons are waves, and energy is the wave frequency.
A localized electron is a wavepacket, and the motion of
an electron is given by the formula for the group veloc-
ity of a wave. An electric field affects an electron by
gradually shifting all the wavevectors in the wavepacket,
and the electron moves because its wave group velocity
changes. Therefore, again, the way an electron responds
to forces is entirely determined by its dispersion relation.
An electron floating in space has the dispersion relation
E=ℏ2 k2 /(2m), where m is the (real) electron mass and
ℏ is reduced Planck constant. In the conduction band
of a semiconductor, the dispersion relation is instead
E=ℏ2 k2 /(2m* ) (m* is the effective mass), so a conduction-
band electron responds to forces as if it had the mass m* .
• Electrons near the top of the valence band behave as
if they have negative mass.[3]
The dispersion relation near the top of the valence band
is E=ℏ2 k2 /(2m* ) with negative effective mass. So elec-
trons near the top of the valence band behave like they
have negative mass. When a force pulls the electrons to
the right, these electrons actually move left. This is solely
due to the shape of the valence band, and is unrelated to
whether the band is full or empty. If you could some-
how empty out the valence band and just put one electron
near the valence band maximum (an unstable situation),
this electron would move the “wrong way” in response to
forces.
• Positively-charged holes as a shortcut for calculating
the total current of an almost-full band.[3]
A perfectly-full band always has zero current. One way
to think about this fact is that the electron states near the
top of the band have negative effective mass, and those
near the bottom of the band have positive effective mass,
4.1. ELECTRON HOLE
so the net motion is exactly zero. If an otherwise-almost-
full valence band has a state without an electron in it, we
say that this state is occupied by a hole. There is a math-
ematical shortcut for calculating the current due to every
electron in the whole valence band: Start with zero cur-
rent (the total if the band were full), and subtract the cur-
rent due to the electrons that would be in each hole state
if it wasn't a hole. Since subtracting the current caused
by a negative charge in motion is the same as adding the
current caused by a positive charge moving on the same
path, the mathematical shortcut is to pretend that each
hole state is carrying a positive charge, while ignoring ev-
ery other electron state in the valence band.
• A hole near the top of the valence band moves the
same way as an electron near the top of the valence
band would move.[3]
This fact follows from the discussion and definition
above. This is an example where the auditorium anal-
ogy above is misleading. When a person moves left in a
full auditorium, an empty seat moves right. But in this
section we are imagining how electrons move through k-
space, not real space, and the effect of a force is to move
all the electrons through k-space in the same direction at
the same time. So a better analogy is a bubble underwa- 4.1.3 See also
ter in a river: The bubble moves the same direction as the
water, not opposite. • Band gap
• Conclusion: Hole as a positive-charge, positive-mass
quasiparticle.
From the above, a hole (A) carries a positive charge, and
(B) responds to electric and magnetic fields as if it had a
positive charge and positive mass. (The latter is because
a particle with positive charge and positive mass responds
to electric and magnetic fields in the same way as a par-
ticle with negative charge and negative mass.) That ex- 4.1.4 References
plains why holes can be treated in all situations as ordi-
[1] For these negative mass electrons, momentum is opposite
nary positively-charged quasiparticles.
Role in semiconductor technology
In some semiconductors, such as silicon, the hole’s effec-
tive mass is dependent on direction (anisotropic), how-
ever a value averaged over all directions can be used for
some macroscopic calculations.
In most semiconductors, the effective mass of a hole is
much larger than that of an electron. This results in lower
91
mobility for holes under the influence of an electric field
and this may slow down the speed of the electronic de-
vice made of that semiconductor. This is one major rea-
son for adopting electrons as the primary charge carriers,
whenever possible in semiconductor devices, rather than
holes.
That said, in many semiconductor devices, both electrons
and holes play an essential role. Examples include p–n
diodes and bipolar transistors.
4.1.2 Holes in quantum chemistry
An alternate meaning for the term electron hole is used
in computational chemistry. In coupled cluster methods,
the ground (or lowest energy) state of a molecule is inter-
preted as the “vacuum state”—conceptually, in this state
there are no electrons. In this scheme, the absence of an
electron from a normally-filled state is called a “hole” and
is treated as a particle, and the presence of an electron in a
normally-empty state is simply called an “electron”. This
terminology is almost identical to that used in solid-state
physics.
• Carrier generation and recombination
• Effective mass
• Electrical resistivity and conductivity
• Semiconductor
to velocity, so forces acting on these electrons cause their
velocity to change in the 'wrong' direction. As these elec-
trons gain energy (moving towards the top of the band),
they slow down.
[2] Weller, Paul F. (1967). “An analogy for elemen-
tary band theory concepts in solids”. J. Chem.
Educ 44 (7): 391. Bibcode:1967JChEd..44..391W.
doi:10.1021/ed044p391.
[3] Kittel, Introduction to Solid State Physics 8th edition, page
194-196
92 CHAPTER 4. BACKGROUND THEORY

4.2 P–n junction in one direction but not in the other (opposite) direction.
This property is explained in terms of forward bias and
reverse bias, where the term bias refers to an application
of electric voltage to the p–n junction.

4.2.2 Equilibrium (zero bias)

In a p–n junction, without an external applied voltage,

an equilibrium condition is reached in which a potential
A p–n junction. The circuit symbol is shown: the triangle corre- difference is formed across the junction. This potential
sponds to the p side.
difference is called built-in potential Vbi .
See also: p–n diode and Diode § Semiconductor diodes After joining p-type and n-type semiconductors, elec-
trons from the n region near the p–n interface tend to
diffuse into the p region. As electrons diffuse, they leave
A p–n junction is a boundary or interface between two
positively charged ions (donors) in the n region. Like-
types of semiconductor material, p-type and n-type, in-
wise, holes from the p-type region near the p–n interface
side a single crystal of semiconductor. It is created by
begin to diffuse into the n-type region, leaving fixed ions
doping, for example by ion implantation, diffusion of
(acceptors) with negative charge. The regions nearby the
dopants, or by epitaxy (growing a layer of crystal doped
p–n interfaces lose their neutrality and become charged,
with one type of dopant on top of a layer of crystal doped
forming the space charge region or depletion layer (see
with another type of dopant). If two separate pieces of
figure A).
material were used, this would introduce a grain bound-
ary between the semiconductors that would severely in-
hibit its utility by scattering the electrons and holes.
p–n junctions are elementary “building blocks” of
most semiconductor electronic devices such as diodes,
transistors, solar cells, LEDs, and integrated circuits; they
are the active sites where the electronic action of the
device takes place. For example, a common type of
transistor, the bipolar junction transistor, consists of two
p–n junctions in series, in the form n–p–n or p–n–p.
The discovery of the p–n junction is usually attributed to
American physicist Russell Ohl of Bell Laboratories.[1]
A Schottky junction is a special case of a p–n junction, Figure A. A p–n junction in thermal equilibrium with zero-bias
where metal serves the role of the p-type semiconductor. voltage applied. Electron and hole concentration are reported
with blue and red lines, respectively. Gray regions are charge-
neutral. Light-red zone is positively charged. Light-blue zone is
4.2.1 Properties of a p–n junction negatively charged. The electric field is shown on the bottom,
the electrostatic force on electrons and holes and the direction in
The p–n junction possesses some interesting properties which the diffusion tends to move electrons and holes.
that have useful applications in modern electronics. A p-
doped semiconductor is relatively conductive. The same The electric field created by the space charge region op-
is true of an n-doped semiconductor, but the junction be- poses the diffusion process for both electrons and holes.
tween them can become depleted of charge carriers, and There are two concurrent phenomena: the diffusion pro-
hence non-conductive, depending on the relative voltages cess that tends to generate more space charge, and the
of the two semiconductor regions. By manipulating this electric field generated by the space charge that tends to
non-conductive layer, p–n junctions are commonly used counteract the diffusion. The carrier concentration pro-
as diodes: circuit elements that allow a flow of electricity file at equilibrium is shown in figure A with blue and red
4.2. P–N JUNCTION 93

lines. Also shown are the two counterbalancing phenom-

ena that establish equilibrium.

PN junction operation in forward-bias mode, showing reducing

depletion width. Both p and n junctions are doped at a 1e15/cm3
doping level, leading to built-in potential of ~0.59 V. Reducing
depletion width can be inferred from the shrinking charge profile,
as fewer dopants are exposed with increasing forward bias.

P-type material repels the holes, while the negative poten-

tial applied to the N-type material repels the electrons. As
electrons and holes are pushed toward the junction, the
distance between them decreases. This lowers the bar-
rier in potential. With increasing forward-bias voltage,
the depletion zone eventually becomes thin enough that
Figure B. A p–n junction in thermal equilibrium with zero-bias the zone’s electric field cannot counteract charge carrier
voltage applied. Under the junction, plots for the charge density, motion across the p–n junction, as a consequence reduc-
the electric field, and the voltage are reported. ing electrical resistance. The electrons that cross the p–n
junction into the P-type material (or holes that cross into
The space charge region is a zone with a net charge pro- the N-type material) will diffuse in the near-neutral re-
vided by the fixed ions (donors or acceptors) that have gion. Therefore, the amount of minority diffusion in the
been left uncovered by majority carrier diffusion. When near-neutral zones determines the amount of current that
equilibrium is reached, the charge density is approxi- may flow through the diode.
mated by the displayed step function. In fact, the re-
gion is completely depleted of majority carriers (leaving Only majority carriers (electrons in N-type material or
a charge density equal to the net doping level), and the holes in P-type) can flow through a semiconductor for a
edge between the space charge region and the neutral re- macroscopic length. With this in mind, consider the flow
gion is quite sharp (see figure B, Q(x) graph). The space of electrons across the junction. The forward bias causes
charge region has the same magnitude of charge on both a force on the electrons pushing them from the N side
sides of the p–n interfaces, thus it extends farther on thetoward the P side. With forward bias, the depletion re-
less doped side in this example (the n side in figures A gion is narrow enough that electrons can cross the junc-
and B). tion and inject into the P-type material. However, they
do not continue to flow through the P-type material in-
definitely, because it is energetically favorable for them
4.2.3 Forward bias to recombine with holes. The average length an electron
travels through the P-type material before recombining is
In forward bias, the p-type is connected with the positive called the diffusion length, and it is typically on the order
terminal and the n-type is connected with the negative of micrometers.[2]
terminal. Although the electrons penetrate only a short distance
With a battery connected this way, the holes in the P- into the P-type material, the electric current continues
type region and the electrons in the N-type region are uninterrupted, because holes (the majority carriers) be-
pushed toward the junction. This reduces the width of gin to flow in the opposite direction. The total current
the depletion zone. The positive potential applied to the (the sum of the electron and hole currents) is constant in
94 CHAPTER 4. BACKGROUND THEORY

space, because any variation would cause charge buildup does so increasingly with increasing reverse-bias voltage.
over time (this is Kirchhoff’s current law). The flow of This increases the voltage barrier causing a high resis-
holes from the P-type region into the N-type region is tance to the flow of charge carriers, thus allowing min-
exactly analogous to the flow of electrons from N to P imal electric current to cross the p–n junction. The in-
(electrons and holes swap roles and the signs of all cur- crease in resistance of the p–n junction results in the junc-
rents and voltages are reversed). tion behaving as an insulator.
Therefore, the macroscopic picture of the current flow The strength of the depletion zone electric field increases
through the diode involves electrons flowing through the as the reverse-bias voltage increases. Once the electric
N-type region toward the junction, holes flowing through field intensity increases beyond a critical level, the p–
the P-type region in the opposite direction toward the n junction depletion zone breaks down and current be-
junction, and the two species of carriers constantly re- gins to flow, usually by either the Zener or the avalanche
combining in the vicinity of the junction. The electrons breakdown processes. Both of these breakdown pro-
and holes travel in opposite directions, but they also have
cesses are non-destructive and are reversible, as long as
opposite charges, so the overall current is in the same di-
the amount of current flowing does not reach levels that
rection on both sides of the diode, as required. cause the semiconductor material to overheat and cause
The Shockley diode equation models the forward-bias thermal damage.
operational characteristics of a p–n junction outside the This effect is used to one’s advantage in Zener diode reg-
avalanche (reverse-biased conducting) region. ulator circuits. Zener diodes have a certain – low – break-
down voltage. A standard value for breakdown voltage is
for instance 5.6 V. This means that the voltage at the cath-
4.2.4 Reverse bias ode can never be more than 5.6 V higher than the voltage
at the anode, because the diode will break down – and
therefore conduct – if the voltage gets any higher. This
in effect regulates the voltage over the diode.
Another application of reverse biasing is Varicap diodes,
where the width of the depletion zone (controlled with
the reverse bias voltage) changes the capacitance of the
diode.

4.2.5 Governing Equations

A silicon p–n junction in reverse bias.
Size of depletion region
Connecting the P-type region to the negative terminal of
the battery and the N-type region to the positive terminal See also: Band bending
corresponds to reverse bias. If a diode is reverse-biased,
the voltage at the cathode is comparatively higher than
For a p–n junction, letting CA (x) and CD (x) be the con-
the anode. Therefore, no current will flow until the diode
centrations of acceptor and donor atoms respectively, and
breaks down. Reverse-bias usually refers to how a diode
letting N0 (x) and P0 (x) be the equilibrium concentra-
is used in a circuit. The connections are illustrated in the
tions of electrons and holes respectively, yields, by Pois-
diagram to the right.
son’s equation:
Because the p-type material is now connected to the neg- 2

ative terminal of the power supply, the 'holes' in the P- − ddxV2 = ρ

type material are pulled away from the junction, caus-
ing the width of the depletion zone to increase. Like-
wise, because the N-type region is connected to the posi-
tive terminal, the electrons will also be pulled away from
the junction. Therefore, the depletion region widens, and
ε = q
ε [(N0 − P0 ) + (CD − CA )]
where V is the electric potential, ρ is the charge density,
ε is permittivity and q is the magnitude of the electron
charge. Letting dp be the width of the depletion region
within the p-side, and letting dn be the width of the de-
pletion region within the n-side, it must be that
4.2. P–N JUNCTION 95

dp CA = dn CD 4.2.6 Summary
because the total charge on either side of the depletion
The forward-bias and the reverse-bias properties of the
region must cancel out. Therefore, letting D and ∆V
p–n junction imply that it can be used as a diode. A p–n
represent the entire depletion region and the potential dif-
junction diode allows electric charges to flow in one di-
ference across it,
∫ ∫ rection, but not in the opposite direction; negative charges
∆V = D qε [(N0 − P0 ) + (CD − CA )] dx dx (electrons) can easily flow through the junction from n to
CA CD 2q
= CA +CD ε (dp + dn )2 p but not from p to n, and the reverse is true for holes.
When the p–n junction is forward-biased, electric charge
Where P0 = N0 = 0 , because we are in the depletion flows freely due to reduced resistance of the p–n junc-
region. And thus, letting d be the total width of the de- tion. When the p–n junction is reverse-biased, however,
pletion region, we get the junction barrier (and therefore resistance) becomes
√
CA +CD greater and charge flow is minimal.
d = 2ε q CA CD ∆V

Where ∆V can be written as ∆V0 + ∆Vext , where we

have broken up the voltage difference into the equilib- 4.2.7 Non-rectifying junctions
rium plus external components. The equilibrium poten-
tial results from diffusion forces, and thus we can cal- In the above diagrams, contact between the metal wires
culate ∆V0 by implementing the Einstein relation and and the semiconductor material also creates metal–
assuming the semiconductor is nondegenerate (i.e. the semiconductor junctions called Schottky diodes. In a
product P0 N0 is independent of the Fermi energy): simplified ideal situation a semiconductor diode would
( ) never function, since it would be composed of several
CA CD
∆V0 = kT q ln P0 N0 diodes connected back-to-front in series. But, in prac-
tice, surface impurities within the part of the semicon-
where T is the temperature of the semiconductor and k is
[3] ductor that touches the metal terminals will greatly re-
Boltzmann constant.
duce the width of those depletion layers to such an ex-
tent that the metal-semiconductor junctions do not act as
diodes. These non-rectifying junctions behave as ohmic
Current across depletion region contacts regardless of applied voltage polarity.

The Shockley ideal diode equation characterizes the cur-

rent across a p-n junction as a function of external voltage 4.2.8 See also
and ambient conditions (temperature, choice of semicon-
ductor, etc.). To see how it can be derived, we must ex- • Diode and junction diode
amine the various reasons for current. The convention
• Diode modelling
is that the forward (+) direction be pointed against the
diode’s built-in potential gradient at equilibrium. • Semiconductor

--- Semiconductor device

• Forward Current ( JF )
--- n-type semiconductor
--- Diffusion Current: current due to local imbal- --- p-type semiconductor
ances in carrier concentration n , via the equa-
• Transistor
tion JD ∝ −q∇n
--- Field-effect transistor
• Reverse Current ( JR ) --- Bipolar junction transistor
∗ Alloy-junction transistor
--- Field Current
∗ p–n–p transistor
--- Generation Current ∗ n–p–n transistor
96 CHAPTER 4. BACKGROUND THEORY

--- Transistor–transistor logic • Understanding the PN Junction - Explains PN junc-

tion in a very easy to understand language.
• Capacitance–voltage profiling
• Deep-level transient spectroscopy
• p–n diode
4.3 Bipolar transistor biasing
• Solar cell Bipolar transistor amplifiers must be properly biased to
operate correctly. In circuits made with individual de-
• Semiconductor detector
vices (discrete circuits), biasing networks consisting of
resistors are commonly employed. Much more elaborate
4.2.9 References biasing arrangements are used in integrated circuits, for
example, bandgap voltage references and current mirrors.
[1] Riordan, Michael; Lillian Hoddeson (1988). Crystal fire: The operating point of a device, also known as bias point,
the invention of the transistor and the birth of the informa-
quiescent point, or Q-point, is the point on the output char-
tion age. USA: W. W. Norton & Company. pp. 88–97.
ISBN 0-393-31851-6.
acteristics that shows the DC collector–emitter voltage
(V ₑ) and the collector current (I ) with no input signal
[2] Hook, J. R.; H. E. Hall (2001). Solid State Physics. John applied. The term is normally used in connection with
Wiley & Sons. ISBN 0-471-92805-4. devices such as transistors.
[3] Luque, Antonio; Steven Hegedus (29 March 2011).
Handbook of Photovoltaic Science and Engineering. John
Wiley & Sons. ISBN 978-0-470-97612-8. 4.3.1 Bias circuit requirements

Signal requirements for Class A amplifiers

4.2.10 Further reading
For analog operation of a Class A amplifier, the Q-point
• Shockley, William (1949). “The Theory of p-n is placed so the transistor stays in active mode (does not
Junctions in Semiconductors and p-n Junction Tran- shift to operation in the saturation region or cut-off re-
sistors”. Bell System Technical Journal 28 (3): 435–
gion) when input is applied. For digital operation, the
489. doi:10.1002/j.1538-7305.1949.tb03645.x. Q-point is placed so the transistor does the contrary -
Retrieved 12 June 2013. switches from the “on” (saturation) to the “off” (cutoff)
state. Often, the Q-point is established near the center
of the active region of a transistor characteristic to allow
4.2.11 External links similar signal swings in positive and negative directions.
The Q-point should be stable; in particular, it should be
• http://www.youtube.com/watch?v=JBtEckh3L9Q insensitive to variations in transistor parameters (for ex-
Educational video on the P-N junction. ample, should not shift if transistor is replaced by another
• “P-N Junction” - PowerGuru, August, 2012. of the same type), variations in temperature, variations in
power supply voltage and so forth. The circuit must also
• Olav Torheim, Elementary Physics of P-N Junctions, be practical; both easily implemented and cost-effective.
2007.
• PN Junction Properties Calculator Thermal considerations
• PN Junction Lab free to use on nanoHUB.org allows
simulation and study of a P-N junction diode with At constant current, the voltage across the emitter–base
different doping and materials. Users can calculate junction VBE of a bipolar transistor decreases 2 mV
current-voltage (I-V) & capacitance-voltage (C-V) (silicon) and 1.8mV (germanium) for each 1 °C rise in
outputs, as well. temperature (reference being 25 °C). By the Ebers–Moll
model, if the base–emitter voltage VBE is held constant
• Theory of P-N Diodes – Dr. Vasileska (2009) and the temperature rises, the current through the base–
4.3. BIPOLAR TRANSISTOR BIASING 97

emitter diode IB will increase, and thus the collector

current IC will also increase. Depending on the bias
point, the power dissipated in the transistor may also in-
crease, which will further increase its temperature and
exacerbate the problem. This deleterious positive feed-
back results in thermal runaway.[1] There are several ap-
proaches to mitigate bipolar transistor thermal runaway.
For example,

• Negative feedback can be built into the biasing cir-

cuit so that increased collector current leads to de-
creased base current. Hence, the increasing collec-
tor current throttles its source.
• Heat sinks can be used that carry away extra heat and
prevent the base–emitter temperature from rising.
• The transistor can be biased so that its collector is
normally less than half of the power supply voltage,
which implies that collector–emitter power dissipa-
tion is at its maximum value. Runaway is then im-
possible because increasing collector current leads
to a decrease in dissipated power; this notion is Fixed bias (Base bias)
known as the half-voltage principle.

The circuits below primarily demonstrate the use of neg- V = IBRB + V ₑ

ative feedback to prevent thermal runaway.
Therefore,
4.3.2 Types of bias circuit for Class A am-
plifiers IB = (V - V ₑ)/RB

The following discussion treats five common biasing cir- For a given transistor, V ₑ does not vary significantly dur-
cuits used with Class A bipolar transistor amplifiers: ing use. As V is of fixed value, on selection of RB, the
base current IB is fixed. Therefore this type is called fixed
1. Fixed bias bias type of circuit.
2. Collector-to-base bias Also for given circuit,
3. Fixed bias with emitter resistor
V = ICRC + V ₑ
4. Voltage divider bias
5. Emitter bias Therefore,

Fixed bias (base bias) V ₑ = V - ICRC

This form of biasing is also called base bias. In the ex-
ample image on the right, the single power source (for
example, a battery) is used for both collector and base of
a transistor, although separate batteries can also be used.
In the given circuit,
The common-emitter current gain of a transistor is an im-
portant parameter in circuit design, and is specified on the
data sheet for a particular transistor. It is denoted as β on
this page.
Because
98 CHAPTER 4. BACKGROUND THEORY

IC = βIB

we can obtain IC as well. In this manner, operating point

given as (V ₑ,IC) can be set for given transistor.
Merits:

• It is simple to shift the operating point anywhere in

the active region by merely changing the base resis-
tor (RB).

• A very small number of components are required.

Demerits:

• The collector current does not remain constant with

variation in temperature or power supply voltage.
Collector-to-base bias
Therefore the operating point is unstable.

Collector Feedback Bias

• Changes in V ₑ will change IB and thus cause IE to
change. This in turn will alter the gain of the stage. This configuration employs negative feedback to prevent
thermal runaway and stabilize the operating point. In this
• When the transistor is replaced with another one, form of biasing, the base resistor R is connected to the
B
considerable change in the value of β can be ex- collector instead of connecting it to the DC source V .
cc
pected. Due to this change the operating point will So any thermal runaway will induce a voltage drop across
shift. the RC resistor that will throttle the transistor’s base cur-
rent.
• For small-signal transistors (e.g., not power transis-
tors) with relatively high values of β (i.e., between From Kirchhoff’s voltage law, the voltage VRb across the
100 and 200), this configuration will be prone to base resistor Rb is
thermal runaway. In particular, the stability factor,
which is a measure of the change in collector cur-
across drop VoltageRc base at Voltage
rent with changes in reverse saturation current, is z }| { z}|{
approximately β+1. To ensure absolute stability of VRb = Vcc − (Ic + Ib )Rc − Vbe .
the amplifier, a stability factor of less than 25 is pre-
ferred, and so small-signal transistors have large sta- By the Ebers–Moll model, Ic = βIb , and so
bility factors.
Ic
Usage: z}|{
VRb = Vcc −( βIb +Ib )Rc −Vbe = Vcc −Ib (β+1)Rc −Vbe .
Due to the above inherent drawbacks, fixed bias is rarely
used in linear circuits (i.e., those circuits which use the From Ohm’s law, the base current Ib = VRb /Rb , and so
transistor as a current source). Instead, it is often used
in circuits where transistor is used as a switch. However,
VRb
one application of fixed bias is to achieve crude automatic z}|{
gain control in the transistor by feeding the base resistor Ib Rb = Vcc − Ib (β + 1)Rc − Vbe .
from a DC signal derived from the AC output of a later
stage. Hence, the base current Ib is
4.3. BIPOLAR TRANSISTOR BIASING 99

Usage: The feedback also decreases the input impedance

Vcc − Vbe of the amplifier as seen from the base, which can be ad-
Ib = vantageous. Due to the gain reduction from feedback,
Rb + (β + 1)Rc
this biasing form is used only when the trade-off for sta-
If Vbe is held constant and temperature increases, then the bility is warranted.
collector current Ic increases. However, a larger Ic causes
the voltage drop across resistor Rc to increase, which in
Fixed bias with emitter resistor
turn reduces the voltage VRb across the base resistor Rb .
A lower base-resistor voltage drop reduces the base cur-
rent Ib , which results in less collector current Ic . Be-
cause an increase in collector current with temperature is
opposed, the operating point is kept stable.
Merits:

• Circuit stabilizes the operating point against varia-

tions in temperature and β (i.e. replacement of tran-
sistor)

Demerits:

• In this circuit, to keep Ic independent of β , the fol-

lowing condition must be met:

β(Vcc − Vbe ) (Vcc − Vbe )

Ic = βIb = ≈
Rb + Rc + βRc Rc

which is the case when

Fixed bias with emitter resistor

βRc ≫ Rb . The fixed bias circuit is modified by attaching an external

resistor to the emitter. This resistor introduces negative
• As β -value is fixed (and generally unknown) for a feedback that stabilizes the Q-point. From Kirchhoff’s
given transistor, this relation can be satisfied either voltage law, the voltage across the base resistor is
by keeping Rc fairly large or making Rb very low.
--- If Rc is large, a high Vcc is necessary, which
increases cost as well as precautions necessary VRb = VCC − Ie Re − Vbe
while handling.
--- If Rb is low, the reverse bias of the collector– From Ohm’s law, the base current is
base region is small, which limits the range of
collector voltage swing that leaves the transis-
tor in active mode. VRb
Ib =
Rb
• The resistor Rb causes an AC feedback, reducing
the voltage gain of the amplifier. This undesirable The way feedback controls the bias point is as follows.
effect is a trade-off for greater Q-point stability. If V ₑ is held constant and temperature increases, emitter
100 CHAPTER 4. BACKGROUND THEORY

current increases. However, a larger Iₑ increases the emit- • In addition to the above, RE causes ac feedback
ter voltage Vₑ = IₑRₑ, which in turn reduces the voltage which reduces the voltage gain of the amplifier.
VR across the base resistor. A lower base-resistor volt-
age drop reduces the base current, which results in less Usage:
collector current because I = β IB. Collector current and The feedback also increases the input impedance of the
emitter current are related by I = α Iₑ with α ≈ 1, so the amplifier when seen from the base, which can be advan-
increase in emitter current with temperature is opposed, tageous. Due to the above disadvantages, this type of
and the operating point is kept stable. biasing circuit is used only with careful consideration of
Similarly, if the transistor is replaced by another, there the trade-offs involved.
may be a change in IC (corresponding to change in β- Collector-Stabilized Biasing
value, for example). By similar process as above, the
change is negated and operating point kept stable.
Voltage divider biasing
For the given circuit,

VCC − Vbe
IB =
RB + (β + 1)RE

Merits:
The circuit has the tendency to stabilize operating point
against changes in temperature and β-value.
Demerits:

• In this circuit, to keep IC independent of β the fol-

lowing condition must be met:

β(VCC − Vbe ) (VCC − Vbe )

IC = βIB = ≈
RB + (β + 1)RE RE

which is approximately the case if

(β + 1)RE ≫ RB
Voltage divider bias
• As β-value is fixed for a given transistor, this relation
can be satisfied either by keeping RE very large, or The voltage divider is formed using external resistors R1
making RB very low. and R2 . The voltage across R2 forward biases the emit-
ter junction. By proper selection of resistors R1 and R2 ,
--- If RE is of large value, high VCC is necessary. the operating point of the transistor can be made inde-
This increases cost as well as precautions nec- pendent of β. In this circuit, the voltage divider holds the
essary while handling. base voltage fixed independent of base current provided
--- If RB is low, a separate low voltage supply the divider current is large compared to the base current.
should be used in the base circuit. Using two However, even with a fixed base voltage, collector cur-
supplies of different voltages is impractical. rent varies with temperature (for example) so an emitter
4.3. BIPOLAR TRANSISTOR BIASING 101

resistor is added to stabilize the Q-point, similar to the • As β-value is fixed for a given transistor, this relation
above circuits with emitter resistor. can be satisfied either by keeping RE fairly large, or
In this circuit the base voltage is given by: making R1 ||R2 very low.

VB = voltage across R2 = Vcc (R1R+R

2
− IB (RR11+R
R2 --- If RE is of large value, high VCC is necessary.
2) 2)
This increases cost as well as precautions nec-
essary while handling.
≈ Vcc (R1R+R
2
provided IB << I2 = --- If R1 || R2 is low, either R1 is low, or R2 is
2)
VB /R2 . low, or both are low. A low R1 raises VB
closer to VC, reducing the available swing in
collector voltage, and limiting how large RC
Also VB = Vbe + IE RE can be made without driving the transistor out
For the given circuit, of active mode. A low R2 lowers V ₑ, reducing
the allowed collector current. Lowering both
resistor values draws more current from the
power supply and lowers the input resistance
VCC
− Vbe of the amplifier as seen from the base.
1+R1 /R2
IB = .
(β + 1)RE + R1 ∥ R2
• AC as well as DC feedback is caused by RE, which
reduces the AC voltage gain of the amplifier. A
Merits:
method to avoid AC feedback while retaining DC
feedback is discussed below.
• Unlike above circuits, only one dc supply is neces-
sary. Usage:
• Operating point is almost independent of β varia- The circuit’s stability and merits as above make it widely
tion. used for linear circuits.

• Operating point stabilized against shift in tempera-

ture. Voltage divider with AC bypass capacitor The stan-
dard voltage divider circuit discussed above faces a draw-
back - AC feedback caused by resistor RE reduces the
Demerits:
gain. This can be avoided by placing a capacitor (CE) in
parallel with RE, as shown in circuit diagram.
• In this circuit, to keep IC independent of β the fol-
lowing condition must be met:
Emitter bias

When a split supply (dual power supply) is available, this

VCC
1+R1 /R2 − Vbe VCC
1+R1 /R2− Vbebiasing circuit is the most effective, and provides zero
IC = βIB = β ≈ bias
, voltage at the emitter or collector for load. The
(β + 1)RE + R1 ∥ R2 RE
negative supply VEE is used to forward-bias the emitter
junction through RE. The positive supply VCC is used
which is approximately the case if to reverse-bias the collector junction. Only two resistors
are necessary for the common collector stage and four
resistors for the common emitter or common base stage.
We know that,
(β + 1)RE >> R1 ∥ R2
VB - VE = V ₑ
where R1 || R2 denotes the equivalent resistance of R1 If RB is small enough, base voltage will be approximately
and R2 connected in parallel. zero. Therefore emitter current is,
102 CHAPTER 4. BACKGROUND THEORY

Voltage divider with capacitor Emitter bias

IE = (VEE - V ₑ)/RE
The operating point is independent of β if RE >> RB/β
Merit:
Good stability of operating point similar to voltage di-
vider bias. (about 1% of maximum possible value). Class AB bias
Demerit: is when the collector current I is about 1/4 of maximum
possible value. The class AB push–pull output amplifier
This type can only be used when a split (dual) power sup- circuit below could be the basis for a moderate-power au-
ply is available. dio amplifier.

4.3.3 Class B and AB amplifiers

Signal requirements

Class B and AB amplifiers employ 2 active devices to

cover the complete 360 deg of input signal flow. Each
transistor is therefore biased to perform over approxi-
mately 180 deg of the input signal. Class B bias is when
the collector current I with no signal is just conducting
4.3. BIPOLAR TRANSISTOR BIASING 103

• Bipolar junction transistor

R1 R3 R4 • MOSFET +V supply

Q3
4.3.5 References
Q4
Input [1] A.S. Sedra and K.C. Smith (2004). Microelectronic Cir-
D1 cuits (5th ed.). New York: Oxford University Press.
Output
p. 397, Figure 5.17, and p.&nb₡sp;1245. ISBN 0-19-
C1 Q1 Q2 D2 514251-9.
Q5
R7 R8 C2
4.3.6 Further reading
R2 R5
R6
• Patil, P.K.; Chitnis, M.M. (2005). Basic Electricity
and Semiconductor0VDevices.
(ground)
Phadke Prakashan.

A practical amplifier circuit

4.3.7 External links
Q3 is a common emitter stage that provides amplifica-
• Bias – from Sci-Tech Encyclopedia
tion of the signal and the DC bias current through D1
and D2 to generate a bias voltage for the output devices. • Electrical Engineering Training Series: Types of
The output pair are arranged in Class AB push–pull, also bias
called a complementary pair. The diodes D1 and D2 pro-
vide a small amount of constant voltage bias for the out-
put pair, just biasing them into the conducting state so
that crossover distortion is minimized. That is, the diodes
push the output stage into class-AB mode (assuming that
the base-emitter drop of the output transistors is reduced
by heat dissipation).
This design automatically stabilizes its operating point,
since overall feedback internally operates from DC up
through the audio range and beyond. The use of fixed
diode bias requires the diodes to be both electrically and
thermally matched to the output transistors. If the out-
put transistors conduct too much, they can easily over-
heat and destroy themselves, as the full current from the
power supply is not limited at this stage.
A common solution to help stabilize the output device
operating point is to include some emitter resistors, typi-
cally an ohm or so. Calculating the values of the circuit’s
resistors and capacitors is done based on the components
employed and the intended use of the amplifier.

4.3.4 See also

• Biasing (electronics)
• Small signal model
Chapter 5

Common Integrated Circuits

5.1 555 timer IC tions. The 555 can be used to provide time delays, as an
oscillator, and as a flip-flop element. Derivatives provide
up to four timing circuits in one package.
Introduced in 1971 by Signetics, the 555 is still in
widespread use due to its ease of use, low price, and sta-
bility. It is now made by many companies in the original
bipolar and also in low-power CMOS types. As of 2003,
it was estimated that 1 billion units are manufactured ev-
ery year.[1]

5.1.1 Design

VCC

NE555 from Signetics in dual-in-line package CONT

THRES OUT

VCC RESET
8 4 TRIG
CONT RESET
5 DISCH
6 R1
THRES 3
R OUT GND
S

Internal schematic

2
TRIG The IC was designed in 1971 by Hans Camenzind under
contract to Signetics, which was later acquired by Philips
7 (now NXP).
DISCH
1 Depending on the manufacturer, the standard 555 pack-
GND
age includes 25 transistors, 2 diodes and 15 resistors on a
silicon chip installed in an 8-pin mini dual-in-line pack-
Internal block diagram age (DIP-8).[2] Variants available include the 556 (a 14-
pin DIP combining two 555s on one chip), and the two
The 555 timer IC is an integrated circuit (chip) used in a 558 & 559s (both a 16-pin DIP combining four slightly
variety of timer, pulse generation, and oscillator applica- modified 555s with DIS & THR connected internally, and

104
5.1. 555 TIMER IC 105

VCC

THRES
RESET
GND 1 8 VCC
CONT
OUT
TRIG 2 7 DIS
TRIG
555
DISCH OUT 3 6 THR
GND

RESET 4 5 CTRL
Internal schematic (CMOS version)

TR is falling edge sensitive instead of level sensitive). Pinout diagram

The NE555 parts were commercial temperature range,
0 °C to +70 °C, and the SE555 part number desig-
nated the military temperature range, −55 °C to +125 Modes
°C. These were available in both high-reliability metal
can (T package) and inexpensive epoxy plastic (V pack- The 555 has three operating modes:
age) packages. Thus the full part numbers were NE555V,
NE555T, SE555V, and SE555T. It has been hypothe- • Monostable mode: In this mode, the 555 func-
sized that the 555 got its name from the three 5 kΩ resis- tions as a “one-shot” pulse generator. Applications
tors used within,[3] but Hans Camenzind has stated that include timers, missing pulse detection, bounce-
the number was arbitrary.[1] free switches, touch switches, frequency divider,
Low-power versions of the 555 are also available, such as capacitance measurement, pulse-width modulation
[4]
the 7555 and CMOS TLC555. The 7555 is designed to (PWM) and so on.
cause less supply noise than the classic 555 and the manu- • Astable (free-running) mode: The 555 can operate
facturer claims that it usually does not require a “control” as an oscillator. Uses include LED and lamp flash-
capacitor and in many cases does not require a decoupling ers, pulse generation, logic clocks, tone generation,
capacitor on the power supply. Those parts should gen- security alarms, pulse position modulation and so
erally be included, however, because noise produced by on. The 555 can be used as a simple ADC, convert-
the timer or variation in power supply voltage might inter- ing an analog value to a pulse length. E.g. selecting
fere with other parts of a circuit or influence its threshold a thermistor as timing resistor allows the use of the
voltages. 555 in a temperature sensor: the period of the out-
put pulse is determined by the temperature. The use
of a microprocessor based circuit can then convert
Pins the pulse period to temperature, linearize it and even
provide calibration means.
The connection of the pins for a DIP package is as fol-
lows: • Bistable mode or Schmitt trigger: The 555 can op-
erate as a flip-flop, if the DIS pin is not connected
Pin 5 is also sometimes called the CONTROL VOLT- and no capacitor is used. Uses include bounce-free
AGE pin. By applying a voltage to the CONTROL latched switches.
VOLTAGE input one can alter the timing characteris-
tics of the device. In most applications, the CONTROL
VOLTAGE input is not used. It is usual to connect a Monostable
10 nF capacitor between pin 5 and 0 V to prevent inter-
ference. The CONTROL VOLTAGE input can be used See also: RC circuit
to build an astable multivibrator with a frequency modu- In the monostable mode, the 555 timer acts as a “one-
lated output. shot” pulse generator. The pulse begins when the 555
106 CHAPTER 5. COMMON INTEGRATED CIRCUITS

VCC The output pulse width of time t, which is the time it takes
to charge C to 2/3 of the supply voltage, is given by

R 4 8

RESET VCC
t = RC ln(3) ≈ 1.1RC
7
DIS
6 3 where t is in seconds, R is in ohms (resistance) and C is
C THR OUT
Out in farads(capacitance).
2
TRIG While using the timer IC in monostable mode, the
Trigger
GND CTRL main disadvantage is that the time span between any
1 5 two triggering pulses must be greater than the RC time
10nF
constant.[6]

GND
Bistable
Schematic of a 555 in monostable mode

VCC

Reset 4 8

RESET VCC
2
TRIG
Trigger 6 3
THR OUT
7
Out
DIS
GND CTRL
1 5
10nF

GND

Schematic of a 555 in bistable mode

The relationships of the trigger signal, the voltage on C and the
pulse width in monostable mode
timer receives a signal at the trigger input that falls be-
low a third of the voltage supply. The width of the output
pulse is determined by the time constant of an RC net-
work, which consists of a capacitor (C) and a resistor (R).
The output pulse ends when the voltage on the capacitor
equals 2/3 of the supply voltage. The output pulse width
can be lengthened or shortened to the need of the specific
application by adjusting the values of R and C.[5]
In bistable (also called Schmitt trigger) mode, the 555
timer acts as a basic flip-flop. The trigger and reset in-
puts (pins 2 and 4 respectively on a 555) are held high
via Pull-up resistors while the threshold input (pin 6) is
simply floating. Thus configured, pulling the trigger mo-
mentarily to ground acts as a 'set' and transitions the out-
put pin (pin 3) to Vcc (high state). Pulling the reset input
to ground acts as a 'reset' and transitions the output pin
to ground (low state). No timing capacitors are required
in a bistable configuration. Pin 5 (control voltage) is con-
nected to ground via a small-value capacitor (usually 0.01
to 0.1 uF); pin 7 (discharge) is left floating.[7]
5.1. 555 TIMER IC 107

2
VCC The power capability of R1 must be greater than
Vcc
.
R1
Particularly with bipolar 555s, low values of R1 must be
R1
avoided so that the output stays saturated near zero volts
4 8
during discharge, as assumed by the above equation. Oth-
RESET VCC erwise the output low time will be greater than calculated
7 above. The first cycle will take appreciably longer than
DIS
6 3 the calculated time, as the capacitor must charge from
THR OUT 0V to 2/3 of VCC from power-up, but only from 1/3 of
R2 2
Out
VCC to 2/3 of VCC on subsequent cycles.
TRIG
GND CTRL To achieve a duty cycle of less than 50% a small diode
(that is fast enough for the application) can be placed
1 5
C
in parallel with R2 , with the cathode on the capacitor
10nF
side. This bypasses R2 during the high part of the cy-
cle so that the high interval depends approximately only
GND
on R1 and C. The presence of the diode is a voltage drop
that slows charging on the capacitor so that the high time
Standard 555 astable circuit is longer than the expected and often-cited ln(2)*R1 C =
0.693 R1 C. The low time will be the same as without the
diode as shown above. With a diode, the high time is
Astable

In astable mode, the 555 timer puts out a continuous ( )

2Vcc − 3Vdiode
stream of rectangular pulses having a specified frequency. high = R1 C · ln
Resistor R1 is connected between VCC and the discharge Vcc − 3Vdiode
pin (pin 7) and another resistor (R2 ) is connected be-
where V ᵢₒ ₑ is when the diode has a current of 1/2 of
tween the discharge pin (pin 7), and the trigger (pin 2) and
V /R1 which can be determined from its datasheet or by
threshold (pin 6) pins that share a common node. Hence
testing. As an extreme example, when V = 5 and V ᵢₒ ₑ=
the capacitor is charged through R1 and R2 , and dis-
0.7, high time = 1.00 R1 C which is 45% longer than the
charged only through R2 , since pin 7 has low impedance
“expected” 0.693 R1 C. At the other extreme, when V =
to ground during output low intervals of the cycle, there-
15 and V ᵢₒ ₑ= 0.3, the high time = 0.725 R1 C which is
fore discharging the capacitor.
closer to the expected 0.693 R1 C. The equation reduces
In the astable mode, the frequency of the pulse stream to the expected 0.693 R1 C if V ᵢₒ ₑ= 0.
depends on the values of R1 , R2 and C:
The operation of RESET in this mode is not well defined,
some manufacturers’ parts will hold the output state to
1 [8]
f = ln(2)·C·(R 1 +2R2 ) what it was when RESET is taken low, others will send
the output either high or low.
The high time from each pulse is given by:
5.1.2 Specifications
high = ln(2) · (R1 + R2 ) · C These specifications apply to the NE555. Other 555
timers can have different specifications depending on the
and the low time from each pulse is given by: grade (military, medical, etc.).

low = ln(2) · R2 · C 5.1.3 Derivatives

where R1 and R2 are the values of the resistors in ohms Many pin-compatible variants, including CMOS ver-
and C is the value of the capacitor in farads. sions, have been built by various companies. Bigger
108 CHAPTER 5. COMMON INTEGRATED CIRCUITS

packages also exist with two or four timers on the same “game paddles” or two joysticks to the host computer. It
chip. The 555 is also known under the following type also used a single 555 for flashing the display cursor.
numbers: A similar circuit was used in the IBM PC.[11] In the joy-
stick interface circuit of the IBM PC, the capacitor (C)
556 Dual timer of the RC network (see Monostable Mode above) was
generally a 10 nF capacitor. The resistor (R) of the RC
network consisted of the potentiometer inside the joy-
stick along with an external resistor of 2.2 kilohms.[12]
The joystick potentiometer acted as a variable resistor.
By moving the joystick, the resistance of the joystick in-
creased from a small value up to about 100 kilohms. The
joystick operated at 5 V.[13]
Software running in the host computer started the pro-
cess of determining the joystick position by writing to a
special address (ISA bus I/O address 201h).[13][14] This
would result in a trigger signal to the quad timer, which
would cause the capacitor (C) of the RC network to begin
charging and cause the quad timer to output a pulse. The
width of the pulse was determined by how long it took the
C to charge up to 2/3 of 5 V (or about 3.33 V), which was
in turn determined by the joystick position.[13][15] The
Die of a 556 dual timer manufactured by STMicroelectronics. software then measured the pulse width to determine the
joystick position. A wide pulse represented the full-right
The dual version is called 556. It features two complete joystick position, for example, while a narrow pulse rep-
555s in a 14 pin DIL package. resented the full-left joystick position.[13]

558 Quad timer 5.1.5 See also

The quad version is called 558 and has 16 pins. To fit • Counter
four 555s into a 16 pin package the power, control volt-
age, and reset lines are shared by all four modules. Each • OpAmp
module’s discharge and threshold circuits are wired to- • Oscillator
gether internally.
• RC circuit

XTR650/651 extended functionality hirel hitemp

(−60°C to 250+ °C) 5.1.6 References

This version includes non-overlapped complementary
outputs, coarse temperature sensor and on-chip 200pF
timing capacitance.[10]
[1] Ward, Jack (2004). The 555 Timer IC – An Interview
with Hans Camenzind. The Semiconductor Museum. Re-
trieved 2010-04-05
[2] van Roon, Fig 3 & related text.

5.1.4 Example applications [3] Scherz, Paul (2000) “Practical Electronics for Inventors”,
p. 589. McGraw-Hill/TAB Electronics. ISBN 978-0-07-
Joystick interface circuit using the 558 quad timer 058078-7. Retrieved 2010-04-05.

[4] Jung, Walter G. (1983) “IC Timer Cookbook, Second

The Apple II microcomputer used a quad timer 558 in Edition”, pp. 40–41. Sams Technical Publishing; 2nd ed.
monostable (or “one-shot”) mode to interface up to four ISBN 978-0-672-21932-0. Retrieved 2010-04-05.
5.2. OPERATIONAL AMPLIFIER 109

[5] van Roon, Chapter “Monostable Mode”. (Using the 555 5.1.8 External links
timer as a logic clock)
• 555 Timer Circuits – the Astable, Monostable and
[6] http://www.national.com/ds/LM/LM555.pdf Bistable
[7] http://www.555-timer-circuits.com/operating-modes. • Simple 555 timer circuits
html
• Java simulation of 555 oscillator circuit
[8] van Roon Chapter: “Astable operation”.
• NE555 Frequency and duty cycle calculator for
[9] http://www.customsiliconsolutions.com/
products-for-ASIC-solutions/standard-IC-products.
astable multivibrators
aspx
• Using NE555 as a Temperature DSP
[10] 15 X-REL Semiconductor Data Sheet, 38100 Grenoble
France • 555 Timer Tutorial

[11] Engdahl, pg 1. • Common Mistakes When Using a 555 Timer

[12] Engdahl, “Circuit diagram of PC joystick interface” • 555 and 556 Timer Circuits

[13] http://www.epanorama.net/documents/joystick/pc_ • 555 using areas and examples circuits

joystick.html
• Working with 555 Timer Circuits Engineers Garage
[14] Eggebrecht, p. 197.
• Analysis and synthesis of a 555 astable multivibrator
[15] Eggebrecht, pp. 197-99 circuit - online calculator

• Online simulations of a 555 astable multivibrator

5.1.7 Further reading circuit - online simulator

• 555 Timer Applications Sourcebook Experiments; H. IC Datasheets

Berlin; BPB Publications; 218 pages; 2008; ISBN
978-8176567909.
• NE555, Single Bipolar Timer, Texas Instruments
• Timer, Op Amp, and Optoelectronic Circuits and
• NE556, Dual Bipolar Timer, Texas Instruments
Projects; Forrest Mims III; Master Publishing; 128
pages; 2004; ISBN 978-0-945053-29-3. • NE558, Quad Bipolar Timer, NXP
• Engineer’s Mini-Notebook – 555 Timer IC Circuits; • LMC555, Single CMOS Timer, Texas Instruments
Forrest Mims III; Radio Shack; 33 pages; 1989; (operates down to 1.5 Volt at 50 uAmp)
ASIN B000MN54A6.
• ICM755x, Single / Dual CMOS Timer, Intersil (op-
• IC Timer Cookbook; 2nd Ed; Walter G Jung; Sams erates down to 2.0 Volt at 60 uAmp)
Publishing; 384 pages; 1983; ISBN 978-0-672-
21932-0. • ZSCT1555, Single CMOS Timer, Diodes Inc (op-
erates down to 0.9 Volt at 74 uAmp)
• 555 Timer Applications Sourcebook with Experi-
ments; Howard M Berlin; Sams Publishing; 158 • TS300x, Single CMOS Timers, Touchstone (oper-
pages; 1979; ISBN 978-0-672-21538-4. ates down to 0.9 Volt at 1.0 uAmp)

• IC 555 Projects; E.A. Parr; Bernard Babani Publish- • XTR65x, HiRel HiTemp Timer, X-REL (operates
ing; 144 pages; 1978; ISBN 978-0-85934-047-2. from −60°C to 230°C)
110 CHAPTER 5. COMMON INTEGRATED CIRCUITS

mode voltages that would destroy an ordinary op-amp),

and negative feedback amplifier (usually built from one
or more op-amps and a resistive feedback network).

5.2.1 Circuit notation

VS+

V+
A Signetics μa741 operational amplifier, one of the most success-
ful op-amps. Vout

5.2 Operational amplifier V−

An operational amplifier (op-amp) is a DC-coupled
high-gain electronic voltage amplifier with a differential
input and, usually, a single-ended output.[1] In this config- V
S−
uration, an op-amp produces an output potential (relative
to circuit ground) that is typically hundreds of thousands
Circuit diagram symbol for an op-amp
of times larger than the potential difference between its
input terminals.[2]
The circuit symbol for an op-amp is shown to the right,
Operational amplifiers had their origins in analog com- where:
puters, where they were used to do mathematical opera-
tions in many linear, non-linear and frequency-dependent
• V₊: non-inverting input
circuits. Characteristics of a circuit using an op-amp
are set by external components with little dependence on • V₋: inverting input
temperature changes or manufacturing variations in the
op-amp itself, which makes op-amps popular building • Vₒᵤ : output
blocks for circuit design.
• VS₊: positive power supply
Op-amps are among the most widely used electronic de-
vices today, being used in a vast array of consumer, in- • VS₋: negative power supply
dustrial, and scientific devices. Many standard IC op-
amps cost only a few cents in moderate production vol-
The power supply pins (VS₊ and VS₋) can be labeled in
ume; however some integrated or hybrid operational am-
different ways (See IC power supply pins). Often these
plifiers with special performance specifications may cost
pins are left out of the diagram for clarity, and the power
over $100 US in small quantities.[3] Op-amps may be
configuration is described or assumed from the circuit.
packaged as components, or used as elements of more
complex integrated circuits.
The op-amp is one type of differential amplifier. Other 5.2.2 Operation
types of differential amplifier include the fully differen-
tial amplifier (similar to the op-amp, but with two out- The amplifier’s differential inputs consist of a non-
puts), the instrumentation amplifier (usually built from inverting input (+) with voltage V₊ and an inverting input
three op-amps), the isolation amplifier (similar to the in- (–) with voltage V₋; ideally the op-amp amplifies only the
strumentation amplifier, but with tolerance to common- difference in voltage between the two, which is called the
5.2. OPERATIONAL AMPLIFIER 111

from the output to either input, this is an open loop cir-

V in cuit acting as a comparator. The circuit’s gain is just the
AOL of the op-amp.
V out
Closed loop

V in
Rg V out

An op-amp without negative feedback (a comparator)

Rf
Rg
differential input voltage. The output voltage of the op-
amp Vₒᵤ is given by the equation:

Vout = AOL (V+ − V− )

An op-amp with negative feedback (a non-inverting amplifier)
where AOL is the open-loop gain of the amplifier (the
term “open-loop” refers to the absence of a feedback loop
from the output to the input). If predictable operation is desired, negative feedback is
used, by applying a portion of the output voltage to the
inverting input. The closed loop feedback greatly re-
Open loop amplifier duces the gain of the circuit. When negative feedback
is used, the circuit’s overall gain and response becomes
The magnitude of AOL is typically very large—100,000 determined mostly by the feedback network, rather than
or more for integrated circuit op-amps—and therefore by the op-amp characteristics. If the feedback network is
even a quite small difference between V₊ and V₋ drives made of components with values small relative to the op
the amplifier output nearly to the supply voltage. Situ- amp’s input impedance, the value of the op-amp’s open
ations in which the output voltage is equal to or greater loop response AOL does not seriously affect the circuit’s
than the supply voltage are referred to as saturation of the performance. The response of the op-amp circuit with its
amplifier. The magnitude of AOL is not well controlled input, output, and feedback circuits to an input is charac-
by the manufacturing process, and so it is impractical to terized mathematically by a transfer function; designing
use an operational amplifier as a stand-alone differential an op-amp circuit to have a desired transfer function is
amplifier. in the realm of electrical engineering. The transfer func-
Without negative feedback, and perhaps with positive tions are important in most applications of op-amps, such
feedback for regeneration, an op-amp acts as a as in analog computers. High input impedance at the in-
comparator. If the inverting input is held at ground (0 put terminals and low output impedance at the output ter-
V) directly or by a resistor R , and the input voltage Vᵢ minal(s) are particularly useful features of an op-amp.
applied to the non-inverting input is positive, the output In the non-inverting amplifier on the right, the presence
will be maximum positive; if Vᵢ is negative, the output of negative feedback via the voltage divider R , R deter-
will be maximum negative. Since there is no feedback mines the closed-loop gain ACL = Vₒᵤ / Vᵢ . Equilibrium
112 CHAPTER 5. COMMON INTEGRATED CIRCUITS

will be established when Vₒᵤ is just sufficient to “reach VS+

around and pull” the inverting input to the same voltage
as Vᵢ . The voltage gain of the entire circuit is thus 1 +
R /R . As a simple example, if Vᵢ = 1 V and R = R ,
Vₒᵤ will be 2 V, exactly the amount required to keep V₋ v+
at 1 V. Because of the feedback provided by the R , R R in Rout
network, this is a closed loop circuit. vin vout
Another way to analyze this circuit proceeds by making
the following (usually valid) assumptions:[4] v− Gvin

• When an op-amp operates in linear (i.e., not satu-

rated) mode, the difference in voltage between the
VS−
non-inverting (+) pin and the inverting (−) pin is
negligibly small.
An equivalent circuit of an operational amplifier that models
• The input impedance between (+) and (−) pins is some resistive non-ideal parameters.
much larger than other resistances in the circuit.
• Infinite input impedance Rᵢ , and so zero input cur-
The input signal Vᵢ appears at both (+) and (−) pins, re- rent
sulting in a current i through R equal to Vᵢ /R .
• Zero input offset voltage

Vin • Infinite voltage range available at the output

i=
Rg • Infinite bandwidth with zero phase shift and infinite
Since Kirchhoff’s current law states that the same current slew rate
must leave a node as enter it, and since the impedance into • Zero output impedance Rₒᵤ
the (−) pin is near infinity, we can assume practically all
of the same current i flows through R , creating an output • Zero noise
voltage
• Infinite Common-mode rejection ratio (CMRR)

( ) • Infinite
( Power
) supply rejection ratio.
Vin Vin × Rf Rf
Vout = Vin +i×Rf = Vin + × Rf = Vin + = Vin 1 +
Rg Rg These ideals can
Rg be summarized by the two “golden rules
By combining terms, we determine the closed-loop gain extquotedbl:
ACL:
I. The output attempts to do whatever is nec-
essary to make the voltage difference between
Vout Rf the inputs zero.
ACL = =1+
Vin Rg II. The inputs draw no current.[5]:177

5.2.3 Op-amp characteristics The first rule only applies in the usual case where the op-
amp is used in a closed-loop design (negative feedback,
Ideal op-amps where there is a signal path of some sort feeding back
from the output to the inverting input). These rules are
An ideal op-amp is usually considered to have the follow- commonly used as a good first approximation
[5]:177
for analyz-
ing properties: ing or designing op-amp circuits.
None of these ideals can be perfectly realized. A real
• Infinite open-loop gain G = vₒᵤ / 'vin op-amp may be modeled with non-infinite or non-zero
5.2. OPERATIONAL AMPLIFIER 113

parameters using equivalent resistors and capacitors in
the op-amp model. The designer can then include these
effects into the overall performance of the final circuit.
Some parameters may turn out to have negligible effect
on the final design while others represent actual limita-
tions of the final performance that must be evaluated.
Real op-amps
Real op-amps differ from the ideal model in various as-
pects.
Low-impedance outputs typically require high quiescent
DC imperfections Real operational amplifiers suffer
from several non-ideal effects:
the amplifier will be significant. Hence, the out-
put impedance of the amplifier limits the maxi-
mum power that can be provided. In configurations
with a voltage-sensing negative feedback, the out-
put impedance of the amplifier is effectively low-
ered; thus, in linear applications, op-amps usually
exhibit a very low output impedance indeed. Neg-
ative feedback can not, however, reduce the limita-
tions that R ₒₐ in conjunction with Rₒᵤ place on the
maximum and minimum possible output voltages; it
can only reduce output errors within that range.
(i.e., idle) current in the output stage and will dissi-
pate more power, so low-power designs may pur-
posely sacrifice low output impedance.

Finite gain Open-loop gain is infinite in the ideal oper- Input current Due to biasing requirements or leakage, a
ational amplifier but finite in real operational am- small amount of current (typically ~10 nanoamperes
plifiers. Typical devices exhibit open-loop DC gain for bipolar op-amps, tens of picoamperes (pA) for
ranging from 100,000 to over 1 million. So long JFET input stages, and only a few pA for MOSFET
as the loop gain (i.e., the product of open-loop and input stages) flows into the inputs. When large re-
feedback gains) is very large, the circuit gain will be sistors or sources with high output impedances are
determined entirely by the amount of negative feed- used in the circuit, these small currents can produce
back (i.e., it will be independent of open-loop gain). large unmodeled voltage drops. If the input cur-
In cases where closed-loop gain must be very high, rents are matched, and the impedance looking out
the feedback gain will be very low, and the low feed- of both inputs are matched, then the voltages pro-
back gain causes low loop gain; in these cases, the duced at each input will be equal. Because the oper-
operational amplifier will cease to behave ideally. ational amplifier operates on the difference between
its inputs, these matched voltages will have no effect
(unless the operational amplifier has poor CMRR,
Finite input impedances The differential input which is described below). It is more common for
impedance of the operational amplifier is defined as the input currents (or the impedances looking out
the impedance between its two inputs; the common- of each input) to be slightly mismatched, and so a
mode input impedance is the impedance from small offset voltage (different from the input offset
each input to ground. MOSFET-input operational voltage below) can be produced. This offset volt-
amplifiers often have protection circuits that ef- age can create offsets or drifting in the operational
fectively short circuit any input differences greater amplifier. It can often be nulled externally; how-
than a small threshold, so the input impedance can ever, many operational amplifiers include offset null
appear to be very low in some tests. However, as or balance pins and some procedure for using them
long as these operational amplifiers are used in a to remove this offset. Some operational amplifiers
typical high-gain negative feedback application, attempt to nullify this offset automatically
these protection circuits will be inactive. The
input bias and leakage currents described below Input offset voltage This voltage, which is what is re-
are a more important design parameter for typical quired across the op-amp’s input terminals to drive
operational amplifier applications. the output voltage to zero,[6][nb 1] is related to the
mismatches in input bias current. In the perfect am-
Non-zero output impedance Low output impedance is plifier, there would be no input offset voltage. How-
important for low-impedance loads; for these loads, ever, it exists in actual op-amps because of imper-
the voltage drop across the output impedance of fections in the differential amplifier that constitutes
114
the input stage of the vast majority of these devices.
Input offset voltage creates two problems: First, due
to the amplifier’s high voltage gain, it virtually as-
sures that the amplifier output will go into satura-
tion if it is operated without negative feedback, even
when the input terminals are wired together. Sec-
ond, in a closed loop, negative feedback configura-
tion, the input offset voltage is amplified along with
the signal and this may pose a problem if high pre-
cision DC amplification is required or if the input
signal is very small.[nb 2]
Common-mode gain A perfect operational amplifier
amplifies only the voltage difference between its
two inputs, completely rejecting all voltages that are
common to both. However, the differential input
stage of an operational amplifier is never perfect,
leading to the amplification of these common volt-
ages to some degree. The standard measure of this
defect is called the common-mode rejection ratio
(denoted CMRR). Minimization of common mode
gain is usually important in non-inverting amplifiers
(described below) that operate at high amplification.
Power-supply rejection The output of a perfect oper-
ational amplifier will be completely independent
from ripples that arrive on its power supply in-
puts. Every real operational amplifier has a spec-
ified power supply rejection ratio (PSRR) that re-
flects how well the op-amp can reject changes in its
supply voltage. Copious use of bypass capacitors
can improve the PSRR of many devices, including
the operational amplifier.
Temperature effects All parameters change with tem-
perature. Temperature drift of the input offset volt-
age is especially important.
Drift Real op-amp parameters are subject to slow
change over time and with changes in temperature,
input conditions, etc.
Noise Amplifiers generate random voltage at the output
even when there is no signal applied. This can be
due to thermal noise and flicker noise of the devices.
For applications with high gain or high bandwidth,
noise becomes a very important consideration.
CHAPTER 5. COMMON INTEGRATED CIRCUITS
AC imperfections The op-amp gain calculated at DC
does not apply at higher frequencies. Thus, for high-
speed operation, more sophisticated considerations must
be used in an op-amp circuit design.
Finite bandwidth All amplifiers have finite bandwidth.
To a first approximation, the op-amp has the fre-
quency response of an integrator with gain. That
is, the gain of a typical op-amp is inversely propor-
tional to frequency and is characterized by its gain–
bandwidth product (GBWP). For example, an op-
amp with a GBWP of 1 MHz would have a gain of
5 at 200 kHz, and a gain of 1 at 1 MHz. This dy-
namic response coupled with the very high DC gain
of the op-amp gives it the characteristics of a first-
order low-pass filter with very high DC gain and low
cutoff frequency given by the GBWP divided by the
DC gain.
The finite bandwidth of an op-amp can be the
source of several problems, including:
• Stability. Associated with the band-
width limitation is a phase difference be-
tween the input signal and the amplifier
output that can lead to oscillation in some
feedback circuits. For example, a sinu-
soidal output signal meant to interfere
destructively with an input signal of the
same frequency will interfere construc-
tively if delayed by 180 degrees form-
ing positive feedback. In these cases,
the feedback circuit can be stabilized
by means of frequency compensation,
which increases the gain or phase mar-
gin of the open-loop circuit. The cir-
cuit designer can implement this com-
pensation externally with a separate cir-
cuit component. Alternatively, the com-
pensation can be implemented within the
operational amplifier with the addition
of a dominant pole that sufficiently at-
tenuates the high-frequency gain of the
operational amplifier. The location of
this pole may be fixed internally by the
manufacturer or configured by the circuit
designer using methods specific to the
op-amp. In general, dominant-pole fre-
quency compensation reduces the band-
width of the op-amp even further. When
5.2. OPERATIONAL AMPLIFIER
the desired closed-loop gain is high, op-
amp frequency compensation is often not
needed because the requisite open-loop
gain is sufficiently low; consequently, ap-
plications with high closed-loop gain can
make use of op-amps with higher band-
widths.
• Noise, Distortion, and Other Effects.
Reduced bandwidth also results in lower
amounts of feedback at higher frequen-
cies, producing higher distortion, noise,
and output impedance and also reduced
output phase linearity as the frequency
increases.
Typical low-cost, general-purpose op-amps
exhibit a GBWP of a few megahertz. Spe-
cialty and high-speed op-amps exist that can
achieve a GBWP of hundreds of megahertz.
For very high-frequency circuits, a current-
feedback operational amplifier is often used.
Input capacitance Most important for high frequency
operation because it further reduces the open-loop
bandwidth of the amplifier.
Common-mode gain See DC imperfections, above.
The input (yellow) and output (green) of a saturated op amp in
an inverting amplifier
Non-linear imperfections
Saturation Output voltage is limited to a minimum
and maximum value close to the power supply
voltages.[nb 3] Saturation occurs when the output of
the amplifier reaches this value and is usually due to:
115
• In the case of an op-amp using a bipolar power
supply, a voltage gain that produces an output
that is more positive or more negative than that
maximum or minimum; or
• In the case of an op-amp using a single sup-
ply voltage, either a voltage gain that pro-
duces an output that is more positive than that
maximum, or a signal so close to ground that
the amplifier’s gain is not sufficient to raise it
above the lower threshold.[nb 4]
Slewing The amplifier’s output voltage reaches its maxi-
mum rate of change, the slew rate, usually specified
in volts per microsecond. When slewing occurs, fur-
ther increases in the input signal have no effect on
the rate of change of the output. Slewing is usu-
ally caused by the input stage saturating; the result
is a constant current i driving a capacitance C in the
amplifier (especially those capacitances used to im-
plement its frequency compensation); the slew rate
is limited by dv/dt=i/C.
Slewing is associated with the large-signal per-
formance of an op-amp. Consider for, exam-
ple an op-amp configured for a gain of 10. Let
the input be a 1 V, 100 kHz sawtooth wave.
That is, the amplitude is 1 V and the period
is 10 microseconds. Accordingly, the rate of
change (i.e., the slope) of the input is 0.1 V
per microsecond. After 10x amplification, the
output should be a 10 V, 100 kHz sawtooth,
with a corresponding slew rate of 1 V per mi-
crosecond. However, the classic 741 op-amp
has a 0.5 V per microsecond slew rate speci-
fication, so that its output can rise to no more
than 5 V in the sawtooth’s 10 microsecond pe-
riod. Thus, if one were to measure the output,
it would be a 5 V, 100 kHz sawtooth, rather
than a 10 V, 100 kHz sawtooth.
Next consider the same amplifier and 100 kHz
sawtooth, but now the input amplitude is 100
mV rather than 1 V. After 10x amplification
the output is a 1 V, 100 kHz sawtooth with a
corresponding slew rate of 0.1 V per microsec-
ond. In this instance the 741 with its 0.5 V per
microsecond slew rate will amplify the input
properly.
116 CHAPTER 5. COMMON INTEGRATED CIRCUITS

Modern high speed op-amps can have slew Limited dissipated power The output current flows
rates in excess of 5,000 V per microsecond. through the op-amp’s internal output impedance,
However, it is more common for op-amps to dissipating heat. If the op-amp dissipates too much
have slew rates in the range 5-100 V per mi- power, then its temperature will increase above
crosecond. For example, the general purpose some safe limit. The op-amp may enter thermal
TL081 op-amp has a slew rate of 13 V per mi- shutdown, or it may be destroyed.
crosecond. As a general rule, low power and
small bandwidth op-amps have low slew rates.
As an example, the LT1494 micropower op- Modern integrated FET or MOSFET op-amps approx-
amp consumes 1.5 microamp but has a 2.7 kHz imate more closely the ideal op-amp than bipolar ICs
gain-bandwidth product and a 0.001 V per mi- when it comes to input impedance and input bias cur-
crosecond slew rate. rents. Bipolars are generally better when it comes to in-
put voltage offset, and often have lower noise. Generally,
Non-linear input-output relationship The output at room temperature, with a fairly large signal, and lim-
voltage may not be accurately proportional to ited bandwidth, FET and MOSFET op-amps now offer
the difference between the input voltages. It is better performance.
commonly called distortion when the input signal
is a waveform. This effect will be very small
in a practical circuit where substantial negative
feedback is used. 5.2.4 Internal circuitry of 741-type op-
amp
Phase reversal In some integrated op-amps, when the
published common mode voltage is violated (e.g. by
one of the inputs being driven to one of the supply Q8 Q9
Q12 Q13 7
VS+
voltages), the output may slew to the opposite polar- Q14

ity from what is expected in normal operation.[7][8] Non-inverting

input
Inverting
input
Q1 Q2 4.5 kΩ
Under such conditions, negative feedback becomes 3 2
Q16 Q17
25 Ω
30 pF
positive, likely causing the circuit to “lock up” in that Q3 Q4
39 kΩ
7.5 kΩ
6
Output
state. Q7
50 Ω

Q20
Q10 Q15

Power considerations 1 Q5
Q6 5
Offset Q11
Q22
Q19
Offset 50 kΩ
null null
1 kΩ 5 kΩ 50 kΩ 50 Ω
1 kΩ
4
Limited output current The output current must be fi- VS−

nite. In practice, most op-amps are designed to limit

the output current so as not to exceed a specified
level – around 25 mA for a type 741 IC op-amp
– thus protecting the op-amp and associated cir-
cuitry from damage. Modern designs are electroni-
cally more rugged than earlier implementations and
some can sustain direct short circuits on their out-
puts without damage.
Output sink current The output sink current is the
maximum current allowed to sink into the output
stage. Some manufacturers show the output voltage
vs. the output sink current plot, which gives an idea
of the output voltage when it is sinking current from
another source into the output pin.
A component-level diagram of the common 741 op-amp. Dotted
lines outline: current mirrors (red); differential amplifier (blue);
class A gain stage (magenta); voltage level shifter (green); output
stage (cyan).
Sourced by many manufacturers, and in multiple similar
products, an example of a bipolar transistor operational
amplifier is the 741 integrated circuit designed by Dave
Fullagar at Fairchild Semiconductor after Bob Widlar's
LM301 integrated circuit design.[9] In this discussion, we
use the parameters of the Hybrid-pi model to character-
ize the small-signal, grounded emitter characteristics of a
transistor. In this model, the current gain of a transistor
is denoted h ₑ, more commonly called the β.[10]
5.2. OPERATIONAL AMPLIFIER 117

Architecture output sink transistor Q20 receives its base drive from
the common collectors of Q15 and Q19; the level-shifter
A small-scale integrated circuit, the 741 op-amp shares Q16 provides base drive for the output source transistor
with most op-amps an internal structure consisting of Q14. Note the similarity between the transistors Q15 and
three gain stages: Q7.
The transistor Q22 prevents this stage from saturating
1. Differential amplifier (outlined blue) — provides by diverting the excessive Q15 base current (it acts as a
high differential amplification (gain), with rejec- Baker clamp).
tion of common-mode signal, low noise, high input
impedance, and drives a
Output amplifier The output stage (Q14, Q20, out-
2. Voltage amplifier (outlined magenta) — provides lined in cyan) is a Class AB push-pull emitter follower
high voltage gain, a single-pole frequency roll-off, amplifier. It provides an output drive with impedance of
and in turn drives the ≈50Ω, in essence, current gain. Transistor Q16 (outlined
3. Output amplifier (outlined cyan and green) — pro- in green) provides the quiescent current for the output
vides high current gain (low output impedance), transistors, and Q17 provides output current limiting.
along with output current limiting, and output short-
circuit protection. Biasing circuits

Additionally, it contains current mirror (outlined red) bias Provide appropriate quiescent current for each stage of
circuitry and a gain-stabilization capacitor (30 pF).
Differential amplifier A cascaded differential ampli-
fier followed by a current-mirror active load, the input
stage (outlined in blue) is a transconductance amplifier,
turning a differential voltage signal at the bases of Q1,
Q2 into a current signal into the base of Q15.
It entails two cascaded transistor pairs, satisfying conflict-
ing requirements. The first stage consists of the matched
NPN emitter follower pair Q1, Q2 that provide high input
impedance. The second is the matched PNP common-
base pair Q3, Q4 that eliminates the undesirable Miller
effect; it drives an active load Q7 plus matched pair Q5,
Q6.
That active load is implemented as a modified Wilson
current mirror; its role is to convert the (differential) in-
put current signal to a single-ended signal without the at-
tendant 50% losses (increasing the op-amp’s open-loop
gain by 3dB).[nb 5] Thus, a small-signal differential cur-
rent in Q3 versus Q4 appears summed (doubled) at the
base of Q15, the input of the voltage gain stage.
Voltage amplifier The (class-A) voltage gain stage
(outlined in magenta) consists of the two NPN transistors
Q15/Q19 connected in a Darlington configuration and
uses the output side of current mirror Q12/Q13 as its col-
lector (dynamic) load to achieve its high voltage gain. The The quiescent currents of Q1/Q3 (resp., Q2/Q4) i1 will
the op-amp.
The resistor (39 kΩ) connecting the (diode-connected)
Q11 and Q12, and the given supply voltage (VS₊−VS₋),
determine the current in the current mirrors, (matched
pairs) Q10/Q11 and Q12/Q13. The collector current of
Q11, i11 * 39 kΩ = VS₊ − VS₋ − 2 VBE. For the typical
VS = ±20 V, the standing current in Q11/Q12 (as well as
in Q13) would be ≈1 mA. A supply current for a typical
741 of about 2 mA agrees with the notion that these two
bias currents dominate the quiescent supply current.
Transistors Q11 and Q10 form a Widlar current mirror,
with quiescent current in Q10 i10 such that ln( i11 / i10 )
= i10 * 5 kΩ / 28 mV, where 5 kΩ represents the emitter
resistor of Q10, and 28 mV is VT, the thermal voltage at
room temperature. In this case i10 ≈ 20 μA.
Differential amplifier The biasing circuit of this stage
is set by a feedback loop that forces the collector currents
of Q10 and Q9 to (nearly) match. The small difference
in these currents provides the drive for the common base
of Q3/Q4 (note that the base drive for input transistors
Q1/Q2 is the input bias current and must be sourced ex-
ternally). The summed quiescent currents of Q1/Q3 plus
Q2/Q4 is mirrored from Q8 into Q9, where it is summed
with the collector current in Q10, the result being applied
to the bases of Q3/Q4.
118
thus be half of i10 , of order ≈ 10 μA. Input bias cur-
rent for the base of Q1 (resp. Q2) will amount to i1 / β;
typically ≈50 nA, implying a current gain h ₑ ≈ 200 for
Q1(Q2).
This feedback circuit tends to draw the common base
node of Q3/Q4 to a voltage V ₒ − 2 * VBE, where V ₒ
is the input common-mode voltage. At the same time,
the magnitude of the quiescent current is relatively insen-
sitive to the characteristics of the components Q1–Q4,
such as h ₑ, that would otherwise cause temperature de-
pendence or part-to-part variations.
Transistor Q7 drives Q5 and Q6 into conduction until
their (equal) collector currents match that of Q1/Q3 and
Q2/Q4. The quiescent current in Q7 is VBE / 50 kΩ,
about 35μA, as is the quiescent current in Q15, with its
matching operating point. Thus, the quiescent currents
are pairwise matched in Q1/Q2, Q3/Q4, Q5/Q6, and
Q7/Q15.
Voltage amplifier Quiescent currents in Q16 and Q19
are set by the current mirror Q12/Q13, which is running
at ≈ 1 mA. Through some (?) mechanism, the collector
current in Q19 tracks that standing current.
Output amplifier In the circuit involving Q16 (vari-
ously named rubber diode or VBE multiplier), the 4.5 kΩ
resistor must be conducting about 100 μA, with the Q16
VBE roughly 700 mV. Then the VCB must be about 0.45
V and VCE at about 1.0 V. Because the Q16 collector is
driven by a current source and the Q16 emitter drives into
the Q19 collector current sink, the Q16 transistor estab-
lishes a voltage difference between Q14 base and Q20
base of ≈ 1 V, regardless of the common-mode voltage
of Q14/Q20 base. The standing current in Q14/Q20 will
be a factor exp(100 mV / VT ) ≈ 36 smaller than the
1 mA quiescent current in the class A portion of the op
amp. This (small) standing current in the output transis-
tors establishes the output stage in class AB operation and Voltage amplifier A current signal i at
reduces the crossover distortion of this stage.
Small-signal differential mode
A small differential input voltage signal gives rise,
through multiple stages of current amplification, to a Output amplifier Output transistors Q14 and Q20 are
much larger voltage signal on output. each configured as an emitter follower, so no voltage gain
CHAPTER 5. COMMON INTEGRATED CIRCUITS
Input impedance Because Q1 and Q3 (resp. Q2 and
Q4) form a Darlington pair, the small-signal differen-
tial input impedance is of order 2hᵢₑh ₑ, where hᵢₑ is the
small-signal input impedance (common emitter) of Q1
and Q3 (resp. Q2 and Q4) and where h ₑ is the transis-
tor small-signal current gain (or β). This contrasts with
what would be the case with a simpler emitter-coupled
pair (long-tailed pair) input stage, where the differential
input impedance is 2hᵢₑ, a factor of β lower. A typical
741 op amp has an input impedance 2–8 MΩ.
Differential amplifier A differential voltage VI at the
op-amp inputs (pins 3 and 2, respectively) gives rise to a
small differential current in the bases of Q1 and Q2 iI ≈
VI / ( 2 hᵢₑ * h ₑ). This differential base current causes a
change in the differential collector current in each leg by
iI * h ₑ. Introducing the transconductance of Q1, gm =
h ₑ / hᵢₑ, the (small-signal) current at the base of Q15 (the
input of the voltage gain stage) is VI * gm / 2.
This portion of the op amp cleverly changes a differen-
tial signal at the op amp inputs to a single-ended signal at
the base of Q15, and in a way that avoids wastefully dis-
carding the signal in either leg. To see how, notice that
a small negative change in voltage at the inverting input
(Q2 base) drives it out of conduction, and this incremen-
tal decrease in current passes directly from Q4 collector
to its emitter, resulting in an decrease in base drive for
Q15. On the other hand, a small positive change in volt-
age at the non-inverting input (Q1 base) drives this tran-
sistor into conduction, reflected in an increase in current
at the collector of Q3. This current drives Q7 further into
conduction, which turns on current mirror Q5/Q6. Thus,
the increase in Q3 emitter current is mirrored in an in-
crease in Q6 collector current, resulting also in a decrease
in base drive for Q15. Besides avoiding wasting 3dB of
gain here, this technique decreases common-mode gain
and feedthrough of power supply noise.
2
Q15’s base
gives rise to a current in Q19 of order i * β (the product
of the h ₑ of each of Q15 and Q19, which are connected in
a Darlington pair). This current signal develops a voltage
at the bases of output transistors Q14/Q20 proportional
to the hᵢₑ of the respective transistor.
5.2. OPERATIONAL AMPLIFIER 119

occurs there; instead, this stage provides current gain, reactive and the closed loop gain is unity or higher.
equal to the h ₑ of Q14 (resp. Q20). By contrast, amplifiers requiring external compensation,
The output impedance is not zero, as it would be in an such as the μA748, may require external compensation
ideal op-amp, but with negative feedback it approaches or closed-loop gains significantly higher than unity.
zero at low frequencies.
Input offset voltage The “offset null” pins may be
Overall open-loop voltage gain The net open-loop used to place external resistors (typically in the form of
small-signal voltage gain of the op amp involves the prod- the two ends of a potentiometer, with the slider connected
uct of the current gain h ₑ of some 4 transistors. In prac- to VS–) in parallel with the emitter resistors of Q5 and
tice, the voltage gain for a typical 741-style op amp is Q6, to adjust the balance of the Q5/Q6 current mirror.
of order 200,000, and the current gain, the ratio of in- The potentiometer is adjusted such that the output is null
put impedance (≈2−6 MΩ) to output impedance (≈50Ω) (midrange) when the inputs are shorted together.
provides yet more (power) gain.
Non-linear characteristics
Other linear characteristics
Input breakdown voltage The transistors Q3, Q4
Small-signal common mode gain The ideal op help to increase the reverse VBE rating: the base-emitter
amp has infinite common-mode rejection ratio, or zero junctions of the NPN transistors Q1 and Q2 break down
common-mode gain. at around 7V, but the PNP transistors Q3 and Q4 have
VBE breakdown voltages around 50 V.[11]
In the present circuit, if the input voltages change in the
same direction, the negative feedback makes Q3/Q4 base
voltage follow (with 2VBE below) the input voltage vari- Output-stage voltage swing and current limiting
ations. Now the output part (Q10) of Q10-Q11 current Variations in the quiescent current with temperature, or
mirror keeps up the common current through Q9/Q8 con- between parts with the same type number, are common,
stant in spite of varying voltage. Q3/Q4 collector cur- so crossover distortion and quiescent current may be sub-
rents, and accordingly the output current at the base of ject to significant variation.
Q15, remain unchanged.
The output range of the amplifier is about one volt less
In the typical 741 op amp, the common-mode rejection than the supply voltage, owing in part to VBE of the out-
ratio is 90dB, implying an open-loop common-mode volt- put transistors Q14 and Q20.
age gain of about 6.
The 25 Ω resistor at the Q14 emitter, along with Q17,
acts to limit Q14 current to about 25 mA; otherwise, Q17
Frequency compensation The innovation of the conducts no current.
Fairchild μA741 was the introduction of frequency com- Current limiting for Q20 is performed in the voltage gain
pensation via an on-chip (monolithic) capacitor, simpli- stage: Q22 senses the voltage across Q19’s emitter resis-
fying application of the op amp by eliminating the need tor (50Ω); as it turns on, it diminishes the drive current
for external components for this function. The 30 pF ca- to Q15 base.
pacitor stabilizes the amplifier via Miller compensation
Later versions of this amplifier schematic may show a
and functions in a manner similar to an op-amp integrator
somewhat different method of output current limiting.
circuit. Also known as 'dominant pole compensation'
because it introduces a pole that masks (dominates) the
effects of other poles into the open loop frequency re- Applicability considerations
sponse; in a 741 op amp this pole can be as low as 10 Hz
(where it causes a −3 dB loss of open loop voltage gain).
Note: while the 741 was historically used in audio and
This internal compensation is provided to achieve un- other sensitive equipment, such use is now rare because
conditional stability of the amplifier in negative feed- of the improved noise performance of more modern op-
back configurations where the feedback network is non- amps. Apart from generating noticeable hiss, 741s and
120 CHAPTER 5. COMMON INTEGRATED CIRCUITS

other older op-amps may have poor common-mode rejec-
tion ratios and so will often introduce cable-borne mains
hum and other common-mode interference, such as switch
'clicks’, into sensitive equipment.
The “741” has come to often mean a generic op-amp IC
(such as μA741, LM301, 558, LM324, TBA221 — or
a more modern replacement such as the TL071). The
description of the 741 output stage is qualitatively simi-
lar for many other designs (that may have quite different
input stages), except:
• Some devices (μA748, LM301, LM308) are not in-
ternally compensated (require an external capacitor
from output to some point within the operational
amplifier, if used in low closed-loop gain applica-
tions).
• Some modern devices have “rail-to-rail output” ca-
pability, meaning that the output can range from
within a few millivolts of the positive supply volt-
age to within a few millivolts of the negative supply
voltage.
• Classification by internal compensation: op-amps
may suffer from high frequency instability in some
negative feedback circuits unless a small compen-
sation capacitor modifies the phase and frequency
responses. Op-amps with a built-in capacitor
are termed extquotedblcompensated extquotedbl, or
perhaps compensated for closed-loop gains down to
(say) 5. All others are considered uncompensated.
• Single, dual and quad versions of many commercial
op-amp IC are available, meaning 1, 2 or 4 opera-
tional amplifiers are included in the same package.
• Rail-to-rail input (and/or output) op-amps can work
with input (and/or output) signals very close to the
power supply rails.
• CMOS op-amps (such as the CA3140E) provide ex-
tremely high input resistances, higher than JFET-
input op-amps, which are normally higher than
bipolar-input op-amps.
• other varieties of op-amp include programmable op-
amps (simply meaning the quiescent current, gain,
bandwidth and so on can be adjusted slightly by an
external resistor).

5.2.5 Classification • manufacturers often tabulate their op-amps accord-

ing to purpose, such as low-noise pre-amplifiers,
Op-amps may be classified by their construction: wide bandwidth amplifiers, and so on.

• discrete (built from individual transistors or 5.2.6 Applications

tubes/valves)

• IC (fabricated in an Integrated circuit) — most com-

mon

• hybrid

IC op-amps may be classified in many ways, including:

• Military, Industrial, or Commercial grade (for ex-

DIP pinout for 741-type operational amplifier
ample: the LM301 is the commercial grade version
of the LM101, the LM201 is the industrial version). Main article: Operational amplifier applications
This may define operating temperature ranges and
other environmental or quality factors.

• Classification by package type may also affect envi- Use in electronics system design
ronmental hardiness, as well as manufacturing op-
tions; DIP, and other through-hole packages are The use of op-amps as circuit blocks is much easier and
tending to be replaced by surface-mount devices. clearer than specifying all their individual circuit ele-
5.2. OPERATIONAL AMPLIFIER 121

ments (transistors, resistors, etc.), whether the amplifiers mains interference and current spikes.
used are integrated or discrete. In the first approximation
op-amps can be used as if they were ideal differential gain
blocks; at a later stage limits can be placed on the accept- Positive feedback applications
able range of parameters for each op-amp.
Another typical configuration of op-amps is with posi-
Circuit design follows the same lines for all electronic cir- tive feedback, which takes a fraction of the output signal
cuits. A specification is drawn up governing what the back to the non-inverting input. An important applica-
circuit is required to do, with allowable limits. For ex- tion of it is the comparator with hysteresis, the Schmitt
ample, the gain may be required to be 100 times, with a trigger. Some circuits may use Positive feedback and Neg-
tolerance of 5% but drift of less than 1% in a specified ative feedback around the same amplifier, for example
temperature range; the input impedance not less than one Triangle wave oscillators and active filters.
megohm; etc.
Because of the wide slew-range and lack of positive feed-
A basic circuit is designed, often with the help of circuit back, the response of all the open-loop level detectors
modeling (on a computer). Specific commercially avail- described above will be relatively slow. External over-
able op-amps and other components are then chosen that all positive feedback may be applied but (unlike internal
meet the design criteria within the specified tolerances at positive feedback that may be applied within the latter
acceptable cost. If not all criteria can be met, the speci- stages of a purpose-designed comparator) this markedly
fication may need to be modified. affects the accuracy of the zero-crossing detection point.
A prototype is then built and tested; changes to meet or Using a general-purpose op-amp, for example, the fre-
improve the specification, alter functionality, or reduce quency of Eᵢ for the sine to square wave converter should
the cost, may be made. probably be below 100 Hz.

Applications without using any feedback Negative feedback applications

That is, the op-amp is being used as a voltage comparator.

Note that a device designed primarily as a comparator
may be better if, for instance, speed is important or a wide
range of input voltages may be found, since such devices
can quickly recover from full on or full off (“saturated”)
states.
A voltage level detector can be obtained if a reference
voltage Vᵣₑ is applied to one of the op-amp’s inputs. This
means that the op-amp is set up as a comparator to de-
tect a positive voltage. If the voltage to be sensed, Eᵢ, is
applied to op amp’s (+) input, the result is a noninverting
positive-level detector: when Eᵢ is above Vᵣₑ , VO equals
+V ₐ ; when Eᵢ is below Vᵣₑ , VO equals −V ₐ . If Eᵢ is
applied to the inverting input, the circuit is an inverting
positive-level detector: When Eᵢ is above Vᵣₑ , VO equals
−V ₐ .
A zero voltage level detector (Eᵢ = 0) can convert, for ex-
ample, the output of a sine-wave from a function genera-
tor into a variable-frequency square wave. If Eᵢ is a sine
wave, triangular wave, or wave of any other shape that
is symmetrical around zero, the zero-crossing detector’s
output will be square. Zero-crossing detection may also
be useful in triggering TRIACs at the best time to reduce
Vin
Vout
R1 R2
An op-amp connected in the non-inverting amplifier configura-
tion
Non-inverting amplifier In a non-inverting amplifier,
the output voltage changes in the same direction as the in-
put voltage.
The gain equation for the op-amp is:
Vout = AOL (V+ − V− )
However, in this circuit V₋ is a function of Vₒᵤ because of
the negative feedback through the R1 R2 network. R1 and
R2 form a voltage divider, and as V₋ is a high-impedance
input, it does not load it appreciably. Consequently:
122 CHAPTER 5. COMMON INTEGRATED CIRCUITS

Rf
V− = β · Vout

where
Rin
Vin
Vout
R1
β=
R1 + R2
Substituting this into the gain equation, we obtain:

An op-amp connected in the inverting amplifier configuration

Vout = AOL (Vin − β · Vout )

This time, V₋ is a function of both Vₒᵤ and Vᵢ due to the
Solving for Vout : voltage divider formed by R and Rᵢ . Again, the op-amp
input does not apply an appreciable load, so:
( )
1
Vout = Vin 1
β + 1/AOL V− = (Rf Vin + Rin Vout )
Rf + Rin
If AOL is very large, this simplifies to Substituting this into the gain equation and solving for
Vout :
( )
Vin Vin R2
Vout ≈ = R1
= Vin 1 + AOL Rf
β R1 +R2
R1 Vout = −Vin ·
Rf + Rin + AOL Rin
The non-inverting input of the operational amplifier If AOL is very large, this simplifies to
needs a path for DC to ground; if the signal source does
not supply a DC path, or if that source requires a given
load impedance, then the circuit will require another re- Rf
sistor from the non-inverting input to ground. When the Vout ≈ −Vin
Rin
operational amplifier’s input bias currents are significant,
A resistor is often inserted between the non-inverting in-
then the DC source resistances driving the inputs should
put and ground (so both inputs “see” similar resistances),
be balanced.[12] The ideal value for the feedback resis-
reducing the input offset voltage due to different voltage
tors (to give minimum offset voltage) will be such that the
drops due to bias current, and may reduce distortion in
two resistances in parallel roughly equal the resistance to
some op-amps.
ground at the non-inverting input pin. That ideal value
assumes the bias currents are well-matched, which may A DC-blocking capacitor may be inserted in series with
not be true for all op-amps.[13] the input resistor when a frequency response down to DC
is not needed and any DC voltage on the input is un-
wanted. That is, the capacitive component of the input
Inverting amplifier In an inverting amplifier, the out- impedance inserts a DC zero and a low-frequency pole
put voltage changes in an opposite direction to the input that gives the circuit a bandpass or high-pass characteris-
voltage. tic.
As with the non-inverting amplifier, we start with the gain The potentials at the operational amplifier inputs remain
equation of the op-amp: virtually constant (near ground) in the inverting configu-
ration. The constant operating potential typically results
in distortion levels that are lower than those attainable
Vout = AOL (V+ − V− ) with the non-inverting topology.
5.2. OPERATIONAL AMPLIFIER 123

Other applications

• audio- and video-frequency pre-amplifiers and

buffers
• differential amplifiers
• differentiators and integrators
• filters
• precision rectifiers
• precision peak detectors
• voltage and current regulators
• analog calculators
• analog-to-digital converters
• digital-to-analog converters
• Voltage clamping
• oscillators and waveform generators

Most single, dual and quad op-amps available have a stan-

dardized pin-out which permits one type to be substituted
for another without wiring changes. A specific op-amp
may be chosen for its open loop gain, bandwidth, noise
performance, input impedance, power consumption, or a
compromise between any of these factors.

5.2.7 Historical timeline

1941: A vacuum tube op-amp. An op-amp, defined
as a general-purpose, DC-coupled, high gain, inverting
feedback amplifier, is first found in U.S. Patent 2,401,779
“Summing Amplifier” filed by Karl D. Swartzel Jr. of GAP/R’s K2-W: a vacuum-tube op-amp (1953)
Bell Labs in 1941. This design used three vacuum tubes
to achieve a gain of 90 dB and operated on voltage rails of
±350 V. It had a single inverting input rather than differ-
R. Ragazzini of Columbia University. In this same paper
ential inverting and non-inverting inputs, as are common
a footnote mentioned an op-amp design by a student that
in today’s op-amps. Throughout World War II, Swartzel’s
would turn out to be quite significant. This op-amp, de-
design proved its value by being liberally used in the M9
signed by Loebe Julie, was superior in a variety of ways.
artillery director designed at Bell Labs. This artillery di-
It had two major innovations. Its input stage used a long-
rector worked with the SCR584 radar system to achievetailed triode pair with loads matched to reduce drift in the
extraordinary hit rates (near 90%) that would not have
output and, far more importantly, it was the first op-amp
been possible otherwise.[14] design to have two inputs (one inverting, the other non-
1947: An op-amp with an explicit non-inverting in- inverting). The differential input made a whole range
put. In 1947, the operational amplifier was first for- of new functionality possible, but it would not be used
mally defined and named in a paper by Professor John for a long time due to the rise of the chopper-stabilized
124 CHAPTER 5. COMMON INTEGRATED CIRCUITS

amplifier.[14] transistor in 1947, and the silicon transistor in 1954, the

1949: A chopper-stabilized op-amp. In 1949, Edwin concept of ICs became a reality. The introduction of the
A. Goldberg designed a chopper-stabilized op-amp.[15] planar process in 1959 made transistors and ICs stable
This set-up uses a normal op-amp with an additional AC enough to be commercially useful. By 1961, solid-state,
amplifier that goes alongside the op-amp. The chopper discrete op-amps were being produced. These op-amps
gets an AC signal from DC by switching between the DC were effectively small circuit boards with packages such
voltage and ground at a fast rate (60 Hz or 400 Hz). This as edge connectors. They usually had hand-selected re-
signal is then amplified, rectified, filtered and fed into sistors in order to improve things such as voltage offset
the op-amp’s non-inverting input. This vastly improved and drift. The P45 (1961) had a gain of 94 dB and ran
the gain of the op-amp while significantly reducing the on ±15 V rails. It was intended to deal with signals in the
output drift and DC offset. Unfortunately, any design range of ±10 V.
that used a chopper couldn't use their non-inverting input1961: A varactor bridge op-amp. There have
for any other purpose. Nevertheless, the much improved been many different directions taken in op-amp design.
characteristics of the chopper-stabilized op-amp made it Varactor bridge op-amps started to be produced in the
the dominant way to use op-amps. Techniques that used early 1960s.[16][17] They were designed to have extremely
the non-inverting input regularly would not be very pop- small input current and are still amongst the best op-amps
ular until the 1960s when op-amp ICs started to show up available in terms of common-mode rejection with the
in the field. ability to correctly deal with hundreds of volts at their in-
1953: A commercially available op-amp. In 1953, vac- puts.
uum tube op-amps became commercially available with
the release of the model K2-W from George A. Philbrick
Researches, Incorporated. The designation on the de-
vices shown, GAP/R, is an acronym for the complete
company name. Two nine-pin 12AX7 vacuum tubes
were mounted in an octal package and had a model K2-P
chopper add-on available that would effectively “use up”
the non-inverting input. This op-amp was based on a de-
scendant of Loebe Julie’s 1947 design and, along with its
successors, would start the widespread use of op-amps in
industry.

GAP/R’s model PP65: a solid-state op-amp in a potted module

(1962)

1962: An op-amp in a potted module. By 1962, sev-

eral companies were producing modular potted packages
that could be plugged into printed circuit boards. These
packages were crucially important as they made the op-
erational amplifier into a single black box which could be
GAP/R’s model P45: a solid-state, discrete op-amp (1961). easily treated as a component in a larger circuit.
1963: A monolithic IC op-amp. In 1963, the first
1961: A discrete IC op-amp. With the birth of the monolithic IC op-amp, the μA702 designed by Bob Wid-
5.2. OPERATIONAL AMPLIFIER 125

lar at Fairchild Semiconductor, was released. Monolithic

ICs consist of a single chip as opposed to a chip and dis-
crete parts (a discrete IC) or multiple chips bonded and
connected on a circuit board (a hybrid IC). Almost all
modern op-amps are monolithic ICs; however, this first
IC did not meet with much success. Issues such as an un-
even supply voltage, low gain and a small dynamic range
held off the dominance of monolithic op-amps until 1965
when the μA709[18] (also designed by Bob Widlar) was
released.
1968: Release of the μA741. The popularity of mono-
lithic op-amps was further improved upon the release of
the LM101 in 1967, which solved a variety of issues,
and the subsequent release of the μA741 in 1968. The
μA741 was extremely similar to the LM101 except that
Fairchild’s facilities allowed them to include a 30 pF com-
pensation capacitor inside the chip instead of requiring
external compensation. This simple difference has made
ADI’s HOS-050: a high speed hybrid IC op-amp (1979)
the 741 the canonical op-amp and many modern amps
base their pinout on the 741s. The μA741 is still in
production, and has become ubiquitous in electronics— tems.
many manufacturers produce a version of this classic
chip, recognizable by part numbers containing 741. The
same part is manufactured by several companies.
1970: First high-speed, low-input current FET de-
sign. In the 1970s high speed, low-input current designs
started to be made by using FETs. These would be largely
replaced by op-amps made with MOSFETs in the 1980s.
During the 1970s single sided supply op-amps also be-
came available.
1972: Single sided supply op-amps being produced.
A single sided supply op-amp is one where the input and
output voltages can be as low as the negative power supply
voltage instead of needing to be at least two volts above it.
The result is that it can operate in many applications with
the negative supply pin on the op-amp being connected to
the signal ground, thus eliminating the need for a separate
negative power supply.
The LM324 (released in 1972) was one such op-amp that An op-amp in a modern mini DIP
came in a quad package (four separate op-amps in one
package) and became an industry standard. In addition Recent trends. Recently supply voltages in analog cir-
to packaging multiple op-amps in a single package, the cuits have decreased (as they have in digital logic) and
1970s also saw the birth of op-amps in hybrid packages. low-voltage op-amps have been introduced reflecting
These op-amps were generally improved versions of ex- this. Supplies of ±5 V and increasingly 3.3 V (some-
isting monolithic op-amps. As the properties of mono- times as low as 1.8 V) are common. To maximize the
lithic op-amps improved, the more complex hybrid ICs signal range modern op-amps commonly have rail-to-rail
were quickly relegated to systems that are required to output (the output signal can range from the lowest supply
have extremely long service lives or other specialty sys- voltage to the highest) and sometimes rail-to-rail inputs.
126
5.2.8 See also
• Operational amplifier applications
• Differential amplifier
• Instrumentation amplifier
• Active filter
• Current-feedback operational amplifier
• Operational transconductance amplifier
• George A. Philbrick
• Bob Widlar
• Analog computer
• Negative feedback amplifier
• Current conveyor
5.2.9 Notes
[1] This definition hews to the convention of measuring op-
amp parameters with respect to the zero voltage point in
the circuit, which is usually half the total voltage between
the amplifier’s positive and negative power rails.
[2] Many older designs of operational amplifiers have offset
null inputs to allow the offset to be manually adjusted
away. Modern precision op-amps can have internal cir-
cuits that automatically cancel this offset using choppers
or other circuits that measure the offset voltage periodi-
cally and subtract it from the input voltage.
[3] That the output cannot reach the power supply voltages is
usually the result of limitations of the amplifier’s output
stage transistors. See Output stage.
[4] The output of older op-amps can reach to within one or
two volts of the supply rails. The output of newer so-called
“rail to rail” op-amps can reach to within millivolts of the
supply rails when providing low output currents.
[5] Widlar used this same trick in μA702 and μA709
5.2.10 References
[1] Maxim Application Note 1108: Understanding Single-
Ended, Pseudo-Differential and Fully-Differential ADC
Inputs – Retrieved November 10, 2007
[2] Analog devices MT-044 Tutorial
CHAPTER 5. COMMON INTEGRATED CIRCUITS
[3] “Burson Opamp”. Parts Connexion. Retrieved 24
September 2012. “BURSON-71484 Dual Discrete Op
Amp Modules, pair pcX Selling Price: $179.95/pr
BURSON-71485 Dual Discrete OpAmp Module, sin-
gle pcX Selling Price: $89.95 each. BURSON-71486
Single Discrete Op Amp Modules, pair pcX Selling
Price: $114.95/pr. Quantity discounts for Modifiers and
OEM’s”
[4] Jacob Millman, Microelectronics: Digital and Analog
Circuits and Systems, McGraw-Hill, 1979, ISBN 0-07-
042327-X, pp. 523-527
[5] Horowitz, Paul; Hill, Winfield (1989). The Art of Elec-
tronics. Cambridge, UK: Cambridge University Press.
ISBN 0-521-37095-7.
[6] D.F. Stout Handbook of Operational Amplifier Circuit De-
sign (McGraw-Hill, 1976, ISBN 0-07-061797-X ) pp. 1–
11.
[7] {{cite web |url=http://www.analog.com/static/
imported-files/tutorials/MT-036.pdf |title=Op Amp
Output Phase-Reversal and Input Over-Voltage
Protection |year=2009 |publisher=Analog Devices
|accessdate=2012-12-27}}
[8] {{cite web |url=http://www.edn.com/contents/images/
45890.pdf |title=Bootstrapping your op amp yields
wide voltage swings |last1=King |first1=Grayson
|last2=Watkins |first2=Tim |date=13 May 1999 |pub-
lisher=Electronic Design News |accessdate=2012-12-
27}}
[9] Lee, Thomas H. (November 18, 2002). “IC Op-Amps
Through the Ages”. Stanford UniversityHandout #18:
EE214 Fall 2002.
[10] Lu, Liang-Hung. “Electronics 2, Chapter 10”. National
Taiwan University, Graduate Institute of Electronics En-
gineering. Retrieved 2014-02-22.
[11] The μA741 Operational Amplifier
[12] An input bias current of 1 µA through a DC source re-
sistance of 10 kΩ produces a 10 mV offset voltage. If
the other input bias current is the same and sees the same
source resistance, then the two input offset voltages will
cancel out. Balancing the DC source resistances may not
be necessary if the input bias current and source resistance
product is small.
[13] http://www.analog.com/static/imported-files/tutorials/
MT-038.pdf
[14] Jung, Walter G. (2004). “Chapter 8: Op Amp History”.
Op Amp Applications Handbook. Newnes. p. 777. ISBN
978-0-7506-7844-5. Retrieved 2008-11-15.
5.3. PHASE-LOCKED LOOP 127

[15] http://www.analog.com/library/analogDialogue/ 5.2.12 External links

archives/39-05/Web_ChH_final.pdf
[16] http://www.philbrickarchive.org/
[17] June 1961 advertisement for Philbrick P2,
http://www.philbrickarchive.org/p2%20and%206033%
20ad%20rsi%20vol32%20no6%20june1961.pdf
[18] A.P. Malvino, Electronic Principles (2nd Ed. 1979. ISBN
0-07-039867-4) p. 476.
• Simple Op Amp Measurements How to measure
offset voltage, offset and bias current, gain, CMRR,
and PSRR.
• Introduction to op-amp circuit stages, second order
filters, single op-amp bandpass filters, and a simple
intercom

• MOS op amp design: A tutorial overview

5.2.11 Further reading • Operational Amplifier Noise Prediction (All Op
Amps) using spot noise
• Design with Operational Amplifiers and Analog In-
tegrated Circuits; 4th Ed; Sergio Franco; McGraw • Operational Amplifier Basics
Hill; 672 pages; 2014; ISBN 978-0078028168.
• History of the Op-amp from vacuum tubes to about
• Op Amps For Everyone; 4th Ed; Ron Mancini;
2002. Lots of detail, with schematics. IC part is
Newnes; 304 pages; 2013; ISBN 978-0123914958.
somewhat ADI-centric.
(Free PDF download of older version)

• Small Signal Audio Design; Douglas Self; Focal • Loebe Julie historical OpAmp interview by Bob
Press; 556 pages; 2010; ISBN 978-0-240-52177-0. Pease

• Op Amp Applications Handbook; Walt G. Jung; • www.PhilbrickArchive.org – A free repository of

Newnes; 896 pages; 2004; ISBN 978-0750678445. materials from George A Philbrick / Researches -
(Free PDF download) Operational Amplifier Pioneer
• Op Amps and Linear Integrated Circuits; James M. • What’s The Difference Between Operational Am-
Fiore; Cengage Learning; 616 pages; 2000; ISBN plifiers And Instrumentation Amplifiers?, Elec-
978-0766817937. tronic Design Magazine
• Operational Amplifiers and Linear Integrated Cir-
cuits; 6th Ed; Robert F Coughlin; Prentice Hall; 529 IC Datasheets
pages; 2000; ISBN 978-0130149916.
• Op-Amps and Linear Integrated Circuits; 4th Ed; • LM301, Single BJT OpAmp, Texas Instruments
Ram Gayakwad; Prentice Hall; 543 pages; 1999; • LM324, Quad BJT OpAmp, Texas Instruments
ISBN 978-0132808682.
• LM741, Single BJT OpAmp, Texas Instruments
• Basic Operational Amplifiers and Linear Integrated
Circuits; 2nd Ed; Thomas L Floyd and David • NE5532, Dual BJT OpAmp, Texas Instruments
Buchla; Prentice Hall; 593 pages; 1998; ISBN 978- (NE5534 is Quad)
0130829870.
• TL072, Dual JFET OpAmp, Texas Instruments
• Troubleshooting Analog Circuits; Bob Pease;
(TL074 is Quad)
Newnes; 217 pages; 1991; ISBN 978-0750694995.
• IC Op-Amp Cookbook; 3rd Ed; Walter G. Jung;
Prentice Hall; 433 pages; 1986; ISBN 978- 5.3 Phase-locked loop
0138896010.
• Engineer’s Mini-Notebook – OpAmp IC Circuits; “PLL” redirects here. For other uses, see PLL (disam-
Forrest Mims III; Radio Shack; 49 pages; 1985; biguation).
ASIN B000DZG196.
128 CHAPTER 5. COMMON INTEGRATED CIRCUITS

A phase-locked loop or phase lock loop (PLL) is a However, if there is an accident, a pace car comes out to
control system that generates an output signal whose set a safe speed. None of the race cars are permitted to
phase is related to the phase of an input signal. While pass the pace car (or the race cars in front of them), but
there are several differing types, it is easy to initially vi-
each of the race cars wants to stay as close to the pace
sualize as an electronic circuit consisting of a variable car as it can. While it is on the track, the pace car is a
frequency oscillator and a phase detector. The oscillator reference, and the race cars become phase-locked loops.
generates a periodic signal. The phase detector compares Each driver will measure the phase difference (a distance
the phase of that signal with the phase of the input peri- in laps) between him and the pace car. If the driver is far
odic signal and adjusts the oscillator to keep the phases away, he will increase his engine speed to close the gap.
matched. Bringing the output signal back toward the in- If he’s too close to the pace car, he will slow down. The
put signal for comparison is called a feedback loop since result is all the race cars lock on to the phase of the pace
the output is 'fed back' toward the input forming a loop. car. The cars travel around the track in a tight group that
Keeping the input and output phase in lock step also is a small fraction of a lap.
implies keeping the input and output frequencies the
same. Consequently, in addition to synchronizing signals, Clock analogy
a phase-locked loop can track an input frequency, or it
can generate a frequency that is a multiple of the input Phase can be proportional to time,[1] so a phase differ-
frequency. These properties are used for computer clock ence can be a time difference. Clocks are, with vary-
synchronization, demodulation, and frequency synthesis. ing degrees of accuracy, phase-locked (time-locked) to a
Phase-locked loops are widely employed in radio, master clock.
telecommunications, computers and other electronic ap- Left on its own, each clock will mark time at slightly dif-
plications. They can be used to demodulate a signal, re- ferent rates. A wall clock, for example, might be fast by
cover a signal from a noisy communication channel, gen- a few seconds per hour compared to the reference clock
erate a stable frequency at multiples of an input frequency at NIST. Over time, that time difference would become
(frequency synthesis), or distribute precisely timed clock substantial.
pulses in digital logic circuits such as microprocessors.
Since a single integrated circuit can provide a complete To keep his clock in sync, each week the owner compares
phase-locked-loop building block, the technique is widely the time on his wall clock to a more accurate clock (a
used in modern electronic devices, with output frequen- phase comparison), and he resets his clock. Left alone,
cies from a fraction of a hertz up to many gigahertz. the wall clock will continue to diverge from the reference
clock at the same few seconds per hour rate.
Some clocks have a timing adjustment (a fast-slow con-
trol). When the owner compared his wall clock’s time to
5.3.1 Practical analogies the reference time, he noticed that his clock was too fast.
Consequently, he could turn the timing adjust a small
Automobile race analogy amount to make the clock run a little slower. If things
work out right, his clock will be more accurate. Over a
For a practical idea of what is going on, consider an auto series of weekly adjustments, the wall clock’s notion of a
race. There are many cars, and the driver of each of them second would agree with the reference time (within the
wants to go around the track as fast as possible. Each lap wall clock’s stability).
corresponds to a complete cycle, and each car will com-
plete dozens of laps per hour. The number of laps per An early electromechanical version of a phase-locked
hour (a speed) corresponds to an angular velocity (i.e. loop was used in 1921 in the Shortt-Synchronome clock.
a frequency), but the number of laps (a distance) corre-
sponds to a phase (and the conversion factor is the dis-
5.3.2 History
tance around the track loop).
During most of the race, each car is on its own and the Spontaneous synchronization of weakly coupled pendu-
driver of the car is trying to beat the driver of every other lum clocks was noted by the Dutch physicist Christiaan
car on the course, and the phase of each car varies freely. Huygens as early as 1673.[2] Around the turn of the
5.3. PHASE-LOCKED LOOP 129

19th century, Lord Rayleigh observed synchronization of • feedback path (which may include a frequency di-
weakly coupled organ pipes and tuning forks.[3] In 1919, vider).
W. H. Eccles and J. H. Vincent found that two electronic
oscillators that had been tuned to oscillate at slightly dif-
ferent frequencies but that were coupled to a resonant cir- Variations
cuit would soon oscillate at the same frequency.[4] Au-
tomatic synchronization of electronic oscillators was de- There are several variations of PLLs. Some terms that
scribed in 1923 by Edward Victor Appleton.[5] are used are analog phase-locked loop (APLL) also re-
ferred to as a linear phase-locked loop (LPLL), digital
Earliest research towards what became known as the phase-locked loop (DPLL), all digital phase-locked loop
phase-locked loop goes back to 1932, when British (ADPLL), and software phase-locked loop (SPLL).[11]
researchers developed an alternative to Edwin Arm-
strong's superheterodyne receiver, the Homodyne or Analog or linear PLL (APLL) Phase detector is an
direct-conversion receiver. In the homodyne or synchro- analog multiplier. Loop filter is active or passive.
dyne system, a local oscillator was tuned to the desired Uses a Voltage-controlled oscillator (VCO).
input frequency and multiplied with the input signal. The
resulting output signal included the original modulation Digital PLL (DPLL) An analog PLL with a digital
information. The intent was to develop an alternative phase detector (such as XOR, edge-trigger JK,
receiver circuit that required fewer tuned circuits than phase frequency detector). May have digital divider
the superheterodyne receiver. Since the local oscilla- in the loop.
tor would rapidly drift in frequency, an automatic cor-
All digital PLL (ADPLL) Phase detector, filter and
rection signal was applied to the oscillator, maintaining
oscillator are digital. Uses a numerically controlled
it in the same phase and frequency as the desired sig-
oscillator (NCO).
nal. The technique was described in 1932, in a paper by
Henri de Bellescize, in the French journal L'Onde Élec- Software PLL (SPLL) Functional blocks are imple-
trique.[6][7][8] mented by software rather than specialized hard-
In analog television receivers since at least the late 1930s, ware.
phase-locked-loop horizontal and vertical sweep circuits Neuronal PLL (NPLL) Phase detector, filter and os-
are locked to synchronization pulses in the broadcast cillator are neurons or small neuronal pools. Uses
signal.[9] a rate controlled oscillator (RCO). Used for track-
When Signetics introduced a line of monolithic ing and decoding low frequency modulations (< 1
integrated circuits such as the NE565 that were complete kHz), such as those occurring during mammalian-
phase-locked loop systems on a chip in 1969, [10]
appli- like active sensing.
cations for the technique multiplied. A few years later
RCA introduced the extquotedblCD4046 extquotedbl
Performance parameters
CMOS Micropower Phase-Locked Loop, which became
a popular integrated circuit. • Type and order
• Lock range: The frequency range the PLL is able to
5.3.3 Structure and function stay locked. Mainly defined by the VCO range.

Phase-locked loop mechanisms may be implemented as
either analog or digital circuits. Both implementations
use the same basic structure. Both analog and digital PLL
circuits include four basic elements:
• Phase detector,
• Low-pass filter,
• Variable-frequency oscillator, and
• Capture range: The frequency range the PLL is able
to lock-in, starting from unlocked condition. This
range is usually smaller than the lock range and will
depend, for example, on phase detector.
• Loop bandwidth: Defining the speed of the control
loop.
• Transient response: Like overshoot and settling time
to a certain accuracy (like 50ppm).
130 CHAPTER 5. COMMON INTEGRATED CIRCUITS

• Steady-state errors: Like remaining phase or timing Clock recovery

error
Some data streams, especially high-speed serial data
• Output spectrum purity: Like sidebands generated streams (such as the raw stream of data from the magnetic
from a certain VCO tuning voltage ripple. head of a disk drive), are sent without an accompanying
• Phase-noise: Defined by noise energy in a certain clock. The receiver generates a clock from an approx-
frequency band (like 10 kHz offset from carrier). imate frequency reference, and then phase-aligns to the
Highly dependent on VCO phase-noise, PLL band- transitions in the data stream with a PLL. This process is
width, etc. referred to as clock recovery. In order for this scheme to
work, the data stream must have a transition frequently
• General parameters: Such as power consumption, enough to correct any drift in the PLL’s oscillator. Typ-
supply voltage range, output amplitude, etc. ically, some sort of line code, such as 8b/10b encoding,
is used to put a hard upper bound on the maximum time
between transitions.
5.3.4 Applications
Phase-locked loops are widely used for synchronization Deskewing
purposes; in space communications for coherent demod-
ulation and threshold extension, bit synchronization, and If a clock is sent in parallel with data, that clock can
symbol synchronization. Phase-locked loops can also be be used to sample the data. Because the clock must be
used to demodulate frequency-modulated signals. In ra- received and amplified before it can drive the flip-flops
dio transmitters, a PLL is used to synthesize new frequen- which sample the data, there will be a finite, and process-
cies which are a multiple of a reference frequency, with , temperature-, and voltage-dependent delay between the
the same stability as the reference frequency. detected clock edge and the received data window. This
Other applications include: delay limits the frequency at which data can be sent. One
way of eliminating this delay is to include a deskew PLL
• Demodulation of both FM and AM signals on the receive side, so that the clock at each data flip-flop
is phase-matched to the received clock. In that type of ap-
• Recovery of small signals that otherwise would be plication, a special form of a PLL called a delay-locked
lost in noise (lock-in amplifier to track the reference loop (DLL) is frequently used.[12]
frequency)
• Recovery of clock timing information from a data Clock generation
stream such as from a disk drive
• Clock multipliers in microprocessors that allow in- Many electronic systems include processors of various
ternal processor elements to run faster than external sorts that operate at hundreds of megahertz. Typically,
connections, while maintaining precise timing rela- the clocks supplied to these processors come from clock
tionships generator PLLs, which multiply a lower-frequency refer-
ence clock (usually 50 or 100 MHz) up to the operating
• DTMF decoders, modems, and other tone decoders, frequency of the processor. The multiplication factor can
for remote control and telecommunications be quite large in cases where the operating frequency is
multiple gigahertz and the reference crystal is just tens or
• DSP of video signals; Phase-locked loops are also
hundreds of megahertz.
used to synchronize phase and frequency to the input
analog video signal so it can be sampled and digitally
processed Spread spectrum
• Atomic force microscopy in tapping mode, to detect
All electronic systems emit some unwanted radio fre-
changes of the cantilever resonance frequency due to
quency energy. Various regulatory agencies (such as the
tip–surface interactions
FCC in the United States) put limits on the emitted en-
• DC motor drive ergy and any interference caused by it. The emitted noise
5.3. PHASE-LOCKED LOOP
generally appears at sharp spectral peaks (usually at the
operating frequency of the device, and a few harmonics).
A system designer can use a spread-spectrum PLL to re-
duce interference with high-Q receivers by spreading the
energy over a larger portion of the spectrum. For exam-
ple, by changing the operating frequency up and down
by a small amount (about 1%), a device running at hun-
dreds of megahertz can spread its interference evenly over
a few megahertz of spectrum, which drastically reduces
the amount of noise seen on broadcast FM radio chan-
nels, which have a bandwidth of several tens of kilohertz.
Clock distribution
Typically, the reference clock enters the chip and drives
a phase locked loop (PLL), which then drives the sys-
tem’s clock distribution. The clock distribution is usually
balanced so that the clock arrives at every endpoint simul-
taneously. One of those endpoints is the PLL’s feedback
input. The function of the PLL is to compare the dis-
tributed clock to the incoming reference clock, and vary
the phase and frequency of its output until the reference
and feedback clocks are phase and frequency matched.
PLLs are ubiquitous—they tune clocks in systems sev-
eral feet across, as well as clocks in small portions of in-
dividual chips. Sometimes the reference clock may not
actually be a pure clock at all, but rather a data stream
with enough transitions that the PLL is able to recover a
regular clock from that stream. Sometimes the reference
clock is the same frequency as the clock driven through
the clock distribution, other times the distributed clock
may be some rational multiple of the reference.
131
Jitter and noise reduction
One desirable property of all PLLs is that the refer-
ence and feedback clock edges be brought into very close
alignment. The average difference in time between the
phases of the two signals when the PLL has achieved lock
is called the static phase offset (also called the steady-
state phase error). The variance between these phases
is called tracking jitter. Ideally, the static phase offset
should be zero, and the tracking jitter should be as low as
possible.
Phase noise is another type of jitter observed in PLLs,
and is caused by the oscillator itself and by elements
used in the oscillator’s frequency control circuit. Some
technologies are known to perform better than others in
this regard. The best digital PLLs are constructed with
emitter-coupled logic (ECL) elements, at the expense of
high power consumption. To keep phase noise low in
PLL circuits, it is best to avoid saturating logic families
such as transistor-transistor logic (TTL) or CMOS.
Another desirable property of all PLLs is that the phase
and frequency of the generated clock be unaffected by
rapid changes in the voltages of the power and ground
supply lines, as well as the voltage of the substrate on
which the PLL circuits are fabricated. This is called sub-
strate and supply noise rejection. The higher the noise
rejection, the better.
To further improve the phase noise of the output, an
injection locked oscillator can be employed following the
VCO in the PLL.
Frequency synthesis
In digital wireless communication systems (GSM,
CDMA etc.), PLLs are used to provide the local os-
cillator up-conversion during transmission and down-
conversion during reception. In most cellular handsets
this function has been largely integrated into a single in-
tegrated circuit to reduce the cost and size of the handset.
However, due to the high performance required of base
station terminals, the transmission and reception circuits
are built with discrete components to achieve the levels
of performance required. GSM local oscillator modules
are typically built with a frequency synthesizer integrated
circuit and discrete resonator VCOs.
132 CHAPTER 5. COMMON INTEGRATED CIRCUITS

5.3.5 Block diagram Phase detector

Main article: phase detector

Block diagram of a phase-locked loop

A phase detector compares two input signals and pro- A phase detector (PD) generates a voltage, which rep-
duces an error signal which is proportional to their phase resents the phase difference between two signals. In a
difference. The error signal is then low-pass filtered and PLL, the two inputs of the phase detector are the ref-
used to drive a VCO which creates an output phase. The erence input and the feedback from the VCO. The PD
output is fed through an optional divider back to the input output voltage is used to control the VCO such that the
of the system, producing a negative feedback loop. If the phase difference between the two inputs is held constant,
output phase drifts, the error signal will increase, driving making it a negative feedback system. There are several
the VCO phase in the opposite direction so as to reduce types of phase detectors in the two main categories of
the error. Thus the output phase is locked to the phase at analog and digital.
the other input. This input is called the reference. Different types of phase detectors have different perfor-
Analog phase locked loops are generally built with an mance characteristics.
analog phase detector, low pass filter and VCO placed in For instance, the frequency mixer produces harmonics
a negative feedback configuration. A digital phase locked that adds complexity in applications where spectral purity
loop uses a digital phase detector; it may also have a di- of the VCO signal is important. The resulting unwanted
vider in the feedback path or in the reference path, or (spurious) sidebands, also called extquotedblreference
both, in order to make the PLL’s output signal frequency spurs extquotedbl can dominate the filter requirements
a rational multiple of the reference frequency. A non- and reduce the capture range and lock time well below
integer multiple of the reference frequency can also be the requirements. In these applications the more complex
created by replacing the simple divide-by-N counter in digital phase detectors are used which do not have as se-
the feedback path with a programmable pulse swallow- vere a reference spur component on their output. Also,
ing counter. This technique is usually referred to as a when in lock, the steady-state phase difference at the in-
fractional-N synthesizer or fractional-N PLL. puts using this type of phase detector is near 90 degrees.
The oscillator generates a periodic output signal. Assume The actual difference is determined by the DC loop gain.
that initially the oscillator is at nearly the same frequencyA bang-bang charge pump phase detector must always
as the reference signal. If the phase from the oscilla- have a dead band where the phases of inputs are close
tor falls behind that of the reference, the phase detec- enough that the detector detects no phase error. For this
tor changes the control voltage of the oscillator so that reason, bang-bang phase detectors are associated with
it speeds up. Likewise, if the phase creeps ahead of the significant minimum peak-to-peak jitter, because of drift
reference, the phase detector changes the control voltage within the dead band. However these types, having out-
to slow down the oscillator. Since initially the oscillator puts consisting of very narrow pulses at lock, are very
may be far from the reference frequency, practical phase useful for applications requiring very low VCO spurious
detectors may also respond to frequency differences, so outputs. The narrow pulses contain very little energy and
as to increase the lock-in range of allowable inputs. are easy to filter out of the VCO control voltage. This
Depending on the application, either the output of the results in low VCO control line ripple and therefore low
controlled oscillator, or the control signal to the oscilla- FM sidebands on the VCO.
tor, provides the useful output of the PLL system. In PLL applications it is frequently required to know
when the loop is out of lock. The more complex digi-
tal phase-frequency detectors usually have an output that
5.3.6 Elements
allows a reliable indication of an out of lock condition.
5.3. PHASE-LOCKED LOOP 133

Filter Feedback path and optional divider

The block commonly called the PLL loop filter (usually

a low pass filter) generally has two distinct functions.
The primary function is to determine loop dynamics, also
called stability. This is how the loop responds to dis-
turbances, such as changes in the reference frequency,
changes of the feedback divider, or at startup. Com-
mon considerations are the range over which the loop
can achieve lock (pull-in range, lock range or capture
range), how fast the loop achieves lock (lock time, lock-
up time or settling time) and damping behavior. Depend-
ing on the application, this may require one or more of An Example Digital Divider (by 4) for use in the Feedback Path
the following: a simple proportion (gain or attenuation), of a Multiplying PLL
an integral (low pass filter) and/or derivative (high pass
filter). Loop parameters commonly examined for this are PLLs may include a divider between the oscillator and
the loop’s gain margin and phase margin. Common con- the feedback input to the phase detector to produce a
cepts in control theory including the PID controller are frequency synthesizer. A programmable divider is par-
used to design this function. ticularly useful in radio transmitter applications, since a
The second common consideration is limiting the amount large number of transmit frequencies can be produced
of reference frequency energy (ripple) appearing at the from a single stable, accurate, but expensive, quartz
phase detector output that is then applied to the VCO crystal–controlled reference oscillator.
control input. This frequency modulates the VCO and Some PLLs also include a divider between the reference
produces FM sidebands commonly called “reference clock and the reference input to the phase detector. If
spurs”. The low pass characteristic of this block can be the divider in the feedback path divides by N and the
used to attenuate this energy, but at times a band reject reference input divider divides by M , it allows the PLL
“notch” may also be useful. to multiply the reference frequency by N /M . It might
The design of this block can be dominated by either of seem simpler to just feed the PLL a lower frequency, but
these considerations, or can be a complex process jug- in some cases the reference frequency may be constrained
gling the interactions of the two. Typical trade-offs are: by other issues, and then the reference divider is useful.
increasing the bandwidth usually degrades the stability Frequency multiplication can also be attained by locking
or too much damping for better stability will reduce the the VCO output to the Nth harmonic of the reference sig-
speed and increase settling time. Often also the phase- nal. Instead of a simple phase detector, the design uses a
noise is affected. harmonic mixer (sampling mixer). The harmonic mixer
turns the reference signal into an impulse train that is rich
in harmonics.[13] The VCO output is coarse tuned to be
close to one of those harmonics. Consequently, the de-
Oscillator sired harmonic mixer output (representing the difference
between the N harmonic and the VCO output) falls within
Main article: Electronic oscillator the loop filter passband.
It should also be noted that the feedback is not limited to
All phase-locked loops employ an oscillator element with a frequency divider. This element can be other elements
variable frequency capability. This can be an analog such as a frequency multiplier, or a mixer. The multiplier
VCO either driven by analog circuitry in the case of an will make the VCO output a sub-multiple (rather than a
APLL or driven digitally through the use of a digital-to- multiple) of the reference frequency. A mixer can trans-
analog converter as is the case for some DPLL designs. late the VCO frequency by a fixed offset. It may also
Pure digital oscillators such as a numerically controlled be a combination of these. An example being a divider
oscillator are used in ADPLLs. following a mixer; this allows the divider to operate at
134 CHAPTER 5. COMMON INTEGRATED CIRCUITS

a much lower frequency than the VCO without a loss in xc (θ) there is a function φ(θ) such that the output G(t)
loop gain. of Filter

5.3.7 Modeling ẋ = Ax + bφ(θr (t) − θc (t)),

Time domain model
The equations governing a phase-locked loop with an
analog multiplier as the phase detector and linear filter
may be derived as follows. Let the input to the phase
detector be xr (θr (t)) and the output of the VCO is
xc (θc (t)) with phases θr (t) and θc (t) . Functions xc (θ)
and xr (θ) describe waveforms of signals. Then the out-
put of the phase detector xm (t) is given by
xm (t) = xr (θr (t))xc (θc (t))
x(0) = x0 ,
G(t) = c∗ x,
in phase domain is asymptotically equal ( the difference
G(t) − xf (t) is small with respect to the frequencies) to
the output of the Filter in time domain model. [14] [15]
Here function φ(θ) is a phase detector characteristic.
Denote by θe (t) the phase difference
θe = θr (t) − θc (t).
Then the following dynamical system describes PLL be-
havior

the VCO frequency is usually taken as a function of the

VCO input g(t) as ẋ = Ax + bφ(θe ),
x(0) = x0 , θe (0) = φ0 .
θ̇e = ωe + gv (c∗ x).

θ̇c (t) = ωc (t) = ωc + gv g(t) Here ωe = ωr − ωc ; ωr is a frequency of reference

oscillator (we assume that ωr is constant).
where gv is the sensitivity of the VCO and is expressed in
Hz / V; ωc is a free-running frequency of VCO.
Example Consider sinusoidal signals
The loop filter can be described by system of linear dif-
ferential equations
xc (θ(t)) = Ac sin(θc (t)), xr (θr (t)) = Ar cos(θr (t))
ẋ = Ax + bxm (t), and simple one-pole RC circuit as a filter. The time-
x(0) = x0 ,
xf (t) = c∗ x, domain model takes the form
where xm (t) is an input of the filter, xf (t) is an output of
the filter, A is n -by- n matrix, x ∈ Rn , b ∈ Rn , c ∈ 1 1
Rn , . x0 ∈ Rn represents an initial state of the filter. ẋ = − x+ Ac Ar sin(θr (t)) cos(θc (t)),
RC RC
The star symbol is a conjugate transpose.
Hence the following system describes PLL θ̇c = ωc + gv (c∗ x)

PD characteristics for this signals is equal[16] to

ẋ = Ax + bxr (θr (t))xc (θc (t)),
x(0) = x0 , θc (0) = φ0 .
θ̇c = ωc + gv (c∗ x) Ac Ar
φ(θr − θc ) = sin(θr − θc )
where φ0 is an initial phase shift. 2
Hence the phase domain model takes form
Phase domain model
1 1 Ac Ar
Consider the input of pll xr (θr (t)) and VCO output ẋ = − x+ sin(θr − θc ),
RC RC 2
xc (θc (t)) are high frequency signals. Then for any piece-
wise differentiable 2π -periodic functions xr (θ) and θ̇c = ωc + gv (c∗ x).
5.3. PHASE-LOCKED LOOP 135

This system of equations is equivalent to the equation of

mathematical pendulum Kp Kv
θo RC
= Kp Kv
θi s2 + s
RC + RC
θ̇c − ωc ωr − θ̇e − ωc
x= = , This is the form of a classic harmonic oscillator. The de-
gv c∗ gv c∗ nominator can be related to that of a second order system:
θ̈c
ẋ = ,
gv c∗
s2 + 2sζωn + ωn2
θr = ωr t + Ψ,
Where
θe = θr − θc ,
• ζ is the damping factor
θ̇e = θ̇r − θ̇c = ωr − θ̇c ,
• ωn is the natural frequency of the loop
1 1 Ac Ar ωc − ωr
θ̈e − θ̇e − sin θe = .
gv c∗ gv c∗ RC 2RC gv c∗ RC For the one-pole RC filter,

Linearized phase domain model √

Kp Kv
ωn =
Phase locked loops can also be analyzed as control sys- RC
tems by applying the Laplace transform. The loop re- 1
sponse can be written as: ζ= √
2 Kp Kv RC
The loop natural frequency is a measure of the response
θo Kp Kv F (s) time of the loop, and the damping factor is a measure of
= the overshoot and ringing. Ideally, the natural frequency
θi s + Kp Kv F (s)
should be high and the damping factor should be near
Where 0.707 (critical damping). With a single pole filter, it is
not possible to control the loop frequency and damping
• θo is the output phase in radians factor independently. For the case of critical damping,

• θi is the input phase in radians

1
• Kp is the phase detector gain in volts per radian RC =
2Kp Kv
• Kv is the VCO gain in radians per volt-second √
ωc = Kp Kv 2
• F (s) is the loop filter transfer function (dimension- A slightly more effective filter, the lag-lead filter includes
less) one pole and one zero. This can be realized with two
resistors and one capacitor. The transfer function for this
The loop characteristics can be controlled by inserting filter is
different types of loop filters. The simplest filter is a one-
pole RC circuit. The loop transfer function in this case
is: 1 + sCR2
F (s) =
1 + sC(R1 + R2 )
This filter has two time constants
1
F (s) =
1 + sRC
The loop response becomes: τ1 = C(R1 + R2 )
136 CHAPTER 5. COMMON INTEGRATED CIRCUITS

τ2 = CR2 shifts prop = 1/128; deriv = 64; for it=1:numiterations %

Simulate a local oscillator using a 16-bit counter phs =
Substituting above yields the following natural frequency
mod(phs + floor(freq/2^16), 2^16); ref = phs < 32768;
and damping factor
% Get the next digital value (0 or 1) of the signal to track
sig = tracksig(it); % Implement the phase-frequency
√ detector rst = ~(qsig & qref); % Reset the “flip-flop” of
Kp Kv the phase-frequency % detector when both signal and
ωn =
τ1 reference are high qsig = (qsig | (sig & ~lsig)) & rst; %
1 ωn τ2 Trigger signal flip-flop and leading edge of signal qref =
ζ= + (qref | (ref & ~lref)) & rst; % Trigger reference flip-flop
2ωn τ1 2
on leading edge of reference lref = ref; lsig = sig; %
The loop filter components can be calculated indepen- Store these values for next iteration (for edge detection)
dently for a given natural frequency and damping factor ersig = qref - qsig; % Compute the error signal (whether
frequency should increase or decrease) % Error signal is
given by one or the other flip flop signal % Implement
Kp Kv a pole-zero filter by proportional and derivative input to
τ1 =
ωn2 frequency filtered_ersig = ersig + (ersig - lersig) * deriv;
% Keep error signal for proportional output lersig =
2ζ 1
τ2 = − ersig; % Integrate VCO frequency using the error signal
ωn Kp Kv
freq = freq - 2^16 * filtered_ersig * prop; % Frequency
Real world loop filter design can be much more complex is tracked as a fixed-point binary fraction % Store the
e.g. using higher order filters to reduce various types or current VCO frequency vcofreq(1, it) = freq / 2^16; %
source of phase noise. (See the D Banerjee ref below) Store the error signal to show whether signal or reference
is higher frequency ervec(1, it) = ersig; end
Implementing a digital phase-locked loop in software
In this example, an array tracksig is assumed to contain
Digital phase locked loops can be implemented in hard- a reference signal to be tracked. The oscillator is imple-
ware, using integrated circuits such as a CMOS 4046. mented by a counter, with the most significant bit of the
However, with microcontrollers becoming faster, it may counter indicating the on/off status of the oscillator. This
make sense to implement a phase locked loop in software code simulates the two D-type flip-flops that comprise a
for applications that do not require locking onto signals phase-frequency comparator. When either the reference
in the MHz range or faster, such as precisely control- or signal has a positive edge, the corresponding flip-flop
ling motor speeds. Software implementation has several switches high. Once both reference and signal is high,
advantages including easy customization of the feedback both flip-flops are reset. Which flip-flop is high deter-
loop including changing the multiplication or division ra- mines at that instant whether the reference or signal leads
tio between the signal being tracked and the output oscil- the other. The error signal is the difference between these
lator. Furthermore, a software implementation is useful two flip-flop values. The pole-zero filter is implemented
to understand and experiment with. As an example of a by adding the error signal and its derivative to the filtered
phase-locked loop implemented using a phase frequency error signal. This in turn is integrated to find the oscillator
detector is presented in MATLAB, as this type of phase frequency.
detector is robust and easy to implement. This example In practice, one would likely insert other operations into
uses integer arithmetic rather than floating point, as such the feedback of this phase-locked loop. For example,
an example is likely more useful in practice. if the phase locked loop were to implement a frequency
% Initialize variables vcofreq = zeros(1, numiterations); multiplier, the oscillator signal could be divided in fre-
ervec = zeros(1, numiterations); % keep track of last quency before it is compared to the reference signal.
states of reference, signal, and error signal qsig = 0; qref
= 0; lref = 0; lsig = 0; lersig = 0; phs = 0; freq = 0; %
Loop filter constants (proportional and derivative) %
Currently powers of two to facilitate multiplication by
5.3. PHASE-LOCKED LOOP
5.3.8 See also
• Direct-digital synthesis
• Costas loop
• Kalman filter
• Direct conversion receiver
• Circle map - a simple mathematical model of the
phase-locked loop showing both mode-locking and
chaotic behavior.
• Carrier recovery
• Delay-locked loop (DLL)
• PLL multibit
• Shortt-Synchronome clock - slave pendulum phase-
locked to master (ca 1921).
5.3.9 References
[1] If the frequency is constant and the initial phase is zero,
then the phase of a sinusoid is proportional to time.
[2] Christiaan Huygens, Horologium Oscillatorium … (Paris,
France: F. Muguet, 1673), pages 18-19. From page 18:
extquotedbl … illudque accidit memoratu dignum, … brevi
tempore reduceret.” ( … and it is worth mentioning, since
with two clocks constructed in this form and which we sus-
pend in like manner, truly the cross beam is assigned two
fulcrums [i.e., two pendulum clocks were suspended from
the same wooden beam]; the motions of the pendulums
thus share the opposite swings between the two [clocks],
as the two clocks at no time move even a small distance,
and the sound of both can be heard clearly together al-
ways: for if the innermost part [of one of the clocks] is
disturbed with a little help, it will have been restored in a
short time by the clocks themselves.) English translation
provided by Ian Bruce’s translation of Horologium Oscil-
latorium … , pages 16-17.
[3] See:
• Lord Rayleigh, The Theory of Sound (London, Eng-
land: Macmillan, 1896), vol. 2. The synchroniza-
tion of organ pipes in opposed phase is mentioned
in §322c, pages 221-222.
• Lord Rayleigh (1907) “Acoustical notes — VII,”
Philosophical Magazine, 6th series, 13 : 316-333.
See “Tuning-forks with slight mutual influence,”
pages 322-323.
137
[4] See:
• Vincent (1919) “On some experiments in which
two neighbouring maintained oscillatory circuits af-
fect a resonating circuit,” Proceedings of the Physi-
cal Society of London, 32, pt. 2, 84-91.
• W. H. Eccles and J. H. Vincent, British Patent Spec-
ifications, 163 : 462 (17 Feb. 1920).
[5] E. V. Appleton (1923) “The automatic synchronization of
triode oscillators,” Proceedings of the Cambridge Philo-
sophical Society, 21 (Part III): 231-248. Available on-line
at: Internet Archive.
[6] Henri de Bellescize, “La réception synchrone,”
L'Onde Électrique (later: Revue de l'Electricité et de
l'Electronique), vol. 11, pages 230-240 (June 1932).
[7] See also: French patent no. 635,451 (filed: 6 Oc-
tober 1931; issued: 29 September 1932); and U.S.
patent “Synchronizing system,” no. 1,990,428 (filed: 29
September 1932; issued: 5 February 1935).
[8] Notes for a University of Guelph course describing the
PLL and early history, including an IC PLL tutorial
[9] “National Television Systems Committee Video Display
Signal IO”. Sxlist.com. Retrieved 2010-10-14.
[10] A. B. Grebene, H. R. Camenzind, “Phase Locking As A
New Approach For Tuned Integrated Circuits”, ISSCC
Digest of Technical Papers, pp. 100-101, Feb. 1969.
[11] Roland E. Best (2007). Phase-Locked Loops: Design,
Simulation and Applications (6th ed.). McGraw Hill.
ISBN 978-0-07-149375-8.
[12] M Horowitz, C. Yang, S. Sidiropoulos (1998-01-01).
“High-speed electrical signaling: overview and limita-
tions”. IEEE Micro.
[13] Typically, the reference sinewave drives a step recovery
diode circuit to make this impulse train. The resulting
impulse train drives a sample gate.
[14] G. A. Leonov, N. V. Kuznetsov, M. V. Yulda-
shev, R. V. Yuldashev (2011). “Computation of
Phase Detector Characteristics in Synchronization Sys-
tems”. Doklady Mathematics 84 (1): 586–590.
doi:10.1134/S1064562411040223.
[15] N.V. Kuznetsov, G.A. Leonov, M.V. Yuldashev, R.V.
Yuldashev (2011). “Analytical methods for compu-
tation of phase-detector characteristics and PLL de-
sign”. ISSCS 2011 – International Symposium on
Signals, Circuits and Systems, Proceedings: 7–10.
doi:10.1109/ISSCS.2011.5978639.
[16] A. J. Viterbi, Principles of Coherent Communication,
McGraw-Hill, New York, 1966
138 CHAPTER 5. COMMON INTEGRATED CIRCUITS

5.3.10 Further reading 5.4 Voltage regulator

• Banerjee, Dean (2006), PLL Performance, Simula-
tion and Design Handbook (4th ed.), National Semi-
conductor.

• Best, R. E. (2003), Phase-locked Loops: Design,

Simulation and Applications, McGraw-Hill, ISBN 0-
07-141201-8

• de Bellescize, Henri (June 1932), “La réception

Synchrone”, L'Onde Electrique 11: 230–240

• Dorf, Richard C. (1993), The Electrical Engineering

Handbook, Boca Raton: CRC Press, ISBN 0-8493-
0185-8
A popular three pin +12 V DC voltage regulator IC.
• Egan, William F. (1998), Phase-Lock Basics, John
Wiley & Sons. (provides useful Matlab scripts for A voltage regulator is designed to automatically main-
simulation) tain a constant voltage level. A voltage regulator may be
a simple extquotedblfeed-forward extquotedbl design or
• Egan, William F. (2000), Frequency Synthesis by may include negative feedback control loops. It may use
Phase Lock (2nd ed.), John Wiley and Sons. (pro- an electromechanical mechanism, or electronic compo-
vides useful Matlab scripts for simulation) nents. Depending on the design, it may be used to regu-
late one or more AC or DC voltages.
• Gardner, Floyd M. (2005), Phaselock Techniques
(3rd ed.), Wiley-Interscience, ISBN 978-0-471- Electronic voltage regulators are found in devices such
43063-6 as computer power supplies where they stabilize the DC
voltages used by the processor and other elements. In au-
• Klapper, J.; Frankle, J. T. (1972), Phase-Locked tomobile alternators and central power station generator
and Frequency-Feedback Systems, Academic Press. plants, voltage regulators control the output of the plant.
(FM Demodulation) In an electric power distribution system, voltage regula-
tors may be installed at a substation or along distribution
• Kundert, Ken (August 2006), Predicting the Phase lines so that all customers receive steady voltage indepen-
Noise and Jitter of PLL-Based Frequency Synthesiz- dent of how much power is drawn from the line.
ers (4g ed.), Designer’s Guide Consulting, Inc.

• Liu, Mingliang (February 21, 2006), Build a 1.5-V 5.4.1 Measures of regulator quality
2.4-GHz CMOS PLL, Wireless Net Design Line. An
article on designing a standard PLL IC for Bluetooth The output voltage can only be held roughly constant; the
applications. regulation is specified by two measurements:
• Wolaver, Dan H. (1991), Phase-Locked Loop Circuit
Design, Prentice Hall, ISBN 0-13-662743-9 • load regulation is the change in output voltage for
a given change in load current (for example: “typ-
• Signal processing and system aspects of all-digital ically 15 mV, maximum 100 mV for load currents
phase-locked loops (ADPLLs) between 5 mA and 1.4 A, at some specified temper-
ature and input voltage”).
• Phase-Locked Loop Tutorial, PLL
• line regulation or input regulation is the degree
• Temporal-code to rate-code conversion by neuronal to which output voltage changes with input (supply)
phase-locked loops. Neural Comput 10:597-650., voltage changes - as a ratio of output to input change
1998 (for example “typically 13 mV/V”), or the output
5.4. VOLTAGE REGULATOR 139

voltage change over the entire specified input volt- the load transient) or input voltage (called the line
age range (for example “plus or minus 2% for input transient) occurs. Some regulators will tend to os-
voltages between 90 V and 260 V, 50-60 Hz”). cillate or have a slow response time which in some
cases might lead to undesired results. This value
Other important parameters are: is different from the regulation parameters, as that
is the stable situation definition. The transient re-
sponse shows the behaviour of the regulator on a
• Temperature coefficient of the output voltage is change. This data is usually provided in the tech-
the change with temperature (perhaps averaged over nical documentation of a regulator and is also de-
a given temperature range). pendent on output capacitance.
• Initial accuracy of a voltage regulator (or simply • Mirror-image insertion protection means that a
“the voltage accuracy”) reflects the error in output regulator is designed for use when a voltage, usually
voltage for a fixed regulator without taking into ac- not higher than the maximum input voltage of the
count temperature or aging effects on output accu- regulator, is applied to its output pin while its input
racy. terminal is at a low voltage, volt-free or grounded.
• Dropout voltage is the minimum difference be- Some regulators can continuously withstand this sit-
tween input voltage and output voltage for which the uation; others might only manage it for a limited
regulator can still supply the specified current. A time such as 60 seconds, as usually specified in the
low drop-out (LDO) regulator is designed to work datasheet. This situation can occur when a three ter-
well even with an input supply only a volt or so above minal regulator is incorrectly mounted for example
the output voltage. The input-output differential at on a PCB, with the output terminal connected to the
which the voltage regulator will no longer maintain unregulated DC input and the input connected to the
regulation is the dropout voltage. Further reduction load. Mirror-image insertion protection is also im-
in input voltage will result in reduced output voltage. portant when a regulator circuit is used in battery
This value is dependent on load current and junction charging circuits, when external power fails or is not
temperature. turned on and the output terminal remains at battery
voltage.
• Absolute maximum ratings are defined for regula-
tor components, specifying the continuous and peak
output currents that may be used (sometimes inter- 5.4.2 Electronic voltage regulators
nally limited), the maximum input voltage, maxi-
mum power dissipation at a given temperature, etc. A simple voltage regulator can be made from a resistor
in series with a diode (or series of diodes). Due to the
• Output noise (thermal white noise) and output logarithmic shape of diode V-I curves, the voltage across
dynamic impedance may be specified as graphs the diode changes only slightly due to changes in cur-
versus frequency, while output ripple noise (mains rent drawn or changes in the input. When precise voltage
“hum” or switch-mode “hash” noise) may be given control and efficiency are not important, this design may
as peak-to-peak or RMS voltages, or in terms of work fine.
their spectra.
Feedback voltage regulators operate by comparing the
• Quiescent current in a regulator circuit is the cur- actual output voltage to some fixed reference voltage.
rent drawn internally, not available to the load, nor- Any difference is amplified and used to control the regu-
mally measured as the input current while no load is lation element in such a way as to reduce the voltage error.
connected (and hence a source of inefficiency; some This forms a negative feedback control loop; increasing
linear regulators are, surprisingly, more efficient at the open-loop gain tends to increase regulation accuracy
very low current loads than switch-mode designs be- but reduce stability. (Stability is avoidance of oscillation,
cause of this). or ringing, during step changes.) There will also be a
trade-off between stability and the speed of the response
• Transient response is the reaction of a regulator to changes. If the output voltage is too low (perhaps due
when a (sudden) change of the load current (called to input voltage reducing or load current increasing), the
140
regulation element is commanded, up to a point, to pro-
duce a higher output voltage–by dropping less of the input
voltage (for linear series regulators and buck switching
regulators), or to draw input current for longer periods
(boost-type switching regulators); if the output voltage is
too high, the regulation element will normally be com-
manded to produce a lower voltage. However, many reg-
ulators have over-current protection, so that they will en-
tirely stop sourcing current (or limit the current in some
way) if the output current is too high, and some regula-
tors may also shut down if the input voltage is outside a
given range (see also: crowbar circuits).
5.4.3 Electromechanical regulators
Circuit design for a simple electromechanical voltage regulator.
A voltage stabilizer using electromechanical relays for switching. discharging back into the generator and attempting to run
In electromechanical regulators, voltage regulation is eas- matically perform this function so that a specific relay is
ily accomplished by coiling the sensing wire to make an not required; this appreciably simplified the regulator de-
CHAPTER 5. COMMON INTEGRATED CIRCUITS
Graph of voltage output on a time scale.
electromagnet. The magnetic field produced by the cur-
rent attracts a moving ferrous core held back under spring
tension or gravitational pull. As voltage increases, so does
the current, strengthening the magnetic field produced by
the coil and pulling the core towards the field. The mag-
net is physically connected to a mechanical power switch,
which opens as the magnet moves into the field. As volt-
age decreases, so does the current, releasing spring ten-
sion or the weight of the core and causing it to retract.
This closes the switch and allows the power to flow once
more.
If the mechanical regulator design is sensitive to small
voltage fluctuations, the motion of the solenoid core can
be used to move a selector switch across a range of re-
sistances or transformer windings to gradually step the
output voltage up or down, or to rotate the position of a
moving-coil AC regulator.
Early automobile generators and alternators had a me-
chanical voltage regulator using one, two, or three relays
and various resistors to stabilize the generator’s output at
slightly more than 6 or 12 V, independent of the engine's
rpm or the varying load on the vehicle’s electrical sys-
tem. Essentially, the relay(s) employed pulse width mod-
ulation to regulate the output of the generator, controlling
the field current reaching the generator (or alternator) and
in this way controlling the output voltage produced.
The regulators used for DC generators (but not alterna-
tors) also disconnect the generator when it was not pro-
ducing electricity, thereby preventing the battery from
it as a motor. The rectifier diodes in an alternator auto-
5.4. VOLTAGE REGULATOR 141

sign. voltage to a predetermined setpoint. When two or more

More modern designs now use solid state technology generators are powering the same system (parallel opera-
(transistors) to perform the same function that the relays tion) the AVR receives information from more generators
perform in electromechanical regulators. to match all output.

Electromechanical regulators are used for mains voltage

stabilisation — see AC voltage stabilizers below. 5.4.5 AC voltage stabilizers

Coil-rotation AC voltage regulator

5.4.4 Automatic voltage regulator

Basic design principle and circuit diagram for the rotating-coil

AC voltage regulator.

This is an older type of regulator used in the 1920s that

Voltage regulator for generators.
To control the output of generators (as seen in ships and
power stations, or on oil rigs, greenhouses and emergency
power systems) automatic voltage regulators are used.
This is an active system. While the basic principle is the
same, the system itself is more complex. An automatic
voltage regulator (or AVR for short) consist of several
components such as diodes, capacitors, resistors and po-
tentiometers or even microcontrollers, all placed on a cir-
cuit board. This is then mounted near the generator and
connected with several wires to measure and adjust the
generator.
uses the principle of a fixed-position field coil and a sec-
ond field coil that can be rotated on an axis in parallel
with the fixed coil, similar to a variocoupler.
When the movable coil is positioned perpendicular to the
fixed coil, the magnetic forces acting on the movable coil
balance each other out and voltage output is unchanged.
Rotating the coil in one direction or the other away from
the center position will increase or decrease voltage in the
secondary movable coil.
This type of regulator can be automated via a servo con-
trol mechanism to advance the movable coil position in
order to provide voltage increase or decrease. A braking
mechanism or high ratio gearing is used to hold the ro-
tating coil in place against the powerful magnetic forces
acting on the moving coil.

How an AVR works: In the first place the AVR moni-

tors the output voltage and controls the input voltage for
the exciter of the generator. By increasing or decreasing
the generator control voltage, the output voltage of the
generator increases or decreases accordingly. The AVR
calculates how much voltage has to be sent to the exciter
numerous times a second, therefore stabilizing the output
Electromechanical
Electromechanical regulators called voltage stabilizers or
tap-changers, have also been used to regulate the voltage
on AC power distribution lines. These regulators operate
by using a servomechanism to select the appropriate tap
142
Magnetic mains regulator
on an autotransformer with multiple taps, or by moving
the wiper on a continuously variable auto transfomer. If
the output voltage is not in the acceptable range, the ser-
vomechanism switches the tap, changing the turns ratio
of the transformer, to move the secondary voltage into
the acceptable region. The controls provide a dead band
wherein the controller will not act, preventing the con-
troller from constantly adjusting the voltage (“hunting”)
as it varies by an acceptably small amount.
PWM Static Voltage Regulator
This is the latest technology of voltage regulation to pro-
vide real-time control of voltage fluctuation, sag, surge
and also to control other power quality issues such as
spikes & EMI/RFI electrical noises. This uses an IGBT
regulator engine generating PWM AC voltage at high
switching frequency. This AC PWM wave is superim-
posed on the main incoming wave through a buck-boost
transformer, to provide precisely regulated AC voltage.
The regulation in this technology is instantaneous, thus
making it suitable for electronic machines which need
precise regulated power.
Constant-voltage transformer
The ferroresonant transformer, ferroresonant regu- Output voltage varies about 1.2% for every 1% change
lator or constant-voltage transformer is a type of sat- in supply frequency. For example, a 2 Hz change in gen-
CHAPTER 5. COMMON INTEGRATED CIRCUITS
urating transformer used as a voltage regulator. These
transformers use a tank circuit composed of a high-
voltage resonant winding and a capacitor to produce a
nearly constant average output voltage with a varying in-
put current or varying load. The circuit has a primary
on one side of a magnet shunt and the tuned circuit coil
and secondary on the other side. The regulation is due to
magnetic saturation in the section around the secondary.
The ferroresonant approach is attractive due to its lack of
active components, relying on the square loop saturation
characteristics of the tank circuit to absorb variations in
average input voltage. Saturating transformers provide a
simple rugged method to stabilize an AC power supply.
Older designs of ferroresonant transformers had an out-
put with high harmonic content, leading to a distorted
output waveform. Modern devices are used to construct a
perfect sine wave. The ferroresonant action is a flux lim-
iter rather than a voltage regulator, but with a fixed supply
frequency it can maintain an almost constant average out-
put voltage even as the input voltage varies widely.
The ferroresonant transformers, which are also known as
Constant Voltage Transformers (CVTs) or ferros, are also
good surge suppressors, as they provide high isolation and
inherent short-circuit protection.
A ferroresonant transformer can operate with an input
voltage range ±40% or more of the nominal voltage.
Output power factor remains in the range of 0.96 or
higher from half to full load.
Because it regenerates an output voltage waveform, out-
put distortion, which is typically less than 4%, is indepen-
dent of any input voltage distortion, including notching.
Efficiency at full load is typically in the range of 89%
to 93%. However, at low loads, efficiency can drop be-
low 60%. The current-limiting capability also becomes
a handicap when a CVT is used in an application with
moderate to high inrush current like motors, transform-
ers or magnets. In this case, the CVT has to be sized to
accommodate the peak current, thus forcing it to run at
low loads and poor efficiency.
Minimum maintenance is required, as transformers and
capacitors can be very reliable. Some units have included
redundant capacitors to allow several capacitors to fail
between inspections without any noticeable effect on the
device’s performance.
5.4. VOLTAGE REGULATOR 143

erator frequency, which is very large, results in an output to dissipate the excess energy. The power supply is de-
voltage change of only 4%, which has little effect for most signed to only supply a maximum amount of current that
loads. is within the safe operating capability of the shunt regu-
It accepts 100% single-phase switch-mode power supply lating device.
loading without any requirement for derating, including If the stabilizer must provide more power, the shunt reg-
all neutral components. ulator output is only used to provide the standard voltage
Input current distortion remains less than 8% THD even reference for the electronic device, known as the voltage
when supplying nonlinear loads with more than 100% stabilizer. The voltage stabilizer is the electronic device,
current THD. able to deliver much larger currents on demand.

Drawbacks of CVTs are their larger size, audible hum-

ming sound, and the high heat generation caused by sat- 5.4.7 Active regulators
uration.
Active regulators employ at least one active (amplifying)
component such as a transistor or operational amplifier.
Commercial use Shunt regulators are often (but not always) passive and
simple, but always inefficient because they (essentially)
Voltage regulators or stabilizers are used to compensate dump the excess current not needed by the load. When
for voltage fluctuations in mains power. Large regula- more power must be supplied, more sophisticated circuits
tors may be permanently installed on distribution lines. are used. In general, these active regulators can be di-
Small portable regulators may be plugged in between sen- vided into several classes:
sitive equipment and a wall outlet. Automatic voltage
regulators are used on generator sets on ships, in emer-
gency power supplies, on oil rigs, etc. to stabilize fluc- • Linear series regulators
tuations in power demand. For example, when a large • Switching regulators
machine is turned on, the demand for power is suddenly
a lot higher. The voltage regulator compensates for the • SCR regulators
change in load. Commercial voltage regulators normally
operate on a range of voltages, for example 150–240 V
or 90–280 V. Servo stabilizers are also manufactured and Linear regulators
used widely in spite of the fact that they are obsolete and
use out-dated technology. Main article: Linear regulator

Voltage regulators are used in devices like air condition-

ers, refrigerators, televisions etc. in order to protect them
from fluctuating input voltage. The major problem faced
is the use of relays in voltage regulators. Relays create
sparks which result in faults in the product.
5.4.6 DC voltage stabilizers
Many simple DC power supplies regulate the voltage us-
ing either series or shunt regulators, but most apply a
voltage reference using a shunt regulator such as a Zener
diode, avalanche breakdown diode, or voltage regulator
tube. Each of these devices begins conducting at a speci-
fied voltage and will conduct as much current as required
to hold its terminal voltage to that specified voltage by
diverting excess current from a non-ideal power source
to ground, often through a relatively low-value resistor
Linear regulators are based on devices that operate in
their linear region (in contrast, a switching regulator is
based on a device forced to act as an on/off switch). In the
past, one or more vacuum tubes were commonly used as
the variable resistance. Modern designs use one or more
transistors instead, perhaps within an Integrated Circuit.
Linear designs have the advantage of very “clean” output
with little noise introduced into their DC output, but are
most often much less efficient and unable to step-up or
invert the input voltage like switched supplies. All lin-
ear regulators require a higher input than the output. If
the input voltage approaches the desired output voltage,
the regulator will “drop out”. The input to output voltage
differential at which this occurs is known as the regula-
tor’s drop-out voltage. Special “low drop-out” regulators
allow an input voltage that can be much lower, often ap-
proaching or equal to that of the desired output.
144 CHAPTER 5. COMMON INTEGRATED CIRCUITS

Entire linear regulators are available as integrated cir- • Switching regulators are required when the only
cuits. These chips come in either fixed or adjustable volt- power supply is a DC voltage, and a higher output
age types. voltage is required.

• At high levels of power (above a few watts), switch-

Switching regulators ing regulators are cheaper (for example, the cost of
removing heat generated is less).
Main article: Switched-mode power supply

SCR regulators
Switching regulators rapidly switch a series device on and
off. The duty cycle of the switch sets how much charge
Regulators powered from AC power circuits can use
is transferred to the load. This is controlled by a simi-
silicon controlled rectifiers (SCRs) as the series device.
lar feedback mechanism as in a linear regulator. Because
Whenever the output voltage is below the desired value,
the series element is either fully conducting, or switched
the SCR is triggered, allowing electricity to flow into the
off, it dissipates almost no power; this is what gives the
load until the AC mains voltage passes through zero (end-
switching design its efficiency. Switching regulators are
ing the half cycle). SCR regulators have the advantages
also able to generate output voltages which are higher
of being both very efficient and very simple, but because
than the input, or of opposite polarity — something not
they can not terminate an on-going half cycle of conduc-
possible with a linear design.
tion, they are not capable of very accurate voltage regula-
Like linear regulators, nearly complete switching regula- tion in response to rapidly changing loads. An alternative
tors are also available as integrated circuits. Unlike lin- is the SCR shunt regulator which uses the regulator out-
ear regulators, these usually require one external compo- put as a trigger, both series and shunt designs are noisy,
nent: an inductor that acts as the energy storage element. but powerful, as the device has a low on resistance.
(Large-valued inductors tend to be physically large rela-
tive to almost all other kinds of componentry, so they are
rarely fabricated within integrated circuits and IC regula- Combination (hybrid) regulators
tors — with some exceptions.[1][2] )
Many power supplies use more than one regulating
method in series. For example, the output from a switch-
Comparing linear vs. switching regulators ing regulator can be further regulated by a linear regula-
tor. The switching regulator accepts a wide range of in-
The two types of regulators have their different advan- put voltages and efficiently generates a (somewhat noisy)
tages: voltage slightly above the ultimately desired output. That
is followed by a linear regulator that generates exactly the
• Linear regulators are best when low output noise desired voltage and eliminates nearly all the noise gener-
(and low RFI radiated noise) is required ated by the switching regulator. Other designs may use
an SCR regulator as the “pre-regulator”, followed by an-
• Linear regulators are best when a fast response to other type of regulator. An efficient way of creating a
input and output disturbances is required. variable-voltage, accurate output power supply is to com-
• At low levels of power, linear regulators are cheaper bine a multi-tapped transformer with an adjustable linear
and occupy less printed circuit board space. post-regulator.

• Switching regulators are best when power efficiency

is critical (such as in portable computers), except
that linear regulators are more efficient in a small
number of cases (such as a 5 V microprocessor of-
ten in “sleep” mode fed from a 6 V battery, if the
complexity of the switching circuit and the junction
capacitance charging current means a high quiescent
current in the switching regulator).
5.4.8 Example linear regulators
Transistor regulator
In the simplest case a common collector transistor (emit-
ter follower) is used with the base of the regulating tran-
sistor connected directly to the voltage reference:
5.4. VOLTAGE REGULATOR 145

U
Regulator with an operational amplifier
CE
+U +U
In Out The stability of the output voltage can be significantly in-
Q creased by using an operational amplifier:
Rv
U
BE
U R U
Out L CE
+U +U
In Out
Rv Q R1
U Dz
Z OA
U
+ BE
R2
U R
_ Out L

U Dz R3
Z

A simple transistor regulator will provide a relatively con-

stant output voltage, Uout, for changes in the voltage of
the power source, Uin, and for changes in load, RL, pro- In this case, the operational amplifier drives the transis-
vided that Uin exceeds Uout by a sufficient margin, and tor with more current if the voltage at its inverting input
that the power handling capacity of the transistor is not drops below the output of the voltage reference at the
exceeded. non-inverting input. Using the voltage divider (R1, R2
and R3) allows choice of the arbitrary output voltage be-
The output voltage of the stabilizer is equal to the zener
tween U and Uᵢ .
diode voltage less the base–emitter voltage of the transis-
tor, UZ − UBE, where UBE is usually about 0.7 V for a
silicon transistor, depending on the load current. If the 5.4.9 See also
output voltage drops for any external reason, such as an
increase in the current drawn by the load (causing a de- • Constant current regulator
crease in the Collector-Emitter junction voltage to ob-
serve KVL), the transistor’s base–emitter voltage (UBE) • DC-to-DC converter
increases, turning the transistor on further and delivering
• Third brush dynamo
more current to increase the load voltage again.
Rv provides a bias current for both the zener diode and • Voltage regulator module
the transistor. The current in the diode is minimum
when the load current is maximum. The circuit designer
must choose a minimum voltage that can be tolerated 5.4.10 References
across Rv, bearing in mind that the higher this voltage
[1] Texas Instruments LM2825 Integrated Power Supply 1A
requirement is, the higher the required input voltage, DC-DC Converter, retrieved 2010-09-19
Uin, and hence the lower the efficiency of the regula-
tor. On the other hand, lower values of Rv lead to higher [2] Linear Technology μModule Regulators, retrieved 2011-
power dissipation in the diode and to inferior regulator 03-08
characteristics.[3] [3] Alley, Charles; Atwood, Kenneth (1973). Electronic En-
VRmin gineering. New York and London: John Wiley & Sons. p.
Rv = IDmin +ILmax /(hF E +1)
534. ISBN 0-471-02450-3.
where VR min is the minimum voltage to be maintained
across Rv
ID min is the minimum current to be maintained through 5.4.11 Further reading
the zener diode
IL max is the maximum design load current • Linear & Switching Voltage Regulator Hand-
hFE is the forward current gain of the transistor, ICollec- book; ON Semiconductor; 118 pages; 2002;
tor / IBase[3] HB206/D.(Free PDF download)
146
5.5 Comparator
For other uses, see Comparator (disambiguation).
In electronics, a comparator is a device that compares
two voltages or currents and outputs a digital signal indi-
cating which is larger. It has two analog input terminals
V+ and V− and one binary digital output Vo . The output
is ideally
{ An operational amplifier (op-amp) has a well balanced
1, if V+ > V−
Vo =
0, if V+ < V−
A comparator consists of a specialized high-gain
differential amplifier. They are commonly used in de-
vices that measure and digitize analog signals, such as
analog-to-digital converters (ADCs), as well as relaxation
oscillators.
5.5.1 Differential Voltage
The differential voltages must stay within the limits spec-
ified by the manufacturer. Early integrated comparators,
like the LM111 family, and certain high-speed compara-
tors like the LM119 family, require differential voltage
ranges substantially lower than the power supply volt-
ages (±15 V vs. 36 V).[1] Rail-to-rail comparators allow
any differential voltages within the power supply range.
When powered from a bipolar (dual rail) supply,
VS− ≤ V+ , V− ≤ VS+
or, when powered from a unipolar TTL/CMOS power
supply:
0 ≤ V+ , V− ≤ Vcc
Specific rail-to-rail comparators with p-n-p input transis-
tors, like the LM139 family, allow input potential to drop
0.3 volts below the negative supply rail, but do not allow
it to rise above the positive rail.[2] Specific ultra-fast com-
parators, like the LMH7322, allow input signal to swing
below the negative rail and above the positive rail, al-
though by a narrow margin of only 0.2 V.[3] Differential
input voltage (the voltage between two inputs) of a mod-
ern rail-to-rail comparator is usually limited only by the
full swing of power supply.
CHAPTER 5. COMMON INTEGRATED CIRCUITS
5.5.2 Op-amp voltage comparator
1 V
Vout
2 V
A simple op-amp comparator
difference input and a very high gain. This parallels the
characteristics of comparators and can be substituted in
applications with low-performance requirements.[4]
In theory, a standard op-amp operating in open-loop con-
figuration (without negative feedback) may be used as a
low-performance comparator. When the non-inverting
input (V+) is at a higher voltage than the inverting input
(V-), the high gain of the op-amp causes the output to sat-
urate at the highest positive voltage it can output. When
the non-inverting input (V+) drops below the inverting in-
put (V-), the output saturates at the most negative voltage
it can output. The op-amp’s output voltage is limited by
the supply voltage. An op-amp operating in a linear mode
with negative feedback, using a balanced, split-voltage
power supply, (powered by ± VS) has its transfer function
typically written as: Vout = Ao (V1 −V2 ) . However, this
equation may not be applicable to a comparator circuit
which is non-linear and operates open-loop (no negative
feedback)
In practice, using an operational amplifier as a compara-
tor presents several disadvantages as compared to using
a dedicated comparator:[5]
1. Op-amps are designed to operate in the linear mode
with negative feedback. Hence, an op-amp typically
has a lengthy recovery time from saturation. Almost
all op-amps have an internal compensation capaci-
tor which imposes slew rate limitations for high fre-
quency signals. Consequently an op-amp makes a
sloppy comparator with propagation delays that can
be as long as tens of microseconds.
2. Since op-amps do not have any internal hysteresis,
an external hysteresis network is always necessary
for slow moving input signals.
3. The quiescent current specification of an op-amp is
valid only when the feedback is active. Some op-
5.5. COMPARATOR 147

amps show an increased quiescent current when the analog to digital converter). If there is a fixed voltage
inputs are not equal. source from, for example, a DC adjustable device in the
signal path, a comparator is just the equivalent of a cas-
4. A comparator is designed to produce well lim- cade of amplifiers. When the voltages are nearly equal,
ited output voltages that easily interface with digi- the output voltage will not fall into one of the logic lev-
tal logic. Compatibility with digital logic must be els, thus analog signals will enter the digital domain with
verified while using an op-amp as a comparator. unpredictable results. To make this range as small as pos-
sible, the amplifier cascade is high gain. The circuit con-
5. Some multiple-section op-amps may exhibit ex-
sists of mainly Bipolar transistors. For very high frequen-
treme channel-channel interaction when used as
cies, the input impedance of the stages is low. This re-
comparators.
duces the saturation of the slow, large P-N junction bipo-
6. Many op-amps have back to back diodes between lar transistors that would otherwise lead to long recovery
their inputs. Op-amp inputs usually follow each times. Fast small Schottky diodes, like those found in bi-
other so this is fine. But comparator inputs are not nary logic designs, improve the performance significantly
usually the same. The diodes can cause unexpected though the performance still lags that of circuits with am-
current through inputs. plifiers using analog signals. Slew rate has no meaning
for these devices. For applications in flash ADCs the dis-
tributed signal across eight ports matches the voltage and
5.5.3 Working current gain after each amplifier, and resistors then be-
have as level-shifters.
The LM339 accomplishes this with an open collector out-
put. When the inverting input is at a higher voltage than
the non inverting input, the output of the comparator con-
nects to the negative power supply. When the non invert-
ing input is higher than the inverting input, the output
is 'floating' (has a very high impedance to ground). The
gain of op amp as comparator is given by this equation
V(out)=V(in)*A
With a pull-up resistor and a 0 to +5 V power supply, the
output takes on the voltages 0 or +5 and can interface with
TTL logic:

Vout ≤ VCC when V+ ≥ V− else 0 .

5.5.4 Key specifications

Several voltage comparator ICs While it is easy to understand the basic task of a compara-
tor, that is, comparing two voltages or currents, several
A dedicated voltage comparator will generally be faster parameters must be considered while selecting a suitable
than a general-purpose operational amplifier pressed into comparator:
service as a comparator. A dedicated voltage comparator
may also contain additional features such as an accurate,
Speed and power
internal voltage reference, an adjustable hysteresis and a
clock gated input. While in general comparators are “fast,” their circuits
A dedicated voltage comparator chip such as LM339 is are not immune to the classic speed-power tradeoff.
designed to interface with a digital logic interface (to a High speed comparators use transistors with larger as-
TTL or a CMOS). The output is a binary state often pect ratios and hence also consume more power.[6] De-
used to interface real world signals to digital circuitry (see pending on the application, select either a comparator
148 CHAPTER 5. COMMON INTEGRATED CIRCUITS

with high speed or one that saves power. For exam- Output type
ple, nano-powered comparators in space-saving chip-
scale packages (UCSP), DFN or SC70 packages such
as MAX9027, LTC1540, LPV7215, MAX9060 and
MCP6541 are ideal for ultra-low-power, portable appli-
cations. Likewise if a comparator is needed to imple-
ment a relaxation oscillator circuit to create a high speed
clock signal then comparators having few nano seconds of
propagation delay may be suitable. ADCMP572 (CML
output), LMH7220 (LVDS Output), MAX999 (CMOS
output / TTL output), LT1719 (CMOS output / TTL out-
put), MAX9010 (TTL output), and MAX9601 (PECL
output) are examples of some good high speed compara-
tors.

Hysteresis
A comparator normally changes its output state when the
voltage between its inputs crosses through approximately
zero volts. Small voltage fluctuations due to noise, al-
ways present on the inputs, can cause undesirable rapid
changes between the two output states when the input
voltage difference is near zero volts. To prevent this out-
put oscillation, a small hysteresis of a few millivolts is in-
tegrated into many modern comparators.[7] For example,
the LTC6702, MAX9021 and MAX9031 have internal
hysteresis desensitizing them from input noise. In place
of one switching point, hysteresis introduces two: one for
rising voltages, and one for falling voltages. The differ-
ence between the higher-level trip value (VTRIP+) and
the lower-level trip value (VTRIP-) equals the hysteresis
voltage (VHYST).
If the comparator does not have internal hysteresis or
if the input noise is greater than the internal hysteresis
then an external hysteresis network can be built using
positive feedback from the output to the non-inverting
input of the comparator. The resulting Schmitt trig-
ger circuit gives additional noise immunity and a cleaner
output signal. Some comparators such as LMP7300,
LTC1540, MAX931, MAX971 and ADCMP341 also
provide the hysteresis control through a separate hystere-
sis pin. These comparators make it possible to add a pro-
grammable hysteresis without feedback or complicated
equations. Using a dedicated hysteresis pin is also conve-
nient if the source impedance is high since the inputs are
isolated from the hysteresis network.[8] When hysteresis
is added then a comparator cannot resolve signals within
the hysteresis band.
A Low Power CMOS Clocked Comparator
Because comparators have only two output states, their
outputs are near zero or near the supply voltage. Bipolar
rail-to-rail comparators have a common-emitter output
that produces a small voltage drop between the output and
each rail. That drop is equal to the collector-to-emitter
voltage of a saturated transistor. When output currents
are light, output voltages of CMOS rail-to-rail compara-
tors, which rely on a saturated MOSFET, range closer to
the rails than their bipolar counterparts.[9]
On the basis of outputs, comparators can also be clas-
sified as open drain or push–pull. Comparators with an
open-drain output stage use an external pull up resistor to
a positive supply that defines the logic high level. Open
drain comparators are more suitable for mixed-voltage
system design. Since the output is high impedance for
logic level high, open drain comparators can also be used
to connect multiple comparators on to a single bus. Push
pull output does not need a pull up resistor and can also
source current unlike an open drain output.
Internal reference
The most frequent application for comparators is the
comparison between a voltage and a stable reference.
Most comparator manufacturers also offer comparators
in which a reference voltage is integrated on to the chip.
Combining the reference and comparator in one chip
not only saves space, but also draws less supply cur-
rent than a comparator with an external reference.[9]
5.5. COMPARATOR
ICs with wide range of references are available such as
MAX9062(200 mV reference), LT6700(400 mV ref-
erence), ADCMP350 (600 mV reference), MAX9025
(1.236 V reference), MAX9040 (2.048 V reference),
TLV3012 (1.24 V reference) and TSM109 (2.5 V ref-
erence).
Continuous versus clocked
A continuous comparator will output either a “1” or a “0”
any time a high or low signal is applied to its input and
will change quickly when the inputs are updated. How-
ever, many applications only require comparator outputs
at certain instances, such as in A/D converters and mem-
ory. By only strobing a comparator at certain intervals,
higher accuracy and lower power can be achieved with a
clocked (or dynamic) comparator structure, also called a
latched comparator. Often latched comparators employ
strong positive feedback for a “regeneration phase” when
a clock is high, and have a “reset phase” when the clock
is low.[10] This is in contrast to a continuous compara-
tor, which can only employ weak positive feedback since
there is no reset period.
5.5.5 Applications
Main article: Comparator applications
Null detectors
A null detector is one that functions to identify when a
given value is zero. Comparators can be a type of am-
plifier distinctively for null comparison measurements. It
is the equivalent to a very high gain amplifier with well-
balanced inputs and controlled output limits. The circuit
compares the two input voltages, determining the larger.
The inputs are an unknown voltage and a reference volt-
age, usually referred to as vᵤ and vᵣ. A reference volt-
age is generally on the non-inverting input (+), while vᵤ
is usually on the inverting input (−). (A circuit diagram
would display the inputs according to their sign with re-
spect to the output when a particular input is greater than
the other.) The output is either positive or negative, for
example ±12 V. In this case, the idea is to detect when
there is no difference between in the input voltages. This
gives the identity of the unknown voltage since the refer-
ence voltage is known.
149
When using a comparator as a null detector, there are
limits as to the accuracy of the zero value measurable.
Zero output is given when the magnitude of the difference
in the voltages multiplied by the gain of the amplifier is
less than the voltage limits. For example, if the gain of the
amplifier is 106 , and the voltage limits are ±6 V, then no
output will be given if the difference in the voltages is less
than 6 μV. One could refer to this as a sort of uncertainty
in the measurement.[11]
Zero-crossing detectors
For this type of detector, a comparator detects each time
an ac pulse changes polarity. The output of the compara-
tor changes state each time the pulse changes its polarity,
that is the output is HI (high) for a positive pulse and LO
(low) for a negative pulse squares the input signal.[12]
Relaxation oscillator
A comparator can be used to build a relaxation oscillator.
It uses both positive and negative feedback. The posi-
tive feedback is a Schmitt trigger configuration. Alone,
the trigger is a bistable multivibrator. However, the slow
negative feedback added to the trigger by the RC circuit
causes the circuit to oscillate automatically. That is, the
addition of the RC circuit turns the hysteretic bistable
multivibrator into an astable multivibrator.[13]
Level shifter
This circuit requires only a single comparator with
an open-drain output as in the LM393, TLV3011 or
MAX9028. The circuit provides great flexibility in
choosing the voltages to be translated by using a suitable
pull up voltage. It also allows the translation of bipolar
±5 V logic to unipolar 3 V logic by using a comparator
like the MAX972.[9]
Analog-to-digital converters
When a comparator performs the function of telling if
an input voltage is above or below a given threshold, it
is essentially performing a 1-bit quantization. This func-
tion is used in nearly all analog to digital converters (such
as flash, pipeline, successive approximation, delta-sigma
modulation, folding, interpolating, dual-slope and others)
150 CHAPTER 5. COMMON INTEGRATED CIRCUITS

in combination with other devices to achieve a multi-bit [8] AN3616, Maxim Integrated Products, Adding Extra Hys-
quantization.[14] teresis to Comparators.

[9] AN886, Maxim Integrated Products, Selecting the Right

Window detectors Comparator.

[10] Pedro M. Figueiredo, João C. Vital (2009). Offset Reduc-

Comparators can also be used as window detectors. In tion Techniques in High-Speed Analog-to-Digital Convert-
a window detector, a comparator used to compare two ers: Analysis, Design and Tradeoffs. Springer. pp. 54–62.
voltages and determine whether a given input voltage is ISBN 978-1-4020-9715-7.
under voltage or over voltage.
[11] Malmstadt, Howard V.; Enke, Christie G.; Crouch, Stan-
ley R. (January 1981), Electronics and Instrumentation for
5.5.6 See also Scientists, The Benjamin/Cummings Publishing Co, pp.
108–110, ISBN 0-8053-6917-1
• Zero crossing threshold detector [12] Electronics and Instrumentation for Scientists. Malmstadt,
• Digital comparator Enke, and Crouch, The Benjamin/Cummings Publishing
Co., In., 1981, p.230.
• Current comparator
[13] Paul Horowitz and Winfield Hill: The Art of Electronics,
• Constant fraction discriminator Cambridge University Press, Second edition, Cambridge
1989, pp.284–285.
• Flash ADC
[14] Phillip Allen and Douglas Holberg: CMOS Analog Circuit
• Sorting network Design, Oxford University Press, Second edition, Oxford
2002.
This article incorporates public domain material from
the General Services Administration document “Federal 5.5.8 External links
Standard 1037C”.
• IC Comparator reference page at home.cogeco.ca
5.5.7 References • Comparator tutorial video with example circuits

[1] LM111/LM211/LM311 datasheet. Texas Instruments. • A Java based resistor value search tool for analysing
August 2003. Retrieved 2014-07-02. an inverting comparator circuit with hysteresis
[2] LM139/LM239/LM339/LM2901/LM3302 datasheet.
Texas Instruments. August 2012. Retrieved 2014-07-02.

[3] LMH7322 datasheet. Texas Instruments. March 2013.

Retrieved 2014-07-02.

[4] Malmstadt, Enke and Crouch, Electronics and Instrumen-

tation for Scientists, The Benjamin/Cummings Publishing
Company, Inc., 1981, ISBN 0-8053-6917-1, Chapter 5.

[5] Ron Mancini, extquotedblDesigning with comparators,”

EDN, March 29, 2001.

[6] Rogenmoser, R.; Kaeslin, H, “The impact of transistor

sizing on power efficiency in submicron CMOS circuits,”
Solid-State Circuits, IEEE Journal of Volume 32, Issue 7,
Jul 1997 Page(s):1142–1145.

[7] Ron Mancini, extquotedblAdding Hysteresis to compara-

tors,” EDN, May 3, 2001.
Chapter 6

Sensors

6.1 Thermistor called a positive temperature coefficient (PTC) thermis-

tor, or posistor. If k is negative, the resistance decreases
A thermistor is a type of resistor whose resistance varies with increasing temperature, and the device is called a
significantly with temperature, more so than in standard negative temperature coefficient (NTC) thermistor. Re-
resistors. The word is a portmanteau of thermal and sistors that are not thermistors are designed to have a k
resistor. Thermistors are widely used as inrush current as close to 0 as possible, so that their resistance remains
limiters, temperature sensors, self-resetting overcurrent nearly constant over a wide temperature range.
protectors, and self-regulating heating elements. Instead of the temperature coefficient k, sometimes the
Thermistors differ from resistance temperature detectors temperature coefficient of resistance αT (alpha sub T) is
[2]
(RTDs) in that the material used in a thermistor is gen- used. It is defined as
erally a ceramic or polymer, while RTDs use pure met-
als. The temperature response is also different; RTDs 1 dR
are useful over larger temperature ranges, while thermis- αT = .
R(T ) dT
tors typically achieve a higher precision within a limited
temperature range, typically −90 °C to 130 °C.[1] This αT coefficient should not be confused with the a
parameter below.

6.1.1 Basic operation

6.1.2 Steinhart–Hart equation
Assuming, as a first-order approximation, that the re-
lationship between resistance and temperature is linear, In practice, the linear approximation (above) works only
then: over a small temperature range. For accurate temperature
measurements, the resistance/temperature curve of the
device must be described in more detail. The Steinhart–
Hart equation is a widely used third-order approximation:
∆R = k∆T

where 1
= a + b ln(R) + c (ln(R))3
T
where a, b and c are called the Steinhart–Hart parame-
∆R
ters, and must be specified for each device. T is the tem-
∆T perature in kelvin and R is the resistance in ohms. To give
resistance as a function of temperature, the above can be
k rearranged into:
Thermistors can be classified into two types, depending
on the classification of k . If k is positive, the resistance y
1 1
−(x+ y2 ) 3
R = e(x− 2 )
3
increases with increasing temperature, and the device is

151
152 CHAPTER 6. SENSORS

where 6.1.4 Conduction model

√( )
a− T1 b 3
y= and x = + y2 NTC
c 3c 4

The error in the Steinhart–Hart equation is generally less
than 0.02 °C in the measurement of temperature over a
200 °C range.[3] As an example, typical values for a ther-
mistor with a resistance of 3000 Ω at room temperature
(25 °C = 298.15 K) are:
a = 1.40 × 10−3
b = 2.37 × 10−4
c = 9.90 × 10−8
6.1.3 B or β parameter equation
Many NTC thermistors are made from a pressed disc,
rod, plate, bead or cast chip of a semiconductor such as a
sintered metal oxide. They work because raising the tem-
perature of a semiconductor increases the number of ac-
tive charge carriers - it promotes them into the conduction
band. The more charge carriers that are available, the
more current a material can conduct. In certain materi-
als like ferric oxide (Fe2 O3 ) with titanium (Ti) doping
a n-type semiconductor is formed and the charge carri-
ers are electrons. In materials such as nickel oxide (NiO)
with lithium (Li) doping a p-type semiconductor is cre-
ated where holes are the charge carriers.[4]
This is described in the formula:

NTC thermistors can also be characterised with the B (or

β) parameter equation, which is essentially the Steinhart– I =n·A·v·e
Hart equation with a = (1/T0 ) − (1/B) ln(R0 ) , b =
I = electric current (amperes)
1/B and c = 0 ,
n = density of charge carriers (count/m³)
A = cross-sectional area of the material (m²)
( )
1 1 1 R v = velocity of charge carriers (m/s)
= + ln e = charge of an electron ( e = 1.602 × 10−19 coulomb)
T T0 B R0
Where the temperatures are in kelvin and R0 is the resis- Over large changes in temperature, calibration is nec-
tance at temperature T 0 (25 °C = 298.15 K). Solving for essary. Over small changes in temperature, if the right
R yields: semiconductor is used, the resistance of the material is
linearly proportional to the temperature. There are many
different semiconducting thermistors with a range from
−B( T1 − T1 ) about 0.01 kelvin to 2,000 kelvins (−273.14 °C to 1,700
R = R0 e 0

°C).
or, alternatively,

R = r∞ eB/T
where r∞ = R0 e−B/T0 .
This can be solved for the temperature:
B
T =
ln (R/r∞ )
The B-parameter equation can also be written as ln R =
B/T + ln r∞ . This can be used to convert the function
of resistance vs. temperature of a thermistor into a lin-
ear function of ln R vs. 1/T . The average slope of this
function will then yield an estimate of the value of the B
parameter.
PTC
Most PTC thermistors are of the “switching” type, which
means that their resistance rises suddenly at a certain crit-
ical temperature. The devices are made of a doped poly-
crystalline ceramic containing barium titanate (BaTiO3 )
and other compounds. The dielectric constant of this
ferroelectric material varies with temperature. Below
the Curie point temperature, the high dielectric constant
prevents the formation of potential barriers between the
crystal grains, leading to a low resistance. In this region
the device has a small negative temperature coefficient.
At the Curie point temperature, the dielectric constant
drops sufficiently to allow the formation of potential bar-
riers at the grain boundaries, and the resistance increases
6.1. THERMISTOR 153

sharply. At even higher temperatures, the material reverts heat energy is transferred to the surrounding environ-
to NTC behaviour. ment. The rate of transfer is well described by Newton’s
Another type of thermistor is a silistor, a thermally sen- law of cooling:
sitive silicon resistor. Silistors employ silicon as the
semiconductive component material. In contrary to the
“switching” type thermistor, silistors have an almost lin- PT = K(T (R) − T0 )
ear resistance-temperature characteristic.[5]
The degaussing coils in many CRT monitors were con- where T(R) is the temperature of the thermistor as a func-
trolled by thermistors bonded to a small heating element. tion of its resistance R, T0 is the temperature of the sur-
The thermistor would be connected in series with the coil roundings, and K is the dissipation constant, usually ex-
across the AC input, with the heater also directly con- pressed in units of milliwatts per degree Celsius. At equi-
nected to the AC input. When cold the thermistor would librium, the two rates must be equal.
allow a large current to flow through but would be quickly
heated by the heating element and the current would trail
to zero. This would degauss the screen every time the PE = PT
power is removed for long enough for the device to cool.
Another device similar in function to PTC thermistor is The current and voltage across the thermistor will depend
the polymer PTC, which is sold under brand names such on the particular circuit configuration. As a simple exam-
as extquotedblPolyswitch extquotedbl “Semifuse”, and ple, if the voltage across the thermistor is held fixed, then
“Multifuse”. This consists of a slice of plastic with carbon by Ohm’s Law we have I = V /R and the equilibrium
grains embedded in it. When the plastic is cool, the car- equation can be solved for the ambient temperature as a
bon grains are all in contact with each other, forming a function of the measured resistance of the thermistor:
conductive path through the device. When the plastic
heats up, it expands, forcing the carbon grains apart, and
causing the resistance of the device to rise rapidly. Like V2
the BaTiO3 thermistor, this device has a highly nonlinear T0 = T (R) −
KR
resistance/temperature response and is used for switch-
ing, not for proportional temperature measurement. The dissipation constant is a measure of the thermal con-
nection of the thermistor to its surroundings. It is gener-
ally given for the thermistor in still air, and in well-stirred
6.1.5 Self-heating effects
oil. Typical values for a small glass bead thermistor are
When a current flows through a thermistor, it will gen- 1.5 mW/°C in still air and 6.0 mW/°C in stirred oil. If
erate heat which will raise the temperature of the ther- the temperature of the environment is known beforehand,
mistor above that of its environment. If the thermistor then a thermistor may be used to measure the value of the
is being used to measure the temperature of the environ- dissipation constant. For example, the thermistor may be
ment, this electrical heating may introduce a significant used as a flow rate sensor, since the dissipation constant
error if a correction is not made. Alternatively, this effect increases with the rate of flow of a fluid past the thermis-
itself can be exploited. It can, for example, make a sensi- tor.
tive air-flow device employed in a sailplane rate-of-climb The power dissipated in a thermistor is typically main-
instrument, the electronic variometer, or serve as a timer tained at a very low level to ensure insignificant tem-
for a relay as was formerly done in telephone exchanges. perature measurement error due to self heating. How-
The electrical power input to the thermistor is just: ever, some thermistor applications depend upon signifi-
cant “self heating” to raise the body temperature of the
thermistor well above the ambient temperature so the
PE = IV sensor then detects even subtle changes in the thermal
conductivity of the environment. Some of these applica-
where I is current and V is the voltage drop across the tions include liquid level detection, liquid flow measure-
thermistor. This power is converted to heat, and this ment and air flow measurement.[2]
154
6.1.6 Applications
• PTC thermistors can be used as current-limiting
devices for circuit protection, as replacements for
fuses. Current through the device causes a small
amount of resistive heating. If the current is large
enough to generate more heat than the device can
lose to its surroundings, the device heats up, caus-
ing its resistance to increase. This creates a self-
reinforcing effect that drives the resistance upwards,
therefore limiting the current.
• PTC thermistors were used as timers in the
degaussing coil circuit of most CRT displays. When
the display unit is initially switched on, current flows
through the thermistor and degaussing coil. The coil
and thermistor are intentionally sized so that the cur-
rent flow will heat the thermistor to the point that
the degaussing coil shuts off in under a second. For
effective degaussing, it is necessary that the magni-
tude of the alternating magnetic field produced by
the degaussing coil decreases smoothly and contin-
uously, rather than sharply switching off or decreas-
ing in steps; the PTC thermistor accomplishes this
naturally as it heats up. A degaussing circuit using
a PTC thermistor is simple, reliable (for its simplic-
ity), and inexpensive.
• PTC thermistors were used as heater in automotive
industry to provide additional heat inside cabin with
diesel engine or to heat diesel in cold climatic con-
ditions before engine injection. 6.1.7
• PTC thermistors are used in temperature compen- The first NTC thermistor was discovered in 1833 by
sated synthesizer voltage controlled oscillators.[6]
• NTC thermistors are used as resistance thermome- behavior of silver sulfide. Faraday noticed that the re-
ters in low-temperature measurements of the order sistance of silver sulfide decreased dramatically as tem-
of 10 K.
• NTC thermistors can be used as inrush-current lim-
iting devices in power supply circuits. They present
a higher resistance initially which prevents large cur-
rents from flowing at turn-on, and then heat up and
become much lower resistance to allow higher cur-
rent flow during normal operation. These thermis-
tors are usually much larger than measuring type
thermistors, and are purposely designed for this 6.1.8 See also
application.[7]
• NTC thermistors are regularly used in automotive
applications. For example, they monitor things like
CHAPTER 6. SENSORS
coolant temperature and/or oil temperature inside
the engine and provide data to the ECU and, indi-
rectly, to the dashboard.
• NTC thermistors can be also used to monitor the
temperature of an incubator.
• Thermistors are also commonly used in modern
digital thermostats and to monitor the temperature
of battery packs while charging.
• Thermistors are often used in the hot ends of 3D
printers; they monitor the heat produced and allow
the printer’s control circuitry to keep a constant tem-
perature for melting the plastic filament.
• NTC thermistors are used in the Food Handling and
Processing industry, especially for food storage sys-
tems and food preparation. Maintaining the correct
temperature is critical to prevent food borne illness.
• NTC thermistors are used throughout the Con-
sumer Appliance industry for measuring tempera-
ture. Toasters, coffee makers, refrigerators, freez-
ers, hair dryers, etc. all rely on thermistors for
proper temperature control.
• NTC thermistors come in bare and lugged forms,
the former is for point sensing to achieve high ac-
curacy for specific points, such as laser diode die,
etc.[8]
History
Michael Faraday, who reported on the semiconducting
perature increased. (This was also the first documented
observation of a semiconducting material.) [9]
Because early thermistors were difficult to produce and
applications for the technology were limited, commer-
cial production of thermistors did not begin until the
1930s.[10] A commercially viable thermistor was in-
vented by Samuel Ruben in 1930.[11]
• Iron-hydrogen resistor
• Thermocouple
6.2. PHOTODIODE 155

6.1.9 References
[1] “NTC Thermistors”. Micro-chip Technologies. 2010.

[2] Thermistor Terminology. U.S. Sensor

[3] “Practical Temperature Measurements”. Agilent Appli-

cation Note. Agilent Semiconductor.

[4] L. W Turner, ed. (1976). Electronics Engineer’s Reference

Book (4 ed.). Butterworths. pp. 6–29 to 6–41. ISBN
0408001682.

[5] “PTC Thermistors and Silistors” The Resistor Guide

[6] Temperature Compensated VCO

[7] Inrush Current Limiting Power Thermistors. U.S. Sensor

[8] “PTC Thermistors Guide- “Publish By Analog Electronic

Technologies”. extquotedbl.

[9] “1833 - First Semiconductor Effect is Recorded”. Com-

puter History Museum. Retrieved 24 June 2014. Three Si and one Ge (bottom) photodiodes

[10] McGee, Thomas (1988). “Chapter 9”. Principles and

Methods of Temperature Measurement. John Wiley &
Sons. p. 203.
Anode Cathode
[11] Jones, Deric P., ed. (2009). Biomedical Sensors. Momen-
tum Press. p. 12.

6.1.10 External links Symbol for photodiode.

• The thermistor at bucknell.edu

Photodiodes are similar to regular semiconductor diodes
• Software for thermistor calculation at Sourceforge except that they may be either exposed (to detect vacuum
UV or X-rays) or packaged with a window or optical fiber
• “Thermistors & Thermocouples:Matching the Tool
connection to allow light to reach the sensitive part of the
to the Task in Thermal Validation” - Journal of Val-
device. Many diodes designed for use specifically as a
idation Technology
photodiode use a PIN junction rather than a p-n junc-
tion, to increase the speed of response. A photodiode is
designed to operate in reverse bias.[1]
6.2 Photodiode
A photodiode is a semiconductor device that converts 6.2.1 Principle of operation
light into current. The current is generated when photons
are absorbed in the photodiode. A small amount of cur- A photodiode is a p-n junction or PIN structure. When a
rent is also produced when no light is present. Photodi- photon of sufficient energy strikes the diode, it creates an
odes may contain optical filters, built-in lenses, and may electron-hole pair. This mechanism is also known as the
have large or small surface areas. Photodiodes usually inner photoelectric effect. If the absorption occurs in the
have a slower response time as its surface area increases. junction’s depletion region, or one diffusion length away
The common, traditional solar cell used to generate elec- from it, these carriers are swept from the junction by the
tric solar power is a large area photodiode. built-in electric field of the depletion region. Thus holes
156 CHAPTER 6. SENSORS

rent of a good PIN diode is so low (<1 nA) that the

Johnson–Nyquist noise of the load resistance in a typi-
cal circuit often dominates.

Other modes of operation

Avalanche photodiodes have a similar structure to reg-

ular photodiodes, but they are operated with much higher
reverse bias. This allows each photo-generated carrier to
be multiplied by avalanche breakdown, resulting in inter-
nal gain within the photodiode, which increases the ef-
fective responsivity of the device.

I-V characteristic of a photodiode. The linear load lines repre-

sent the response of the external circuit: I=(Applied bias voltage-
Diode voltage)/Total resistance. The points of intersection with
the curves represent the actual current and voltage for a given
bias, resistance and illumination.

move toward the anode, and electrons toward the cathode,

and a photocurrent is produced. The total current through
the photodiode is the sum of the dark current (current that
is generated in the absence of light) and the photocurrent,
so the dark current must be minimized to maximize the
sensitivity of the device.[2]

Photovoltaic mode

When used in zero bias or photovoltaic mode, the flow

of photocurrent out of the device is restricted and a volt-
age builds up. This mode exploits the photovoltaic effect,
which is the basis for solar cells – a traditional solar cell
is just a large area photodiode.

Photoconductive mode Electronic symbol for a phototransistor

In this mode the diode is often reverse biased (with the A phototransistor is a light-sensitive transistor. A com-
cathode driven positive with respect to the anode). This mon type of phototransistor, called a photobipolar tran-
reduces the response time because the additional reverse sistor, is in essence a bipolar transistor encased in a
bias increases the width of the depletion layer, which de-
transparent case so that light can reach the base-collector
creases the junction’s capacitance. The reverse bias alsojunction. It was invented by Dr. John N. Shive (more fa-
increases the dark current without much change in the mous for his wave machine) at Bell Labs in 1948,[4]:205
photocurrent. For a given spectral distribution, the pho-but it wasn't announced until 1950.[5] The electrons that
tocurrent is linearly proportional to the illuminance (and
are generated by photons in the base-collector junction
to the irradiance).[3] are injected into the base, and this photodiode current
Although this mode is faster, the photoconductive mode is amplified by the transistor’s current gain β (or h ₑ).
tends to exhibit more electronic noise. The leakage cur- If the emitter is left unconnected, the phototransistor
6.2. PHOTODIODE 157

becomes a photodiode. While phototransistors have a

higher responsivity for light they are not able to detect
low levels of light any better than photodiodes. Photo-
0.6
transistors also have significantly longer response times.

response [A/W]
Field-effect phototransistors, also known as photoFETs, 0.5
are light-sensitive field-effect transistors. Unlike photo-
0.4
bipolar transistors, photoFETs control drain-source cur-
rent by creating a gate voltage. 0.3

0.2
6.2.2 Materials 0.1

The material used to make a photodiode is critical to 0

defining its properties, because only photons with suf- 400 600 800 1000
ficient energy to excite electrons across the material’s wavelength [nm]
bandgap will produce significant photocurrents.
Materials commonly used to produce photodiodes Response of a silicon photo diode vs wavelength of the incident
include:[6] light
Because of their greater bandgap, silicon-based photodi-
odes generate less noise than germanium-based photodi- Dark current The current through the photodiode in
odes. the absence of light, when it is operated in photo-
conductive mode. The dark current includes pho-
Unwanted photodiode effects tocurrent generated by background radiation and the
saturation current of the semiconductor junction.
Any p-n junction, if illuminated, is potentially a photo- Dark current must be accounted for by calibration
diode. Semiconductor devices such as transistors and if a photodiode is used to make an accurate optical
ICs contain p-n junctions, and will not function cor- power measurement, and it is also a source of noise
rectly if they are illuminated by unwanted electromag- when a photodiode is used in an optical communi-
netic radiation (light) of wavelength suitable to produce cation system.
a photocurrent;[7][8] this is avoided by encapsulating de-
vices in opaque housings. If these housings are not com- Response time A photon absorbed by the semiconduct-
pletely opaque to high-energy radiation (ultraviolet, X- ing material will generate an electron-hole pair
rays, gamma rays), transistors and ICs can malfunction [9] which will in turn start moving in the material un-
due to induced photo-currents. Background radiation der the effect of the electric field and thus generate a
[10]
from the packaging is also significant. Radiation hard- current. The finite duration of this current is known
ening mitigates these effects. as the transit-time spread and can be evaluated by
using Ramo’s theorem. One can also show with this
theorem that the total charge generated in the exter-
6.2.3 Features nal circuit is well e and not 2e as might seem by the
presence of the two carriers. Indeed the integral of
Critical performance parameters of a photodiode include: the current due to both electron and hole over time
must be equal to e. The resistance and capacitance
Responsivity The Spectral responsivity is a ratio of of the photodiode and the external circuitry give rise
the generated photocurrent to incident light power, to another response time known as RC time constant
expressed in A/W when used in photoconductive τ = RC . This combination of R and C integrates
mode. The wavelength-dependence may also be ex- the photoresponse over time and thus lengthens the
pressed as a Quantum efficiency, or the ratio of the impulse response of the photodiode. When used in
number of photogenerated carriers to incident pho- an optical communication system, the response time
tons, a unitless quantity. determines the bandwidth available for signal mod-
158 CHAPTER 6. SENSORS

ulation and thus data transmission. of a mechanical obstruction to the beam (slotted optical
switch), or to couple two digital or analog circuits while
Noise-equivalent power (NEP) The minimum input maintaining extremely high electrical isolation between
optical power to generate photocurrent, equal to the them, often for safety (optocoupler).
rms noise current in a 1 hertz bandwidth. NEP is Photodiodes are often used for accurate measurement of
essentially the minimum detectable power. The re- light intensity in science and industry. They generally
lated characteristic detectivity ( D ) is the inverse have a more linear response than photoconductors.
of NEP, 1/NEP. There is also the specific detectiv-
ity ( D⋆ ) which is the detectivity multiplied by the They are also widely used in various medical applica-
square root tions, such as detectors for computed tomography (cou-
⋆
√ of the area ( A ) of the photodetector, ( pled with scintillators), instruments to analyze samples
D = D A ) for a 1 Hz bandwidth. The specific
detectivity allows different systems to be compared (immunoassay), and pulse oximeters.
independent of sensor area and system bandwidth; PIN diodes are much faster and more sensitive than p-
a higher detectivity value indicates a low-noise de- n junction diodes, and hence are often used for optical
vice or system.[11] Although it is traditional to give ( communications and in lighting regulation.
D⋆ ) in many catalogues as a measure of the diode’s
P-N photodiodes are not used to measure extremely low
quality, in practice, it is hardly ever the key param-
light intensities. Instead, if high sensitivity is needed,
eter.
avalanche photodiodes, intensified charge-coupled de-
vices or photomultiplier tubes are used for applications
When a photodiode is used in an optical communication such as astronomy, spectroscopy, night vision equipment
system, all these parameters contribute to the sensitivity and laser rangefinding.
of the optical receiver, which is the minimum input power
required for the receiver to achieve a specified bit error Pinned photodiode is not a PIN photodiode, it has p+/n/p
rate. regions in it. It has a shallow P+ implant in N type dif-
fusion layer over a P-type epitaxial substrate layer. It is
used in CMOS Active pixel sensor.[12]
6.2.4 Applications
P-N photodiodes are used in similar applications to other Comparison with photomultipliers
photodetectors, such as photoconductors, charge-coupled [13]
devices, and photomultiplier tubes. They may be used Advantages compared to photomultipliers:
to generate an output which is dependent upon the illu-
mination (analog; for measurement and the like), or to 1. Excellent linearity of output current as a function of
change the state of circuitry (digital; either for control incident light
and switching, or digital signal processing).
2. Spectral response from 190 nm to 1100
Photodiodes are used in consumer electronics devices nm (silicon), longer wavelengths with other
such as compact disc players, smoke detectors, and the semiconductor materials
receivers for infrared remote control devices used to con-
trol equipment from televisions to air conditioners. For 3. Low noise
many applications either photodiodes or photoconductors
may be used. Either type of photosensor may be used 4. Ruggedized to mechanical stress
for light measurement, as in camera light meters, or to
5. Low cost
respond to light levels, as in switching on street lighting
after dark. 6. Compact and light weight
Photosensors of all types may be used to respond to in-
7. Long lifetime
cident light, or to a source of light which is part of the
same circuit or system. A photodiode is often combined 8. High quantum efficiency, typically 60-80% [14]
into a single component with an emitter of light, usually a
light-emitting diode (LED), either to detect the presence 9. No high voltage required
6.2. PHOTODIODE 159

Disadvantages compared to photomultipliers: 6.2.6 See also

• Electronics
1. Small area
• Band gap
2. No internal gain (except avalanche photodiodes, but
• Infrared
their gain is typically 102 –103 compared to up to
108 for the photomultiplier) • Optoelectronics

3. Much lower overall sensitivity • Optical interconnect

• Light Peak
4. Photon counting only possible with specially de-
signed, usually cooled photodiodes, with special • Interconnect bottleneck
electronic circuits
• Optical fiber cable
5. Response time for many designs is slower • Optical communication
• Parallel optical interface
6. latent effect
• Opto-isolator
• Semiconductor device
6.2.5 Photodiode array
• Solar cell
• Avalanche photodiode
• Transducer
• LEDs as Photodiode Light Sensors
• Light meter
• Image sensor
• Transimpedance amplifier

6.2.7 References
This article incorporates public domain material from
the General Services Administration document “Federal
Standard 1037C”.

[1] James F. Cox (26 June 2001). Fundamentals of linear

A 2 x 2 cm photodiode array chip with more than 200 diodes
A one-dimensional array of hundreds or thousands of
photodiodes can be used as a position sensor, for exam-
ple as part of an angle sensor.[15] One advantage of pho-
todiode arrays (PDAs) is that they allow for high speed
parallel read out since the driving electronics may not be
built in like a traditional CMOS or CCD sensor.
electronics: integrated and discrete. Cengage Learning.
pp. 91–. ISBN 978-0-7668-3018-9. Retrieved 20 Au-
gust 2011.
[2] Filip Tavernier, Michiel Steyaert High-Speed Optical Re-
ceivers with Integrated Photodiode in Nanoscale CMOS
Springer, 2011 ISBN 1-4419-9924-8, Chapter 3 From
Light to Electric Current - The Photodiode
[3] “Photodiode slide”.
160 CHAPTER 6. SENSORS

[4] Michael Riordan and Lillian Hoddeson. Crystal Fire: The 6.2.8 External links
Invention of the Transistor and the Birth of the Information
Age. ISBN 9780393318517. • Technical Information Hamamatsu Photonics

[5] “The phototransistor”. Bell Laboratories RECORD. May • Using the Photodiode to convert the PC to a Light
1950. Intensity Logger

[6] Held. G, Introduction to Light Emitting Diode Technol- • Design Fundamentals for Phototransistor Circuits
ogy and Applications, CRC Press, (Worldwide, 2008).
• Working principles of photodiodes
Ch. 5 p. 116. ISBN 1-4200-7662-0

[7] Z. Shanfield et al, 1988, Investigation of ra-

diation effects on semiconductor devices 6.3 Photoresistor
and integrated circuits, DNA-TR-88-221,
www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA210165

[8] Krzysztof Iniewski, ed, 2010, Radiation Effects in Semi-

conductors, CRC Press, ISBN 978-1-4398-2694-2

[9] H.R. Zeller, 1995, Cosmic ray induced failures in high

power semiconductor devices, www.sciencedirect.com/
science/article/pii/0038110195000825

[10] T. May and M. Woods, Alpha-particle-induced soft er-

rors in dynamic memories”, IEEE Trans. Elec. Dev.,
vol. 26, No. 1, pp 2, Jan 1979, cited in R. C. Baumann
2004, Soft errors in commercial integrated circuits, Int.
J. High Speed Electronics and Systems, Vol. 14, No 2
(2004) 299-309: “alpha particles emitted from the nat-
ural radioactive decay of uranium, thorium, and aughter
isotopes present as impurities in packaging materials were
found to be the dominant cause of [soft error rate] in [dy- The symbol for a photoresistor[1]
namic random-access memories].”

[11] Graham Brooker, Introduction to Sensors for Rang- A photoresistor or light-dependent resistor (LDR)
ing and Imaging, ScitTech Publishing, 2009 ISBN or photocell is a light-controlled variable resistor. The
9781891121746 page 87 resistance of a photoresistor decreases with increas-
ing incident light intensity; in other words, it exhibits
[12] http://electronics.stackexchange.com/questions/83018/ photoconductivity. A photoresistor can be applied in
difference-between-buried-photodiode-and-pinned-photodiode
light-sensitive detector circuits, and light- and dark-
activated switching circuits.
[13] Photodiode Technical Guide on Hamamatsu website
A photoresistor is made of a high resistance
[14] Knoll, F.G. (2010). Radiation detection and measurementsemiconductor. In the dark, a photoresistor can
−4th ed. p. 298. Wiley, Hoboken, NJ. ISBN 978-0-470- have a resistance as high as a few megaohms (MΩ),
13148-0 while in the light, a photoresistor can have a resistance
as low as a few hundred ohms. If incident light on
[15] Wei Gao (2010). Precision Nanometrology: Sensors and a photoresistor exceeds a certain frequency, photons
Measuring Systems for Nanomanufacturing. Springer. pp. absorbed by the semiconductor give bound electrons
15–16. ISBN 978-1-84996-253-7.
enough energy to jump into the conduction band. The
resulting free electrons (and their hole partners) conduct
• Gowar, John, Optical Communication Systems, 2 ed., electricity, thereby lowering resistance. The resistance
Prentice-Hall, Hempstead UK, 1993 (ISBN 0-13- range and sensitivity of a photoresistor can substantially
638727-6) differ among dissimilar devices. Moreover, unique
6.3. PHOTORESISTOR 161

photoresistors may react substantially differently to Photoresistors also exhibit a certain degree of latency be-
photons within certain wavelength bands. tween the moment they are hit by light and the subse-
A photoelectric device can be either intrinsic or extrin- quent moment that their resistance actually drops. This
sic. An intrinsic semiconductor has its own charge car- drop is usually about 10 milliseconds. Also, they exhibit a
riers and is not an efficient semiconductor, for example, much longer latency when going from lighted to dark en-
silicon. In intrinsic devices the only available electrons vironments, often of as long as one second. This property
are in the valence band, and hence the photon must have makes them unsuitable for sensing rapidly flashing lights
enough energy to excite the electron across the entire (unless the flashing is so frequent that the [3]
diode cannot
bandgap. Extrinsic devices have impurities, also called detect the drop in light between flashes).
dopants, added whose ground state energy is closer to the
conduction band; since the electrons do not have as far to 6.3.2 Applications
jump, lower energy photons (that is, longer wavelengths
and lower frequencies) are sufficient to trigger the device. Photoresistors come in many types. Inexpensive
If a sample of silicon has some of its atoms replaced by cadmium sulphide cells can be found in many consumer
phosphorus atoms (impurities), there will be extra elec- items such as camera light meters, street lights, clock ra-
trons available for conduction. This is an example of an dios, alarm devices, night lights, outdoor clocks, solar
extrinsic semiconductor.[2] street lamps and solar road studs, etc.
They are also used in some dynamic compressors to-
gether with a small incandescent or neon lamp, or light-
emitting diode to control gain reduction. A common us-
age of this application can be found in many guitar am-
plifiers that incorporate an onboard tremolo effect, as the
oscillating light patterns control the level of signal run-
ning through the amp circuit.
The use of CdS and CdSe[4] photoresistors is severely re-
stricted in Europe due to the RoHS ban on cadmium.
Lead sulphide (PbS) and indium antimonide (InSb)
LDRs (light-dependent resistors) are used for the mid-
infrared spectral region. Ge:Cu photoconductors are
among the best far-infrared detectors available, and are
used for infrared astronomy and infrared spectroscopy.
The internal components of a photoelectric control for a typical
American streetlight. The photoresistor is facing rightwards, and
controls whether current flows through the heater which opens 6.3.3 References
the main power contacts. At night, the heater cools, closing the
power contacts, energizing the street light. [1] British Standard 3939

[2] Diffenderfes, Robert (2005). Electronic Devices: System

and Applications. New Delhi: Delimar. p. 480. ISBN
6.3.1 Design considerations 978-1401835149.

Photoresistors are less light-sensitive devices than photo- [3] http://www.resistorguide.com/photoresistor/

diodes or phototransistors: the two latter components are [4] “Silonex: TO-18 photocells on ceramic substrate extquot-
true semiconductor devices, while a photoresistor is pas- edbl (PDF). Retrieved 17 October 2013.
sive component and does not have a PN-junction. The
photoresistivity of any photoresistor may vary widely de-
pending on ambient temperature, making them unsuit- 6.3.4 See also
able for applications requiring precise measurement of
or sensitivity to light. • Optoelectronics
162 CHAPTER 6. SENSORS

• Photodiode • off-resistance: the resistance when switched off.

This is typically a number of megohms or gigaohms.
• Phototransistor
• signal range: the minimum and maximum voltages
allowed for the signal to be passed through. If these
6.3.5 External links are exceeded, the switch may be destroyed by ex-
cessive currents. Older types of switches can even
• Using a photoresistor to measure nocturnal light lev- latch up, which means that they continue to conduct
els - The Moonlight Sensor excessive currents even after the faulty signal is re-
moved.
• Using a photoresistor to track light
• charge injection. This effect causes the switch to
• Connecting a photoresistor to a circuit inject a small electric charge into the signal when
it switches on, causing a small spike or glitch. The
• Photoresistor overview - detailing operation, struc- charge injection is specified in coulombs.
ture and circuit information
6.4.1 See also
6.4 Analogue switch • Analog-to-digital converter
• Telecommunications equipment
The analogue (or analog) switch, also called the bi-
lateral switch, is an electronic component that behaves
in a similar way to a relay, but has no moving parts.
The switching element is normally a pair MOSFET
transistors, one an N-channel device, the other a P-
channel device. The device can conduct analog or dig-
ital signals in either direction when on and isolates the
switched terminals when off. Analogue switches are usu-
ally manufactured as integrated circuits in packages con-
taining multiple switches (typically two, four or eight).
These include the 4016 and 4066 from the 4000 series.
The control input to the device may be a signal that
switches between the positive and negative supply volt-
ages, with the more positive voltage switching the device
on and the more negative switching the device off. Other
circuits are designed to communicate through a serial port
with a host controller in order to set switches on or off.
The signal being switched must remain within the bounds
of the positive and negative supply rails which are con-
nected to the P-MOS and N-MOS body terminals. The
switch generally provides good isolation between the con-
trol signal and the input/output signals. They are not used
for high voltage switching.
Important parameters of an analogue switch are:

• on-resistance: the resistance when switched on.

This commonly ranges from 5 ohms to a few hun-
dred ohms.
Chapter 7
The Decibel
7.1 Decibel
This article is about the unit of level. For other uses, see ample, dBm indicates a reference level of one milliwatt,
Decibel (disambiguation).
The decibel (dB) is a logarithmic unit used to express
the ratio between two values of a physical quantity, of-
ten power or intensity. One of these quantities is often a
reference value, and in this case the decibel can be used
to express the absolute level of the physical quantity, as
in the case of sound pressure. The number of decibels
is ten times the logarithm to base 10 of the ratio of two
power quantities,[1] or of the ratio of the squares of two
field amplitude quantities. One decibel is one tenth of
one bel, named in honor of Alexander Graham Bell. The
bel is seldom used without the deci- prefix.
The definition of the decibel is based on the measure-
ment practices in telephony of the early 20th century in
the Bell System in the United States. Today, the unit is
used for a wide variety of measurements in science and
engineering, most prominently in acoustics, electronics,
and control theory. In electronics, the gains of amplifiers,
attenuation of signals, and signal-to-noise ratios are often
expressed in decibels. The decibel confers a number of
advantages, such as the ability to conveniently represent
very large or small numbers, and the ability to carry out
multiplication of ratios by simple addition and subtrac-
tion.
A change in power by a factor of 10 corresponds to a 10
dB change in level. A change in power by a factor of two
approximately corresponds to a 3 dB change. A change
in voltage by a factor of 10 results in a change in power
by a factor of 100 and corresponds to a 20 dB change. A
change in voltage ratio by a factor of two approximately
corresponds to a 6 dB change.
163
The decibel symbol is often qualified with a suffix that in-
dicates which reference quantity has been used or some
other property of the quantity being measured. For ex-
while dBu is referenced to approximately 0.775 volts
RMS.[2]
In the International System of Quantities, the decibel is
defined as a unit of level or level difference, equal to one-
tenth of a bel. The bel is then defined in terms of the
neper, an alternative unit of level of root-power quan-
tities, applicable when the natural logarithm (base e) is
used to define the level.[3]
7.1.1 History
The decibel originates from methods used to quantify
signal losses in telephone circuits. These losses were
originally measured in units of Miles of Standard Cable
(MSC), where 1 MSC corresponded to the loss of power
over a 1 mile (approximately 1.6 km) length of standard
telephone cable at a frequency of 5000 radians per second
(795.8 Hz), and roughly matched the smallest attenuation
detectable to the average listener. Standard telephone ca-
ble was defined as “a cable having uniformly distributed
resistance of 88 ohms per loop mile and uniformly dis-
tributed shunt capacitance of .054 microfarad per mile”
(approximately 19 gauge).[4]
The transmission unit (TU) was devised by engineers of
the Bell Telephone Laboratories in the 1920s to replace
the MSC. 1 TU was defined as ten times the base-10 loga-
rithm of the ratio of measured power to a reference power
level.[5] The definitions were conveniently chosen such
that 1 TU approximately equaled 1 MSC (specifically,
1.056 TU = 1 MSC).The threshold of hearing is 25 dB[6]
In 1928, the Bell system renamed the TU the decibel,[7]
164
being one tenth of a newly defined unit for the base-10
logarithm of the power ratio. It was named the bel, in
honor of their founder and telecommunications pioneer
Alexander Graham Bell.[8] The bel is seldom used, as the
decibel was the proposed working unit.[9]
The naming and early definition of the decibel is de- 7.1.2
scribed in the NBS Standard’s Yearbook of 1931:[10]
Since the earliest days of the telephone,
the need for a unit in which to measure the
transmission efficiency of telephone facilities
has been recognized. The introduction of ca-
ble in 1896 afforded a stable basis for a con-
venient unit and the “mile of standard” cable
came into general use shortly thereafter. This
unit was employed up to 1923 when a new unit
was adopted as being more suitable for modern
telephone work. The new transmission unit is
widely used among the foreign telephone orga-
nizations and recently it was termed the “deci-
bel” at the suggestion of the International Advi-
sory Committee on Long Distance Telephony.
The decibel may be defined by the state-
ment that two amounts of power differ by 1
decibel when they are in the ratio of 100.1 and
any two amounts of power differ by N decibels
when they are in the ratio of 10N(0.1) . The num-
ber of transmission units expressing the ratio of
any two powers is therefore ten times the com-
mon logarithm of that ratio. This method of
designating the gain or loss of power in tele-
phone circuits permits direct addition or sub-
traction of the units expressing the efficiency
of different parts of the circuit...
Standards
In April 2003, the International Committee for Weights
and Measures (CIPM) considered a recommendation for
the decibel’s inclusion in the International System of
Units (SI), but decided not to adopt the decibel as an SI
unit.[11] However, the decibel is recognized by other in-
ternational bodies such as the International Electrotech-
nical Commission (IEC) and International Organization
for Standardization (ISO).[12] The IEC permits the use
of the decibel with field quantities as well as power and
this recommendation is followed by many national stan-
dards bodies, such as NIST, which justifies the use of
the decibel for voltage ratios.[13] The term field quantity
CHAPTER 7. THE DECIBEL
is deprecated by ISO, which favors root-power. Neither
IEC nor ISO permit the use of modifiers such as dBA or
dBV; such units, though widely used, are not defined by
international standards.
Definition
The decibel (dB) is one tenth of the bel (B): 1B = 10dB.
The bel is (1/2) ln(10) nepers.
The bel represents a ratio between two power quanti-
ties of 10:1, and a ratio between two field quantities
of √10:1.[14] A field quantity is a quantity such as volt-
age, current, pressure, electric field strength, velocity, or
charge density, the square of which in linear systems is
proportional to power.[15] A power quantity is a power
or a quantity directly proportional to power, e.g., energy
density, acoustic intensity and luminous intensity.
The method of calculation of a ratio in decibels depends
on whether the measured property is a power quantity or
a field quantity.
Two signals that differ by one decibel have a power ratio
1
of 10 10 which is approximately 1.25892, and an ampli-
√ 1
tude (field) ratio of 10 10 (1.12202).[16][17]
Although permissible, the bel is rarely used with other SI
unit prefixes than deci. It is preferred to use hundredths
of a decibel rather than millibels.[18]
Conversions
The bel is defined by ISO Standard 80000-3:2006 as (1/2)
ln(10) nepers (Np), where ln denotes the natural loga-
rithm. Because the decibel is one tenth of a bel, it follows
that 1 dB = (1/20) ln(10) Np. The same standard defines
1 Np as equal to 1 (thereby relating all of the units as
nondimensional natural log of field-quantity ratios, 1 dB
= 0.11513..., 1 B = 1.1513...). Since logarithm differ-
ences measured in these units are used to represent power
ratios and field ratios, the values of the ratios represented
by each unit are also included in the table.
Power quantities
When referring to measurements of power or intensity, a
ratio can be expressed in decibels by evaluating ten times
the base-10 logarithm of the ratio of the measured quan-
tity to the reference level. Thus, the ratio of a power value
7.1. DECIBEL 165

P 1 to another power value P 0 is represented by L B, that where V 1 is the voltage being measured, V 0 is a specified
ratio expressed in decibels,[19] which is calculated using reference voltage, and G B is the power gain expressed in
the formula: decibels. A similar formula holds for current.
The term root-power quantity is introduced by ISO Stan-
( ) dard 80000-1:2009 as a substitute of field quantity. The
P1
LdB = 10 log10 term field quantity is deprecated by that standard.
P0

The base-10 logarithm of the ratio of the two power levels Examples
is the number of bels. The number of decibels is ten times
the number of bels (equivalently, a decibel is one-tenth of All of these examples yield dimensionless answers in dB
a bel). P 1 and P 0 must measure the same type of quantity, because they are relative ratios expressed in decibels. The
and have the same units before calculating the ratio. If P 1 unit dBW is often used to denote a ratio for which the
= P 0 in the above equation, then L B = 0. If P 1 is greater reference is 1 W, and similarly dBm for a 1 mW reference
than P 0 then L B is positive; if P 1 is less than P 0 then point.
L B is negative.
Rearranging the above equation gives the following for- • Calculating the ratio of 1 kW (one kilowatt, or 1000
mula for P 1 in terms of P 0 and L B: watts) to 1 W in decibels yields:
( )
1000 W
LdB GdB = 10 log10 ≡ 30 dB
P1 = 10 10 P0 1W
√
• The ratio of 1000 V ≈ 31.62 V to 1 V in decibels
Field quantities
is
When referring to measurements of field amplitude, it is ( )
31.62 V
usual to consider the ratio of the squares of A1 (mea- GdB = 20 log10 ≡ 30 dB
1V
sured amplitude) and A0 (reference amplitude). This is
because in most applications power is proportional to the (31.62 V/1 V)2 ≈ 1 kW/1 W , illustrating the conse-
square of amplitude, and it is desirable for the two deci- quence from the definitions above that GdB has the same
bel formulations to give the same result in such typical value, 30 dB , regardless of whether it is obtained from
cases. Thus, the following definition is used: powers or from amplitudes, provided that in the specific
system being considered power ratios are equal to ampli-
tude ratios squared.
( ) ( )
A21 A1
LdB = 10 log10 = 20 log10 .
A20 A0 • The ratio of 1 mW (one milliwatt) to 10 W in deci-
bels is obtained with the formula
The formula may be rearranged to give
( )
0.001 W
GdB = 10 log10 ≡ −40 dB
LdB 10 W
A1 = 10 20 A0
• The power ratio corresponding to a 3 dB change in
Similarly, in electrical circuits, dissipated power is typi- level is given by
cally proportional to the square of voltage or current when
the impedance is held constant. Taking voltage as an ex- 3
G = 10 10 × 1 = 1.99526... ≈ 2
ample, this leads to the equation:
A change in power ratio by a factor of 10 is a change of
10 dB. A change in power ratio by a factor of two is ap-
( )
V1 proximately a change of 3 dB. More precisely, the factor
GdB = 20 log10 is 103/10 , or 1.9953, about 0.24% different from exactly
V0
166
2. Similarly, an increase of 3 dB implies an increase in
√
voltage by a factor of approximately 2 , or about 1.41, an
increase of 6 dB corresponds to approximately four times
the power and twice the voltage, and so on. In exact terms
the power ratio is 106/10 , or about 3.9811, a relative error
of about 0.5%.
7.1.3 Properties
The decibel has the following properties:
• The logarithmic nature of the decibel means that a
very large range of ratios can be represented by a
convenient number, in a similar manner to scientific
notation. This allows one to clearly visualize huge
changes of some quantity. See Bode plot and semi-
log plot. For example, 120 dB SPL may be clearer
than a “a trillion times more intense than the thresh-
old of hearing”, or easier to interpret than “20 pas-
cals of sound pressure”.
• The unit is an additive function, which means that
the overall gain of a multi-component system, such
as a series of amplifier stages, can be calculated
by summing the gains in decibels of the individual
components, rather than multiply the amplification
factors; that is, log(A × B × C) = log(A) + log(B)
+ log(C). Practically, this means that, armed only
with the knowledge that 1 dB is approximately 26%
power gain, 3 dB is approximately 2× power gain,
and 10 dB is 10× power gain, it is possible to de-
termine the power ratio of a system from the gain
in dB with only simple addition and multiplication.
For example:
A system consists of 3 amplifiers in
series, with gains (ratio of power
out to in) of 10 dB, 8 dB, and 7
dB respectively, for a total gain of
25 dB. Broken into combinations
of 10, 3, and 1 dB, this is:
25 dB = 10 dB + 10 dB +
3 dB + 1 dB + 1 dB
With an input of 1 watt, the output
is approximately
1 W x 10 x 10 x 2 x 1.26
x 1.26 = ~317.5 W
CHAPTER 7. THE DECIBEL
Calculated exactly, the output is 1
W x 1025/10 = 316.2 W. The ap-
proximate value has an error of
only +0.4% with respect to the ac-
tual value which is negligible given
the precision of the values supplied
and the accuracy of most measure-
ment instrumentation.
7.1.4 Advantages and disadvantages
Advantages
• According to Mitschke,[20] “The advantage of using
a logarithmic measure is that in a transmission chain,
there are many elements concatenated, and each has
its own gain or attenuation. To obtain the total, addi-
tion of decibel values is much more convenient than
multiplication of the individual factors.”
• The human perception of the intensity of, for exam-
ple, sound or light, is more nearly linearly related to
the logarithm of intensity than to the intensity itself,
per the Weber–Fechner law, so the dB scale can be
useful to describe perceptual levels or level differ-
ences.
Disadvantages
According to several articles published in Electrical En-
gineering[21] and the Journal of the Acoustical Society of
America,[22][23][24] the decibel suffers from the following
disadvantages:
• The decibel creates confusion.
• The logarithmic form obscures reasoning.
• Decibels are more related to the era of slide rules
than that of modern digital processing.
• Decibels are cumbersome and difficult to interpret.
Hickling concludes “Decibels are a useless affectation,
which is impeding the development of noise control as
an engineering discipline”.[23]
Another disadvantage is that quantities in decibels are not
necessarily additive,[25][26] thus being “of unacceptable
form for use in dimensional analysis extquotedbl.[27]
7.1. DECIBEL 167

For the same reason that decibels excel at multiplicative that the ear can hear is greater than or equal to 1 tril-
operations (e.g., antenna gain), they are awkward when lion (1012 ).[30] Such large measurement ranges are con-
dealing with additive operations. Peters (2013, p. 13)[28] veniently expressed in logarithmic units: the base-10 log-
provides several examples: arithm of 1012 is 12, which is expressed as a sound pres-
sure level of 120 dB re 20 micropascals. Since the hu-
• “if two machines each individually produce a [sound man ear is not equally sensitive to all sound frequencies,
pressure] level of, say, 90 dB at a certain point, then noise levels at maximum human sensitivity, somewhere
when both are operating together we should expect between 2 and 4 kHz, are factored more heavily into
the combined sound pressure level to increase to 93 some measurements using frequency weighting. (See also
dB, but certaintly not to 180 dB! extquotedbl Stevens’ power law.)

• “supposed that the noise from a machine is mea- Further information: Examples of sound pressure and
sured (including the contribution of background sound pressure levels
noise) and found to be 87 dBA but when the ma-
chine is switched off the background noise alone is
measured as 83 dBA. ... the machine noise [level
Electronics
(alone)] may be obtained by 'subtracting' the 83
dBA background noise from the combined level of
In electronics, the decibel is often used to express power
87 dBA; i.e., 84.8 dBA.”
or amplitude ratios (gains), in preference to arithmetic
• “in order to find a representative value of the sound ratios or percentages. One advantage is that the total
level in a room a number of measurements are taken decibel gain of a series of components (such as amplifiers
at different positions within the room, and an aver- and attenuators) can be calculated simply by summing the
age value is calculated. (...) Compare the logarith- decibel gains of the individual components. Similarly, in
mic and arithmetic averages of ... 70 dB and 90 dB: telecommunications, decibels denote signal gain or loss
logarithmic average = 87 dB; arithmetic average = from a transmitter to a receiver through some medium
80 dB.” (free space, waveguide, coaxial cable, fiber optics, etc.)
using a link budget.
The decibel unit can also be combined with a suffix to
7.1.5 Uses create an absolute unit of electric power. For example,
it can be combined with “m” for “milliwatt” to produce
Acoustics
the extquotedbldBm extquotedbl. Zero dBm is the level
corresponding to one milliwatt, and 1 dBm is one decibel
The decibel is commonly used in acoustics as a unit
greater (about 1.259 mW).
of sound pressure level, for a reference pressure of 20
micropascals in air[29] and 1 micropascal in water. The In professional audio specifications, a popular unit is the
reference pressure in air is set at the typical threshold of dBu. The suffix u stands for unloaded, and was probably
perception of an average human and there are common chosen to be similar to lowercase v, as dBv was the older
comparisons used to illustrate different levels of sound name for the same unit. It was changed to avoid con-
pressure. Sound pressure is a field quantity, therefore the fusion with dBV. The dBu is a root mean square (RMS)
field version of the unit definition is used: measurement of voltage that uses as its reference approx-
( ) imately 0.775 VRMS. Chosen for historical reasons, the
prms
Lp = 20 log10 pref dB reference value is the voltage level which delivers 1 mW
of power in a 600 ohm resistor, which used to be the stan-
where pᵣₑ is equal to the standard reference dard reference impedance in telephone circuits.
sound pressure level of 20 micropascals in air
or 1 micropascal in water.
Optics
The human ear has a large dynamic range in audio recep-
tion. The ratio of the sound intensity that causes perma- In an optical link, if a known amount of optical power, in
nent damage during short exposure to the quietest sound dBm (referenced to 1 mW), is launched into a fiber, and
168 CHAPTER 7. THE DECIBEL

the losses, in dB (decibels), of each electronic component In cases such as this, where the numerical value of the
(e.g., connectors, splices, and lengths of fiber) are known, reference is explicitly and exactly stated, the decibel mea-
the overall link loss may be quickly calculated by addition surement is called an “absolute” measurement, in the
and subtraction of decibel quantities.[31] sense that the exact value of the measured quantity can
In spectrometry and optics, the blocking unit used to be recovered using the formula given earlier. If the nu-
measure optical density is equivalent to −1 B. merical value of the reference is not explicitly stated, as
in the dB gain of an amplifier, then the decibel measure-
ment is purely relative.
Video and digital imaging The SI does not permit attaching qualifiers to units,
whether as suffix or prefix, other than standard SI pre-
In connection with video and digital image sensors, deci- fixes. Therefore, even though the decibel is accepted for
bels generally represent ratios of video voltages or dig- use alongside SI units, the practice of attaching a suffix to
itized light levels, using 20 log of the ratio, even when the basic dB unit, forming compound units such as dBm,
the represented optical power is directly proportional to dBu, dBA, etc., is not.[13] The proper way, according to
the voltage or level, not to its square, as in a CCD imager the IEC 60027-3,[12] is either as Lx (re xᵣₑ ) or as Lx/xᵣₑ ,
where response voltage is linear in intensity.[32] Thus, a where x is the quantity symbol and xᵣₑ is the value of the
camera signal-to-noise ratio or dynamic range of 40 dB reference quantity, e.g., LE (re 1 μV/m) = LE/₍₁ μV/ ₎ for
represents a power ratio of 100:1 between signal power the electric field strength E relative to 1 μV/m reference
and noise power, not 10,000:1.[33] Sometimes the 20 log value.
ratio definition is applied to electron counts or photon
counts directly, which are proportional to intensity with- Outside of documents adhering to SI units, the practice
out the need to consider whether the voltage response is is very common as illustrated by the following examples.
linear.[34] There is no general rule, with various discipline-specific
practices. Sometimes the suffix is a unit symbol (“W”,
However, as mentioned above, the 10 log intensity con- extquotedblK”, extquotedblm”), sometimes it’s a translit-
vention prevails more generally in physical optics, includ- eration of a unit symbol (“uV” instead of μV for micro
ing fiber optics, so the terminology can become murky volt), sometimes it’s an acronym for the units name (“sm”
between the conventions of digital photographic technol- for m2 , “m” for mW), other times it’s a mnemonic for
ogy and physics. Most commonly, quantities called “dy- the type of quantity being calculated (“i” for antenna gain
namic range” or “signal-to-noise” (of the camera) would w.r.t. an isotropic antenna, extquotedblλ extquotedbl for
be specified in 20 log dBs, but in related contexts (e.g. anything normalized by the EM wavelength), or other-
attenuation, gain, intensifier SNR, or rejection ratio) the wise a general attribute or identifier about the nature of
term should be interpreted cautiously, as confusion of the the quantity (“A” for A-weighted sound pressure level).
two units can result in very large misunderstandings of the The suffix is often connected with a dash (dB-Hz), with
value. a space (dB HL), with no intervening character (dBm),
Photographers also often use an alternative base-2 log or enclosed in parentheses, dB(sm).
unit, the f-stop, and in software contexts these image level
ratios, particularly dynamic range, are often loosely re- Voltage
ferred to by the number of bits needed to represent the
quantity, such that 60 dB (digital photographic) is roughly Since the decibel is defined with respect to power, not
equal to 10 f-stops or 10 bits, since 103 is nearly equal to amplitude, conversions of voltage ratios to decibels must
210 . square the amplitude, or use the factor of 20 instead of
10, as discussed above.
7.1.6 Suffixes and reference values dBV

Suffixes are commonly attached to the basic dB unit in dB(VRMS) – voltage relative to 1 volt, regard-
order to indicate the reference value against which the less of impedance.[2]
decibel measurement is taken. For example, dBm indi-
cates power measurement relative to 1 milliwatt. dBu or dBv
7.1. DECIBEL
A schematic showing the relationship between dBu (the voltage of 20, and the measures of power (e.g. dB SIL and dB
source) and dBm (the power dissipated as heat by the 600 Ω SWL) use the factor of 10.
resistor)
√
RMS voltage relative to 0.6 V ≈ dB SPL (sound pressure level) – for sound in
0.7746 V ≈ −2.218 dBV .[2] Originally
dBv, it was changed to dBu to avoid confusion
with dBV.[35] The “v” comes from “volt”,
while “u” comes from “unloaded”. dBu can be
used regardless of impedance, but is derived
from a 600 Ω load dissipating 0 dBm (1
mW). The reference√ voltage comes from the
computation V = 600 Ω · 0.001 W .
In professional audio, equipment may be cali-
brated to indicate a “0” on the VU meters some
finite time after a signal has been applied at an
amplitude of +4 dBu. Consumer equipment
will more often use a much lower “nominal”
signal level of −10 dBV.[36] Therefore, many
devices offer dual voltage operation (with dif-
ferent gain or “trim” settings) for interoperabil-
ity reasons. A switch or adjustment that cov-
ers at least the range between +4 dBu and −10
dBV is common in professional equipment.
dBmV
dB(mVRMS) – voltage relative to 1 millivolt
across 75 Ω.[37] Widely used in cable television
networks, where the nominal strength of a sin-
gle TV signal at the receiver terminals is about
0 dBmV. Cable TV uses 75 Ω coaxial cable, so
0 dBmV corresponds to −78.75 dBW (−48.75
dBm) or ~13 nW.
dBμV or dBuV
dB(μVRMS) – voltage relative to 1 microvolt.
Widely used in television and aerial amplifier
specifications. 60 dBμV = 0 dBmV.
169
Acoustics
Probably the most common usage of “decibels” in refer-
ence to sound level is dB SPL, sound pressure level ref-
erenced to the nominal threshold of human hearing:[38]
The measures of pressure (a field quantity) use the factor
dB SPL
air and other gases, relative to 20 micropascals
(μPa) = 2×10−5 Pa, approximately the quietest
sound a human can hear. For sound in water
and other liquids, a reference pressure of 1 μPa
is used.[39]
An RMS sound pressure of one pascal corresponds to a
level of 94 dB SPL.
dB SIL
dB sound intensity level – relative to 10−12
W/m2 , which is roughly the threshold of hu-
man hearing in air.
dB SWL
dB sound power level – relative to 10−12 W.
dB(A), dB(B), and dB(C)
These symbols are often used to denote the use
of different weighting filters, used to approx-
imate the human ear’s response to sound, al-
though the measurement is still in dB (SPL).
These measurements usually refer to noise and
noisome effects on humans and animals, and
are in widespread use in the industry with re-
gard to noise control issues, regulations and en-
vironmental standards. Other variations that
may be seen are dBA or dBA. According to
ANSI standards, the preferred usage is to write
LA = x dB. Nevertheless, the units dBA and
dB(A) are still commonly used as a shorthand
for A-weighted measurements. Compare dBc,
used in telecommunications.
170
dB HL or dB hearing level is used in audiograms as a
measure of hearing loss. The reference level varies with
frequency according to a minimum audibility curve as de-
fined in ANSI and other standards, such that the resulting
audiogram shows deviation from what is regarded as 'nor-
mal' hearing.
dB Q is sometimes used to denote weighted noise level, dBsm
commonly using the ITU-R 468 noise weighting
Audio electronics
dBm
dB(mW) – power relative to 1 milliwatt. In au-
dio and telephony, dBm is typically referenced
relative to a 600 ohm impedance,[40] while in Radio power, energy, and field strength
radio frequency work dBm is typically refer-
enced relative to a 50 ohm impedance.[41] dBc dBc – relative to carrier—in telecommunications,
dBFS
dB(full scale) – the amplitude of a signal com-
pared with the maximum which a device can
handle before clipping occurs. Full-scale may
be defined as the power level of a full-scale
sinusoid or alternatively a full-scale square
wave. A signal measured with reference to
a full-scale sine-wave will appear 3dB weaker
when referenced to a full-scale square wave,
thus: 0 dBFS(ref=fullscale sine wave) = −3
dBFS(ref=fullscale square wave).
dBTP
dB(true peak) - peak amplitude of a signal
compared with the maximum which a device
can handle before clipping occurs.[42] In dig-
ital systems, 0 dBTP would equal the highest
level (number) the processor is capable of rep-
resenting. Measured values are always nega-
tive or zero, since they are less than or equal to
full-scale.
Radar
dBZ
CHAPTER 7. THE DECIBEL
dB(Z) – decibel relative to Z = 1 mm6 m−3 :[43]
energy of reflectivity (weather radar), related
to the amount of transmitted power returned
to the radar receiver. Values above 15–20 dBZ
usually indicate falling precipitation.[44]
dB(m2 ) – decibel relative to one square meter:
measure of the radar cross section (RCS) of
a target. The power reflected by the target is
proportional to its RCS. “Stealth” aircraft and
insects have negative RCS measured in dBsm,
large flat plates or non-stealthy aircraft have
positive values.[45]
this indicates the relative levels of noise or sideband
power, compared with the carrier power. Compare
dBC, used in acoustics.
dBJ dB(J) – energy relative to 1 joule. 1 joule = 1 watt
second = 1 watt per hertz, so power spectral density
can be expressed in dBJ.
dBm dB(mW) – power relative to 1 milliwatt. Tradi-
tionally associated with the telephone and broad-
casting industry to express audio-power levels ref-
erenced to one milliwatt of power, normally with a
600 ohm load, which is a voltage level of 0.775 volts
or 775 millivolts. This is still commonly used to
express audio levels with professional audio equip-
ment.
In the radio field, dBm is usually referenced to a 50 ohm
load, with the resultant voltage being 0.224 volts.
dBμV/m or dBuV/m dB(μV/m) – electric field
strength relative to 1 microvolt per meter. Often
used to specify the signal strength from a television
broadcast at a receiving site (the signal measured at
the antenna output will be in dBμV).
dBf dB(fW) – power relative to 1 femtowatt.
dBW dB(W) – power relative to 1 watt.
dBk dB(kW) – power relative to 1 kilowatt.
7.1. DECIBEL
Antenna measurements
dBi
dB(isotropic) – the forward gain of an antenna
compared with the hypothetical isotropic an-
tenna, which uniformly distributes energy in all
directions. Linear polarization of the EM field
is assumed unless noted otherwise.
dBd
dB(dipole) – the forward gain of an antenna
compared with a half-wave dipole antenna. 0
dBd = 2.15 dBi
dBiC
dB(isotropic circular) – the forward gain of
an antenna compared to a circularly polarized
isotropic antenna. There is no fixed conversion
rule between dBiC and dBi, as it depends on
the receiving antenna and the field polarization.
dBq
dB(quarterwave) – the forward gain of an an-
tenna compared to a quarter wavelength whip.
Rarely used, except in some marketing mate-
rial. 0 dBq = −0.85 dBi
dBsm
dB(m2 ) – decibel relative to one square meter:
measure of the antenna effective area.[46]
dBm−1
dB(m−1 ) – decibel relative to reciprocal of me-
ter: measure of the antenna factor.
Other measurements
dB-Hz
dB(Hz) – bandwidth relative to one hertz. E.g.,
20 dB-Hz corresponds to a bandwidth of 100
171
Hz. Commonly used in link budget calcula-
tions. Also used in carrier-to-noise-density ra-
tio (not to be confused with carrier-to-noise ra-
tio, in dB).
dBov or dBO
dB(overload) – the amplitude of a signal (usu-
ally audio) compared with the maximum which
a device can handle before clipping occurs.
Similar to dBFS, but also applicable to analog
systems.
dBr
dB(relative) – simply a relative difference from
something else, which is made apparent in con-
text. The difference of a filter’s response to
nominal levels, for instance.
dBrn
dB above reference noise. See also dBrnC
dBrnC
dBrnC represents an audio level measure-
ment, typically in a telephone circuit, rela-
tive to the circuit noise level, with the mea-
surement of this level frequency-weighted by a
standard C-message weighting filter. The C-
message weighting filter was chiefly used in
North America. The Psophometric filter is
used for this purpose on international circuits.
See Psophometric weighting to see a compar-
ison of frequency response curves for the C-
message weighting and Psophometric weight-
ing filters.[47]
dBK
dB(K) – decibels relative to kelvin: Used to
express noise temperature.[48]
dB/K
dB(K−1 ) – decibels relative to reciprocal of
kelvin [49] -- not decibels per kelvin: Used for
the G/T factor, a figure of merit utilized in
satellite communications, relating the antenna
gain G to the receiver system noise equivalent
temperature T.[50][51]
172
7.1.7 Related units
mBm mB(mW) – power relative to 1 milliwatt, in mil-
libels (one hundredth of a decibel). 100 mBm =
1dBm. This unit is in the Wi-Fi drivers of the Linux
kernel[52] and the regulatory domain sections.[53]
Np or cNp
Another closely related unit is the neper (Np)
or centineper (cNp). Like the decibel, the
neper is a unit of level.[54] The linear approxi-
mation 1cNp =~ 1% for small percentage dif-
ferences is widely used finance.
1 Np = 20 log10 e dB ≈ 8.685889638 dB
7.1.8 Fractions
Attenuation constants, in fields such as optical fiber com-
munication and radio propagation path loss, are often ex-
pressed as a fraction or ratio to distance of transmission.
dB/m means decibels per meter, dB/mi is decibels per
mile, for example. These quantities are to be manipu-
lated obeying the rules of dimensional analysis, e.g., a
100-meter run with a 3.5 dB/km fiber yields a loss of
0.35 dB = 3.5 dB/km × 0.1 km.
7.1.9 See also
• Apparent magnitude
• Cent in music
• dB drag racing
• Equal-loudness contour
• Noise (environmental)
• Phon
• Richter magnitude scale
• Signal noise
• Sone
CHAPTER 7. THE DECIBEL
7.1.10 Notes and references
[1] IEEE Standard 100 Dictionary of IEEE Standards Terms,
Seventh Edition, The Institute of Electrical and Electron-
ics Engineering, New York, 2000; ISBN 0-7381-2601-2;
page 288
[2] Analog Devices : Virtual Design Center : Interactive De-
sign Tools : Utilities : VRMS / dBm / dBu / dBV calcu-
lator
[3] “ISO 80000-3:2006”. International Organization for
Standardization. Retrieved 20 July 2013.
[4] Johnson, Kenneth Simonds (1944). Transmission Circuits
for Telephonic Communication: Methods of Analysis and
Design. New York: D. Van Nostrand Co. p. 10.
[5] Don Davis and Carolyn Davis (1997). Sound system engi-
neering (2nd ed.). Focal Press. p. 35. ISBN 978-0-240-
80305-0.
[6] Bell Labs (1925). Transmission Circuits for Telephonic
Communication.
[7] R. V. L. Hartley (Dec 1928). extquotedbl'TU' becomes
'Decibel' extquotedbl. Bell Laboratories Record (AT&T)
7 (4): 137–139.
[8] Martin, W. H. (January 1929). “DeciBel—The New
Name for the Transmission Unit”. Bell System Technical
Journal 8 (1).
[9] 100 Years of Telephone Switching, p. 276, Robert J. Cha-
puis, Amos E. Joel, 2003
[10] William H. Harrison (1931). “Standards for Transmis-
sion of Speech”. Standards Yearbook (National Bureau
of Standards, U. S. Govt. Printing Office) 119
[11] Consultative Committee for Units, Meeting minutes, Sec-
tion 3
[12] “Letter symbols to be used in electrical technology –
Part 3: Logarithmic and related quantities, and their
units”, IEC 60027-3 Ed. 3.0, International Electrotech-
nical Commission, 19 July 2002.
[13] Thompson, A. and Taylor, B. N. sec 8.7, “Logarithmic
quantities and units: level, neper, bel”, Guide for the Use
of the International System of Units (SI) 2008 Edition,
NIST Special Publication 811, 2nd printing (November
2008), SP811 PDF
[14] “International Standard CEI-IEC 27-3 Letter symbols to
be used in electrical technology Part 3: Logarithmic quan-
tities and units”. International Electrotechnical Commis-
sion.
7.1. DECIBEL
[15] Brian C.J. Moore (1995). Hearing. Academic Press. p.
11. ISBN 9780080533865.
[16] Mark, James E., Physical properties of polymers hand-
book, Springer, 2007, p 1025: extquotedbl… the decibel
represents a reduction in power of 1.258 times.”
[17] Yost, William, Fundamentals of hearing: an introduction,
Holt, Rinehart and Winston, 1985, p 206: extquotedbl…
a pressure ratio of 1.122 equals +1.0 dB”
[18] Fedor Mitschke, Fiber Optics: Physics and Technology,
Springer, 2010 ISBN 3642037038.
[19] David M. Pozar (2005). Microwave Engineering (3rd ed.).
Wiley. p. 63. ISBN 978-0-471-44878-5.
[20] Fiber Optics (Springer, 2010)
[21] C W Horton, “The bewildering decibel”, Elec. Eng., 73,
550-555 (1954).
[22] C S Clay (1999), Underwater sound transmission and SI
units, J Acoust Soc Am 106, 3047
[23] R Hickling (1999), Noise Control and SI Units, J Acoust
Soc Am 106, 3048
[24] D M F Chapman (2000), Decibels, SI units, and stan-
dards, J Acoust Soc Am 108, 480
[25] Nicholas P. Cheremisinoff (1996) Noise Control in Indus-
try: A Practical Guide, Elsevier, 203 pp, p.
[26] Andrew Clennel Palmer (2008), Dimensional Analysis
and Intelligent Experimentation, World Scientific, 154
pp, p.13
[27] J.C. Gibbings, Dimensional Analysis, p.37, Springer,
2011 ISBN 1849963177.
[28] R J Peters, Acoustics and Noise Control, Routledge, Nov
12, 2013, 400 pages
[29] “Electronic Engineer’s Handbook” by Donald G. Fink,
Editor-in-Chief ISBN 0-07-020980-4 Published by Mc-
Graw Hill, page 19-3
[30] National Institute on Deafness and Other Communica-
tions Disorders, Noise-Induced Hearing Loss (National In-
stitutes of Health, 2008).
[31] Bob Chomycz (2000). Fiber optic installer’s field manual.
McGraw-Hill Professional. pp. 123–126. ISBN 978-0-
07-135604-6.
[32] Stephen J. Sangwine and Robin E. N. Horne (1998). The
Colour Image Processing Handbook. Springer. pp. 127–
130. ISBN 978-0-412-80620-9.
173
[33] Francis T. S. Yu and Xiangyang Yang (1997).
Introduction to optical engineering. Cambridge Uni-
versity Press. pp. 102–103. ISBN 978-0-521-57493-8.
[34] Junichi Nakamura (2006). “Basics of Image Sensors”. In
Junichi Nakamura. Image sensors and signal processing
for digital still cameras. CRC Press. pp. 79–83. ISBN
978-0-8493-3545-7.
[35] What is the difference between dBv, dBu, dBV, dBm, dB
SPL, and plain old dB? Why not just use regular voltage
and power measurements? – rec.audio.pro Audio Profes-
sional FAQ
[36] deltamedia.com. “DB or Not DB”. Deltamedia.com. Re-
trieved 2013-09-16.
[37] The IEEE Standard Dictionary of Electrical and Electron-
ics terms (6th ed.). IEEE. 1996 [1941]. ISBN 1-55937-
833-6.
[38] Jay Rose (2002). Audio postproduction for digital video.
Focal Press,. p. 25. ISBN 978-1-57820-116-7.
[39] Morfey, C. L. (2001). Dictionary of Acoustics. Academic
Press, San Diego.
[40] Bigelow, Stephen. Understanding Telephone Electronics.
Newnes. p. 16. ISBN 978-0750671750.
[41] Carr, Joseph (2002). RF Components and Circuits.
Newnes. pp. 45–46. ISBN 978-0750648448.
[42] ITU-R BS.1770
[43] “Glossary: D’s”. National Weather Service. Retrieved
2013-04-25.
[44] “Radar FAQ from WSI”. Archived from the original on
2008-05-18. Retrieved 2008-03-18.
[45] “Definition at Everything2”. Retrieved 2008-08-06.
[46] EW 102: A Second Course in Electronic Warfare - David
Adamy - Google Livros. Books.google.com.br. Retrieved
2013-09-16.
[47] dBrnC is defined on page 230 in “Engineering and Op-
erations in the Bell System,” (2ed), R.F. Rey (technical
editor), copyright 1983, AT&T Bell Laboratories, Mur-
ray Hill, NJ, ISBN 0-932764-04-5
[48] Satellite Communication: Concepts And Applications -
K. N. Raja Rao - Google Livros. Books.google.com.br.
2013-01-31. Retrieved 2013-09-16.
[49] Comprehensive Glossary of Telecom Abbreviations
and Acronyms - Ali Akbar Arabi - Google Livros.
Books.google.com.br. Retrieved 2013-09-16.
174
[50] The Digital Satellite TV Handbook - Mark E. Long - Google
Livros. Books.google.com.br. Retrieved 2013-09-16.
[51] Reference Data for Engineers: Radio, Electronics, Com-
puters and Communications - Mac E. Van Valkenburg -
Google Livros. Books.google.com.br. 2001-10-19. Re-
trieved 2013-09-16.
[52] setting the TX power for a Wi-Fi device in Linux showing
units in mBm
[53] kernel notification of change in regulatory domain show-
ing units in mBm
[54] “ISO 80000-3:2006”. International Organization for
Standardization. Retrieved 20 July 2013.
7.1.11 External links
• What is a decibel? With sound files and animations
• Conversion of sound level units: dBSPL or dBA to
sound pressure p and sound intensity J
• OSHA Regulations on Occupational Noise Expo-
sure
7.2 Noise (electronics)
This article is about fundamental noise processes. For
noise arising from outside sources, see electromagnetic
compatibility and electromagnetic interference. While noise is generally unwanted, it can serve a useful
In electronics, noise is a random fluctuation in an elec- purpose in some applications, such as random number
generation or dithering.
Analog display of random fluctuations in voltage: e.g., pink
noise.
CHAPTER 7. THE DECIBEL
trical signal, a characteristic of all electronic circuits.[1]
Noise generated by electronic devices varies greatly,
as it can be produced by several different effects.
Thermal noise is unavoidable at non-zero temperature
(see fluctuation-dissipation theorem), while other types
depend mostly on device type (such as shot noise,[1][2]
which needs steep potential barrier) or manufacturing
quality and semiconductor defects, such as conductance
fluctuations, including 1/f noise.
In communication systems, noise is an error or undesired
random disturbance of a useful information signal in a
communication channel. The noise is a summation of
unwanted or disturbing energy from natural and some-
times man-made sources. Noise is, however, typically
distinguished from interference, (e.g. cross-talk, delib-
erate jamming or other unwanted electromagnetic inter-
ference from specific transmitters), for example in the
signal-to-noise ratio (SNR), signal-to-interference ratio
(SIR) and signal-to-noise plus interference ratio (SNIR)
measures. Noise is also typically distinguished from
distortion, which is an unwanted systematic alteration of
the signal waveform by the communication equipment,
for example in the signal-to-noise and distortion ratio
(SINAD). In a carrier-modulated passband analog com-
munication system, a certain carrier-to-noise ratio (CNR)
at the radio receiver input would result in a certain signal-
to-noise ratio in the detected message signal. In a digi-
tal communications system, a certain E /N 0 (normalized
signal-to-noise ratio) would result in a certain bit error
rate (BER).
7.2.1 Noise types
Thermal noise
Main article: Johnson–Nyquist noise
Johnson–Nyquist noise[1] (sometimes thermal, Johnson
or Nyquist noise) is unavoidable, and generated by
the random thermal motion of charge carriers (usually
electrons), inside an electrical conductor, which happens
regardless of any applied voltage.
Thermal noise is approximately white, meaning that its
power spectral density is nearly equal throughout the
7.2. NOISE (ELECTRONICS) 175

frequency spectrum. The amplitude of the signal has sively within the material; the electrons do not have dis-
very nearly a Gaussian probability density function. A crete arrivial times. Shot noise has been demonstrated
communication system affected by thermal noise is often in mesoscopic resistors when the size of the resistive el-
modeled as an additive white Gaussian noise (AWGN) ement becomes shorter than the electron-phonon scatter-
channel. ing length.[3]
The root mean square (RMS) voltage due to thermal noise
vn , generated in a resistance R (ohms) over bandwidth Flicker noise
Δf (hertz), is given by
Main article: Flicker noise
√
vn = 4kB T R∆f
Flicker noise, also known as 1/f noise, is a signal or pro-
where kB is Boltzmann’s constant (joules per kelvin) and cess with a frequency spectrum that falls off steadily into
T is the resistor’s absolute temperature (kelvin). the higher frequencies, with a pink spectrum. It occurs
in almost all electronic devices, and results from a variety
As the amount of thermal noise generated depends upon of effects, though always related to a direct current.
the temperature of the circuit, very sensitive circuits such
as preamplifiers in radio telescopes are sometimes cooled
in liquid nitrogen to reduce the noise level. Intermodulation noise

This type of noise is caused when signals of different fre-

Shot noise quencies share the same medium.
Main article: Shot noise
Crosstalk
Shot noise in electronic devices results from unavoid-
able random statistical fluctuations of the electric current This is unwanted coupling of signals.
when the charge carriers (such as electrons) traverse a
gap. The current is a flow of discrete charges, and the Impulse noise
fluctuation in the arrivals of those charges creates shot
noise. Shot noise is similar to the noise created by rain short peak of noise example Lightning, electrical distur-
falling on a tin roof. The flow of rain may be relatively bances,flaws in communication system
constant, but the raindrops arrive discretely.
The root-mean-square value of the shot noise current in
is given by the Schottky formula Interference

Contamination by various signals from human sources

√ example Power lines transmitters It does not disappear
in = 2Iq∆B
when signal is switched off.
where I is the DC current, q is the charge of an electron,
and ΔB is the bandwidth in hertz.
Burst noise
The shot noise assumes independent arrivals. Vacuum
tubes have shot noise because the electrons randomly Main article: Burst noise
leave the cathode and arrive at the anode (plate). A tube
may not exhibit the full shot noise effect: the presence of
Burst noise consists of sudden step-like transitions be-
a space charge tends to smooth out the arrival times (and
tween two or more levels (non-Gaussian), as high as sev-
thus reduce the randomness of the current). eral hundred microvolts, at random and unpredictable
Conductors and resistors typically do not exhibit shot times. Each shift in offset voltage or current lasts for sev-
noise because the electrons thermalize and move diffu- eral milliseconds, and the intervals between pulses tend
176
to be in the audio range (less than 100 Hz), leading to the fluorescent lamps cause industrial noise. These noises are
term popcorn noise for the popping or crackling sounds it produced by the discharge present in all these operations.
produces in audio circuits.
Transit-time noise
If the time taken by the electrons from traveling from
emitter to collector becomes comparable to the period
of the signal being amplified, that is, at frequencies above
VHF and beyond, so-called transit-time effect takes place
and noise input admittance of the transistor increases.
From the frequency at which this effect becomes signif-
icant it goes on increasing with frequency and quickly
dominates over other terms.
Avalanche noise
Avalanche noise is the noise produced when a junction
diode is operated at the onset of avalanche breakdown, a
semiconductor junction phenomenon in which carriers in
a high voltage gradient develop sufficient energy to dis-
lodge additional carriers through physical impact, creat-
ing ragged current flows.
7.2.2 Coupled noise
See also: Electromagnetic compatibility
Energy external of the receiver can couple noise, also by change a noisy altered output signal to a more theoretical
energy conversion. Generally this is done by fundamental output signal.
interaction, in electronics mainly by Inductive coupling
and/or capacitive coupling.
Atmospheric noise (static noise)
Main article: Atmospheric noise
This noise is also called static noise and it is the natural
source of disturbance caused by lightning discharge of
in thunderstorm and the natural(electrical) disturbances
occurring in the nature.
Industrial noise
Sources such as automobiles, aircraft, ignition elec-
tric motors and switching gear, High voltage wires and
CHAPTER 7. THE DECIBEL
Extraterrestrial noise
Noise from outside the Earth includes:
Solar noise Noise that originates from the Sun is called
solar noise. Under normal conditions there is constant
radiation from the Sun due to its high temperature. Elec-
trical disturbances such as corona discharges, as well as
sunspots can produce additional noise.
Cosmic noise Main article: Cosmic noise
Distant stars generate noise called cosmic noise. While
these stars are too far away to individually affect terres-
trial communications systems, their large number leads
to appreciable collective effects. Cosmic noise has been
observed in a range from 8 MHz to 1.43 GHz.
Reduction of noise coupling
In many cases noise found on a signal in a circuit is un-
wanted. When creating a circuit, one usually wants a
true output of what the circuit has accomplished. There
are many different noise reduction techniques that can
1. Faraday cage – A Faraday cage is a good way to re-
duce the overall noise in a complete circuit. The
Faraday cage can be thought of as an enclosure that
separates the complete circuit from outside power
lines and any other signal that may alter the true sig-
nal. A Faraday cage will usually block out most elec-
tromagnetic and electrostatic noise.
2. Capacitive coupling – A current through two resis-
tors, or any other type of conductor, close to each
other in a circuit can create unwanted capacitive
coupling. If this happens an AC signal from one
part of the circuit can be accidentally picked up in
another part. The two resistors (conductors) act like
a capacitor thus transferring AC signals. There may
be other reasons for which capacitive coupling is
7.2. NOISE (ELECTRONICS)
wanted but then it would not be thought of as elec- 7.2.3
tronic noise.
3. Ground loops – When grounding a circuit, it is im-
portant to avoid ground loops. Ground loops occur
when there is a voltage drop between the two ground
potentials. Since ground is thought of as 0V, the
presence of a voltage is undesirable at any point of
a ground bus. If this is the case, it would not be a
true ground. A good way to fix this is to bring all
the ground wires to the same potential in a ground
bus.
4. Shielding cables – In general, using shielded cables
to protect the wires from unwanted noise frequen-
cies in a sensitive circuit is good practice. A shielded
wire can be thought of as a small Faraday cage for a
specific wire as it uses a plastic or rubber enclosing
the true wire. Just outside of the rubber/plastic cov-
ering is a conductive metal that intercepts any noise
signal. Because the conductive metal is grounded,
the noise signal runs straight to ground before ever
getting to the true wire. It is important to ground
the shield at only one end to avoid a ground loop on
the shield.
5. Twisted pair wiring – Twisting wires very tightly to-
gether in a circuit will dramatically reduce electro-
magnetic noise. Twisting the wires decreases the
loop size in which a magnetic field can run through
to produce a current between the wires. Even if
the wires are twisted very tightly, there may still be
small loops somewhere between them, but because
they are twisted the magnetic field going through the Noise power is measured in watts or decibels (dB) relative
smaller loops induces a current flowing in opposite to a standard power, usually indicated by adding a suffix
ways in each wire and thus cancelling them out.
6. Notch filters – Notch filters or band-rejection filters dBrn(144-line).
are essential when eliminating a specific noise fre-
quency. For example, in most cases the power lines
within a building run at 60 Hz. Sometimes a sen-
sitive circuit will pick up this 60 Hz noise through
some unwanted antenna (could be as simple as a
wire in the circuit). Running the output through a
notch filter at 60 Hz will amplify the desired signal
without amplifying the 60 Hz noise. So in a sense
the noise will be lost at the output of the filter.
177
Quantification
The noise level in an electronic system is typically mea-
sured as an electrical power N in watts or dBm, a
root mean square (RMS) voltage (identical to the noise
standard deviation) in volts, dBμV or a mean squared er-
ror (MSE) in volts squared. Noise may also be charac-
terized by its probability distribution and noise spectral
density N 0 (f) in watts per hertz.
A noise signal is typically considered as a linear ad-
dition to a useful information signal. Typical signal
quality measures involving noise are signal-to-noise ratio
(SNR or S/N), signal-to-quantization noise ratio (SQNR)
in analog-to-digital conversion and compression, peak
signal-to-noise ratio (PSNR) in image and video cod-
ing, E /N 0 in digital transmission, carrier to noise ratio
(CNR) before the detector in carrier-modulated systems,
and noise figure in cascaded amplifiers.
Noise is a random process, characterized by stochastic
properties such as its variance, distribution, and spectral
density. The spectral distribution of noise can vary with
frequency, so its power density is measured in watts per
hertz (W/Hz). Since the power in a resistive element is
proportional to the square of the voltage across it, noise
voltage (density) can be described by taking the square
root of the noise power density, resulting in volts per
√
root hertz ( V/ Hz ). Integrated circuit devices, such as
operational amplifiers commonly quote equivalent input
noise level in these terms (at room temperature).
after dB. Examples of electrical noise-level measurement
units are dBu, dBm0, dBrn, dBrnC, and dBrn(f 1 − f 2 ),
Noise levels are usually viewed in opposition to signal lev-
els and so are often seen as part of a signal-to-noise ratio
(SNR). Telecommunication systems strive to increase the
ratio of signal level to noise level in order to effectively
transmit data. In practice, if the transmitted signal falls
below the level of the noise (often designated as the noise
floor) in the system, data can no longer be decoded at the
receiver. Noise in telecommunication systems is a prod-
uct of both internal and external sources to the system.
178 CHAPTER 7. THE DECIBEL

7.2.4 Dither • Scherz, Paul. (2006, Nov 14) Practical Electronics

for Inventors. ed. McGraw-Hill.
If the noise source is correlated with the signal, such as
in the case of quantisation error, the intentional introduc-
tion of additional noise, called dither, can reduce overall 7.2.8 Further reading
noise in the bandwidth of interest. This technique allows
retrieval of signals below the nominal detection thresh- • Sh. Kogan (1996). Electronic Noise and Fluctua-
old of an instrument. This is an example of stochastic tions in Solids. Cambridge University Press. ISBN
resonance. 0-521-46034-4.

7.2.5 See also 7.2.9 External links

• Discovery of cosmic microwave background radia-
tion
• Generation–recombination noise
• Matched filter for noise reduction in modems
• Noise reduction and noise cancellation for audio and
images
• Error correction for digital signals subject to noise. opened and closed. Usually, non-overlapping signals are
• Phonon noise
7.2.6 Notes
[1] Motchenbacher, C. D.; Connelly, J. A. (1993). Low-noise
electronic system design. Wiley Interscience.
• Active Filter (Sallen & Key) Noise Study
7.3 Switched capacitor
A switched capacitor is an electronic circuit element
used for discrete time signal processing. It works by mov-
ing charges into and out of capacitors when switches are
used to control the switches, so that not all switches are
closed simultaneously. Filters implemented with these
elements are termed “switched-capacitor filters,” and de-
pend only on the ratios between capacitances. This
makes them much more suitable for use within integrated
circuits, where accurately specified resistors and capaci-
tors are not economical to construct.[1]

[2] Kish, L. B.; Granqvist, C. G. (November 2000). “Noise

in nanotechnology”. Microelectronics Reliability (El- 7.3.1 The switched capacitor resistor
sevier) 40 (11): 1833–1837. doi:10.1016/S0026-
2714(00)00063-9.

[3] Steinbach, Andrew; Martinis, John; Devoret, Michel

(1996-05-13). “Observation of Hot-Electron Shot
Noise in a Metallic Resistor”. Phys. Rev. Lett.
76 (20): 38.6–38.9. Bibcode:1996PhRvL..76...38M.
doi:10.1103/PhysRevLett.76.38.

7.2.7 References
• White noise calculator, thermal noise - Voltage in
microvolts, conversion to noise level in dBu and
dBV and vice versa
• This article incorporates public domain mate- Switched-capacitor resistor
rial from the General Services Administration docu-
ment “Federal Standard 1037C” (in support of MIL- The simplest switched capacitor (SC) circuit is the
STD-188). switched capacitor resistor, made of one capacitor C and
7.3. SWITCHED CAPACITOR 179

two switches S1 and S2 which connect the capacitor with So the equivalent resistance R (i.e., the voltage–current
a given frequency alternately to the input and output of relationship) is:
the SC. Each switching cycle transfers a charge q from
the input to the output at the switching frequency f . Re-
call that the charge q on a capacitor C with a voltage V V 1
R= = .
between the plates is given by: I CS f
Thus, the SC behaves like a resistor whose value depends
on capacitance CS and switching frequency f.
q = CV
The SC resistor is used as a replacement for simple resis-
where V is the voltage across the capacitor. Therefore, tors in integrated circuits because it is easier to fabricate
when S1 is closed while S2 is open, the charge stored in reliably with a wide range of values. It also has the ben-
the capacitor CS is: efit that its value can be adjusted by changing the switch-
ing frequency (i.e., it is a programmable resistance). See
also: operational amplifier applications. q = CV This
same circuit can be used in discrete time systems (such
qIN = CS VIN .
as analog to digital converters) as a track and hold cir-
When S2 is closed, some of that charge is transferred out cuit. During the appropriate clock phase, the capacitor
of the capacitor, after which the charge that remains in samples the analog voltage through switch one and in the
capacitor CS is: second phase presents this held sampled value to an elec-
tronic circuit for processing.

qOUT = CS VOUT . 7.3.2 The Parasitic Sensitive Integrator

Thus, the charge moved out of the capacitor to the output
is:

q = qIN − qOUT = CS (VIN − VOUT )

Because this charge q is transferred at a rate f, the rate of

transfer of charge per unit time is:

A Simple Switched Capacitor Parasitic-Sensitive Integrator

I = qf.
Often switched capacitor circuits are used to provide ac-
Note that we use I, the symbol for electric current, for
curate voltage gain and integration by switching a sam-
this quantity. This is to demonstrate that a continuous
pled capacitor onto an op-amp with a capacitor Cf b in
transfer of charge from one node to another is equivalent
feedback. One of the earliest of these circuits is the
to a current. Substituting for q in the above, we have:
parasitic-Sensitive integrator developed by the Czech en-
gineer Bedrich Hosticka.[2] Let us analyze what happens
in this case. Denote by T = 1/f the switching period.
I = CS (VIN − VOUT )f Recall that in capacitors
Let V be the voltage across the SC from input to output.
So:
charge = capacitance × voltage

Then, at the instant when S1 opens and S2 closes, we have

V = VIN − VOUT . the following:
180 CHAPTER 7. THE DECIBEL

1) Because Cs has just charged: Use In Discrete-Time Systems

The delaying parasitic insensitive integrator has a wide

Qs (t) = Cs · Vs (t) use in discrete time electronic circuits such as biquad fil-
ters, anti-alias structures, and delta sigma data converters.
2) Because the feedback cap, Cf b , is suddenly charged This circuit implements the following z-domain function:
with that much charge (by the opamp, which seeks a vir-
tual short circuit between its inputs):
1
H(z) =
z−1
Qf b (t) = Qs (t − T ) + Qf b (t − T )

Now dividing 2) by Cf b : 7.3.4 The Multiplying Digital to Analog

Converter

Qs (t − T )
Vf b (t) = + Vf b (t − T )
Cf b

And inserting 1):

Cs
Vf b (t) = · Vs (t − T ) + Vf b (t − T )
Cf b

This last equation represents what is going on in Cf b - it

increases (or decreases) its voltage each cycle according
to the charge that is being “pumped” from Cs (due to the A 1.5 bit Multiplying Digital to Analog Converter
op-amp).
However, there is a more elegant way to formulate this One useful characteristic of switched capacitor circuits
fact if T is very short. Let us introduce dt ← T and is that they can be used to perform many circuit tasks at
dVf b ← Vf b (t)−Vf b (t−dt) and rewrite the last equation the same time, which is difficult with non-discrete time
divided by dt: components. The multiplying digital to analog converter
(MDAC) is an example as it can take an analog input,
add a digital value to it, and multiply this by some factor
dVf b (t) Cs based on the capacitor ratios. The output of the MDAC
=f · Vs (t)
dt Cf b is given by the following:
Therefore, the op-amp output voltage takes the form:
Vi · (C1 + C2 ) − (d − 1) · Vr · C2 + Vos · (C1 + C2 + Cp )
∫ VOut = (C1 +C2 +Cp )
1 C1 +
VOU T (t) = −Vf b (t) = − 1 Vs (t)dt A

f Cs Cf b The MDAC is a common component in modern pipeline

Note that this is an integrator with an “equivalent resis- analog to digital converters as well as other precision ana-
tance” Req = f C 1
. This allows its on-line or runtime log electronics and was first created in the form[3]above by
adjustment (if we manage to make the switches oscillate Stephen Lewis and others at Bell Laboratories.
s

according to some signal given by e.g. a microcontroller).

7.3.5 See also
7.3.3 The Parasitic Insensitive Integrator
• Nyquist–Shannon sampling theorem
7.4. H BRIDGE 181

• Aliasing 7.4.1 General

• Switched-mode power supply H bridges are available as integrated circuits, or can be
• Charge pump built from discrete components.[1]
The term H bridge is derived from the typical graphical
representation of such a circuit. An H bridge is built
7.3.6 References with four switches (solid-state or mechanical). When
the switches S1 and S4 (according to the first figure) are
[1] Switched Capacitor Circuits, Swarthmore College course
closed (and S2 and S3 are open) a positive voltage will be
notes, accessed 2009-05-02
applied across the motor. By opening S1 and S4 switches
[2] B. Hosticka, R. Brodersen, P. Gray, “MOS Sampled Data and closing S2 and S3 switches, this voltage is reversed,
Recursive Filters. ng Switched Capacitor Integrators,” allowing reverse operation of the motor.
IEEE Journal of Solid State Circuits, Vol SC-12, No.6,
December 1977. Using the nomenclature above, the switches S1 and S2
should never be closed at the same time, as this would
[3] Stephen H. Lewis et al., “A 10-bit, 20Msample/s Analog cause a short circuit on the input voltage source. The
to Digital Converter, IEEE Journal of Solid State Circuits, same applies to the switches S3 and S4. This condition is
March 1992 known as shoot-through.

• Mingliang Liu, Demystifying Switched-Capacitor

Circuits, ISBN 0-7506-7907-7 7.4.2 Operation

7.4 H bridge

The two basic states of an H bridge

The H-bridge arrangement is generally used to reverse

the polarity of the motor, but can also be used to 'brake'
the motor, where the motor comes to a sudden stop, as the
motor’s terminals are shorted, or to let the motor 'free run'
to a stop, as the motor is effectively disconnected from the
circuit. The following table summarises operation, with
S1-S4 corresponding to the diagram above.
Structure of an H bridge (highlighted in red)

An H bridge is an electronic circuit that enables a volt- 7.4.3 Construction

age to be applied across a load in either direction. These
circuits are often used in robotics and other applications Relays
to allow DC motors to run forwards and backwards.[1]
Most DC-to-AC converters (power inverters), most A way to build[2]an H bridge is use an array of relays from
AC/AC converters, the DC-to-DC push–pull converter, a relay board.
most motor controllers, and many other kinds of power A “double pole double throw” relay can generally achieve
electronics use H bridges. In particular, a bipolar stepper the same electrical functionality as an H bridge (con-
motor is almost invariably driven by a motor controller sidering the usual function of the device). However a
containing two H bridges. semiconductor-based H bridge would be preferable to the
182 CHAPTER 7. THE DECIBEL

channel MOSFETs. This requires a more complex de-

Structure of an H bridge (highlighted in red)
sign since the gates of the high side MOSFETs must be
driven positive with respect to the DC supply rail. Many
integrated circuit MOSFET gate drivers include a charge
pump within the device to achieve this.
Alternatively, a switched-mode DC–DC converter can be
used to provide isolated ('floating') supplies to the gate
drive circuitry. A multiple-output flyback converter is
well-suited to this application.
Another method for driving MOSFET-bridges is the use
of a specialised transformer known as a GDT (Gate Drive
Transformer), which gives the isolated outputs for driv-
ing the upper FETs gates. The transformer core is usually
a ferrite toroid, with 1:1 or 4:9 winding ratio. However,
this method can only be used with high frequency sig-
nals. The design of the transformer is also very impor-
tant, as the leakage inductance should be minimized, or
cross conduction may occur. The outputs of the trans-
former also need to be usually clamped by Zener diodes,
because high voltage spikes could destroy the MOSFET
gates.

Variants

A common variation of this circuit uses just the two tran-

sistors on one side of the load, similar to a class AB am-
L298 dual H bridge motor driver plifier. Such a configuration is called a “half bridge”.[3]
The half bridge is used in some switched-mode power
supplies that use synchronous rectifiers and in switching
relay where a smaller physical size, high speed switching, amplifiers. The half-H bridge type is commonly abbrevi-
or low driving voltage (or low driving power) is needed, ated to “Half-H” to distinguish it from full (“Full-H”) H
or where the wearing out of mechanical parts is undesir- bridges. Another common variation, adding a third 'leg'
able. to the bridge, creates a three-phase inverter. The three-
phase inverter is the core of any AC motor drive.
N and P channel semiconductors A further variation is the half-controlled bridge, where
the low-side switching device on one side of the bridge,
A solid-state H bridge is typically constructed using op- and the high-side switching device on the opposite side of
posite polarity devices, such as PNP BJTs or P-channel the bridge, are each replaced with diodes. This eliminates
MOSFETs connected to the high voltage bus and NPN the shoot-through failure mode, and is commonly used to
BJTs or N-channel MOSFETs connected to the low volt- drive variable or switched reluctance machines and actu-
age bus. ators where bi-directional current flow is not required.

N channel-only semiconductors Commercially available

The most efficient MOSFET designs use N-channel There are many commercially available inexpensive sin-
MOSFETs on both the high side and low side because gle and dual H-bridge packages, and L293x series are the
they typically have a third of the ON resistance of P- most common ones. Few packages, like L9110,[4] have
7.5. HALL EFFECT SENSOR 183

built-in flyback diodes for back EMF protection. Projects

• Tutorial: Build a 5A H-Bridge motor controller

7.4.4 Operation as an inverter
A common use of the H bridge is an inverter. The ar-
rangement is sometimes known as a single-phase bridge
inverter.
The H bridge with a DC supply will generate a square
wave voltage waveform across the load. For a purely in-
ductive load, the current waveform would be a triangle
wave, with its peak depending on the inductance, switch-
ing frequency, and input voltage.
• Building an H-bridge-controlled motor with photo-
cells to track light
• H-bridge motor control with 4017 (in Turkish)
• Using the HIP4081A for H-bridge control
• Using the L293D H bridge for DC motor control
• A simple circuit designed around L293D motor
driver IC

7.4.5 See also 7.5 Hall effect sensor

• Active rectification

• Commutator (electric)

7.4.6 References
[1] Al Williams (2002). Microcontroller projects using the Ba-
sic Stamp (2nd ed.). Focal Press. p. 344. ISBN 978-1-
57820-101-3.

[2] wordpress.com

[3] “The Half-bridge Circuit Revealed” Tom Ribarich (2012)

[4] wordpress.com

A wheel containing two magnets passing by a Hall effect sensor

7.4.7 External links
A Hall effect sensor is a transducer that varies its output
• H-Bridge Theory and Practice voltage in response to a magnetic field. Hall effect sen-
sors are used for proximity switching, positioning, speed
• Brief H-Bridge Theory of Operation detection, and current sensing applications.[1]
• H-bridge tutorial discussing various driving modes In its simplest form, the sensor operates as an analog
and using back-EMF transducer, directly returning a voltage. With a known
magnetic field, its distance from the Hall plate can be de-
• PWM DC Motor Controller Using MOSFETs and termined. Using groups of sensors, the relative position
IR2110 H-Bridge Driver of the magnet can be deduced.

• H-Bridges on the BEAM Robotics Wiki Frequently, a Hall sensor is combined with circuitry that
allows the device to act in a digital (on/off) mode, and
• Derivation of formulas to estimate H-bridge con- may be called a switch in this configuration. Com-
troller current (Vex, JAGUAR,Victor). Discusses monly seen in industrial applications such as the pictured
why some H-bridges used in robotics have non- pneumatic cylinder, they are also used in consumer equip-
linear current and speed responses. ment; for example some computer printers use them to
184 CHAPTER 7. THE DECIBEL

Engine fan with Hall effect sensor

The magnetic piston (1) in this pneumatic cylinder will cause the Commonly used circuit symbol
Hall effect sensors (2 and 3) mounted on its outer wall to activate
when it is fully retracted or extended.
7.5.1 Hall probe

A Hall probe contains an indium compound

semiconductor crystal such as indium antimonide,
mounted on an aluminum backing plate, and encap-
sulated in the probe head. The plane of the crystal is
detect missing paper and open covers. When high relia- perpendicular to the probe handle. Connecting leads
bility is required, they are used in keyboards. from the crystal are brought down through the handle to
Hall sensors are commonly used to time the speed of the circuit box.
wheels and shafts, such as for internal combustion engine When the Hall probe is held so that the magnetic field
ignition timing, tachometers and anti-lock braking sys- lines are passing at right angles through the sensor of
tems. They are used in brushless DC electric motors to the probe, the meter gives a reading of the value of
detect the position of the permanent magnet. In the pic- magnetic flux density (B). A current is passed through
tured wheel with two equally spaced magnets, the voltage the crystal which, when placed in a magnetic field has
from the sensor will peak twice for each revolution. This a extquotedblHall effect extquotedbl voltage developed
arrangement is commonly used to regulate the speed of across it. The Hall effect is seen when a conductor
disk drives. is passed through a uniform magnetic field. The natu-
7.5. HALL EFFECT SENSOR
ral electron drift of the charge carriers causes the mag-
netic field to apply a Lorentz force (the force exerted on
a charged particle in an electromagnetic field) to these
charge carriers. The result is what is seen as a charge
separation, with a buildup of either positive or negative
charges on the bottom or on the top of the plate. The
crystal measures 5 mm square. The probe handle, being
made of a non-ferrous material, has no disturbing effect
on the field.
A Hall probe should be calibrated against a known value
of magnetic field strength. For a solenoid the Hall probe
is placed in the center.
7.5.2 Working principle
When a beam of charged particles passes through a mag-
netic field, forces act on the particles and the beam is de-
flected from its straight line path. The beam of charged
particles refers to the electrons flowing through a con-
ductor. When a current carrying conductor is placed in
a magnetic field perpendicular to the path of the elec-
trons, the electrons are deflected from its straight line
path. Therefore, one side of the conductor becomes neg-
ative portion and the other side becomes positive one.
The transverse voltage is measured and is known as Hall
Voltage.[2]
The charge separation continues until the force on the
charged particles from the electric field balances the force
produced by magnetic field. If the current is constant,
then the Hall voltage is a measure of the magnetic flux
density. There are two forms of Hall Effect Sensors. One
is linear where the output voltage linearly varies with the
magnetic flux density. The other is known as threshold
where there is a sharp drop of output voltage at a partic- 7.5.5 Advantages
ular magnetic flux density.
A Hall effect sensor may operate as an electronic switch.
7.5.3 Materials for Hall effect sensors
The key factor deciding on sensitivity of Hall effect sen-
sors is high electron mobility. As a result following ma-
terials are especially suitable for Hall effect sensors:
• gallium arsenide (GaAs)
• indium arsenide (InAs)
• indium phosphide (InP)
185
• indium antimonide (InSb)
• graphene [3]
7.5.4 Signal processing and interface
Hall effect sensors are linear transducers. As a result such
sensors require linear circuit for processing of the sensor’s
output signal. Such a linear circuit:
• provides constant driving current to the sensors
• amplifies the output signal
In some cases linear circuit may cancel the offset volt-
age of Hall effect sensors. Moreover, AC modulation of
driving current may reduce the influence of offset voltage
on characteristics of Hall effect sensor.
Hall effect sensors with linear transducers are commonly
integrated with digital electronics. This enables advanced
corrections of the sensor’s characteristics (e.g. tempera-
ture coefficient corrections) and digital interfacing to mi-
croprocessor systems. In some solutions of IC Hall ef-
fect sensors DSP is implemented. The simplest digital
processing circuit is an electronic comparator enabling
switch on/off operation, which is useful in industrial au-
tomation applications.
The Hall effect sensors interfaces may include input di-
agnostics, fault protection for transient conditions, and
short/open circuit detection. It may also provide and
monitor the current to the Hall effect sensor itself. There
are precision IC products available to handle these fea-
tures.
• Such a switch costs less than a mechanical switch
and is much more reliable.
• It can be operated up to 100 kHz.
• It does not suffer from contact bounce because a
solid state switch with hysteresis is used rather than
a mechanical contact.
• It will not be affected by environmental contami-
nants since the sensor is in a sealed package. There-
fore it can be used under severe conditions.
186
In the case of linear sensor (for the magnetic field strength
measurements), a Hall effect sensor:
• can measure a wide range of magnetic fields
• is available that can measure either North or South
pole magnetic fields
• can be flat
7.5.6 Disadvantages
Hall effect sensors provide much lower measuring accu-
racy than fluxgate magnetometers or magnetoresistance
based sensors. Moreover, Hall effect sensors drift signif-
icantly, requiring compensation.
7.5.7 Applications
Position sensing
Sensing the presence of magnetic object (connected with
the position sensing) is the most common industrial ap-
plication of Hall effect sensors, especially these operating
in the switch mode (on/off mode). The Hall effect sen-
sors are also used in the brushless DC motor to sense the
position of the rotor and to switch the transistor in the
right sequence.
DC current transformers
Hall effect sensors may be utilized for contactless mea-
surements of DC current in current transformers. In such
a case the Hall effect sensor is mounted in the gap in mag-
netic core around the current conductor.[4] As a result, the
DC magnetic flux can be measured, and the DC current
in the conductor can be calculated.
Automotive fuel level indicator
The Hall sensor is used in some automotive fuel level In-
dicators. The main principle of operation of such indica-
tor is position sensing of floating element.[5]
This can either be done by using a vertical float magnet
or a rotating lever sensor.
• In a vertical float system a permanent magnet is
mounted on the surface of a floating object. The
CHAPTER 7. THE DECIBEL
current carrying conductor is fixed on the top of the
tank lining up with the magnet. When the level of
fuel rises, an increasing magnetic field is applied on
the current resulting in higher Hall voltage. As the
fuel level decreases, the Hall voltage will also de-
crease. The fuel level is indicated and displayed by
proper signal condition of Hall voltage.
• In a rotating lever sensor a diametrically magne-
tized ring magnet rotates about a linear hall sensor.
The sensor only measures the perpendicular (verti-
cal) component of the field. The strength of the field
measured correlates directly to the angle of the lever
and thus the level of the fuel tank.
7.5.8 References
[1] Ed Ramsden (2006). Hall-effect sensors: theory and appli-
cations (2, illustrated ed.). Elsevier. ISBN 0-7506-7934-
4.
[2] R. S. Popović (2004). Hall effect devices (2, illustrated
ed.). CRC Press. ISBN 0-7503-0855-9.
[3] Petruk, O.; Szewczyk, R.; Ciuk, T. et al. (2014).
“Sensitivity and Offset Voltage Testing in the Hall-
Effect Sensors Made of Graphene”. Advances in In-
telligent Systems and Computing (Springer) 267: 631.
doi:10.1007/978-3-319-05353-0_60.
[4] Petruk, O.; Szewczyk, R.; Salach, J.; Nowicki, M. (2014).
“Digitally Controlled Current Transformer with Hall Sen-
sor”. Advances in Intelligent Systems and Computing
(Springer) 267: 641. doi:10.1007/978-3-319-05353-
0_61.
[5] https://www.infineon.com/dgdl/AppNote_
Liquid_Level_Sensing_Rev.1.0.pdf?folderId=
db3a30431ce5fb52011d4cae1f582dad&fileId=
db3a30432313ff5e0123a385f3b2262d
7.5.9 Further reading
• A. Baumgartner et al., “Classical Hall effect in scan-
ning gate experiments”, Phys. Rev. B, 74, 165426
(2006), doi:10.1103/PhysRevB.74.165426
Chapter 8
Filters
8.1 Low-pass filter
A low-pass filter is a filter that passes signals with
a frequency lower than a certain cutoff frequency and
attenuates signals with frequencies higher than the cut-
off frequency. The amount of attenuation for each fre-
quency depends on the filter design. The filter is some-
times called a high-cut filter, or treble cut filter in au-
dio applications. A low-pass filter is the opposite of a
high-pass filter. A band-pass filter is a combination of a
low-pass and a high-pass filter.
Low-pass filters exist in many different forms, includ-
ing electronic circuits (such as a hiss filter used in
audio), anti-aliasing filters for conditioning signals prior
to analog-to-digital conversion, digital filters for smooth-
ing sets of data, acoustic barriers, blurring of images, and
so on. The moving average operation used in fields such
as finance is a particular kind of low-pass filter, and can
be analyzed with the same signal processing techniques as
are used for other low-pass filters. Low-pass filters pro-
vide a smoother form of a signal, removing the short-term
fluctuations, and leaving the longer-term trend.
An optical filter can correctly be called a low-pass fil- synthesisers. See subtractive synthesis.
ter, but conventionally is called a longpass filter (low fre-
quency is long wavelength), to avoid confusion.
8.1.1 Examples
Acoustics
A stiff physical barrier tends to reflect higher sound fre-
quencies, and so acts as a low-pass filter for transmitting
sound. When music is playing in another room, the low
notes are easily heard, while the high notes are attenuated.
187
Electronics
In an electronic low-pass RC filter for voltage signals,
high frequencies in the input signal are attenuated, but
the filter has little attenuation below the cutoff frequency
determined by its RC time constant. For current signals,
a similar circuit, using a resistor and capacitor in parallel,
works in a similar manner. (See current divider discussed
in more detail below.)
Electronic low-pass filters are used on inputs to
subwoofers and other types of loudspeakers, to block high
pitches that they can't efficiently reproduce. Radio trans-
mitters use low-pass filters to block harmonic emissions
that might interfere with other communications. The tone
knob on many electric guitars is a low-pass filter used to
reduce the amount of treble in the sound. An integrator
is another time constant low-pass filter.[1]
Telephone lines fitted with DSL splitters use low-pass and
high-pass filters to separate DSL and POTS signals shar-
ing the same pair of wires.[2][3]
Low-pass filters also play a significant role in the sculpt-
ing of sound created by analogue and virtual analogue
8.1.2 Ideal and real filters
An ideal low-pass filter completely eliminates all frequen-
cies above the cutoff frequency while passing those be-
low unchanged; its frequency response is a rectangular
function and is a brick-wall filter. The transition region
present in practical filters does not exist in an ideal fil-
ter. An ideal low-pass filter can be realized mathemati-
cally (theoretically) by multiplying a signal by the rectan-
gular function in the frequency domain or, equivalently,
convolution with its impulse response, a sinc function, in
188
sin(πx)
πx
1.0
0.8
0.6
0.4
0.2
x
-6 -4 -2 2 4 6 −10
-0.2
The sinc function, the impulse response of an ideal low-pass fil-
ter.
the time domain.
However, the ideal filter is impossible to realize without
also having signals of infinite extent in time, and so gen-
erally needs to be approximated for real ongoing signals,
because the sinc function’s support region extends to all
past and future times. The filter would therefore need to
have infinite delay, or knowledge of the infinite future and
past, in order to perform the convolution. It is effectively
realizable for pre-recorded digital signals by assuming ex-
tensions of zero into the past and future, or more typically
by making the signal repetitive and using Fourier analy-
sis.
Real filters for real-time applications approximate the
ideal filter by truncating and windowing the infinite im-
pulse response to make a finite impulse response; apply-
ing that filter requires delaying the signal for a moderate
period of time, allowing the computation to “see” a little
bit into the future. This delay is manifested as phase shift.
Greater accuracy in approximation requires a longer de-
lay.
An ideal low-pass filter results in ringing artifacts via
the Gibbs phenomenon. These can be reduced or wors-
ened by choice of windowing function, and the design
and choice of real filters involves understanding and min-
imizing these artifacts. For example, “simple truncation
[of sinc] causes severe ringing artifacts,” in signal recon-
struction, and to reduce these artifacts one uses window
functions “which drop off more smoothly at the edges.”[4]
The Whittaker–Shannon interpolation formula describes
how to use a perfect low-pass filter to reconstruct a
continuous signal from a sampled digital signal. Real
CHAPTER 8. FILTERS
digital-to-analog converters use real filter approxima-
tions.
8.1.3 Continuous-time low-pass filters
10
Cutoff frequency
0
−3.01 dB
Slope: −20 dB/decade
−20
Gain (dB)
−30
−40
−50
Passband Stopband
−60
0.001 0.01 0.1 1 10 100 1000
Angular frequency (rad/s)
The gain-magnitude frequency response of a first-order (one-
pole) low-pass filter. Power gain is shown in decibels (i.e., a 3 dB
decline reflects an additional half-power attenuation). Angular
frequency is shown on a logarithmic scale in units of radians per
second.
There are many different types of filter circuits, with dif-
ferent responses to changing frequency. The frequency
response of a filter is generally represented using a Bode
plot, and the filter is characterized by its cutoff frequency
and rate of frequency rolloff. In all cases, at the cutoff
frequency, the filter attenuates the input power by half or
3 dB. So the order of the filter determines the amount
of additional attenuation for frequencies higher than the
cutoff frequency.
• A first-order filter, for example, reduces the signal
amplitude by half (so power reduces by a factor of
4), or 6 dB, every time the frequency doubles (goes
up one octave); more precisely, the power rolloff ap-
proaches 20 dB per decade in the limit of high fre-
quency. The magnitude Bode plot for a first-order
filter looks like a horizontal line below the cutoff
frequency, and a diagonal line above the cutoff fre-
quency. There is also a “knee curve” at the bound-
ary between the two, which smoothly transitions be-
tween the two straight line regions. If the transfer
function of a first-order low-pass filter has a zero
as well as a pole, the Bode plot flattens out again,
at some maximum attenuation of high frequencies;
8.1. LOW-PASS FILTER 189

such an effect is caused for example by a little bit of Laplace notation

the input leaking around the one-pole filter; this one-
pole–one-zero filter is still a first-order low-pass. See Continuous-time filters can also be described in terms
Pole–zero plot and RC circuit. of the Laplace transform of their impulse response, in
a way that lets all characteristics of the filter be easily an-
alyzed by considering the pattern of poles and zeros of
• A second-order filter attenuates higher frequencies
the Laplace transform in the complex plane. (In discrete
more steeply. The Bode plot for this type of filter
time, one can similarly consider the Z-transform of the
resembles that of a first-order filter, except that it
impulse response.)
falls off more quickly. For example, a second-order
Butterworth filter reduces the signal amplitude to For example, a first-order low-pass filter can be described
one fourth its original level every time the frequency in Laplace notation as:
doubles (so power decreases by 12 dB per octave, or
40 dB per decade). Other all-pole second-order fil-
Output 1
ters may roll off at different rates initially depending =K
on their Q factor, but approach the same final rate Input 1 + sτ
of 12 dB per octave; as with the first-order filters, where s is the Laplace transform variable, τ is the filter
zeroes in the transfer function can change the high- time constant, and K is the filter passband gain.
frequency asymptote. See RLC circuit.

8.1.4 Electronic low-pass filters

• Third- and higher-order filters are defined similarly.
In general, the final rate of power rolloff for an Passive electronic realization
order- n all-pole filter is 6n dB per octave (i.e., 20n

Vin
dB per decade).

On any Butterworth filter, if one extends the horizontal

R
line to the right and the diagonal line to the upper-left
(the asymptotes of the function), they intersect at exactly
the cutoff frequency. The frequency response at the cutoff
frequency in a first-order filter is 3 dB below the horizon-

Vout
tal line. The various types of filters (Butterworth filter,
Chebyshev filter, Bessel filter, etc.) all have different-
looking knee curves. Many second-order filters have
“peaking” or resonance that puts their frequency response
at the cutoff frequency above the horizontal line. Further-
more, the actual frequency where this peaking occurs can
be predicted without calculus, as shown by Cartwright[5]
C
et al. For third-order filters, the peaking and its frequency
of occurrence can too be predicted without calculus as
recently shown by Cartwright[6] et al. See electronic filter
for other types. Passive, first order low-pass RC filter
The meanings of 'low' and 'high'—that is, the cutoff fre-
quency—depend on the characteristics of the filter. The One simple low-pass filter circuit consists of a resistor
term “low-pass filter” merely refers to the shape of the fil- in series with a load, and a capacitor in parallel with the
ter’s response; a high-pass filter could be built that cuts off load. The capacitor exhibits reactance, and blocks low-
at a lower frequency than any low-pass filter—it is their frequency signals, forcing them through the load instead.
responses that set them apart. Electronic circuits can be At higher frequencies the reactance drops, and the capac-
devised for any desired frequency range, right up through itor effectively functions as a short circuit. The combina-
microwave frequencies (above 1 GHz) and higher. tion of resistance and capacitance gives the time constant
190 CHAPTER 8. FILTERS

of the filter τ = RC (represented by the Greek letter tau). C

The break frequency, also called the turnover frequency
or cutoff frequency (in hertz), is determined by the time
constant:
R2

fc =
1
=
1 vin R1
2πτ 2πRC vout
or equivalently (in radians per second):

1 1
ωc = =
τ RC An active low-pass filter

This circuit may be understood by considering the time

the capacitor needs to charge or discharge through the Active electronic realization
resistor:
Another type of electrical circuit is an active low-pass fil-
ter.
• At low frequencies, there is plenty of time for the ca-
pacitor to charge up to practically the same voltage In the operational amplifier circuit shown in the figure,
as the input voltage. the cutoff frequency (in hertz) is defined as:

• At high frequencies, the capacitor only has time to

charge up a small amount before the input switches 1
fc =
direction. The output goes up and down only a small 2πR 2C
fraction of the amount the input goes up and down. or equivalently (in radians per second):
At double the frequency, there’s only time for it to
charge up half the amount.
1
ωc =
Another way to understand this circuit is through the con- R 2C
cept of reactance at a particular frequency:
The gain in the passband is −R2 /R1 , and the stopband
drops off at −6 dB per octave (that is −20 dB per decade)
• Since direct current (DC) cannot flow through the as it is a first-order filter.
capacitor, DC input must flow out the path marked
Vout (analogous to removing the capacitor).
Discrete-time realization
• Since alternating current (AC) flows very well
For another method of conversion from continuous- to
through the capacitor, almost as well as it flows
discrete-time, see Bilinear transform.
through solid wire, AC input flows out through
the capacitor, effectively short circuiting to ground
(analogous to replacing the capacitor with just a Many digital filters are designed to give low-pass charac-
wire). teristics. Both infinite impulse response and finite im-
pulse response low pass filters as well as filters using
fourier transforms are widely used.
The capacitor is not an “on/off” object (like the block or
pass fluidic explanation above). The capacitor variably
acts between these two extremes. It is the Bode plot and Simple infinite impulse response filter The effect of
frequency response that show this variability. an infinite impulse response low-pass filter can be simu-
8.1. LOW-PASS FILTER 191

lated on a computer by analyzing an RC filter’s behavior That is, this discrete-time implementation of a simple RC
in the time domain, and then discretizing the model. low-pass filter is the exponentially-weighted moving av-
erage
R
∆T
yi = αxi +(1−α)yi−1 where α≜
RC + ∆T
By definition, the smoothing factor 0 ≤ α ≤ 1 . The ex-
vin C vout pression for α yields the equivalent time constant RC in
terms of the sampling period ∆T and smoothing factor α
:

( )
A simple low-pass RC filter 1−α
RC = ∆T
α
From the circuit diagram to the right, according to
Kirchhoff’s Laws and the definition of capacitance: If α = 0.5 , then the RC time constant is equal to the sam-
pling period. If α ≪ 0.5 , then RC is significantly larger
than the sampling interval, and ∆T ≈ αRC .
The filter recurrence relation provides a way to determine
the output samples in terms of the input samples and the
preceding output. The following pseudocode algorithm
simulates the effect of a low-pass filter on a series of dig-
ital samples:
where Qc (t) is the charge stored in the capacitor at time // Return RC low-pass filter output samples, given input
t . Substituting equation Q into equation I gives i(t) = samples, // time interval dt, and time constant RC func-

d t , which can be substituted into equation V so that:

C dv out tion lowpass(real[0..n] x, real dt, real RC) var real[0..n]
y var real α := dt / (RC + dt) y[0] := x[0] for i from 1 to
n y[i] := α * x[i] + (1-α) * y[i-1] return y
d vout
vin (t) − vout (t) = RC The loop that calculates each of the n outputs can be
dt refactored into the equivalent:
This equation can be discretized. For simplicity, assume
for i from 1 to n y[i] := y[i-1] + α * (x[i] - y[i-1])
that samples of the input and output are taken at evenly-
spaced points in time separated by ∆T time. Let the sam- That is, the change from one filter output to the next
ples of vin be represented by the sequence (x1 , x2 , ..., xn ) , is proportional to the difference between the previous
and let vout be represented by the sequence (y1 , y2 , ..., yn ) output and the next input. This exponential smooth-
, which correspond to the same points in time. Making ing property matches the exponential decay seen in the
these substitutions: continuous-time system. As expected, as the time con-
stant RC increases, the discrete-time smoothing param-
eter α decreases, and the output samples (y1 , y2 , ..., yn )
yi − yi−1 respond more slowly to a change in the input samples
xi − yi = RC
∆T (x1 , x2 , ..., xn ) ; the system has more inertia. This filter
is an infinite-impulse-response (IIR) single-pole low-pass
And rearranging terms gives the recurrence relation
filter.

contribution Input output previous from Inertia

z ( }| Finite impulse response Finite-impulse-response fil-
){ z ( }| ){
∆T RC ters can be built that approximate to the sinc function
yi = xi + yi−1 . time-domain response of an ideal sharp-cutoff low-pass
RC + ∆T RC + ∆T
192 CHAPTER 8. FILTERS

filter. In practice, the time-domain response must be time 8.2 High-pass filter
truncated and is often of a simplified shape; in the sim-
plest case, a running average can be used, giving a square This article is about an electronic component. For the
time response.[7] Australian band, see High Pass Filter (band).

8.1.5 See also A high-pass filter (HPF) is an electronic filter that

passes high-frequency signals but attenuates (reduces the
• Baseband amplitude of) signals with frequencies lower than the
cutoff frequency. The actual amount of attenuation for
• DSL filter each frequency varies from filter to filter. A high-pass
filter is usually modeled as a linear time-invariant system.
It is sometimes called a low-cut filter or bass-cut fil-
8.1.6 References ter.[1] High-pass filters have many uses, such as blocking
DC from circuitry sensitive to non-zero average voltages
[1] Sedra, Adel; Smith, Kenneth C. (1991). Microelectronic
Circuits, 3 ed. Saunders College Publishing. p. 60. ISBN
or RF devices. They can also be used in conjunction with
0-03-051648-X. a low-pass filter to make a bandpass filter.

[2] “ADSL filters explained”. Epanorama.net. Retrieved

2013-09-24. 8.2.1 First-order continuous-time imple-
[3] “Home Networking – Local Area Network”. Pcwee-
mentation
nie.com. 2009-04-12. Retrieved 2013-09-24.

[4] Mastering Windows: Improving Reconstruction C

[5] K. V. Cartwright, P. Russell and E. J. Kaminsky,”Finding Vin Vout
the maximum magnitude response (gain) of second-order
filters without calculus,” Lat. Am. J. Phys. Educ. Vol. 6,
No. 4, pp. 559-565, 2012.
R
[6] Cartwright, K. V.; P. Russell and E. J. Kaminsky (2013).
“Finding the maximum and minimum magnitude re-
sponses (gains) of third-order filters without calculus”.
Lat. Am. J. Phys. Educ. 7 (4): 582–587.

[7] Signal recovery from noise in electronic instrumentation Figure 1: A passive, analog, first-order high-pass filter, realized
– T H Whilmshurst by an RC circuit

The simple first-order electronic high-pass filter shown

8.1.7 External links in Figure 1 is implemented by placing an input voltage
across the series combination of a capacitor and a resistor
• Low-pass filter
and using the voltage across the resistor as an output. The
• Low Pass Filter java simulator product of the resistance and capacitance (R×C) is the
time constant (τ); it is inversely proportional to the cutoff
• ECE 209: Review of Circuits as LTI Systems, a frequency fc, that is,
short primer on the mathematical analysis of (elec-
trical) LTI systems.
1 1
• ECE 209: Sources of Phase Shift, an intuitive expla- fc = = ,
2πτ 2πRC
nation of the source of phase shift in a low-pass fil-
ter. Also verifies simple passive LPF transfer func- where fc is in hertz, τ is in seconds, R is in ohms, and C
tion by means of trigonometric identity. is in farads.
8.2. HIGH-PASS FILTER 193

I(t)
z ( }| ){ ( )
d Vin d Vout d Vin d Vout
Vout (t) = C − R = RC −
dt dt dt dt
This equation can be discretized. For simplicity, as-
sume that samples of the input and output are taken
at evenly-spaced points in time separated by ∆T time.
Let the samples of Vin be represented by the sequence
Figure 2: An active high-pass filter (x1 , x2 , . . . , xn ) , and let Vout be represented by the se-
quence (y1 , y2 , . . . , yn ) which correspond to the same
points in time. Making these substitutions:
Figure 2 shows an active electronic implementation of a
first-order high-pass filter using an operational amplifier.
( )
In this case, the filter has a passband gain of -R2 /R1 and xi − xi−1 yi − yi−1
has a corner frequency of yi = RC −
∆T ∆T
And rearranging terms gives the recurrence relation
1 1
fc = = ,
2πτ 2πR1 C
inputs prior from contribution Decaying input in change from Contribution
z }| { }| z {
Because this filter is active, it may have non-unity pass- RC RC
band gain. That is, high-frequency signals are inverted yi = yi−1 + (xi − xi−1 )
RC + ∆T RC + ∆T
and amplified by R2 /R1 .
That is, this discrete-time implementation of a simple
continuous-time RC high-pass filter is
8.2.2 Discrete-time realization
RC
For another method of conversion from continuous- to yi = αyi−1 +α(xi −xi−1 ) where α≜
discrete-time, see Bilinear transform. RC + ∆T
By definition, 0 ≤ α ≤ 1 . The expression for parameter
Discrete-time high-pass filters can also be designed. α yields the equivalent time constant RC in terms of the
Discrete-time filter design is beyond the scope of this ar- sampling period ∆T and α :
ticle; however, a simple example comes from the con-
version of the continuous-time high-pass filter above to ( )
a discrete-time realization. That is, the continuous-time α
RC = ∆T
behavior can be discretized. 1−α
From the circuit in Figure 1 above, according to If α = 0.5 , then the RC time constant equal to the
Kirchhoff’s Laws and the definition of capacitance: sampling period. If α ≪ 0.5 , then RC is significantly
smaller than the sampling interval, and RC ≈ α∆T .


Vout (t) = I(t) R (V)
Algorithmic implementation
Qc (t) = C (Vin (t) − Vout (t)) (Q)


I(t) = ddQtc (I) The filter recurrence relation provides a way to determine
the output samples in terms of the input samples and the
where Qc (t) is the charge stored in the capacitor at time preceding output. The following pseudocode algorithm
t . Substituting Equation (Q) into Equation (I) and then will simulate the effect of a high-pass filter on a series of
Equation (I) into Equation (V) gives: digital samples:
194 CHAPTER 8. FILTERS

// Return RC high-pass filter output samples, given in- is built into a loudspeaker cabinet it is normally a passive
put samples, // time interval dt, and time constant RC filter that also includes a low-pass filter for the woofer and
function highpass(real[0..n] x, real dt, real RC) var so often employs both a capacitor and inductor (although
real[0..n] y var real α := RC / (RC + dt) y[0] := x[0] very simple high-pass filters for tweeters can consist of
for i from 1 to n y[i] := α * y[i-1] + α * (x[i] - x[i-1]) a series capacitor and nothing else). As an example, the
return y formula above, applied to a tweeter with R=10 Ohm, will
The loop which calculates each of the n outputs can be determine the capacitor value for a cut-off frequency of
refactored into the equivalent: 5 kHz. C = 2πf1 R = 6.28×5000×10
1
= 3.18 × 10−6 , or
approx 3.2 μF.
for i from 1 to n y[i] := α * (y[i-1] + x[i] - x[i-1])
An alternative, which provides good quality sound with-
However, the earlier form shows how the parameter α out inductors (which are prone to parasitic coupling, are
changes the impact of the prior output y[i-1] and current expensive, and may have significant internal resistance)
change in input (x[i] - x[i-1]). In particular, is to employ bi-amplification with active RC filters or ac-
tive digital filters with separate power amplifiers for each
• A large α implies that the output will decay very loudspeaker. Such low-current and low-voltage line level
slowly but will also be strongly influenced by even crossovers are called active crossovers.[1]
small changes in input. By the relationship between Rumble filters are high-pass filters applied to the removal
parameter α and time constant RC above, a large of unwanted sounds near to the lower end of the audible
α corresponds to a large RC and therefore a low range or below. For example, noises (e.g., footsteps, or
corner frequency of the filter. Hence, this case cor- motor noises from record players and tape decks) may be
responds to a high-pass filter with a very narrow stop removed because they are undesired or may overload the
band. Because it is excited by small changes and RIAA equalization circuit of the preamp.[1]
tends to hold its prior output values for a long time,
it can pass relatively low frequencies. However, a High-pass filters are also used for AC coupling at the in-
constant input (i.e., an input with (x[i] - x[i-1])=0) puts of many audio power amplifiers, for preventing the
will always decay to zero, as would be expected with amplification of DC currents which may harm the ampli-
a high-pass filter with a large RC . fier, rob the amplifier of headroom, and generate waste
heat at the loudspeakers voice coil. One amplifier, the
• A small α implies that the output will decay quickly professional audio model DC300 made by Crown Inter-
and will require large changes in the input (i.e., (x[i] national beginning in the 1960s, did not have high-pass
- x[i-1]) is large) to cause the output to change much. filtering at all, and could be used to amplify the DC sig-
By the relationship between parameter α and time nal of a common 9-volt battery at the input to supply 18
constant RC above, a small α corresponds to a small volts DC in an emergency for mixing console power.[2]
RC and therefore a high corner frequency of the fil- However, that model’s basic design has been superseded
ter. Hence, this case corresponds to a high-pass fil- by newer designs such as the Crown Macro-Tech series
ter with a very wide stop band. Because it requires developed in the late 1980s which included 10 Hz high-
large (i.e., fast) changes and tends to quickly forget pass filtering on the inputs and switchable 35 Hz high-
its prior output values, it can only pass relatively high pass filtering on the outputs.[3] Another example is the
frequencies, as would be expected with a high-pass QSC Audio PLX amplifier series which includes an in-
filter with a small RC . ternal 5 Hz high-pass filter which is applied to the inputs
whenever the optional 50 and 30 Hz high-pass filters are
turned off.[4]
8.2.3 Applications
Mixing consoles often include high-pass filtering at each
Audio channel strip. Some models have fixed-slope, fixed-
frequency high-pass filters at 80 or 100 Hz that can be
High-pass filters have many applications. They are used engaged; other models have 'sweepable HPF'—a high-
as part of an audio crossover to direct high frequencies pass filter of fixed slope that can be set within a specified
to a tweeter while attenuating bass signals which could frequency range, such as from 20 to 400 Hz on the Midas
interfere with, or damage, the speaker. When such a filter Heritage 3000, or 20 to 20,000 Hz on the Yamaha M7CL
8.2. HIGH-PASS FILTER 195

A 75 Hz “low cut” filter from an input channel of a Mackie 1402

mixing console as measured by Smaart software. This high-pass
filter has a slope of 18 dB per octave. Example of high-pass filter applied to the right half of a pho-
tograph. Left side is unmodified, Right side is with a high-pass
filter applied (in this case, with a radius of 4.9)

digital mixing console. Veteran systems engineer and live

sound mixer Bruce Main recommends that high-pass fil- 8.2.4 See also
ters be engaged for most mixer input sources, except for
those such as kick drum, bass guitar and piano, sources • DSL filter
which will have useful low frequency sounds. Main writes
that DI unit inputs (as opposed to microphone inputs) • Band-stop filter
do not need high-pass filtering as they are not subject to • Band-pass filter
modulation by low-frequency stage wash—low frequency
sounds coming from the subwoofers or the public ad- • Bias tee
dress system and wrapping around to the stage. Main
indicates that high-pass filters are commonly used for di- • Low-pass filter
rectional microphones which have a proximity effect—a • Differentiator
low-frequency boost for very close sources. This low fre-
quency boost commonly causes problems up to 200 or
300 Hz, but Main notes that he has seen microphones 8.2.5 References
that benefit from a 500 Hz HPF setting on the console.[5]
[1] Watkinson, John (1998). The Art of Sound Reproduction.
Focal Press. pp. 268, 479. ISBN 0-240-51512-9. Re-
trieved March 9, 2010.
Image
[2] Andrews, Keith; posting as ssltech (January 11, 2010).
High-pass and low-pass filters are also used in digital “Re: Running the board for a show this big? extquotedbl.
Recording, Engineering & Production. ProSoundWeb.
image processing to perform image modifications, en-
Retrieved 9 March 2010.
hancements, noise reduction, etc., using designs done in
either the spatial domain or the frequency domain.[6] [3] “Operation Manual: MA-5002VZ”. Macro-Tech Series.
Crown Audio. 2007. Retrieved March 9, 2010.
A high-pass filter, if the imaging software does not have
one, can be done by duplicating the layer, putting a gaus- [4] “User Manual: PLX Series Amplifiers”. QSC Audio.
sian blur, inverting, and then blending with the original 1999. Retrieved March 9, 2010.
layer using an opacity (say 50%) with the original layer.[7]
[5] Main, Bruce (February 16, 2010). “Cut 'Em Off
The unsharp masking, or sharpening, operation used in At The Pass: Effective Uses Of High-Pass Filtering”.
image editing software is a high-boost filter, a general- Live Sound International (Framingham, Massachusetts:
ization of high-pass. ProSoundWeb, EH Publishing).
196 CHAPTER 8. FILTERS

[6] Paul M. Mather (2004). Computer processing of remotely

sensed images: an introduction (3rd ed.). John Wiley and
L L
Sons. p. 181. ISBN 978-0-470-84919-4. C C
[7] “Gimp tutorial with high-pass filter operation”.
Vi Vo
L C
8.2.6 External links
• Common Impulse Responses

• ECE 209: Review of Circuits as LTI Systems, a A medium-complexity example of a band-pass filter.
short primer on the mathematical analysis of (elec-
trical) LTI systems.

• ECE 209: Sources of Phase Shift, an intuitive expla-

nation of the source of phase shift in a high-pass fil- Bandpass is an adjective that describes a type of filter or
ter. Also verifies simple passive LPF transfer func- filtering process; it is to be distinguished from passband,
tion by means of trigonometric identity. which refers to the actual portion of affected spectrum.
Hence, one might say “A dual bandpass filter has two
passbands.” A bandpass signal is a signal containing a
8.3 Band-pass filter band of frequencies not adjacent to zero frequency, such
as a signal that comes out of a bandpass filter.[2]
An ideal bandpass filter would have a completely flat
0 dB passband (e.g. with no gain/attenuation throughout) and
would completely attenuate all frequencies outside the
passband. Additionally, the transition out of the pass-
−3 dB band would be instantaneous in frequency. In practice,
no bandpass filter is ideal. The filter does not attenuate
all frequencies outside the desired frequency range com-
pletely; in particular, there is a region just outside the in-
B tended passband where frequencies are attenuated, but
not rejected. This is known as the filter roll-off, and it
f is usually expressed in dB of attenuation per octave or
fL f0 fH decade of frequency. Generally, the design of a filter
seeks to make the roll-off as narrow as possible, thus al-
Bandwidth measured at half-power points (gain −3 dB, √2/2, or lowing the filter to perform as close as possible to its in-
about 0.707 relative to peak) on a diagram showing magnitude tended design. Often, this is achieved at the expense of
transfer function versus frequency for a band-pass filter. pass-band or stop-band ripple.
The bandwidth of the filter is simply the difference be-
A band-pass filter is a device that passes frequencies tween the upper and lower cutoff frequencies. The shape
within a certain range and rejects (attenuates) frequen- factor is the ratio of bandwidths measured using two
cies outside that range. different attenuation values to determine the cutoff fre-
quency, e.g., a shape factor of 2:1 at 30/3 dB means the
bandwidth measured between frequencies at 30 dB atten-
8.3.1 Description
uation is twice that measured between frequencies at 3 dB
An example of an analogue electronic band-pass filter attenuation.
is an RLC circuit (a resistor–inductor–capacitor circuit). Optical band-pass filters are common in photography and
These filters can also be created by combining a low-pass theatre lighting work. These filters take the form of a
filter with a high-pass filter.[1] transparent coloured film or sheet.
8.3. BAND-PASS FILTER 197

8.3.2 Q-factor 8.3.5 References

A band-pass filter can be characterised by its Q-factor.
The Q-factor is the inverse of the fractional bandwidth.
A high-Q filter will have a narrow passband and a low-Q
filter will have a wide passband. These are respectively
referred to as narrow-band and wide-band filters.
[1] E. R. Kanasewich (1981). Time Sequence Analysis in Geo-
physics. University of Alberta. p. 260. ISBN 0-88864-
074-9.
[2] Belle A. Shenoi (2006). Introduction to digital signal pro-
cessing and filter design. John Wiley and Sons. p. 120.
ISBN 978-0-471-46482-2.

[3] Norman Stuart Sutherland (1979). Tutorial Essays in Psy-

8.3.3 Applications chology. Lawrence Erlbaum Associates. p. 68. ISBN
0-470-26652-X.
Bandpass filters are widely used in wireless transmitters
and receivers. The main function of such a filter in a 8.3.6 External links
transmitter is to limit the bandwidth of the output sig-
nal to the band allocated for the transmission. This pre-
vents the transmitter from interfering with other stations.
In a receiver, a bandpass filter allows signals within a se-
lected range of frequencies to be heard or decoded, while
preventing signals at unwanted frequencies from getting
through. A bandpass filter also optimizes the signal-to-
noise ratio and sensitivity of a receiver.
In both transmitting and receiving applications, well-
designed bandpass filters, having the optimum bandwidth
for the mode and speed of communication being used,
maximize the number of signal transmitters that can exist
in a system, while minimizing the interference or compe-
tition among signals.
Outside of electronics and signal processing, one exam-
ple of the use of band-pass filters is in the atmospheric
sciences. It is common to band-pass filter recent meteo-
rological data with a period range of, for example, 3 to
10 days, so that only cyclones remain as fluctuations in
the data fields.
In neuroscience, visual cortical simple cells were first
shown by David Hubel and Torsten Wiesel to have re-
sponse properties that resemble Gabor filters, which are
band-pass.[3]

8.3.4 See also

• Atomic line filter

• Audio crossover

• Band-stop filter
Chapter 9

Text and image sources, contributors, and

licenses

9.1 Text
• Transistor Source: http://en.wikipedia.org/wiki/Transistor?oldid=625165861 Contributors: Mav, The Anome, Taw, Rjstott, Jkominek,
Sandos, Youssefsan, RAD, Little guru, Mudlock, Ray Van De Walker, SimonP, Maury Markowitz, Ellmist, Gbraad, Heron, RTC,
JohnOwens, Michael Hardy, Tim Starling, Cprompt, Nixdorf, Ixfd64, Ahoerstemeier, Cyp, ZoeB, Stevenj, Suisui, Iammaxus, Александър,
Kaeslin, Julesd, Glenn, Bogdangiusca, Cyan, Nikai, Tristanb, Jiang, Lommer, HolIgor, Wikiborg, Reddi, Stone, Dfeuer, Andrewman327,
Gutza, Zoicon5, PeterGrecian, Timc, Tpbradbury, Marshman, Maximus Rex, Grendelkhan, Omegatron, ReciprocityProject, Thue,
Stormie, Bloodshedder, Raul654, Dpbsmith, Flockmeal, Ldo, Phil Boswell, Maheshkale, Robbot, Pigsonthewing, Jakohn, Owain, Fredrik,
Pjedicke, Babbage, Jondel, Bkell, Hadal, UtherSRG, Galexander, Jleedev, Alan Liefting, David Gerard, Enochlau, Wjbeaty, Ancheta Wis,
Giftlite, DavidCary, Mat-C, Ferkelparade, Brian Kendig, COMPATT, Fleminra, Capitalistroadster, Dratman, Chowbok, Gadfium, Plutor,
Sonjaaa, Antandrus, Mako098765, Jossi, Untifler, Avihu, Dcandeto, Qdr, Jimaginator, Mike Rosoft, Vesta, Mindspillage, Zed, Discospin-
ster, Rich Farmbrough, Rhobite, Rmalloy, Pjacobi, ArnoldReinhold, Xezbeth, Mani1, Dmeranda, Dyl, Kbh3rd, Klenje, Plugwash, Jindrich,
Srivatsaaithal, CanisRufus, Sfahey, El C, Lankiveil, Barfooz, Sietse Snel, Neilrieck, Spoon!, Bobo192, EricBarbour, R. S. Shaw, Elipongo,
Matt Britt, Mikel Ward, Jojit fb, Kjkolb, Wikinaut, DanB, Haham hanuka, Hooperbloob, Nsaa, Nazli, Alansohn, Orimosenzon, Jared81,
Interiot, Eric Kvaalen, Barium, Atlant, WTGDMan1986, Ashley Pomeroy, Mr snarf, Brinkost, Snowolf, Blobglob, Oneliner, Wtshyman-
ski, Knowledge Seeker, Cburnett, Suruena, Cal 1234, TenOfAllTrades, DV8 2XL, Gene Nygaard, MIT Trekkie, Redvers, TheCoffee,
Ahseaton, HenryLi, Flying fish, Begemotv2718, Veemonkamiya, Polyparadigm, Matijap, MONGO, Pyrosim, Cbdorsett, Eyreland, Bar0n,
Zzyzx11, CPES, Palica, Msiddalingaiah, Graham87, Magister Mathematicae, Haikupoet, Snafflekid, Coneslayer, JVz, Mjm1964, Bernard
van der Wees, Tangotango, Colin Hill, Vegaswikian, DonSiano, Ligulem, LjL, Rbeas, Yamamoto Ichiro, FlaBot, Naraht, Arnero, Ysangkok,
Nihiltres, AJR, Gparker, RexNL, Gurch, DavideAndrea, RobyWayne, Alvin-cs, Kri, JonathanFreed, Jidan, Chobot, Krishnavedala, DVdm,
Cornellrockey, Bubbachuck, YurikBot, Wavelength, Marginoferror, Hairy Dude, Jimp, SpuriousQ, Stephenb, Gaius Cornelius, Yyy, Shad-
dack, Brejc8, Pseudomonas, NawlinWiki, Rohitbd, ONEder Boy, RazorICE, Jpbowen, Speedevil, Scs, Misza13, Scottfisher, DeadEyeAr-
row, Bota47, Jeh, Searchme, Light current, 21655, Ninly, Theda, Closedmouth, Arthur Rubin, Vdegroot, Cronostvg, Emc2, Wbrameld,
Katieh5584, Kungfuadam, GrinBot, Zvika, ModernGeek, Elliskev, That Guy, From That Show!, Minnesota1, Attilios, Siker, SmackBot,
YellowMonkey, RockMaestro, Dovo, Reedy, Thorseth, Delldot, StephenJMuir, Unforgettableid, Magwich77, Gilliam, Simoxxx, Andy
M. Wang, Lindosland, QEDquid, Master Jay, Avin, @modi, Thumperward, Oli Filth, EncMstr, Papa November, SEIBasaurus, DHN-
bot, Squibman, Audriusa, WDGraham, Foogod, HeKeRnd, Can't sleep, clown will eat me, Writtenright, Sephiroth BCR, KaiserbBot,
Lantrix, Yidisheryid, Rrburke, VMS Mosaic, Chcknwnm, Nakon, Valenciano, MichaelBillington, BWDuncan, Repairscircuitboards, Jk-
lin, DMacks, Rspanton, Ligulembot, Ohconfucius, The undertow, SashatoBot, Kuru, NeilUK, Danorux, Lazylaces, Evenios, JorisvS, Sce-
toaux, IronGargoyle, CyrilB, Loadmaster, MarkSutton, Slakr, Optimale, George The Dragon, Rogerbrent, Dicklyon, Waggers, Mets501,
EEPROM Eagle, Softice6, Caiaffa, Tsolosmi, Kvng, KJS77, Cmcginnis, Iridescent, Drlegendre, Yves-Laurent, Paul Foxworthy, DarkCell,
Aeons, IanOfNorwich, Tawkerbot2, Daniel5127, G-W, Chetvorno, Elekas, Compy 386, David Carron, ThisIsMyUsername, CmdrObot,
Irwangatot, Chrumps, Ilikefood, JohnCD, Rohan2kool, Zureks, Old Guard, Casper2k3, Cydebot, Verdy p, Tawkerbot4, DumbBOT, Ed-
itor at Large, Splateagle, Charlvn, Malleus Fatuorum, 6pence, Jessemonroy650, Epbr123, Pcu123456789, Headbomb, Electron9, Gerry
Ashton, Nezzadar, Leon7, CboneG5, Natalie Erin, Escarbot, AntiVandalBot, Luna Santin, Firespray, EarthPerson, Scientific Ameri-
can, RapidR, Dvandersluis, Farosdaughter, Rico402, JAnDbot, Xhienne, Dan D. Ric, Em3ryguy, Harryzilber, MER-C, CosineKitty, Eri-
coides, Dagnabit, Britcom, Dricherby, Snowolfd4, PhilKnight, Denimadept, Acroterion, I80and, Bongwarrior, VoABot II, Verkhovensky,
BigChicken, Robcotton, Schily, Sub40Hz, Bleh999, Allstarecho, Canyouhearmenow, Clipjoint, Matt B., Species8471, Cocytus, Gjd001,

198
9.1. TEXT 199

VMMK, MartinBot, Chrismon, Tamer ih, DatasheetArchive, Compnerd09, Kostisl, R'n'B, Galootius, Esolbiz, LedgendGamer, Transisto,
J.delanoy, Hans Dunkelberg, ChrisfromHouston, Uncle Dick, Kevin aylward, Ginsengbomb, Darth Mike, Dfries, Tarotcards, SJP, Big-
dumbdinosaur, Mermadak, Imchandan, Jamesontai, Zuban42, Hmsbeagle, Ale2006, JonS117, Idioma-bot, Reelrt, IFly, Chinneeb, King
Lopez, VolkovBot, TreasuryTag, ABF, HeckXX, Constant314, Ryan032, Philip Trueman, TXiKiBoT, Jomasecu, The Original Wild-
bear, Davehi1, FDominec, Rei-bot, Axonn77, Soldior60, CanOfWorms, Supertask, LeaveSleaves, Wickedclown29, Cameronled, Hell-
cat fighter, Randers1, Enviroboy, RaseaC, Spinningspark, Bucko1992, Northfox, Symane, Jimmi Hugh, Logan, Kbrose, Anirak1337,
Area51david, SieBot, Dwandelt, Tescoid, WereSpielChequers, Jonnic1, Toghome, Vanished User 8a9b4725f8376, Chmyr, Guillermo90r,
Jayzor123, Oda Mari, Davidperson, Lightmouse, Poindexter Propellerhead, Boots232, Apsrobillo, Bludude90210, GAMER 20999, Alf
loves chocolate, Nibol, Dolphin51, Denisarona, CodyARay, C0nanPayne, Asher196, Explicit, Loren.wilton, Martarius, ClueBot, The Thing
That Should Not Be, Rodhullandemu, Pakaraki, Mattgirling, Garyzx, Mild Bill Hiccup, Boing! said Zebedee, Blanchardb, Urb4nn1nj4,
Puchiko, 718 Bot, Masterpiece2000, Masoud691, Habibi 66, Mahya42, Kurdestan, Morristanenbaum, Faranak moradipoor, Rahmaty,
PixelBot, Pmronchi, Conical Johnson, Geniusinfranceman, RedSHIFT, Sun Creator, Brews ohare, Rakins007, Tayyabarif, 07mahmooda,
Dekisugi, The Red, Carriearchdale, ChrisHodgesUK, Chaosdruid, Thingg, Wstorr, Aitias, NorthernNerd, SoxBot III, Rtellason, DumZi-
BoT, Dorit82, Scapler, Delt01, XLinkBot, Spitfire, WikHead, Noctibus, Dmillard10, Kbdankbot, Addbot, Pyfan, DOI bot, Captain-tucker,
Ronhjones, Fieldday-sunday, CanadianLinuxUser, Fluffernutter, SpillingBot, MrOllie, Mentisock, Download, Glane23, Favonian, 84user,
Numbo3-bot, Semiwiki, Lightbot, Hhcox, Zorrobot, Jackelfive, PlankBot, Luckas-bot, OrgasGirl, JSimmonz, Ptbotgourou, Fraggle81,
Cc2po, Crispmuncher, DJ LoPaTa, THEN WHO WAS PHONE?, , Mike1975, TestEditBot, Commissar Mo, OregonD00d,
AnomieBOT, Efa, Captain Quirk, Ulric1313, Flewis, LiuyuanChen, Materialscientist, Kirachinmoku, Citation bot, Carlsotr, Carl086,
Frankenpuppy, Xqbot, Hammack, Capricorn42, Nokkosukko, Magnus0re, GrouchoBot, Nedim Ardoğa, Yoganate79, Der Falke, Maitchy,
Henk.muller, Richard BB, Shadowjams, David Nemati, Mike Dill, A. di M., Some standardized rigour, Forrest575757, Prari, Fres-
coBot, Tobby72, Tiramisoo, Lonaowna, Jc3s5h, Dooley3956, Hébus, Firq, Dman223, Roman12345, E1m1o1, Citation bot 1, Kient123,
AstaBOTh15, Pinethicket, Spidey104, Quantumsilverfish, RedBot, SpaceFlight89, Jamesinderbyshire, Mikespedia, Tcnuk, Tyler-willard,
Bgpaulus, Lissajous, Circuitsmith, Sensitivo, TobeBot, Surendhar Murugan, MarkGT, Privatise, Michael9422, Dinamik-bot, Vrenator,
Thomaskutty, Stephen2zidang, Tbhotch, DARTH SIDIOUS 2, SSgator15, DASHBot, EmausBot, John of Reading, Jovianconflict, Immu-
nize, Beatnik8983, RA0808, Mchaiiann, Olof nord, Smappy, AG SILVER92, Tommy2010, REMspectrum, Freetall, Serketan, Markk01,
Joao.pimentel.ferreira, Mkratz, MigueldelosSantos, Rails, 2n3055, Sbmeirow, L Kensington, Anonimski, ChuispastonBot, RockMagnetist,
LeAwesome0001, TYelliot, 28bot, Profurp, ClueBot NG, Wikivinyl, Dratini0, Wikishotaro, Vergamiester, Akjahid, Kjece, Vishal.vnair,
Helpful Pixie Bot, J caraball, Tejasvi.ts, Wbm1058, Lowercase sigmabot, Zzyxzaa26, Vagobot, ISTB351, Mondeepsaikia, PearlSt82,
Tritomex, Tiscando, Colin5555, Vivek7de, Anamatsu, Sunshine Warrior04, Bluefire272, Klilidiplomus, Simeondahl, Dinesh.lei, Hebert
Peró, Chromastone1998, Miguelmadruga, Khazar2, EuroCarGT, JYBot, Dexbot, Havabighed, Dhanmantee, YaganZ, Vahid alpha, I am
One of Many, Joeymank, Historianbuff, DavidLeighEllis, Wamiq, Ugog Nizdast, Wordpressstar, Deedmonds, Gokul.gk7, Mh akbarpour,
Pri88yank, Monkbot, BatManFascination, JaunJimenez, DaveeBlahBlah and Anonymous: 1113
• Bipolar junction transistor Source: http://en.wikipedia.org/wiki/Bipolar_junction_transistor?oldid=627800109 Contributors: Axel-
Boldt, Sandos, Mudlock, Heron, RTC, Michael Hardy, Tim Starling, Ahoerstemeier, Kaeslin, Glenn, Stefan, Nikai, Dcoetzee, The Anome-
bot, Maximus Rex, Omegatron, Josh Cherry, Jondel, Giftlite, Christopher Parham, DavidCary, Mat-C, MathKnight, Everyking, Leonard
G., Micru, Vadmium, Chowbok, Mako098765, TedPavlic, Pjacobi, Kbh3rd, Plugwash, Sfahey, Bdieseldorff, Rgdboer, Bookofjude, Matt
Britt, Hooperbloob, Alansohn, RobertStar20, Pion, Wtshymanski, Dirac1933, Gene Nygaard, Capecodeph, Dan100, Kenyon, Unixxx,
Mário, Jftsang, Pink-isnt-well, Lincher, Graham87, Yurik, Pharlap, Nanite, Snafflekid, Rjwilmsi, DonSiano, Lor772, FlaBot, Ian Pitch-
ford, Arnero, Alfred Centauri, Nimur, Fresheneesz, Physchim62, Jidan, Chobot, YurikBot, Borgx, Cookie4869, SpuriousQ, John2kx,
Hydrargyrum, Zimbricchio, Shaddack, Rohitbd, Justin Eiler, Steven Hepting, Dhollm, Searchme, Light current, Phil Holmes, Ancla-
tion, Allens, Thorney¿?, Tom Morris, Chowwinston, RTKfan, SmackBot, Pennywisdom2099, Gilliam, Chris the speller, DHN-bot, Darth
Panda, Chendy, Trekphiler, Chlewbot, OrphanBot, Easwarno1, Jon Awbrey, Luís Felipe Braga, Rspanton, Ohconfucius, Pramod.s, Sasha-
toBot, JoshuaZ, CyrilB, Rogerbrent, Dicklyon, Onionmon, Pgadfor, Yves-Laurent, Pelotas, Quodfui, Xcentaur, Mikiemike, CmdrObot,
Editor at Large, Omicronpersei8, Thijs!bot, Epbr123, Enter The Crypt, AuKNiFe, Frozenport, Bmunden, DmitTrix, Electron9, Gerry
Ashton, Widefox, Guy Macon, DarkAudit, Waerloeg, Squantmuts, Em3ryguy, MER-C, CosineKitty, Mccamant, VoABot II, Catslash,
Robcotton, LorenzoB, Matt B., Gwern, Jackson Peebles, Entner, Glrx, R'n'B, DrKiernan, Dmillimono, PhirePhly, Kevin aylward, Lone
Skeptic, Iverson2, Tarotcards, Daniele.tampieri, Zedh, SidShakal, Bonadea, Useight, Funandtrvl, Deor, King Lopez, VolkovBot, ICE77,
Rclocher3, Constant314, A4bot, Hqb, JayC, LeaveSleaves, Inductiveload, TelecomNut, Andy Dingley, Spinningspark, AlleborgoBot,
EmxBot, SieBot, Daphreak, Jonnic1, Cwkmail, Jp314159, Flyer22, Eigenpirate, Steven Zhang, Int21h, StaticGull, Anchor Link Bot,
ImageRemovalBot, Sfan00 IMG, Elassint, ClueBot, H.E. Hall, Sabbah67, MichaelVernonDavis, Mild Bill Hiccup, Somwk, Alkamid,
Brews ohare, Maarschalk, 7, Addbot, Furiousgreencloud, Мыша, Jncraton, MrOllie, Lightbot, Heinzelmann, Biezl, Legobot, Luckas-bot,
Yobot, Andreasmperu, Kamran engineer, AnomieBOT, Ciphers, ^musaz, Kingpin13, Пика Пика, Materialscientist, Citation bot, Franken-
puppy, LilHelpa, Xqbot, Gilo1969, Isheden, Panagea, GrouchoBot, Prunesqualer, Javirosa, Sophus Bie, Dougofborg, FrescoBot, Luke831,
Jc3s5h, Roman12345, Austria156, Adlerbot, SpaceFlight89, Inderpeet singh, Lissajous, Vertpox, Aesthe, Extra999, MrSnoot, Qtipium,
AndyHe829, EmausBot, John of Reading, Beatnik8983, Dltwaddell, Tommy2010, Matthewbeckler, Traxs7, MigueldelosSantos, Dffgd,
Kgsbot, Jbergste, 28bot, Rocketrod1960, Mikhail Ryazanov, ClueBot NG, Zelpld, Satellizer, Dywindel, Cntras, Mehtablocham, Helpful
Pixie Bot, Wbm1058, Trunks ishida, Patwal.manish, CitationCleanerBot, Piet De Pauw, MrBill3, Pratyya Ghosh, Tkaret, Dexbot, Mogism,
Manish cfc, Makecat-bot, Lightfoot54, Ndikumana, Michipedian, Jianhui67, Shipandreceive, Teerthram and Anonymous: 416
• Field-effect transistor Source: http://en.wikipedia.org/wiki/Field-effect_transistor?oldid=626735374 Contributors: Sandos, Mudlock,
Heron, Michael Hardy, Tim Starling, Dgrant, Stw, Stevenj, Kaeslin, Glenn, Nikai, EdH, Rob Hooft, Lommer, Colin Marquardt, Ozuma,
Omegatron, Jerzy, Donarreiskoffer, Robbot, Jakohn, Altenmann, Moink, Wikibot, Wjbeaty, Nunh-huh, Cantus, Jaan513, Glengarry, Bob-
blewik, Mako098765, Supaari, Simson, Sillydragon, Klemen Kocjancic, Clemwang, Qdr, TedPavlic, Rmalloy, Pjacobi, Clawed, Deelkar,
200 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

Duk, Matt Britt, Colin Douglas Howell, Timl, Kjkolb, Alphax, Hooperbloob, Nsaa, Pinar, Atlant, Bantman, Versageek, Kfitzner, Joriki,
Prophile, Daniel Case, Cscott, Tabletop, Isnow, Gimboid13, Graham87, BD2412, Nanite, Snafflekid, Rjwilmsi, Antonen, Amrik, Joeythe-
hobo, Arnero, Ewlyahoocom, Antikon, Chobot, Jaraalbe, YurikBot, Borgx, RobotE, Adam1213, Shell Kinney, Shaddack, Wiki alf,
Zwobot, Bota47, Trygvis, Bakkster Man, Light current, Lt-wiki-bot, SmackBot, KocjoBot, Dexter73, Ohnoitsjamie, Hmains, Cesari-
ous, Lindosland, Chris the speller, Sadads, Audriusa, Ian Burnet, Can't sleep, clown will eat me, Tvaughn05, Er Komandante, Pinktulip,
ArglebargleIV, JorisvS, CyrilB, Cikicdragan, Rogerbrent, Robert Bond, Dicklyon, MTSbot, Hu12, Dead3y3, Yves-Laurent, Menswear,
Johnlogic, Smiling man, Thijs!bot, Epbr123, Martin Hogbin, Coldcuts, Yibo0121, Squantmuts, JAnDbot, Xhienne, Dicentra, Magiola-
ditis, AtticusX, CodeCat, Wal1964, Tboschman, Bissinger, Transisto, J.delanoy, Nigholith, Holyone2, AntiSpamBot, DadaNeem, The
Original Wildbear, Alex rosenberg35, Phirosiberia, Andy Dingley, SieBot, Matthew Yeager, Happysailor, Steven Zhang, Ax716, OKBot,
Dravecky, SallyForth123, Kelvinite, ClueBot, Pierebean, Rememberlands, Mumiemonstret, Hugozair, Shinkolobwe, Brews ohare, Stickee,
SilvonenBot, Iranway, Addbot, MatthewJHerbert, Favonian, Tide rolls, Hhcox, Legobot, Fraggle81, Cepheiden, DJ LoPaTa, Mike1975,
KDS4444, ^musaz, Materialscientist, Thisara.d.m, Citation bot, Xqbot, Nedim Ardoğa, Fixentries, Dave3457, GliderMaven, Adlerbot,
MJ94, Lissajous, Inbamkumar86, Yunshui, Patrickwilkerson, MrSnoot, EmausBot, Tpudlik, PascalD25, Nobelium, GianniG46, Sonygal,
Donner60, Ashu2748, ClamDip, ClueBot NG, Tylermw, Alex Nico, Zak.estrada, Widr, Wbm1058, DBigXray, BG19bot, Niteshns.elec,
PrathameshD.elec, Tropcho, Trevayne08, Jgm47, Ulidtko, Tungstic, Osiris, RichardMills65, StarryGrandma, Compwb, Cerabot, Frosty,
Graphium, Little green rosetta, Telfordbuck, Gokki2012, My name is not dave, Mithoon, Skr15081997, Tinkerer02, Anubhavawasthi05
and Anonymous: 231

• Silicon Source: http://en.wikipedia.org/wiki/Silicon?oldid=627479860 Contributors: AxelBoldt, CYD, Mav, The Anome, Stephen Gilbert,
Jeronimo, Rjstott, LA2, Josh Grosse, XJaM, PierreAbbat, Ortolan88, William Avery, Ktsquare, DrBob, Heron, Fonzy, Stevertigo, Dwmy-
ers, RTC, Kchishol1970, Erik Zachte, Ixfd64, Miciah, Eric119, Shimmin, Dgrant, Card, Looxix, Ahoerstemeier, Cyp, Mac, Suisui, Poor
Yorick, Andres, Mxn, Andfarm, MedievalFreak, Schneelocke, Emperorbma, Tantalate, Stone, David Latapie, DJ Clayworth, Vancouver-
guy, Grendelkhan, Omegatron, Shafei, Denelson83, Donarreiskoffer, Robbot, Hankwang, Moriori, Jotomicron, Romanm, Naddy, Modu-
latum, Smallweed, Bkell, Hadal, Raeky, JerryFriedman, Carnildo, Alan Liefting, Giftlite, Inter, Lupin, Timpo, No Guru, Gracefool, Al-
istairMcMillan, Solipsist, Khalid hassani, Brockert, Darrien, Jackol, Glengarry, Delta G, Golbez, Wmahan, Knutux, Quadell, Antandrus,
Ravikiran r, Lvl, Icairns, Karl-Henner, Nielmo, Wyllium, Jh51681, Deglr6328, Trevor MacInnis, Canterbury Tail, Ultratomio, Discospin-
ster, FT2, Vsmith, Spundun, Paul August, DcoetzeeBot, Bender235, ESkog, Sunborn, Plugwash, Brian0918, RJHall, JustinWick, Canis-
Rufus, MBisanz, El C, Kwamikagami, Hayabusa future, Chairboy, Remember, Femto, Bobo192, Mike Schwartz, Smalljim, Matt Britt,
SpeedyGonsales, Trevj, Rje, Merope, HasharBot, Melah Hashamaim, Jumbuck, Etrigan, Storm Rider, Jérôme, Danski14, Alansohn, Gary,
PaulHanson, Patrick Bernier, Arthena, Atlant, Rd232, Sl, Redfarmer, Bantman, Snowolf, Twisp, RainbowOfLight, Dirac1933, TenOfAll-
Trades, Skatebiker, Gene Nygaard, Iustinus, HenryLi, Adrian.benko, Rossheth, Dismas, Megan1967, Isfisk, Woohookitty, Mindmatrix,
TigerShark, Myleslong, Benbest, Polyparadigm, Fbriere, WadeSimMiser, Miss Madeline, Schzmo, Z303, Terence, Sengkang, GregorB,
Mr. Qwert, Alec Connors, Palica, Dysepsion, Mandarax, DePiep, Jclemens, Saperaud, Rjwilmsi, Coemgenus, MZMcBride, Tawker, Sean-
Mack, Daano15, Dar-Ape, Jobarts, Robert Fraser, FlaBot, Musser, Nihiltres, RexNL, Gurch, TimSE, TheDJ, Chobot, Jaraalbe, Bgwhite,
Gwernol, YurikBot, Wavelength, Sceptre, Huw Powell, Fabartus, Sillybilly, Kirill Lokshin, Stephenb, Yyy, Shaddack, Draeco, NawlinWiki,
Complainer, Buster79, Jaxl, Brandon, Syrthiss, BOT-Superzerocool, DeadEyeArrow, Karl Andrews, Nlu, Robotics1, Ms2ger, Tetracube,
Zargulon, Theodolite, Chaos syndrome, Dspradau, Petri Krohn, Droidus, Murphyjd, CWenger, Phil Holmes, HereToHelp, Anclation,
Kajerm, GrinBot, Elliskev, Bwiki, Vanka5, True Pagan Warrior, SmackBot, Sagie, Heroismic, Olorin28, Bomac, Davewild, Delldot, Kin-
tetsubuffalo, Steffen Kaufmann, Edgar181, Gilliam, Bluebot, Dr bab, Jprg1966, Thumperward, Miquonranger03, Deli nk, SEIBasaurus,
TheLeopard, CSWarren, DHN-bot, Sbharris, Darth Panda, Theilert, Gyrobo, Kotra, Can't sleep, clown will eat me, Shalom Yechiel,
Keith Lehwald, Kr5t, Gurps npc, Reengler, Addshore, SundarBot, PsychoCola, Dali, Cyhatch, Aldaron, Reid A., Cybercobra, Makemi,
Nakon, Savidan, RandomP, PokeTIJeremy, Polonium, Imaginateca, Doodle77, Airwolf, PeterJeremy, Bartholomeyczik, Ck lostsword,
Pilotguy, Ohconfucius, Chymicus, SashatoBot, Bcasterline, DA3N, Archimerged, Kuru, Khazar, John, Heimstern, Linnell, AB, JorisvS,
KristianMolhave, Ckatz, Jamix, Slakr, Beetstra, Haasl, Condem, Jose77, Ndittert, Caiaffa, Pejman47, Quaeler, Sfgagnon, Iridescent, Paul
Koning, JayZ, Tmangray, Shoeofdeath, Igoldste, Thricecube, Tawkerbot2, Daniel5127, Poolkris, JForget, Stifynsemons, Thermochap,
The Librarian, Dycedarg, JohnCD, R9tgokunks, Logical2u, Reywas92, A876, Mike Christie, Astrochemist, Hyperdeath, Rifleman 82,
Gogo Dodo, Farzaneh, DumbBOT, Jay32183, FastLizard4, Omicronpersei8, EvocativeIntrigue, Thijs!bot, Epbr123, Barticus88, NJPhar-
ris, Norman Yarvin, TheFearow, Communisthamster, Headbomb, Marek69, DmitTrix, West Brom 4ever, John254, E. Ripley, EdJohn-
ston, CharlotteWebb, Dawnseeker2000, Natalie Erin, SparhawkWiki, Escarbot, Porqin, AntiVandalBot, Opelio, AlexOvShaolin, Danger,
Michael.j.sykora, Gökhan, Res2216firestar, JAnDbot, Deflective, Leuko, Plantsurfer, Planetary, Mcorazao, Instinct, Jabam, 03272, An-
donic, Hut 8.5, Time3000, .anacondabot, Karlhahn, Bongwarrior, VoABot II, AuburnPilot, Rhadamante, Dichrra, CTF83!, SwiftBot, Ani-
mum, Adrian J. Hunter, Ciaccona, Allstarecho, P.B. Pilhet, User A1, The Real Marauder, Vssun, DerHexer, Kraxler, Cocytus, Ratherhavea-
heart, Hdt83, MartinBot, ChemNerd, ItaniuMatrix, R'n'B, Dgwohu, AlexiusHoratius, Leyo, Siliconov, Peterheis, Watch37264, J.delanoy,
Captain panda, Abby, NightFalcon90909, Pursey, 12dstring, Thaurisil, Eskimospy, Acalamari, Shawn in Montreal, Mahewa, Gman124,
Oldboltonian, Coppertwig, Hessammehr, AntiSpamBot, Warut, NewEnglandYankee, Touch Of Light, Juliancolton, Cometstyles, STBotD,
Vanished user 39948282, Sarregouset, Squids and Chips, CardinalDan, Idioma-bot, Wikieditor06, VolkovBot, CWii, ABF, Science4sail,
Luzheng, Jeff G., Ryan032, Philip Trueman, TXiKiBoT, Zidonuke, The Original Wildbear, Malinaccier, Miranda, Park70, Ann Stouter,
Kumorifox, Seraphim, Axiosaurus, Kirsten07734, Jackfork, LeaveSleaves, ^demonBot2, Psyche825, Cremepuff222, Cbaker9552, Hast-
ings007, Smartdude122, Vaubin, Lamro, Redyoshi49q, Go-in, Falcon8765, Enviroboy, Sevela.p, AlleborgoBot, TheBendster, Lando5,
LuigiManiac, Petergans, EmxBot, CMBJ, SieBot, PlanetStar, WereSpielChequers, Ktulu6, ToePeu.bot, Cwkmail, Yintan, Goron1130,
Oda Mari, It.franciscus, Oxymoron83, Nuttycoconut, Pooeater69, RooZ, Steven Zhang, Helikophis, Colboi, JackSchmidt, Hobartimus,
Hak-kâ-ngìn, Robertan, Dzukman2000, StaticGull, Double Vigie, Squirmymcphee, Nergaal, Denisarona, Martarius, ClueBot, Andrew
Nutter, WilliamRoper, Naaa127, The Thing That Should Not Be, Zach4636, Arakunem, Mild Bill Hiccup, Doseiai2, CounterVandal-
9.1. TEXT 201

ismBot, Blanchardb, Neverquick, DragonBot, Souseiseki42, Excirial, Alexbot, CrazyChemGuy, LeoCRogers, Aubriexoxo, Bloodofjing,
Aristotle28, Hunt ice, Abrech, Dspark76, Yim00, Arjayay, The Red, Kakofonous, Enricoros, Kariboo, Plasmic Physics, Versus22, SoxBot
III, Glacier Wolf, Crazy Boris with a red beard, Usaf2222, XLinkBot, Rror, Little Mountain 5, Skarebo, Noctibus, MystBot, Hexa-
Chord, Willking1979, Some jerk on the Internet, Twaz, Fieldday-sunday, Biskuvi, SpillingBot, Download, CarsracBot, Zomgadonggs,
Numbo3-bot, VASANTH S.N., Tide rolls, David0811, HerculeBot, LuK3, FSIM, Alfie66, Luckas-bot, Yobot, 2D, Lethalgeek, Frag-
gle81, Cflm001, Davduff, KamikazeBot, Elizgoiri, Eudialytos, Daniel 1992, Synchronism, AnomieBOT, Pkenriquez, Marauder40, Kristen
Eriksen, Master of Pies, Rubinbot, Galoubet, AdjustShift, Richnotts, Crystal whacker, Flewis, Materialscientist, Citation bot, Djtrimz,
45Factoid44, Rockoprem, Frankenpuppy, LilHelpa, Andrewmc123, Xqbot, Zad68, Sketchmoose, Kojiki1976, Sionus, Capricorn42,
Br77rino, Delphonic, Supermansbutt, Srich32977, Ubcule, GrouchoBot, Call me Bubba, MC Steel, Doulos Christos, Shadowjams, Erik9,
Nixón, R8R Gtrs, FrescoBot, Thatsjustnotcricket, Eldin raigmore, Citation bot 1, DrilBot, Pinethicket, A8UDI, Fentlehan, Xeworlebi,
Jauhienij, FoxBot, Double sharp, TobeBot, Ticklewickleukulele, Intrr, Hbec, Mattiee'x, Doggy69, Ritter17055, Alteo255, Askud99,
RjwilmsiBot, EmausBot, Cricobr, ITshnik, Dewritech, GoingBatty, RA0808, Solarra, P. S. F. Freitas, Chuck Baggett, Liquidmetalrob,
Shuipzv3, StringTheory11, Chemicalinterest, 1 of 6818119630, Wikfr, GianniG46, Wayne Slam, ChemMater, Bulkhardon, L1A1 FAL,
Donner60, Mentibot, ChuispastonBot, Peter Karlsen, DASHBotAV, Xanchester, ClueBot NG, Michael2707, Matthiaspaul, This lousy
T-shirt, Andrewsailer, Parcly Taxel, Kasirbot, TheFame123, Pilularis, Diyar se, Helpful Pixie Bot, Geo7777, JohnSRoberts99, HMSSo-
lent, Wbm1058, Bibcode Bot, DBigXray, Lowercase sigmabot, M0rphzone, PTJoshua, Rijinatwiki, MusikAnimal, Alanmcruickshank,
AdventurousSquirrel, Doggydodo, Too0003, Sam881030, Azurelcorrupted, Bplppl, Glacialfox, Brendonwalsh27, TheFinniest, Trollin-
boss, BattyBot, DarafshBot, Th4n3r, Ellebellyjelly, NitRav, BrightStarSky, Dexbot, JZNIOSH, Lugia2453, Gwitz13, Graphium, Sowlos,
Jnargus, The Anonymouse, Roman Champ, Reatlas, Hermespeed, Mr funky clouds, Percyjvalgal, Epicgenius, Mcatee.jesse, Gruvinnz,
Wedgeline, Ankhsoprah, Ramendoctor, Rahuljohn77, Ebag7125, Lolydolly, Username469, Cinaeth1, Rotaryphone111, Keelyellenmarie,
BethNaught, Lalalalalalalala1231231231234, Dorkakiin and Anonymous: 1025
• Germanium Source: http://en.wikipedia.org/wiki/Germanium?oldid=623792123 Contributors: AxelBoldt, CYD, Mav, Bryan Derksen,
LA2, Josh Grosse, PierreAbbat, William Avery, DrBob, David spector, Bdesham, RTC, Tim Starling, Erik Zachte, Gdarin, Shellreef,
Graue, Minesweeper, Ahoerstemeier, Mac, Jimfbleak, Mgimpel, Suisui, Poor Yorick, Schneelocke, Emperorbma, Tantalate, Stone, David
Latapie, Greenrd, Tpbradbury, Grendelkhan, SEWilco, Thue, Jerzy, Donarreiskoffer, Robbot, Chris 73, Romanm, Naddy, Hadal, Carnildo,
Giftlite, Graeme Bartlett, AJim, Guanaco, Yekrats, Darrien, Jaan513, Jackol, Delta G, R. fiend, OldZeb, Antandrus, Qleem, Thincat,
Icairns, Gscshoyru, B.d.mills, Joyous!, Trevor MacInnis, Plexust, Discospinster, Guanabot, Vsmith, Nvj, Paul August, DcoetzeeBot, Ben-
der235, Sunborn, RJHall, CanisRufus, El C, Joanjoc, Kwamikagami, Remember, Art LaPella, Triona, Femto, Krellis, Jumbuck, Alansohn,
Abe Lincoln, Sl, PhazZ, Staeiou, Bsadowski1, Woohookitty, Linas, StradivariusTV, Benbest, Sengkang, Graham87, BD2412, DePiep,
Grammarbot, Sjö, Saperaud, Rjwilmsi, WCFrancis, XLerate, NeonMerlin, FlaBot, Ground Zero, Nihiltres, Srleffler, King of Hearts,
Chobot, Jaraalbe, YurikBot, Wavelength, Mukkakukaku, Petiatil, Hellbus, Hydrargyrum, Yyy, Shaddack, Alex Bakharev, Bovineone,
NawlinWiki, Janke, The Ogre, Ospalh, Bota47, Ms2ger, Tetracube, Lt-wiki-bot, E Wing, GraemeL, Geoffrey.landis, Anclation, Curpsbot-
unicodify, GrinBot, Luk, ChemGardener, Itub, Amalthea, SmackBot, Alexdeh, KocjoBot, Delldot, Edgar181, Eloil, Kdliss, Kurykh, Per-
sian Poet Gal, Thumperward, MalafayaBot, Deli nk, Dlohcierekim’s sock, Sbharris, Gyrobo, Tsca.bot, Can't sleep, clown will eat me, Ser-
gio.ballestrero, Kaimiddleton, Rrburke, SundarBot, PsychoCola, Cybercobra, Nibuod, Smokefoot, Doodle77, DMacks, PeterJeremy, Cap-
tainbeefart, SashatoBot, Archimerged, John, Buchanan-Hermit, Edwy, Beetstra, Avs5221, Jopusbob, Novangelis, Jose77, Sifaka, KJS77,
Iridescent, LeyteWolfer, Igoldste, FelisSchrödingeris, Thricecube, Cryptic C62, Poolkris, Zubaexy, JForget, Stifynsemons, CmdrObot,
R9tgokunks, FlyingToaster, MarsRover, Intelligentguest, Christian75, Roberta F., Thijs!bot, Epbr123, Barticus88, Irishleprechaun, Runch,
Kablammo, Headbomb, WillMak050389, Phopon, D.H, BlytheG, Escarbot, Eleuther, Mentifisto, Ju66l3r, AntiVandalBot, WinBot, Ope-
lio, JAnDbot, Deflective, Plantsurfer, East718, Acroterion, Magioladitis, WolfmanSF, Karlhahn, Bongwarrior, VoABot II, Keithpoole,
Verkhovensky, TARBOT, Ling.Nut, Aka042, Giggy, LorenzoB, Somearemoreequal, STBot, Rricci, ChemNerd, Trevor f, Watch37264,
J.delanoy, Dshenai, Gman124, Warut, NewEnglandYankee, Ljgua124, Potatoswatter, STBotD, Ja 62, Squids and Chips, GrahamHardy,
Vranak, VolkovBot, ABF, Butwhatdoiknow, Mightyhansa, Philip Trueman, TXiKiBoT, Oshwah, GimmeBot, Oudegeest, Qxz, Retiono
Virginian, Axiosaurus, Vladsinger, Falcon8765, AlleborgoBot, LuigiManiac, Legoktm, EmxBot, Biscuittin, SaltyBoatr, SieBot, Planet-
Star, WereSpielChequers, Keilana, Scorpion451, Avnjay, Hak-kâ-ngìn, The Hemp Necktie, StaticGull, Sean.hoyland, Mygerardromance,
Nergaal, Laburke, MarsmanRom, Church, ClueBot, Knepflerle, DanielDeibler, CounterVandalismBot, Blanchardb, Piledhigheranddeeper,
Ottava Rima, Acolorpink1, DragonBot, AssegaiAli, Lugh23, Juanathan, Ottre, Tylerdmace, NuclearWarfare, Jotterbot, New4325, Larry-
MorseDCOhio, Ember of Light, The Red, Thingg, Aitias, Plasmic Physics, Versus22, Dana boomer, RexxS, Badinfinity, TravisAF, Sky-
Lined, Mr0t1633, Willking1979, Some jerk on the Internet, DOI bot, NjardarBot, CarsracBot, PranksterTurtle, Loveandprotect, Novel
tubes, Alchemist-hp, Numbo3-bot, Tide rolls, Lightbot, Trevas, Margin1522, Luckas-bot, Yobot, EchetusXe, Lethalgeek, Berkay0652,
II MusLiM HyBRiD II, Obscuranym, Cjp24, Tempodivalse, Magog the Ogre, AnomieBOT, Jim1138, Galoubet, Commander Shepard,
Kingpin13, Hcaz11, Materialscientist, Digitalfear, Citation bot, Frankenpuppy, ArthurBot, LovesMacs, Obersachsebot, Andrewmc123,
Xqbot, TheAMmollusc, Ywaz, Jslefo, JimVC3, Capricorn42, Srich32977, Abce2, RibotBOT, Nedim Ardoğa, Fromjarod, FrescoBot,
Scott940603, Hgjghjh, Hedgerhedger, Wik1ped1a is meant 2 be vanda1ised, Barathrumm, Spindocter123, Finalius, Citation bot 1,
Pinethicket, A8UDI, RedBot, Phearson, Wikitanvir, Jhbuk, Jauhienij, ActivExpression, FoxBot, Double sharp, TobeBot, Vrenator, Dracae-
naFragrans, Tbhotch, Slipknot1018, AXRL, RjwilmsiBot, Regancy42, DASHBot, EmausBot, Rbaselt, Richard.danylyuk, Racerx11, K6ka,
Kaimakides, Peterindelft, HiW-Bot, Daonguyen95, StringTheory11, Alpha Quadrant (alt), Sthubertus, Aschwole, Donner60, Chuispaston-
Bot, Zenhomeenergy, Mikhail Ryazanov, ClueBot NG, Feedintm, Leeroyjenkins1996, Mouse20080706, Helpful Pixie Bot, Bibcode Bot,
Charouili, ElphiBot, Supernerd11, Zombieslayer523, Shisha-Tom, Murcielago7, ChrisGualtieri, CarrieVS, Dexbot, TrollTrollTrollTroll,
Frosty, Jamesx12345, T42N24T, Jnargus, Eyesnore, Handoman123210, Kogge, Ramendoctor, Partymarty21, Prof.Haddock, Scooby1961,
Monkbot, Hardkhora, Mtbrandon and Anonymous: 393
• Gallium arsenide Source: http://en.wikipedia.org/wiki/Gallium_arsenide?oldid=617335450 Contributors: Maury Markowitz, Hep-
202 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

haestos, Edward, Tim Starling, Collabi, Dgrant, Mac, Strebe, Julesd, Samw, HolIgor, Tantalate, Stone, Andrewman327, Denelson83,
Donarreiskoffer, Gentgeen, Hankwang, Psychonaut, Giftlite, Eequor, Jabowery, Gadfium, Karl-Henner, N328KF, Cacycle, Femto, Matt
Britt, Keenan Pepper, Benjah-bmm27, Walkerma, Ynhockey, Suruena, Gene Nygaard, Japanese Searobin, Firsfron, DuncanWidman,
Benbest, Polyparadigm, Alan Canon, CronoDAS, Nanite, Rjwilmsi, TheGWO, FlaBot, SchuminWeb, Jeepo, Fresheneesz, Physchim62,
Jaraalbe, YurikBot, Conscious, Alexmorgan, Rada, Shaddack, DragonHawk, ArséniureDeGallium, BOT-Superzerocool, RyanJones,
Calaschysm, Sbyrnes321, Attilios, SmackBot, Ququ, Thorseth, The Photon, Jrockley, Lainagier, Commander Keane bot, Hugo-cs, Blue-
bot, Thumperward, Oli Filth, Plotnick, AndrewBuck, Miquonranger03, Papa November, Gruzd, Rrburke, Phudga, Smokefoot, Doodle77,
Mion, Chymicus, Beetstra, Norm mit, Thricecube, Eastlaw, DangerousPanda, CmdrObot, Corp1117, Thijs!bot, Barticus88, Peter gk, Bri-
anthegiant, Rehnn83, JAnDbot, MER-C, Plantsurfer, Flippin42, VoABot II, Keithpoole, R'n'B, T.vanschaik, R!SC, LordAnubisBOT, Billr
wiki, Atropos235, Jim Swenson, VolkovBot, Bry9000, Nicholasnice, Axiosaurus, Razvan NEAGOE, Why Not A Duck, Kbrose, AIMet-
alsResearcher, Da Joe, Debussy Agutter, Chem-awb, Swisskitt, JSVickers, CohesionBot, ChardonnayNimeque, Wilso418, Wikimedes,
Plasmic Physics, Addbot, Mortense, Frankhindle, Cantaloupe2, Some jerk on the Internet, Laser Razor, Hudavendigar, Lightbot, Legobot,
Luckas-bot, Bunnyhop11, CheMoBot, KamikazeBot, Materialscientist, Peterdx, GrouchoBot, Pinethicket, Tiananmen 8888, RedBot,
MastiBot, Mikespedia, RjwilmsiBot, EmausBot, Dcirovic, Chemicalinterest, Sthubertus, Tolly4bolly, Joeblanda, Mikhail Ryazanov, Clue-
Bot NG, Rycecube57, Snotbot, Dsperlich, Ntansu, Helpful Pixie Bot, Wbm1058, BG19bot, Moonx, CeraBot, Senor Spandex, Monkbot,
La2O3 and Anonymous: 91
• Voltage-controlled oscillator Source: http://en.wikipedia.org/wiki/Voltage-controlled_oscillator?oldid=623210713 Contributors: Glenn,
Radiojon, Omegatron, Phil Boswell, DavidCary, Ds13, MFNickster, Ary29, Mike Rosoft, Hooperbloob, Atlant, Wtshymanski, Miq, Ve-
gaswikian, FlaBot, Chobot, YurikBot, PinothyJ, Gaius Cornelius, Alynna Kasmira, Brandon, Light current, Donald Albury, JLaTondre,
SmackBot, Telestylo, Maliaydin, Chris the speller, Mailmerge, Vina-iwbot, Wmattis, Kvng, Chetvorno, Chrumps, Boardhead, Electron9,
AkosSzoboszlay, EP111, JAnDbot, Pi.1415926535, Nikevich, HL-SDK, Japo, Bissinger, Glrx, Cuddlyable3, Spinningspark, Allebor-
goBot, SieBot, BotMultichill, A. Carty, Zdilli, Vahid avr, IR-TCI, Firefly322, PlantTrees, Denisarona, Dp67, BOTarate, Dim4ik, Thingg,
XLinkBot, AgnosticPreachersKid, WikHead, Addbot, Ravindra 20, Redheylin, Zorrobot, Ronaldo.nunez, Luckas-bot, Yobot, Ptbotgourou,
Fraggle81, AnomieBOT, ^musaz, Materialscientist, Comt Till, Febert, Berrinkursun, Gbalasandeep, Anitauky, J. in Jerusalem, EmausBot,
ZéroBot, Aportnoy, Akerans, 28bot, Vishalshindeelect, Rushikeshbansodeelec, Z-communications and Anonymous: 63
• Frequency-shift keying Source: http://en.wikipedia.org/wiki/Frequency-shift_keying?oldid=599753694 Contributors: The Anome,
Michael Hardy, Ellywa, CatherineMunro, Glenn, Robbot, Ktims, Karn, Ssd, Ary29, Vishahu, Sonett72, ChrisJ, Mcpusc, Smalljim, Bsad-
owski1, Dan100, Camw, Jonnabuz, Murat40, BD2412, HappyCamper, FlaBot, Fresheneesz, Roboto de Ajvol, Willpo, Deville, David Bid-
dulph, SDS, SmackBot, Moeron, KelleyCook, Oli Filth, Harumphy, Dougmc, Dicklyon, Tawkerbot2, Xcentaur, Wafulz, Chrumps, Tawker-
bot4, Kozuch, Thijs!bot, Epbr123, Electron9, Tarnjp, Kauczuk, Alphachimpbot, Jim.henderson, Microsloth, Glrx, Haffner, J.delanoy,
Dhaluza, Brianonn, CanOfWorms, LeaveSleaves, Stoneygirl45, Sv1xv, RFdave007, Djkryptyk, Addbot, Cuaxdon, Ramkumarvecsrv,
Тиверополник, Legobot, Yobot, AnomieBOT, Living001, Jeffrey Mall, Omnipaedista, FrescoBot, 2A4Fh56OSA, Yahia.barie, DARTH
SIDIOUS 2, RjwilmsiBot, RenamedUser01302013, Fæ, ‫بدر الإسلام‬, ClueBot NG, MerlIwBot, Wbm1058, BG19bot, Mbpaz, Srinathkr3,
ChrisGualtieri, Ginsuloft, MitchRandall and Anonymous: 82
• Amplifier Source: http://en.wikipedia.org/wiki/Amplifier?oldid=627607080 Contributors: AxelBoldt, Eloquence, Mav, Ray Van De
Walker, SimonP, Waveguy, Heron, Kku, Ixfd64, Delirium, Docu, Kingturtle, Glenn, Nikai, GRAHAMUK, Jengod, Ww, Wik, Jessel, Max-
imus Rex, Omegatron, Bevo, Raul654, Lumos3, Friedo, RedWolf, Donreed, Dave Bass, Hcheney, David Gerard, Centrx, Giftlite, DocWat-
son42, Lunkwill, DavidCary, Laudaka, Lupin, Vk2tds, Markus Kuhn, Jcobb, AJim, Maroux, Jason Quinn, Nayuki, Wmahan, Chowbok,
Sam Hocevar, Jcorgan, Abdull, Rich Farmbrough, TedPavlic, Guanabot, Pmsyyz, Pt, Meggar, Timl, Hooperbloob, Watsonladd, Malo, Os-
modiar, Wtshymanski, Twisp, Crosbiesmith, Woohookitty, Uncle G, Pol098, CaptainTickles, BD2412, FreplySpang, Snafflekid, Koavf,
Quiddity, Oblivious, Brighterorange, RobertG, Arnero, Margosbot, Alfred Centauri, Kolbasz, 121a0012, Bgwhite, Ahpook, The Rambling
Man, Nol Aders, Matt512, Epolk, Bergsten, Chaser, Rohitbd, Bjf, Bou, Welsh, Howcheng, Thiseye, Dhollm, Speedevil, DeadEyeArrow,
Searchme, Light current, Mattg2k4, Deville, Kungfuadam, Mebden, Jer ome, Kf4bdy, SmackBot, Reedy, Unyoyega, Freestyle, David-
david, Lindosland, Amatulic, Chris the speller, Bluebot, TimBentley, Cadmium, Thumperward, Papa November, Szidomingo, Sajendra,
OrphanBot, Seduisant, Evilspoons, SnappingTurtle, DMacks, Pilotguy, Bn, Shields020, Breno, Minna Sora no Shita, CyrilB, Rogerbrent,
Dicklyon, 2006mba, Kvng, Politepunk, OnBeyondZebrax, Iridescent, Walton One, Mihitha, Yves-Laurent, Chetvorno, JohnTechnologist,
Xcentaur, CmdrObot, Chrumps, Nczempin, Lenilucho, Anoneditor, Doctormatt, Tubenutdave, Red Director, HermanFinster, Australian
audio guy, FredYork, Gionnico, Editor at Large, Enter The Crypt, Pjvpjv, Saimhe, Guy Macon, Mccartyp, CPMartin, CosineKitty, TAn-
thony, MegX, Jahoe, Magioladitis, VoABot II, Askari Mark, JNW, JamesBWatson, Faizhaider, MichaelSHoffman, Black Stripe, Ngwill,
MartinBot, Sigmundg, Jim.henderson, Anaxial, Nono64, Masisnr1, M samadi, DrKiernan, AntiSpamBot, SophieCat, Vspengen, Colorbow,
Ale2006, Mlewis000, Funandtrvl, Joeinwap, Meiskam, ICE77, Philip Trueman, The Original Wildbear, Zuperman, Smcreator, Henrydask,
Anonymous Dissident, Afluent Rider, Someguy1221, Monkey Bounce, Don4of4, Jackfork, Billinghurst, Kilmer-san, Dragonkillernz, Spin-
ningspark, Internetexploder, Biscuittin, Audioamp, Krawi, Hiddenfromview, Henry Delforn (old), Lightmouse, Nitram cero, StaticGull,
Denisarona, Asher196, Thinkingatoms, ClueBot, Binksternet, The Thing That Should Not Be, GeoffreyHale, Jan1nad, GreenSpigot, An-
nArborRick, Blanchardb, Linan0827, Gtstricky, Brews ohare, Arjayay, Versus22, Johnuniq, XLinkBot, Alexius08, Revancher, Srcloutier,
Pedro magalhaes86, Addbot, Mortense, Olli Niemitalo, Avobert, Yobot, Jordsan, Bestiasonica, Dleger, P1ayer, Sarukum, AnomieBOT,
Piano non troppo, B137, Materialscientist, Citation bot, LilHelpa, Justanothervisitor, Ubcule, Maitchy, Uusijani, GliderMaven, FrescoBot,
Gog182, Jc3s5h, Nickw2066, Gdje je nestala duša svijeta, Icontech, I dream of horses, TechnoDanny, Anooshg, Jujutacular, Hessamnia,
Orenburg1, Theo10011, Belledonne, Qianchq, John of Reading, Kodabmx, Cmavr8, TuomTuo, GoingBatty, Solarra, AnonymousNarrator,
The Nut, ChunkyPastaSauce, Tuborama, Peterh5322, Lowkyalur, Jefffolly, Lakkasuo, Petrb, ClueBot NG, Jaanus.kalde, MelbourneStar,
Piast93, Andreas.Persson, Historikeren, Robsuper, MerlIwBot, Helpful Pixie Bot, HMSSolent, Bibcode Bot, Supersam654, Citation-
CleanerBot, 1292simon, Braun walter, ChrisGualtieri, Dexbot, Frosty, Mark viking, Epicgenius, Acrislg, Jamesmcmahon0, Brzydalski,
9.1. TEXT 203

Spyglasses, Rewa, AddWittyNameHere, Jbolton07, Gerbenvaneerten, Barefootwhistler, MasterTriangle12, Grsh90 and Anonymous: 404
• Electron hole Source: http://en.wikipedia.org/wiki/Electron_hole?oldid=613374475 Contributors: Maury Markowitz, Tim Starling,
Lkesteloot, Omegatron, Bevo, Robbot, Ojigiri, Wjbeaty, LiDaobing, Karol Langner, DragonflySixtyseven, Roo72, Tirthajyoti, PhilHibbs,
Nigelj, Robotje, Ranveig, Capi crimm, Palica, Mandarax, Salleman, FlaBot, Arnero, Chobot, YurikBot, RobotE, Bambaiah, Archelon,
Shaddack, Spike Wilbury, Ninly, Sbyrnes321, The Photon, Shai-kun, Betacommand, Jcarroll, Bluebot, Jprg1966, Lagrangian, DMacks,
Jaganath, Robofish, JorisvS, Mgiganteus1, Noah Salzman, Tawkerbot2, Jh12, Chetvorno, Gogo Dodo, Delta Spartan, Envy0, Thijs!bot,
Barticus88, Salgueiro, JAnDbot, Britcom, Savant13, Sir Link, Sqush101, Aboutmovies, Andejons, Kurosa, Captainlavender, JhsBot, Jack
Naven Rulez, SieBot, Extremecircuitz, No such user, Addbot, Out of Phase User, Bob K31416, OlEnglish, ‫דוד שי‬, Uroboros, Meisam,
Legobot, Luckas-bot, LilHelpa, Xqbot, Leonardo Da Vinci, Erik9bot, Citation bot 1, Lissajous, EmausBot, K6ka, ClueBot NG, Crazy-
monkey1123, Bibcode Bot, Williammathew30, Iplaycards, Vanquisher.UA, YimmyYohnson and Anonymous: 59
• P–n junction Source: http://en.wikipedia.org/wiki/P–n_junction?oldid=627108074 Contributors: AxelBoldt, Waveguy, RTC, Dmd, Mac,
HolIgor, Auric, Wjbeaty, Ancheta Wis, Rafaelgr, Armandino, Mako098765, Abdull, Jfraser, Matt Britt, Foobaz, Timl, Storm Rider,
Keenan Pepper, Wtshymanski, Tebbb, Marudubshinki, Nanite, Amr Ramadan, Vegaswikian, LjL, Prgo, Alfred Centauri, Kolbasz, Tomer
Ish Shalom, Chobot, YurikBot, Sceptre, Gaius Cornelius, Shaddack, NawlinWiki, Bota47, Light current, Chaiken, Katieh5584, Attil-
ios, SmackBot, Jacek Kendysz, Mauls, JAn Dudík, Bluebot, Pieter Kuiper, MalafayaBot, Darth Panda, Apocryphite, Radagast83, Dr-
philharmonic, DMacks, Catani, Vriullop, Intellectnfun, JorisvS, CyrilB, Cikicdragan, Dicklyon, Filelakeshoe, Chetvorno, SkyWalker,
Christian75, Maque, Thijs!bot, Headbomb, Electron9, Gerry Ashton, AntiVandalBot, Email4mobile, Dukebody, Kskowron, Gresszilla,
TheNoise, MartinBot, Bissinger, Glrx, CommonsDelinker, LordAnubisBOT, NewEnglandYankee, Cmichael, DorganBot, PowerWill500,
VolkovBot, Larryisgood, Scholzilla, Someguy1221, Lerdthenerd, Andy Dingley, AlleborgoBot, Nagy, SieBot, VVVBot, Delu 85, Pratik
mallya, Nopetro, Wilson44691, Arjen Dijksman, Siyamraj, Anchor Link Bot, ClueBot, Brews ohare, Vboo-belarus, XLinkBot, Terry0051,
MystBot, Zinger0, Addbot, Mortense, Napy1kenobi, ProperFraction, Download, Jamesrei, Shrikul joshi, ScAvenger, Cesaar, Luckas-bot,
Yobot, Senator Palpatine, Choij, Materialscientist, Citation bot, Darcovian, DSisyphBot, Igorpark, Raffamaiden, Rickproser, Jangirke,
FrescoBot, Jc3s5h, BenzolBot, Youarefunny, MJ94, SpaceFlight89, Lowrybob, Javaidphy, ‫علی ویکی‬, TheGrimReaper NS, MrSnoot,
Bhawani Gautam, EmausBot, Beatnik8983, Dewritech, Monterey Bay, TyA, Xiutwel-0003, Noophilic, ClueBot NG, Starshipenterprise,
Jbolte, Widr, Helpful Pixie Bot, Wbm1058, Helloakshaypoddar, Metricopolus, Satishb.elec, Tarunselec, Ulidtko, C susil, Aloysius314,
IngenieroLoco, Ginsuloft, Mattkevmd, Jadecatz, Kirasan5 and Anonymous: 186
• Bipolar transistor biasing Source: http://en.wikipedia.org/wiki/Bipolar_transistor_biasing?oldid=608294834 Contributors: Brian-
Willoughby, Natrij, TedPavlic, Wtshymanski, Bruce1ee, Toffile, RadioFan, Moe Epsilon, Phil Holmes, Derek Andrews, Steve carlson,
Ohconfucius, Robofish, Rogerbrent, Dicklyon, Iridescent, Jaksmata, Xcentaur, Circuit dreamer, Alaibot, Qwyrxian, Marek69, Deficit,
Dr. Blofeld, Magioladitis, GreenSpigot, Brews ohare, Spitfire, Paushali, WikHead, Airplaneman, Addbot, Anypodetos, ThermalCat, Nas-
nema, Shadowjams, FrescoBot, Enery the 8th, Vrenator, Retro917, Peter Karlsen, 28bot, ClueBot NG, 220 of Borg, ChrisGualtieri, Ekren,
Azeezur rahman and Anonymous: 61
• 555 timer IC Source: http://en.wikipedia.org/wiki/555_timer_IC?oldid=627217654 Contributors: Damian Yerrick, AxelBoldt, Scipius,
Heron, RTC, Stw, Ahoerstemeier, Glenn, Nikai, Tomv, PeterGrecian, Omegatron, Huangdi, Alan Liefting, Giftlite, Brouhaha, Leonard G.,
Nielmo, Sonett72, Abdull, Grunt, ThreeE, NathanHurst, Discospinster, Pak21, Pmsyyz, Jcmaco, Alistair1978, Kaisershatner, Kwamik-
agami, Bobo192, Longhair, Towel401, Hooperbloob, Nazli, Alansohn, Keenan Pepper, Wtmitchell, Wtshymanski, Mikeo, SteveLetwin,
Gene Nygaard, Kay Dekker, Mindmatrix, Sdschulze, Jacj, Palica, Pfalstad, Mandarax, Dubkiller, Pbeens, Brighterorange, Ptdecker, FlaBot,
Nihiltres, RexNL, Turidoth, Gwernol, YurikBot, Adam1213, RussBot, Hellbus, Toffile, Hyjwei, Mikebest, Mikeblas, Speedevil, Ninly,
Nkendrick, Ataub2qf, SmackBot, Elonka, The Photon, Stefan506, Colin99, Bromskloss, Gilliam, Hlovdal, MrDrBob, Thumperward, Oli
Filth, Papa November, Audriusa, Can't sleep, clown will eat me, Frap, Aznium, Weregerbil, Flsp70, DireWolf, IronGargoyle, Llamadog903,
CyrilB, Dicklyon, MTSbot, Dl2000, Hu12, Jachim, Tawkerbot2, Mikiemike, Ubernerd, Circuit dreamer, WeggeBot, **mech**, Bill
(who is cool!), A876, After Midnight, Click23, Foil166, Sprhodes, Dtgriscom, Marek69, Electron9, Mallred, Dgies, Mihtjel, Escar-
bot, Hmrox, AntiVandalBot, Luna Santin, Ndyguy, MER-C, Wser, Hut 8.5, RebelRobot, .anacondabot, VoABot II, Mondebleu, Chkno,
ArmadilloFromHell, Welle4, Mermaid from the Baltic Sea, Keith D, Jerry teps, The Canadian Roadgeek, J.delanoy, Jcurie, Vesa Linja-
aho, NightFalcon90909, Flicovent, Kraftlos, Pundit, Potatoswatter, Bonadea, Ripper-b, VolkovBot, ICE77, Lexein, Quentonamos, Philip
Trueman, Hqb, Olly150, Jack1993jack, Inductiveload, Suriel1981, PeterEasthope, Truthanado, Jtcampbell, Josh the Nerd, Yintan, Fdu-
raibi, Jp314159, Oda Mari, JSpung, Ebarnett, Allmightyduck, XU-engineer, Shooke, Superbeecat, Prasanthv88, Bekuletz, ClueBot, Tim
Forcer, Avenged Eightfold, Snigbrook, The Thing That Should Not Be, WaltBusterkeys, SuperHamster, Kiu77, (void*), Somno, Excirial,
Rswarbrick, RexxS, Against the current, Rror, AbstractBeliefs, Actam, Avoided, Thatguyflint, Addbot, Technicalpic, Ronhjones, Jncraton,
Eivindbot, Glane23, Favonian, Heshamdiab116, Numbo3-bot, Ettrig, SLourenco, Quadrescence, Yobot, CSSINC, Motif user, Materi-
alscientist, Xqbot, Ywaz, Bubble-boy-115, JWBE, Jmundo, SassoBot, Jacksonmiss, Prari, FrescoBot, 6hug99ko, ZenerV, Tetraedycal,
42murr42, DrilBot, Kimphill, A8UDI, Btilm, MinimanDragon32, Mikespedia, Jrkemerer, Lissajous, ApusChin, Segal’sLaw, Techwet-
paintwiki, Vikasjois, Nikhilpatel4488, DARTH SIDIOUS 2, Gloomofdom, EmausBot, WikitanvirBot, Immunize, Sophie, VOG-NevaDA,
Colin555, John Cline, East of Borschov, Sbmeirow, Raghavendrabsrg, DASHBotAV, Petrb, ClueBot NG, Terry caborn, 3000farad, Reify-
tech, Mtmcdaid, Ercrt, Jupiter Kasparov, ChrisGammell, MusikAnimal, Satishb.elec, Colin5555, Trevayne08, Mattsains, Hcamen, Lor-
dOider, Dhx1, Mogism, Techdude3331, Junbert hular, Guanta37201, AnthonyRobinson123, Sravan75, Nattsukhdeep, Sureshkumar.suraj,
Wasdichsoveraenderthat, ElectronicKing888, Jelabon123 and Anonymous: 335
• Operational amplifier Source: http://en.wikipedia.org/wiki/Operational_amplifier?oldid=627316180 Contributors: Mav, The Anome,
WillWare, Malcolm Farmer, Heron, Edward, RTC, Michael Hardy, Mahjongg, Wapcaplet, Julesd, Glenn, Ghewgill, GRAHAMUK,
Bemoeial, Dysprosia, Andrewman327, Omegatron, Wernher, AnonMoos, AlexPlank, Hankwang, RedWolf, Donreed, Smither, Pengo,
Giftlite, DavidCary, Mintleaf, Inkling, Ds13, CyborgTosser, Leonard G., Frencheigh, Rpyle731, Mboverload, Foobar, Mike R, Aulis
204 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

Eskola, DRE, Eranb, KeithTyler, Clemwang, M1ss1ontomars2k4, Adashiel, TedPavlic, ArnoldReinhold, Sn0wflake, ESkog, Plugwash,
Dpotter, CanisRufus, Bdieseldorff, Shanes, Bobo192, Nigelj, .:Ajvol:., Foobaz, Hooperbloob, Musiphil, Neonumbers, Atlant, Keenan
Pepper, Wtmitchell, Wtshymanski, Gene Nygaard, Alai, DSatz, Unixxx, Weyes, Woohookitty, Mindmatrix, Pol098, Cbdorsett, Gim-
boid13, Cataclysm, Msiddalingaiah, Snafflekid, Bvankuik, Sjakkalle, Arabani, Alejo2083, Chris Pressey, Ground Zero, Margosbot, Ef-
ficacy, Alfred Centauri, RexNL, Enon, Fresheneesz, Srleffler, YurikBot, Borgx, Gaius Cornelius, Rohitbd, Synaptidude, Grafen, 48v,
Sparkinark, TDogg310, Voidxor, Supten, DeadEyeArrow, Elkman, Searchme, Bakkster Man, Light current, Super Rad!, Closedmouth,
Mike1024, Mebden, Luk, Peranders, SmackBot, Igtorque, Thelukeeffect, Speight, Man with two legs, Jwestbrook, Lindosland, QEDquid,
KD5TVI, Chris the speller, Bluebot, Oli Filth, EncMstr, Papa November, DHN-bot, FredStrauss, Audriusa, Royboycrashfan, Can't sleep,
clown will eat me, LouScheffer, CanDo, TWINE006, Henning Makholm, Evlekis, Bejnar, Ohconfucius, Missamo80, Archimerged, Aw-
bliven, JohnWittle, Antireconciler, CyrilB, Rogerbrent, Dicklyon, Novangelis, Kvng, Pgadfor, Yves-Laurent, Phil Christs, JForget, Circuit
dreamer, WeggeBot, Myasuda, Anoneditor, Naspilot, Pyrilium, ZHENG Jiajun, Viscious81, HermanFinster, Christian75, Khattab01,
Raidfibre, Josemiotto, Editor at Large, Saintrain, Serych, Barticus88, Sunny sin2005, Electron9, Vasurak, Jonnie5, Nick Number, Es-
carbot, AntiVandalBot, Luna Santin, Seaphoto, Lovibond, MikeLynch, JAnDbot, .K, CosineKitty, Magioladitis, Secret Squïrrel, Acu7,
Jerome Baum, Hypergeek14, SwiftBot, Clankypup, Crunchy Numbers, First Harmonic, Allstarecho, Nodekeeper, Cocytus, Manavbhard-
waj, Mårten Berglund, Sigmundg, Glrx, CommonsDelinker, Jascii, J.delanoy, Jcurie, Sakthivel30, Zen-in, Szzuk, SJP, Rumpelstiltskin223,
TottyBot, Homer Landskirty, Pmoseley, Red Thrush, VolkovBot, ICE77, Philip Trueman, DoorsAjar, Draurbilla, Ianr44, Bizarro Bull, In-
ductiveload, Andy Dingley, Spinningspark, SieBot, Caltas, Charles.small, Roy hu, Jjwilkerson, Jp314159, Opamp, M Puddick, OsamaBin-
Login, Evaluist, Faradayplank, PHermans, OscarJuan, PerryTachett, ClueBot, Smart Viral, Binksternet, ArthurOgawa, Mild Bill Hiccup,
Srinivasbt, Alexbot, Jusdafax, Brews ohare, Sldghmmr, La Pianista, Wstorr, Aitias, Johnuniq, SoxBot III, Analogkidr, Salam32, Alexius08,
ZooFari, Skonieczny, Gggh, Addbot, Mortense, Some jerk on the Internet, Olli Niemitalo, Theleftorium, Sudeepa123, Ppj4, Punkguitar,
AnnaFrance, Favonian, LinkFA-Bot, Bruno gouveia rodrigues, Supav1nnie, Eng general, OlEnglish, Pietrow, Zorrobot, RobertTanzi,
Luckas-bot, Yobot, Ptbotgourou, Annomination, AnomieBOT, Paulthomas2, Jim1138, JackieBot, Piano non troppo, LiuyuanChen, Ma-
terialscientist, Citation bot, Akilaa, Xqbot, Sellyme, XZeroBot, ‫دانقولا‬, Dprabhu, Maitchy, Endothermic, GliderMaven, FrescoBot, Lu-
cienBOT, Ong saluri, Berrinkursun, Anitauky, Wikigayburgers, Roman12345, Vhann, Gdje je nestala duša svijeta, I dream of horses,
Jonesey95, Hoo man, Ezhuttukari, Mikespedia, Abhishekchavan79, Æneas, Dinamik-bot, Overjive, PleaseStand, DARTH SIDIOUS 2,
Teravolt, Breezeboy, EmausBot, John of Reading, Kasper Meerts, Torturella, Tawsifkhan, Solarra, Meht7860, Fæ, East of Borschov,
Aflafla1, Thine Antique Pen, Rcsprinter123, Sbmeirow, Peterh5322, L Kensington, Danielop-NJITWILL, VictorianMutant, Petrb, Sud-
heerp99, ClueBot NG, Jaanus.kalde, Grottolese, Muon, Iinvnt, Widr, Lain.cai, Oddbodz, Helpful Pixie Bot, Wbm1058, Czar44, So-
daant, Minsbot, Yogirox234, Hghyux, MarinSwimmer, ChrisGualtieri, SD5bot, Dexbot, Jamesx12345, Visitor01, Monkbot, Pcrengnr,
TerryAlex, MorganBEAST, AntonKrugerAtUiowa, Madphysics and Anonymous: 445
• Phase-locked loop Source: http://en.wikipedia.org/wiki/Phase-locked_loop?oldid=626157396 Contributors: Bryan Derksen, The Anome,
Ap, PierreAbbat, Heron, Michael Hardy, Karada, CesarB, PingPongBoy, Mac, Glenn, Raven in Orbit, Technopilgrim, Dcoetzee,
Lkesteloot, Omegatron, Maheshkale, Robbot, Jotomicron, Naddy, Mirv, Iain.mcclatchie, Cutler, Alan Liefting, Giftlite, Brouhaha, Neffk,
Sarex, Tietew, BrianWilloughby, Abdull, RevRagnarok, D6, JGeld, Rich Farmbrough, Xezbeth, CanisRufus, Cmdrjameson, Timl, Ilt-
seng, Hooperbloob, Tom Yates, Arthena, Wtshymanski, Brholden, DV8 2XL, Gene Nygaard, Alai, Linas, Sburke, Ruud Koot, Burgher,
Jonnabuz, Msiddalingaiah, Snafflekid, Kotukunui, Rjwilmsi, Leeyc0, Tawker, Mbutts, FlaBot, Chris Pressey, Mel Gibson, Gurch, Roby-
Wayne, WriterHound, YurikBot, Encyclops, Prometheus235, Toffile, Hydrargyrum, Brandon, Guerberj, Mikeblas, Mysid, Light cur-
rent, Deville, Ninly, Alanb, LeonardoRob0t, Whaa?, Zvika, Pankkake, SmackBot, H2eddsf3, Reedy, Unyoyega, Fulldecent, Larrykoen,
Lindosland, Chris the speller, Bluebot, Oli Filth, Papa November, Can't sleep, clown will eat me, Frap, OrphanBot, Shanid, Nakon,
Wirbelwind, Dr. Crash, Viyh, Kvng, Paul Koning, Yenhsrav Keviv, Tudy77, Blehfu, Chetvorno, Emote, Eastlaw, Philshea, Chrumps, Re-
question, Johnlogic, Mattisse, Alex Forencich, DmitTrix, Electron9, Prolog, Hmo, Beabroad, User A1, Glrx, R'n'B, Huzzlet the bot,
Kar.ma, Rod57, Ganymedstanek, Vanished user 47736712, Jrolston, Rex07, Frodo avr, Squids and Chips, Anton Rakitskiy, TXiKi-
BoT, ElinorD, ChooseAnother, Jpat34721, Billgordon1099, Wangyiliu99, Dirkbb, Faduman, AlleborgoBot, Plan10, SieBot, Cwkmail,
Mwaisberg, EngineerSteve, HURRICANE1415, WWStone, Vahid avr, IR-TCI, WakingLili, Dp67, ClueBot, PipepBot, Yegorius, 718
Bot, Stevenmyan, Alexbot, PixelBot, Jimfordbroadcom, Peter.C, SchreiberBike, DumZiBoT, Apps guy, Feinoha, MystBot, J 0JFCfr-
mAyw59oVFk, Dsimic, Addbot, AVand, Duketron, H92Bot, Mitch feaster, Lightbot, Wireless friend, Sechinsic, Legobot, Luckas-bot,
OrgasGirl, Castagna, AnomieBOT, LilHelpa, Obersachsebot, Grim23, Devanney, Feldhaus, Jangirke, Prosaicpat, MastiBot, Merlinsorca,
Renatyv, HitiABC, Cetsurfer, EmausBot, John of Reading, Tolly4bolly, Orange Suede Sofa, ClueBot NG, Daithiob, Helpful Pixie Bot,
Emresearch13, Neøn, LeonovGA, Kuznetsov N.V., Frogging101, Rabachand, 08Peter15, Dexbot, Bizet74, Vicentealvarez2, Ehud.ahissar,
Avinash701, Drjimbonobo, Lizbel123, AntonKrugerAtUiowa, Crystallizedcarbon and Anonymous: 217
• Voltage regulator Source: http://en.wikipedia.org/wiki/Voltage_regulator?oldid=627175051 Contributors: BlckKnght, Heron, Bdonlan,
Glenn, GRAHAMUK, Zoicon5, Maximus Rex, Omegatron, Hankwang, Henrygb, DavidCary, Scott MacLean, Sonett72, Jcmaco, Rubi-
con, Jaberwocky6669, Plugwash, RoyBoy, Nigelj, Timl, Hooperbloob, Jakew, Mareino, 1-1111, Tpikonen, Atlant, Wtshymanski, Pol098,
Mandarax, Tizio, Brighterorange, Yamamoto Ichiro, Margosbot, Alfred Centauri, Srleffler, Jidan, Visor, DMahalko, Toffile, Gaius Cor-
nelius, Rohitbd, Ethan, Spike Wilbury, Mikeblas, LeoNerd, Hirak 99, Ninly, SmackBot, Unyoyega, Gilliam, Lindosland, Chris the speller,
Oli Filth, Spacemeng, Audriusa, Bsilverthorn, Frap, Lagrangian, KLLvr283, CyrilB, Slakr, Kvng, TW2, Yves-Laurent, Nathanbriggs,
Chetvorno, Circuit dreamer, TheTito, Fairsing, Cydebot, Ebraminio, Al Lemos, JAnDbot, PhilKnight, Bongwarrior, VoABot II, J.P.Lon,
Chris G, Manavbhardwaj, Nono64, Silverxxx, Atropos235, Almazi, Philip Trueman, Dsignoff, Paverider, Biscuittin, Keilana, Sriramk750,
Allmightyduck, Harry the Dirty Dog, Hamiltondaniel, Dolphin51, Floodamanny, OldenGray, ClueBot, SuperHamster, No such user, Di-
agramma Della Verita, DeltaQuad, Pardo bsso, SoxBot III, XLinkBot, Mifter, Oboylej, Badgernet, MystBot, Addbot, MrOllie, Glass
Sword, Zaphodikus, Semiwiki, Gail, Arsenalboi20, Götz, Jim1138, Materialscientist, RibotBOT, Amaury, Maitchy, Chongkian, Shadow-
jams, MZIrish, FrescoBot, Þjóðólfr, Jaymie94, Pinethicket, Kl2217, Boobarkee, Rxgvhqkrywq6cxdjoar, Tucvbif, Di gama, Yusiang1998,
9.1. TEXT 205

Lee A. Hart, Eirik1231, DexDor, AndyHe829, Wyatte Gillette, EmausBot, Orphan Wiki, Dcirovic, ZéroBot, Mkratz, Pgarg78, Sbmeirow,
Mohsen.1987, Donner60, ClueBot NG, Akashacharyak, Ulrich67, Rezabot, MerlIwBot, Helpful Pixie Bot, Aholyokeb, Wbm1058, Matare-
sephotos, Mpalframan, Tiscando, Vydeoatpict, Jschnabs, Jamietwells, Nilay.pant, Tentinator, Buntybhai, Monkbot, GinAndChronically,
RoyPijnenburg, Mario Castelán Castro, ChamithN and Anonymous: 205
• Comparator Source: http://en.wikipedia.org/wiki/Comparator?oldid=627647264 Contributors: Heron, Edward, Ahoerstemeier, Glenn,
Smack, Schneelocke, Omegatron, Dbroadwell, Stuuf, Jaan513, Discospinster, Sietse Snel, Hooperbloob, Andrewpmk, Cburnett, Jannex,
Matthew Platts, Mandarax, Maxim Razin, Arnero, Crazycomputers, Stassats, Guerberj, Katieh5584, Bmearns, Lindosland, Bluebot, Day-
dreamer302000, Af1218, Barney Stratford, S Roper, Jeff Wheeler, Paul Rako, Avé, 16@r, Dicklyon, Daharde, Chetvorno, Mikiemike, Cir-
cuit dreamer, TAB, Thijs!bot, CosineKitty, Adilsm, Scottr9, R'n'B, Amikake3, Grantdj, Miwanya, Inductiveload, AlleborgoBot, Flyer22,
Sokari, ClueBot, The Thing That Should Not Be, Optics guy07, VQuakr, Mild Bill Hiccup, Somno, Three-quarter-ten, ChardonnayN-
imeque, PixelBot, SpikeToronto, Analogkidr, Addbot, Mortense, Yobot, Xqbot, NSK Nikolaos S. Karastathis, FrescoBot, Febert, Smur-
fettekla, Berrinkursun, Cannolis, I dream of horses, Calmer Waters, Chotugubbi, Eguru37, Wikieditorz, Reconsider the static, EmausBot,
ZéroBot, East of Borschov, Derekleungtszhei, ChuispastonBot, ClueBot NG, Helpful Pixie Bot, CitationCleanerBot, BattyBot, Jaspritsgill,
Imneerajgarg3, Monkbot, Ksumatrob and Anonymous: 109
• Thermistor Source: http://en.wikipedia.org/wiki/Thermistor?oldid=626052477 Contributors: Sodium, Bryan Derksen, Zundark, Szopen,
Ray Van De Walker, Heron, Tim Starling, Tango, Ellywa, Zoicon5, Maximus Rex, Omegatron, Carbuncle, Jni, Twang, Chuunen Baka,
Donarreiskoffer, Robbot, Hankwang, Tonsofpcs, Sho Uemura, Alan Liefting, Buster2058, Centrx, BenFrantzDale, Tsca, Leonard G., Bob-
blewik, Wmahan, DougEngland, Zondor, Grunt, Plugwash, Kjkolb, Hooperbloob, Alansohn, PAR, Velella, RainbowOfLight, Henry W.
Schmitt, Gene Nygaard, Vadim Makarov, Jimgeorge, Ruud Koot, Graham87, War, Toki, Brighterorange, Unfocused, Allen Moore, FlaBot,
Margosbot, Srleffler, Zotel, Chobot, Karch, Mhking, Cookie4869, CambridgeBayWeather, Shaddack, Veledan, Howcheng, Raven4x4x,
RUL3R, Searchme, Light current, Morcheeba, Closedmouth, Tabby, Sbyrnes321, Finell, Grend3l, Waulfgang, SmackBot, Thorseth, Iopq,
Papep, Ohnoitsjamie, Oscarthecat, Bluebot, Oli Filth, SchfiftyThree, Милан Јелисавчић, BullRangifer, Doodle77, Mircealutic, Mion,
Shirifan, Dicklyon, Komeil, Tawkerbot2, Bemasher, CmdrObot, Zureks, Requestion, NickFr, Ussensor, Yaris678, PureGenius, NascarEd,
N5iln, James086, Drigz, Greg L, Nemilar, Stannered, AntiVandalBot, Quintote, Myanw, JAnDbot, Txomin, Rob Kam, Mdurante, Magi-
oladitis, Diablod666, VoABot II, Rivertorch, Chakri srivatsa, Madmanguruman, MartinBot, Manavbhardwaj, Neeners, Jim.henderson,
Tgmsfu, EliV, Choihei, Ashtead Tutor, Dhaluza, Halmstad, Ibdelfest, Funandtrvl, Nasanbat, VolkovBot, ABF, Constant314, TXiKi-
BoT, Oh Snap, Salvar, Brian Helsinki, Broadbot, Jackfork, Madhero88, Triesault, AlleborgoBot, Neparis, SieBot, God Emperor, Mas-
gatotkaca, Seraphal, ClueBot, Janz94b, Mardetanha, Awickert, Plainman, Estirabot, Dittos12, Paulienator, Karpouzi, Teslaton, Rob-bob7-
0, SkyLined, Kbdankbot, Addbot, Mortense, Willking1979, Captain-tucker, AkhtaBot, West.andrew.g, Jnmurfin, Lou Mueller, Cesaar,
Legobot, Luckas-bot, Yobot, Ptbotgourou, Amirobot, QueenCake, Tempodivalse, AnomieBOT, Rubinbot, Ndgrahams, Aditya, Mate-
rialscientist, ArthurBot, Xqbot, Vichou, Nedim Ardoğa, Yoganate79, Jangirke, IShadowed, Idyllic press, Pinethicket, SpaceFlight89,
Mayank2507, TjBot, EmausBot, Dewritech, GoingBatty, Tommy2010, Hhhippo, Back0ut, Jwortzel, SamuelFreli, ClueBot NG, Mel-
bourneStar, Lochlan1, Widr, Helpful Pixie Bot, Ramaksoud2000, Piguy101, Klilidiplomus, Mrt3366, MadCowpoke, Yukichen, Isarra
(HG), BeaumontTaz, Sihuapilapa, Chestercheryl, RogerDulhunty, Epicgenius, R.Mann66, Jodosma, Yardimsever, 99kmg365, Zeniu17,
AMWEI Thermistor and Anonymous: 254
• Photodiode Source: http://en.wikipedia.org/wiki/Photodiode?oldid=627766575 Contributors: Css, Perry Bebbington, Heron, Freckle-
foot, Bdesham, RTC, Tim Starling, Rabin, Liftarn, Sannse, Glenn, Nikai, Andres, Epo, Reddi, Omegatron, Francs2000, Robbot, Fredrik,
Roscoe x, Giftlite, Ferkelparade, Mako098765, Rdsmith4, Nickptar, Deglr6328, Murtasa, Shanes, Bookofjude, Giraffedata, BigRiz,
Hooperbloob, Atlant, Wtshymanski, DV8 2XL, Linas, Pol098, Palica, RuM, Snafflekid, Rjwilmsi, KaiMartin, Weihao.chiu, Srleffler,
Chobot, Jaraalbe, YurikBot, Jeffthejiff, Hede2000, Jhchang, Kkmurray, Morcheeba, Pb30, GrinBot, KnightRider, SmackBot, Terry1944,
The Photon, Fulldecent, Ohnoitsjamie, Vercalos, Lindosland, HenrikS, Oli Filth, A. B., Audriusa, LouScheffer, DinosaursLoveExis-
tence, Nakon, Xagent86, Rcopley, DJIndica, Anachron, NANOIDENT, Breno, Minna Sora no Shita, MadScientistVX, Beetstra, Dicklyon,
Spiel496, Kvng, Moonkey, Ddcc, Tarchon, Nilfanion, Shadwstalkr, Dancter, Mitjaprelovsek, Thijs!bot, Electron9, Tellyaddict, Big Bird,
KrakatoaKatie, Whogue, Gh5046, Em3ryguy, Vanished user ty12kl89jq10, Vssun, DerHexer, Kevglynn, MartinBot, Ctroy36, J.delanoy,
Pharaoh of the Wizards, Jstahley, Zen-in, Dlegros, Sgc2002, Philip Trueman, Srdhrp vict, Daisydaisy, Adam C C, LeaveSleaves, Silver-
bone, Tmaull, Andy Dingley, AledJames, WingkeeLEE, Cyfal, ClueBot, Robertmuil, Lucas the scot, Nrnkpeukdzr, AlanM1, H0dges,
Addbot, Mathieu Perrin, A0602336, Download, Shkirenko edik, Favonian, Bwrs, Semiwiki, Yobot, Amirobot, Fotodiod, SwisterTwister,
AnomieBOT, Jim1138, Materialscientist, Xqbot, Erud, Capricorn42, RibotBOT, Nedim Ardoğa, Chn Andy, FrescoBot, Arande2, Evan
124, Tóraí, Scwks, Devindunseith, Texasron, EmausBot, John of Reading, WikitanvirBot, Giant12111, Sbmeirow, Karthikndr, Atcold,
Tls60, Puffin, Intellec7, Staticd, ClueBot NG, Fulvio314, Helpful Pixie Bot, OAnimosity, Djjhs, Himu.arif, Fathpour, Phadd, JaunJimenez,
GoLiang, Jogi.sreeram and Anonymous: 218
• Photoresistor Source: http://en.wikipedia.org/wiki/Photoresistor?oldid=624078563 Contributors: Heron, Bdesham, RTC, Tim Starling,
Egil, Iain, Julesd, Glenn, Andres, Bemoeial, ElusiveByte, Omegatron, Raul654, Eugene van der Pijll, Robbot, Hankwang, Aechols, Glenn
Koenig, Adashiel, Mike Rosoft, Jiy, Neko-chan, Gilgamesh he, Femto, Billymac00, Giraffedata, Samadam, Hooperbloob, Alansohn,
Gary, Atlant, Evil Monkey, Gene Nygaard, UFu, Cbdorsett, Cyberman, K3wq, Rjwilmsi, Ademkader, Alfred Centauri, Kolbasz, Srl-
effler, YurikBot, Adam1213, Hydrargyrum, Gaius Cornelius, Havok, NawlinWiki, Wiki alf, Mssetiadi, Elkman, Ikonen, Tabby, Donald
Albury, Kevin, Neilgravir, Sbyrnes321, KnightRider, Gigs, Erbbysam, Relaxing, Wheatleya, Oli Filth, Pevarnj, Akendall, CyrilB, OZOO,
E-Kartoffel, Yves-Laurent, Dancter, Thijs!bot, Frozenport, Jojan, Marek69, Zaiken, AntiVandalBot, Luna Santin, Âme Errante, Ioeth,
JAnDbot, MER-C, CosineKitty, Hut 8.5, VoABot II, Kharkless, MartinBot, Aragorn450, Gzkn, Paulvermillion, Jim Swenson, Jennavecia,
Amikake3, Ulfbastel, Clarince63, Melsaran, Yk Yk Yk, KyleSinghSaini, SieBot, Flyer22, Richgoddard, Cameron7935, C4udy, ClueBot,
PaulLowrance, Heracletus, DragonBot, PixelBot, BOTarate, HumphreyW, DumZiBoT, Teslaton, Addbot, Mortense, Willking1979, Ele-
ment16, Ronhjones, Sploshy100, Glane23, Chzz, Semiwiki, Luckas-bot, TaBOT-zerem, Muranesenema, KDS4444, Ciphers, Poetman22,
206 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

Cachao, Duhos, RibotBOT, Prari, FrescoBot, Jc3s5h, MGA73bot, PigFlu Oink, Kutayzorlu, Pinethicket, SpaceFlight89, Dinamik-bot,
Soulhack, ‫دالبا‬, EmausBot, Go82102, Mkratz, HawkMcCain, TYelliot, Mikhail Ryazanov, ClueBot NG, O.Koslowski, Widr, Help-
ful Pixie Bot, HMSSolent, Vaibhavsg.elec, Quickshivam, FynnM, Chip123456, Phillip mcfuzz, Lxdcn, SFK2, Arunwebber, Tentinator,
Ruprecht27, JaunJimenez and Anonymous: 190
• Analogue switch Source: http://en.wikipedia.org/wiki/Analogue_switch?oldid=624623726 Contributors: SimonP, Waveguy, Heron,
Colin Marquardt, Zoicon5, Omegatron, Robbot, Mintleaf, Everyking, Glogger, Hooperbloob, Gene Nygaard, Uncle G, Josh Parris,
KaiMartin, Bmicomp, Light current, SmackBot, Oli Filth, Mion, Catapult, Pi.1415926535, Mild Bill Hiccup, Addbot, Erik9bot, Febert,
Berrinkursun, Gbalasandeep, Anitauky, EmausBot, Kuashio, Rangoon11, Thawk59 and Anonymous: 10
• Decibel Source: http://en.wikipedia.org/wiki/Decibel?oldid=628262911 Contributors: Tobias Hoevekamp, DrBob, Heron, Spiff, Edward,
Patrick, Michael Hardy, Iluvcapra, Ahoerstemeier, Hpa, BAxelrod, Smack, Timwi, Pheon, Furrykef, Omegatron, Indefatigable, Pakaran,
Hankwang, Pigsonthewing, Peak, Academic Challenger, Rsduhamel, Pabouk, Giftlite, DocWatson42, DavidCary, Mintleaf, BenFrantz-
Dale, Zigger, Karn, Abqwildcat, Jason Quinn, Mboverload, Rchandra, AlistairMcMillan, Antandrus, Cihan, Icairns, Sonett72, Brian-
Willoughby, Mormegil, Jayjg, Abar, Rich Farmbrough, Guanabot, Florian Blaschke, Xezbeth, Alistair1978, Bender235, ZeroOne, A
purple wikiuser, Plugwash, Petersam, Mark R Johnson, Alansohn, Vslashg, Duffman, RPaschotta, Andrewpmk, SlimVirgin, !melquiades,
Kotasik, Hu, Benjamin.Heasly, Hessi, Wtshymanski, Cburnett, Stephan Leeds, Gene Nygaard, Peter Hanes, Nuno Tavares, Mário, LOL,
Pol098, Walker44, Kristaga, MONGO, Kgrr, Pyrosim, Cbdorsett, Scm83x, Jwoodger, E090, Deltabeignet, BD2412, Rjwilmsi, Joffan,
Aechris, Wesley R. Elsberry, N0YKG, Titoxd, Ianharvey, RexNL, Ayla, DrVeghead, Fresheneesz, Skierpage, Srleffler, Glenn L, Epit-
ome83, King of Hearts, Chobot, DVdm, Crovax, Oh2mqk, Whosasking, Rmo13, Wolfmankurd, RussBot, Muchness, Peter S., Xihr,
Splash, JabberWok, BlongerBros, Hydrargyrum, Manop, Sixteen Left, Burek, Accurrent, Wiki alf, Razer64, Dhollm, Zephalis, Asbl, Eu-
rosong, Ke6jjj, Light current, Noclip, Open2universe, Knotnic, Nikkimaria, Pb30, Esprit15d, David Biddulph, Thesleepwalker, GrinBot,
Bo Jacoby, Cmglee, Sbyrnes321, DVD R W, CIreland, SmackBot, Tgoose, Bobet, Xephael, CelticJobber, Hydrogen Iodide, Fuchsia Groan,
Cutter, RichGibson, GraemeMcRae, Bromskloss, Commander Keane bot, Gilliam, Skizzik, JAn Dudík, Lindosland, PeterCooperJr, Blue-
bot, Kurykh, Oli Filth, Greatgavini, Nbarth, Colonies Chris, Dual Freq, Emurphy42, GeeksHaveFeelings, Microfrost, Rrburke, Spectro-
gram, Cybercobra, MichaelBillington, Kneale, Gurnec, Fitzhugh, Romanski, Daniel.Cardenas, Mostlyharmless, Ugur Basak Bot, Rklawton,
Straif, Hefo, Jidanni, Henry’s Cat, Beta34, JustinSmith, Dicklyon, Tobyw87, SolarAngel, Dr.K., Noleander, Kvng, Samstayton, Dl2000,
ILovePlankton, Smsaladi, Emote, TheHorseCollector, Bungledb, Myncknm, Desrt, Tarchon, Cydebot, MC10, JJC1138, Zginder, Rajki-
ran g, DumbBOT, Ehudzel, Centuriono, Thijs!bot, Yhevhe, Epbr123, Mdberg1, Headbomb, Dtgriscom, Sobreira, Marek69, Chris01720,
Neil916, Electron9, CharlotteWebb, Uruiamme, Auralcircuitry, AntiVandalBot, Luna Santin, RapidR, Rutuag, Chill doubt, Alphachimp-
bot, Ingolfson, JackSparrow Ninja, JAnDbot, MER-C, Magioladitis, Bongwarrior, VoABot II, Catslash, SHCarter, Stotan, Ling.Nut,
MarcusMaximus, Enquire, Waninge, Pan Dan, WLU, The Sanctuary Sparrow, Amphedu, MartinBot, IgorSF, Ephracis, Jim.henderson,
Rettetast, Anaxial, R'n'B, Bjerke, SSSidhu, M samadi, Mange01, Ali, Poopface666, Nolaiz, Floaterfluss, Pcfjr9, Kraftlos, Robinparfitt,
Juliancolton, Davecrosby uk, Mlewis000, Sapphire12758, Jeff G., Toddles29, EchoBravo, Boute, Philip Trueman, The Original Wildbear,
Oric.gr, Link, Vipinhari, A4bot, Hqb, Pediacycle, Tonyjeffs, Billydee84, J.C. Roberts, Dendodge, Greg searle, Hanjabba, Jhawkinson,
Alexjliu, Spinningspark, Insanity Incarnate, NMTPhysics, Tbmoney1000, Thunderbird2, Mr. PIM, Markwads, Bluemouse2306, Thomas
EL Smith, Kbrose, Bohemian simian, Caltas, Yintan, Ygramul, Permacultura, Bsherr, Antonio Lopez, Kielhofer, AMCKen, OscarJuan,
Kudret abi, Gomertrash, Dellr, Jjefferp, ClueBot, Rumping, Binksternet, Eric Wester, EoGuy, Thubing, Meekywiki, Uncle Milty, Tim-
mmmCam, Klenod, Sun Creator, Dirt Tyrant, Madkaugh, Jonverve, Fmiser, GeoffMacartney, DumZiBoT, JKeck, AlanM1, Avoided,
Superbirk, Subversive.sound, Kace7, HexaChord, Addbot, Caden, Ocdnctx, Fgnievinski, Computer Guy 990, GyroMagician, Canadi-
anLinuxUser, C6H3N3O3, NjardarBot, Redheylin, Glane23, LemmeyBOT, Numbo3-bot, Ehrenkater, Fundamentisto, Didie, Legobot,
Yobot, AnomieBOT, Götz, Jeni, Bently64, Ipatrol, Materialscientist, Citation bot, Foreman1280, Richard Jay Morris, Nifky?, Oshasafety,
Jeffrey Mall, Heddmj, Nasa-verve, WaysToEscape, Dave3457, Dogbert66, Jc3s5h, Euc, Oalp1003, Citation bot 1, Pinethicket, Princes-
sofLlyr, 10metreh, RedBot, SpaceFlight89, Dtrx, North8000, Vrenator, Fastilysock, Goph'r, Marie Poise, DARTH SIDIOUS 2, Caf-
feein, Jaguarondi, EmausBot, Immunize, Mordgier, Dewritech, Roshamboz, O'DaveY, Wikipelli, BrianServis, Josve05a, Shofus, Don-
dervogel 2, H3llBot, Glockenklang1, ChemMater, Mltinus, Divineale, Mxctor, ClueBot NG, DieSwartzPunkt, Widr, Helpful Pixie Bot,
Kurtdebille, Rsercher, Kamerrill, BG19bot, Gurt Posh, Emayv, Yuanquan74, Daleshirk, J991, Planetary Chaos Redux, Mark Arsten,
CitationCleanerBot, NotWith, Aisteco, BattyBot, Pad3509, Brownb 22, ChrisGualtieri, GoShow, Garamond Lethe, Khazar2, Ntmbeast,
Riccardo4444, Smit 58, Reatlas, Hoodecho, Faizan, Epicgenius, Seanhalle, BreakfastJr, Frozenprakash, Lolguy9, Babitaarora, Spyglasses,
Quadlou, Jackmcbarn, W. P. Uzer, Coreyemotela, 22lsufan, MrsBentonsclass, Swarhili, Jayakumar RG, Simplexdave and Anonymous:
635
• Noise (electronics) Source: http://en.wikipedia.org/wiki/Noise_(electronics)?oldid=628056538 Contributors: William Avery, Heron,
Omegatron, Smither, Alan Liefting, Ancheta Wis, Giftlite, DavidCary, Vadmium, Keenan Pepper, Penwhale, Gene Nygaard, TheAr-
madillo, Kosher Fan, Pol098, Rjwilmsi, Bruce1ee, Lmatt, Splintercellguy, Toffile, Daniel C, Light current, Deville, Tevildo, Kle0012,
SmackBot, Ssbohio, Macintosh User, Lindosland, Oli Filth, SchfiftyThree, Il palazzo, Scwlong, Dreadstar, RickO5, Rogerbrent, Dick-
lyon, Waggers, Chetvorno, TristanJ, Tbone2001, Cydebot, Martin Hedegaard, JAnDbot, Mytomi, Rich257, STBot, Glrx, Mange01,
RSRScrooge, Maurice Carbonaro, Rod57, Jeepday, Funandtrvl, Jamelan, Improve, Yngvarr, SieBot, Pitoutom, Flyer22, Paolo.dL, Chem-
awb, Vcaeken, YSSYguy, Arjayay, Johnuniq, Rror, DOI bot, Fyrael, GyroMagician, Yobot, AnomieBOT, Valueyou, Materialscientist,
Nasa-verve, RibotBOT, Nedim Ardoğa, Aandroyd, Miyagawa, FrescoBot, Nageh, Michael93555, Throwaway85, DexDor, Ok2ptp, Ri-
ichrd, Hhhippo, ClueBot NG, Helpful Pixie Bot, Bibcode Bot, Htavroh, Ahmedmustafamahmoud, BattyBot, Mcginnsc, ChrisGualtieri,
Tagremover, Frosty, Atul hbk and Anonymous: 61
• Switched capacitor Source: http://en.wikipedia.org/wiki/Switched_capacitor?oldid=611735228 Contributors: HelgeStenstrom, Heron,
TedPavlic, Magister Mathematicae, Guerberj, Jaycarlson, SmackBot, Katanzag, Bluebot, Savidan, Jeff Wheeler, Reza mirhosseini,
9.1. TEXT 207

EdGl, Yves-Laurent, Biscay, Mikiemike, CmdrObot, Underpants, Thijs!bot, RobBrisbane, Mubed, Phosphoricx, User A1, RockMFR,
VolkovBot, Deki mg, Rob hagan, Treekids, Mreza v, Addbot, Willplatts, EconoPhysicist, Redheylin, Yobot, Theorih, Catwikionline,
EmausBot, Hanjifi, ClueBot NG, Snotbot, KLBot2, BrunoHaider and Anonymous: 37

• H bridge Source: http://en.wikipedia.org/wiki/H_bridge?oldid=625428917 Contributors: RTC, DavidCary, BenFrantzDale, Vadmium,

Sarex, Publunch, Roo72, Jcmaco, Fuxx, Sanctum, Nasukaren, Magetoo, Marudubshinki, Vegaswikian, Ademkader, FlaBot, Arnero,
DVdm, WriterHound, Hellbus, Speedevil, !jrb, SmackBot, Oli Filth, OrphanBot, CyrilB, Dicklyon, Wizard191, Brwilliams, Amalas,
Hans.vdb, Pro crast in a tor, CosineKitty, Donny3000, Americanhero, Matt B., Glrx, Smial, Leyo, TSullivan, Roboto1148, SoCal-
SuperEagle, Ikalogic, Noformation, Spinningspark, Biscuittin, Envergure, Cupzonia, Spuzzdawg, Treekids, ClueBot, Snigbrook, Pelesl,
DumZiBoT, XLinkBot, Ijayasin, Sisera, Addbot, Mortense, Semiwiki, Luckas-bot, Yobot, Fraggle81, Xqbot, GliderMaven, HRoestBot,
Steve2011, John of Reading, WikitanvirBot, Bamyers99, John Smith 104668, ‫خالقیان‬, ClueBot NG, Kubing, Clampower, Helpful Pixie
Bot, Wbm1058, BG19bot, JYBot, Vamfun, Tony Mach, Creation911, Jodosma, Spyglasses, Nchiazza, 8BitTRex and Anonymous: 82

• Hall effect sensor Source: http://en.wikipedia.org/wiki/Hall_effect_sensor?oldid=627806864 Contributors: Magnus Manske, Glenn,

Smack, Omegatron, BenFrantzDale, Geeoharee, Ot, Karol Langner, Sam Hocevar, Cacycle, Android79, Bobo192, Remuel, Enric Naval,
A-Day, Hooperbloob, Alansohn, Hohum, Evil Monkey, IMeowbot, Angr, Alvis, Sjakkalle, Mbutts, Anthony81212, Srleffler, Yurik-
Bot, RussBot, Sillybilly, Stephenb, Janke, Mikeblas, Attilios, SmackBot, Bluebot, Tsca.bot, DMacks, Mion, Es330td, Yves-Laurent,
Vanisaac, Thijs!bot, Adams13, JAnDbot, Cubaexplorer, Rewolff, Nyq, Nposs, DadaNeem, Inwind, Funandtrvl, TXiKiBoT, David Con-
drey, SieBot, Antzervos, ClueBot, Danpeirce, Mralik, ALLurGroceries, Pocketrocket24, Little Mountain 5, Addbot, Kongr43gpen, Gra-
hamatwp, OlEnglish, AnomieBOT, Materialscientist, Xqbot, GrouchoBot, Trurle, RibotBOT, Zhangzhe0101, Ksk2005, John of Reading,
Jnanadevm, Joe Gazz84, Josve05a, , Aeonx, Jwortzel, Sbmeirow, Djapa84, Bcaulf, RockMagnetist, ClueBot NG, JohnTsams, Muon,
Rezabot, Dsperlich, Helpful Pixie Bot, BG19bot, Alangar Manickam, CitationCleanerBot, Samhuntvid, Riyasmydheen, Magnetic models
and Anonymous: 70

• Low-pass filter Source: http://en.wikipedia.org/wiki/Low-pass_filter?oldid=623165965 Contributors: Mav, The Anome, Rjstott, Heron,
Patrick, JohnOwens, Lexor, David Martland, Glenn, Palfrey, Dysprosia, Furrykef, Omegatron, ThereIsNoSteve, Giftlite, DavidCary, Wolf-
keeper, Bensaccount, Vadmium, LucasVB, Antandrus, BrianWilloughby, Moxfyre, Rfl, Rich Farmbrough, TedPavlic, Mecanismo, ESkog,
Teorth, Foobaz, Cavrdg, Hangjian, Hooperbloob, Dragoljub, Wtshymanski, Cburnett, Flying fish, Davidkazuhiro, Pol098, Akavel, Pfal-
stad, Torquil, Mikm, Alfred Centauri, Kri, Krishnavedala, Borgx, PinothyJ, Toffile, Gaius Cornelius, Brandon, Mikeblas, Searchme, Light
current, Mickpc, Deville, Petri Krohn, LeonardoRob0t, Phil Holmes, RG2, Mejor Los Indios, EXonyte, KnightRider, Mitchan, Steve
carlson, Pgk, Niehaus, Nbarth, RoysonBobson, Zvar, Soundsop, IE, P.o.h, Elzair, Dog Eat Dog World, Minna Sora no Shita, Roger-
brent, Dicklyon, Kvng, Ss181292, Unmitigated Success, Myasuda, Paddles, Editor at Large, Epbr123, Sobreira, Bobblehead, Brichcja,
Majorly, Danroa, Lovibond, Ekkanant, JAnDbot, Xhienne, Drizzd, Time3000, Bongwarrior, VoABot II, Dics, Eus Kevin, Parijata, Kayau,
MartinBot, Renski, Tgeairn, RockMFR, Mange01, Slamedsilver, LLcopp, VolkovBot, Inductiveload, Ahmedsaieed, Spinningspark, Anoko
moonlight, Kbrose, Tetos, Tugjob, Dp67, ClueBot, Binksternet, Brews ohare, Thingg, 7, Dusen189, Johnuniq, XLinkBot, Mm40, ZooFari,
Addbot, Howard Landman, Jojhutton, Redheylin, Parvejkhan, Nocal, Tide rolls, Gail, Legobot, Bdb112, Floquenbeam, Jim1138, B137,
Materialscientist, Citation bot, Xqbot, Armstrong1113149, Pontificalibus, Christopherley, RibotBOT, Rb88guy, GliderMaven, ICEAGE,
Jonesey95, RedBot, Piandcompany, December21st2012Freak, The Utahraptor, Mgclapé, Astro89, WikitanvirBot, Immunize, Dewritech,
Catshome2000, Zueignung, Teapeat, Dweymouth, ClueBot NG, Satellizer, Widr, Varun varshney12, OceanEngineerRI, Kizzlebot, JYBot,
Kroq-gar78, CsDix, My name is not dave, Quenhitran, Meteor sandwich yum, Monkbot and Anonymous: 221

• High-pass filter Source: http://en.wikipedia.org/wiki/High-pass_filter?oldid=611538221 Contributors: The Anome, Rjstott, PierreAb-

bat, Lexor, Glenn, Charles Matthews, Omegatron, Eugene van der Pijll, ThereIsNoSteve, Tonsofpcs, BenFrantzDale, Vadmium, Rfl,
TedPavlic, Bobo192, Rbj, Foobaz, Hooperbloob, Dragoljub, RJFJR, Zawersh, Gene Nygaard, Bruce89, Thryduulf, Robert K S, Gre-
gorB, Waldir, Pfalstad, Rjwilmsi, Lockley, Arnero, PinothyJ, Alynna Kasmira, Mikeblas, Attilios, KnightRider, SmackBot, Vina-iwbot,
P.o.h, Elzair, Soumyasch, Dicklyon, Shaunwhite000, Shoez, Myasuda, Scoofy, Sobreira, Drizzd, AndyBloch, .anacondabot, Magioladitis,
Faizhaider, Baccyak4H, Katalaveno, Joerglwitsch, Ziounclesi, Spinningspark, Tresiden, Fibo1123581321, Bekuletz, ClueBot, Binksternet,
Estirabot, Gciriani, XLinkBot, Addbot, Jojhutton, Olli Niemitalo, Redheylin, Legobot, Yobot, Fraggle81, Amirobot, EnTerr, Gianno,
Citation bot, ShornAssociates, Armstrong1113149, RibotBOT, Mnmngb, Maitchy, Ll1324, JMS Old Al, Toriicelli, DARTH SIDIOUS 2,
EmausBot, Xiutwel-0003, ChuispastonBot, ClueBot NG, Rezabot, Helpful Pixie Bot, Gfoltz9, AvocatoBot, ChrisGualtieri, Txnhockey3,
CsDix, Monkbot and Anonymous: 79

• Band-pass filter Source: http://en.wikipedia.org/wiki/Band-pass_filter?oldid=625817556 Contributors: The Anome, Maury Markowitz,

Patrick, Angela, Nanobug, Glenn, Poor Yorick, BAxelrod, Emmjade, Guaka, Omegatron, ThereIsNoSteve, Giftlite, Sword, Zowie, Rfl,
CALR, Billlion, Shanes, Hooperbloob, SidP, Cburnett, OwenX, Cbdorsett, Pfalstad, Zbxgscqf, Sango123, Ianthegecko, Arnero, Antikon,
DVdm, Martin Hinks, YurikBot, Splash, PinothyJ, Toffile, Brandon, Hakeem.gadi, Deville, KNfLrPnKNsT, LeonardoRob0t, Poulpy,
Machtzu, RG2, Henrikb4, Binarypower, Commander Keane bot, Oli Filth, Vina-iwbot, Clicketyclack, Robofish, Mofomojo, Dicklyon,
Tawkerbot2, Nalvage, Sobreira, AlienBlancmange, CosineKitty, Email4mobile, RisingStick, STBot, Mange01, Acalamari, VolkovBot,
Cuddlyable3, Inductiveload, Spinningspark, Benjwgarner, Dp67, Binksternet, PipepBot, ChrisHodgesUK, Johnuniq, Addbot, Alexandra
Goncharik, Redheylin, OlEnglish, B137, Citation bot, GrouchoBot, Ebrambot, Lorem Ip, ClueBot NG, Rezabot, Helpful Pixie Bot, Fly-
guy53, Ankitd.elec, Chetan.meshram, Omegaoptical, Forestrf, CsDix, Ugog Nizdast, Joshua Mahesh Inayathullah and Anonymous: 63
208 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

9.2 Images
• File:1st_Order_Lowpass_Filter_RC.svg Source: http://upload.wikimedia.org/wikipedia/commons/e/e0/1st_Order_Lowpass_Filter_
RC.svg License: Public domain Contributors: Own work Original artist: Inductiveload
• File:555_Astable_Diagram.svg Source: http://upload.wikimedia.org/wikipedia/commons/3/3d/555_Astable_Diagram.svg License:
Public domain Contributors: Own drawing, made in Inkscape 0.43 Original artist: jjbeard
• File:555_Bistable.svg Source: http://upload.wikimedia.org/wikipedia/commons/7/78/555_Bistable.svg License: CC0 Contributors: Own
work Original artist: AbstractBeliefs
• File:555_Monostable.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/19/555_Monostable.svg License: Public domain
Contributors: Own work Original artist: Inductiveload
• File:555_Pinout.svg Source: http://upload.wikimedia.org/wikipedia/commons/c/c7/555_Pinout.svg License: Public domain Contributors:
Own work Original artist: Inductiveload
• File:75_Hz_HPF_on_Smaart.jpg Source: http://upload.wikimedia.org/wikipedia/commons/4/4d/75_Hz_HPF_on_Smaart.jpg License:
CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: Own work Original artist: Binksternet
• File:7812_voltage_regulator.jpg Source: http://upload.wikimedia.org/wikipedia/commons/c/ce/7812_voltage_regulator.jpg Li-
cense: CC-BY-3.0 Contributors: http://traveler.com.br/photos/matarese/photo/435-a-7812-voltage-regulator-chip/ Original artist:
Mataresephotos
• File:AFSK_1200_baud.ogg Source: http://upload.wikimedia.org/wikipedia/commons/2/27/AFSK_1200_baud.ogg License: CC-BY-
SA-3.0 Contributors: ? Original artist: ?
• File:Active_Highpass_Filter_RC.png Source: http://upload.wikimedia.org/wikipedia/commons/8/87/Active_Highpass_Filter_RC.png
License: Public domain Contributors: Own work by Toriicelli Original artist: Toriicelli
• File:Active_Lowpass_Filter_RC.svg Source: http://upload.wikimedia.org/wikipedia/commons/5/59/Active_Lowpass_Filter_RC.svg
License: Public domain Contributors: http://en.wikipedia.org/wiki/File:Active_Lowpass_Filter_RC.svg Original artist: Inductiveload
• File:Amplifier_Circuit_Small.svg Source: http://upload.wikimedia.org/wikipedia/commons/f/fd/Amplifier_Circuit_Small.svg Li-
cense: Public domain Contributors: This vector image was created with Inkscape. Original artist: Twisp
• File:Approximated_Ebers_Moll.svg Source: http://upload.wikimedia.org/wikipedia/commons/e/eb/Approximated_Ebers_Moll.svg
License: CC-BY-SA-3.0 Contributors: File:Ebers-Moll model schematic (NPN).svg Original artist: Original uploader was Krishnavedala
at en.wikipedia derivative work: Inductiveload (talk)
• File:BJT_NPN_symbol.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/66/BJT_NPN_symbol.svg License: CC-BY-
SA-3.0-2.5-2.0-1.0 Contributors: Created by User:Omegatron
Original artist: Omegatron

• File:BJT_NPN_symbol_(case).svg Source: http://upload.wikimedia.org/wikipedia/commons/c/cb/BJT_NPN_symbol_%28case%29.

svg License: Public domain Contributors: Own work Original artist: Zedh
• File:BJT_PNP_symbol.svg Source: http://upload.wikimedia.org/wikipedia/commons/9/9b/BJT_PNP_symbol.svg License: CC-BY-
SA-3.0-2.5-2.0-1.0 Contributors: Created by User:Omegatron
Original artist: Omegatron

• File:BJT_PNP_symbol_(case).svg Source: http://upload.wikimedia.org/wikipedia/commons/a/ab/BJT_PNP_symbol_%28case%29.

svg License: CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: Own work Original artist: Zedh
• File:BJT_h-parameters_(generalised).svg Source: http://upload.wikimedia.org/wikipedia/commons/9/9f/BJT_h-parameters_
%28generalised%29.svg License: CC-BY-SA-3.0 Contributors:
• H-parameters.gif Original artist: H-parameters.gif: Original uploader was Rohitbd at en.wikipedia
• File:BandDiagram-Semiconductors-E.PNG Source: http://upload.wikimedia.org/wikipedia/commons/8/85/
BandDiagram-Semiconductors-E.PNG License: CC-BY-SA-2.5 Contributors: ? Original artist: ?
• File:Bandpass_Filter.svg Source: http://upload.wikimedia.org/wikipedia/commons/b/bd/Bandpass_Filter.svg License: Public domain
Contributors: Own work Original artist: Inductiveload
• File:Bandstruktur_GaAs_en.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/67/Bandstruktur_GaAs_en.svg License:
CC-BY-SA-3.0 Contributors:
• Bandstruktur_GaAs.svg Original artist: Bandstruktur_GaAs.svg: Cepheiden.
• File:Bandwidth_2.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/6b/Bandwidth_2.svg License: Public domain Contrib-
utors: Own work Original artist: Inductiveload
9.2. IMAGES 209

• File:Bardeen_Shockley_Brattain_1948.JPG Source: http://upload.wikimedia.org/wikipedia/commons/c/c2/Bardeen_Shockley_

Brattain_1948.JPG License: Public domain Contributors: eBay item Original artist: AT&T; photographer: Jack St. (last part of name not
stamped well enough to read), New York, New York.
• File:Butterworth_response.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/60/Butterworth_response.svg License: CC-
BY-SA-3.0-2.5-2.0-1.0 Contributors:
• Butterworth_response.png Original artist:
• derivative work: Krishnavedala (talk)
• File:C555_Internal_Circuit.svg Source: http://upload.wikimedia.org/wikipedia/commons/2/29/C555_Internal_Circuit.svg License:
CC-BY-3.0 Contributors: Own work Original artist: Wdwd
• File:ClassG.svg Source: http://upload.wikimedia.org/wikipedia/commons/4/48/ClassG.svg License: CC-BY-SA-3.0 Contributors:
• ClassG.GIF Original artist:
--- Tiger66
--- derivative work:Henrydask
• File:ClassH.svg Source: http://upload.wikimedia.org/wikipedia/commons/f/f1/ClassH.svg License: CC-BY-SA-3.0 Contributors:
• ClassH.GIF Original artist: Tiger66
• File:Classe_E.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/11/Classe_E.svg License: CC-BY-SA-3.0 Contributors:
own work/travail personnel Original artist: Yves-Laurent Allaert
• File:Clutch_with_Hall_Effect_sensor.jpg Source: http://upload.wikimedia.org/wikipedia/commons/4/4f/Clutch_with_Hall_Effect_
sensor.jpg License: CC-BY-2.0 Contributors: Flickr Original artist: Ð¤Ð¸Ð³Ñ Ñ ÐºÐ¸
• File:Collector_to_base_bias.PNG Source: http://upload.wikimedia.org/wikipedia/commons/7/79/Collector_to_base_bias.PNG Li-
cense: Public domain Contributors: Transferred from en.wikipedia; transferred to Commons by User:Sfan00_IMG using CommonsHelper.
Original artist: Original uploader was Xcentaur at en.wikipedia
• File:Common_Hall_Sensor_Symbol.png Source: http://upload.wikimedia.org/wikipedia/commons/1/1e/Common_Hall_Sensor_
Symbol.png License: CC-BY-SA-3.0 Contributors: Created in gEDA and exported
Previously published: http://www.turbo-electric.com Original artist: Grahamatwp
• File:Commons-logo.svg Source: http://upload.wikimedia.org/wikipedia/en/4/4a/Commons-logo.svg License: ? Contributors: ? Original
artist: ?
• File:Comparators_stuuf.jpg Source: http://upload.wikimedia.org/wikipedia/commons/1/1d/Comparators_stuuf.jpg License: CC-BY-
SA-3.0 Contributors: Own work Original artist: Stuuf
• File:Cscr-featured.svg Source: http://upload.wikimedia.org/wikipedia/en/e/e7/Cscr-featured.svg License: ? Contributors: ? Original
artist: ?
• File:Current-Voltage_relationship_of_BJT.png Source: http://upload.wikimedia.org/wikipedia/commons/3/33/Current-Voltage_
relationship_of_BJT.png License: CC-BY-SA-3.0 Contributors: Own work Original artist: AlexHe34
• File:Cylinders_with_Hall_sensors.png Source: http://upload.wikimedia.org/wikipedia/commons/9/9c/Cylinders_with_Hall_sensors.
png License: CC-BY-SA-3.0 Contributors: Own work by the original uploader Original artist: User:IMeowbot
• File:Darlington_transistor_MJ1000.jpg Source: http://upload.wikimedia.org/wikipedia/commons/d/d9/Darlington_transistor_
MJ1000.jpg License: Attribution Contributors: Transferred from de.wikipedia; transferred to Commons by User:Wdwd using
CommonsHelper.
Original artist: thomy_pc.
• File:Dependent_Sources.PNG Source: http://upload.wikimedia.org/wikipedia/commons/6/6a/Dependent_Sources.PNG License: CC-
BY-SA-3.0-2.5-2.0-1.0 Contributors: Own work Original artist: Brews ohare
• File:Diamond_Cubic-F_lattice_animation.gif Source: http://upload.wikimedia.org/wikipedia/commons/2/22/Diamond_Cubic-F_
lattice_animation.gif License: CC0 Contributors: http://www.msm.cam.ac.uk/phase-trans/2003/MP1.crystals/MP1.crystals.html ; English
Wikipedia Original artist: original uploader: Brian0918
• File:Discrete_opamp.png Source: http://upload.wikimedia.org/wikipedia/commons/8/85/Discrete_opamp.png License: CC-BY-SA-3.0
Contributors: Op Amp Applications Handbook Original artist: Analog Devices
• File:Divide_4.png Source: http://upload.wikimedia.org/wikipedia/commons/6/64/Divide_4.png License: CC-BY-3.0 Contributors: Own
work Original artist: Jon Guerber
• File:Dynamic_Comparator.png Source: http://upload.wikimedia.org/wikipedia/commons/c/c1/Dynamic_Comparator.png License:
CC-BY-SA-3.0 Contributors: Transferred from en.wikipedia. Original artist: Guerberj at en.wikipedia
• File:ECC83_Glow.jpg Source: http://upload.wikimedia.org/wikipedia/commons/5/5a/ECC83_Glow.jpg License: CC0 Contributors:
Own work Original artist: Olli Niemitalo
210 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

• File:EMRI_LXCOS_Voltage_Regulator.jpg Source: http://upload.wikimedia.org/wikipedia/commons/5/53/EMRI_LXCOS_

Voltage_Regulator.jpg License: CC-BY-SA-3.0 Contributors: Own work Original artist: RoyPijnenburg
• File:Early_effect_(NPN).svg Source: http://upload.wikimedia.org/wikipedia/commons/4/49/Early_effect_%28NPN%29.svg License:
Public domain Contributors: Own work Original artist: Inductiveload
• File:Ebers-Moll_model_schematic_(NPN).svg Source: http://upload.wikimedia.org/wikipedia/commons/8/8a/Ebers-Moll_model_
schematic_%28NPN%29.svg License: CC-BY-SA-3.0 Contributors:
• Ebers-Moll_Model_NPN.PNG Original artist: Ebers-Moll_Model_NPN.PNG: Original uploader was Krishnavedala at en.wikipedia
• File:Ebers-Moll_model_schematic_(PNP).svg Source: http://upload.wikimedia.org/wikipedia/commons/e/ec/Ebers-Moll_model_
schematic_%28PNP%29.svg License: CC-BY-SA-3.0 Contributors:
• Ebers-Moll_Model_PNP.PNG Original artist: Ebers-Moll_Model_PNP.PNG: Original uploader was Krishnavedala at en.wikipedia
• File:Edit-clear.svg Source: http://upload.wikimedia.org/wikipedia/en/f/f2/Edit-clear.svg License: ? Contributors: ? Original artist: ?
• File:Electron-hole.svg Source: http://upload.wikimedia.org/wikipedia/commons/d/d6/Electron-hole.svg License: Public domain Con-
tributors: Own derivative work of wikipedia:Image:ElectronHole.JPG Original artist: User:Sir_Link
• File:Electronic_Amplifier_Class_A.png Source: http://upload.wikimedia.org/wikipedia/commons/9/9b/Electronic_Amplifier_Class_
A.png License: CC-BY-SA-3.0 Contributors: en:Image:Electronic_Amplifier_Class_A.png Original artist: GRAHAMUK with modifica-
tion of Yves-Laurent Allaert
• File:Electronic_Amplifier_Class_B_fixed.png Source: http://upload.wikimedia.org/wikipedia/commons/b/b6/Electronic_Amplifier_
Class_B_fixed.png License: CC-BY-SA-3.0 Contributors: Transferred from en.wikipedia; transferred to Commons by User:Logan using
CommonsHelper. Original artist: Nitram cero at en.wikipedia
• File:Electronic_Amplifier_Class_C.png Source: http://upload.wikimedia.org/wikipedia/commons/e/e1/Electronic_Amplifier_Class_
C.png License: CC-BY-SA-3.0 Contributors: en:Image:Electronic_Amplifier_Class_C.png Original artist: GRAHAMUK with modifica-
tion of Yves-Laurent Allaert
• File:Electronic_Amplifier_Push-pull.svg Source: http://upload.wikimedia.org/wikipedia/commons/b/b4/Electronic_Amplifier_
Push-pull.svg License: CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: File:Electronic Amplifier Push-pull.png Original artist: Lakkasuo
• File:Emitter_bias.PNG Source: http://upload.wikimedia.org/wikipedia/commons/3/39/Emitter_bias.PNG License: Public domain Con-
tributors: Transferred from en.wikipedia; transferred to Commons by User:Sfan00_IMG using CommonsHelper. Original artist: Original
uploader was Xcentaur at en.wikipedia
• File:FET_comparison.png Source: http://upload.wikimedia.org/wikipedia/commons/c/c0/FET_comparison.png License: Public do-
main Contributors: ? Original artist: ?
• File:Ferrosilicon.JPG Source: http://upload.wikimedia.org/wikipedia/commons/5/51/Ferrosilicon.JPG License: CC-BY-SA-3.0-2.5-
2.0-1.0 Contributors: Own work Original artist: FocalPoint
• File:Fixed_bias.PNG Source: http://upload.wikimedia.org/wikipedia/commons/7/79/Fixed_bias.PNG License: Public domain Contrib-
utors: Transferred from en.wikipedia; transferred to Commons by User:Sfan00_IMG using CommonsHelper. Original artist: Original
uploader was Xcentaur at en.wikipedia
• File:Fixed_bias_with_emitter_resistor.PNG Source: http://upload.wikimedia.org/wikipedia/commons/b/b2/Fixed_bias_with_
emitter_resistor.PNG License: Public domain Contributors: Transferred from en.wikipedia; transferred to Commons by User:Sfan00_IMG
using CommonsHelper. Original artist: Original uploader was Xcentaur at en.wikipedia
• File:Folder_Hexagonal_Icon.svg Source: http://upload.wikimedia.org/wikipedia/en/4/48/Folder_Hexagonal_Icon.svg License: ? Con-
tributors: ? Original artist: ?
• File:Fotodio.jpg Source: http://upload.wikimedia.org/wikipedia/commons/e/e6/Fotodio.jpg License: CC-BY-SA-3.0 Contributors: ?
Original artist: ?
• File:Fsk.svg Source: http://upload.wikimedia.org/wikipedia/commons/3/39/Fsk.svg License: CC-BY-SA-3.0 Contributors: ? Original
artist: ?
• File:GHS-pictogram-pollu.svg Source: http://upload.wikimedia.org/wikipedia/commons/b/b9/GHS-pictogram-pollu.svg License: ?
Contributors: EPS file pollu.eps from UNECE web site converted with ImageMagick convert and with potrace, edited in inkscape Original
artist: Torsten Henning
• File:GHS-pictogram-skull.svg Source: http://upload.wikimedia.org/wikipedia/commons/5/58/GHS-pictogram-skull.svg License: ?
Contributors: EPS file skull.eps from UNECE web site converted with ImageMagick convert and with potrace, edited in inkscape Orig-
inal artist: Torsten Henning
• File:Gallium-arsenide-unit-cell-3D-balls.png Source: http://upload.wikimedia.org/wikipedia/commons/2/25/
Gallium-arsenide-unit-cell-3D-balls.png License: Public domain Contributors: ? Original artist: ?
• File:Gallium_Arsenide_(GaAs)_2extquotedbl{}_wafer.jpg Source: http://upload.wikimedia.org/wikipedia/commons/2/22/Gallium_
Arsenide_%28GaAs%29_2%22_wafer.jpg License: CC-BY-SA-3.0 Contributors: Own work Original artist: La2O3
9.2. IMAGES 211

• File:Gallium_arsenide_crystal.jpg Source: http://upload.wikimedia.org/wikipedia/commons/6/6f/Gallium_arsenide_crystal.jpg Li-

cense: CC-BY-SA-3.0 Contributors: http://woelen.homescience.net/science/index.html Original artist: W. Oelen
• File:General_Microwave_VCO.png Source: http://upload.wikimedia.org/wikipedia/commons/5/5b/General_Microwave_VCO.png Li-
cense: CC-BY-SA-3.0 Contributors: http://www.herley.com/index.cfm?act=product&prd=481 Original artist: en:Herley Industries
• File:Generic_741_pinout_top.png Source: http://upload.wikimedia.org/wikipedia/commons/7/72/Generic_741_pinout_top.png Li-
cense: CC-BY-SA-3.0 Contributors: Own work by uploader (derived from other own (Creative Commons) work from uploader at
http://www.tedpavlic.com/teaching/osu/ece209/support/part_pinouts.pdf) Original artist: TedPavlic
• File:Germane-2D-dimensions.png Source: http://upload.wikimedia.org/wikipedia/commons/c/ca/Germane-2D-dimensions.png Li-
cense: Public domain Contributors: Own work Original artist: Ben Mills
• File:Gnome-mime-sound-openclipart.svg Source: http://upload.wikimedia.org/wikipedia/commons/8/87/
Gnome-mime-sound-openclipart.svg License: Public domain Contributors: Own work. Based on File:Gnome-mime-audio-openclipart.
svg, which is public domain. Original artist: User:Eubulides
• File:H_bridge.svg Source: http://upload.wikimedia.org/wikipedia/commons/d/d4/H_bridge.svg License: CC-BY-SA-3.0 Contributors:
Own work Original artist: Cyril BUTTAY
• File:H_bridge_operating.svg Source: http://upload.wikimedia.org/wikipedia/commons/f/f2/H_bridge_operating.svg License: CC-BY-
SA-3.0 Contributors: own work, made using inkscape Original artist: Cyril BUTTAY
• File:Hall_sensor_tach.gif Source: http://upload.wikimedia.org/wikipedia/commons/7/7e/Hall_sensor_tach.gif License: CC-BY-SA-3.0
Contributors: ? Original artist: ?
• File:Hazard_N.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/6a/Hazard_N.svg License: Public domain Contributors:
Converted from EPS file at http://forum.cptec.org/index.php?showtopic=368 Original artist: See historic
• File:Hazard_T.svg Source: http://upload.wikimedia.org/wikipedia/commons/3/39/Hazard_T.svg License: Public domain Contributors: ?
Original artist: ?
• File:High_Pass_Filter_Example.jpg Source: http://upload.wikimedia.org/wikipedia/commons/c/ca/High_Pass_Filter_Example.jpg
License: CC0 Contributors: Own work Original artist: Ll1324
• File:High_pass_filter.svg Source: http://upload.wikimedia.org/wikipedia/commons/f/fe/High_pass_filter.svg License: Public domain
Contributors: ? Original artist: ?
• File:Hybrid-pi_detailed_model.svg Source: http://upload.wikimedia.org/wikipedia/en/2/2f/Hybrid-pi_detailed_model.svg License: ?
Contributors: ? Original artist: ?
• File:Hybrid_opamp.png Source: http://upload.wikimedia.org/wikipedia/commons/8/84/Hybrid_opamp.png License: CC-BY-SA-3.0
Contributors: Op Amp Applications Handbook Original artist: Analog Devices
• File:IGFET_N-Ch_Dep_Labelled.svg Source: http://upload.wikimedia.org/wikipedia/commons/e/e8/IGFET_N-Ch_Dep_Labelled.
svg License: Public domain Contributors: From Scratch in Inkcape 0.43 Original artist: jjbeard
• File:IGFET_N-Ch_Enh_Labelled.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/62/IGFET_N-Ch_Enh_Labelled.
svg License: Public domain Contributors: ? Original artist: ?
• File:IGFET_N-Ch_Enh_Labelled_simplified.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/61/IGFET_N-Ch_
Enh_Labelled_simplified.svg License: CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: Based on Image:IGFET N-Ch Enh Labelled.svg, Created
by User:Omegatron
Original artist: User:Omegatron

• File:IGFET_P-Ch_Dep_Labelled.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/1b/IGFET_P-Ch_Dep_Labelled.

svg License: Public domain Contributors: From Scratch in Inkcape 0.43 Original artist: jjbeard
• File:IGFET_P-Ch_Enh_Labelled.svg Source: http://upload.wikimedia.org/wikipedia/commons/0/0c/IGFET_P-Ch_Enh_Labelled.
svg License: Public domain Contributors: From Scratch in Inkcape 0.43 Original artist: jjbeard
• File:IGFET_P-Ch_Enh_Labelled_simplified.svg Source: http://upload.wikimedia.org/wikipedia/commons/c/c4/IGFET_P-Ch_Enh_
Labelled_simplified.svg License: CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: Based on Image:IGFET P-Ch Enh Labelled.svg, Created by
User:Omegatron
Original artist: User:Omegatron

• File:Inside_of_a_Boss_Audio_DD3600_Class_D_mono_block_amp.jpg Source: http://upload.wikimedia.org/wikipedia/commons/

2/26/Inside_of_a_Boss_Audio_DD3600_Class_D_mono_block_amp.jpg License: CC-BY-3.0 Contributors: Transferred from
en.wikipedia; transferred to Commons by User:Logan using CommonsHelper.
Original artist: Daniel Christensen (talk). Original uploader was Daniel Christensen at en.wikipedia
• File:Inverting_Amplifier_Signal_Clipping.png Source: http://upload.wikimedia.org/wikipedia/commons/9/9f/Inverting_Amplifier_
Signal_Clipping.png License: CC0 Contributors: Own work Original artist: Super Rad!
212 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

• File:JFET_N-Channel_Labelled.svg Source: http://upload.wikimedia.org/wikipedia/commons/4/46/JFET_N-Channel_Labelled.svg

License: Public domain Contributors: From Scratch in Inkcape 0.43 Original artist: jjbeard
• File:JFET_N-dep_symbol.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/6d/JFET_N-dep_symbol.svg License: CC-
BY-SA-3.0-2.5-2.0-1.0 Contributors: Own work Original artist: Zedh
• File:JFET_P-Channel_Labelled.svg Source: http://upload.wikimedia.org/wikipedia/commons/0/09/JFET_P-Channel_Labelled.svg
License: Public domain Contributors: From Scratch in Inkcape 0.43 Original artist: jjbeard
• File:JFET_n-channel_en.svg Source: http://upload.wikimedia.org/wikipedia/commons/5/50/JFET_n-channel_en.svg License: CC-BY-
SA-3.0-2.5-2.0-1.0 Contributors:
• JFET_n-channel.svg Original artist: JFET_n-channel.svg: Phirosiberia
• File:K2-w_Vacuum_Tube_Op-amp.jpg Source: http://upload.wikimedia.org/wikipedia/commons/1/15/K2-w_Vacuum_Tube_
Op-amp.jpg License: CC-BY-SA-3.0 Contributors: Own work Original artist: Bdieseldorff
• File:Konstanze.jpg Source: http://upload.wikimedia.org/wikipedia/commons/f/f0/Konstanze.jpg License: Public domain Contributors:
Own work Original artist: Ulfbastel
• File:L298_IMGP4533_wp.jpg Source: http://upload.wikimedia.org/wikipedia/commons/2/23/L298_IMGP4533_wp.jpg License: FAL
Contributors: Own work Original artist: Smial
• File:LM741CN.jpg Source: http://upload.wikimedia.org/wikipedia/commons/9/95/LM741CN.jpg License: CC0 Contributors: Own
work Original artist: Olli Niemitalo
• File:Lateral_mosfet.svg Source: http://upload.wikimedia.org/wikipedia/commons/7/79/Lateral_mosfet.svg License: CC-BY-SA-3.0
Contributors: ? Original artist: ?
• File:MDAC.png Source: http://upload.wikimedia.org/wikipedia/en/3/3c/MDAC.png License: ? Contributors: ? Original artist: ?
• File:Mergefrom.svg Source: http://upload.wikimedia.org/wikipedia/commons/0/0f/Mergefrom.svg License: Public domain Contributors:
? Original artist: ?
• File:MidSTAR-1.jpg Source: http://upload.wikimedia.org/wikipedia/commons/a/a0/MidSTAR-1.jpg License: Public domain Contribu-
tors:
• MidSTAR.jpg Original artist: MidSTAR.jpg: United States Naval Academy
• File:Modular_opamp.png Source: http://upload.wikimedia.org/wikipedia/commons/9/97/Modular_opamp.png License: CC-BY-SA-
3.0 Contributors: Op Amp Applications Handbook Original artist: Analog Devices
• File:Monokristalines_Silizium_für_die_Waferherstellung.jpg Source: http://upload.wikimedia.org/wikipedia/commons/2/23/
Monokristalines_Silizium_f%C3%BCr_die_Waferherstellung.jpg License: CC-BY-SA-3.0 Contributors: ? Original artist: ?
• File:Moving_Coil_Voltage_Regulator.png Source: http://upload.wikimedia.org/wikipedia/en/7/79/Moving_Coil_Voltage_Regulator.
png License: ? Contributors: ? Original artist: ?
• File:NE555_Bloc_Diagram.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/17/NE555_Bloc_Diagram.svg License:
Public domain Contributors: Own work Original artist: BlanchardJ
• File:NE555_Internal_Circuit.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/1c/NE555_Internal_Circuit.svg License:
CC-BY-3.0 Contributors: Own work Original artist: Wdwd
• File:NE555_Monotable_Waveforms_(English).png Source: http://upload.wikimedia.org/wikipedia/commons/d/dc/NE555_
Monotable_Waveforms_%28English%29.png License: Public domain Contributors:
• NE555_Monostable_Waveforms_fr.png Original artist: NE555_Monostable_Waveforms_fr.png: BlanchardJ
• File:NPN_BJT_(Planar)_Cross-section.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/6b/NPN_BJT_%28Planar%
29_Cross-section.svg License: Public domain Contributors: Own work Original artist: Inductiveload
• File:NPN_BJT_-_Structure_&_circuit.svg Source: http://upload.wikimedia.org/wikipedia/commons/4/49/NPN_BJT_-_Structure_
%26_circuit.svg License: CC-BY-SA-3.0 Contributors:
• Based on File:Npn-structure.png, by User:Heron at the English Wikipedia Original artist: Inductiveload
• File:NPN_BJT_Basic_Operation_(Active).svg Source: http://upload.wikimedia.org/wikipedia/commons/1/13/NPN_BJT_Basic_
Operation_%28Active%29.svg License: Public domain Contributors: Own drawing, done in Inkscape Original artist: Inductiveload
• File:NPN_Band_Diagram_Active.svg Source: http://upload.wikimedia.org/wikipedia/commons/4/4b/NPN_Band_Diagram_Active.
svg License: Public domain Contributors: Own drawing, done in Inkscape Original artist: Inductiveload
• File:NPN_Band_Diagram_Equilibrium.svg Source: http://upload.wikimedia.org/wikipedia/commons/a/a3/NPN_Band_Diagram_
Equilibrium.svg License: Public domain Contributors: Own drawing, done in Inkscape Original artist: Inductiveload
• File:NPN_common_emitter_AC.svg Source: http://upload.wikimedia.org/wikipedia/commons/8/8c/NPN_common_emitter_AC.svg
License: CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: Own work Original artist: Zedh
9.2. IMAGES 213

• File:NTC_bead.jpg Source: http://upload.wikimedia.org/wikipedia/commons/3/3b/NTC_bead.jpg License: CC-BY-SA-2.0-de Contrib-

utors: photo taken with Canon PowerShot G3 Original artist: Ansgar Hellwig
• File:Npn_heterostructure_bands.svg Source: http://upload.wikimedia.org/wikipedia/commons/b/b1/Npn_heterostructure_bands.svg
License: CC-BY-SA-3.0 Contributors: Own work Original artist: Matthewbeckler
• File:NucleophilicAdditionWithOrganogermanium.png Source: http://upload.wikimedia.org/wikipedia/commons/e/e3/
NucleophilicAdditionWithOrganogermanium.png License: CC-BY-SA-3.0 Contributors: Transferred from en.wikipedia; trans-
ferred to Commons by User:T.vanschaik using CommonsHelper. Original artist: Original uploader was V8rik at en.wikipedia. Later
version(s) were uploaded by Mahahahaneapneap at en.wikipedia.
• File:Nuvola_apps_edu_science.svg Source: http://upload.wikimedia.org/wikipedia/commons/5/59/Nuvola_apps_edu_science.svg Li-
cense: LGPL Contributors: http://ftp.gnome.org/pub/GNOME/sources/gnome-themes-extras/0.9/gnome-themes-extras-0.9.0.tar.gz
Original artist: David Vignoni / ICON KING
• File:Nuvola_apps_ksim.png Source: http://upload.wikimedia.org/wikipedia/commons/8/8d/Nuvola_apps_ksim.png License: LGPL
Contributors: http://icon-king.com Original artist: David Vignoni / ICON KING
• File:Office-book.svg Source: http://upload.wikimedia.org/wikipedia/commons/a/a8/Office-book.svg License: Public domain Contribu-
tors: This and myself. Original artist: Chris Down/Tango project
• File:Op-Amp_Comparator.svg Source: http://upload.wikimedia.org/wikipedia/commons/0/0d/Op-Amp_Comparator.svg License:
Public domain Contributors: Own work Original artist: Inductiveload
• File:Op-Amp_Internal.svg Source: http://upload.wikimedia.org/wikipedia/commons/0/0d/Op-Amp_Internal.svg License: Public do-
main Contributors: Own work Original artist: Inductiveload
• File:Op-Amp_Inverting_Amplifier.svg Source: http://upload.wikimedia.org/wikipedia/commons/4/41/Op-Amp_Inverting_
Amplifier.svg License: Public domain Contributors: Own work Original artist: Inductiveload
• File:Op-Amp_Non-Inverting_Amplifier.svg Source: http://upload.wikimedia.org/wikipedia/commons/4/44/Op-Amp_
Non-Inverting_Amplifier.svg License: Public domain Contributors: Own work Original artist: Inductiveload
• File:Op-amp_open-loop_1.svg Source: http://upload.wikimedia.org/wikipedia/commons/8/8e/Op-amp_open-loop_1.svg License: CC-
BY-SA-3.0 Contributors: my drawing Original artist: Ong saluri
• File:Op-amp_symbol.svg Source: http://upload.wikimedia.org/wikipedia/commons/9/97/Op-amp_symbol.svg License: CC-BY-SA-
3.0-2.5-2.0-1.0 Contributors: Own work Original artist: User:Omegatron
• File:OpAmpTransistorLevel_Colored_Labeled.svg Source: http://upload.wikimedia.org/wikipedia/commons/e/e0/
OpAmpTransistorLevel_Colored_Labeled.svg License: CC-BY-2.5 Contributors: redrawn png file (from User:Omegatron), datasheet
Original artist: Daniel Braun
• File:Operational_amplifier_noninverting.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/66/Operational_amplifier_
noninverting.svg License: CC-BY-SA-3.0 Contributors: my drawing Original artist: Ong saluri
• File:P45N02LD.jpg Source: http://upload.wikimedia.org/wikipedia/commons/a/af/P45N02LD.jpg License: CC-BY-SA-3.0 Contribu-
tors: ? Original artist: ?
• File:PD-icon.svg Source: http://upload.wikimedia.org/wikipedia/en/6/62/PD-icon.svg License: ? Contributors: ? Original artist: ?
• File:PLL,usage.png Source: http://upload.wikimedia.org/wikipedia/commons/0/0f/PLL%2Cusage.png License: CC-BY-SA-3.0 Con-
tributors: Transferred from en.wikipedia; transfer was stated to be made by User:IngFrancesco. Original artist: Original uploader was
Iain.mcclatchie at en.wikipedia Later version(s) were uploaded by Vahid avr at en.wikipedia.
• File:PLL_generic_inline_optional_N.png Source: http://upload.wikimedia.org/wikipedia/commons/6/60/PLL_generic_inline_
optional_N.png License: CC-BY-SA-4.0 Contributors: Own work Original artist: Crystallizedcarbon, Iain.mcclatchie, Mysid, Yegorius
• File:PNP_BJT_-_Structure_&_circuit.svg Source: http://upload.wikimedia.org/wikipedia/commons/9/90/PNP_BJT_-_Structure_
%26_circuit.svg License: CC-BY-SA-3.0 Contributors:
• Based on File:Pnp-structure.png, by User:Heron at the English Wikipedia Original artist: Inductiveload
• File:PN_Junction_in_Reverse_Bias.png Source: http://upload.wikimedia.org/wikipedia/commons/0/01/PN_Junction_in_Reverse_
Bias.png License: GFDL 1.2 Contributors: ? Original artist: ?
• File:PN_band.gif Source: http://upload.wikimedia.org/wikipedia/commons/2/22/PN_band.gif License: CC-BY-3.0 Contributors:
GIF:http://nanohub.org/resources/8797/ , Tool link: https://nanohub.org/tools/pntoy/ Original artist: Saumitra R Mehrotra & Gerhard
Klimeck
• File:PN_diode_with_electrical_symbol.svg Source: http://upload.wikimedia.org/wikipedia/commons/7/79/PN_diode_with_
electrical_symbol.svg License: CC-BY-SA-3.0 Contributors: Own work Original artist: Raffamaiden
• File:Padlock-silver.svg Source: http://upload.wikimedia.org/wikipedia/commons/f/fc/Padlock-silver.svg License: ? Contributors: http:
//openclipart.org/people/Anonymous/padlock_aj_ashton_01.svg Original artist: This image file was created by AJ Ashton. Uploaded from
English WP by User:Eleassar. Converted by User:AzaToth to a silver color.
214 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

• File:Parasitic_Sensative_Inverter.png Source: http://upload.wikimedia.org/wikipedia/en/7/76/Parasitic_Sensative_Inverter.png Li-

cense: ? Contributors: ? Original artist: ?
• File:Pdms.png Source: http://upload.wikimedia.org/wikipedia/commons/6/6a/Pdms.png License: CC-BY-SA-3.0 Contributors: selfmade
with ChemDraw and Photoshop Original artist: Jesse
• File:Pet_Flasche.JPG Source: http://upload.wikimedia.org/wikipedia/commons/e/ea/Pet_Flasche.JPG License: CC-BY-SA-3.0-2.5-
2.0-1.0 Contributors: ? Original artist: ?
• File:Philco_Surface_Barrier_transistor=1953.jpg Source: http://upload.wikimedia.org/wikipedia/commons/e/e8/Philco_Surface_
Barrier_transistor%3D1953.jpg License: CC-BY-SA-3.0 Contributors: Own work Original artist: Historianbuff
• File:Photodiode_array_chip.jpg Source: http://upload.wikimedia.org/wikipedia/commons/c/ce/Photodiode_array_chip.jpg License:
CC-BY-SA-3.0 Contributors: Own work Original artist: Fulvio314
• File:Photodiode_operation.png Source: http://upload.wikimedia.org/wikipedia/commons/0/06/Photodiode_operation.png License:
CC-BY-SA-3.0 Contributors:
• Kennlinie_Photodiode_1.png Original artist: Kennlinie_Photodiode_1.png: Gregor Hess (Ghe42)
• File:Photodiode_symbol.svg Source: http://upload.wikimedia.org/wikipedia/commons/2/2b/Photodiode_symbol.svg License: CC-BY-
SA-3.0-2.5-2.0-1.0 Contributors: ? Original artist: ?
• File:Photoresistor.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/13/Photoresistor.svg License: Public domain Contrib-
utors: Own work Original artist: Горовато
• File:PhototransistorSymbol.png Source: http://upload.wikimedia.org/wikipedia/commons/5/57/PhototransistorSymbol.png License:
CC-BY-3.0 Contributors: WikiProject Wikipedia Original artist: myself
• File:Pink.noise.png Source: http://upload.wikimedia.org/wikipedia/commons/7/72/Pink.noise.png License: Public domain Contributors:
Transferred from de.wikipedia; transferred to Commons by User:Leyo using CommonsHelper. Original artist: Original uploader was
Bautsch at de.wikipedia
• File:Pn-junction-equilibrium-graphs.png Source: http://upload.wikimedia.org/wikipedia/commons/f/fa/
Pn-junction-equilibrium-graphs.png License: CC-BY-SA-3.0 Contributors: en:Image:Pn-junction-equilibrium-graphs.png Origi-
nal artist: en:User:TheNoise
• File:Pn-junction-equilibrium.png Source: http://upload.wikimedia.org/wikipedia/commons/d/d6/Pn-junction-equilibrium.png Li-
cense: CC-BY-SA-3.0 Contributors: Transferred from en.wikipedia; Transfer was stated to be made by User:Mu301. Original artist:
Original uploader was TheNoise at en.wikipedia
• File:Polycrystalline_silicon_rod.jpg Source: http://upload.wikimedia.org/wikipedia/commons/5/59/Polycrystalline_silicon_rod.jpg
License: CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: Own work Original artist: Warut Roonguthai
• File:Pwm_amp.svg Source: http://upload.wikimedia.org/wikipedia/commons/4/4c/Pwm_amp.svg License: CC-BY-SA-3.0 Contributors:
English Wikipedia Original artist: Rohitbd
• File:Quartz,_Tibet.jpg Source: http://upload.wikimedia.org/wikipedia/commons/1/14/Quartz%2C_Tibet.jpg License: CC-BY-SA-2.5
Contributors: Own work Original artist: JJ Harrison (jjharrison89@facebook.com)
• File:Question_book-new.svg Source: http://upload.wikimedia.org/wikipedia/en/9/99/Question_book-new.svg License: ? Contributors:
? Original artist: ?
• File:RC_Divider.svg Source: http://upload.wikimedia.org/wikipedia/commons/3/3b/RC_Divider.svg License: Public domain Contribu-
tors: SVG work by uploader Original artist: ZooFari
• File:Radiolaria3434.JPG Source: http://upload.wikimedia.org/wikipedia/commons/c/c1/Radiolaria3434.JPG License: CC-BY-SA-3.0
Contributors: ? Original artist: ?
• File:Relationship_between_dBu_and_dBm.png Source: http://upload.wikimedia.org/wikipedia/commons/c/c7/Relationship_
between_dBu_and_dBm.png License: CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: ? Original artist: ?
• File:Renierit.JPG Source: http://upload.wikimedia.org/wikipedia/commons/e/ea/Renierit.JPG License: CC-BY-SA-2.0-de Contrib-
utors: Own work Original artist: Alchemist-hp (<a href='//commons.wikimedia.org/wiki/User_talk:Alchemist-hp' title='User talk:
Alchemist-hp'>talk</a>) (www.pse-mendelejew.de)
• File:Replica-of-first-transistor.jpg Source: http://upload.wikimedia.org/wikipedia/commons/b/bf/Replica-of-first-transistor.jpg Li-
cense: Public domain Contributors: ? Original artist: ?
• File:Response_silicon_photodiode.svg Source: http://upload.wikimedia.org/wikipedia/commons/4/41/Response_silicon_photodiode.
svg License: CC-BY-SA-3.0 Contributors: Own work Original artist: KaiMartin
• File:SI_base_unit.svg Source: http://upload.wikimedia.org/wikipedia/commons/c/c8/SI_base_unit.svg License: CC-BY-SA-3.0 Contrib-
utors: I (Dono (talk)) created this work entirely by myself. Base on http://www.newscientist.com/data/images/archive/2622/26221501.jpg
Original artist: Dono (talk)
• File:SMD_Cystal_Oscillator_TCXO.png Source: http://upload.wikimedia.org/wikipedia/commons/2/2c/SMD_Cystal_Oscillator_
TCXO.png License: CC-BY-3.0 Contributors: Own work Original artist: Appaloosa
9.2. IMAGES 215

• File:STM-NE556-HD.jpg Source: http://upload.wikimedia.org/wikipedia/commons/c/c3/STM-NE556-HD.jpg License: CC-BY-3.0

Contributors: http://zeptobars.ru/en/read/open-microchip-asic-what-inside-II-msp430-pic-z80 Original artist: ZeptoBars
• File:Schematic_of_class_H_current_amplifier.gif Source: http://upload.wikimedia.org/wikipedia/commons/a/aa/Schematic_of_
class_H_current_amplifier.gif License: CC-BY-SA-3.0 Contributors: Own work Original artist: Braun walter
• File:Signetics_NE555N.JPG Source: http://upload.wikimedia.org/wikipedia/commons/2/21/Signetics_NE555N.JPG License: CC-BY-
SA-3.0 Contributors: Own work Original artist: de:User:Stefan506
• File:Silicon-unit-cell-3D-balls.png Source: http://upload.wikimedia.org/wikipedia/commons/f/f1/Silicon-unit-cell-3D-balls.png Li-
cense: Public domain Contributors: Own work Original artist: Ben Mills
• File:Silicon_wafer_with_mirror_finish.jpg Source: http://upload.wikimedia.org/wikipedia/commons/5/5d/Silicon_wafer_with_
mirror_finish.jpg License: Public domain Contributors: NASA Glenn Research Center Original artist: NASA Glenn Research Center
• File:Silizium_pulver.jpg Source: http://upload.wikimedia.org/wikipedia/commons/4/4a/Silizium_pulver.jpg License: CC-BY-SA-3.0
Contributors: eigenes Bild; de:Bild:Silizium pulver.jpg Original artist: Silane
• File:Simple_electromechanical_regulation.PNG Source: http://upload.wikimedia.org/wikipedia/en/c/ca/Simple_electromechanical_
regulation.PNG License: ? Contributors: ? Original artist: ?
• File:Simple_electromechanical_voltage_regulator.PNG Source: http://upload.wikimedia.org/wikipedia/en/3/38/Simple_
electromechanical_voltage_regulator.PNG License: ? Contributors: ? Original artist: ?
• File:Sinc_function_(normalized).svg Source: http://upload.wikimedia.org/wikipedia/commons/d/d4/Sinc_function_%28normalized%
29.svg License: CC-BY-SA-3.0-2.5-2.0-1.0 Contributors: ? Original artist: ?
• File:Singlemode_fibre_structure.svg Source: http://upload.wikimedia.org/wikipedia/commons/8/84/Singlemode_fibre_structure.svg
License: CC-BY-SA-3.0 Contributors: derivative work by uploader Original artist: Original by Bob Mellish, SVG derivative by Benchill
• File:Stabilizer.JPG Source: http://upload.wikimedia.org/wikipedia/commons/d/d5/Stabilizer.JPG License: Public domain Contributors:
Originally uploaded by copyright holder Sriramk750 at English Wikipedia to http://en.wikipedia.org/wiki/File:Stabilizer.JPG Original
artist: Sriramk750 at English Wikipedia
• File:Streetlight_control.jpg Source: http://upload.wikimedia.org/wikipedia/commons/c/c8/Streetlight_control.jpg License: CC-BY-
SA-3.0 Contributors: Own work Original artist: MyName (Atlant)
• File:Switching_capacitor_schematic.PNG Source: http://upload.wikimedia.org/wikipedia/commons/e/e3/Switching_capacitor_
schematic.PNG License: Public domain Contributors: en.wikipedia Original artist: en:Reza mirhosseini
• File:Symbol_book_class2.svg Source: http://upload.wikimedia.org/wikipedia/commons/8/89/Symbol_book_class2.svg License: CC-
BY-SA-2.5 Contributors: Mad by Lokal_Profil by combining: Original artist: Lokal_Profil
• File:Text_document_with_red_question_mark.svg Source: http://upload.wikimedia.org/wikipedia/commons/a/a4/Text_document_
with_red_question_mark.svg License: Public domain Contributors: Created by bdesham with Inkscape; based upon Text-x-generic.svg
from the Tango project. Original artist: Benjamin D. Esham (bdesham)
• File:Thermistor.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/10/Thermistor.svg License: Public domain Contributors:
Own drawing, made in Inkscape 0.43 Original artist: jjbeard
• File:Transbauformen.jpg Source: http://upload.wikimedia.org/wikipedia/commons/e/e1/Transbauformen.jpg License: CC-BY-SA-3.0
Contributors: Own work Original artist: Ulfbastel
• File:Transistor-die-KSY34.jpg Source: http://upload.wikimedia.org/wikipedia/commons/e/e2/Transistor-die-KSY34.jpg License:
Public domain Contributors: ? Original artist: ?
• File:Transistor_Simple_Circuit_Diagram_with_NPN_Labels.svg Source: http://upload.wikimedia.org/wikipedia/commons/9/91/
Transistor_Simple_Circuit_Diagram_with_NPN_Labels.svg License: CC-BY-SA-3.0 Contributors: I created a postscript file, and con-
verted it to SVG using the pstoedit program. Original artist: Michael9422
• File:Transistor_as_switch.svg Source: http://upload.wikimedia.org/wikipedia/commons/5/5d/Transistor_as_switch.svg License: Public
domain Contributors: Own work Original artist: FDominec
• File:Transistor_on_portuguese_pavement.jpg Source: http://upload.wikimedia.org/wikipedia/commons/3/38/Transistor_on_
portuguese_pavement.jpg License: CC-BY-SA-3.0 Contributors: Own work Original artist: Joao.pimentel.ferreira
• File:Transistorer_(croped).jpg Source: http://upload.wikimedia.org/wikipedia/commons/2/2c/Transistorer_%28croped%29.jpg Li-
cense: CC-BY-SA-3.0 Contributors: Own work Original artist: Transisto at en.wikipedia
• File:Transparent.gif Source: http://upload.wikimedia.org/wikipedia/commons/c/ce/Transparent.gif License: Public domain Contribu-
tors: Own work Original artist: Edokter
• File:Ua741_opamp.jpg Source: http://upload.wikimedia.org/wikipedia/commons/4/43/Ua741_opamp.jpg License: CC-BY-3.0 Contrib-
utors: Transferred from en.wikipedia; transferred to Commons by User:BokicaK using CommonsHelper.
Original artist: Teravolt (talk). Original uploader was Teravolt at en.wikipedia
• File:VCO.jpg Source: http://upload.wikimedia.org/wikipedia/commons/a/af/VCO.jpg License: CC-BY-SA-3.0-2.5-2.0-1.0 Contribu-
tors: Own work Original artist: ^musaz
216 CHAPTER 9. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

• File:Voltage_divider_bias.PNG Source: http://upload.wikimedia.org/wikipedia/commons/f/f7/Voltage_divider_bias.PNG License:

Public domain Contributors: Transferred from en.wikipedia; transferred to Commons by User:Sfan00_IMG using CommonsHelper. Orig-
inal artist: Original uploader was Xcentaur at en.wikipedia
• File:Voltage_divider_with_cap.PNG Source: http://upload.wikimedia.org/wikipedia/commons/1/13/Voltage_divider_with_cap.PNG
License: Public domain Contributors: Transferred from en.wikipedia; transferred to Commons by User:Sfan00_IMG using
CommonsHelper. Original artist: Original uploader was Xcentaur at en.wikipedia
• File:Voltage_modulation_Class_H.jpg Source: http://upload.wikimedia.org/wikipedia/commons/7/7b/Voltage_modulation_Class_H.
jpg License: CC-BY-SA-3.0 Contributors: Own work Original artist: Braun walter
• File:Voltage_stabiliser_OA,_IEC_symbols.svg Source: http://upload.wikimedia.org/wikipedia/commons/f/fb/Voltage_stabiliser_
OA%2C_IEC_symbols.svg License: CC-BY-SA-3.0 Contributors: http://commons.wikimedia.org/wiki/File:Stab_ov.svg Original artist:
German verison: Appaloosa, Translated by Eirik1231
• File:Voltage_stabiliser_transistor,_IEC_symbols.svg Source: http://upload.wikimedia.org/wikipedia/commons/c/c3/Voltage_
stabiliser_transistor%2C_IEC_symbols.svg License: CC-BY-SA-3.0 Contributors: http://commons.wikimedia.org/wiki/File:Ser_stab.svg
Original artist: German verison: Appaloosa, Translated by Eirik1231
• File:Wikibooks-logo-en-noslogan.svg Source: http://upload.wikimedia.org/wikipedia/commons/d/df/Wikibooks-logo-en-noslogan.
svg License: CC-BY-SA-3.0 Contributors: Own work Original artist: User:Bastique, User:Ramac et al.
• File:Wikibooks-logo.svg Source: http://upload.wikimedia.org/wikipedia/commons/f/fa/Wikibooks-logo.svg License: CC-BY-SA-3.0
Contributors: Own work Original artist: User:Bastique, User:Ramac et al.
• File:Wikiversity-logo.svg Source: http://upload.wikimedia.org/wikipedia/commons/9/91/Wikiversity-logo.svg License: ? Contributors:
Snorky (optimized and cleaned up by verdy_p) Original artist: Snorky (optimized and cleaned up by verdy_p)
• File:Wiktionary-logo-en.svg Source: http://upload.wikimedia.org/wikipedia/commons/f/f8/Wiktionary-logo-en.svg License: Public
domain Contributors: Vector version of Image:Wiktionary-logo-en.png. Original artist: Vectorized by Fvasconcellos (talk · contribs),
based on original logo tossed together by Brion Vibber
• File:X_mark.svg Source: http://upload.wikimedia.org/wikipedia/commons/a/a2/X_mark.svg License: Public domain Contributors: Own
work Original artist: User:Gmaxwell
• File:Yes_check.svg Source: http://upload.wikimedia.org/wikipedia/en/f/fb/Yes_check.svg License: ? Contributors: ? Original artist: ?

9.3 Content license

• Creative Commons Attribution-Share Alike 3.0